Drsk

A grammar

2017-12-13
 i

Copyright © 2017 Isoraķatheð. Permission is granted to copy, distribute

and/or modify this document under the terms of the GNU Free Documenta-
tion License, Version 1.3 or any later version published by the Free Software
Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-
Cover Texts. A copy of the license is included in the section entitled “GNU
Free Documentation License”.
ii
 " #
11 6 8 11
O :=
19 20 11 13
" #
3 15 6 3
V := O +
−10 −6 −10 −4
" #
1 9 2 9 9
C := + 10J
2 −20 −2 2 9
D := O; V ; C
iv
Contents

I Preface xiii

Administrative xv

Introduction xvi

II Context 1

1 The world and its languages 3

1.1 A brief history of Pasaru . . . . . . . . . . . . . . . . . . . . 3
1.2 The way of the tongue . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Structure of a language . . . . . . . . . . . . . . . . . . . . . 5
1.4 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 The way of the mind . . . . . . . . . . . . . . . . . . . . . . . 9
1.6 Cultural power . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.7 Language lineage . . . . . . . . . . . . . . . . . . . . . . . . . 11

2 The region 15
2.1 Grade (1) inhabitants . . . . . . . . . . . . . . . . . . . . . . 15
2.2 Grade (2) inhabitants . . . . . . . . . . . . . . . . . . . . . . 19
2.3 Vassalship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Restitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Grade (3) inhabitants . . . . . . . . . . . . . . . . . . . . . . 25

3 The country 29
3.1 The Flag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2 A quick history after independence . . . . . . . . . . . . . . . 31
3.3 Provinces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.4 Currency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.5 Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.6 Rights and liberties . . . . . . . . . . . . . . . . . . . . . . . 44
3.7 Government . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.8 Unorderable hierarchy . . . . . . . . . . . . . . . . . . . . . . 46
3.9 Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.10 Minority and neighbouring languages . . . . . . . . . . . . . 50

v
vi CONTENTS

3.11 The political status of Drsk . . . . . . . . . . . . . . . . . . . 54

4 The culture 55
4.1 Daily life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.2 Food . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.3 Climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.4 Clothing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.5 Housing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.6 Music . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.7 Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

III Vocabulary & Etymology 69

5 Neighbouring and Related Languages 71

5.1 Rattssaw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.2 xRron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.3 Nnn Heeel, Genbarg and Ṕeri . . . . . . . . . . . . . . . . . . 73
5.4 Egako . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.5 Egonyota Pasaru . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.6 Internal languages . . . . . . . . . . . . . . . . . . . . . . . . 76
5.7 Other spoken languages . . . . . . . . . . . . . . . . . . . . . 77
5.8 Nonspeaking languages . . . . . . . . . . . . . . . . . . . . . 77

6 The External History of Drsk 79

6.1 Chronology and the Feature Change Rate . . . . . . . . . . . 80
6.2 General principles of language contact . . . . . . . . . . . . . 85
6.3 Phase 1: Inheritance . . . . . . . . . . . . . . . . . . . . . . . 88
6.4 Phase 2: Rattssaw invasion . . . . . . . . . . . . . . . . . . . 89
6.5 Phase 3: Rattssaw set-in . . . . . . . . . . . . . . . . . . . . 89
6.6 Self-modification . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.7 Phase 4: Grade (3) importation . . . . . . . . . . . . . . . . 89

7 The Internal History of Drsk 91

7.1 Periods of Drsk . . . . . . . . . . . . . . . . . . . . . . . . . . 91
7.2 Early Drsk . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
7.3 Middle Drsk . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
7.4 Late Drsk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

8 Vocabulary from other languages 93

8.1 Inherited words . . . . . . . . . . . . . . . . . . . . . . . . . . 93
8.2 Rattssaw words . . . . . . . . . . . . . . . . . . . . . . . . . . 93
8.3 New imports . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
CONTENTS vii

IV Phonology 97

9 Phonetic inventory 101

9.1 Divisions of the phoneme grid . . . . . . . . . . . . . . . . . . 103
9.2 Squares . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
9.3 Dialectal (non-)variations . . . . . . . . . . . . . . . . . . . . 106

10 Phonoruns 107
10.1 Phonoruns in Rattssaw . . . . . . . . . . . . . . . . . . . . . 107
10.2 Terminators and Non-Terminators . . . . . . . . . . . . . . . 110
10.3 Phonological and lexical words . . . . . . . . . . . . . . . . . 113
10.4 Prosody . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
10.5 Stress and Intonation . . . . . . . . . . . . . . . . . . . . . . 115
10.6 Syllabic non-denial and reconciliation . . . . . . . . . . . . . 118

11 Phonotactics 121
11.1 Adjacency rules for phonemes . . . . . . . . . . . . . . . . . . 122
11.2 Adjacency rules for phonoruns . . . . . . . . . . . . . . . . . 124
11.3 The case of the middle phonemes . . . . . . . . . . . . . . . . 125
11.4 Word and sentence boundaries . . . . . . . . . . . . . . . . . 125
11.5 Order of application of phonotactic rules . . . . . . . . . . . 125

12 Vowel insertion 127

12.1 Quantification of vowel insertion . . . . . . . . . . . . . . . . 127
12.2 Vowel insertion machine . . . . . . . . . . . . . . . . . . . . . 129
12.3 Inserted vowels . . . . . . . . . . . . . . . . . . . . . . . . . . 132
12.4 Mergers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
12.5 Insertion location . . . . . . . . . . . . . . . . . . . . . . . . . 133
12.6 An Example of a Syllabic Form . . . . . . . . . . . . . . . . . 135
12.7 The Phonorun Form . . . . . . . . . . . . . . . . . . . . . . . 136
12.8 The rebirth of syllables . . . . . . . . . . . . . . . . . . . . . 137
12.9 Local variation of quantities . . . . . . . . . . . . . . . . . . 138
12.10 Table of dialects and their insertion strategy . . . . . . . . . 141

V Orthography 145

13 Written orthography 147

13.1 Romanisation . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
13.2 Rattssaw script . . . . . . . . . . . . . . . . . . . . . . . . . . 149
13.3 B.-Pasaru script . . . . . . . . . . . . . . . . . . . . . . . . . 150
13.4 Letter number . . . . . . . . . . . . . . . . . . . . . . . . . . 151
13.5 Letter names . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
13.6 Punctuation . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
13.7 A brief history of the orthography . . . . . . . . . . . . . . . 159
viii CONTENTS

14 Other orthographies 161

14.1 The state of alternate orthographies . . . . . . . . . . . . . . 161
14.2 The shape alphabet – an artistic code . . . . . . . . . . . . . 162
14.3 Tap code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
14.4 The spoken orthography – a “phonetic alphabet” . . . . . . . 166
14.5 Usage of alternate orthographies . . . . . . . . . . . . . . . . 166

VI Morphology 169

15 Morphology overview 171

15.1 The word type hierarchy . . . . . . . . . . . . . . . . . . . . 171
15.2 Colour scheme . . . . . . . . . . . . . . . . . . . . . . . . . . 172
15.3 Words and wordlets . . . . . . . . . . . . . . . . . . . . . . . 173
15.4 General comments on morphology . . . . . . . . . . . . . . . 174

16 Nouns 175
16.1 Total localisation . . . . . . . . . . . . . . . . . . . . . . . . . 175
16.2 General structure . . . . . . . . . . . . . . . . . . . . . . . . 176
16.3 Internal words vs. full nouns . . . . . . . . . . . . . . . . . . 178
16.4 Genders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
16.5 Transformation of Genders . . . . . . . . . . . . . . . . . . . 184
16.6 The noun suffix . . . . . . . . . . . . . . . . . . . . . . . . . . 186
16.7 Omission of noun components . . . . . . . . . . . . . . . . . 189
16.8 Multiple-Root nouns . . . . . . . . . . . . . . . . . . . . . . . 193
16.9 Unit names – the full story . . . . . . . . . . . . . . . . . . . 194
16.10 Place names . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
16.11 Other names . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
16.12 Name-based pronouns . . . . . . . . . . . . . . . . . . . . . . 202
16.13 Colours . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
16.14 The calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

17 Numbers 211
17.1 The digit state machine . . . . . . . . . . . . . . . . . . . . . 211
17.2 Origins and Cognates of the Digit State Machine . . . . . . . 213
17.3 Different bases . . . . . . . . . . . . . . . . . . . . . . . . . . 214
17.4 Larger numbers . . . . . . . . . . . . . . . . . . . . . . . . . . 217
17.5 Non-integers . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
17.6 Numbers and affices . . . . . . . . . . . . . . . . . . . . . . . 223
17.7 Ordinal numbers . . . . . . . . . . . . . . . . . . . . . . . . . 223
17.8 Quantities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224

18 Verbs 229
18.1 Verb delocalisation . . . . . . . . . . . . . . . . . . . . . . . . 229
CONTENTS ix

18.2 Affix localisation . . . . . . . . . . . . . . . . . . . . . . . . . 229

18.3 Noun-Verb coupling . . . . . . . . . . . . . . . . . . . . . . . 229
18.4 Crutches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
18.5 Canes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
18.6 Process, Product, Facilitator . . . . . . . . . . . . . . . . . . 229
18.7 Sporadics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
18.8 Motion and Position . . . . . . . . . . . . . . . . . . . . . . . 229
18.9 Conjunctions as verbs . . . . . . . . . . . . . . . . . . . . . . 229
18.10 Copula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

19 Dredge 231
19.1 Dredge delocalisation . . . . . . . . . . . . . . . . . . . . . . 231
19.2 Owned and ownerless dredge . . . . . . . . . . . . . . . . . . 231
19.3 Relational dredge . . . . . . . . . . . . . . . . . . . . . . . . 231
19.4 Purpose of dredge . . . . . . . . . . . . . . . . . . . . . . . . 231
19.5 Examples of dredge . . . . . . . . . . . . . . . . . . . . . . . 231
19.6 Inflection of dredge . . . . . . . . . . . . . . . . . . . . . . . . 231

20 Tense, Aspect & Mood 233

20.1 Noun-like, verb-like and phrase-like markers . . . . . . . . . . 233
20.2 Tense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
20.3 Aspect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
20.4 Mood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233

21 All other words 235

21.1 Prowords . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
21.2 Escapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
21.3 Answer words . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
21.4 Particles and Exclamations . . . . . . . . . . . . . . . . . . . 240
21.5 Camera morphemes . . . . . . . . . . . . . . . . . . . . . . . 240
21.6 Letter words . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
21.7 Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
21.8 Incomparisons . . . . . . . . . . . . . . . . . . . . . . . . . . 240
21.9 Loanwords . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
21.10 Quasinouns . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240

22 Conversions by Morphology 241

22.1 Typing words . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
22.2 Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
22.3 Gender patriation and depatriation . . . . . . . . . . . . . . . 241

VII Syntax 243

23 Word order 245

x CONTENTS

23.1 Nouns and verbs . . . . . . . . . . . . . . . . . . . . . . . . . 245

23.2 Internal words . . . . . . . . . . . . . . . . . . . . . . . . . . 247
23.3 Quasinoun detachment . . . . . . . . . . . . . . . . . . . . . 249
23.4 Modifiers and the Modified . . . . . . . . . . . . . . . . . . . 250
23.5 Noun phrases and complexes . . . . . . . . . . . . . . . . . . 252
23.6 Verb complexes . . . . . . . . . . . . . . . . . . . . . . . . . . 253
23.7 Noun-verb complexes . . . . . . . . . . . . . . . . . . . . . . 253
23.8 Idiomatic verb placement . . . . . . . . . . . . . . . . . . . . 253

24 Sentence structure 255

24.1 Additional arguments . . . . . . . . . . . . . . . . . . . . . . 255
24.2 Additional verbs . . . . . . . . . . . . . . . . . . . . . . . . . 255
24.3 TAM words . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
24.4 Questions and responses . . . . . . . . . . . . . . . . . . . . . 255
24.5 Narratives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
24.6 Small contexts . . . . . . . . . . . . . . . . . . . . . . . . . . 255
24.7 Incomplete sentences . . . . . . . . . . . . . . . . . . . . . . . 255
24.8 Dates, times and intervals . . . . . . . . . . . . . . . . . . . . 255

25 Repetition and Recursion 257

25.1 Relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . 257
25.2 Compound sentences . . . . . . . . . . . . . . . . . . . . . . . 257
25.3 Set phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
25.4 Large-scale structure . . . . . . . . . . . . . . . . . . . . . . . 257
25.5 Conversions by Syntax . . . . . . . . . . . . . . . . . . . . . . 257
25.6 Definition lists . . . . . . . . . . . . . . . . . . . . . . . . . . 257
25.7 Sentence to Verbs . . . . . . . . . . . . . . . . . . . . . . . . 257

VIII Pragmatics 259

26 Conversations 261
26.1 A status game . . . . . . . . . . . . . . . . . . . . . . . . . . 261
26.2 Casual conversation . . . . . . . . . . . . . . . . . . . . . . . 261
26.3 Formal conversation . . . . . . . . . . . . . . . . . . . . . . . 261
26.4 Multiple conversants . . . . . . . . . . . . . . . . . . . . . . . 261
26.5 Announcements . . . . . . . . . . . . . . . . . . . . . . . . . 261

27 Large-scale communication 263

27.1 Formalising communications . . . . . . . . . . . . . . . . . . 263
27.2 Brief messages . . . . . . . . . . . . . . . . . . . . . . . . . . 264
27.3 Letters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
27.4 Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
27.5 Articles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
CONTENTS xi

27.6 Books . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268

27.7 Poëtry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
27.8 Dictionaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268

28 Registers 271
28.1 Simple registers . . . . . . . . . . . . . . . . . . . . . . . . . 271
28.2 Diagrams as register . . . . . . . . . . . . . . . . . . . . . . . 271
28.3 Tables as register . . . . . . . . . . . . . . . . . . . . . . . . . 271

29 Storytelling 273
29.1 The birth of fiction . . . . . . . . . . . . . . . . . . . . . . . . 273
29.2 Relations to reälity . . . . . . . . . . . . . . . . . . . . . . . . 276
29.3 Worldstate (“International”) literature . . . . . . . . . . . . . 277
29.4 Graded literature . . . . . . . . . . . . . . . . . . . . . . . . . 283
29.5 Fact-telling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
29.6 Series recital . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
29.7 Literature and media . . . . . . . . . . . . . . . . . . . . . . 292
29.8 Literature distribution . . . . . . . . . . . . . . . . . . . . . . 293
29.9 The Battle of the Bodies . . . . . . . . . . . . . . . . . . . . 294

IX Drsk in use 297

30 The train is approaching 299

31 The *Horse1 and the Stone 301

32 A conversation at the market 303

33 Regarding a lost *sheep3 305

X Additional articles 307

A Bugscript 309
A.1 History and rationale . . . . . . . . . . . . . . . . . . . . . . 310
A.2 Basic structure . . . . . . . . . . . . . . . . . . . . . . . . . . 311
A.3 Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
A.4 Typographical terms . . . . . . . . . . . . . . . . . . . . . . . 315
A.5 Simple phonemes . . . . . . . . . . . . . . . . . . . . . . . . . 319
A.6 Compound phonemes and belts . . . . . . . . . . . . . . . . . 325
A.7 Modifier-based phonemes . . . . . . . . . . . . . . . . . . . . 327
A.8 Non-oral phonemes and syllabic denial . . . . . . . . . . . . . 327
A.9 Archiphonemes and indistict phonemes . . . . . . . . . . . . 327
A.10 Beyond phonetic notation . . . . . . . . . . . . . . . . . . . . 327
xii CONTENTS

A.11 Bugscript derivatives . . . . . . . . . . . . . . . . . . . . . . . 327

A.12 Glossary of bugscript terms . . . . . . . . . . . . . . . . . . . 327
A.13 Programmatic generation of bugscript . . . . . . . . . . . . . 327

B Octovexillology 329
B.1 Components of an ovecs . . . . . . . . . . . . . . . . . . . . . 329
B.2 Mounting a flag . . . . . . . . . . . . . . . . . . . . . . . . . 336
B.3 Making a flag . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
B.4 Representative power of a flag . . . . . . . . . . . . . . . . . 343
B.5 Glossary of ovecs terms . . . . . . . . . . . . . . . . . . . . . 343

XI Appendices 345

C Lexicon 347
C.1 Etymological information . . . . . . . . . . . . . . . . . . . . 347
C.2 Entry samples . . . . . . . . . . . . . . . . . . . . . . . . . . 348
C.3 Key to abbreviations . . . . . . . . . . . . . . . . . . . . . . . 349

D Reference 351
Book 23 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
Book 24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
Books 25 to 27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
Book 28 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Book 29 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Books 30 and 31 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Books 32 to 34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354

E The language skeleton 355

F Useful lists 357

Month names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357

G Colour list 359

XII Postface 361

History 363

Acknowledgements 365

Afterword 367

GNU Free Documentation License 369

 Part I

Preface

xiii
Administrative

• This version of the Drsk grammar:-

– was generated at 2021-01-15 [Link], and

– is version M-202101.

• The topic symbol for Drsk is circle-P, triangle-D.

• The topic string is “Dr”.

• Where any conflict between this grammar and the Book of Conworlds
exists, the latter takes precedence unless:-

– A reference exists from this book to the Book of Conworlds, in

which case this grammar takes precedence, or
– It is otherwise indicated

• This book is not intended to:-

– Make, substantiate, endorse or otherwise implicate any pronounce-

ments on linguistics.
– Make, substantiate, endorse, or otherwise implicate any state-
ment of fact about languages on Earth.
– Incite, create, or plan any revolution in any context.
– Cure or treat any disease or condition.

• This document has editable source code in .org format (along with an
additional [Link] file for miscellaneous formatting) and is avail-
able upon request, or the reader can retrieve it himself from the official
repository of this grammar, located at [Link]
isoraqathedh/drsk-grammar. The URL may be subject to change
at a future date without further notice.

xv
Introduction

Drsk /drɨsk/ is a language that is spoken in a foreign planet. Though the

physical forms of the speakers are mostly human, their underlying psychol-
ogy is somewhat different. In spite of all the similarities regarding most of
the superficial trappings of society in general, they have values, morals and
outlooks on society that are sufficiently different to contemporary human
norms that any description of Drsk would be incomplete without them, and
be harder to understand for the lack of them.
Therefore, this grammar contains an initial section, part II, which lays
bare the context behind the civilisation that spawned this language. We
shall start with the planet as a whole, then drill down to the region that
speaks Drsk. At the same time, we will also move forward in time from the
beginning of civilisation to a time that should be familiar to most of the
readers of this grammar. However, we will not be going into the older forms
of the language, and only briefly touch on the history of the language.
As such, this grammar gives an overview of the language at all levels of
detail in one particular point in time, from the very beginnings in phonology
all the way up to short passages. The whole text would be built upon
a linguistic theory that’s completely unusable on Earth languages and is
invented mostly for describing languages on this planet where it enjoys much
more success, by fiat. Any pronouncements of a linguistic fact should be
regarded as being for this planet only, unless indicated otherwise.
At times we will call upon the original books that described this lan-
guage, which is typically called the “Book of Conworlds”. This will serve as
our primary (and indeed only) citation in the work. Where the citation is
required, the book and page number (and subpage letter, if necessary) is
provided in the form (Book | Page and subpage).
Wherever possible, translations are provided in the most natural English
that still distinguishes the relevant grammatical features. Certain culture-
specific words are avoided in favour of more universal concepts, but where
this is not possible, a sensible English closest equivalent is used, which is
marked with a preceding star (*), and in some cases a subscript where more
than one term corresponds to the same English word.
This happens a lot with animals and certain names, where it is genuinely
not possible to substitute familiar examples. For some names it is not par-

xvi
 xvii

ticularly difficult to leave them as-is as is standard for the literary tradition
in English; however for animals we shall “translate” them to their nearest
functional equivalents on Earth, which inevitably requires subscripting as it
is not guaranteed that an animal on one world is a drop-in replacement for
one in the other.
Other terms that would have this star treatment include technical terms
for certain cultural concepts, names for materials, and certain more unusual
divisions of familiar spaces. Names of units (an individual in society), certain
other technical terms for certain other cultural concepts, and place names
will not use the star and instead be rendered as they are, where technical
limitations permit.
As a project, the Drsk language is still undergoing some development,
though it is largely stable as it is. This document will continue to grow as
areas of the language receives attention, but anything that is already here
would not likely change.
xviii INTRODUCTION
Part II

Context

1
Chapter 1

The world and its languages

1.1 A brief history of Pasaru

Pasaru is a planet orbiting a sun named Setura. It is very similar to Earth
in many ways, including size, mass and distance to the parent star, although
Setura is somewhat weaker than the Sun and Pasaru orbits somewhat closer
in. Pasaru contains all the trappings that an Earthlike planet would have.
Relevant to Drsk is that it has things like mountains, rivers, tectonic plates,
and similar things, all of which are concordant to the laws of physics.
It is inhabited by a species known as kilis, which superficially look like
humans and share most of the physical characteristics but have some fairly
significant differences which we will get on to describing later.
The history of civilisation in Pasaru is much more certain than that
of Earth’s, and the individual stages are fairly easy to understand if a bit
drawn out – on average a societal change that would take 1 year on Earth
would take about 10 times as long in Pasaru, with only a modest change in
the average lifespan in contemporary times. We start our story a little bit
later however, when the first of the kilis arrive in what will soon be called
Apurhagat, a large peninsula in the northern hemisphere that plays host to
the nations that will one day speak Drsk.
These civilisations are in broad strokes similar to those of contemporary
Earth’s, modulo a handful of tiny settings that have been tweaked somehow.
The civilisation is social, has a broad sense of hierarchy (although in some
cases this hierarchy might not have total or partial order as it does on
Earth), finds some things good and other things abhorrent, and sometimes
these might even be explicitly expressed as affecting a quality named “order”.
As the thousands of years roll by, this single civilisation splits apart as
they usually do, becoming branches of a since-forgotten ancestor. Current
scholars call this old civilisation the Soþip, which also happens to be the
name of the language that it speaks for reasons that will be made clear later.
These daughter civilisations, which soon become individual proto-nations

3
4 CHAPTER 1. THE WORLD AND ITS LANGUAGES

and other related names for countries, soon begin the usual “international
drama”, which is expressed quantitatively: that is, the gain and loss of such
items as territory, influence, and similar.
This back-and-forth of motion continues for some (finite!) amount of
time, and while it would not last forever, it does last until at least the
period in which the Drsk as described in this grammar is spoken, which
would in the roughest terms describe a technology level comparable to the
50 years surrounding 2000 on Earth. The exact date does not matter for
reasons which we will get to in section 3.2.
Since this is a grammar of a particular language inside this planet, we
will only go into as much detail as is required to accurately and faithfully
describe the features of Drsk and how it has gotten to this stage. This can
be done using a two-pronged approach; in section 1.2, we shall describe the
way that languages in general work in this planet, and in section 1.5, we
shall describe how these languages evolve throughout the ages, and also a
little bit of the reasoning behind these differences, including in no small part
a large amount of mysticism presented as a part of reälity itself.
It should be noted here that apart from the name of the language itself,
none of the names in this section are in Drsk. This should be an indication
as to the language’s overall importance.

1.2 The way of the tongue

The languages of Pasaru are very similar to the languages of Earth, yet at
times the former can have features that would be very hostile to human
mouths and therefore won’t be possible in practise for the latter. Naturally
this derives from the fact that the speakers of languages such as Drsk are not
exactly human in the head even as their physical forms are similar enough
that IPA remains as useful as it is for languages of Earth, and perhaps even
more so.
The primary difference in the languages of the two planets are exemplified
in the way that they are structured. It is somewhat hopeless to attempt to
describe the overall “structure” of all languages on Earth, but in Pasaru
such a project is not unfeasible – languages there are subject to a single
underpinning theory known as features, which are essentially one very core
concept expressible using a single sentence or thereabouts that is expressed
by way of tables and formulae. This difference makes it very amenable to
being described in the form of a grammar. A further discussion of feature
theory is shown in section 1.3.
Otherwise, another clear difference in languages in Pasaru is that it is
otherwise very regular compared to languages on Earth. That is to say,
formulae that predict the formation of a particular system tend to fail a lot
less often, whether measuring by word or by use.
1.3. STRUCTURE OF A LANGUAGE 5

1.3 Structure of a language

Languages in Pasaru are described using feature theory, a series of axioms
that model languages as abstract mathematical objects that have several
fixed properties.
Languages, as a general, abstract object, contain three components.

Identity This component captures the idea that two units that might have
different internal grammars might be speaking the same language. Its
main role is to control language taxonomy and itself consists of a
unique identifier – a genuine name (though a more accurate treat-
ment of how names work will be provided later) – and an “evolution
parameter”, which is a single letter that changes more or less randomly
once a certain number of sound changes are accumulated.

Skeleton This component is common to all languages. It provides a struc-

ture upon which features can stick to and build on. Its form can be
considered to be the “heart” of a given language.
The language skeleton is displayed as a single graph, which is displayed
in Figure 1.1. (A | 3)

Paradigms In practise languages cannot remain abstract forever and have

to eventually “descend” to a practical plane. The conversion from
abstract skeleton to physical objects is handled by paradigms.
This component contains a handful of other objects, but its exact
subcomponent is dependent on its final medium. For most languages
it will include things like inflection tables and vocabulary lists, with
spoken languages also having phonemic inventories and similar.
Generally speaking, paradigms are placed as “nodes” in a skeleton and
are controlled by the eponymous features in feature theory.

One can go quite far in imagining the skeleton as the “model”, contain-
ing lots of other abstract objects and arrows that link between them; and
paradigms as the “realisation” of that model, which control the expression
of these in a more physical model.
Within each of the three objects noted above, there can be more sub-
objects placed within. For spoken languages, we can also add as generic
properties the following items:

Phonology This is a paradigm that transforms a word into tangible ob-

jects, through an intermediate representation known as “phonemes”.
In spoken languages the tangible object in question are sound waves,
and this is what phonology specifically refers to.
6 CHAPTER 1. THE WORLD AND ITS LANGUAGES

Orthography Alternatively, a paradigm that transforms a word into tan-

gible written objects, usually called “graphs” and “graphemes”, is called
an orthography. This may have something to do with the phonology,
but there is no guarantee that it would do so. A spoken language has
no requirement for this component, but it comes up regularly enough
to describe it here.

It should be noted that romanisation would be included here as well.

Morphology This is a part of the skeleton that contains the “word list”.
Intuitively, as most monoglots would have you believe, this is the heart
of a language. In feature theory, it is considered somewhat unusual in
that features are the least structured, but otherwise it is a perfectly
normal part of the language skeleton.

Lexicon Here the results of morphology – morphemes – are connected to-

gether into “dictionary words”. This is also a part in which most non-
linguists would say is an essential part of the language, but it is even
less of one here than for morphology. However, most dictionaries and
similar work on this level, omitting a large part of the mechanisms
lurking in the morphology.

Syntax This is the part of the skeleton where individual words are combined
to form sentences. This includes word order for the most part, but
inflation tables, which for some languages do the same job, are also
included here. This part of the language is what most others would
call the “grammar”.

Pragmatics This is the part where we assemble sentences together to form

entire utterances and passages. It is the most abstract level, not only
in the sense that it is very high up in the language skeleton, but also
that it is able to express so many things that it is essentially insulated
from reälity in all ways.

Putting it all together, we can create the language skeleton for spoken
language, which looks like Figure 1.1.
Each of the objects in the skeleton depend on the objects below it, with
the root nodes being free to be anything the Will chooses it to be (or more
carefully, is not predicted from the theory). Notice that some of the com-
ponents are nodes, and others are edges. Very roughly speaking, nodes
represent objects that are meant to be altered by other objects, which are
represented by edges.
1.3. STRUCTURE OF A LANGUAGE 7

Passages

Pragmatics

Sentences

Script
Semantics Grammar Syntax
(orthography)

Lexicon

Morphology

Phonemes
Phonotactics Morphemes
(phonology)

Figure 1.1: The skeleton for every language in Pasaru. Uncoloured nodes
represent places where mechanisms alter things, whereas the coloured nodes
represent paraidgms and lists. coloured nodes are coloured according to
their abstraction level, with red being the lowest and blue being the highest.
8 CHAPTER 1. THE WORLD AND ITS LANGUAGES

1.4 Features
With the skeleton in place, we can now talk about features.1
Like most other things, features contain a number of substructures, which
are listed below:

Core This is a single sentence, which may or may not have blanks in it.
For example:

• “Limit the number of … to …”

• “Divide the semantic space regarding … to an unusual level”
• “Reuse structures that used to describe …”
• “Do not include … which usually is a consequence of …, a thing
that is included”
• “Pathological topologies” (of any sort).

Occasionally however a feature can be sufficiently complex that it can-

not be described using a single sentence. These are called “irregular
features” and are treated as just regular features that merely lack the
summary.

Parameters Parameters represent two things: the first being the words
that fill in the blanks in the core, and any other tweaks that are re-
quired for the feature to fit in the language. For the feature “Reuse
structures that used to describe …”, an example of the first type is sim-
ply the phrase “ordinal numbers”, and the second type would specify
how much of the structures of ordinal numbers to be included.

Paradigms This final component lists the words that are necessary to bring
the feature to the language. Usually this comes in the form of tables
and word lists, which is why it is called a paradigm. For the fea-
ture “Reuse structures that used to describe ordinal numbers”, the
word lists would definitely include things like names for ω, addition,
multiplication and exponentiation, with inflections perhaps being less
suitable.

We can then bond the feature to one node of the language skeleton.
This is usually done via the paradigm, but one feature can also be explicitly
dependent on another feature. All of this can be manipulated directly using
diachronics.
Features are then combined together in a controlling feature called a
system. Systems help corral large amounts of features and to control their
1
Elsewhere in this book, the word “feature” will be used in a context-dependent manner,
and may have multiple meanings – this is very important to note, such as when we discuss
phonological features such as [+voiced].
1.5. THE WAY OF THE MIND 9

numbers, and are probably “real” in the sense that linguistic processes (di-
achronic and synchronic) do operate on them occasionally, but on the other
hand some linguistic processes do not.
Furthermore, individual features can also be connected to each other via
the “mechanism-pathway” system. In this case, we speak of a “mechanism” –
an object that reflects a change in some level of the language skeleton, such
as an inflectional paradigm, or a mutation of some kind – and a “pathway”,
which is a trigger for that mechanism to occur. There’s no requirement
that they match up one-to-one, and this can make language analysis a lit-
tle bit less straightforward than what a naïve assumption might provide.
This mechanism too can be operated on directly but this time only using
diachronic processes.
Though we speak at length about features here, it is not an explicit
part of Pasaru linguistics, though it is helpful to keep it in mind to help
organise the language as an object of study. This grammar is in fact partially
structured so that features and systems are linked up together.

1.5 The way of the mind

As previously mentioned the mind of a kilis is very different from that of a
typical human, even though the two might look fairly similar to each other.
These differences can be a little bit unwieldy to get one’s head around but
for the most part unless the situation does involve society in some way one
can more or less ignore it. Nevertheless the mind does play a big role in
language change and evolution so we shall briefly describe it here.
The most important thing that one must know regarding the psychology
of a kilis and the languages it speaks is that the two are equals; instead of
having language be a subordinate of the mind, the language has a “mind” of
its own as well, independent of any individual speaker and in some cases can
appear to dictate its own fate, though this so-called “Will of the Tongue” has
no physical form and is not a scientific concept. A slightly less mystical way
of expressing this is that whatever is usually indicated as being an expression
of the Will is simply not predicted by the facts that are accommodated by
feature theory.
Nevertheless, the most visceral way that language change occurs in these
two planets is that in Pasaru, language change happens quite literally overnight,
as a discrete application of a specific change over a given population. Com-
plications can still occur, however, as there is no requirement that the pop-
ulations are previously named, well-known, or have any sort of societal im-
portance. (They do however have the requirement of being compact geo-
graphically.)
That particular change can be effected by various agents, and one of the
most peculiar behaviours is the fact that celebrities – linguistic celebrities,
10 CHAPTER 1. THE WORLD AND ITS LANGUAGES

even – has the ability to, in effect, temporarily become the Will of the Tongue
and enforce language change essentially just by proclaiming it is. This is the
most visible way that a Will can manifest itself in the real world, however
briefly, and it shows that in Pasaru, prescriptivists can very well become
change agents of the language. (16 | 52) (16 | 53)
Regardless of its actual existence, the Will of the Tongue exerts a no-
ticeable effect on the real world, and in some cases exhibit some seriously
threatening behaviour. In particular, languages are typically “maps”, as op-
posed to “territory”; they are a particular representation of a reälity that
would not bend to inelegant divisions of the semantic space – except, on
Pasaru languages have a somewhat nasty habit of ignoring that and liter-
ally becoming the territory, somehow exerting enough force upon both the
physical reälity and the perception of that reälity by its speakers to change
both to suit its own ends. This is not inherently worrisome, no more than
a magical language would be, but as of yet the boundaries that limit this
reälity-changing mechanism are a scratch too far from the dangerous end of
the Tinkerbell effect.
It is currently known that languages cannot change physical quantities
to a measurable degree, and that one cannot define physical laws out of ex-
istence, but slightly less “close-to-the-metal” properties like philosophy, psy-
chology and politics have been known to change on the drop of a coin simply
by a convenient language change passing by. Psychology on this planet is
therefore somewhat limited because any particular phenomenon carries an
appreciable risk of being invalidated on the virtue of its own publication.
Grammars however are somewhat safer as they only describe what actually
happens without making any predictions and prescriptions, and Wills cur-
rently are not known to violate causality by changing history. However, the
publication of any such grammar can cause languages to change course in
some way or another, usually by affirming any ad-hoc or implied structures
that was used to explain one language or another.

1.6 Cultural power

Like languages, a culture also has a Will. In principle this Will is of equal
power to a language’s, and in the currently popular theory they are placed
as such, below what is termed the “heart of the culture” (25 | 30).
In practise the two languages aren’t exactly equal; though in some slightly
vague terms they are equal given enough time, over the short term one may
overpower the other, and the reason why the notion of “overpowering” is
vague is that the Will of the Language alone has the power to alter reälity
directly, whereas the Will of the Culture can only do so by altering cultural
parameters. In this sense, the Will of the Language is stronger than the
Will of the Culture.
1.7. LANGUAGE LINEAGE 11

It should be noted while this may look like a personification, Wills are
decidedly not so; Wills do not look like anything, nor are their behaviour
in any way like either humans or kilis. In fact, a distinguishing feature of
Wills is that they do not even have a consistent sense of identity; as befits
the objects that they represent, they are prone to split and combine (but
mostly split). They may be able to “choose”, in some abstract ways, which
rules to follow and which rules not to, but the fact that they do not carry
a persistent identity throughout their lifetimes makes them very different
from real, living things.
A very notable difference between living things, Wills and the Heart is
that the latter two have two options toward no longer existing: they can die,
in which case things behave much like a typical death of a living thing; or
they can be succeeded, in which case the object in question merely morphs
into one or more other objects continuously in such a way that the boundary
is often hard to notice in the moment, or even after the fact. More relevant
to languages, language death is when a language’s last speakers dies out,
but a language succession is when a language smoothly evolves into one or
more daughter languages. The latter designation usually applies to language
families, which are explicitly identified with the proto-language. The idea
of succession is very nicely translated because not only is the language has
had a successor, but that its job as a communication tool has had success
and can therefore be retired.

1.7 Language lineage

Languages in Pasaru evolve from three proto-Worlds; Drsk comes from the
northernmost one of the three, proto-Soþip. Drilling down from there, Drsk
descends from proto-Soþip through Kebois and Rxut into a family of lan-
guages called the Remmsp, (14 | 60) from which it is a direct descendent
of. Figure 1.2 shows a more complete version of the full family tree that
eventually birthed Drsk. Interestingly, there is also a geographical version
of this development, and they line up fairly well.
However, it should be here that languages do not carry descriptors
such as “Early”, “Middle”, “Late”, “Modern”, “Upper”, “Lower”, &c., to de-
note either space or time distribution. For time in particular languages go
through eras that are identified by a single letter (or syllable, or any singular
grapheme in the language’s script, or some other civilisation’s script.) This
means that there’s no way to determine the age of a particular language
from its age letter, as that single grapheme is known.
In our case, Drsk starts off from Drsk S which branches off from Remmsp
E, and S slowly follows through F, D, a period where Rattssaw suppressed
Drsk, a resulting creole known as R-Drsk A (and the corresponding Drsk
12 CHAPTER 1. THE WORLD AND ITS LANGUAGES

era P), the successor of Drsk RI , Lˡ, T, and so on. (23 | 54) The current age
letter in effect is Drsk M.
Age letters can also be used to describe other things, such as ages of
instability where a strong wave of language change occurs. In Drsk, these
typically mask the transitional period between age letters, so for example a
difficult time between S and F is typically named VE .
The existence of such waves are an interesting feature of how languages
change, and can be used to actually used to identify the start and end-points
of a language in a diachronic sense. Since languages are relatively stable
between waves, continuity is easy to establish and therefore the lifespan of
a language can easily be determined down to decades. (36 | 33)
Dialectal variations are also named using a single letter sometimes, but
since most of them have a very strong geographical connection they are also
named using toponyms instead.
Era letters can be used to track how they change over time. More infor-
mation of how fast, and in what ways they change, are in chapter 6.
1.7. LANGUAGE LINEAGE 13

Ancient languages

Proto-Soþip

Kebois

New Kebois Spaargri Rxut

Proto-Remmsp

Drsk Claybæpp xRorran

Drsk S Gułryi Egako xRron Rattssaw

The Remmsp

Drsk F Aslatt

Drsk D

Drsk Ri

Drsk T

Drsk (detail)

Figure 1.2: A history of languages that are related to Drsk, as well as other
major languages. Major language families are in coloured blocks; Drsk itself
is outlined in red. Time generally increases from top to bottom. The proto-
language is in an egg shape.
14 CHAPTER 1. THE WORLD AND ITS LANGUAGES
Chapter 2

The region

For the most part the continent of Apurhagat (see figure 2.1) has cultural
assumptions that are most similar to contemporary European cultures, at
least when it comes to naming and splitting the semantic space. For instance,
for the most part names for “units” – an arbitrary object with magnitude one,
usually taken to be some kind of living object participating in at least one
widely-recognised society – are drawn from a finite and small pool, compass
headings are understood as having four points, and so on and so forth.
Nevertheless, some properties remain genuinely alien, such as toponymy
and certain societal values which would be somewhat distasteful to mention
in a grammar.
As will be mentioned later, the region’s influence is largely held by one
̲
particular country, the Empire of Senlis (EP Jesdicsenlis, hereafter J.-Senlis).
Its language, Rattssaw, is therefore fairly widespread in the language, al-
though as with some other features, it is unwise to synonymise it with Latin
and Drsk with (e.g. English) as the feature transfer of the two languages is
very different.
We shall now cover the history of this particular region by considering
what inhabitants live here at different times. Such categorisation of inhab-
itants are generally called “grades”, though they are usually unnamed. In
this section, we shall refer to them using parenthesised numbers, with 1 be-
longing to the earliest known grade. The grade analysis works because they
happen infrequently enough that the consequences of one invading group
more or less stabilise before the next group comes in. (28 | 35)

2.1 Grade (1) inhabitants

So called because they are the first settlers of the region, they come mostly
from the south, that is, from the opposite side of where the J.-Senlis inhabi-
tants came from. This is largely evidenced by language movement patterns,
where the proto-Soþip languages come in from the north whereas the proto-

15
16 CHAPTER 2. THE REGION

Figure 2.1: A map of the world of Pasaru, at the present day. The continent
of Apurhagat is highlighted in yellow and blue; The region that we are
interested in, and therefore the region that speaks Drsk, is highlighted in
blue. Labels are in E.-Pasaru.

Wans languages come in from the direction for which it is named. They
aren’t homogeneous, but they aren’t exactly bustling with diversity either,
with about five or six distinct civilisations making up the grade in its en-
tirety. The names of these civilisations have been lost for the most part, but
they have been given names reconstructed from Rattssaw data and they do
maintain the collective name derived from the Remmsp.
This grade of inhabitants therefore share a great deal of cultural qualities
with the neighbours to the south, and in some sense “belong” to this region.
Culture however is heavily influenced by later grades to the point where
separating culture from this grade from culture imported from later grades
can be somewhat tricky, but it is doable.
We will now briefly speak of some pertinent qualities of the culture sur-
rounding Grade (1) inhabitants, specifically those that can be gathered by
briefly mentioning small quantities. These are directly relevant to the lan-
guage, as one would readily observe.

2.1.1 Names(1)
Units receive more than one name at any one time, in a scheme known as
“equatorial naming” (22 | 48). This naming scheme, endemic to this region,
has names change periodically to satisfy a relation that involves far more
2.1. GRADE (1) INHABITANTS 17

than just the identity of the unit to be named. It can also depend on the
time of day, the time of year, temperature, local wind direction, success of
one’s business, and so on. What is involved and what isn’t depends on the
subcultures.
In the case of Drsk speakers, the naming system is rather mild, with
the only time-variable objects in the name sectioned off into a segmentable
portion of the name and limited to eleven variable objects. More details will
be provided in section 16.9.
Each unit only has one name at a time, and do not otherwise carry family
names, tribe names, honorary names, or any such similar items, though there
are communal names and status markers which are part of the name but not
frequently expressed. The exception comes with those with great societal
power who have, in addition to this variable Equatorial name, a self-declared
Regal name, which is also one word long.
Names are generally formulaic but otherwise meaningless, as with the
rest of the planet. Any valid combination of phonemes that does not make
a word when the name was first brought up is a valid name. Name reuse is
technically allowed but is highly disfavoured, but changing the colour in the
time-independent quantities status as well as starting with a different “salt”
can help with this.
A full treatment of names is available in section 16.9.

2.1.2 Ages(1)
Units have two separate time periods before adulthood, and two afterward.
The exact age boundaries are vague, but the first two largely correspond to
childhood and teenagers while the last two are seniors and “super-seniors”.
Each class has its own expected duties and benefits, and there is a system
of ritualised contact between these age groups.

2.1.3 Navigation(1)
Compass roses for grade (1) culture, like cultures around this region, tend
to have four points: north, east, south and west. Other cultures don’t have
this though, preferring six points to the compass, one for each 60°. (22 | 47)

2.1.4 Calendar(1)
Grade (1) inhabitants track not single days, but treats collections of three
days as the unit of timekeeping that all other timekeeping units are based on.
This three-day period is called a triad, mainly for reasons of convenience.
The year is divided into 238 of these triads, with key triads named and
the rest numbered from those named triads. The named triads refer to
important times for an agrarian culture such as harvest, sowing or chopping
down planted trees, and there are 36 of them. As a result they do not
18 CHAPTER 2. THE REGION

share an equal amount of triads between them, but they don’t even try; the
smallest gap between two named triads, the triad of planting and the triad
of breeding of *cows, has only 3 triads between them.
Triads themselves are culturally significant, as they take the same social
role as the day does in other regions. Something that foreigners might
consider a part of some “daily routine” such as exercise, working hours, and
social hours would instead be assigned to one of the three days in a triad
instead, in effect making the social day inside the Northern hemisphere three
times longer than the social day outside. This has some pretty intense
consequences. (27 | 34)

2.1.5 Religion(1)
Grade (1) inhabitants generally involve themselves in a semi-public display
of polytheism. The total number of gods is overwhelming, perhaps more
than three unique gods per individual worshipper, and as times go on the
number inflates greatly. In a sense they are more like cards in a trading card
game where you can make up your own cards, than objects of worship in
their own right. Were there any other religions that have been believed in
Grade (1), they would have been subsumed into this one, for reasons that
we shall see later.
Nevertheless there about 100 gods that escape this endless divine infla-
tion, and are well known throughout the reign of the Grade (1) inhabitants.
They are not believed to limit themselves to one or more duties; instead the
belief is that only their incapabilities are enumerable, and they are free to
do almost anything outside of such. Hence, there’s no “god of fertility”, or
“god of fortune, wisdom and having four fingers”; if you wish to pray for
some situation, you are only restricted from requesting from some gods, not
restricted to requesting from them.
(This remains true for all the other many millions of gods that are present
in a more private venue but because of the private nature this is not always
apparent.)
Worship largely revolves around having the bewildering pantheon turn
on itself so that they do not band together and somehow think of destroying
the living world. This can be done by essentially naming more gods that can
counter the possible behaviours of the existing ones, but often this happens
by naming existing gods – possibly from other faiths – and essentially locking
them up in a fighting network by some incredibly tortuous story that leaves
all of them stuck with each other for an indefinite period of time. However,
there are some powers that are simply that powerful, and to that the only
solution is to invent more restraining gods to keep that one busy. It’s possible
to pick up traces of older religions in the more established deities, because
for whatever reason their powers seem to be somewhat weak compared to
contemporary ones, and they are particularly tight-knit in their own wars.
2.2. GRADE (2) INHABITANTS 19

There is a sense that the individual gods are not at worship here, and
the thing that is revered is the never-ending network of red string and names
on card that this endless game of oneupmanship produces as a byproduct.
This is an astute observation, and one that will eventually bring us to the
end of this particular faith, and to all faiths in general on this planet, for
which see Section 3.2.

2.1.6 Language legacy(1)

This layer of inhabitants contributes the most to the development of the
language. Most linguistic metaphors spoken from day to day draw from ex-
perience in this layer, especially with regards to the single rapidly-expanding
religion explained above.
Despite this a lot of the phrases and constructions in this region are
overshadowed by similar ones in the next layer, which is partly by design
and partly by accident, as many languages do.

2.2 Grade (2) inhabitants

Invasion of the J.-Senlis

At some point the empire to the east took a bit of a liking to the lands that
the Grade (1) inhabitants that currently inhabit the land. After a short
and frighteningly efficient time, they did manage, and the J.-Senlis became
the effective rulers of the region. Though they have a consistent culture
and a well-known name (the Rattssaw speakers), we shall refer to them for
consistency as Grade (2) inhabitants.
Qualitatively speaking the Grade (2) inhabitants are from the same fam-
ily as the Grade (1) inhabitants, though of course this has never implied any
particular feelings of brotherhood between the ruler and the ruled. Never-
theless, their cultural qualities are very similar:

2.2.1 Names(2)
In Grade (2), names use the so-called Northern-naming technique, (19 | 1)
where no one has a name, and everyone is given a name, and probably many
of them. Each time a unit encounters or joins a social group, he is always
given a new name, which is decided on by the group with no further input
by the joining unit. This includes the government, which gives everyone a
name, and also the parents, who gives every child a name.
It is not unusual for the given name to be “rescinded” and replaced with
another name when situations arise that require it, usually at times of hard-
ship, or at major societal milestones. It is also not particularly unusual to
have names that are insulting, and again it should be noted that the receiver
20 CHAPTER 2. THE REGION

usually has no recourse to that, though this is very rarely done and is a ma-
jor taboo if misused. Otherwise, once assigned this name does not change,
come rain or snow, come day or night, come summer or winter.
Names are drawn from a pool of known “good” names, which can vary
from place to place and from time to time. Unlike many other places in
Pasaru, names are not randomly generated and cannot just be any old com-
bination of phonemes that fits the phonology of the language. An exception
is made for names given by the government, which due to several issues
regarding making enough names now instead issues a randomly-generated
number for its names, making it resemble more of an ID than a name.2
Again, each society gives one name, and one name only, to each member.
There is a tradition of keeping government, family and professional names
together to form a “full name” of sorts (in that order), but this is an external
effect.
Some units are famous enough that they can break through this layer and
end up being called the same in all circles (bar small enough communities).
These are very important units indeed, and they are known just by this.
This is rare enough that not even all kings and emperors can attain this
rank.

2.2.2 Ages(2)

Unlike Grade (1), Grade (2) only distinguishes between child and adult, with
nothing in between. However, there is a distinguishing made after death:
there’s the recently-dead, and the ancient dead. The distinction between
the last two largely revolves around what can be said about them, and how
honest one can be about their behaviour.
Naturally, the more long ago someone has died, the more honest one can
afford to be regarding the subject. The threshold is of course a bit fuzzy,
but it usually is about a third as long as the subject had lived or two years,
whichever is longer.

2.2.3 Navigation(2)

The navigation system however is the same as in Grade (1), with four points
to the compass.

2
Indeed, some “international IDs” issued by the J.-Senlis use the Government’s name
as the ID number of the unit (plus perhaps a random salt for some level of privacy) and
some other convenient name that everyone has as the “actual name” for the international
community. It is occasional that some foreign communities that largely do not engage in
this naming scheme humour them in this case.
2.2. GRADE (2) INHABITANTS 21

2.2.4 Calendar(2)
Like in Grade (1) triads are the basic unit of measurement for the calendar,
but this time above the layer of triad there are some more divisions. Seven
triads together make a month, and there are thirty-four of these, all with
the same number of triads. These are arranged in four seasons, as would be
expected on Earth, but this is not the general case in Pasaru, as the general
international standard is to have three seasons: hot, wet and dry.

2.2.5 Religion(2)
The Grade (2) religions are spearheaded by Pnfy, which is a monotheïstic
religion whose major deïty is named the same. While it is not alone in itself,
it is heavily dominating, and most other religions live in the shadow in it.
Pnfy did not really permeate into the language all that much at least
not compared to Grade (1), so its description here shall be kept brief. Pnfy,
as a god, is not particularly interesting, as he is largely considered to be
little more than a “source”, albeit one that is particularly important. In this
theology the world is considered to have always existed, without any cause
or fate, and the role of Pnfy is to play around with it and “make sparks”, as
it were.
To do this Pnfy largely handles things by the creation of automata that
help make the process of creating and managing structures in the world a
simpler affair than it would otherwise be. It is these automata, not Pnfy
himself, that create and destroy, but since it is Pnfy that creates and main-
tains them, he is considered the object of worship.
Worship here consists mainly hoping for the creation or the encountering
of new automata that would further one’s goals toward whatever. While he
is technically the source of morals and values in the world, this portion of the
religion is usually downplayed due to the somewhat uncontrollable epidemic
of hypocrisy that permeates the positions of power in Grade (2).

2.2.6 Values(2)
In lieu of Pnfy providing this, there is a third council that is neither clergy
nor ministry that is given power to create and in some rare cases enforce
them. In some ways they resemble academia, but there is another institution
in Grade (2) culture that is actually academia so it is not normally considered
as such.
Generally speaking, Grade (2) culture values the stability in forming
structures, but while such hierarchies is largely enshrined in the system,
orderedness is not, so what happens is that transitivity of power is not
preserved in the large scale.
However, the values demand equality in other respects, such in large-
scale economics and other quantitative items. These items are inherently
22 CHAPTER 2. THE REGION

ordered because they depend on real numbers, which are, of course, ordered.
Such order is incompatible with the values, so they must be the same.

2.2.7 Language legacy(2)

The language legacy of Grade (2) is immediately visible – it is this Grade
that has given the language the gift of writing, as even today the Rattssaw
script (as seen in Section 13.2) is the primary orthography of the language.
However, beyond this obvious layer the impact of Grade (2) is lacking
in some key areas, syntax in particular. The greatest impact on the actual
language itself by this grade is in single words, connotations and things like
the formal messages. Indirect effects on the language involve the instillation
of many Grade (2) values into Drsk, including the idea of triads and the
differing role each day in a triad has.

2.3 Vassalship
After some time of direct rule, the region is left to become a tributary
state. This means that although notionally independent, the entire country
is still largely under the thumb of the J.-Senlis. However, at this point the
structure of the country is determined for the present day, and the language
boundaries are determined at roughly this point.
Language-wise, Drsk is not particularly vibrant. Though largely toler-
ated and even taught in a limited basis in the official education system, the
language is not used in the highest levels of government, which of course is
one of the only sources of prestige in the area. In some cases, the language
is functionally extinct on a provincial level, with of course the ones near the
border with the J.-Senlis being the case.
However, there are still some bright spots for the language. In particular,
the free cities near the coast are still majority Drsk-speaking and act as a
source of the language’s strength. For this reason, the so-called “standard
dialect” of Drsk is still the one spoken in the Free City of Nffek, despite the
capital being in Trokier.
The culture of the region at this moment is almost the same as the one
during direct rule, i.e. Grade (2), but now more cultural trappings of Grade
(1) are allowed to be openly expressed. The system is still extremely biased
to Grade (2) culture though, as that is what a vassalship tends to become.
During this time, the paradigms of storytelling have changed a little more
to become more Grade (2) like. The old storytelling tradition slowly died
out, but a large number of famous new stories were created at this time,
such as Twenty-Six Variations on the Behaviour of a King and A Landslip
at Dd ms, which are indeed all written by Grade (1) inhabitants.
Also, at this point, the first real efforts at trying to describe Drsk’s
grammar has been initiated. While the language has always been very well
2.4. RESTITUTION 23

documented before, as is standard for other countries, the collected “data”

in the form of a massive corpus and a catalogue of phonological information
is only analysed at this stage, thanks to the rise of modern linguistics at or
about this stage.
At this time, the region is officially named after the Rattssaw name for
Grade (1) cultures in general: “Oerds” \[RDᵈS\] (see Figure 2.2 for an exam-
ple of how it is used). This term is strictly Rattssaw and is not particularly
flattering to Grade (1), but there is an uptake amongst some of the grade,
mainly those in the southwest and southeast of current-day Klzdmk.

2.4 Restitution
As the years of the vassalship wore on the empire slowly became corrupt and
eccentric, and it also became more and more authoritarian and expansionist.
After a while, one of the emperors finally snapped and decided it is an
excellent idea to start to take over the world.
The result is something called the Final War, or the Cause-of-Restitution,
and similar appellations that make its status as one of the last wars of its
kind. As a language, however, the war is actually more or less inconsequen-
tial while it was going on as the whole country was basically held into stasis
during it, not only linguistically but also in other fields such as industry and
culture. So we’ll skip ahead when the dust has settled and a large majority
of the world delivered the message that what the J.-Senlis was doing is a big
no-no to the relevant parties by way of a metaphorical big stick.
The most important thing that happened in terms of the language is
that Klzdmk was granted its independence, with the territory that it had
during vassalship, although at that point Drsk no longer had full dominion
over its territory and most of its eastern provinces are mostly monolingual
Rattssaw. This is the main benefit that the post-war reparation negotiations
– called the Restitution – had to the region, and therefore the language.
Quickly after this Drsk is given a secured area for it to thrive, and in a
nice turn-around the total area that is Drsk-speaking at this point is greater
than it was at any previous point in history. However, with this status, it
also has to be for the first time more mindful of the languages that it has
replaced in its stead, of which only a small handful has survived. Most of
them were and still are considered “paraphyletic linguistic clades” of Drsk,
(which essentially amounts to calling them somewhere between dialect and
sister language) which makes their status ambiguous and therefore are in
danger. Linguistically this means they are given era letters that refer to
Drsk, but they also get their own names, which are the “paraphyletic” part
of the saying – that these are not actual languages because they are just
“paraphyletic” clades branching off from Drsk.
24 CHAPTER 2. THE REGION

Figure 2.2: A map of the J.-Senlis just after Restitution. The flags of each
of the four cultures that form the empire are included at the bottom, as well
as the national flag of the J.-Senlis. The newly liberated Klzdmk is seen to
the west, labelled Oerds.
2.5. GRADE (3) INHABITANTS 25

The somewhat tricky situation here is not simplified by the fact that
Rattssaw remains a healthy linguistic community inside Klzdmk, and is still
highly popular for migration toward the still-very-rich J.-Senlis.

2.5 Grade (3) inhabitants

Globalised immigration

The J.-Senlis is not the only threat to Drsk after the war. As trade routes
open up, increasingly greater and greater amounts of units get more and
more cosmopolitan and migratory, and they bring their language with them.
In the history of Klzdmk, these immigration patterns are considered a
“third grade” of inhabitants, such are their numbers and their impact to
the region. Unlike the Grade (2) inhabitants and to a lesser extent the
Grade (1) inhabitants, these are clearly not a homogeneous group but are
a vast collection of groups that are united only to the extent that they
are foreigners. Their values then, and consequently their other “cultural
parameters”, are generalised and not specific to any one culture.
Nevertheless, Grade (3) is still said to have some cultural parameters –
this is mainly because by this point the balance of power has moved south,
and the capital of the world has at this point already been established as
such. Therefore, Grade (3) most strongly reflects the values and parameters
that originate from the planetary capital, which is named Kalendiyë, located
in the equatorial continent of Ordžojan and also the national capital of
Vohalyo.3 And so here they are.
A few other details are present in Section 3.2 and Section 29.1.

2.5.1 Names(3)
In contrast to native names, Grade (3) names typically use the Southern
naming technique, which is much more familiar to most readers than the
other naming traditions present in the country.
In this naming convention, names are defined by the one who has them,
and remains the same no matter what society or group you belong to. While
this sounds entirely conventional, keep in mind that this is also true when
the unit has no name as well, as is the case when he is a baby. Names are
not recorded until a subject is old enough to give one to himself, and before
that he is usually referred to indirectly, e.g. by number in the family or
some sort of serial number.
3
It’s not a straightforward relationship, but the long and short of it is that most of the
motivation toward this immigration comes through Vohalyo because it is the global capital,
and the population then disperses from there. While Klzdmk still gets local immigration, it
now also gets immigration from far across the world by way of Vohalyo.
26 CHAPTER 2. THE REGION

2.5.2 Ages(3)
There are also two main stages for ages in Grade (3), although in more recent
times the gap between them has become a separate age range in itself.
Per our rules regarding translating names, this period is called being
a *teenager, but while the behaviours of this period is similar the exact
time period this transition period lasts is not the same as on Earth, with
the general period being approximately one twenty-fifth of one’s natural
lifecycle. By the beginning of the influence of Grade (3) this is about 10
years or more.

2.5.3 Navigation(3)
The six-point compass is popular in most of the world, and as might be
expected has one point pointing north and the rest at 60° intervals from
each other. Most civilisations also have specialised terms to refer to east
and west, but they are not common and are different from field to field,
reïnforcing their secondary nature.
The remaining four words that represent the 60° directions are imported
for use into Drsk more or less as directly as any foreign word would be, with
relatively little implementation.

2.5.4 Calendar(3)
The Grade (3) calendar is as expected the international calendar, and does
not have the concept of triads. Instead the more familiar concept of weeks
and months are used in this calendar.
The 714 days in the year are divided into 21 months of 34 days. It is
also divided into 1080 yearly degrees, which then goes through the minutes
and seconds in the expected manner.
As the calendar was developed in the subtropical areas on the subtropical
areas west coast of the Mirada Ocean – that is to say, the homeland of the
E.-Pasaru speakers, in Calendia – there is also season-based divisions in
it. Rather than the more temperate-based spring, summer, autumn and
winter,4 or the more tropical based wet and dry seasons, the 21 months are
divided into three seasons: wet, dry and hot. There’s no doubt that this
division is partially due to the fact that 21 is divisible by 3 and not 2 or 4.
For the names of the months in this calendar, which is used elsewhere in
the book, see appendix F.

4
The four seasons exist and are recognised as a common concept in Pasaru but they are
by no means universal or inviolable. For instance, Drsk, amongst other languages mostly
spoken in the far north, tend to use the same word for spring and autumn.
2.5. GRADE (3) INHABITANTS 27

2.5.5 Religion(3)
At the point of the introduction of Grade (3), the religious scene has had a
vast amount of unusual developments that makes it so that the end result is
not particular recognisable to humans as religion. The details are explained
in section 29.1, but the summary is that religion at this time has evolved
from being a source of morality or ethics to the beginnings of fiction, which
have hitherto been undiscovered.

2.5.6 Values(3)
Unlike values(2) , values(3) is not a list of moral imperatives but a generalised
set of ideas that are brought in with Grade (3) civilisation. This includes
things that lack any kind of moral or ethical value such as names for new
elements, words for modern economics and trade, and concepts relating to
mathematics, computing and management, though of course there are also
some other items of ethical interest that are included in the system.
The “modern” (i.e. Grade (3)) set of values is that of a seemingly pre-
carious balance between freedom and hierarchy, and it strikes at the very
heart of a fundamental difference between humans and kilis. This difference
is that the average kilis is much weaker and less of an individual than a
typical human is, and they are heavily dependent on the existence of some
hierarchy which relays instructions up and down (and sometimes around,
because of the possibility of unrankable hierarchies) the chain of command.
Nevertheless the individual still does exist as an independent object, and an
individual can still – with great difficulty – “act independently” as a home-
steader might do.
It is this independence that Grade (3) civilisation hopes to eventually
achieve, as impossible as a full implementation might be.
28 CHAPTER 2. THE REGION
Chapter 3

The country

Name (Rattssaw) Klzdmk

Name (Drsk) Klssrdmk

Population 75 × 106 National Flag

Drives on left, between lines

Voltage standard A

Time zone −120°

Government Monarchy

Vassal of J.-Senlis

Nationality makeup

65% S. Apurhagat
15% Senlis
9% International
4% Esħ Pez

Figure 3.1: A brief overview of Klssrdmk.

Klssrdmk is a former territory of the nearby J.-Senlis, a massive country

that lies to the east and is the region’s major cultural influence. Because of
this, and the fact that Klssrdmk borders the J.-Senlis, Drsk too is heavily
influenced by Rattssaw, to the point where the former gives way to the
latter. Out of its seven provinces, only two of them are monolingual Drsk;
one of them is monolingual Rattssaw and the rest are bilingual. (23 | 1)

29
30 CHAPTER 3. THE COUNTRY

Though it is heavily influenced by things beyond its borders, it is not

entirely without its own culture. The by-now-cliché “patchwork of identities”
is, though trite, a perfectly adequate description of what the nation is now.
To describe how Drsk works, we need to describe roughly how the country
that speaks it works. As the language is not spoken very far outside of
Klssrdmk, we do not need to discuss much outside of the area. This is mainly
because the language and the country are very closely related to each other,
and therefore the two parts of the Heart of the Culture are almost completely
equal to each other. This has some very interesting consequences, which will
discuss later.
Because of that, the general collective memories of such citizens form a
rich base of experiences that are drawn from by the language. Therefore,
after a brief discussion of the country and some common national trappings
in this chapter. Parts of Drsk that are more tied to the culture than the
country itself that doesn’t necessarily tie itself to the country in chapter 4.
As we go through, we will also start introducing Drsk words as required into
the mix, and give them appropriate glosses.

3.1 The Flag

Although the flag of Klzdmk might be a little bit unusual at first glance,
this is not out of the ordinary as the general “graphical group identification
device” – that is, a thing that you can see that represents some individuals
put together – is in fact an ovecs, the study of which is called “octovexillol-
ogy”. These are octagonal pieces of wood or metal, hence the name, which
are placed on top of poles. A full description of how ovexes work is beyond
the scope of this grammar, but more information may be sought in appendix
B.
What is relevant to the language however is that it has a flag, which is
not exactly the same as the flag of the country. The important distinction
here is that the visible surface of an ovecs consists of two parts: the octag-
onal face that is easily recognisable, called a banner, and a rectangular face
underneath it, called a legend. The latter is an integral part of the flag and
cannot be omitted without some extenuating circumstance such as lack of
space. It is this legend that allows one to make distinctions like that between
country and language. (34 | 23)
In Figure 3.1, the flag of Klzdmk is the banner of Klzdmk along with
the legend which contains two letters in the Rattssaw script: K and Z. In
contrast, the flag of Drsk, the language, is the banner of Klzdmk along with
a legend that contains the letters D and R, also in the Rattssaw script (figure
3.2). That difference illustrates the usage of the legend in modern times,
which originally started as a method disambiguation between two nations
or similar whose banners are too similar in a diplomatic situation.
3.2. A QUICK HISTORY AFTER INDEPENDENCE 31

This usage is only really possible when the

language and the nation is very closely related
to each other, which is true for the case of Drsk
and Klzdmk but not so for other languages.
What these languages do vary from place to
place, but it is beyond the scope of this gram-
mar.

3.2 A quick history after inde-

pendence
3.2.1 Immediately after indepen-
dence
After the Restitution Klzdmk received a large Figure 3.2: The flag of
amount of natural resources which was there all Drsk.
along but is usually diverted away from it due
to the centralised nature of an empire. Though
by independence most of the ability for the land
to produce any more is mostly exhausted the
remaining stockpile is generally enough to keep
it running for at least three times as long as it actually needed to, though
at this point we are getting ahead of ourselves.
The remaining locations still has lingering tensions, especially consid-
ering that the capital of the fledgling country was placed deep into the
Rattssaw-speaking area. This is a little crass as it contains the majority of
loyalists that aren’t entirely happy with the partition (as Figure 3.3 shows,
the Trokier region doesn’t even have a majority of Drsk speakers, but the
Restitution gave it to Klzdmk because that was how it was divided during
the vassalship era) but a compromise was made in that a brief area around
the capital is given official bilingual status.
After a brief cultural outrage moment most of the tensions died down,
and the country as a whole started to connect with each other, with the vast
majority of the country’s infrastructure being planned out and built at or
around this time.

3.2.2 Division into language communities

The division of provinces and the rise of the individual language regions (see
section 3.3.2) is interrelated, but which way the arrow of causality goes is
more than a little bit complex. Nevertheless, the provinces are very strictly
related to the language regions, as the extreme similarity of the political
map in Figure 3.3 and the region map in 3.4 show. Indeed, one can see the
32 CHAPTER 3. THE COUNTRY

latter as a layer atop the former, because the only difference between the
two is the layer of links.
The actual relation is mostly that of mutual enforcement. The provinces
were both a reflection of historical nations and modern demographics, which
provides the initial seed for the language communities to form. However,
linguistically speaking the language regions are mostly artificial, as the entire
nation assumes a weak dialect continuum that, as with all other dialect
continua, resists division.
Nevertheless, the popular conscious of the moment does not really inter-
nalise the idea of dialect continua and education continues to cut the dialects
of the language
(An aside: the dialects in Nffek, especially the ones near the city, are
genuinely different from the dialects in a radical way and it does make sense
to count it out of the continuum for the most part, but of course contact be-
tween it and Ntpeasdr still makes for a blurred line. The province boundary
is still a good boundary line however.)

3.2.3 The Mirror

A half a millennium after independence, a species-wide event occurred es-
sentially overnight that caused a paradigm shift in every civilisation. This
event is frequently known as The Mirror, because it what happened was,
essentially, that the entire world gained an intense introspection skill.
Such a momentous and literal change of mind of every consciousness in
the species is as of yet a complete mystery. Not only is there no forewarning,
there is also nothing that happened in this magnitude since, and they have
been waiting for a long while now. That having been said, the magnitude
of this change is so large that there isn’t much that can conceivably top or
even equal it.
The most important item that occurred with this sudden change of mind
is the rise of the Heart of the Things system which was explained earlier in
section 1.5. The system has always been present in some way or another in
history, but The Mirror not only makes it more visible, but also in limited
situations influence the system and adjust its parameters.
Also at or about this point, the final stages of religion come into play,
as it slowly turns into modern fiction as we know it today. The details of
this transformation is noted in Section 29.1, but one of the key items that
caused this effect to happen is indeed The Mirror, as it allowed the mind to
become flexible enough to be able to better grasp the many different ways
something can deviate from the truth – and in some cases, how truth itself
can deviate from itself.
Linguistically the primary effect of The Mirror is to speed up the devel-
opment of language to a stop. Essentially, with everyone having the ability
and know-how to change his own language on the fly, or even in response to
3.3. PROVINCES 33

scientific findings, linguistics as a field is greatly threatened, and the ideas of

particular language celebrities become unnervingly important. Eventually
the languages of the world manage to stabilise in the long run – while many
individual innovations are created, they all roughly cancel each other out so
that over a period of thousands of years the language only evolves a small
bit compared to what it was before. Though not quantifiable the slowdown
is present, pervasive and persistent.
Finally the technological development of the world in general also be-
comes more cautious and therefore slow, with much of the resulting tech-
nologies being correspondingly less impacting to things like the environment
or moral commitments.

3.2.4 Drsk and Klzdmk in the Modern Day

At the point when the language described in this document is spoken,
Klzdmk has become more of a sleepy outpost country, which it has always
been since its days as a vassal country. This relative stability in its period
also contributes to the language being stabilised in the long run as well.
The effects of The Mirror makes this essentially the “final version” of the
language, though it continues to evolve anyway in the long history of the
world, just very, very slowly. (32 | 50) Nevertheless, decade-to-decade it is
correct enough to say that the language described here is the definitive Drsk.

3.3 Provinces
As previously mentioned, Drsk the language is almost entirely localised
within Klzdmk, and this has as consequences an interesting topology in
the language.
The entire country is divided into seven provinces, each named after
their capital city. Because of factors previously mentioned, the capital of
the country is not in the centre, but off to the east of the country, deep in
the Rattssaw speaking part.

3.3.1 The provinces, in detail

Each of the seven provinces in the country have a different makeup of speak-
ers in the countries. The situation is briefly explained in Table 3.1. Note
that due to other languages, the three numbers do not add up to the fourth.
Trokier (Čkʲr) is the capital province and is the only region that doesn’t
have any official recognition of Drsk with the exception of the actual capital.
It is the second most-populous province in the list, and is overall the net
cultural exporter of the country, being usually a handful of decades ahead
in legislation compared to the rest. It remains very friendly to the J.-Senlis,
and continues to have a special relationship with it.
34 CHAPTER 3. THE COUNTRY

Figure 3.3: A political map of Klzdmk, with the language’s rough sphere of
influence marked.
3.3. PROVINCES 35

Table 3.1: Summary of the language situation of Klzdmk, by

province.

Name Speakersa Total

Drsk Rattssaw Drb Rsc Both populationa
čkʲr Trokier 4600 6850 4650 15150
ntpsd Ntpeasdr 2600 950 2000 5700
nᵘfk Nffek 17500 550 6050 24400
ðvᵉþtł Eamttel 6500 650 750 8400
dmᵘz Woróli 2700 1100 500 4400
bzmᵘlˡ Bssreoml 3850 1250 700 5800
smᵘsˢv Iefiól 4300 3350 350 11050
Total 42050 14700 15000 74900
a All numbers are in thousands, rounded to the nearest
50 000.
b Drsk
c Rattssaw

The capital can also be called trgr in Drsk, and this is in fact its of-
ficial, formal and older name. The new name, čkʲr, is a back-formation from
the Rattssaw name Trokier. In conventional speech the two are considered
interchangeable, but strictly speaking trgr is the city, and čkʲr is the country
or province. The difference is subtle enough that some mapmakers make the
mistake. (24 | 70)
Ntpeasdr (Ntpsd) is the a vast province, the largest, but is largely rural
and therefore is very loosely populated. However, for historical reasons a
large amount of the province is bilingual.
Nffek (Nᵘfk) is by far the stronghold of the Drsk language, and is largely
considered to be the “purest” surviving Drsk variant. Indeed, by many counts
it is the most conservative dialect, but of course being a last scion of sorts
imparts a lot of changes that a mere successor of a language would not have.
The province is divided into two by the free city of Eamttel, but the true
division is in the more innovative shoreline region in the south and the more
rural and conservative region in the main body of the province.
The city itself, nᵘfk, is a mercantile city on several islands, divided into
several language sub-communities based on its wards, (33 | 79) which is not
commonly seen elsewhere.
The free city of Eamttel (Ðvᵉþtł) is so named because it is the only city
that’s also an entire province. It is one of the more populous cities, with
a total population of about 8½ million. It grew itself as a port of call for
36 CHAPTER 3. THE COUNTRY

travellers from the ocean to the west and voyages between the southwestern
peninsula and the northern isles.
Woróli (Dmᵘz), the central province, is also intensely agrarian and con-
sidered the heartland of the province. It is largely considered the cultural
centre of the culture surrounding Drsk. This is also one of the more conserva-
tive areas of the language, with many older words still present in particular.
Nevertheless, due to its central location it still has a significant amount of
immigration involved and as such a lot of innovations from other provinces
make it less “pure” than that of northern Nffek.
Bssreoml (Bzmᵘlˡ) is another coastal province, and is in some cases
called the back garden of the country. Though most of it is temperate
marshland, it does have some of the greenest plains in the land, and a good
area of it is under a nature reserve, although the reserve is frequently cut
up by roads.
Iefiól (Smᵘsˢv) is one of the provinces where a large part of the country
is not majority Drsk speaking. But a good pack of this group is also not
Rattssaw speaking either, and instead speaks the language which the nation
of Earmo2 l2 supports. This gives it a unique mix of culture that is seen
nowhere else in the nation.

3.3.2 Language regions and provinces

North Nffek Ntpeasdr

nᵘfk-z ntpsd

Eamttel Woróli Trokier

ðkvᵉþtł ðmᵘz čkʲr

South Nffek Bssreoml Iefiól

nᵘfk-l nbzmᵘlˡ smᵘsˢv

Figure 3.4: The language regions of Klzdmk. Solid lines represent a solid
continuum of dialects between the two regions exist; dashed lines indicate a
more tenuous relationship; and the relationship between the two halves of
Nffek are indicated with a dotted line.
3.3. PROVINCES 37

The language’s physical presence is still very much geographically influ-

enced by the borders of Klzdmk. Each of the provinces can be understood as
one language region, six (or seven, if you count the two parts Nffek Province),
of which form a ring around a central province, Woróli, which is the true
centre of the language. (28 | 42)
The seven “peripheral language regions” of Drsk, then, are from Trokier
anticlockwise: Trokier, Ntpeasdr, North Nffek, Eamttel, South Nffek, Bss-
reoml and Iefiól. Each of these seven regions surround the “central language
region”, which as we can see in figure 3.4 is Woróli.
As with all things with language, the regions are not as well-defined and
there is a lot of mixing in the places where two regions touch. The analogy
is that of the colour wheel: Around each of the peripheral language regions,
the colours are the that of the rainbow, cycling and blending with each
other. Coming closer to the central language region, all the colours slowly
blend into white. There is however a small irregularity in the analogy in that
North Nffek is somewhat more abberant relative to the rest of the nation
and so they would be assigned a colour similar to bright pink – outside of
the normal colour wheel, and something a little bit of its own.
Additionally, cities tend to have their own layer atop the general geo-
graphical regions, and their classification situation is slightly complicated.
In some ways, they each live in their own language region, informed by their
surrounding language region, so for instance Drsk in the mᵘrk (Umrok) Lan-
guage Region is similar to the that in the Ntpeasdr Language Region that
surrounds it. On the other hand, each individual Urban Language Region
also share a large amount of similarities with each other, so they also in some
way they also belong a Language Region together, which is usually dubbed
the Inter-Urban Language Region.5 While originally this language region
is not very well-established, it has become more so ever since inter-urban
communication was increased during the invention of mass communication.
In the colour analogy, they would represent bright or dark spots in the lan-
guage community, with the common element of luminosity corresponding to
the inter-urban similarities.
The colour analogy is not an idle comparison; it is officially endorsed as
part of the language, and although the exact colours are not used in Drsk
something basically analogous to this have been provided. It should be noted
that the analogy was not developed as part of a native Drsk construction –
section 16.13 should give a very good reason as to why – but an external effect
that was imported and internalised into the native Drsk-speaking community
and given a life of its own.

5
Where this does not introduce ambiguity, the Inter-Urban Language Region can also be
referred to as the Urban Language Region.
38 CHAPTER 3. THE COUNTRY

3.3.3 Four Language Centres

The language regions provide a perfect set of objects to describe the four
language centres. They represent a handy collection and categorisation of
dialects that allow for certain labels to be placed on them.
In the context of Drsk, the centres are assigned to the language regions
rather than individual dialects due to them being in a dialect continuum.
They are:

Normal The dialect spoken in the literal, geographic centre of the area in
which the language is spoken. Also known as a “central” dialect.

Traditional The dialect which has changed the least over a period of time
relevant to the context. This is also called a “conservative”.

Standard The dialect that carries the most social power, and promulgated
over other dialects. This is also the prestige or authoritative dialect.

Reference The dialect that is studied the most often, and have been doc-
umented most heavily.

Or, in summary:

Centre In reference to Most… Least… Holder

Normal Geography Central Periphery Woróli
Traditional Diachronics Conservative Innovative Nffek
Standard Politics Prestigious – Trokier
Reference Synchronics Studied – Eamttel

In most other languages, some or all of these four will be one and the
same, but the unique circumstances of Drsk makes them not so. The geo-
graphical, highly objective nature of the central dialect makes it a bit of an
odd one out, so it’s not too surprising that it’s different from the rest, but
during the country’s vassalship period, the language’s centres diverged as
different cities (and subsequently language regions) receive different types of
attention – the city of Nffek has had the least amount of interference with
the meddling hands of the J.-Senlis, and it just so happens to be not very
innovative historically which robs it of its change; most traders and more
linguistically-inclined societies tend to cluster around the port city of Eamt-
tel, so it gets studied more often. When the country got itself back together
again, it’s politically expedient that the dialect in the capital gets the most
recognition and value, so it’s now the standard dialect. And finally, the
language regions circle itself around Woróli, which further cements its role
as the central dialect even though it hasn’t really got anything going for it
otherwise.
3.4. CURRENCY 39

This grammar generally concerns itself with a vague generalisation of

all of them, which in the end looks like a mix between the traditional and
standard dialects.

3.4 Currency
The Drsk national currency is not the single legal currency in the country.
As described in (36 | 3), there is a plethora of local currencies ranging from
city- to province-wide, and all of those are legitimate alternatives to the
national currency. The only privilege that the national currency has over
the local currencies is that it cannot be refused to settle a debt. This has
linguistic implications, which we will explore in section 17.8.2.
Regardless of local or national currency however, as with many places
influenced by the J.-Senlis, the non-(duo-)decimal Xt-M system (24 | 77)
is the model for all currencies in the country. This implies the usual ar-
rangement of seven layers of currencies, with different combinations of the
quantity, pipes, and units.
Below is the system of units:

Name Symbol Equivalent

ry Mǀ1
dʲgs M1 Mǀ4
ðʲħy 1M M4
ŋrw Xt 1 432M
ħt 1 Xt Xt 432
tᵗħt TX 1 1728 Xt

The ry is the smallest unit of currency and is primarily used as a unit of

account rather than a minted coin. It is built to be small enough that it can
subdivide the ħt with a round power of twelve, so that it can pretend to be a
(duo-)decimal currency after all. In this case, 1 Xt = M|126 , which one might
think to be a little excessive for a division – six digits! – but recall that a
strong social stratification makes it so that each individual level underneath
is reasonably important, just not all at once to any individual member of
society.
Four ry make a dʲgs, which is the smallest coin minted. There are only
two coins that are denominated in this unit: the M1 and the M2.
The most popular unit of currency that most of the lower classes use is
called the ðʲħy, for each one is composed of four dʲgs. Amounts in this class
typically refer to all kinds of everyday objects that one purchases on a daily
basis. As with the case of most of the regions around this location, notes
are only used with a certain denomination and up, and ðʲħy are not in this
division. Hence, the coins range uncharacteristically wide in this part of the
40 CHAPTER 3. THE COUNTRY

world, ranging from 1, 3, 6, 12, 36, 72, 144 and 288M. Some banks issue
notes for the 144M and the 288M, referring to this new unit as the mtr
(with symbol 1|M, so for instance 150M2 becomes 1|6M2) so as to keep the
rule that notes and coins are named separately. All these coins have specific
names amongst the populace too, and of course they all have some regional
selection.
The definition of the ðʲħy is different from the definition used in its Ratts-
saw cognate the awddewa, which has 40 of the next level down instead of
just 4. The extra factor of 10 in the Rattssaw version is to assist in divisions
of five (which occur somewhat commonly), but when Klzdmk adopted the
system they preferred the regularity of having an almost duodecimal system
over 5-divisibility.
The ŋrl is the designated “big” unit amongst the elite and the rich, and
it takes 432M to make one ŋrl, which as the symbol Xt 1. This unit of
currency is still within reach of the lower classes, but only the lower values,
and most of the amounts here are used primarily in the context of long-term
earnings and expenditures. The rich, of course, can spend this like the poor
do the ðʲħy. Notes here are as in the ðʲħy coins, but they omit the 6 in the
preferred series of 1, 3, 6, 12.
The largest unit commonly seen in the common world, rich or poor, is the
ħt, which is the unit that everyone aspires to, whether or not he can reach
it or not. Once again, Xt 432 make 1 Xt, but this time there are no notes as
there is not usually any demand for prices in this range to be paid in cash.
Items like houses, mansions and (later) vehicles are quoted in this unit of
currency. Conversion factors are also linked to this unit, though within the
sphere of influence of the J.-Senlis, conversion factors with the ðʲħy is used
instead, which is significantly more in reach with the general population but
still retains universal familiarity within the sphere of influence.
The final unit, the tᵗħt, is a unit of account, much like the smallest unit,
but its utility is much less utilitarian. The purpose of the tᵗħt, which converts
to 1728 ħt, is to bring down the lofty prices that governmental policies and
similar long-term commitments to numbers more understandable by the
typical brain which cannot comprehend these amounts. Where chopping
of three zeroes does not suffice for “fun-sized” numbers, there’s also the
much lesser-used tᵗtᵗħt (symbol 1 TX), which chops off 9 zeroes from the
ħt amount, and the even-lesser-used tᵘtᵗtᵗħt (symbol 1|TX), which chops
off 12 zeroes. Unlike the tᵗħt, these two have no official recognition and are
only used informally in news reports and similar.
These symbols can be strung together using relatively simple rules. No-
tice how the order of the units matter: 1M 6= M1. The rules for stringing
together a set of units is as such:
1. Set apart all the quantities with the same alphanumeric characters.
2. For each quantity with the same alphanumeric characters:
3.5. TRANSPORT 41

(a) Put the units where the number is on the left on the left, and
similarly for the right, e.g. “2M, M3”.
(b) If the symbol has a pipe in it, then it goes at the very end of the
list, with the above rule, e.g. “M3, M|1”.
(c) Merge the pipes together with the unpiped units. This is eas-
ier done than said; in the above example, M3 and M|1 combine
together to form M3|1.
(d) Finally, merge the units-on-right quantities with the units-on-
left quantities in that order, discarding the duplicated unit. For
instance, “2M, M3|1” becomes “2M3|1”.
3. Finally combine each of the units with a space separating them, in this
order: TX, Xt, M.

So for instance, a sum of 3 Xt, Xt 100, 50M, M2 and M|1 becomes

3 Xt 100 50M2|1.
In practise not all units would appear in the same quantity and some of
them would be dropped in different contexts.

3.5 Transport
Transit and journeys in Klzdmk is largely patterned after the historical
divisions of the way that Grade (1) divides travel. In general however, this
division is mostly sensible and international, with only a small amount of
adjustment in the way that the divisions are made.

3.5.1 Unassisted travel

“Unassisted travel” means travel without help of motor or machine, with all
motion being done via muscle power. Items such as backpacks or drawn
wagons are typically included, and in later times animals are also included
as motorised transport grew on.
Such methods are typically given a special position in both the culture
and the engineering in the nation, as it is considered one of the more funda-
mental parts of freedom. Generally speaking, save for some critical military
infrastructure, there is a general consensus that those on foot or with an
animal should be allowed to roam free within the country wherever a roof
is not present. Even where an area is fenced off, some path must be pro-
vided for pedestrians to pass through, though this can manifest as a simply
maintained path at the edge of a property.
Overall, a large number of rural pedestrians still roam the back roads,
justifying the dense foot-track network that covers the entire network. They
are numerous and frequent, and every track and road that is dedicated to any
other mode of traffic has to also provide it somewhere nearby to pedestrians.
42 CHAPTER 3. THE COUNTRY

3.5.2 Road

Figure 3.5: A road map of Klzdmk.

Motorised traffic is fairly new, but due to the early adoption of them
the road network in Klzdmk is organised and greatly hierarchical, which
is surprising given the general non-orderable theme that the rest of the
countries run on. A map of the highest speed motorways is given in Figure
3.5.
Like many places in the world, Klzdmk drives on the left. Furthermore,
like many places in the sphere of influence of the J.-Senlis, roads are given
numbers, rather than destinations. And like most everywhere except a con-
tinent on the other side of the world, one must drive between the lines.
Major highways in the nation are all hub-based, with the capitals of each
province being the hub of each item. However, the hub highways all connect
up in such a way that they connect in a more or less imperfect grid, with one
3.5. TRANSPORT 43

capital’s radial connecting up the other. An additional set of roads circle

the country’s borders, and there are additional high-speed roads to relieve
locations that would otherwise lack them.
Lower-class roads are chaotic as they always are, and most of the major
roads that were once lower class have been upgraded to higher-class ones.
As one might expect they exist to serve any number of smaller communities
and other places that don’t require the service of the higher-speed roads.

3.5.3 Rail
Railways in Klzdmk are similar to the road infrastructure, but they are
not required to have a pedestrian right of way due to safety grounds (and
that sitting on the train counts as walking, in an obtuse legal sense). Like
roads, trains drive on the left, but unlike roads and like most other railway
traditions, trains follow lines (which are painted between the two tracks)
rather than having lane separators.
The country is divided into two gauges, one for the native-majority
1777 mm gauge, and the Rattssaw-derived Rattssaw gauge, which is 1851.5 mm.
Although this seems somewhat inconsequential, this is exploited in Drsk
statements, which one might see in (translated) expressions such as

(1) ??
Making a dual gauge
lit. forced redundancy

In general, rail is used in small towns and villages and long-range intercity
transport, but its role is greatly diminished inside large cities and inside
suburban areas and is entirely nonexistent in a city centre. At least, that’s
the densely packed organic cities in the west – rail has a higher penetration
in the planned and more spread out cities further east where there is heavy
Rattssaw influence.
Inside a city, city-wide metro is a classic staple in all of the major
Rattssaw-speaking areas, and in the largest eastern cities such as Nffek and
Eamttel. Again this seemingly irrelevant it does show up in the language,
though in this case it is less focused on figure of speech and more on associ-
ations of various concepts with the provinces.

3.5.4 Air
Travel by aeroplane is at its infancy, but the nation has already developed a
large number of airports, which for the most part handles the longest-range
domestic traffic and international traffic, as short-scale traffic is well-handled
by rail and road.
The most notable airports are in each of the provincial capitals, as well
as in the inland city of ktᵘc. Trokier, Nffek and Eamttel all have multiple
44 CHAPTER 3. THE COUNTRY

airports, which divide the air traffic amongst domestic and international
airports.
Interestingly, in the majority of airports, flights are divided into “domes-
tic” (tʲ.mᵘl.dᵘk.tʲ), “international” (tʲ.mᵘl.mᵘd.tʲ) and “exotic” (tʲ.mᵘ[Link].tʲ)
sections. The domestic section is as expected, but what we would normally
consider “international” is divided into two: whether or not the destination
belongs to one of a specified list of countries that are considered “cultur-
ally related” to Klzdmk. This largely consists of countries that are in the
sphere of the J.-Senlis, which is largely due to historical reasons as outlined
in previous sections.

3.5.5 Ferry
Ferries crisscross the waterways and seas around the nation. They handle
short-range connections, as well as movements around the coast and along
rivers.
The country has one major navigable river and two minor ones, and
there is a modest collection of canals to bridge such items. Boat traffic is
largely a historical deal and does not have much effect on the modern world,
or even in the colonial and vassalship era.

3.6 Rights and liberties

As with much of the world, the ideas of civil liberties and what counts as
being free or not free in the Klzdmk is largely a mixed bag compared to a
typical, “Western-civilisation”-based ideal. This is as previously explained
partially due to the wildly differing psychologies between humans and kilis,
of which the latter are much less likely to want to assert one’s individuality.
This has resulted in a corresponding lack of civil liberties; however one
should also note that due to the general consensus that knowledge is free to
find and exploit, and the fact that those few that do seek individuality and
striking their own path do so with frightening efficiency and effectiveness,
that there are also other parts of civil society that seem lawless in comparison
to the ideal.
One of those places where civil liberties are in (perhaps over-)abundance
is the idea of speech. Speech is not only information and knowledge, and
therefore already a terminal value to be allowed to be as free as possible,
but also that it is one of those ways that society in general has decided is the
way to keep new ideas coming in. Additionally, this is because that to some
extent a kilis’s speech is considered to be a part of its self as well; while any
speech is able to be disowned by its owner – and it frequently is, to the point
where such events go by without much fanfare or even anyone noticing – if it
hasn’t it’s considered more or less a part of a particular framework roughly
understood as “the intrinsically valuable contribution to the ‘marketplace of
3.7. GOVERNMENT 45

King
House
• Armed forces general
• Overrides decisions
• Overrides decisions made by the King in
made by House in some other sectors, in partic-
sectors ular political issues

• Issues pardons • Runs day-to-day policies

• Can select one special in- • Runs courts

terest field to assume di-
• Immune to dismissals
rect control over, once in
from the King
a while
Figure 3.6: The powers of the King and the House.

ideas’ that constitute an inalienable part of society”, and therefore is in some

very real and social sense protected. This applies even in the case where the
speech is repugnant or even anti-social.
This results in a language development where socio-linguistic factors are
greatly diminished in some regions, and others greatly exaggerated. We’ve
previously mentioned the idea of linguistic celebrities in section 3.2.3, but
there are other effects that specifically come from the political situation
which flows from above. For example, there is a relative lack of restraint
when it comes to forbidden words – any such are quick to change and mostly
local in nature, and are frequently moved out of the way, with virtually no
words that are recognised as vulgar throughout all time and space.

3.7 Government
The government of Klzdmk is headed by a king, which in accordance to
the general culture generally abdicates after 48 years from coronation. The
current dynasty has been in power for approximately twenty to thirty gen-
erations and does wield appreciable amounts of actual power, on top of
actually using it.
Below royal family is a House, which contains the majority of the no-
blemen that wield and exercise the remainder of the political power in the
country. The House is comparable to a legislature in a democracy, or a coun-
cil of dukes, guild masters and other landlords in a feudal society. To some
extent the House is some combination of both, as well as a small amount
of an “assembly of ‘the Workers” ’ which might be seen in a more Earthling
contemporaneous society – that would be equivalent to the guild masters,
which are selected to represent industry and industry member interests.
46 CHAPTER 3. THE COUNTRY

Klzdmk is not, in most ordinary senses, a democracy. Historically, the

theory of power being derived from the general population has never took
much hold within civilisation and for reasons that are not particularly clear
citizens – ordinary subjects – tend to vote to relinquish this power when
they are granted it. (20 | 39) The institution of voting is therefore not well
established and there is no pretence at all of the government taking orders
from the citizenry.
Nevertheless, voting is used in formal contexts to inform the government
of what the general population thinks. Unlike a true democracy however,
the government is by no means obligated to accept the results of the vote
and the results can and are frequently discarded in favour of whatever the
King or the House thinks (the importance of the animosity between the
King and the House and how in practice one does not outrank the other in
such a clear-cut way as it is described nominally in ensuring that the general
populace can have their interests represented is demonstrated here.)
This means that regardless of the fact that Klzdmk is not a democracy,
there remains a well-established infrastructure for voting and polls regardless
of the fact that it is not occupied by a robust political institution. Instead
the voting organisations are largely occupied by units with a strong interest
in statistics and “gauging the social environment” (i.e. analytics, before such
a thing was popular). Voting rates are not as high as they might be in a
fully democratic society with a similar social environment, but the rates are
kept high and the results kept relevant through the general population also
interested in examining themselves.
Linguistically speaking, this is largely reflected in semantics of particular
words such as tʲ.msdᵘ.tʲ “vote, poll, measurement, observation”. However,
in a deeper level, the surprising high amounts of democracy in a democracy-
lacking nation does show up in how language change occurs. Linguistic
celebrities, as described in section 3.2.3, also do a similar thing with polls
when deciding to champion one change over another.

3.8 Unorderable hierarchy

As with most of the region, there is a highly stratified class system in place,
with each class itself having a complicated system of outranking that does
not create an overall “highest of the class”.
Indeed, within each class, this system of outranking would create an
organisational chart where there can be no “rank” as the relation < (where
a < b means “b outranks a”) is not necessarily transitive – i.e. a king can
be outranked by a pawn. Such a setting subsequently creates a system of
equality within each class, because anyone can call on someone else to pull
rank and somehow manage to do so. Social power and capital is not strong
3.9. ECONOMY 47

enough to organise general power toward its own, though to be fair this is
not for lack of trying.
Nevertheless, there is an interest to keeping this going, and that is the
fact that this “unorderable hierarchy” is only for a limited contexts; there
is a traditional power hierarchy at the highest level where the King – the
King of Klzdmk – is at the top and there is no particular individual that
can outrank him.
Generally speaking, the exact government structure does not contribute
greatly to the overall form of the language, largely for reasons of the lan-
guage having some form of sovereignty of its own. Even in a more direct
effect of how idiolects vary across social strata, the confusing non-orderable
hierarchy in each class allow the language to mix and homogenise, resulting
in a largely geographical spread in dialectal variation. However, the class
system does allow for a limited amount of class-based dialectal variation,
however subdued.

3.9 Economy
At the present time, Klzdmk is a mixed economy, with a roughly even mix
of service and production-based economy. The production-based economy is
however mostly primary, with much of the country being based on farming
and agrarian. Industry is rare and not seen in most of the country, mostly
concentrated in the Rattssaw-speaking east.
The country itself is not a particularly high-performing country in an
economic sense – it’s mostly overshadowed by the J.-Senlis, as might be
expected from historical concerns.
Figure 3.7 displays the GDP distribution of the country as well as the
approximate geographical extent and primary economic output of each sec-
tor, (36 | 5) and table 3.2 displays some of the key economic indicators of
each sector. (A “sector”, in this case, is a statistical subdivision of a province
that has roughly the same economic output.)
Language-wise the economic information informs how certain dialects
and language regions set boundaries against each other; there is also the
usual stereotyping in terms of what one does and how much one earns to
how one speaks, to the point where to some extent there is also a “Poverty
Language Region” (which to be clear is always subordinate to the geograph-
ical language regions). It’s not always easy to determine which way the
arrow of causation goes when it comes to socio-economic status and how the
language regions form; to some extent they reïnforce each other.
Diachronically however economics-based language variation becomes a
force to be reckoned with; this is because of two competing factors: very
rich societies tend to generate prescriptivists that end up being active change
agents of the language; meanwhile impoverished societies tend to generate
48 CHAPTER 3. THE COUNTRY

Key to symbols
Consumer goods Entertainment Durables/intangibles Production
11 Nffek urban
Textile Music Communications Construction & Agriculture
Maintenance
Retail Visual arts Marine (??) Mining
Finance
Food service New games Education Housing
Healthcare
Vehicles Old games Conservation
12 Nffek suburban Tech Publication
Conservation

13 Nffek South 14 sˢfmᵘrⁱ 71 North coast 63 North coast

15 rⁱscᶜ 76 stdᵘt 73 Central 62 R-A Corridor

20 Eamttel 72 stdᵘt 61 Trokier

74 sdkⁱ Banks

16 Exclave 75 R-A Corridor 64 South

30 Urban 36 Rural 50 City 51 North

42 Urban coast 45 North inland

41 cᶜmr coast 44 Urban inland 46 Central inland 52 South

43 Rural coast 47 South inland

Figure 3.7: The economic output and resources of each sector, as well as the
sectors themselves.
3.9. ECONOMY 49

Table 3.2: Vital economic data for each sector. Monetary units are in Xt;
all values are per capita, per year, where applicable. WPR is the Workforce
Participation Rate.

Province Sector GDP Income Emp. Rt. W.P.R.

Trokier Trokier 4420 8640 96.4% 68.2%
North Coast 3890 7250 97.1% 70.0%
sdkⁱ Banks 3420 5350 98.2% 72.2%
South 2890 3790 94.0% 74.5%
Nteaspdr North Coast 3400 7580 96.2% 74.6%
stdᵘt 2860 4450 93.5% 72.0%
Other central 2200 4890 94.0% 66.9%
sdkⁱ Banks 4500 5020 95.1% 71.5%
R-A Corridor 2520 3520 88.4% 70.2%
North 1980 2980 86.1% 56.1%
Nffek Nffek Urban 6420 12220 98.6% 66.5%
Nffek Ext. Urban 6550 12640 98.8% 68.2%
Nffek F.C. South 4000 9990 94.4% 75.5%
sˢfmᵘrⁱ 3810 8500 95.9% 62.0%
rⁱscᶜ 5340 7150 93.1% 58.9%
South exclave 3250 4050 92.0% 64.6%
Eamttel – 6510 9150 95.1% 62.2%
Woróli Urban 3820 7250 86.1% 54.0%
Rural 2230 6000 91.2% 64.0%
Bssreoml cᶜmr banks 6940 8630 89.4% 56.4%
Urban coastal 6240 7520 90.4% 59.4%
Other coastal 5600 6440 92.1% 62.2%
Urban inland 4400 3500 88.6% 52.2%
North inland 2650 4990 91.1% 72.2%
Central inland 3550 4460 93.4% 73.9%
South inland 3840 4620 92.2% 70.0%
Iefiól Urban 6250 5580 94.2% 59.6%
Rural north 5020 5890 97.4% 72.2%
Rural south 5660 7250 99.0% 74.1%
National 5090 8350 95.9% 66.9%
50 CHAPTER 3. THE COUNTRY

language change in a more “organic” (read: natural) way which may or may
not overwhelm prescriptivists through sheer numbers.
Furthermore, individuals in lower socio-economic statuses have a partic-
ular advantage to them unavailable to those higher – vocabulary control.
That is to say, poorer parts of the country, by virtue of having by far a more
varied experience of the country, tend to have more control over what words
get assigned to what experiences. The rich may attempt to use status to
cover over those terms, but these attempts rarely succeed in this culture.
This rather surprising result still has no particular good explanations at-
tached to it, but some early hypotheses about how the poor have some kind
of quality typically known as “authenticity” have been raised, though testing
such elusive qualities have proven difficult.

3.10 Minority and neighbouring languages

It is important to describe how Drsk and Klzdmk exist in the context of
other languages and countries respectively. Chapter 5 will describe this in
more detail, while in this chapter we will be more interested in describing
how the country deals with the languages.
In this section we will introduce these languages in brief, which is im-
portant for the full understanding of how the language evolves over time.
Figure 3.8 shows the geographical extent of these languages with relation
to Klzdmk.

3.10.1 Recognised language categories

The government recognises some categories that certain languages fall into.
This usually confers to the language some kind of privilege or other spe-
cial treatment compared to those that don’t fall into any category. These
categories are:

Graded languages “Graded” here is a euphemism of “of the invaders that

have eventually managed to conquer us in some way and have their way
of life imposed on us”, referring specifically to the numbered grades of
invaders that conquered the area now occupied by the country. These
are the languages that are imposed onto Klzdmk (or its predecessor
states) from without and it now finds difficult, inconvenient or unrea-
sonable to remove.

Indigenous languages It’s a little strange for a Drsk-speaking polity to

designate other languages as “indigenous”, because to some extent Drsk
is “indigenous” – it has never spread very far beyond its borders after
it has been formed, and of course it has been suppressed from the
graded languages. Nevertheless, the government did once have a fairly
3.10. MINORITY AND NEIGHBOURING LANGUAGES 51

Figure 3.8: A map of Apurhagat near Klzdmk, with dominant languages of

that region marked.
52 CHAPTER 3. THE COUNTRY

aggressive approach to suppressing other languages within its borders6

even as it is being suppressed by Rattssaw. As a result, in more modern
times, what once were peer languages – both in the sense that they
are close to each other geographically and that they are close to each
other diachronically – these are labelled as “indigenous languages” and
receive special protection. The only wriggle here is that for some
reason Drsk itself also includes itself as an indigenous language, which
makes sense in a historical sense but not so much in a political sense.

Nonspeaking languages Languages for those that cannot hear are called
“nonspeaking languages”, because the only concrete thing that they all
have in common is that they do not use sound in encoding abstract
ideas to concrete objects. However, what they do use is not always the
same – while a majority of them use some sort of hand signs far more
also adds to it a number of props or common tools which are readily
available.

Historical and heritage languages Finally, languages that were once spo-
ken in Klzdmk but are now either extinct or succeeded7 are also recog-
nised, though their treatment differs from other recognised languages.
Generally speaking, old versions of Drsk or Rattssaw are not included.

Languages that are recognised typically enjoy certain privileges that

other languages do not. For example, it may be accepted as a language
for use for government communication (even though the government does
not actually speak the language actively) and receive automatic approval
to post on things like road signs or posters if requested (only Rattssaw and
Drsk are required to be on such writing).
Table 3.3 is a listing of languages that have such recognition.

3.10.2 Treatment of minority and neighbouring languages

Drsk as a language, and Klzdmk as a country is something that is termed a
“transmission country” – that is to say, while it has been placed under the
auspices of a colonising power at some point in the past, it has also been
6
There is an additional wriggle to this, and it is that the language itself, via the Will
(see section 1.6) can also in some sense bully other languages into submission. This is not
a mechanism that is in effect in Klzdmk, but on the other side of the world, in the extreme
shores of Banřello, this is very much in place. (30 | 12)
7
In the traditional reckoning, a language can have three statuses: alive, extinct and
succeeded. (24 | 39) While most would be familiar with the idea of an alive and an extinct
language (though an alive language is usually called an “extant” language), the idea of a
“succeeded” language is that it is an extinct language that is only extinct because it has
evolved into one or more successor languages through ordinary linguistic evolution rules.
Such a language is “succeeded” because it has successors, but it could also be claimed to have
“succeeded” in its purpose and has spawned daughter languages to continue its duty.
3.10. MINORITY AND NEIGHBOURING LANGUAGES 53

Table 3.3: Recognised languages of Klzdmk. Languages marked * have not

had their properties determined yet and so their names are indicated in
E.-Pasaru.

Graded Indigenous Nonspeaking Historical

1 E.-Pasaru E.-Fdtrpbë* 6G-J E.-Donas*
2 Rattssaw E.-Knfrd* Trokier Sign* E.-Kesš*
3 E.-Ðvtl* Nffek & Eamttel Sign*
4 E.-Deskis*
5 E.-Tliť*
6 E.-Tmoar*
7 &c.

the colonising power at other times, or even contemporaneous to when it is

being colonised. In other words, from the historical record it is clear that it
has suppressed a lot of the languages that it can while itself being suppressed
by the graded language of the time (Rattssaw).
Keep in mind that while we say “suppress”, we don’t really mean a large-
scale prevention of the suppressed language from being spoken – though
there is some level of that kind of suppression in action, it’s never as widespread
primarily because of the common value amongst all kilis that languages are
always something that are worth preserving.
The suppression, therefore, involves preserving the language in such a
strict way that speakers are encouraged to move off it by allowing more
freedom for the preferred language to change and adapt. This kind of sup-
pression is the most common way a civilisation can stop another one from
speaking a particular language.
In any case, toward more modern days, languages are treated in higher
regard than ever before, and this is reflected in general governance – each
of the recognised languages listed in table 3.3 of course has a large number
of protections listed on to them, but there are many languages that don’t
make that list but are protected nonetheless in other ways. There’s the usual
preservation and documentation efforts, as well as active encouragement
toward learning these languages and passive acceptance of these languages
in government documents, for instance.
For languages that are not recognised, these accommodations are still
provided though not on an obligatory basis. There aren’t that many lan-
guages that are relevant here, though there is a number of languages that do
get significant numbers of speakers in Klzdmk yet are not recognised. These
are usually immigrant groups that don’t speak one of the graded languages,
and for those there is usually a support network for the languages and the
people that speak them.
54 CHAPTER 3. THE COUNTRY

3.11 The political status of Drsk

The country, the culture and the language are, unlike most more widely-
spoken languages and even most neighbouring languages, are mostly corre-
lated and exclusive to each other, in a manner very much familiar to con-
temporary Earthling creatures as the idea of a nation-state. Keep in mind
that although Klzdmk does have this behaviour, this is not by any means
a universal feature in the whole world, least of all in this area where right
next door there is an empire (in the national sense) that spans multiple such
languages and cultures.
As such Drsk enjoys a particular status in the country unparalleled by
any other language there and unparalleled in any other countries. For exam-
ple, although there are many recognised languages, if some law is written in
Drsk it is typically considered to be the standard and all other languages are
therefore translations of it. This does not amount to “supremacy” – that is
to say, the concept that the Drsk copy of a law prevails over all other trans-
lations of that law – with recognised languages being referenced whenever
there is a found ambiguity in Drsk, and vice-versa.
This kind of officialdom and power can be contrasted with how the lan-
guage and its speakers have been treated over the centuries, where at some
point the language has been as elevated as being a regional language, and at
times it was entirely suppressed, which we detail in part in chapters 2 and
6.
Chapter 4

The culture

Most of the remaining parts of life in Klzdmk relevant to the language is

described here. The “present time”, as previously mentioned, is the time in
which the language as it is described in this document is spoken. At all
times, we will relate this back to how the language developed.

4.1 Daily life

Throughout the history of the country, there had always been a diverse array
of different ways how one divides up the day into work, play and sleep. This
is exploited by both the economy and the government, for example resulting
in a population where there’s always an appreciable proportion which is
awake, for instance.
Two such schedules are old enough to be counted as indigenous, the
northern, mainstream, schedule, and the southern, periphery, schedule. These
two are described below, along with some of the more modern schedules and
how all of them interact with each other.
As explained in section 2.1.4, for some of these systems it is not the day
but a combination of three days that is divided to form a typical “daily”
routine. Over these three days the cycle will contain the usual combination
of work, leisure, social, family and unitary maintenance. The way they are
spread over three days however make it a little bit trickier than usual.
First, we should note the length of the triad: it is three days long, which
converts to 3240° (native Pasaru units) or 132.8 ks (SI units). This does not
actually matter too much in terms of the daily routine, because at the time
such a routine is established there is no such fine division in the time of day:
there’s only six rough time periods for each rotation of the planet, four in
the daytime (morning, noon, afternoon, evening) and two at night (night,
predawn). However, we can still divide the triad as such, using these rough
divisions, as we do in figure 4.1.

55
56 CHAPTER 4. THE CULTURE

1
1

1
oo
ng

g
1

n
rn
ni

ni
n

e
or

oo

ve
ft
M

Night 1 Predawn 1
N

0° 216° 324° 432° 540° 810°

2
2

2
o
ng

ng
no
2
ni

ni
n

er
or

oo

ve
ft
M

Night 2 Predawn 2
N

1080° 1296° 1404° 1512° 1620° 1890°

3
3

on

3
g

g
no
in

in
n

er

n
or

oo

ve
ft
M

Night 3 Predawn 3
N

2160° 2376° 2484° 2592° 2700° 2970°

Figure 4.1: The divisions of the triad, which is used in table 4.1.
4.1. DAILY LIFE 57

4.1.1 Northern schedule

Though the actual schedule has long since been abandoned with the excep-
tion of a few holdouts in the most traditional places, much of the language
still persists in the Grade (1) calendar system (for which see section 2.1.4),
which of course also includes the timing. The divisions of the triads are
most important for this old system, so this is what this section will describe
in more detail.
With the division of the triad in figure 4.1 in mind, the regular schedule
looks roughly as table 4.1. Notice how roughly each time slot can accom-
modate one or two activities.

Table 4.1: The rough daily schedule of the traditional life.

Day Time Action Drsk

1 Sunrise Awake kʲ.pls.tʲ
Morning Breakfast (1) [Link]ˢr.tʲ
Morning Work þ.ħlr.tʲ
Afternoon Rest rⁱ.sˢk.tʲ
Sunset Night work þ.ħłlˡ.tʲ
Midnight Dinner (1) [Link]ᵘkⁱ.tʲ
Predawn Rest
2 Sunrise Work
Morning Lunch (1) f.nᵘtᵘb.tʲ
Noon Sleep (2) ð.nᵘsˢdᵘ.tʲ
Late afternoon Awake kʲ.dypls.tʲ
Sunset Work
Predawn Rest
3 Mid-morning Work
Early afternoon Lunch (2) [Link]ᵘ.tʲ
Afternoon Recreation rⁱ.tʲ
Evening Reflection ð.tʲkⁱlł.tʲ
Night Sleep (1) ð.zl.tʲ

Already this immediately reveals that the triad has four meals and two
sleep periods, both of each have words that correspond to them – however in
the modern day they have obviously drifted in meaning to converge better
with the modern, Grade (3) civilisation’s norms.

4.1.2 Southern schedule

Further to the south, where triads are less prominent, there is an alternate
regime that comes in a different set of words. This is the comparatively
58 CHAPTER 4. THE CULTURE

newer system, which although not originating from the Grade (2) system is
greatly promoted by and used by them. (27 | 39)
The system can be characterised as being a halfway house between the
traditional system as described in the previous section and the modern day-
based system, in that the cycle between work and sleep is based on the triad
but the day is still respected and sleep is always at night. Roughly speaking,
the distribution of meal and sleep is as follows:

Day Meal Sleep Drsk Rattssaw

1 Breakfast (B1) [Link].tʲ ffoa
Dinner (B1) f.sˢdrr.tʲ* ssdroar
Sleep (B1) ð.ksˢrv.tʲ* yoámwà
2 Breakfast (B2) f.vᵉvᵉf.tʲ* éoyéf
Lunch (B3) [Link].tʲ kfoá
Sleep (B2) ð.drk.tʲ psoer
3 Breakfast (B3) f.tᵗsˢts.tʲ ttssayó
Lunch (B3) [Link].tʲ* eards
Dinner (B3) f.mᵘŋ.tʲ mom
Sleep (B3) ð.mᵘlˡp.tʲ eloa

Words marked with * in the table are obsolete in this sense and the roots
have either changed in meaning or have disappeared altogether in a process
which we will discuss in section 8.2.5.
There are also generic words for meals and sleep:

English Drsk P8 Drsk T Rattssaw

Meal [Link].tʲ [Link].·.tʲ oabps
Sleep ð.tmpl.tʲ ð.tm.·.tʲ tmbery

It should be noted that these two words are obsolete with these meanings,
in much the same way historically: the roots are retired in the form as shown
here, replaced with the empty root rule (see section 16.7.2) and a CAT MAJ
that is based on that root.
As with the northern schedule, the first two days are mostly filled with
work, and the last one with rest, though some people might work with an
even schedule with the second day being a rest day after lunch (B2).
This daily schedule is mostly obsolete with the abandonment of its
strongest supporters, the Grade (2) civilisation, but the impact on the cul-
ture and the language remains. For the latter, the impact is largely based
on vocabulary, with much of the words that described its activities being
8
For old forms of the language, the morphemes are translated back into modern (Drsk T)
language without affecting their order. Additionally, modern genders and suffices are added
back on.
4.1. DAILY LIFE 59

passed on to the modern language, as we have demonstrated above. Some

of these have also evolved into verbs. For example, a generic verb for “sleep”
is also tmpl.

4.1.3 Collision of the two schedules

The two ways to divide up the triads certainly do have their conflict points,
and this has been exploited in various ways over the many millennia.
In the old days, there were the two schedules that we have outlined in
sections 4.1.1 and 4.1.2, which already is a ripe stage for conflicts. with
adherents of neither schedule wanting to or can be made to change, the area
where Klzdmk now resides is largely split into three economic forces: one
for each of the schedules and a small class of go-betweens that coördinate
between the two. This eventually gave way to a more geography-based
distinction, where individuals are expected to yield to the local schedule.
With the invasion of the Grade (2), things changed quite a lot. The most
critical of these changes is of course the fact that one of the schedules are
now heavily supported by the dominant Grade (2) – the Southern one – so
the northern schedule is beaten back to small or peripheral areas. Though
that is the case, the old native schedule retains some grip in some of the most
peripheral locations, such as Nffek, but even then eventually the city and
its regions have largely been pacified out of their usage of the old schedule
purely out of economic might. A few holdouts do remain and this is the
basis on which the old words survived (though they would have survived
regardless, via metaphor and semantic drift).
After the Restitution, the incoming Grade (3) civilisation introduced
the day-based schedule, which has in turn has challenged both triad-based
schedules with even greater economic might. However, because of the histor-
ical baggage of Grade (2), the Southern schedule took much more damage to
its hegemony than the Northern one, which actually took a small recovery
(though ultimately not much).
The day-based system ultimately won out over all of the other schedules
in most respects, though the holidays are still based on triads and to some
extent the triads are now treated as “sub-weeks” on top of the normal weeks
of eight or nine days. The old system of having go-betweens for each of the
schedules have been revived for the modern era for much the same purpose:
to liaise between individuals that are living on different schedules. This is
also mitigated using the system of “parallel holidays”, where some of the
population take some holidays on an offset (usually at about 13 yr either side
of the actual holiday), which makes go-betweens slightly less necessary than
in the past. (? | ?)
60 CHAPTER 4. THE CULTURE

4.2 Food
We will now describe the basic food pattern of the country on a meal-to-
meal basis. First, we must note that there is a fair amount of variation
of what food is eaten in what meal where and by whom; as might what
be expected, there is significant variation of food intake across geographical
and class boundaries. These particular meal plans represent that of a typical
middle-class individual with no particular dietary restrictions in three areas:

• Nffek, a coastal, mostly urban area with a relatively small palette but
with no particular pressing concern for food supply;

• Woróli, an inland, mostly rural area with some food supply limitations;
and

• Trokier, an inland, mostly urban area with large amounts of food,

mostly imported, supplemented by fishing in area’s rivers

4.2.1 Breakfast
This meal is generally considered to be the second largest meal, and largely
consists of three main components, plus drink: a meat component ([Link].tʲ), a
vegetable component (f.mᵘdᵘ.tʲ) and a small amount of something known as
[Link]ⁱ.tʲ, which is a surprisingly highly processed food product largely derived
from liquids such as *milk and *syrup that provides a large amount of energy
for the day ahead. [Link]ⁱ.tʲ can be compared to be some kind of *coffee.
In Trokier, breakfast is largely curtailed in favour of a smaller meal con-
taining only the drink and fzlⁱtʲ, which may or may not be compensated for
a third lunch known as [Link]ᵘ.tʲ, whereas in Woróli the meal is considered
the most important and is a little bit larger in all portions than in other
places, with the exception of the meat component, as meat is in slightly
shorter supply there.
A list of common dishes for breakfast for each of the three places we have
described is in table 4.2.

4.2.2 Dinner
Dinner in the traditional sense is largely meat-based with some staple food
attached to it. It is intended to allow for through-the-night working, and in
some sense is more lunch-like than the two lunches.
This is also the time when a small amount of alcohol is traditionally
drunk, in the form of a *wine of sorts (that is to say, fermented *grapes1 ),
but this is not frequently done in urban areas, and in any case does not
amount to much intoxication (though due to differing tolerances this turns
out to be about 20 to 30 per cent more than the average humans – though
4.2. FOOD 61

Table 4.2: Typical menus for breakfasts at selected locations.

Loc
Nffek Woróli Trokier
C

Entree ... —
[Link].tʲ
*bread
f.n.dʲs.tʲ
*berry1

Meat
f.dʲlⁱ.tʲ [Link].tʲ *mut- f f.k.tʲ [Link].tʲ
*chicken1 with ton (bites) *mutton bites
*potato sauce and *rice3
and *rice3
[Link].tʲ
*beef2 with
marrow sauce
[Link].tʲ *mut-
ton (bites)

Vegetable ...
f.kⁱzd.tʲ f.kⁱzd.tʲ
*potato *potato
(baked) (baked)
f.rčk.tʲ f.rčk.tʲ
*bush5 *bush5
f.dᵈkv.tʲ
*leaf-ball

*Coffee
[Link]ⁱ.tʲ *coffee

Drink
[Link]ⁱt.tʲ water [Link]ⁱt.tʲ water [Link]ⁱt.tʲ water
[Link].tʲ fruit [Link].tʲ fruit
juice juice
[Link]ᵘ.tʲ
*wine
62 CHAPTER 4. THE CULTURE

there is a significant urban–rural difference for alcohol tolerance). The *wine

is stronger in Woróli however.
There is an old tradition that states that dinner should be eaten with
everyone facing the sunset, or west if the sunset is not otherwise visible.
Hence the term

(2) kls ð.nb.t

eat sunset
“Eating the sunset”

for this particular meal, which persists even today.

4.2.3 Lunch (1)

The first lunch is a relatively light meal that mostly sits between breakfast
and dinner in terms of food; however, the composition is mostly grains and
staple foods, with dishes tending to be plain, simple and formulaic. Common
dishes that is eaten include several types of savoury pastries, a clear soup
with a chunk of *chicken in it, and *eggs2 and *eggs3 9 mixed together with
a thin slice of *ham. This remains true throughout all regions with little
variation; even in the most impoverished areas, the first lunch normally
contains some or all of these dishes, with little else thrown in.
The formulaic meals of this meal make it an ideal platform to hold cook-
ing competitions, when standardisation is to be emphasised, though there
are competitions for both creative and standardised cooking that are com-
mon throughout Klzdmk. Language-wise the meal is typically used as a
metaphor for routine and simple items, and is a common reference point for
things that are normally as such but has gone beyond it and is now eccentric
and strange.
Even in ancient times, snacks may be eaten at this time instead of a
full meal. Closer to the modern day, with the day becoming the dominant
timekeeping era, this meal is ever the more marginalised, with much of its
properties combined and subsumed with second lunch.

4.2.4 Lunch (2)

The second lunch is the largest and most important meal in the entire triad.
There is relatively little in terms of restrictions on what can and cannot
9
*Eggs1 , that of a particular type of meat-bearing land animal known as the [Link]ᵉ.tʲ, is
generally used as an ingredient for further processing and is not considered a food without
such processing. It’s not easy to provide an English translation for such an animal because
its properties don’t really match any Earth-based animal even as a rough approximation.
There’s also an *eggs4 and *eggs5 , of which the former is formally considered inedible,
which are produced by various types of *fish. *Eggs4 is typically used more for dyes and
glue, but cultures elsewhere do consume it.
4.2. FOOD 63

be eaten during this meal, only that it is to be fairly lavish and consists of
mainly savoury dishes. Staples are a common, but by no means universal,
sight for this meal, more commonly seen in Woróli and Trokier than in Nffek
where the meal tends to consist largely of meat and vegetables.
This is also one of the few meals that frequently consist of multiple
courses, normally two: the main course, and a dessert. The dessert is, unlike
some desserts common on Earth, also savoury, with sweet dishes being only
a minor component of the dessert. However, the more aristocratic members
of Trokier do eschew the savoury dessert for a sweet dessert more familiar
with Rattssaw culture, both back then and currently. Some locales may also
add an entree to the second lunch, making it a three course meal.
As one can imagine, this is the meal where the greatest feasts are gen-
erally held. It also helps that this is the meal eaten at the end of a triad,
which helps give it the momentum it deserves. The combination of these
two facts make it so that the second lunch has the same position as a dinner
is in contemporary Earthling society.
In the modern day, with the single day being the primary cycle, the
word for the second lunch has drifted to mean any feast that is, for example,
used to commemorate some event or another. This can be reversed with
additional specifications, of which see… Also, as previously mentioned, in
the modern day some of the characteristics of other meals may or may not
be combined into it, in particular the west-facing requirements of the dinner
and some of the formulaic dishes from first lunch.

4.2.5 Snacks
Whether or not someone has snacks during the day outside of these meal
times is a matter of choice, and is one that is roughly evenly distributed
between approval and disapproval. Regardless, there are many small dishes
that are largely regarded as only snacks despite requiring quite a lot of effort
to prepare and serve them.
Snacks in the region come in three forms:

• Hard, long-lasting, room-temperature foodstuffs that take a significant

amount of time to chew and bite down into, are largely analogous to
hard candy, and are mostly intended to keep one’s mouth shut more
than sustenance;

• Soft, perishing, hot or room-temperature foodstuffs that resemble a

small version of a normal meal, including flowing liquids that can make
eating a bit messy, analogous to dim sum in small quantities; and

• Soft, long-lasting, cold or room-temperature foodstuffs that are cov-

ered in sugar and salt and is intended to provide a flavour hit as a
64 CHAPTER 4. THE CULTURE

primary goal. This is analogous to other types of candy and even

waffle-type desserts.

…
Though these categories are similar, there is a large variety of snacks
that are all fairly regional, so it would not be fair to merely list those of the
three regions that we have been using as points of comparisons.

4.3 Climate
To describe clothes, we first need to see why anyone would want to wear
clothes, which is normally dictated by climatic requirements.
First, we need to clarify some important global concepts that are true
throughout the entire planet. The first and most important thing about
Pasaru in terms of understanding the climate of Klzdmk is that the atmo-
sphere is significantly thicker, with the mean air pressure at sea level being
1540 hPa. This changes the boiling point of water, raising it to 115 ◦C. How-
ever, the freezing point of water remains near 0 ◦C, so we don’t have to worry
about that.
The increased air pressure also means that the wind and the rain are a
little more severe in terms of impact for the same net amount of wind and
rainfall.
We first indicate the climate of the rural areas near three major cities:

• Trokier, representing an inland forested climate;

• Nffek, representing a coastal climate; and

• Keutkk, representing a more arid inland climate.

These three cities are chosen for their representative power in the general
climate of the country, which we will see later as we describe the climates in
depth.
The climate charts are displayed in figures 4.2 to 4.4. In these figures,
each of the 21 months of the modern calendar, as described in section 2.5.4
– with the names written in E.-Pasaru, as is customarily done in climate
charts in Pasaru10 – has vital statistics indicated: average rainfall, record
low, average low, average mean, average high, and record high temperatures.
These are measured over a period of about 50 years in the contemporary
year of 15 000 PDN. The seasons are indicated to help with orientation, but
as the names of the seasons are meaningless with regards to Klzdmk they
are replaced with generic names.
10
For the full month names, refer to appendix F.
4.3. CLIMATE 65

Additionally, it’s also at this time where how the year is arranged be-
comes important: the beginning of the year represents autumn in the south-
ern hemisphere, as it is the tradition that the year be ended in a big rest
period, which is what summer typically is for a civilisation that rarely needs
to tend to fields in the summer but does have a very intense harvest season.
As Klzdmk is in the northern hemisphere however, this means that the end
of the year tends to be marked by the end of winter, which in turn means
that hill-shaped temperature curve is shifted to the left in this graph relative
to what Earthlings might expect, resulting in a more tilde (~)-like shape.
Such readers may understand this as having your winter holiday celebrations
at or around mid-October.
First, let’s discuss the commonalities between all three areas. Being in
the temperate zone, there is a clear-cut display of four seasons, and in this
case because of the way the year boundary is placed they are also aligned
with the year boundary. In almost the entire country it gets cold enough
to snow, though this rarely actually happens in most areas, with the plain
where Trokier is sited in does see regular snow, which is referenced in certain
metaphors.
Furthermore, there is a relatively low amount of temperature variation
throughout the year. This is partially due to the thicker atmosphere mod-
erating things. Nevertheless, in inland areas there is still potential for large
temperature swings both daily and annual.
Now that the things in common have been discussed, let’s talk about the
differences.

Year Upper Third Year Middle Third Year Lower Third

Temperatures in °C
Percipitation in mm
34 33 34
31 31 30
27 28 28
28 27 28 25
25 26 23
22 22 23 23 23 23 24 21
21 22 21 19
18 19 19 20 19 19 17 17
17 18 17 18 18 18 17 15 15
15 16 15 15 14
14 15 13 13 13 13
12 11 12 11 13 14 11 12 12
10 10 11 11 9 10 10 9
7 8 8 7 8
6 7 8 7 5 6 5
5 4 4 2 3 2
2 1 0
1 1 0 -2
-3 -2 -2
-3 -4 -5
-7

53 60 56 37 40 70 84 88 101 98 86 94 86 102 93 63 44 56 35 47 28
Yo Ma Ďo Ka Ir Po L De Ro St Aď Üs Se E Ha Yi Ȝe At Je As O

Figure 4.2: Climate chart of Trokier. Further details in text.

First, Trokier: the capital of Klzdmk it might be, but the city’s climate
is stereotypically J.-Senlis, with the division between each of the seasons
66 CHAPTER 4. THE CULTURE

being very pronounced. Due to its position in the plain, it does receive a
fair amount of wind and rain throughout the year, but the strongest winds
and the heaviest rains tends to be near the autumn.11 Trokier’s weather
represents the country as it was in the vassalship days as a stereotype, and
this is a common shortcut many works use as a way to indicate the time
of year and the year it is in, regardless of whether or not the place in the
setting could plausibly get that weather at that time of year at that time.
Such clues are typically not that unambiguous though as merely snowing in
the winter is of course a common occurrence in more places than Trokier.

Year Upper Third Year Middle Third Year Lower Third

Temperatures in °C
Percipitation in mm

31 30
29 29
26 26 27 27 27 27 26
25 24 26 26 26 25
24 25 25 24 23
21 22 23 22 22 22 23 22 23 21
20 21 21 20 20 20 20 18
17 18 18 19 19 19 18 18 18 16
15 16 17 17 16 17 17 16
15 14 16 14 14
13 13 13 13 14 15 13 12 13 13
11 11 11 13 12 13 10
8 9 8 9 10 10 11 9 8
9 8
4 7 7 7 5
2 4 5 5 2
1
-2
-4
-7

130 134 130 126 108 135 143 132 128 125 116 124 136 130 136 149 155 114 155 126 125
Yo Ma Ďo Ka Ir Po L De Ro St Aď Üs Se E Ha Yi Ȝe At Je As O

Figure 4.3: Climate chart of Nffek. Further details in text.

Nffek on the other hand is closer to the sea, and that therefore gives a
far more moderating effect to the climate. Therefore, spring, summer and
autumn are not particularly distinct from each other.
In fact, a lot of the climate of Nffek is because the sea surrounds the city
on three sides. For example, the sea also gives what can charitably be called
“a summer” a huge lag, making it so that the city has its warmest month
some one-fifth of a year later than the rest of the country.
The only thing that is not caused by the sea is something that’s called
the Big Cold Wind, which is when during the winter the cold wind from
the polar regions cause the city to suffer intense, bitterly cold, yet very brief
winters. This cold wind is also referred to as part of the national vocabulary,
for example:

11
In this section, we will use conventional English words for seasons in the way that most
Earthlings expect: summer is at the hottest point in the year, winter is in the coldest point
in the year, spring is after winter but before summer, and autumn is after summer but before
the next winter.
4.3. CLIMATE 67

(3) cbnnᵘfkybkʲmᵘtʲ
cⁱ-bn- nᵘfk-y- bpkʲmᵘ- tʲ
:[divine]-:[miracle]- Nffek-QN- hit- CITE
A sudden and unpleasant surprise

(For more information on how the gloss works and what the divisions
are, see section 16.4.)

Year Upper Third Year Middle Third Year Lower Third

Temperatures in °C
Percipitation in mm 39
35 34 35
32
30 31 31 29
28 28 30 28
28 27 28 28
23 26 25 25 24
23 24 24
22 22 21 20
18 19 20 20 19 20 18 17
18 18 16
15 16 16 16 17 16 16 14
13 14 14 16 14 12
12 12 11 12
11 12 12 10 9
11 8 9 9 9 8
7 8 7 7
7 6 7 5 4
2 3 3 2 2
3 2 0 1
-1 -1
-2 -2 -3 -3
-3 -4 -6
-7 -8
-8 -9 -8
-11 -10 -11
-16

103 94 46 19 37 35 49 26 15 9 36 59 44 67 101 167 263 296 203 144 153

Yo Ma Ďo Ka Ir Po L De Ro St Aď Üs Se E Ha Yi Ȝe At Je As O

Figure 4.4: Climate chart of Keutkk. Further details in text.

Finally, Keutkk is the representative of a more rural, rustic, “indigenous”

(insofar as the country can have an indigenous population) side of Klzdmk.
The difference between summer and winter is highly pronounced, with spring
and autumn being also mostly a linear affair between cool and hot. There
is also a pronounced dry and wet progression, although in this case this
is mostly due to other geographical concerns – a persistent frontal system
moves north and south each year and it is during the winter when the it just
so happens to pass over the city, causing poor weather.
This cold and windy winter is also referred to in Drsk:

(4) (text)
...
...
A very poor situation lit. “Not as bad as a winter in Keutkk”

In which we also notice the euphemistic tradition of this particular con-

struction in Drsk.
68 CHAPTER 4. THE CULTURE

4.4 Clothing

4.5 Housing
The traditional house of the nation is a general synthesis of the ancient,
Grade (1) one-storey courthouse and the slightly less ancient, Grade (2)
three-storey “tower block”. In either case, we will have to appeal to the
climate again to describe various aspects of why the house is the way it
is. In particular, we will also appeal to the local mineral situation to help
understand the building materials needed. This will happily give us the
opportunity to name some materials in Drsk too.

4.6 Music
The culture of music in the modern world is largely converged to the in-
ternational culture of music, but there is some remaining cultural influence
from the J.-Senlis days. Much of the musical influence of the old, Grade (1)
culture has more or less been entirely erased.
This section largely deals with non-lyrical music, as there is in many
cases a very tight combination between poëtry and music, as one might
expect. The combination of the two will be treated in more detail in section
27.7.

4.7 Games
A number of games are also important contributors to the language, espe-
cially in vocabulary and fixed phrases, so we’ll describe them in brief detail
here.

4.7.1 Digitex
This is a game of numbers that requires some understanding of the workings
of the digit state machine, which is deeply embedded into Drsk as part of
its number naming system (see section 17.1). As such it can be presumed
to have been invented contemporaneously with the formation of the state
machine, but there is a fair amount of ambiguity to which one of the two
strictly came first. Nevertheless, it is very much a characteristic part of
Klzdmk life so it is worth it to describe it here. (36 | 8)
 Part III

Vocabulary & Etymology

69
Chapter 5

Neighbouring and Related

Languages

In section 3.10 we have mentioned how the country, Klzdmk, handles neigh-
bouring and minority languages, but we haven’t actually introduced what
they are and how they affect Drsk. As these languages have a direct and
notable effect on Drsk, it is helpful to introduce these languages on their
own terms.
The languages are arranged roughly in order of influence and closeness,
which is in turn in order of how they are categorised (see section 3.10.1 for
recognition categories).

5.1 Rattssaw
Even after the Restitution stripped it of much of its strength and territory,
the J.-Senlis remains a significant and credible power in the region, and in
the case of Klzdmk is never too far, being just across the border to the
east. Accordingly, the language it speaks, Rattssaw #rattssá /ʀat+ ts+ au̯/
[ʀat͡θ̆t͡s̆au̯] continues to be an enduring influence on the current influence on
Drsk, over and past its original reach being the prestige language of region
when Klzdmk was just a vassal of the J.-Senlis.
Today, as previously explained, Rattssaw is official in Klzdmk and every
province within it except Nffek and Eamttel. It is also the sole official
language of the province named Trokier, with the exception of the capital
Trokier, which is bilingual for largely political reasons (actual Drsk speaking
rates in the capital isn’t significantly higher than the rest of the province).
Rattssaw is, by most intuitive measures, the reason why Drsk has so
many of the properties that it has today: it’s a syllable-denying language,
which Drsk has explicitly inherited (see chapter 10); its vocabulary is by
far the largest single stock that Drsk vocabulary takes from aside from
its inherited stock (see chapter 8.2); and while arguable, the propensity

71
72 CHAPTER 5. NEIGHBOURING AND RELATED LANGUAGES

of Drsk to have largely atomic verbs (see chapter 18) and highly structured
nouns (section 16.2) is a reflection of how Rattssaw has highly structured
verbs (14 | 64) and largely atomic nouns (15 | 9). In particular, the idea of
phonoruns originated from Rattssaw and is heavily tied to it and its culture,
to the point where any description of phonoruns is not complete without
some discussion of Rattssaw phonoruns. This is what we will do in section
10.1, for example.
There still remains parts where the languages clearly don’t have much
influence against each other though; for example, the number system of
Rattssaw and Drsk (for which see section 17.1) are completely unrelated
to each other. Additionally, other more superficial properties of Drsk that
Rattssaw has pushed onto it do exist and are of course “inherited” strongly,
largely relating to orthography, and in particular the letter order.
As a language of its own Rattssaw is not particularly bizarre, though how
much of it is because it doesn’t have much in the way of strange features and
how much of it is because of simple familiarity due to exposure is a separate
story. It’s a language that is largely isolating, like most of the languages in
the region.
Visible Rattssaw vocabulary is rare in ordinary Drsk sentences, though
they do exist in obscured form, covered by layers and layers of Drsk grammar
and morphology. They’re mostly naturalised into ordinary Drsk words and
are indistinguishable from them, with no particular connotations associated
with them.

5.2 xRron
To the south and southwest of Drsk is the kingdom of Earmol, which speaks
a relative of Drsk called xRron /#ʀõ/. It is a relative in the sense that
both of these are Remmsp languages and they do share a common ancestor
(see figure 1.2) and the fact that the Oerds and Śeltr ethnicities are also
strongly related makes the languages considered to be closer to each other
than Drsk is to Rattssaw, which is also a Remmsp language in part but has
been hybridised with languages further to the east and therefore has less
connection to the Remmsp.
In a few words xRron is almost like Drsk if it had not been influenced
so heavily by Rattssaw by the way of political and linguistic takeover via
vassalship. Much of the vocabulary between the two languages are recog-
nisably related even if they do not look similar and requires a bit of analysis
to actually correlate one to another. Here are some examples:

English Drsk xRron

… … …
5.3. NNN HEEEL, GENBARG AND ṔERI 73

Grammatically speaking, the two languages remain somewhat similar

to each other, but xRron is far more inflection-heavy and maintains many
historical roots dropped by Drsk. However, there are some changes that did
affect both languages, such as the formation of the digit state machine.
In this grammar we will refer to xRron both as a proxy for their common
ancestor, proto-Remmsp, and as simply a neighbouring language.

5.3 Nnn Heeel, Genbarg and Ṕeri

To the north of Klzdmk are a federation of several quasi-countries that are
collectively known as the Esħ Pez /esx pez/, which exist on two island
named Řallona (the large one to the east) and Balogen (the small one to the
east). Exactly what the Esħ Pez is, its history and its own story in general
is largely beyond the scope of this grammar, but it is important to describe
the languages it speaks and how they affect Drsk.

Figure 5.1: A map of Řallona and Balogen. The spheres of influence of each
language is marked.

The three languages of the Esħ Pez are Nnn Heeel, Genbarg and Ṕeri, of
which only the first have a large impact on Drsk. The other two languages
are insulated from Drsk by Nnn Heeel, and are also isolated genealogically:-

• Genbarg /geNbaRg/ [gẽba˞k] is a descendent of Âagenzbèe, a language

clear on the other side of the northernmost parts of the Agelta Ocean,
and therefore a part of the Siťhanhor branch of the Soþip languages,
74 CHAPTER 5. NEIGHBOURING AND RELATED LANGUAGES

i.e. essentially completely unrelated. Its closest relatives is Cindri,

further to the north, and Âagenzbèe itself, which has recently been
revived as a common language in historical home on Agenbi.

• Ṕeri /p͡feɾi(ː)/ (Nnn Heeel “Pperi” /p͡feri/, Genbarg “Pheliy” /pʰelii/),

is an international language, but specifically only for Genbarg and
Nnn Heeel. It was primarily created in order to facilitate transfer of
information between the two language spheres, which as is depicted
in figure 5.1 is geographical in nature. It was constructed largely by
taking random features of both languages and then mashing them to-
gether to form a compromise languages; though because of previous
contact of the two languages their grammar and vocabulary has con-
verged enough to make the compromise just about work, with some
effort. For irreconcilable differences, some compromise is made, usu-
ally taking Rattssaw words as a neutral but domineering third party.

So that leaves Nnn Heeel /nN heEl/ [nə ̃ hẽ̃əl], the third of the three
languages in the island. This is again one of the languages that have been
affected a lot by Rattssaw to a large extent. In fact it’s in many ways a
clone of Drsk, as it follows much of its developments throughout the years,
and it has a similar origin as well, though there are clear differences between
them. Chief among them is that Nnn Heeel is one of the very few languages
where names actually mean something, and its own name is no exception:

(*) nnn heeel

our thought

It’s not a particularly remarkable language, but it did give much of its
phonological features which kick-started Drsk’s long journey into vowel-free
territory – in particular the former’s special vowel /E/, indicated in orthogra-
phy by tripling the previous letter, is widely regarded as the most important
factor that caused the collapse of the Drsk vowel system to nothing.
Nnn Heeel is also partially responsible for how numbers are expressed in
Drsk. What that means is that Nnn Heeel’s numeral system also contains
the digit state machine. We will discuss the digit state machine as it applies
in Drsk further in section 17.1, but here we will summarise it as such: digits
are named using a limited number of roots – much fewer than the number
of digits that are needed for a particular base – and those roots always
consist of “commands” that alter an initial number to the desired value.
For instance, in Nnn Heeel, these are the only four “number words” used to
construct digits:

1 2 C D
ħ tt taam e
+1 +2 ×4 −1
5.4. EGAKO 75

To construct the digit 7, for example, you will have to use the command
sequence “+2, ×4, −1”, which becomes “tttaame”. (18 | 22) The Nnn Heeel
number system is actually more complicated than this when it comes to
larger numbers but it is beyond the scope of this grammar to describe that.
In Klzdmk it is a minority language spoken most frequently in the north-
ern shore, except in the areas near Nffek where its presence is not as tol-
erated as in the rural areas (though it maintains a small presence there).
The frequent ferry services to the countries in Balogen help keep the di-
alects in these two areas mixed together and indistinct; similarly, there is a
minor Drsk presence in the corresponding countries, and they are normally
considered to be part of the Nffek Language Community.

5.4 Egako
This is another relative of Drsk that has a lot of influence over it, even though
the bulk of its speakers is in the J.-Senlis, further to the east. There remains
however a sizeable population of Egako speakers in Klzdmk, particularly to
the southeast.
Egako is related to xRron as seen in figure 1.2, and in fact the two are
fairly similar in many ways. That in turn makes it related to Drsk with more
or less the same distance, making its true effect on the language relatively
small or obscured with those from xRron. In fact one can easily state that
Egako is Drsk but with more influence from Rattssaw and less from other
languages. So for instance, while Drsk uses a variant of phonoruns using a
termina
Egako also has a digit state machine, in a far more extreme extent than
either those of Drsk or Nnn Heeel.
…

5.5 Egonyota Pasaru

Egonyota Pasaru, usually abbreviated to “E.-Pasaru”12 is a kind of gerry-
mandered or internationalised dialect of Egonyota Vohalyosun, which is in
turn a language from the central continent named Ordžojan which is spo-
ken in the place that due to the World War because the single undisputed
capital of the world. And because of that position, the language spread
everywhere, but especially in Apurhagat, where immigration from Ordžojan
are particularly encouraged to attempt to dilute the influence of Rattssaw,
and so the J.-Senlis.
As a language, it’s also not particularly remarkable; it’s a fairly mun-
dane language with the only notable feature being that it uses a VSO word
12
There is a series of abbreviations that represent standard abbreviations for things like
“Language”, “Kingdom”, “City” and so on. These are seen in part in (16 | 55).
76 CHAPTER 5. NEIGHBOURING AND RELATED LANGUAGES

order, which is not uncommon but is one of the least geographically bound
features (it tends to show up everywhere, but not in huge swathes). It is
also extremely isolating in some sense, with very few bound morphemes per
word, but it does make heavy use of compounding to the point where its
script is tricameral – an upper case, a lower case and a “middle case” which
is used to indicate morpheme boundaries within words.
There’s also a distinct and somewhat rarely used “pragma” syntax which
allows the language to specify its own rules and change how the grammar
works. This is a deep-reaching change and can alter any part of the language,
even the vocabulary to some extent. It’s normally used to accommodate
second-language users but it can be used to demonstrate features of other
languages without technically breaking the rules of E.-Pasaru itself.
As an international language, it’s frequently referred to as a visiting or
invading language and is a characteristic of the Grade (3) culture. In more
recent days, there had been more and more loanwords that come from E.-
Pasaru into Drsk, though not in the sense that the words are just borrowed
in as it is, with at most minor variations due to phonology – while that
is generally the case for verbs, for nouns the common practise is to calque
compounds in E.-Pasaru back into equivalent Drsk roots (or the closest
equivalent thereof) and then reconstruct it with the appropriate native Drsk
genders. Certain other forms are also imported at this stage, such as the
relative clause structure but in a restricted context.
Overall, E.-Pasaru words in a Drsk sentence tend to evoke a sense of
modernity in when used liberally. It’s not a common occurrence in normal
speech, but when discussing modern technology or cultural terms such terms
are mostly inevitable. In that sense it’s very similar to English in the modern
day on Earth.

5.6 Internal languages

A language is “internal” according the Klzdmk government when its extent
is mostly inside its borders and it isn’t Drsk. Such languages are sometimes
called minority languages in other countries, but the Klzdmk government
does not use this term primarily because certain other countries do.
Regardless, the description “minority languages” is mostly apt both be-
cause only a few individuals and groups speak them, and that in a political
sense they are “minors” – that is to say, they are sometimes hard to dis-
tinguish from Drsk, due to the way that languages evolve. There are some
exceptions to this rule, however, which makes the government’s decision not
entirely political in nature.
These languages are typically fairly close relatives of Drsk, and in some
cases they can be mistaken (deliberately or not) for dialects because of how
5.7. OTHER SPOKEN LANGUAGES 77

similar they are. Typically they differ from Drsk in the following manners,
in order of most to least obvious.
The clearest difference is the phonology of these languages. Local mi-
nority languages tend to adopt most of the trappings of the Drsk language
region that surrounds them, and there is ample evidence that this is bidi-
rectional – the minority language influences the language region, which in
turn reïnforces the minority language’s phonology.13 This is only possible
because the two languages are very closely related in the first place, so there
is still a chance for the phonology of both languages to re-converge on each
other.

5.7 Other spoken languages

Like any other vibrant civilisation, there are hundreds of thousands of lan-
guages that exist throughout the history of Pasaru from the start of history
to the time where Drsk as described here is spoken, of which about one
third are extant at that latter time. In turn a good deal of them are also
languages of diaspora that don’t merely exist within Klzdmk’s borders, but
are instead spread across a large region, or perhaps even the entire world,
making them not internal languages as per the definition thereof.

5.8 Nonspeaking languages

Signed languages in Pasaru are more accurately called “nonspeaking lan-
guages”, because the only thing in common that each one has is that none
of them uses sound as a method of encoding abstract ideas into tangible
objects – in terms of the language skeleton described in section 1.3, this is
the language’s phonology and are explicitly identified as such. Most of the
languages require the use of some prop or other speaking aid to articulate
entirely, with only a minority requiring only a speaker’s four (or two) hands
and their positions.
There are 18 recognised nonspeaking languages that are recognised as
being at least marginally used within the borders of Klzdmk, the largest
three of which are:

• 6G-J, the globally recognised “international” nonspeaking language

which has much the same power of E.-Pasaru. This is a nonspeaking
language that requires a speaking rod in order to articulate correctly,
which gives it its English nickname “rodlang”. The rod can be touched
at up to two points, rotated, presented, and moved as part of the
13
Note that this can also be used offensively, in the sense that sufficiently motivated actors
can use this convergence to claim that the minority language is “just a dialect” and therefore
is not worth protecting as a language.
78 CHAPTER 5. NEIGHBOURING AND RELATED LANGUAGES

language’s phonology. It is typically known as “E.-Dďeyš” for those

that do speak.

• Trokier Sign Language or West Senlis Sign Language is a lan-

guage that uses something called a “5×5”, which is a grid of five squares
by five squares that speakers drag or tap their fingers along in order
to express words and phrases.

• Nffek & Eamttel Sign Language is also a 5×5 language, and is

fairly similar to Trokier Sign Language in terms of grammar. However,
its syntax in particular diverges significantly from Trokier SL and has
overtures of Nnn Heeel Southern Sign Language.

Sign languages typically use only one of a limited number of orthogra-

phies, and those are distinct from those of the spoken languages. The most
common orthography is called Gloss gðmᵘsytʲ, which in fact has a(n admit-
tedly small) effect on Drsk, in the way that in Drsk, a common nickname
for nonspeaking languages is

(5) gðmᵘltʲ
g- ð- mᵘl- tʲ
[Link]- nonspeaking- [Link] cite

which derives from the fact that one in Text, the word for unspeaking
languages is typically indicated by a pentagram,14 unlike all other words
which are indicated using gloss words from a surrounding metalanguage (or
E.-Pasaru if there is none).15
These nonspeaking languages also have a divergent grammar from spo-
ken languages, although in some cases the difference is simply a matter of
labelling. Nevertheless, many nonspeaking languages use 6G-J’s model of
space-like and time-like words. (24 | 36) That is to say, words are not di-
vided into nouns and verbs but whether or not the concept has a definite
extent in space and time. Sentences are composed by combining one time-
like word and one or more space-like words (or the other way around) so
that they specify a particular event happening in a particular space and
time. (24 | 38)

14
In Latin text, Gloss is written within angle brackets and the pentagram is replaced with
a filled star, like this: hFi
15
Gloss is largely used to embed nonspeaking languages into text of speaking language;
another script called Text is used to write 6G-J in its own terms, and a third script called
Notation is used to indicate words phonetically in a fairly bulky notation resembling a music
staff (both in Latin form and in the original native format).
Chapter 6

The External History of

Drsk

[Insert E.-Pasaru text here]

The language has proven to be easily dissected after a few years
of careful analysis, but the things that have come out of the
breakdown has been thoroughly wonderful, collecting some of
the rarest and inspiring elements of languages from across the
world… it’s as if the language comes from the old gods them-
selves.
— Aħuverbt Tiknærė, Professor of Language Analysis at the
Hyper-University of the Edrensano, 11335 PDN 942°Y

In this chapter, we will examine how the language’s vocabulary evolves,

from words that have been inherited from the language’s descendents to
words that have been given to it through the various waves of invasion and
immigration above. Additionally, we will also examine how Drsk eventually
evolves each of these strata in its own way, apart from everywhere else.
Inevitably therefore there will be some overlap with the more general
summary of the history of the region and the country which have been
covered in sections 2.1, 2.2 and 2.5, but this time there will be more focus on
words and diachronic evolution. There will also be a more detailed discussion
on the evolution of Drsk before the present time and how those changes affect
words.
Several of these changes will also have to refer to concepts of the lan-
guage, which we have not actually described yet. As such we will have to
refer to future chapters in this chapter as appropriate.
This chapter complements Chapter 7, which describes some of these
and other changes of the language exclusively through changes that aren’t
borrowing features and other objects from other languages.

79
80 CHAPTER 6. THE EXTERNAL HISTORY OF DRSK

6.1 Chronology and the Feature Change Rate

The events described in chapters 2 and to a smaller extent 3 occurred in
the time period depicted in figure 6.1. As previously mentioned in section
1.7, language evolution in Pasaru occurs more or less in staggered steps
creating distinct eras which are indicated with letters. The start of a new
era of a language typically indicates some minor or major overhaul to the
language, though which ones get a letter and which ones don’t is largely
a contention of politics, not linguistics in general (though there are some
linguistic considerations that do override at some point, but that is largely
because at that point history will start backing the linguistic claims.)

6.1.1 The ages of Drsk

Broadly speaking, when discussing the history of Drsk, we use the six major
ages of Drsk: S, F, P, Ri , Lˡ and T. There’s also the “before Drsk” times,
which we will label pre-S (it’s also Remmsp E but pre-S is a better label when
referring strictly to Drsk only), and a period of time where the language has
been suppressed, which is when it is called P but also spawned a creole la-
belled A. These eras can be classified in table 6.1, which demonstrate that
the eras are specific combinations of “internal” (endogenous) change com-
pilations and “external” (exogenous) change compilations. Internal change
compilations are bundled into “periods” and given ordinal labels like they
are in traditional Pasaru linguistics16 , while exogenous change compilations
are marked with “phases” and numbered using cardinal labels.

Table 6.1: Ages, phases and periods of Drsk. See main text for further
details.

Age Phase Period

Pre-S Phase 1 –
S Phase 1 First
F Phase 2 First
P Phase 3 Second
Ri Phase 3 Third
Lˡ Phase 3 Third
T Phase 4 Third
M Phase 4 Third

Eras, in this context, are merely the combination of phases and periods
in such a way that each era roughly corresponds to one phase and one period.
This is considered a sufficient requirement for a change in age letters, any
16
In deference to the traditional naming styles on Earth, though, the first period can always
be referred to as the “Early” period, so for First Drsk, we can say “Early Drsk” instead.
 6.1. CHRONOLOGY AND THE FEATURE CHANGE RATE 81

-12000 4000
-9800 -1800 -70

Rattssaw occupation Vassalship

-8500 -5000
R
U Ss

-11000 -6700 -3800 1080 3210

S F -5500 P Ri T D
D
-11900 -9800 -1800 260
I
(1) (2) Lesser (2) (3)
-2000 500
-1800 -70

Vassalship

-1900 -1250 -850

R
T Þþ Ff

-1800 -500
D
Lˡ Full suppression

-1800 260
I
Lesser (2) G (3)

Figure 6.1: Approximate timeline of the events described in this and previous
sections. The timeline below is an expansion of the vassalship period on the
top timeline. Each row in the entries represent a category of events, indicated
by a letter: R for a Rattssaw language era; D for a Drsk language era; and
I for various grade-based invasions. Years are in PDN.
82 CHAPTER 6. THE EXTERNAL HISTORY OF DRSK

major change in either phase or period will trigger an age change, because
it has also changed the language enough at that point to warrant an age
change anyway. However, the two don’t have to be in sync with each other,
and there are additional considerations that also can change the era letter
also.

6.1.2 The Timescale of Change

The rough timescale that these eras last on is determined by something
called a feature change rate – the mean number of features per unit time
per (active, mutating) speaker that gets added, subtracted or modified. This
value can be quantified in a rigorous way, but there is a number of factors
that have to be taken into account in order for us to actually understand
this number, which is symbolised f˙, we need to clear up a number of details.
First, this measure is partially derived from a number of different sources
that all arrive at roughly the same number. We can measure the feature
change rate directly, in some sense, by merely observing the objects that
denote the dimensions of the quantity: Sample a number of eligible speakers,
note the number of changed features for each speaker, average them out and
divide by the time. However, we can also measure this in a more macro view
by observing the mean era change time τe for a historical period. We can
see that from Figure 6.1 that τe = 2.8(8) kyr. The two are connected by a
simple inverse relationship known as the Shedding Law:
p
τe f˙ = R(t) + R( ds ) + R(ρ) + smaller terms (6.1)
The proportionality “constant”, which is summarised as√R, is composed
of a time-varying background level R(t), a component R( ds ) which is a
function of the distance from the nearest navigable waterway ds , a compo-
nent R(ρ) based on the population density and some additional terms which
take on other factors based on social contact, use of other languages, and
even the nature of the speakers themselves (e.g. older speakers tend to have
a slower rate of change), but these effects are typically not as large and most
social effects can in fact be predicted through just modelling ρ and ds .
It is interesting to note that the background level changes over time, and
in the modern period is in a low that it turns out will never recover. At very
late times, we write:
t−t0
−
R(t) = Rt,0 e τd
(6.2)
which, with the understanding that Rt,0 , t0 and τd all being constants,
is the classic exponential decay formula.
A careful reading of equation 6.1 reveals that the dimensions of R is “fea-
ture changes per speaker”, which has a number of complicated implications.
In particular, the rate of change of a particular language is broadly related
6.1. CHRONOLOGY AND THE FEATURE CHANGE RATE 83

to the total number of speakers, but the exact relationship is not strictly
linear (note that “speakers” are dimensionless and therefore can have any
number of mathematical operations done on it and the final result can still
be dimensionally consistent) but it very nearly is, and it can be modelled as
a piece-wise linear function called Nelk’s Law:

R̂ = N (P )R (6.3)
where R̂ is the real feature change rate, P is the population of speakers,
and N (P ) is Nelk’s function, defined as



 P 2, P < P0
log P, P < P < P

0 1
N (P ) ∝ (6.4)


 1, P 1 < P < P2

P, P2 < P
where in turn Pn for integer n are constants. There is a small amount
of uncertainty for all of them to allow for a smooth transition between the
pieces.
Second, the actual value of τe appears very high to most Earthlings,
and therefore the value of f˙ is correspondingly very low. (Consider that in
2000 yr, Latin has evolved into several descendent languages already, com-
pared to this situation where the time between early Drsk and when Drsk
is succeeded exceeds the breadth of this particular civilisation.) This might
seem very long, but recall that generally speaking, the “social development
speed” of Pasaru is very slow compare to modern Earthling situation. As a
general rule of thumb, one year of social development in Pasaru only corre-
sponds to one-tenth of a year of social development on Earth. What “social
development” is can be a bit of a difficult thing to define, but it always
involved society and other life activities.
It is also fairly common to see a single language take the mantle of being
the language’s successor and also its name, compared to on Earth, where
this might happen for only a small number of other languages (e.g. Chinese
up until 1500, Romanian). Here, the situation is slightly less clear-cut; it’s
hard to say that Early Drsk and Drsk T are “the same language”, to the point
where the two are nearly entirely unintelligible with each other, but the two
languages remain very closely related and it doesn’t take too much training
for speaker of one of these eras to easily get used to speaking in another
era’s language. This can be down to a better intuition of the underlying
linguistics by the average speaker but it may also be a genuine connection
that makes the languages less “mother and daughter” and more “continuous
flow from one to another”, despite how the language actually evolves in steps.
Third, notice that the shedding law uses the feature change rate, which
measures the raw number of features changed, regardless of how many fea-
84 CHAPTER 6. THE EXTERNAL HISTORY OF DRSK

tures the language actually has. It is helpful at this point to roughly note
that the number of features a language has doesn’t change very quickly or
very much over time, and there is also very little variation in the number of
features across many languages, even in disparate areas.
Therefore, all languages also has the rough language change rate per fea-
ture, and there is broadly no correlation between the two, with the relation
between the f˙ per feature and raw f˙ itself being completely swamped by
the variance inside each of those two quantities. Nevertheless it is interest-
ing to note that the effect is real; languages with more features are, absent
mitigating details, more stable than languages with fewer features. On a
theoretical basis, this is because features tend to reïnforce each other, but
there are other effects that make this relation not ironclad.
Fourth, the feature change rate is done per speaker per unit time. This
is done in net, that is to say, the feature changes that are measured are the
ones that manage to propagate all the way to the majority of the population,
and that means that those that only manage to extend locally, or those
where the mutation eventually died out are not counted. The rate at which
mutations are generated, regardless of their ultimate fate, is indicated by the
quantity f˙∗ , which is related to f˙ by more complicated dynamical laws that
are beyond the scope of this book.

6.1.3 The Timescale of Drsk

For the case of Drsk in particular, as previously mentioned, τe = 2.8(8) kyr,
which means that we can then calculate:

R = · · · features un−1 (6.5)

R
f˙ = (6.6)
τe
· · · features un−1
= (6.7)
2.8(8) kyr
= 3.4(2) × 10−5 features kyr−1 un−1 (6.8)

This value is the reference timescale of Drsk.

Overall, this is not a remarkable number, being fairly close to the median
and the mean of all languages, both globally and regionally. Keep in mind
that the variation across languages is also not very high and for any given
time most languages have very similar R. That is to say, in terms of the
Shedding Law, R(t) dominates all the other terms quite considerably even
up to very late times.
The implication here is that to some extent, the changes of Drsk “sync
up” with the changes in other languages, so if one language changes era, the
other ones start changing as well.
6.2. GENERAL PRINCIPLES OF LANGUAGE CONTACT 85

6.2 General principles of language contact

We cannot genuinely discuss how Drsk changes over time without discussing
how languages in Pasaru, in general, interact with others and how they
change. The interaction between languages is governed by feature theory,
specifically, by the exchange, cloning, convergence and divergence of fea-
tures.
Recall from section 1.3 that languages in Pasaru are conceptualised as
consisting of a skeleton, common to all languages; adorned by features, an
object which is normally depicted as a single sentence which describes a
particular quirk a language has, which in turn might have blanks that have
to be filled in by the language; and fulfilled by paradigms, which are es-
sentially tables that express how to implement the feature in terms of less
abstract objects. This means that the language can be modelled as a graph
of coloured nodes and edges, which is mostly tree-like17 – this can in fact be
drawn, and for Drsk is explicitly created in chapter E.
With this in mind we can now describe how language change occurs when
two languages come into contact with each other.
Language change in feature theory is modelled exactly as such: a fea-
ture in the language’s skeleton is altered by changing one of its parts, e.g.
its paradigm, one of the parameters that is needed to complete the fea-
ture, or even complete shedding, adding or altering of the feature itself.
When it comes to contact, there are three major methods that another lan-
guage (more specifically its tree) can affect the original tree via some fairly
common-sense ways: copying, assimilation and dissimulation. We will ex-
amine each of these, with some examples drawn from Drsk.
Keep in mind that there are other ways that a language can change its
features, not the least through simply doing it by itself. They still use the
same methods as mentioned here.

6.2.1 Definition of contact

Two languages are in contact with each other when two populations of speak-
ers are in the same place at the same time. The “place” does not have to be
physical though it usually is in early times; just when the two languages are
spoken in a context where both can easily observe the other.
In modern times this means that big languages are almost always in
contact with every other language, though this is not reciprocal as small
languages lack the reach to actually reach to the core of the big language
(there are exceptions of course). In a more historical sense however this
is a rarer occurrence, though by no means impossible; one of the easiest
examples to make in the context of Drsk is how Rattssaw managed to be
in contact with Drsk without the reverse happening primarily because the
17
Specifically, this graph is a forest if we only consider edges of one colour at a time.
86 CHAPTER 6. THE EXTERNAL HISTORY OF DRSK

former is in a position of power that the latter lacks and therefore is able
to influence it more than the other way around. This is the “imperial-style”
contact as it is frequently called, an appellation that has continued to the
present day to describe non-mutual contacting.
In the diagrams below contact is usually depicted literally: two language
trees are first drawn separately, then drawn closer to “contact” them. Dotted
lines are used to make that contact explicit, and show what items the contact
has affected.

6.2.2 Copying
A language copies a feature from another language when it, unsurprisingly,
gains a new feature that is identical or near-identical to a feature of a lan-
guage that it is in close contact with.
Because of concerns from other levels of the skeleton, this copying is
not always perfect, but sufficiently abstracted features are usually copied in
directly. Frequently, the primary difference that is generated when copying
come in when paradigms are copied – as this usually involves features further
below the language skeleton (i.e. the phonology) which may forbid entries
in the paradigm from existing in the language where the features are copied
to, the entries are instead copied with modifications to comply with the
requirements those features in the destination language.
Take for instance the introduction of the phoneme /E/ into Drsk from
Nnn Heeel:
…

6.2.3 Assimilation
If the two languages in contact have features that perform similar semantics
one might start copying the other. This is a process called assimilation
and it’s very similar to copying; the primary difference is that the feature
exists in some form on both languages to start with. It happens much
more commonly than copying because making two existing objects behave
more similarly to each other is much easier than having to copy from one to
another from scratch.
As with copying the assimilation is rarely perfect, and now that there is
a base to work from there is also the factor of inertia to think about, which
would resist any large change done to any part of a feature. This means
that despite it being the most common type of language change involving
language contact, it’s also one of the most conservative changes a language
can undergo.
An example of this is…:
…
6.2. GENERAL PRINCIPLES OF LANGUAGE CONTACT 87

6.2.4 Dissimulation
The opposite of assimilation can also happen, because of any number of
factors, not the least because of political reasons. Here, the process is similar
to assimilation, just that when the two features come close to each other they
start mutating so they look different from each other. As we are discussing
language change in a particular language, in the ideal case it’s the subject
language that does all the change and the other language doesn’t change
at all, but in reality both languages’ features usually dissimulate from each
other.
The main problem with modelling dissimulation over the other changes
is in which direction the features change in; with assimilation, “toward each
other” is roughly well-defined, even if not entirely well-defined. Normally
the actual direction of change is “random” – that is to say, there’s no fixed
direction for any particular instance, and there’s no particular factor that
favours one direction over another.
Because of this dissimulation is not as common as assimilation; though
they are not particularly rare some languages show encounter rates less than
8 times that of other types of change involving more than one language. This
1

is the same with Drsk, of which no notable example of dissimulation exists,

though there are some very related examples with regard to vocabulary from
nearby languages:
…

6.2.5 The causation of change

When discussing language change from contact, it’s always difficult to de-
termine which party, if either, initiated the change. Strictly speaking, the
answer is “neither”: languages are not sentient and therefore can’t make de-
cisions about anything, much less changing themselves or other languages.
As previously mentioned in section 1.6 however, it is helpful to model a
language as having a “Will” of sorts, which abstracts away the real – if
beyond-awareness – decisions that individuals make and how they compete
with each other into a black box that acts like it’s sentient but is actually
not.
Even with that abstraction it’s not always clear which language “did” the
change through contact, though the abstraction does help with some of them.
In the case of Drsk, most of the language change can in fact be attributed
to Rattssaw if the contacting language is Rattssaw, solely because of the
out-of-linguistics factors – specifically, that something is that Rattssaw is
politically more prestigious than Drsk, so it (its speakers) tend to influence
Drsk (or its speakers) while the other way round is basically impossible.
This of course doesn’t rule out the ability for Drsk to just change on its own
without any outside influence of course, but that’s a different story.
88 CHAPTER 6. THE EXTERNAL HISTORY OF DRSK

Otherwise, the general principle behind which side “caused” the language
change is a relatively simple logic: given that the languages have been in
contact for a long time, it’s generally considered the language that did not
sustain the change is the one that caused the change. This simple logic is not
always applicable, as has been previously emphasised, but as a first-order
approximation it is a good enough statement.

6.3 Phase 1: Inheritance

The beginning of Drsk is similar to the beginning of most languages in

Pasaru, with the language being split from an existing language, and it did
not get to keep the name.18 The language in question is proto-Remmsp, and
this is where the bulk of the vocabulary of Drsk comes from. Even today,
with much of the vocabulary thrown out due to various internal and external
influences, the core vocabulary list, and the gender system that grew out of
it, is heavily based on proto-Remmsp.
This can be easily seen by observing how the proto-Remmsp words even-
tually evolve into some Drsk genders (see table 16.2 for a full list):

English Proto-Remmsp Drsk S Drsk M

Food /pəfoχɪθ/ /p͡foʁiθ/ f.tʲ
Units /midχɪθ/ /midʁiθ/ n.tʲ
Place /konaχɪθ/ /konaʁiθ/ k.tʲ

The resemblance is perhaps more obvious if we study the verbs, which

have not eroded as much as the nouns have:

English Proto-Remmsp Drsk S Drsk M

Eat
Burn
Irrigate

18
There are two well-studied other ways languages form in Pasaru: pidginisation, when a
language is formed out of combining multiple languages together with some innovation of
its own, where the ancestry does not involve one clear parent, and genesis, which describes
the formation pathway for the three proto-Worlds. The latter is not so much a formation
pathway as it is a black box term that describes how non-languages can evolve into languages,
but relative to contemporary Earthlings as of the current time progress on this field is much
more advanced on Pasaru.
6.4. PHASE 2: RATTSSAW INVASION 89

6.4 Phase 2: Rattssaw invasion

6.5 Phase 3: Rattssaw set-in

6.6 Self-modification

6.7 Phase 4: Grade (3) importation

90 CHAPTER 6. THE EXTERNAL HISTORY OF DRSK
Chapter 7

The Internal History of Drsk

In the previous chapter we have focused on how Drsk has evolved by looking
at how other languages have influenced it. Of course, being a language, Drsk
also has entirely endogenous changes that should be described separately.
This chapter documents such changes using the era letters as a reference of
the progression of time, and each era we will discuss three different types of
changes: phonological, lexical and grammatical, in that order.
Changes in other parts of the language skeleton are, while not always
exogenous, usually so, as they’re not “part of the language” in the strictest
sense. For example, the highest levels of the language skeleton, dealing
with large-scale structures like letter-writing and story-writing, generally
are abstract enough so that the specific language doesn’t matter, so features
on this level from other languages can easily transfer over, and because
of the status of Drsk is transferred over from other languages (primarily
Rattssaw) There are exceptions to the abstract nature of high-level parts of
the language skeleton of course, such as poëtry, and those are significantly
more resistant to external influence, but are in turn not as fast to change as
the rest due to traditions being a more powerful conservative force as well.

7.1 Periods of Drsk

As previously mentioned, there are three or four distinct periods of Drsk,
which are labelled using ordinal numbers. These periods are actually not
the primary focus; instead, it is the transition between them that actually
define the periods.
Specifically, each transition between the periods has a particularly large
suite of language changes which significantly changes the language. The
transitions are large in scale, but this is basically the only thing that is in
common for all of them – they differ in what they affect and how they affect
things.

91
92 CHAPTER 7. THE INTERNAL HISTORY OF DRSK

7.2 Early Drsk: S and F

7.3 Second Drsk: P

7.4 Third Drsk: Ri and T

Chapter 8

Vocabulary from other

languages

8.1 Inherited words

8.1.1 The Remmsp and cognates
8.1.2 Nnn Heeel and other distant languages

8.2 Rattssaw words

8.2.1 Layers of import
8.2.2 Hidden imports
There are some imported words that consists of entirely native parts. These
are the calqued terms and they happen quite a lot especially with regard to
Rattssaw words, due to just how well-entrenched the language is to Drsk’s
development.
One of the prototypical examples of hidden imports are the names of
months, which are entirely based on the month names in Rattssaw, but
none of the words are actually Rattssaw words. (28 | 49) Instead, the gen-
eral principle behind each Rattssaw word is understood, and then broadly
translated into Drsk, then used normally thereafter.
First, keep in mind that unlike most other places, the months are num-
bered starting from 12, and given that there are 34 months in the traditional
calendar, the highest-numbered month is 45. Next, the gender for traditional
months is sˢ.l.ðm.
Now, let’s have a look at the logic behind the naming system for the
months work. In the traditional calendar, there are three seasons: sum-
mer (-f-), winter (-kᵏ-), and the transition between the two which are not
distinguished at this level.

93
94 CHAPTER 8. VOCABULARY FROM OTHER LANGUAGES

We set aside the first and last months and name them separately, so that
we have an even eight nameable months per season. The two months are
named “prologue” and “epilogue” for reasons that will be come obvious soon:

Month Gloss Rattssaw Root

12 Prologue realen ts
45 Epilogue headss fs

Summer and winter are generally considered to be times when events

happen, and they are therefore named as if they were part of a story.19
Corresponding summer and winter months are therefore named identically,
distinguished using the gender system:

Summer Winter Gloss Rattssaw Root

21 37 Begin uereddt lˡt
22 38 Rise ksshll rⁱp
23 39 False peak sscg sˢcᶜg
24 40 Lull er dᵈdᵘ
25 41 Recovery bsseal dᵘr
26 42 Climax krcss cᶜg
27 43 Denouement bngt bmᵘt
28 44 Conclusion ttrrls tᵘs

The transitional months are named in reference to the temperature of

the month in question. They are also named in pairs, but in reverse order
to each other because of the nature of the things that they are named after.
Sometimes, the names don’t match up exactly:

Spr. Aut. Gloss Rattssaw Root

13 36 Cold ssbvff mk
14 35 Thaw–Freeze ein–ffdoar rs–mvŋ
15 34 Arise–Asleep bbnd–uərm rⁱpt–ćrⁱs
16 33 Blooming–Shedding vdtt–ngsp vdð–nᵘsp
17 32 Storms niərddm nⁱdm
18 31 Sun uluat kⁱt
19 20 Wind–Muggy ulle–əarwá znᵘv–tᵗd
20 29 Hot audds bŋ

In this table, where tee spring name is different from the autumn name,
the former is before the dash, and the latter after it.
19
There is a slight complication in that the concept of a plot is a more regional and less
universal concept than one might expect from a purely Earthling perspective. See section
29.4 for details.
8.3. NEW IMPORTS 95

That they don’t match exactly is what is used to distinguish which is

spring and which is autumn, even now when these month names are no
longer in use.
One important thing to note here is that all of these roots are native, but
they all are directly translated from a corresponding Rattssaw root. There
are a few that are also loanwords, with them filling a lexical gap in the native
vocabulary.
This level of import is exclusive to the graded languages and primarily
Rattssaw at that, though there are some exceptions given for some neigh-
bouring peer languages (i.e. xRron for the most part).

8.2.3 Standard cognates

8.2.4 Layer (2) imports
8.2.5 Root collapse

8.3 New imports

8.3.1 Imports from E.-Pasaru
8.3.2 Imports from Sturp
96 CHAPTER 8. VOCABULARY FROM OTHER LANGUAGES
 Part IV

Phonology

97
 99

A historical note

The native method of discussing the phonology in the language is to

write the phones and phonemes in Bugscript. (20 | 55) However,
in the interests of readability to this book’s intended audience they
are replaced with IPA. Refer to appendix A for more information
on this script.
100
Chapter 9

Phonetic inventory

The name of the country of Klmzdk should already raise alarm bells as to
implications on the phonology of Drsk. This would become explicit once we
have seen the phonetic inventory.
However, in general terms, the phonology of Drsk is largely unsurprising
given the context as written in earlier parts, with the majority of its qualities
being mostly predictable from the areal effects of nearby languages as well
its heritage of previous languages.
The table in figure 9.1 shows the phonetic inventory of Drsk. (23 | 2) For
ease of typesetting, and also for historical reasons, the table is rotated from
what it would be in a traditional IPA chart, and where things are placed
regarding manner and place of articulations are reärranged somewhat to
conform to a past version of the language. The abbreviated headings are

NT non-terminal phoneme

T terminal phoneme

NV unvoiced phoneme

V voiced phoneme

where the term “terminal” is defined in section 10.2.

Conspicuous by their absence is any vowel phonemes, which is of course
what was alluded to all this time. This doesn’t mean that the language
lacks any vowel whatsoever; however, it does mean that any vowels that do
exist have no ability to distinguish words, not even their presence or absence
in a particular position. For this reason, Drsk is described as a vowel-free
language, which is distinguished from a vowel-less language in that the latter
genuinely lacks any vowel in any capacity, except perhaps in paralinguistic
situations.
This particular feature is indeed unique to Drsk, though again it’s not
particularly unexpected given the trends that other languages nearby. In

101
102 CHAPTER 9. PHONETIC INVENTORY

Nasals Orals Linguals

T NT NV V T NT
T NT T NT
Labial m̰ m p f-ɸ b v-β
Dental n̰ n t s d z-ʒ ɾ lɫ
tʰ dʰ ɬ
tʲ ʃ dʲ ɹ
θ̆ θ ð̆ ð
θ̆ʲ θʲ ð̆ʲ ðʲ
Dental affricate v͡n t͡θ
t͡ɹ˔̠̊
t͡s d͡z
Palatal ɲ c ç ɟ
cʰ
Velar ŋ̰ ŋ k x g
kʲ gʲ ʎ
kʰ
ɣ
Velar affricate k͡x
Laryngeal ʡ ħ

n.b. /t͡ɹ/̠̊ is a “historical artefact” and would be better notated as /t͡ɹ/.

Figure 9.1: The phoneme inventory for Drsk.

9.1. DIVISIONS OF THE PHONEME GRID 103

particular, if we consider how Rattssaw handles phonoruns, (as seen in sec-

tion 10.1), it’s not too much of a stretch to consider that another language
which picks up the feature from it – as was done in Drsk – would alter it
in its own idiosyncratic way, and further alter the already unusual phonetic
systems away from the norms of language, however relaxed it might be in
Pasaru.
Some phonemes don’t have a consistent NT/T categorisation; they are
marked as spanning both NT and T columns. Whether they are NT or T will
depend on whatever one is convenient when discussing a word’s signature,
which we shall go further into in section 10.2.

9.1 Divisions of the phoneme grid

Though the phoneme grid is the most traditional ways to show the phonemes
in Drsk, there is a lot of detail that is needed to explain it. Furthermore,
there are some details that the phoneme grid missed, and we’ll discuss some
of those things as well.

9.1.1 Terminators and non-terminators, an introduction

The primary division of the two phonemes is the distinction between ter-
minals and non-terminal phonemes, which is a property owned by each
phoneme controlling the way phonemes cluster together into phonoruns,
which would be later explained in Chapter 10.
Essentially, the idea is that some phonemes would terminate phonoruns,
and some won’t, hence the name. Characteristically these phonemes would
be considered “finishers” in some way, which means that typically they are
characterised as short, halting, concluding, and perhaps some combination
of these.
The exact role of the these phonemes are explained further in section
10.2.

9.1.2 Mixing place and manner of articulations

The somewhat unusual arrangement of place and manner of articulation in
the traditional phoneme grid is as previously mentioned a historical artefact,
but let’s give this statement into a closer inspection.
First, the inversion of axes compared to standard Earth phonetics is a
simple difference of choice of convention; there is no particular reason as to
why the place of articulation has to be in the horizontal and the manner in
the vertical. As so it turns out, the current configuration is chosen so that
it would fit in predominantly vertically-oriented tables that are preferred
in most print of the time (see sections 27.6 and 28.3 for a more detailed
explanation of the effect of medium in the language.)
104 CHAPTER 9. PHONETIC INVENTORY

Significantly more interesting and nontrivial is the shuffling of the man-

ner of articulation across both axes, which are as previously expressed is a
historical artifact. Specifically, the placement of affricates into the place of
articulation is a reflection of their late appearance in the language’s devel-
opment.

9.1.3 Extensible and non-extensible phonemes

A related concept to the idea of terminal and non-terminal phonemes are
the extensible and non-extensible phonemes, which was once used in place
of the former. As it stands this current division is unused in the language,
but retains some prestige in certain circles.
The primary difference between the two is their behaviour under gemi-
nation. Although gemination is not part of the language, consonant length-
ening is still a part of everyday speech, as it is the way in which sounds
are emphasised or made “louder”, consonants being noted to be hard to ac-
tually make loud. Extensible phonemes are those that can be maintained
indefinitely, while non-extensible ones are not. In practise this means that
non-terminal phonemes that are also not plosives or affricates are extensible,
and all of the remainder are not.
The paralinguistic situations that use this division of phonemes are sim-
ply those that require emphasis as outlined in the previous paragraph. To
make a word “louder” or more important, one preferentially makes the exten-
sible phonemes longer. It is considered impossible to lengthen non-extensible
phonemes, so they are always left alone in this process.

9.1.4 Adorned and plain phonemes

This division of phonemes is completely unrelated to Drsk in any way, but
it is extremely relevant to Drsk’s relation to Rattssaw. Specifically, some
of the letters used in writing Drsk are “adorned”, in that to properly write
them one requires two Rattssaw letters. In the Latin orthography we have
chosen for Drsk, it is normally, but not always, the case that one Latin letter
represents one Drsk letter subject to some minor conditions.
It’s immediately obvious that this genuinely has no relation to the lan-
guage, just how it is written, but because of the relative importance of
Rattssaw against Drsk, it has continued to receive much attention in certain
contexts such as poetry and lipograms.

9.1.5 Non-phonemes imported into Drsk

Finally there are also loan phones that have arrived to Drsk with the intro-
duction of other grades of civilisation to its language region. As the whole
of the Rattssaw alphabet is also included in the Drsk alphabet, invariably
certain phonemes that Rattssaw has would be imported in Drsk as well.
9.2. SQUARES 105

The most notable of these is the Rattssaw phoneme /#/,20 representing a

prevocalisation, i.e. the mismatch between the tongue moving into position
and the vocal cords beginning to vibrate. It is often heard as a very brief
[ŋ] or [ə].
This phone is not phonemic to Drsk, but it has been imported to it as
a phone. In some cases it is visibly present in Rattssaw loanwords. It is
in this case recognised as a phone, [#], which also has the same phonetic
realisation.

9.2 Squares
The “past version” of the language alluded to earlier has a much smaller
number of phonemes, but they still conform to the same grid. In fact, this
is the original form of the Drsk consonant table: (28 | 54)

Nasal Oral (NV) Oral (V) Lingual

Labial •
Dental • • • •
Palatal • • •
Velar • • •
Laryngeal •

Each bullet • here represents either one or two phonemes in the ancient
form of the language, depending on additional factors that are fairly com-
plex. Essentially what happens here is that each of these twelve single or
double phonemes – called “squares” – are part of the most primitive form
of Drsk’s phonology21 and is cemented as the “real” phonology regardless of
any changes that might have come since.
How this simple phoneme table eventually absorbed all the vowels and
became the phonetic inventory of today is a well-documented if slightly
bewildering process. The entire development would depend on adding on
how the vowel system has slowly collapsed over the many centuries, but the
short version is that each time a vowel is lost, the inventory first doubles
the number of consonants, then reshuffles and merges the remainder down
in complicated ways until it is ready for another vowel to be absorbed. This
pattern can be seen most clearly in the loss of the final two vowels, */a/
(penultimate) and */i/ (final), which respectively created the affricates and
the terminal/non-terminal split. Most of the remainder are obscured under
20
It is written /#/ because of the word boundary symbol, #, in other parts of linguistics.
21
This does not necessarily mean that it is the oldest form, i.e. Drsk S from section 1.7:
Instead it is an amalgam of various inventories from all times and places that are centred
on Drsk S and therefore represents no single time or place but rather displays the most
stereotypical features of all of them. Additionally the language at the time still has vowels,
but they obviously do not display in the current iteration of the language.
106 CHAPTER 9. PHONETIC INVENTORY

many centuries of splitting and merging,22 and so it is fairly difficult to

actually retrace the steps of collapse from the current inventory.
Regardless of their current irrelevance to the phonology, these squares
have had real impact over the language over the years, especially with regard
to both artistic concerns, such as poetry and numerical divination (numerol-
ogy); and practical ones, like collation and labelling things using letters. For
the latter, each of the twelve squares have a nominal letter which is also a
plain Rattssaw letter, and all remaining letters are considered “sub-types” of
that nominal letter.
There is no particular order specified to both the squares and the letters
within any particular square, so they belong into Type 3 in the list hierarchy
mentioned in (19 | 57), which is different from the standard alphabetical
order seen in section 13.4, which is Type 4.

9.3 Dialectal (non-)variations

While virtually impossible by Earth standards, dialect variations in phoneme
inventory in Pasaru is almost always structure-preserving. That is to say,
dialects tend to change a single phone to another phone, occasionally to an-
other one that is not in the “standard” dialect but otherwise retain the same
structure as the rest. In particular, no dialect merges or splits phonemes in
such a way that homophones form or disappear.
No particular examples are provided here, but some might be implied in
later chapters.
However, it should be noted that a lot of the fixed nature of the inventory
is “bought” (by some conversation principle) by a dizzying array of other di-
alectal variations that don’t involve altering the consonant inventory. That
is, the bulk of the dialectal variation comes from discussing vowel inser-
tion tactics, and this is the primary way that dialects tend to distinguish
themselves. More on this is discussed in Chapter 12, and Section 12.9 in
particular.

22
The total collapse of the vowel system is very similar to a similar development where
the lexical inventory of Drsk has steadily merged down in favour of a gender system of
categories and residuals which are described synchronically in section 16.4 and diachronically
in (32 | 15). These two systems are contemporaneous, and there is good reason to believe
that they derive from a common cause.
Chapter 10

Phonoruns

A lack of vowels is not the only phonetic irregularity that Drsk has. Sylla-
bles too are somewhat missing, and in their place are phonoruns, which are
a simpler structure. While phonoruns are also a combination of phonemes
into more accessible and pronounceable groups, the difference is that the
boundaries between phonoruns are determined in a fundamentally differ-
ent way than syllables. Specifically, for Drsk, structures that are termed
“explicit-terminator phonoruns” best describe their behaviour.
In this chapter we shall denote phonoruns by the symbol ρ, and sylla-
bles with σ. We shall discuss in section 10.1 the behaviour of phonoruns
using the prototypical phonorun-only language Rattssaw. Then, in section
10.2, we discuss how the way explicit-terminator phonoruns work, and how
they apply to Drsk in particular. We shall then explore the implications of
phonoruns and prosody in section 10.4, and finish with a reconciliation of
Drsk phonoruns and Drsk syllables in section 10.6.
It should be noted that while we speak of “Drsk syllables”, the phonorun
has always been the more important and fundamental part of Drsk phonol-
ogy. Syllables, while playing an important role in some contexts, are highly
dependent on vowel insertion (for which see chapter 12) and are therefore
dialect-dependent; on the other hand, the signature (which is to a phonorun
as a syllabification is to a syllable) of a word is relatively more stable cross-
dialectally and is therefore the one that gets used in contexts like lexicogra-
phy and education.
Where it does not introduce ambiguity, the word “phonorun” may be
abbreviated to the word “run”.

10.1 Phonoruns in Rattssaw

Phonoruns first appeared in Rattssaw, before being spread as an areal fea-
ture to Drsk. As such it is productive to briefly touch upon its phonorun sys-

107
108 CHAPTER 10. PHONORUNS

tem, and to give ourselves a little bit of an understanding of how phonoruns

work in general.
In Rattssaw, every phoneme carries an enumeration feature, [Open] and
[Closed]. All vowels are open, all plosives are closed, and all the other
phonemes are one or the other, though sometimes this can depend on the
previous phoneme. Since this is a Drsk grammar, not a Rattssaw grammar,
we will work with Rattssaw orthography only, which is to a near-perfect
approximation Rattssaw phonetics.
So for our purposes, the phonemes in Rattssaw are as in table 10.1, along
with their open and closed status. (14 | 54)

Table 10.1: Table of phonemes in Rattssaw by their open/closed status.

English Symbol Rattssaw List

Open o Þhaər a e i o u á é ó ay wá yé h
Closed c Skkdþ p b t d pp bb tt dd þ f ff k g kk j ss #
Inheritable i Fflayer l r rr þþ z s f x ś m n ng

Now, the important part about phonoruns in Rattssaw is that they al-
ways consist of one or more of exactly one type of phoneme. Symbolically
we say that a phonorun ρ consists of
(
o + (o|i)∗
ρ= (10.1)
c + (c|i)∗
where + means string concatenation and | means alternation.
For instance, these Rattssaw words are “syllabified” as such:

1. Rattssá
ω

ρ ρ ρ ρ

i→c o c c o

R a tt ss á

Here we have a fairly simple word with four runs. As is demonstrated

here runs can be as little as one phoneme long. Furthermore, an i
becomes a c if it appears word initially as the initial phonemes is
always c unless indicated otherwise, in anticipation of the honorific
prefix #.
In some dialects, the initial r becomes an i → o instead, which means
that the word now has three runs instead of four. Here the reason
10.1. PHONORUNS IN RATTSSAW 109

is that the goal is to reduce the number of runs and therefore the
“inheriting” part of the inheritable phoneme can also inherit from the
right if the phoneme on the left is unavailable.

2. Mssdn
ω

c c c c

m ss d n

This is in some way the opposite of (1) – instead of having lots of

short runs, we have one very long run. The prosody here is somewhat
different as a result.

3. Eahmsrkdoryé
ω

ρ ρ ρ

o o o i c i c c o i o

e a h m s r k d o r yé

This is a complicated word with multiple runs that contain multiple

phonemes.

4. Yéadeams
ω

ρ ρ ρ ρ

o o c o o i c

yé a d e a m s

And to round it off, here is a word that has both complex and simple
runs. This should take more or less the same time as in (3), in spite
of some of the objects being simpler.

In general though, apart from a difference in prosody and the simpler

structure, phonoruns are more or less the same as syllables.
110 CHAPTER 10. PHONORUNS

The important part to take away here is that a phonorun is controlled

by largely regular means and therefore can be predicted in its entirety even
programmatically. The trade off is that sometimes they can get very long,
an that is a fact that remains true as we transition to discussing Drsk
phonoruns.

10.2 Terminators and Non-Terminators

The major difference then between Rattssaw phonoruns and Drsk phonoruns
is that instead of the boundary being an implicit gap caused by a switch
between two types of phoneme, it is determined by a very explicit property
of particular phonemes.
In other words, certain phonemes carry the property [+Terminator],
which signals that a phonorun that precedes it ends with it, and a new
one begins after it. (23 | 4) Hence, one can describe a phonorun using the
following formula:

ρ = n∗ + t (10.2)
where n is a non-Terminal, t is a terminal, and ∗ is a Kleene star.
And then, a phonological word ω is defined using the auxiliary object ω 0
as

ω 0 = ρ∗ + ρ (10.3)
∗ ∗ ∗
= (n + t) + n + t (10.4)

in which case ω is defined to be any continuous substring of ω 0 which

contains at least one t and ends in a t. Note that sometimes a word can
end with an n, in which case an epenthetic terminator is inserted. This
terminator can be implicit and is generally assimilated to the preceding
consonant, or [k] otherwise.
For example, in the word tmrlħð “Rudder-TOP”, the final morpheme is
an n, as seen in the analysis in item 3. The entire word is pronounced /[Link]-
lħð/. In order to have the last phoneme be a complete run, the epenthetic
phoneme must be inserted, so for instance [tmrlħðk], or to apply an assim-
ilation rule [tmrlħðd]. Other considerations may limit assimilation options,
such as the fact that adding [t] would make things needlessly ambiguous
because it would be the same as one of the suffices t /t/ [t] “TRANS” (see
section 16.6). Furthermore, since we have not accounted for vowel insertion
(see section 12), the final narrow transcription would be more complex than
this.
As with Rattssaw phonoruns, Drsk phonoruns are also entirely regular
and can be predicted programmatically, but this time run delimiters are
10.2. TERMINATORS AND NON-TERMINATORS 111

explicitly marked out using specific phonemes. There are two main analyses
that explain this particular phenomenon:
1. The single-class hypothesis claims that there has only ever been
one class of phoneme, as most other languages that use phonoruns
typically require that consonants be able to cluster together to a single
class; and in the case of Drsk there are only consonants so naturally
there’s only one “extant” class. (Older versions of Drsk indeed still
have vowels.) In this case there are pressures arising from overlong
phonoruns that is then resolved by creating a terminator structure,
explicitly terminating a run at set point.
2. The restricted second-class hypothesis claims that the terminal
phonemes have always been part of a separate class – it isn’t unheard
of that consonants are broken up in phonorun classes – and that the
length of the run in this particular class is then restricted to 1.
It can be argued that the truth is somewhere in between these options,
with a large amount of phoneme classes either melting away or combining
into one, all the way up until the last few phonemes remain in a hold-out
class, which promptly was restricted down to runs of 1 long. This situation
is then reänalysed into the situation we have today.
A quick historical note here is that the T/NT distinction is fairly recent;
it was formed when the last of the vowels were removed from Drsk (see sec-
tion 9.2). While this seems like evidence against the single class, this is not
necessarily the case; in particular the T/NT distinction was formed quickly
after the single remaining class (of vowels) disappeared and the restrictions
explained under the restricted second-class hypothesis could be simply a re-
flection of the fact that the vowel remnants normally don’t appear next to
each other.
Regardless, here are some examples of how “syllabification” works in
Drsk, using the orthography mentioned in section 13.1. The t phoneme
that is required for a word to be valid is coloured as such. Phonologically
speaking, there is no particular justification as to which of the terminals can
be the requirement, only that there is one, though overall the choice of what
counts as the requirement is the first terminal in the root word.
1. ksbftʲ “Seasonal pond”
ω

ρ ρ ρ

t n t n t

k s b f tʲ
112 CHAPTER 10. PHONORUNS

A relatively straightforward word consisting of three runs with a mod-

est amount of phonemes each.
2. dhkⁱmdᵘčssrtʲ “*Wool1 ”
ω

ρ ρ ρ

t n n t n n n n n t

d h kⁱ m dᵘ č s s r tʲ
A slightly more complex word with long runs. In principle, there is no
upper bound to how long a run can be, though of course practically
they are bounded by how much one can remember at any given time.
That is to say, in Drsk there is no “maximal phonorun”.
3. tmrlħð “Rudder-TOP”
ω

ρ ρ ρ

t t n n n n t

t m r l ħ ð [k]

Ω
This word is special in that the phonological word ω is not the same
as the lexical word Ω, which means that the final [k] that was added
on was merely a phonotactical consideration – it does not form part
of the lexical word at all. However, both can be putative “roots” of
the word, so they are drawn as trees with shared leaves here. More
examples of this kind of detachment can be seen in section 10.3.
The examples above show a great difference between phonoruns and
syllables; not only are phonoruns allowed to go on arbitrarily long, they
also have relatively little internal structure (as previously mentioned), which
implies that most if not every possible combination of phonemes up to the
definition of the phonorun can be one. Other languages that use phonoruns
don’t have this unusual leniency, though in effect their restrictions tend to
be adjacency limits (including adjacencies with the boundaries) as that is
all that phonoruns can afford.
If a terminator starts a word, it is called the “incipator”. Phonologically
this does not change the behaviour of the phoneme in isolation, and it re-
tains virtually all of the qualities that a terminator would have. However,
10.3. PHONOLOGICAL AND LEXICAL WORDS 113

incipators can sometimes become part of a “false word” (a phonorun that

does not belong to any particular word) if it is preceded by a word that does
not end in a terminator. All words shown here are incipators, and in fact a
large amount of words have one, again due to morphological considerations
on nouns.

10.3 Phonological and lexical words

As seen in example (3) in section 10.2, phonological words ω may not be the
same as lexical words Ω, though the two are closely related.
As so it happens most ordinary nouns are guaranteed to have Ω = ω, as
due to the suffix mechanism (see section 16.6) they all in their dictionary
form have the suffix tʲ which is class t and therefore closes the run right
where it is most desirable, i.e. at the end of Ω. While its most common
alternate forms replace the tʲ with other phonemes, the replacements are
also frequently closed when it terminates a word (e.g. t, p or tᵗ), which
allows us to confidently have Ω = ω. When this doesn’t happen, we have
several ways to resolve the problem. This behaviour is fairly complex and
interacts strongly with runs and syllables.
When Ω 6= ω, the situation is considered unstable and some dialects
behave strangely. In particular, the phonemes that “stick out” of a lexical
word can be swept up by a terminator in the next word, causing a ρ to stick
out to the next Ω, seemingly ignoring any phoneme that has been inserted
stop up the phonorun at the end of Ω). This can result in a case where
the phonoruns as they are pronounced not lining up with word boundaries.
A concrete example of this is given in section 10.6, where we will also see
common ways to resolve this instability.
The case where more than two ω fit inside one Ω is somewhat rare but
is not unheard of even in ordinary speech. This happens when an Ω is long,
which can be somewhat annoying to read as a single phonological word; in
this case the lexical word is subdivided into more manageable chunks. This
mainly happens with nouns, which conveniently does have segmentable parts
– gender–root-sequence–suffix – so in this case it is used as a division.
The easiest way to demonstrate this is with numbers, as thus:

10.4 Prosody
Generally speaking, prosody using phonoruns are handled by the time it
takes to go through one of them, and all of them in a single word. This
timing can be accurately predicted using a simple equation, which is known
as the prosody equation.
First, let’s see the equation itself, which is Equation 10.5.
114 CHAPTER 10. PHONORUNS

T = KnL+νn (10.5)
Here, K, L and ν are language- and dialect-specific constants, T is the
total amount of time required to pronounce the phonorun, and n is the
number of phonemes that are in that run. K in particular controls the
overall speed of the speech, and since that is in turn controlled by various
individual effects (as well as the choice of units used to measure T ) this
turns out to be basically impossible to predict. On the other hand, L and ν
are much more predictable and stable from region to region, and from year
to year.
L is called the “starting constant”. It says that as a run goes on for
longer, the successive consonants take longer and longer to come out. There
is a small correction factor called ν, which is called the “corrective constant”.
This further penalises large runs.
Generally speaking, exponentials are bad as they quickly grow out of
control, so for this reason we often have 1 < L ≤ 54 and 0 ≤ ν ≤ 40 1
, and
similarly it gives a soft cap as to the maximum number of phonemes in a
run, namely, n . 6.
For example, a specific dialect of Rattssaw has the following quantities
for use in the Prosody Equation:
5 1
L = ;ν = (10.6)
4 48
For all Drsk dialects, the distribution of L and ν can be summarised as
a roughly normal distribution with the following parameters:

L = 1.15 ± 0.04 (10.7)

ν = 0.0139 ± 0.0019 (10.8)

Geographically speaking, this distribution is largely continuous, with

neighbouring regions having roughly equal values of L and ν. It is usually
higher toward the south, and lower toward the north and near the border
with the J.-Senlis.
Other qualities, such as stress, tone, length and hiatus, work basically as
in syllables. The only major trap to watch out for is that sometimes it may
not be possible to apply such qualities to certain runs, due to them lacking
key features such as vowels or voiced phonemes, as some runs lack the former
and some lack both. Since none of these are phonemic whatsoever in Drsk,
however, and the idea of vowel-free runs make them somewhat troublesome
to deal with, they are not usually even considered amongst speakers to even
exist; however with vowel insertion these features become a lot more feasible,
and they do exist in Drsk, without too much controversy, at least when stress
is involved.
10.5. STRESS AND INTONATION 115

Other run-languages may lack stress altogether though, perhaps due to

restrictions put to their phonology (lacking a vowel insertion rule where
necessary, for instance), and in those cases often other phonetic features
may be used as a substitute to the otherwise-universal stress patterns, such
as length or in some cases tone.
The final thing to note about this system is that it means the entire
language is phoneme-timed, which is surprisingly low-level, with a subordi-
nate timing related to phonorun timing. In Drsk, the Prosody Equation, if
pushed beyond n ∼ 6, L is capped to some fixed amount, which we can set
to 1, which in turn determines the value of K for each dialect.

10.5 Stress and Intonation

As previously mentioned, the ideas of standard phonology still carry over to
phonoruns mostly unscathed, though of course there are some changes that
have to be made to accommodate the fact that runs generally have fewer
guarantees that new features can stack on top of.
With that in mind, we can say that Drsk does indeed have stress on the
run level, and not only that, it would make more sense to analyse such stress
on the run level (plus any associated vowel insertion rules) than it would
be to analyse them on the syllable level (i.e. after the associated vowels are
inserted). It is true that some expressions of stress and intonation require
the vowels to be inserted afterward, but this is considered to be a restraint
on vowel insertion, rather than taking it as a syllable-level feature. However,
it is not phonemic; there is no minimal pair that distinguishes two words
only from stress. It is in fact more accurate to call it a type of focus marking,
as it is generally used to break up monotony in an utterance.
Though historically derived from stress, the current form of the language
does not actually use stress to indicate stress, but instead uses pre-hiatus,
i.e. a small pause inserted before the next run is spoken.

10.5.1 Intra-word stress

In general, words in Drsk has the stress in three locations, which may or
may not be in the same place:

• The incipator carries stress 1, which is the strongest of them all.

– Where an incipator does not exist, the first run of a word carries
stress 3.

• The third run on the end, where it exists, carries stress 3. In some
dialects this is not included and is instead just another ordinary run.
116 CHAPTER 10. PHONORUNS

• The final run in a word carries stress 2. This does not apply to nouns
that have a suffix.

(Stress 0 can be considered unstressed in all cases.)

Stresses 1, 2 and 3 are similar to primary, secondary and tertiary stress in
several respects, but their behaviour is also slightly different. In particular,
there is some complex defaulting behaviour going on in smaller words where
there isn’t enough room for so many stress levels. In the end they still boil
down phonetically to two levels of stress: stressed and unstressed.
In this section, this colour represents Stress 1, this colour represents
Stress 2 and this colour represents Stress 3. The stress number is also
indicated beneath the relevant runs for those who cannot see colours.

(6) nᵘfnsyzŋr “he, an elusive, unknown animal”

nᵘ fnsy zŋr
1 2
/ˈn/ /fnsʡ/ /zŋɹ/
A fairly basic word with three runs. Here the application of the rules
is fairly straightforward. As expected, the run with stress 1, i.e. /n/,
gets the stress.

(7) ðfkᵏm “cohabit”

ðf kᵏ m
3 2
/ðf/ /kʰ/ /ˈm/
Another simple word. However, this one lacks an incipator, so no run
gets stress 1. In this case, the stress falls on the one with the next
number, so here it is on the /m/.

(8) gvᵉlmᵘfyďčtᵗdᵈrⁱtʲ “a proclamatory song about salt”

g vᵉlmᵘ fy ď čtᵗ dᵈ rⁱ tʲ
1 3
͡ ͡
/ˈg/ /vnlm̰/ /fʡ/ /ð̆/ /tɹ˔̠̊ tʰ/ /dʰ/ /ɾ/ /tʲ/
This very long word has an incipator, so the stress falls on it. The tʲ
/tʲ/ is a suffix, so it does not receive any stress at all.
The word is long enough that one might be tempted to bring in the
/dʰ/ as a “secondary stress” of some kind. This is sometimes done in
the southern areas, but is not considered standard.
10.5. STRESS AND INTONATION 117

Once the run with the stress has been identified, we label it as the “key
run”, which will, if selected, finally receive the stress as required. This means
a “key run” does not necessarily receive stress, but instead it must satisfy
another requirement to do so.

10.5.2 Inter-word stress

Meanwhile, beyond the word level, there are additional stress levels, which
like the word-level stress levels are numbered, but this time using numbers
4 and 5, just to make things clear which is which. Again it all boils down
to which syllables get stress and which doesn’t.
Generally speaking, there are two key words inside each sentence, which
give a hint as to what the topic is about. Changing the stress of a sentence
generally doesn’t change the meaning, but can have some pragmatic effects
and provides contextual clues.
Inter-word stress is more “powerful” than intra-word stress, in that a
word that has inter-word stress will take over stress from other words that
only have intra-word stress. Schematically, consider the following set of runs,
that have the following run structure.

(ρρρ)(ρρ)(ρρρρρρ)(ρρρρρ) (10.9)

Here, individual words ω are wrapped in brackets, (), whereas any stress
levels are marked as superscript numbers, e.g. ρn . Let’s assume the following
stress pattern is determined:

(ρ3 ρR2 )(R1 ρ)(R1 ρρρρρ2 )(R1 ρρ3 ρρ2 ) (10.10)

This would naturally cause the runs that have become capital Rs to be
stressed, given the rules discussed earlier.
Now let’s consider the introduction of inter-word stress on these four
words.

(ρ3 ρR2 )(R1 ρ)4 (R1 ρρρρρ2 )(R1 ρρ3 ρρ2 )5 (10.11)

This causes the stress pattern to reädjust as follows:

(ρ3 ρρ2 )(R1 ρ)4 (ρ1 ρρρρρ2 )(R1 ρρ3 ρρ2 )5 (10.12)

Notice how in the third word the ρ1 has been “robbed” of its stress,
because the previous word has inter-word stress, and so all the words outside
must be de-stressed in order to emphasise that. Similar arguments apply
to the ρ2 in the first word, but in some dialects the ρ3 in that word would
become stressed in its stead.
118 CHAPTER 10. PHONORUNS

10.6 Syllabic non-denial and reconciliation

After vowel insertion, the vowels form nuclei, and from there we have recre-
ated syllables, σ. The ability of Drsk to somehow retain the syllable even
after phonoruns have supplanted it as the main way to divide up a word
is remarkable, although circumstances has never conspired again to see if
another language would do the same under similar conditions. This makes
Drsk a non-denial language – it does not deny that syllables exist in itself
(which is definitely the case in Rattssaw, where there’s no sign of a syllable
anywhere.) As previously mentioned however, where and what vowels are
inserted is highly dialectal and therefore cannot be made consistent through-
out the language.
Generally speaking, a Drsk syllable prefers to be in the form described
by section 12.8. While they can also be considered units of prosody, it
is not generally done for one simple reason: their relationship with both
types of words can get very counterintuitive and unusual. Compare this
to phonoruns, which has a much nicer behaviour when it comes to word
boundaries.
All ω have an obvious parent Ω such that no ω can simultaneously belong
to two Ω. That is to say, we can construct a forest where all the roots are
objects in Ω and all the leaves are objects in ω. Graphically, it would look
like this:
ω:
Ω:
where each line is an instance of an Ω or an ω depending on which
row it is on, and their boundaries are indicated. Notice how while every
Ω boundary coïncides with an ω boundary, there can be times where an ω
slips out, due to the requirements of an ω needing to terminate at a t, not
seen in this level of abstraction.
As for the most part we have Ω = ω, most of the lines in the two rows
line up exactly. Since they’re mostly plain nouns, we can therefore say that
most nouns have a phonetic signature that allows them to be detected even
without much external context. The caveat is that this is not a fact that is
transparent to most speakers and the education system does not particularly
seize upon this point.
In the cases where Ω 6= ω, things get a little bit more involved, so let’s
provide an example phrase. (23 | 46)

(9) drsk mŤSRGÐ plmᵘð

/dɹsk mθ̆sɹgð plm̰ð/
[dɹɨsk mɨθ̆sɨɹgðɨb plɨm̰ðt]

Note that we have already inserted vowels (here, the constant [ɨ]) in the
narrow transcription, and opted for a certain assimilation rule to round off
10.6. SYLLABIC NON-DENIAL AND RECONCILIATION 119

d ɹ ɨ s k m ɨ θ̆ s ɨ ɹ g ð ɨ b p l ɨ m̰ ð t

ρ:
σ:
Ω:
ω:

Figure 10.1: The divisions of the phrase drsk mŤSRGÐ plmᵘð in runs (ρ),
syllables (σ), lexical (Ω) and phonological (ω) words. On the phonoruns row,
the terminators are indicated using dots. Grey lines connect each phoneme
to the rows to ease reading.

the phonoruns (here marked as such). Let’s draw in the diagram indicating
where all the syllables σ, runs ρ, phonological words ω and lexical words Ω
are. This diagram is in figure 10.1.
We can see that in this phrase, the requirement that no ω can be part of
two or more Ω holds, and in fact there’s usually a one-to-one correspondence
between them. This is the usual case, and having a many-to-one relationship
is exceedingly rare.
Notice that for the most part, Drsk syllables will connect any isolated
Terminal phonemes to the next run, stopping at somewhat unpredictable
locations (see, for instance, how Drsk ends up being one syllable, with the
incipating d being pushed in to the rest of the word which is one run long.)
Indeed, most foreign speakers unaccustomed to phonoruns would use that
many syllables to pronounce it.
An interesting effect comes when considering how the [p] in plmᵘð in-
teracts with the epenthetic consonant [b]. In dialects that utilise constant
epenthesis to stop up phonoruns (using [k], for instance), it and the [p] would
be combined to create a coärticulated consonant, [k͡p], or sometimes an un-
released [k̚p]. In any case, the run then ends in between words. This can
result in a somewhat confusing situation where a ρ can span more than one
ω and Ω, but there is no problem since ρs are determined on a phrase level,
whereas the rule that they must stick within one ω or Ω is only determined
on a single-Ω level.
However, when it comes to syllables, the system breaks down somewhat.
Notice the awkward placement of the syllable boundary; it is because there’s
no consensus as to which place the boundary should be, especially if the
dialect does not assimilate the epenthetic consonant. This means that it is
possible that the syllable boundary can stick in-between the two consonants,
causing an awkward situation where there is a time interval where both
syllables are in effect, which isn’t particularly favoured. Syllables are in fact
120 CHAPTER 10. PHONORUNS

only tacked on to Drsk analysis in some way, and they don’t have anything
hanging off of them.
As a result, syllable analysis is not particular common with Drsk, and
phonorun analysis is favoured instead.
Chapter 11

Phonotactics

Despite23 the amount of flexibility provided by vowel insertion, sometimes

due to other factors consonants can still not manage to cluster in a single set
of values, everywhere. (23 | 6) Vowel insertion is however still a huge part
of Drsk phonology however, and are covered in the next chapter, Chapter
12. This chapter will cover the remainder of what counts as phonotactics in
Drsk, which isn’t very much.
One factor of the limited amount of phonotactics is the correspondingly
limited depth of structure of phonoruns. The whole thing about phonoruns
after all is that each individual phoneme is more or less egalitarian except
for the terminator; all of the other phonemes are basically interchangeable
in role with each other, because there really only is one role. For instance,
there really isn’t any room in phonoruns to forbid a particular phoneme to
not be in front of a particular run, because otherwise it would be categorised
as a separate type of run.
Because of the high level of freedom that this allows the phonemes, there
aren’t that many things that one can discuss other than how phonemes
are restricted from put together, and how they might affect each other if
they are placed together. Essentially, what happens is that the extreme
freedom makes any discussion about the values mostly pointless – there’s no
linguistics in it.
Nevertheless, the relationship between nearby characters are still con-
strained, and therefore we can discuss them meaningfully. As such, this
chapter will mostly focus on adjacency rules, what happens when conso-
nants come too close to each other; middle phonemes, consonants that
are at once terminal and nonterminal; and how the whole all of this interacts
with the word and sentence boundaries.

23
It’s not really “despite”, but in principle one can possibly imagine that the vowels manage
the consonants in such a way that they somehow stabilise them.

121
122 CHAPTER 11. PHONOTACTICS

11.1 Adjacency rules for phonemes

As previously mentioned the primary mechanism for limiting or interacting
phonemes is by way of adjacency rules. When considering phonemes, the
key item to consider is that they’re mostly atomic, with no structure in
between them, so they are mostly inflexible as to how they give and take.
They still do have some give with regards to how they are expressed, which
include a direct change in their values.

11.1.1 Two of a kind

The most common causes for phonotactical considerations is when two iden-
tical phonemes manages to be placed next to each other. For example, /tt/,
/ss/ /ll/. These exist and they are the main reason why the romanisation
uses superscript letters to distinguish digraphs from those two phonemes
being placed next to each other, so for instance /ss/ is ss and /ʃ/ is sˢ.
Nevertheless, a phoneme sequence like /ss/ is in some dialects quite un-
savoury and therefore they are pronounced differently in practise. The exact
change depends on the dialect, and the tactics are surprisingly diverse; and
furthermore, the exact tactic is different depending on the exact phoneme.
In places where vowel insertion is common, such as in the southeast, the
vowel insertion rules may be suspended to specifically handle the case of the
two-of-a-kind. For instance, in the ktᵘc Language Region, pairs of identical
sibilants such as /ss/ and /ʃʃ/ usually have a vowel inserted between them,
in this case a specific vowel as in [sʌs] and [ʃʌʃ];24 in dialects closer to nᵘfk
the key is to alter the articulation slightly, which amounts here to including
additional outflow into the second consonant, which we might notate as
[sꜜs],25 or as a brief pause, so [s(.)s].
Sibilants are special in that they have a more or less consistent treatment
throughout dialects. Other phonemes are much less clear-cut. Terminating
plosives, for instance, tend to get split apart into separate phonoruns, for
instance:

1. /tt/ = [tθ]

2. /dd/ = [dð]

But sometimes, they just assimilate or combine into each other:

1. /ss/ = [ʒ] 26

2. /cc/ = [k]
24
The appropriate vowel for /s/ is usually [ɨ].
25
Not an entirely accurate notation, but “increased outflow” has no good symbol in the
IPA so it’ll have to do.
26
For dialects that do not insert vowels between two sibilants.
11.1. ADJACENCY RULES FOR PHONEMES 123

3. /kk/ = [k]

4. /ll/ = [lː]
This one in particular is important as it shows that sometimes the
two-of-a-kind uses some distinguishing feature.

11.1.2 Three-of-a-kind and higher

Should a triple phoneme appear such as /zzz/ or /ddd/, as would normally
happen in genders (see, for instance, the zzz gender seen in punctuation in
section 13.6), several other rules are in place.
For some of the dialects, these are treated as two-of-a-kind with an ex-
tra copy hanging to the right, and therefore there’s nothing new going on.
However, there are some language regions that do handle these differently,
which are common in Woróli and other central language regions.
It is extremely rare to see more than three phonemes adjacent to each
other, but it could happen, such as for instance, when a three-of-a-kind gen-
der meets up with a root that starts with that same letter, e.g. fff.ffdsn.tʲ,
which names a certain type of complicated dish. This word is pronounced
/[Link].tʲ/, which has five /f/ in a row, one of the longest sequences in
common speech.

11.1.3 Double articulation

Things are even more complicated for two phonemes that aren’t the same.
Now there is an additional concern where the resulting articulation can ex-
press a consonant that’s completely different from the two that spawned
them, on top of having additional rules that may prevent them from inter-
acting altogether.
As a general principle, two-of-a-kind or higher and other similar special
rules such as preferred clusters take precedence over this general double-
articulation rule.
One of these pairs is special in that it appears particularly often: with
numbers, which are further detailed in section 17.1, there is a common pair of
phonemes that are placed together, which are /tʰ/ and one of /tʲ t θ̆/. This
pair (actually three pairs) happen often enough that there’s a resolution
that’s common to all dialects: the /tʰ/ is removed. So common is this
adjacency rule that it is frequently indicated in orthography, though it is
possible to explain this disappearance without appealing to phonetics.
In the case of numbers, this is reflected in the orthography, but in other
cases this is optional and in fact not commonly done because it does strange
things to the transparent orthography.
124 CHAPTER 11. PHONOTACTICS

11.2 Adjacency rules for phonoruns

Phonoruns too have adjacency rules, but they are always subordinate to
phoneme adjacency rules. That is to say, all phoneme adjacency rules are
applied first, and then phonorun adjacency rules are considered with the
inserted or deleted phonemes under the phoneme adjacency rules now being
considered phonemic in this case. (34 | 22).
Here are some common cross-dialectal adjacency rules.

11.2.1 Lone or Double Terminators

A phonorun is allowed to be a single phoneme long, that is, just the T. This
means that it is possible for two terminating phonemes to come together,
which might be unexpected. If terminators are bunched next to each other
then some dialects may decide to add [s] between them to break them up.
Examples:

(10) a. zntʲ
b. rtkb /ɾ.t.k.b/ [rstksb] (plus vowel insertion)

11.2.2 Incipators
Some language regions distinguish a terminal phoneme that starts a word. In
this case, it is called an incipator, and normally it receives special treatment.
For instance, take the following word, which has an incipator:

(11) tnᵘrd /t.n̰ɹd/

For regions that treat incipators differently, the following rules broadly
apply:

1. The incipator rejects all vowels that could have appeared in front of
it, and frequently immediately behind it as well.

2. Incipators are always pronounced as their nominal value.

3. Incipators are treated as part of the phonorun that follows them.

From a diachronic point of view, incipators are the final vestiges of the
old, syllabic form of the language, being the first syllable’s onset if it turns
out to evolve into a terminal phoneme. This is the explanation for several
of these properties above.
A word where there is only one phoneme which is a terminator does not
count as an incipator. This means that that incipators cannot appear next
to each other. There aren’t that many words that are one phoneme long,
11.3. THE CASE OF THE MIDDLE PHONEMES 125

and there are virtually no situations where two such words are placed next
to each other, which implies that for the most part the word with the single
phoneme is treated as a lone phonorun with nothing special to it.

11.2.3 Phonetic waves

A phonetic wave is a particular feature of a phonorun-based language where
the phonemes of two distinct classes alternate, with the same number of
phonemes per run.
In Drsk, this feature is very marginal, as there are only two classes and
one of them is restricted to appearing only once per run. Nevertheless, this is
enough to have a sequence of, for instance, “NT, T, NT, T” to be considered
a phonetic wave, and so many runs of length 2 placed next to each other are
“phonetic waves” in Drsk.
For phonetic waves, the run-timed nature of the language becomes al-
tered slightly. Each individual run is now preceded by a hiatus, and the
non-terminal phoneme is now pronounced slightly shorter. Additionally,
inserted vowels are liable to be dropped again at this stage.

11.3 The case of the middle phonemes

On the phoneme chart, there are some phonemes that are not T or NT
and are instead placed a cell that spans both of these columns. These
phonemes are therefore called “middle phonemes”, and their behaviour is, as
their position indicates, somewhere in the middle between being a terminal
and a non-terminal phoneme.
Middle phonemes are not a third class of phoneme however; when spo-
ken they are always either T or NT. What a particular middle phoneme
becomes is done on a case-by-case and dialect-by-dialect situation, which is
all governed by a general overarching principle: minimise the total number
of phonoruns while ensuring that the longest phonoruns are not too long.
For instance, take the phoneme /t͡θ/, which is a middle phoneme.

11.4 Word and sentence boundaries

11.5 Order of application of phonotactic rules

126 CHAPTER 11. PHONOTACTICS
Chapter 12

Vowel insertion

A large part of Drsk phonology revolves around the insertion of vowels in the
large thicket of consonants that the phonemes create by necessity. There are
two important parts to vowel insertion in Drsk: what to insert, and where.

12.1 Quantification of vowel insertion

Vowel insertion is quantifiable and this is frequently done to identify and
range dialects. There are two major quantities that are used to that end:
the number of vowels that can be inserted, and how often they are inserted.

12.1.1 Inserted vowel count

The number of vowels that are inserted is typically called Nv , and the set of
vowels chosen is NvS , with the constraints

Nv = |NvS | (12.1)
1 ≤Nv ≤ 9 (12.2)
Nv ∈ Z (12.3)

Furthermore, the elements in NvS are themselves an ordered pair of the

form (v, C), where v is the vowel that can be inserted and C is the set of
consonants that can follow that vowel. Normally, if context does not require
C in any form, we can omit and just write v instead.
It should be noted that the number of vowel quantities that are inserted,
Nv , is an indicator of dialect usages, but for traditional reasons it is not
used to do so. We can do it anyway, and the resulting chart would look very
similar to the vowel insertion density map seen in Figure 12.4, just with
less detail as the amount of values are limited. However, averaging the Nv
for any individual dialect over a particular region can restore the fidelity to
some extent.

127
128 CHAPTER 12. VOWEL INSERTION

Although Nv may obtain a value as high as 9, it is usually much lower,

with peaks at 3 and especially 1. Nv = 9 is theoretical and not seen in any
one dialect, but contains every possible distinction that is made by every
common dialect. Those that have a smaller Nv will undergo mergers, and
their nominal vowel will also change slightly.
Any particular Nv has only a handful of associated NvS , which is far less
than the expected number would be if every combination is able to appear.
As such, some of the vowel lists have their own labels. For instance, the
unique set where Nv = 1 is usually labelled 1Sv and is the set {[ɨ]}. There
are only two sets where Nv = 2, and those are labelled 2Sv 1 = {[ɨ], [ə]} and
2Sv 2 = {[ɨ], [ɯ]}. The labels 1 and 2 don’t necessarily mean anything, but if
there are few sets of that size they are usually in order of most speakers.
Nv is highly correlated with ρv (see section 12.9). Briefly, the number
of vowels that can be inserted is roughly correlated with how often they are
inserted. This is somewhat expected as more vowels usually demand more
diversity.

12.1.2 Vowel density

Vowel density, more accurately named the vowel insertion density, is, roughly
speaking, how dense vowels are in a particular insertion strategy. We can
define the vowel density as such:

nv (S)
ρv = lim (12.4)
l(S)→∞ nc (S)
nv
≈ (12.5)
nc
In words, we take some passage or block of text S, and count the number
of vowels nv and consonants nc in S. We then let the length of S, l(S), grow
arbitrarily large. If we dispose of most of the mathematical bookkeeping,
we can simply say that, in general, it is the amount of consonants divided
by the number of vowels. Vowel clusters are prohibited, so ρv < 1 for all
dialects, and this is also why we chose to use this definition and not the
perhaps more natural definition of density

nv (S)
ρ0v = lim (12.6)
n
l(S)→∞ c (S) + nv (S)
nv
≈ (12.7)
nc + nv
We can then measure this quantity throughout the entire Drsk-speaking
area, subject to the same smoothing procedure as in section 12.1.3 and come
to an isodense map. This map has several nice properties: it is everywhere
12.2. VOWEL INSERTION MACHINE 129

smooth, has no sharp rises or drops anywhere, and resembles a height map.
Note that as with all population-based maps, things get very uncertain in
depopulated areas. The map is shown in Figure 12.4.

12.1.3 Smoothing
These quantities are very well-defined for an individual speaker – at least as
well-defined as any quantity could be in linguistics – but generalising them
to dialects present a small difficulty. This is done by the following method.
Each idiolect’s Nv is assessed, and then the speaker of that idiolect has
his location tracked and identified in a small region. With enough idiolects,
one can then build up an average Nv over a small (say 10 km × 10 km)
region. This is fine enough for most dialect studies, and is therefore what is
displayed in contour graphs such as Figure 12.4.

12.2 Vowel insertion machine

The idea of vowel insertion can be modelled as an elaborate machine which is
composed of three smaller machines each of which has two common functions
and structure and one of them having just a little extra. They are wired
in a particular way that expresses how the language, and more specifically
each of its speakers, approach vowel insertion.
Conceptually the goal of the combined machine is to take in a series
of consonants which already have consonant-based phonotactical concerns
applied to it, and provide the final spoken sequence, which is studded with
vowels – which is also the output’s official name, “vowel studded sequence”.
The general structure of the (abstract) machine is depicted as in figure
12.1 with each of the smaller machines being depicted as in figure 12.2.
(34 | 45) These figures summarise the general journey words go through in
order to turn from abstract phonemes to spoken sequences, which we will
go through in text here.
All words have to go through at least one of the small components,
meaning that most words will have some vowels inserted into it and all
words will at least be considered for insertion (only a small number of words,
normally very small words with two phonemes, can escape with zero vowels
inserted, and then only in certain dialects.) They then go through the
machine, perhaps going through a bypass in some cases, and eventually
land as a vowel-studded sequence at the end.
The typical route a word follows the machines in this order:

1. The dialect (i.e. language region) gets first priority with the word.
Normally, this is also the machine that makes the largest changes to
the word and the machine that can actively bypass all other machines.
However, this is also the only machine that can be bypassed by other
130 CHAPTER 12. VOWEL INSERTION

Phonemes

Parochial
Dialectal rules override

Individual
Parochial rules override

Irregular Parochial
words bypass Individual rules
bypass (taboo)

Vowel-studded sequence

Figure 12.1: Overview of the vowel insertion machine.

In

PO
IO

Irr IL IL
IL

VD
VD VD

DB PO IO PB Out
(a) Dialect (b) Parochial (c) Individual

Figure 12.2: Detail of the machines inside the vowel insertion machine.
Abbreviations are explained in text.
12.2. VOWEL INSERTION MACHINE 131

machines, which is a key part of keeping the language’s vowel inser-

tion strategy from being too homogeneous. Such bypasses are called
“overrides”.

2. The next machine is (conceptually) owned by a small group which

the speaker belongs to, typically called in this context the “clique”
or the “small circle”, with the adjective “parochial”. This represents
the social group influence on vowel insertion and dialects and is fairly
hard to delineate as it’s different for every group except in the general
structure. It’s also got a bypass mechanism for the next machine down
but it is not frequently used due to bad social implications.

3. The final machine is owned by the speaker himself, and as the last
hurdle for a particular word cannot by overridden. This is even harder
to analyse as each speaker has his own and its existence, operation and
change is normally beyond the individual’s awareness, but with some
careful analysis and enough samples in a sufficiently small time span.

Therefore, the regular route words take are is marked in bold red in
figure 12.1. Note that in most cases, the word is only affected by a single
machine, and the rest just pass through normally, mainly because absent
irregular words all machines roughly choose the same positions for insertion
and the same vowels to insert.
Within each machine a simple mechanism is used:

1. If a word is selected to be overridden or bypassed, then this happens

here.

2. Places for vowel insertion are selected. This is conceptually done via
the Insertion Locator, abbreviated IL.

3. What vowels to insert are selected. This is conceptually done via the
Vowel Decider, abbreviated VD.

4. The insertion goes through.

5. The word is then passed on to the next machine, or is outputted as

the vowel-studded sequence.

Figure 12.2a also shows that for the dialectal machine, the bypass in-
cludes an additional provision for certain words to be given their own in-
sertion location, which may be different from the ones that the dialectal
IL would have found. This alternative IL is labelled “Irr” in the diagram.
Additionally, figure 12.2b emphasises the fact that its bypass is placed after
the insertion has gone through.
132 CHAPTER 12. VOWEL INSERTION

12.3 Inserted vowels

Generally speaking, the vowel that is inserted into the word is entirely de-
pendent on the consonant that follows it. As will be mentioned in section
12.9 and implied in section 12.2, the exact rules will be somewhat diverse
over the entire region where Drsk is spoken, but the statements produced
here will be more or less true throughout the Drsk-speaking world. How-
ever, keep in mind that as with most things in linguistics, there are always
exceptions, and there is never a rule to account for them.
Throughout all dialects, there are at most 9 vowels that are candidates
for insertion. Consonants are divided into these nine vowels as such: (23 | 3)

i • tʲ θʲ ɨ • Others ɯ• x ʡ ħ ɣ
ɪ • t tʰ t͡θ θ̆ ɯ̞ • k g c ɟ
ɤ• ɫ ʎ
ə• ɬ ɹ

ɛ • t͡s d͡z t͡ɹ˔̠̊

ɐ• m n ŋ ɲ

In this diagram, the vowels are on the left, whereas the consonants that
the vowel can precede are on the right. Note that none of the vowels are
displayed as rounded, even ones that one might normally expect are due
to their position. This is true for some dialects but others may choose to
arbitrarily round some of them. [ɨ] is the special all-purpose phone that can
precede any consonant, even ones that are listed for other vowels.
Additionally, there is no dialect that uses any vowel lower than /ɐ/; this
might appear to be a violation of the rule that the vowel space must be
maximally extended where there is any extension in that direction at all,
but a comparison of the dialects and the marginal vowels [a] and [ɐ] seen in
the phonetic alphabet for the numerals 1 and 2 (see section 14.4) show that
the vast majority of dialects use the latter and not the former. Indeed [a]
is only seen in the numeral 1 in the spoken orthography outlined in section
14.4, and so it could be said that /a/, if it exists at all, is under ownership of
a single niche concept and can so be ignored for the purposes of most users.
This lack of [a] in general Drsk contributes to a stereotype that such
speakers, and those that look like them, cannot open their mouths very
well.
12.4. MERGERS 133

12.4 Mergers
Smaller Nv are produced by way of mergers, which are exactly as they sound
– the consonants that belong under one vowel are merged together with the
consonants that belong under another vowel, to one of the original vowels
or in some cases a different third vowel. Mergers are expressed in merger
formulae, which are in the form:

p1 + p2 + p3 + · · · + pi → f (12.8)
where pi are individual phones, and f is the vowel they are merged into.
The merger means that all pi are now pronounced f , and all the consonants
that are allowed to be under any individual pi are now allowed to be preceded
by f .
In this form, the most common mergers are:

[i] + [ɪ] → [ɪ] (12.9)

[ɛ] + [ə] + [ɤ] → [ə] (12.10)
[ɯ̞] + [ɯ] → [ɯ] (12.11)
[ɯ] + [ɤ] → [ɤ] (12.12)
[i] + [ɛ] → [e] (12.13)
x → [ɨ] (12.14)

Note that for formula 12.10 the three terms on the left show a triple
merger. There is nothing unusual about that, but it does also imply that
there are additional rules that involve fewer terms. In particular, for this
rule either [ɤ] or [ɛ] can be omitted from the merger.
Formula 12.14 is a basic license to arbitrarily merge any single vowel into
the vowel [ɨ].
Any combination that results in no conflict can be used in a single dialect.
Additionally, these rules can be conditionalised to certain environments such
as only restricting some of these rules to only apply to the beginning or end
of a word (specifically a phonological word, see section 10.6) or the exact
consonants involved can be reassigned to other vowels during the merger.
Therefore a more exacting merger formula should use the full (v, C) tuple
instead of just the plain v.

12.5 Insertion location

The other part of the vowel insertion system is where to place it, and the
general idea is generally to allow insertion almost wherever. However, there
are some general rules to where they are usually expected and where they
are usually not expected.
134 CHAPTER 12. VOWEL INSERTION

A particular method for installing vowels into the series of consonants

is called a “Form”. Each dialect has approximately one or two forms, and
they number in the tens to hundreds. These also tend to vary across smaller
areas such as towns, neighbourhoods or even individuals, and this is one of
the ways that one can identify an individual. Note that which vowels are
being inserted is not part of a form.
There are two major “types” of forms: one that results in the creation of
pseudo-syllables and one that does not. (34 | 18) These are called Syllabic
Forms and Phonorun Forms. Of the two, the former is more numerous
and common, while the latter is concentrated in the Woróli and Eamttel
Language Areas.
Generally speaking, a form consists of a single sequence that will be
the general guideline to where vowels may (not must) be inserted. The
rule is typically simple and can be easily applied without doing too much
calculation. However, forms are a general rule and certain words, especially
common ones, may not follow its prescribed rule. This happens, for example,
with the special pronouns #p and #v, as well as a smattering of common
words in most dialects, with the most common ones being ntʲ and pdᵘ.
Additionally, individuals may decide to deviate from the rule themselves,
adding to the general diversity of forms.
Some pairs of clusters like to stick together and refuse to admit a vowel
between them. These pairs of consonants are called “preferred clusters”, and
while there are some that are genuinely universal, which pairs are preferred
and which aren’t are usually dependent on dialects. So strong is this relation
that it is used as the primary tool to distinguish dialects from each other.
Where a form requires that one insert a vowel in a preferred cluster, a vowel
is never inserted, but whether the vowel becomes simply lost or it is
delayed to another location is dependent on dialect.
We can make use of the wheel of Drsk dialects defined earlier in section
3.3.2 to indicate these preferred clusters, by unwrapping the wheel – that
is, the coloured regions in Figure 3.4 – into a flat line and then using the
perpendicular axis to indicate popularity and prevalence of a single pair of
preferred cluster.
The result, Figure 12.3, is drawn as a trapezium to indicate there are
few universally-accepted preferred consonants.
As it implies, the majority of these preferred clusters are preferred around
the long annulus that surrounds the central dialect; Woróli’s dialects usually
have very low tolerance for vowels so virtually every pair of consonants is
preferred, and it would be impractical to show every pair. A similar story
goes for North Nffek, as it is similarly intolerant of vowels. As one ventures
further down the blue (“ultra-local”) end of the spectrum, the amount of
preferred clusters become extreme and fragmented; at this point the diagram
partially breaks down as the two-dimensional nature of geography starts to
show but the general idea remains.
12.6. AN EXAMPLE OF A SYLLABIC FORM 135

Figure 12.3: A list of consonant pairs that prefer not to have a vowel inserted
between them sorted by dialect and prevalence. Brackets placed around a
superscripted character mean that both the plain consonant and the con-
sonant with the additional superscript letter can be paired with the other
letter to formed a preferred cluster; if the character is replaced by a · this
applies to any possible letter that has the same base character.

12.6 An Example of a Syllabic Form

Syllabic forms usually create a form for a “pseudo-syllable” to form. In this
section, we will discuss syllabic forms using the pseudo-syllable structure
used in Nteaspdr, which is

P P vLP P ∪ LT

where:

• P is a plosive;

• L is a liquid;

• a ∪ b is the union of the set of phonemes in a or b;

• T is a terminator;

• x is the complement of the set of phonemes in x, i.e. any phoneme not

in set x.

• v is the inserted vowel; and

• The overall expression is formed by concatenating the elements in the

formula in the order they appear in the expression.
136 CHAPTER 12. VOWEL INSERTION

Apart from the inserted vowel, which is required by virtue of vowel inser-
tion, only one of the following items must appear in the “syllable”: the initial
plosive, initial non-plosive or the terminator. Other elements are optional
and as with other languages with real syllables usually omitted as the full
syllable structure is very large and unwieldy.
So for instance:

(12) d ɹ s k
d ɹ ɨ s k
P P v P T

(13) k s ð̆ m · ɹ l ʃ θ · d͡z k
͡
ɨ k s ð̆ ɨ m · ɹ ɨ l ʃ ɨ θ · dz ɨ k
v T P v L P P v T
P P L v T P v T

(Word boundaries are shown by middle dots.)

In both (12) and (13) we set Nv = 1 (and therefore NvS = 1Sv = {[ɨ]})
to keep things simple regarding insertion. The pseudo-syllables that have
been formed as a result of the insertion is also shown below. Notice that
unlike real syllables, pseudo-syllables can overlap with each other (though
they don’t have to) and although not displayed here an isolated consonant
may belong to no pseudo-syllable at all.

12.7 The Phonorun Form

The “phonorun normal form” is another system of vowel insertion. This is a
particular form of insertion that adheres a bit more strongly to the idea of
a phonorun than the insertion rule we have previously described. This isn’t
visible in the quantitative metrics as they insert roughly the same quantity
of vowels anyway.
The normal form is so named because it is the form of insertion used
in the normal dialect (the one in Wóroli). There are also other forms
of phonorun-based insertion rules, such as the Phonorun Reference Form,
which behaves similarly but with subtle differences.
In all phonorun-based insertion vowels are inserted at fixed intervals from
the terminator. Specifically, in the Normal Form, vowels may be inserted in
the following positions left of the terminator: (34 | 18)

{0} ∪ (3N+ − 1) = {0, 2, 5, 8, 11, · · · }

This means, for instance,
12.8. THE REBIRTH OF SYLLABLES 137

(14) d ɹ s k
ɨ d . ɹ s ɨ k
0 T 2 1 0 T

and this is normally notated as writing the number of phonemes that

are required between vowels, i.e. (2, 3, 3), where the overline here represents
that the last item repeats forever. (In practise, one can anticipate that no
run will exceed 20 phonemes, so the notation can be truncated.)
Here we have three opportunities to insert a vowel but we only actually
inserted two, under our own discretion. Remember that it is a may-insert,
not a must-insert.
The other phonorun-based form that is still extant is the Reference
Form, which is more commonly seen near Eamttel. The Reference Form
uses the pattern (1, 2, 3, 2),27 which instead allows insertion at positions
{1, 2, 4, 7, 9, 12, · · · }, or in set notation {1} ∪ (5N+ − 3) ∪ (5N+ − 1).
Keep in mind that incipators are sometimes treated differently when
inserting vowels. For instance, some of the dialects in the rural areas tend
to add an extra vowel of some sort after the incipator, if there is no other
vowels that would be inserted and it does not separate a preferred cluster.
This may mean that the resulting realisation might end up being parsed by
foreigners as crossing phonorun boundaries.

12.8 The rebirth of syllables

The syllable analysis of Drsk is borne from vowel insertion, and from the
fact that vowel insertion happens in specific places. The “syllable structure”
from section 12.5 are taken as genuine syllables, and from there the Drsk
syllable is born.
However, for various reasons discussed in section 10.6, this analysis does
not necessarily translate very well to certain parts of the phonology. It does
however influence dialects inland, where the grip of syllables is stronger.
The false syllables are also taken as real syllables when Drsk words are
exported into other languages. However, different languages may take to dif-
ferent syllables, owing to the fact that any individual consonant can belong
to zero, one or two syllables.
Take for instance name of the language itself: drsk. In full, with com-
bination with its genders and suffix, the name is gŋdrsktʲ, which in turn is
pronounced /gŋ.d.ɹsk.tʲ/, or in “syllables” [gɨŋ.ɨd.ɹɨsk.ɨtʲ].
This is loaned into other languages as such:

• In E.-Pasaru, the country is named “J.-Odrsi” /...o.dɹs.i/, derived from

drsk as /d.ɹsk/ expanded to [ɤd.ɹ.sɨk̚] from one particular dialect near
27
There’s no particular reason why the repetitions are not folded as far as they can be;
this is down to historical considerations.
138 CHAPTER 12. VOWEL INSERTION

Eamttel. The transformation from /ɨ/ to /i/ is a later invention; the

prefix “J.-”28 is the ordinary prefix for kingdoms and empires.

• Also in E.-Pasaru, the language is named “E.-Ďǐrsǐk” /...ðɨɹ.sɨk/, again

from /d.ɹsk/ becoming [dɨɹ.sɨk]. This one is from another dialect, this
time closer to Nffek, that treats incipators differently when inserting
vowels. Similarly, the prefix E.-12 is the ordinary prefix for languages.

• In Sturp, Drsk is named “Ginetresuk” /gi.nət.ʁə.suk/, or more notably

just “Ginet” /[Link]/ as a nickname of sorts. Sturp prohibits final /d/
so /t/ is substituted. The vowels are shifted somewhat arbitrarily
due to dialectal variation, which is reconstructed to be a dialect that
inserts them as *[giŋ.əd.ɹesk].

• The smaller language of Toipia (EP E.-Teifie) names Drsk “Goundlih-

siat moon” /gouɁ[email protected]Ɂ.../,29 which is realised as [gʌ[Link]ə.̯ set...].30
Toipia’s double vowel nucleus obscures the details here a little bit, but
some things are still visible: The vowels /ou ia/ are all manifestations
of the same dialect that also derived “Odrsi” (as seen in their actual
manifestations [ʌ e] < [o i]; the extra /i@/ is a phonotactic requirement
of Toipia’s.) The word “moon”31 simply means “language” here.

A lot of other languages borrow from these languages (or Rattssaw’s

/d.ɹs.k/) to name the Drsk language, but the point where the syllables
are “false” is made here: Sturp says /gŋ.t.ɹsk/ but Toipia says /[Link]/.
Only languages where there is a phonorun gets to faithfully replicate the
phonological structure of the word.
The examples should further illustrate that the vowel insertion rules
described here is only one of many and there are alternate positions that
can be used that would not be predicted by the system above.

12.9 Local variation of quantities

There are two types of variations in Drsk language communities, which
together give them their identity: their exact insertion strategy (which we
will describe in section 12.10, and their quantitative metrics, which we will
describe here. Given a combination of these two, one can uniquely identify
28
“Jesdic” /Ɂ[Link]/
29
The phoneme /@/, in Teifie, is a “clip” vowel; when placed at the second slot, the other
vowel (which must be one of Teifie’s normal vowels, /i a o u/) will be clipped slightly and
ended with either [ə]̯ or just be glottalised, so /a@/ is realised as [aə]̯ or [a̰]. Various other
vowel shenanigans are described in (34 | 26).
30
Additionally, the closed syllable phoneme /-Ɂ/ has a highly variable pronunciation in
Toipia, being pronounced [n] if next to /d/. However, the adoption of the final [t] is only for
sympathetic purposes with the language, and not specifically part of Toipia phonotactics.
31
“Moon”, phonemic orthography “mooj” /mooɁ/ [moon]
12.9. LOCAL VARIATION OF QUANTITIES 139

most language communities; if also given a rough geographic distribution,

then all can be identified uniquely.
As previously explained in Section 12.1, dialects and their determination
in Drsk is a largely quantitative affair, which means that dialect continua
are produced naturally instead of discrete dialects.
The quantities in use for determining such criteria are the values men-
tioned in that section: Nv and ρv . Together with a largely qualitative ex-
pression of how innovative a dialect is (see section 3.3.3), the full dialect
situation of Drsk is revealed.
First, let’s consider ρv by applying the smoothing procedure (see section
12.1.3) to the entire Drsk-speaking area.

Figure 12.4: The vowel insertion density isodense map, with the density of
the provincial capitals marked.

ρv runs through almost the entire range where it is allowed, and settle
in very characteristic patterns with correlations to a large number of other
qualities. In particular:

1. Areas that have large amounts of contacts with other languages, in

particular over land, tend to have a higher ρv .

2. Areas deep inland have a higher ρv , whereas coastal regions have lower
ρv .
140 CHAPTER 12. VOWEL INSERTION

3. Urban areas have higher ρv . Areas with very low population density
tend to have lower ρv , but of course the quantity is undefined where
there are no speakers.

4. The central part of the language region has very low ρv , due to a
confluënce of several factors listed above but also because the region
has faced a sort of stereotype threat where it has a lower ρv because
it is expected to have one. Such is the joy of having feedback loops in
your linguistics.

ρv is also known to have a strong correlation as to how conservative or

innovative a language is. In particular, at the centre of the language region,
not only is ρv at its lowest at ρv = 0.165±0.04, the language at that location
also preserves many ancient roots that have now been lost, and subsequently
contains a lot of bare-root speech as well.
It also has a reasonable correlation with Nv , which is noticeably stronger
if one adds on a subjective judgement on the innovation of any particular
dialect.

Figure 12.5: A comparison of Nv (generalised over language regions, allowing

it to take non-integer values) and ρv over a division of dialects over two parts:
innovative and conservative dialects. Also see text for further discussion.
(Error bars have been omitted).

The results, in Figure 12.5, are two lines, one for each of the dialect
classifications. In the most general sense, the idea is that more conservative
dialects tend to use fewer vowels than innovative ones, and lower ρv largely
corresponds to higher Nv . In other words, when the vowels get far apart,
they tend to become more distinctive.
It would appear that the major reason why the correlation is in this
direction is because of some unclear conservation law: one can either have
a steady streams of similar vowels or they can be very few and be varied.
12.10. TABLE OF DIALECTS AND THEIR INSERTION STRATEGY141

These are both responses to the enormous stresses created by massive con-
sonant clusters: one either controls it by way of inserting vowels often but
making them all distinct – the more conservative way of doing it – or by
attempting to keep the consonants together but adding additional distin-
guishing features to the following vowel to keep the information content of
any particular combination of phonemes high enough on redundancy.
The one extra feature visible in this graph is the Foreigner’s Bump,
where the number of vowels are lightly increased in innovative dialects when
a lot of vowels are inserted. This feature is, as its name suggests, created
by foreigners adding their usual four- to six-vowel systems from their native
language to the insertion system.

12.10 Table of dialects and their insertion strat-

egy
We will now examine a number of dialects and how they deal with a selection
of particularly complex words.
The dialects are selected to display a maximum amount of diversity,
which is done to show how far the system can stretch. It should also serve
to make it more obvious that the two vowel insertion strategies that are
described above are not universal. They are displayed in table 12.1.
Table 12.1 uses a number of abbreviations to keep to space constraints.
First, the names of the dialects (actually language communities) are given.
The next section then includes some pertinent information about that dialect
when discussing vowel insertion:

• “Strategy” describes the fundamental mechanism that vowel insertion

is performed in. They are:-

S1 The syllable-based insertion rules, as described in section 12.6.

S2 Another set syllable-based insertion rules, this time using a second
“syllabic” structure, specifically:

C LV N

or consonant-liquid-vowel-nasal.
PN The Phonorun Normal Form, as described in section 12.7.

• NvS is the list of available vowels and the consonants that can follow
them, arranged in the vowel chart with the corresponding figure num-
ber.

• “ρv ” is the vowel insertion density.

142 CHAPTER 12. VOWEL INSERTION

Table 12.1: A comparison of how different dialects (language commu-

nities) each deal with a sample list of words. See text for descriptions
of the labels.

Dialect
Eamttel Semreal Mskre Dsnptt
Property
Strategy S1 S1 S2 PN
NvS 12.6a 12.6b 12.6c 12.6d
ρv (Low) (High) (Medium) (Low)
Location Eamttel Bssreoml Trokier Woróli
Preferred Clusters Lots A few A few Some
Regular Words
Word
fbnħpdᵘcᶜtʲ [-] [-] [-] [-]
ntʲ [-] [-] [-] [-]
W3 [-] [-] [-] [-]
W4 [-] [-] [-] [-]
W5 [-] [-] [-] [-]
W6 [-] [-] [-] [-]
W7 [-] [-] [-] [-]
W8 [-] [-] [-] [-]
W9 [-] [-] [-] [-]
W10 [-] [-] [-] [-]
W11 [-] [-] [-] [-]
W12 [-] [-] [-] [-]
Irregular Words
pdᵘ [pɨd͡z] [ip.ɨd͡z] [ɨp.d͡z] [ɨp.ɤd͡z]
sm [sɨm] [səm] [sɨm] [sem]
#p(a) [ʌp] [ʌp] [ɨp] [əp]
#v(b) [və] [vɯ] [və] [vɤ]
dᵈd [dʰɨd] [ɪdʰ.ɪd] [ɨdʰ.d] [edʰ.əd]
(a) #p is always to be realised as [Vp].
(b) #v is always to be realised as [vV].
12.10. TABLE OF DIALECTS AND THEIR INSERTION STRATEGY143

ɨ • Others ɯ• k g c ɟ x ɣ ʎ
i • t &c. θ &c.

e • t͡s d͡z t͡ɹ ɬ m n ŋ ɲ

(a) Eamttel (ðvᵉþtł) Language Community

ɨ • Others
ɪ • t d &c. θ ð &c.
ɤ• ɫ k c ɣ
e • t͡s d͡z t͡ɹ ɬ

ɐ• m n ŋ ɲ

(b) Semreal (smrl) Language Community

ɨ • All

(c) Mskre (mskr) Language Community

ɨ • Others ɯ• k g c ɟ x ɣ ʎ ɹ
i • t͡s d͡z t͡ɹ t d &c. θ ð &c.

ə• m n ŋ ɲ ɬ ɫ

(d) Dsnptt (dtᵘprⁱ) Language Community

Figure 12.6: Vowel correspondences for the dialects.

144 CHAPTER 12. VOWEL INSERTION

Table 12.2: Words used in table 12.1 and the phonemes that make them up.

Orthography IPA
fbnħpdᵘcᶜtʲ /fb.nħp.d͡zcʰ.tʲ/
ntʲ /ntʲ/
W3 I3
W4 I4
W5 I5
W6 I6
W7 I7
W8 I8
W9 I9
W10 I10
W11 I11
W12 I12
pdᵘ /p.d͡z/
sm /sm/
#p /p/
#v /v/
dᵈd /dʰ.d/

Then come the list of words that test and demonstrate the insertion
rules. The word is first given in orthography, then in phonemes, with dots
separating runs. Then the corresponding dialect’s realisations are listed.
 Part V

Orthography

145
Chapter 13

Written orthography

Common amongst most languages in Pasaru, the written orthographies in

Drsk are transparent and correspond perfectly to the phonetics. This re-
mains true despite the fact that some of these languages are very old indeed;
however in the case of Drsk the orthography is very new and the average
linguist need not be concerned about exact implications of an orthography
that somehow manages to keep up with the times.
(But if one is interested, the effect is that of an active if somewhat clan-
destine relationship between a speaker and the Will of the Language – when
the Will promulgates a change in, e.g. pronunciation, the speaker immedi-
ately respells it to the same effect, regardless of how others speak. Keep
in mind that in this case Wills can very well be in the forms of celebrities
and other frequently-in-contact units that merely needs to declare that a
language change is popular before it is effected to a large population near-
instantly.)
The common written orthography is typically called an alphabet, but
since the language doesn’t have vowels the difference between an alphabet,
an abugida and an abjad is more or less nil for this language. It is appro-
priate to call it an alphabet however because all the “native” scripts that it
uses are alphabets when used to write other languages.

13.1 Romanisation
The table in figure 13.1 shows the Latin orthography of Drsk, which is used
throughout the document. (24 | 5)
Note here the usage of a superscripted letter to allow for unambiguous
digraph division.
Strictly speaking, this is not exactly correct, as what is here labelled as
ťᵗ should really be tť , where the caron should belong to the superscripted t
and not the regular t. Because the superscript ť does not appear in Unicode,
in this grammar it is replaced with tˣ.

147
148 CHAPTER 13. WRITTEN ORTHOGRAPHY

Nasals Orals Linguals

T NT NV V T NT
T NT T NT
Labial mᵘ m p f b v
Dental nᵘ n t s d z rⁱ l lˡ
tᵗ dᵈ ł
tʲ sˢ dʲ r
ť þ ď ð
ťʲ þʲ ďʲ ðʲ
Dental affricate vᵉ tˣ
č
tᵘ dᵘ
Palatal nⁱ c ć cⁱ
cᶜ
Velar ŋᵘ ŋ k h g
kʲ gʲ lⁱ
kᵏ
gᵍ
Velar affricate kⁱ
Laryngeal y ħ

Figure 13.1: The Romanisation scheme for Drsk, as arranged in its phoneme
inventory table.
13.2. RATTSSAW SCRIPT 149

It is customary in the Romanisation that capital letters are not used in

general running text, as the native script does not have a case distinction.
Where used, capital letters indicate something has changed with the letter,
which in the case of Drsk means that it is either preceded by a #, as in the
case of P and V, or has been escaped, as in the case of everything else. For
example, names are always escaped, so the name \M would have a capital M,
and similarly the escaped sequence \[ŤSRGÐ\] would also be in all capitals.
In some cases the backslashes can be removed, which leaves the capitals as
the only indication that this is an escaped sequence. Place names are a
possible exception, where they sometimes are capitalised as in conventional
English, especially in text where the body text is mostly English anyway.

13.2 Rattssaw script

The Rattssaw script, named grⁱtᵗsˢħbztʲ, is the usual mode of writing Drsk;
it not having a writing system to begin with, simply borrowed it from Ratts-
saw, though the historical situation is more along the lines of having the
script be foisted upon it. It has an essentially one-to-one relationship with
the Romanisation (and hence the IPA), as seen in Figure 13.2.

Figure 13.2: The glyph list for the B.-Rattssaw script for Drsk. Where there
are two variant glyphs for the same letter, the less common of the two are
in a lighter shade of blue.

The script has been notably altered in some cases. Notice how a few
extra letters have been included, and there are a few new diacritics to help
with denoting some systematic changes as well. Existing letters are also
reused for new purposes. These are indicated in Figure 13.3. The new
glyphs still conform to the general æsthetic of the Rattssaw script: no more
than three strokes, with one of them being a large swooping curve.
150 CHAPTER 13. WRITTEN ORTHOGRAPHY

Figure 13.3: Left: a list of new glyph forms unique to Drsk. Right: novel
diacritics for the Drsk alphabet.

There are some characters that have two variants, which in Figure 13.2
are indicated by a different colour. This occurs mainly because there are
two different periods as to which the script was standardised.
Note that the letters m and mᵘ have ligatures with the spaces, which is
indicated in figure 13.4.

Figure 13.4: The ligatures of m and mᵘ with the two spaces.

13.3 B.-Pasaru script

As with all other languages in the Jesdic Pasaru, there exists a transliteration
scheme to the world’s official script, Bez Pasaru. It is used when importing
Drsk words into the world’s official language, Egonyota Pasaru. It too has
a one-to-one relationship with the Romanisation and the IPA, and it is seen
in Figure 13.5.
Notice that some of the glyphs in this script is unique to Drsk and is not
used outside of it. This seems unusual for something that is the international
script, but not only is this commonly done, it is basically expected for several
reasons:

• Giving a language unique letters allows the civilisation to retain its

own identity even in the face of homogenisation of script.
13.4. LETTER NUMBER 151

Figure 13.5: The glyph list for the B.-Pasaru script for Drsk.

• There is now provision for the language to change the meaning of those
letters later if it desires for whatever reason without any coördination
from other languages.

• Individual letters now have an identity that keeps it from being mis-
used in other languages.

• B.-Pasaru does not like using diacritical marks, which would otherwise
be used freely across languages.

13.4 Letter number

Every letter in Drsk is also associated with a number. This number is
frequently used to easily encode the alphabet to another orthography, or
as an initial step in transmission that does not permit the sending of text,
only numbers, and integers at that. The number represents the alphabetical
order as well, though which way the arrow of causation goes is a little bit of
a mystery. The code and the corresponding letters are shown in figure 13.6.
(27 | 28)
Originally this code was built for Rattssaw, and increases by one for each
letter in the Rattssaw alphabet. It is adapted for Drsk by simply appending
an extra digit, which is usually 0, but increases as smoothly as possible for
letters that don’t appear in the Rattssaw alphabet. This means that if there
is a five-to-one correspondence, the digits line up as 0-2-4-6-8, i.e. as if there
is a six-to-one correspondence but with the last one missing.
152 CHAPTER 13. WRITTEN ORTHOGRAPHY

Number Dr Rs Number Dr Rs Number Dr Rs

10 s s 126 mᵘ 238 gᵍ
20 y y 130 ð þþ 240 l l
30 p p 133 ðʲ 243 lⁱ
40 tᵗ tt 136 ď 246 lˡ
46 tˣ 139 ďʲ 249 ł
50 sˢ ss 140 r r 250 (iᵉ) yé
60 ŋ ng 146 rⁱ 260 t t
66 ŋᵘ 150 nⁱ i 264 tʲ
70 þ þ 160 (yᵉ) ə 268 tᵘ
73 þʲ 170 (j ) j 270 f f
76 ť 180 (á ) á 276 v
79 ťʲ 190 (ś ) ś 280 (uᵃ) wá
80 (a ) a 1A0 d d 290 (o ) o
90 k k 1A4 dʲ 2A0 b b
94 kⁱ 1A8 dᵘ 2B0 kᵏ kk
98 kʲ 1B0 ħ x 300 z z
A0 (fᶠ) ff 200 (rʳ) rr 310 h h
B0 c c 210 (ó ) ó 320 (u ) u
B2 ć 220 # # 330 n n
B4 cⁱ 226 (@ ) @ 336 nᵘ
B6 cᶜ 227 \ 340 (w ) w
B8 č 229 \[ 350 vᵉ e
100 (bᵇ) bb 22A \] 360 (aʸ) ay
110 (pᵖ) pp 230 g g 370 dᵈ dd
120 m m 234 gʲ 380 (é ) é

Figure 13.6: The complete alphabet and alphabetical order for Drsk, to-
gether with its letter code. It should be noted that the code is in duodeci-
mal, using the letters A and B to stand in for ten and eleven respectively.
Brackets surrounding a letter means that it is inserted to match the Ratts-
saw alphabet. Lines are drawn every time the twelves digit reaches 1 or
7.
13.5. LETTER NAMES 153

Furthermore, letters that do not appear in Drsk appear in the alphabet

anyway, which is useful mainly to write Rattssaw words and very little else.
In some cases the alphabet may omit these letters, which are parenthesised,
but the letter code always considers these as a proper part of the alphabet.
An interesting effect of this behaviour is that the two Rattssaw letters
# and @ (which is also #́) are imported into Drsk, the former of which
is subsequently used in two words: #p “[Link]” and #v “in-
[Link]”. This is usually hidden in the romanisation by
using the capital letters P and V, but in the B.-Rattssaw orthography the
# is there, clear as day. @ however remains more or less unused.
As a result analogous punctuation that Drsk has for this purpose are also
included in the alphabet, after the two Rattssaw ones. The exact purpose
of these letters are further explained in Section 21.6.
This makes the numbering for the whole # series of letters difficult, as
it is specified that @ takes the halfway position (i.e. -6.) The result is that
the three native punctuation are squeezed awkwardly in the remaining space
in pairs.
It should be noted that the correspondences have an impact on the ro-
manisation of Drsk. In particular, the fact that in Rattssaw, h represents
/h ~ x/, doesn’t make much of an effect to h, which is always /x/. However,
in some cases the letters don’t match, mainly to attempt to reuse some of
letters that represent non-existent phonemes, particularly vowels. The main
examples are y (Rattssaw: /ɪ/, Drsk: /ʡ/) and the tt/tᵗ family (Rattssaw:
/t+ /,32 Drsk: /tʰ/). # is a special case where the original value, also /#/
(for which see section 9.1.5) is instead replaced with nothing at all, making
a silent letter from the beginning.

13.5 Letter names

Each letter also has a name, which informs the orthographies and the translit-
erations. The entire letter name list is shown in Figure 13.7.
It is quite clear that there is a pattern to this list, and it greatly informs
how the transliteration is handled. The rules for deriving letter names after
the basic letters are handled are as such:

• If the transliteration has an ʲ, the letter would have the name of its
plain variant with the suffix dʲ.

• If the transliteration has a doubled letter, the letter’s name is that of

the single letter with a suffix č.
32
This notation means that the phoneme is “leaky”, which means that airflow escapes the
mouth after a plosive, but only just. One may consider it to be a combination of an affricate
and an aspirated phoneme.
154 CHAPTER 13. WRITTEN ORTHOGRAPHY

Dr Name Drsk Name Drsk Name

s sbst mᵘ sˢmd gᵍ grč
y yl ð ðs l nᵘl
p psn ðʲ ðsdʲ lⁱ dᵘnᵘl
tᵗ tfdsˢ ď mᵘďddʲ lˡ nᵘlč
tˣ ŋᵘtˣ ďʲ mᵘďd ł dᵘłn
sˢ sbstč r rl (iᵉ) čsh
ŋ rŋ rⁱ dᵘrl t tfdsˢ
ŋᵘ sˢrŋ nⁱ dᵘsčn tʲ tfdsˢdʲ
þ rþ (yᵉ) ys tᵘ sˢtfdsˢ
þʲ rþdʲ (j ) ðs f rv
ť vᵉsd (á ) svᵉs v vsn
ťʲ vᵉsddʲ (ś ) ssbst (uᵃ) sˢsvᵉ
(a ) vᵉs d dzgl (o ) tᵘs
k skpł dʲ dzgldʲ b bsn
kⁱ sskpł dᵘ sˢdzgl kᵏ skpłč
kʲ skpłdʲ ħ sħt z zlfd
(fᶠ) rvč (rʳ) rlč h sˢl
c zck (ó ) stᵘs (u ) nᵘs
ć szck # #sp n sčn
cⁱ dᵘzck (@ ) s#sp nᵘ sˢsčn
cᶜ zckč \ dsv (w ) þᵘs
č csn \[ dsvr vᵉ hvᵉr
(bᵇ) bsnč \] dsvp (aʸ) vᵉys
(pᵖ) psnč g gr dᵈ dzgl
m md gʲ grdʲ (é ) shvᵉr

Figure 13.7: The names of each letter in the Drsk alphabet. As all words
have the gender-and-suffix combination gvᵉ.·.tʲ, they are omitted here.
13.6. PUNCTUATION 155

• If the transliteration has an ᵘ, prefix sˢ.

• If the transliteration has an acute or an ⁱ, prefix dᵘ or s. There is some

confusion as to which goes to which. Note that the symbol @ follows
this rule, as it is a # with an acute accent.

This even applies to letters that are outside of the normal Drsk alphabet,
which are derived by analogy from existing letters or by simply taking an
approximation of its value plus s.
Otherwise, the origins of most of the letter names lie again in Rattssaw,
which explains their somewhat spotty connection to the actual value of the
letter. However, there are letters that are genuinely native-derived, and
they can be seen by their repetitive and systematic nature, which today
only survives in the sets bsn, vsn, psn, csn and rv, rþ, rl. Of these, only
bsn and vsn remain from the original incarnation of this alphabet; the rest
are replacements for a pattern that has long survived its members. (23 | 64)
The names of the letters have some significance beyond being mere
names, because by escaping single letters, one instructs the reader to read
the name of the letter instead of pronouncing its value. For instance, these
two single roots are pronounced differently (spaces have been added to em-
phasise the expansion):

(15) d f s
d f s

(16) d \F s
d rv s

Clearly this is more likely to happen for some letters than others, but
often when it comes to naming objects, a clever name would only consist of
a few escaped letters like this.

13.6 Punctuation
Beyond the simple matter of writing words, the script also has a handful
of characters that handle punctuation. Only the most common of these are
given numbers, and only the escape characters are given alphabetical order.
This is as with scripts that are influenced by or are derived from B.-Senlis,
but “international practise” usually puts all of these in the alphabetical order.
As a compromise, the grammar places them here, as part of the orthography
but away from the main alphabetical listing.
Despite their non-inclusion, they can still form a part of words and can
be used as such, though no native word uses punctuation as a letter in
156 CHAPTER 13. WRITTEN ORTHOGRAPHY

the same way that they use letters in the alphabet as letters. To do this
one simply needs to escape the punctuation mark, in essence “escaping” the
punctuation mark’s normal meaning and treating it as a simple letter. More
detail of this behaviour will be described in sections 13.6.3 (for the escape
sequences’ presentation), 16.9 (for their role in names) and 21.2 (for their
other semantic roles).

13.6.1 Small and large spaces

There are two types of inter-word spacing in Drsk, the small space and
the large space. However, they are not an empty space, but are instead
interpuncts: in the Rattssaw script, they are written as “·” and “:”; and in
the Bez Pasaru script, they are written as “ ” and “-”. Both spaces separate
words, but only the large space separates sentences. The small and large
spaces are also called the single and double space, after their shape in B.-
Rattssaw. All other punctuation characters have ligatures for the two spaces,
which are explicitly laid out in Figure 13.8.
There are no other sentence-ending punctuation, with the behaviour of
the Latin script question mark, comma, semicolon &c. being handled mostly
by context or in some cases by simply refusing to use constructs that require
them. The latter is used for the non-usage of the comma, as there is gener-
ally no phonetic distinction between subordinate sentences/clauses/&c. and
regular ones in the spoken language.
Because spaces are visible in B.-Rattssaw, there is a particular possibility
that leading and trailing spaces can have semantic meaning in the language.
And indeed, they do, at least sometimes. To have the additional semantic
meanings, spaces have to either double up or start a line with no written
objects before it. Let’s examine how these spaces work.
Starting a paragraph with a large space is done for three reasons:

1. As a sort of indentation, though real indentation using an empty space

(and not a large space) is preferred for this reason. This is normally
done in more informal texts of some length, and in places where the
character set is limited and there is no “real space”.

2. As a title marker. The large space is placed before all other body
text, and normally also with no indentation. It should always be on
the beginning of the line. In contexts where there is colour, the large
space is customarily in an emphasis colour, and that colour is red on a
bright background and yellow on a dark background. It’s also expected
that the title not end with a large space either, regardless of whether
or not the sentence is complete.

3. As a way to emphasise the entire paragraph. This is the intended

meaning if a paragraph that is not the first one starts with a large
13.6. PUNCTUATION 157

space. Where colour is used this large space should not be emphasised
with colour. In formal texts this is used to mark summary or intro-
ductory paragraphs, whereas in less formal texts this is used in a way
similar to how Latin texts may use boldface for an entire paragraph.
In informal texts when the large space is used for this purpose they
can be repeated for further emphasis.

Starting a paragraph with a small space is not usually done, at least not
with this formulation. Instead, adding an additional small space before any
single word is done for emphasis, again in a similar manner to how boldface
is used in Latin texts. Naturally this can include the first word in a sentence
or a paragraph, which means that a small space can appear in at the start
of a paragraph and after a large space. More than one small space can be
inserted for further emphasis.
A line consisting of only spaces of either type is considered to be a section
separator which is analogous to the horizontal rule in Latin texts. Unlike
the horizontal rule, they are aligned always to the beginning of the line.
In terms of spoken word, these extra usage of spaces are actually visible
as well, and in fact these are the originally what these spaces are used for.
In later iterations of the script, the meanings of these symbols are altered
primarily for ease of reading and the alternate meanings are the result of
the shifting of the semantic space.

13.6.2 Quotation and parenthetical characters

These are imported more or less directly from the Rattssaw script. They
are seen in Figure 13.8. (23 | 41)
All names for quotation marks have the gender zzz.·.tʲ, and so they are
omitted here.

Unit quote (lk) Indicates that this is the name of a unit. This is not often
used and is often replaced with capitals.

Title quote (fm) Indicates that this is a title of a work. “Works” gener-
ally include large bodies of text, images, sounds, or any combination
thereof.

Non-unit quote (fsm) This indicates a name that does not belong to any
object that would otherwise use a more appropriate quotation mark.

Place quote (kⁱl) This indicates a place name. Its symbol in the B.-Senlis
script doubles as a position marker, à la “X marks the spot”.

Botany quote (z) This indicates that it is the name for a type of plant.
It does not indicate that this is a name for a specific plant. In some
cases it may also indicate the name for any species of living thing.
158 CHAPTER 13. WRITTEN ORTHOGRAPHY

Figure 13.8: Left: the quotation marks of Drsk, written in the two native
scripts. Right: ligatures for the combination of quotation marks and spaces.
Top to bottom, in order: Unit, Title, Non-unit, Place, Botany, True, Verba-
tim, Verbatim-and-silent, Unusual usage and brackets.

True quote (tˣh) This is the quotation mark that claims that the words
inside are spoken by someone else.

Verbatim quote (skħ) This is the quotation mark that claims that the
words inside are spoken by someone else exactly.

Verbatim and silent quote (dsˢz) This mark indicates a quotation mark
that claims that the words inside are what a particular speaker means,
though he might not have said them explicitly.

Unusual usage quote (mᵘrⁱ) This surrounds words that are used in an
unusual manner, such as giving it a new meaning or to juxtapose it
against some other word.

Brackets (gᵍ) These brackets not only show parentheticals, they also are
used to explicitly resolve syntactical ambiguities wherever they arise.

13.6.3 Escape characters

The three special letters \, \[ and \] are what is called “escape characters”.
There are two types of escapes:

Individual letter escape One single character is escaped. This single

character is preceded by \, and can be any character at all, even itself.
13.7. A BRIEF HISTORY OF THE ORTHOGRAPHY 159

Group letter escape A sequence of letters are escaped. The beginning of

the sequence is marked with \[, and the end of the sequence is marked
with \]. Generally speaking, the letters are also set in capitals. This
may or may not be the same as escaping each letter individually.

These sequences are used in a variety of ways, and some are quite versa-
tile. However, the exact specifics are handled in section 21.2, and their role
in naming is specifically handled in section 16.9.4.

13.7 A brief history of the orthography

As previously mentioned the “gift” of writing was unceremoniously pushed
onto the Grade (1) civilisation by the Grade (2) civilisation. Like with most
other languages at the time, the rate of change can be considered to be
quite slow at that time, so the orthography is a good match to the spoken
language even after hundreds of years.
That this script was put upon the Drsk speakers involuntarily has some
fairly obvious consequences to it: primarily, the script is unprepared to
distinguish things that the spoken language does easily, which means that
lots of interesting innovations has to be made to the script in order for it
to fit. The scholars in Grade (2) are not at all prepared to understand
the ramifications of orthography – as so the saying goes, a good writing
script is hard to make – so there are a handful of rough spots in the initial
orthography.
In later times, however, knowledge of Drsk amongst Grade (2) increased
and the alphabet as shown in this document is promulgated. This script,
unlike the original version, has specific graphemes that fill in the worst of the
shortcomings of the Rattssaw alphabet with regards to Drsk (specifically, ť,
ď, \, ť and c), which were previously indicated using the special adjoining
marker in the Rattssaw script.
For the most part, the orthography then stabilised over time, and man-
ages to track the growth of the language quite well. This continued for a long
while until political unrest closer to the Restitution caused a small break in
the tendering of the orthography.
Sometime before the Restitution a transliteration of the language to B.-
Pasaru (then B.-Vohalyosun) is developed by linguists further south. This
extra script is mostly academic and used to import words specific to Drsk
into E.-Pasaru (again, then E.-Vohalyosun) until the end of the war and the
Restitution began, when the independence of Klzdmk was introduced.
The situation with the orthography in particular as well as the language
in general mostly calmed down during this war (see section 2.4 for details).
However, things picked up again shortly after independence, as an idea was
floated to rewrite Drsk into B.-Pasaru in order to distance itself from the
J.-Senlis (as Grade (1) is pre-literate, there is no “native script” to drop
160 CHAPTER 13. WRITTEN ORTHOGRAPHY

to, and any attempt to create one as part of a propaganda effort didn’t
take off). In the end, this does not take off, but it has influenced a lot of
policy decisions, such as the location of the capital and the insistence of the
country’s bilingualism. Nevertheless, in some places the B.-Pasaru script
is used in some diaspora communities that live in places that are not as
familiar with the B.-Rattssaw script.
At this point the standard orthography is more or less stable and after a
minor spelling reform its particular letters are included into the equivalent of
Unicode in Pasaru. This has a series of consequences to its future evolution:
primarily, that any change is now a little bit harder; and secondarily, that
any changes that do occur will want to follow characters that are already
in *Unicode. Prior to this standardisation system, the pressure was there
but is vastly smaller. Nevertheless, there remains lots of novelty characters
that were created for the purposes of respelling the language, including more
than a good dozen attempts to standardise or describe the vowel insertion
patterns of the language.
Meanwhile a plethora of alternate orthographies have sprung up during
the period of suppression of the language, which at the time was mostly
in the form of sneaking the language in by disguising it as something that
is not writing. These alternate orthographies will be treated specifically in
Chapter 14 as they form an integral part to the identity of the language,
even if they don’t otherwise figure very much into the linguistic part.
One of the more obscure consequences of the standard orthography is the
lack of indication of vowels, even though vowel letters are available in the B.-
Rattssaw script. The main reason for this is that, as demonstrated in section
12.10, neither the position nor the value of the vowels can be held steady
throughout the language. In light of that, and the at the time important
desire to have a single, unifying orthography for all the land (and all the
language), vowels are dropped from the orthography to avoid implying that
any word is pronounced with vowels placed in any particular way. (34 | 25)
Chapter 14

Other orthographies

14.1 The state of alternate orthographies

For as long as there is a way to write, there are attempts to allow for this new
system to use as as many media as possible, as it is not always desireable
or practicable to put in explicit text. For this reason, there are several
“alternate orthographies” (often termed “deuterographies” or “diagraphies”)
which each cater to a niche where ordinary writing is not possible.
These alphabets are treated as if they are just another way to write Drsk,
even if some of these were originally designed (and in some cases, still are) as
steganographic systems. They are not considered to be a type of encryption,
nor are they thought to be “lesser” in some particular manner. There is of
course still a explicit rule that the standard orthography is to be used where
restrictions that brought upon these alternate orthographies don’t apply,
though what counts as not applying is somewhat weakly construed and there
are cases where deuterographies are used where the orthography might be
more efficient.
(Naturally, the orthography tends to be the most efficient of them all.)
The inclusion of deuterographies as part of the language is fairly unusual
in the planet, but it is standard practise in most of the Remmsp languages.
This is largely a case of individual preference writ large, but there are prac-
tical reasons why this is done as well: (27 | 32)

• Alternate orthographies can act as a level of steganography, in which

places which would not otherwise accept text can be convinced to
nevertheless write the language.

– From this, we can also have “bilingualism by stealth” – by having

the other language use its standard orthography and “sneaking in”
Drsk by using its alternate orthographies that are less visible, one
can often put Drsk in places where the other language is heavily
encouraged to the exclusion of all others.

161
162 CHAPTER 14. OTHER ORTHOGRAPHIES

– More practically it allows for the standards of encoding Drsk to

be taught in an explicit or standardised manner, which is helpful
when the public needs to understand why situations that require
the use of these alternate orthographies sometimes results in sur-
prising behaviour.

• Alternatively, by making it part of the language one can thwart at-

tempts at using them for actual, serious encryption, which was a goal
for at least some of these.

• The artists demand them, or rather, they got ahold of them and now
they won’t let go. Artists are well known for never surrendering free-
dom once they learnt of it.33

• And of course, it’s also partially tradition, as this is common through-

out the Remmsp, and so it can be surmised that it was a feature of
proto-Remmsp.

No matter what the justification, it is commonly taken that the more

frequently-seen alternate orthographies are uncontroversially “part of the
language”, in that a proper description of Drsk is not considered complete
without it, regardless of whether or not it is actually part of the language as
a system to assign meanings to more physical representations. As a result
they are described here, as equals to each others, and only just a subordinate
to the “real” orthography.

14.2 The shape alphabet – an artistic code

The shape alphabet is one of the more striking of the codes, and it is one of
those that started off as steganography. It is also well known to be highly
deficient in glyph count – all the letters in the Drsk alphabet are squeezed
into just 9 distinct glyphs, meaning that each figure would have to play the
duty of almost a dozen other letters.
The reason as to this heavily reduced glyph count however is because
there is an immense freedom one has to alter those glyphs to fit in other
media, such as a drawing or an animation. Because of this, the lowered glyph
count is not very damaging as there is often other non-script factors that
can help with disambiguation. In particular two similar words like -dks-
“fort” and -dᵈkⁱsˢ- “non-potable water” can still be disambiguated simply by
drawing the object near the words, or more cleverly hiding the script within
the drawing.
The full list of glyphs, and the letters that they represent, are seen in
figure 14.1. (27 | 33)
33
unless it’s the absolute anarchy that maths provides, of course. But see section 29.3.2.
14.2. THE SHAPE ALPHABET – AN ARTISTIC CODE 163

Figure 14.1: The shape alphabet.

164 CHAPTER 14. OTHER ORTHOGRAPHIES

These basic glyphs spell words using the usual manner of writing one
after another, but the important part here is that the writing direction is
not fixed and can take any line as desired.
Not displayed here is the myriad of systems that attempt to calm the
massive ambiguity by adding distinguishing marks and hatches. Such schemes
are generally idiosyncratic and are not considered part of the language as the
standard way to deal with the ambiguity is by consulting context, whether
it be the more standard linguistic context of nearby words, or by the other
non-linguistic cues that surround or hide the word.
Shape code has several other irregularities that are missing in other or-
thographies. Due to problems regarding amount of available space, it is
often the case that genders and suffices are missing from nouns, in a method
sometimes called “bare-root writing”.
Bare-root writing is interesting because it hearkens back to an older form
of the language, when the massive lexical collapse occurred that caused many
existing roots to coälesce. In fact, so severe was the collapse that the shape
alphabet has an alternate mode to get around the resulting ambiguity – the
shape logography.
In the logography, the letters no longer represent individual sounds, but
instead have to be taken as individual words to be learned separately. Some
of these words do have fairly transparent spellings, but other words are
loaned from nearby languages, such as Rattssaw. On Earth, a similar thing
has happened with Book Pahlavi, so this isn’t entirely surprising, though
of course the mechanisms behind these two situations are still different.
(27 | 40)

14.3 Tap code

The tap code is one part the deaf version of the language and one half Morse
code. (27 | 30) It associates each letter in the alphabet to a number using
the letter number described in section 13.4, and then associates each number
with a slightly unusual octal number.
To tap out an octal number, one uses three fingers, and assign each one
a power of two, viz. 1, 2, 4. The fingers that are tapped will then add its
value to the number in question. Digits are separated by rhythm, so each
digit takes constant time to sign.
An example of how rhythm-based divisions work is displayed in figure
14.2, which is a piano-roll representation of the value

150350 712301 132760 600707 134TAP

As with a standard piano-roll, each vertical line in the background grid takes
the same amount of time to cross, so it’s entirely like a typical rhythm game
where the black dots are where you need to land your fingers down.
14.3. TAP CODE 165

4:
2:
1:

Figure 14.2: A piano-roll representation of tapping out the amount

150350 712301 132760 600707 134TAP in tap code.

The transformation from letter number to tap code is not straightfor-

ward, however. It depends on whether the units digit in the duodecimal
representation is 0 or not. If it is zero, discard the zero, and then tap the
remaining digits out in standard octal.

3012 312
(14.1)
3TAP
16012 1612
(14.2)
22TAP
2B012 2B12
(14.3)
43TAP

If it is not zero, things get a little bit more complicated. As before one
removes the units digit and sign that part regularly. However, before that,
the units digit is signed by converting it to octal and then adding 608 to its
octal value. It is then tapped out as itself:

BA12 A12 · B12

(128 + 608 )TAP · 13TAP (14.4)
7213TAP
1A412 412 · 1A12
(48 + 608 )TAP · 24TAP (14.5)
6424TAP

Here the symbol is used instead of the more conventional =, in order

to avoid distracting technicalities of defining equality between these values.
However, it is true that the relationship is two-way (as it must be for a code
to make sense) – BA12 = 7213TAP , 2B012 = 43TAP , and so on.
The gap between letters is tapped out as 0. Word separators are tapped
00, but there is no higher division.
As previously mentioned the tap code is partly for the deaf and partly
for the ability to convert letter numbers into a binary code of sorts. The
unusual usage of octal comes from the “deaf” part of the equation – the
Eastern Klzdmk Language of the Deaf uses eight “spare phonemes” to spell
166 CHAPTER 14. OTHER ORTHOGRAPHIES

out Drsk words whose signs are too obscure or nonexistent. However, they
were significantly more efficient than tap code, because there are specialised
combinations of spare phonemes that allow for multiple digits to be trans-
mitted at once. The limitations brought forth by more recent inventions of
the *telegraph makes it so that only one digit can be transmitted at a time.

14.4 The spoken orthography – a “phonetic al-

phabet”
The tap code and the phonetic alphabet are very closely related to each
other. It is a very simple system that is created by simply associating each
word with a phrase that includes both the name of the letter (as in section
13.5) and the tap code.
The letter name is spoken mostly as-is, with minor adjustments for di-
acritics; the tap code however has a special encoding which is also used for
numbers. These octal numbers are pronounced:

Number Pronunciation
0 [hə]
1 [da]
2 [tɐ]
3 [kʌ]
4 [sɯ]
5 [ŋ̰ɨ]
6 [ʡy]
7 [ðo]

Notice that vowels are significant here, partially for consistency with the
rest of the planet and partially for redundancy.

14.5 Usage of alternate orthographies

In modern days these alternate orthographies generally don’t have much
practical use, and aren’t used often in practice; they are unwieldy and have
severe deficiencies that make them unsuitable for long-running text. There
are exceptions, but they generally revolve around having another medium
of expression that is more amenable to them, such as the digit code.
Historically though, as previously mentioned, these deuterographies are
important in keeping the language alive as they evade the (at the time)
inattentive eyes of those that would rather see the language disappear. By
quietly teaching the existence of these codes, and being able to deftly hide
them amongst other less conspicuous text, writings in Drsk can continue to
14.5. USAGE OF ALTERNATE ORTHOGRAPHIES 167

thrive even amongst some fairly heavy hiding. Eventually a sort of equilib-
rium came about where Drsk is tolerated but only in terms of these alternate
orthographies, which is fortunate because it is only so long before anyone
sufficiently motivated will decode and stamp them out.
Back in the modern day, with their historical role now invalid, deutero-
graphies are placed into a more decorative or subtext role. Thus, they are
still playing to their strengths even while the context has changed.
For example,
168 CHAPTER 14. OTHER ORTHOGRAPHIES
 Part VI

Morphology

169
Chapter 15

Morphology overview

The structure of words in Drsk is varied and elaborate, but somewhat simple
in that it there is only a small number of individual forms that need to be
explained. In this chapter, we will give a brief overview of categorising words
and how they fit with each other.

15.1 The word type hierarchy

Words

Bound
Morphemes

Inflectional Other
Nouns Verbs Sundry Quasinouns Dredge
Morphemes Morphemes

Subdivided by various genders Ownerless Owned

Content

Other Other
Relational
Ownerless Owned

Structure

Figure 15.1: The word type hierarchy.

Words in Drsk are divided into a variety of small pieces, which for the
most part belong in a neatly-partitioned tree. This tree is displayed in Figure
15.1, and we will be discussing these in full in later chapters.
This particular form of the tree is the traditional form, although it is not
usually presented as a tree but instead as a circle, with the root node “Words”
in the centre and all the other objects surrounding them in an onion-skin

171
172 CHAPTER 15. MORPHOLOGY OVERVIEW

manner. This is why the two large boxes labelled “Content” and “Structure”
seem to cross tree branches.
Many of the word types shown in this diagram are familiar to those who
have even a most cursory understanding of language, and for the most part
they behave exactly as one would expect things labelled with those names to
behave. These categories include nouns, verbs and most of the inflectional
morphemes. All the other categories are a little bit more obscure but they
otherwise still behave simply and predictably. The only complications these

15.2 Colour scheme

The general method of explicitly indicating the function of each word in the
Pasaru language tradition is to use Orbscript, which functions as interlinear
gloss but somewhat more limited in that it primarily concerns itself with the
function and properties of each morpheme rather than combining it with the
lexical meaning of the word itself.
A full treatment of Orbscript is beyond the scope of this grammar34 as
we will be using standard interlinear gloss, but its role of also indicating
the grammatical part of speech of the word will be used in augmenting the
standard gloss text.
The primary innovation in Drsk glossing rules is the usage of colour to
indicate a part of speech. Although the individual colours used are somewhat
different throughout the region and time, and there is a difference as to how
colours are detected across the two different species, the general idea is
easy enough for us to figure out what it is. For this reason, we shall make
our own colour scheme here, one that is mostly æsthetic enough for human
consumption.
Here, the colour scheme we will use throughout the book is as such:

• Nouns (including quasinouns):

– For the genders:

* CAT SMAJ;
* CAT MAJ; and
* CAT MIN
– The root (which is uncoloured); and
– The suffix

• Verbs (to be used sparingly as it is quite hard to read; otherwise may

be this colour but with the background set instead);

• Dredge:
34
But apparently a whole lot of other things are not!
15.3. WORDS AND WORDLETS 173

– Owned dredge; and

– Ownerless dredge

• Other words that don’t fit with the above divisions (Sundry)

Any words that require colour but actually colouring it as required may
be distracting would be instead formatted as such. Other words that don’t
fit are not coloured.
The point to this colouring is not just to add unnecessary strain to
printers, but importantly it helps one navigate around the morphology of
the language, as they have been designed (if a little haphazardly) such that
the colours are evenly spread around every major morphology type with
a few exceptions. It also allows rapid identification of the part of speech,
which is customary in the traditional linguistics practise that surrounds the
language.

15.3 Words and wordlets

One of the most important divisions for languages is the division between
agglutinative languages and isolating languages. There’s also the third horse
in this particular comparison, which are the fusional languages, but such
features do not appear frequently in the list of languages in Pasaru, so we
will not consider them here.
Within traditional Pasaru linguistics, which can be seen in the proto-
typical “international language”, Egonyota Pasaru, the distinction between
morphemes and words can get a little bit vague as they both have an inde-
pendent visual existence.
Let’s consider the two Drsk words:

(17) a. zzzlktʲ
b. fd

These two words are lexical, as in, they represent a single coherent con-
cept. Yet, as the first of these is a noun, it’s very structured, and we can
split it up:

(18) z- z- z- lk- tʲ
grammar- graphemes- punctuation- [Link]- CITE

These morphemes are not normally considered words in conventional

linguistics – they’re morphemes – but in the Pasaru linguistic theory they
may be. In this case, we call them wordlets (zdsħmᵘdtʲ), as opposed to the
normal word zmᵘdtʲ. These “wordlets” generally just means that it’s small
174 CHAPTER 15. MORPHOLOGY OVERVIEW

fragment of a word, and makes no distinction to whether or not they can

stand on their own. That distinction is normally qualified using an adjective.
As such the distinction between the agglutinative language and an iso-
lating one is normally down to the phonological word, which remains well-
defined for most languages under consideration.
Whereas the verb is just fd, with no particular structure to them. There
is nothing inside the individual letters f and d that makes up their combined
meaning, viz. “adapt, adopt, admit”.
We can then finally make the case that in this framework, the nouns
are highly agglutinative, due to their large amounts of structure, and nouns
are highly isolating, because they are primarily atomic with no or very lit-
tle structure inside them.35 This contrast between nouns and verbs are a
recurring areal feature in languages around the continent,36 and are shared
by them regardless of their relationship with each other.

15.4 General comments on morphology

Despite the orthography, Drsk is like many of the languages nearby it ex-
tremely isolating, in a sense, which is indeed an areal feature influenced by
Rattssaw and other (genealogically related) languages. Most of the deriva-
tion with the language comes from simple concatenation, with a marginal
amount of it requiring a little bit of reordering. The contents of the wordlets
(as defined in section 15.3) do not appreciably change through most of these
derivations.
That has been the norm of the languages in this area since “ancient
times”, but there is historical record of previous versions of the language
that display a different past to the language. One of the most compelling
records of the language changing over time is the development of the Digit
State Machine, which is the modern way that numbers are formed in Drsk,
evolving from an older, slightly less state machine-like system. The full
details of this evolution is given in section 17.2, but in this section we want
to focus on one item in particular: the actual shapes of those wordlets that
are behind those symbols, and how we can tell they have changed.

35
Some verbs may have a small amount of structure to them as they have related compo-
nents, or may be derived from smaller verbs. However, the individual verb roots are allowed
to occupy the entire phonological word, which is not the case with nouns which at the very
least has to share with a suffix and normally has to share with many other things that aren’t
part of the root itself.
36
Due to the huge influence of the J.-Senlis, this particular contrast is seen all around the
northern continent, including the culturally distinct Sitħanor over to the east.
Chapter 16

Nouns

Nouns in Drsk are very structured objects, and like many languages on the
planet encompass more than just a single root – things like noun phrases
frequently also appears as “nouns” in a typical grammar. The highly struc-
tured nature of Drsk nouns facilitate this, although this is not as blatant as
in other languages.
In this grammar, we will continue to follow convention by indicating all
nouns and noun-related items in one chapter.

16.1 Total localisation

While technically a matter of syntax, it is nevertheless an important mor-
phological point that nouns never move within a phrase or sentence; their
locations are totally fixed and cannot be altered in any normal manner.
Even in literary formats considered laxer than the rest, the language still
puts up a lot of resistance to having its nouns moved around.
This extreme rigidity in word order is called “localisation”, and it is one
of the ways that one can identify nouns in the sentence. It generally permits
a word order to be determined in all cases. Further details can be found in
section 24.
It should be noted that while a noun can be localised, this is not the
same as saying that the name is monolithic. The difference is simply that of
other words: suppose that a noun is composed of some components, which
we will simplified by a simple list, (ab1 b2 · · · bn c), where the parenthesis here
represents the word boundary. Localisation means that the word can be
altered in such a way that it can be rewritten as any combination of the
symbols in the original word, and also one or more pairs of parenthesis,
e.g. (b1 b2 )(ab3 b4 · · · bn c). What is not permissible is for any other parts
of words to intrude into it, so for instance it is not permissible to have
*(b1 b2 )d(ab3 b4 · · · bn c) or *(ab1 db2 · · · bn c), if d is an object that is not part
of the original noun.

175
176 CHAPTER 16. NOUNS

CAT SMAJ

Gender CAT MAJ

CAT MIN

I-Root
Noun Internal words*
I-Suffix
Nucleus
Root

Suffix

Figure 16.1: Full structure of a single Drsk noun.

16.2 General structure

Every noun has a large amount of internal structure to them, which is highly
present amongst all nouns. In the most general sense, the structure of a noun
is broken down into the three parts: Gender, Nucleus and Suffix. Two of
these parts have further substructure to them, so a more detailed expansion
of nouns would look as in figure 16.1, with the tree rotated in order to save
horizontal space:
The “internal words” themselves have additional structure that is based
on what they have semantically, but they all have the same structure: a
single monolithic root, followed by an Internal Suffix, which also carries the
same colour as normal word suffices. Different types of internal words may
contain more than one repetition of this basic structure, and determination
of such would be indicated semantically.
Further discussion of how the divisions work and how they behave are
further detailed in Sections 16.4 and 16.8. The suffix is in contrast entirely
monolithic and its possible values are seen in Sections 16.6.
Keep in mind that this structure is only for common nouns; names and
other so-called “special nouns” get a specialised structure. For instance,
names of units typically have the structure as seen in Figure 16.2, whereas
the structure for names of vehicles are seen in Figure 16.3. In these figure
the notation λ.n indicate that there are to be at most n letters.
Regardless of the actual form here, all values shown are considered max-
imal; there are usually ways one can prune the tree into something smaller,
either to derive new names or to create a nickname of sorts. Further details
on the structure of unit names will be discussed in Section 16.9.
The distinction between a noun being localised and a noun being mono-
lithic plays a role in deciding whether or not a noun is a genuine lexeme.
The parts of a noun that can theoretically detach off from the base word
16.2. GENERAL STRUCTURE 177

Name

Status Origin Commune \[ Unit \]

λ.2 λ.2 λ.3 λ.5

Figure 16.2: The full structure of a name of a unit.

Noun root
Generic
Group root λ.3

Model λ.6
Specific
Submodel λ.2
Name
\[

Identity λ.5
Identity
Nickname λ.2

\]

Figure 16.3: The full structure of a name of a vehicle, rotated to save hori-
zontal space.
178 CHAPTER 16. NOUNS

are internal words, and those are typically not considered to be part of a
lexemic noun.

16.3 Internal words vs. full nouns

One of the more contentious issues in Drsk nouns is the existence of “internal
words” – these are items, structurally similar to the so-called “full nouns”,
that can sometimes be included into full nouns by grammatical means. This
is not at all an uncommon occurrence in the languages of Pasaru, but ev-
ery language that does distinguish between them treats them differently, so
below is how Drsk handles them.
As seen in figure 16.1, an internal word and a root are almost exactly
identical in structure – a single blob followed by a suffix of some description.
The distinction between them is clear, but not easily definable; most of
the ways that are proposed to distinguish between them have the arrow of
causation go the wrong way. The best way to distinguish between them is
piecewise: (29 | 9)

1. Nouns that refer to “concrete objects” or “first order abstractions” typ-

ically can be roots of full nouns. “Concrete objects” refer to tangible
items such as stones, stars and substrates, whereas “first order ab-
stractions” cover most objects that are generalised from these concrete
objects, such as numbers, colours and organisations.

2. Nouns that refer to abstractions of any order can be roots of internal

words. This means that some words can become both internal and full
words.

3. Quasinouns are internal-only, regardless of how concrete they are.

This three-step division are also used in determining what suffix they
use, as seen in section 16.6.2.

16.4 Genders
Although named “genders”, this is partially a taxomonic prefix system that
was developed in the view of a rapidly-decreasing amount of nouns that was
used as time went on. Reälity, however, resists such categorisation, and this
is why today the system remains partial, and since it was a natural growth
of the language rather than something that has been instituted by someone,
it remains somewhat awkward, but is very serviceable otherwise.
Because of the its taxomonic system, genders can and are subdivided
into three layers: CAT SMAJ, CAT MAJ and CAT MIN, as well as a single
alternate system called CAT STAR (which can also be spelt CAT*). For
16.4. GENDERS 179

historical reasons, these gender types are called Categories, with abbrevia-
tion CAT. All these categories cover all common nouns, with some proper
nouns also covered.
When genders are indicated by themselves, they are usually written with
a dot after in the Romanisation, e.g. k.. This can combine with the rule
that says that suffices alone are prefixed with a dot, e.g. .t, to create a word
that only has a suffix and a gender, which is written, e.g. k.·.t, where the
interpunct marks the missing root.
Because the number of genders is far too large to indicate using the
standard MORPHEME RENDERING, a special syntax is used to indicate them
in the gloss. While we could use colours to indicate them, we will instead
keep the glosses monochrome and use italics to indicate gender. As there
are two ways to gloss gender – one to indicate the gender indicated by all
three components, and one to indicate the breakdown into the three types of
category – we shall prepend the italics with colons: two colons for ::[overall
gender from all three categories], and one colon for :[the meaning of a single
category type]. If it is a CAT STAR, the colon(s) become(s) F instead.
Table 16.1 summarises how each of these types complement each other
in terms of simple binary properties.

Table 16.1: An overview of the four category types and how they are distin-
guished from each other.

Category STAR SMAJ MAJ MIN

Top-down X X
Bottom-up X X
Comprehensive X X
Partial X X
Tree-like X X X
Grid-like X
Order 1 1 2 3

16.4.1 CAT SMAJ

The Super-Major Category type, abbreviated to CAT SMAJ, is a top-level
division of the semantic space. Using the letters in the Drsk alphabet, the
space is divided into a few dozen parts, which are then labelled somewhat
arbitrarily with the letters. Not all letters are used, and this cannot be
predicted either – except for ħ, which is the designated name for a non-
CAT SMAJ.
CAT SMAJ is considered a top-down division because the entire space is
considered as a whole, and then subdivided into anything necessary. Because
180 CHAPTER 16. NOUNS

of this, and the fact that reälity cannot be easily segmented, there are several
“catch-all” categories, which are further explained in Section 16.4.6.
A list of meanings are indicated in table 16.2.

16.4.2 CAT STAR

The Special Category type, which is called CAT STAR or CAT*, is an alter-
nate division of the semantic space that specifically restricts itself to names,
usually of units and places. As with the ordinary Super-Major Category
types of genders, some letters are not used, and this is not easily predictable,
and furthermore there is no guarantee that a letter that is used in a CAT
SMAJ is not used in a CAT STAR, and vice versa, and the same applies for
non-usage.
Furthermore, although there are some entries where the two gender types
are thematically similar, (e.g. ð, k) there are also others where the two types
are entirely unrelated (e.g. v, p), and these are in fact the norm. There’s
again no obvious pattern to this.
A list of meanings are indicated in Table 16.2. Notice that the cases
where there is a CAT* for the males of a particular class of objects usually
have no corresponding non-male objects; in this case the missing class is
either given a CAT* that would be the next best fit or use a generic category
such as “friends” (n).
CAT STAR, unlike CAT SMAJ, can have either one or two letters, but
not any other number. (23 | 11)

Table 16.2: A list of genders, both super-major (CAT) and special (CAT*).
(23 | 13) A suffix of (m) restricts the class to males; a suffix of (nm) restricts
the class to non-males.

Letter CAT CAT*

mᵘ unfathomables area
m ships, trees & stones
nᵘ fords
n units in general friends
vᵉ *birds
nⁱ flat objects crossroads
ŋᵘ unseen animals fields, parks, resources
ŋ domesticated animals caves & canyons
p other animals dead males
f foods
b pets (m)
v plants & fungi royalty (nm)
Continued on next page
16.4. GENDERS 181

Continued from previous page

Letter CAT CAT*
t fabrics tunnels
s reptiles paths, gangplanks
d materials drivers, pilots
z grammar, ABS 3 artists
tᵗ services servants (m)
dᵈ distinguished citizens
tʲ royalties divine & royalty (m)
sˢ crops farmland
dʲ *fish1
þ economics
ť fire, heat bridges
ď beasts of burden beaches
ð celestial objects stars, stellar remnants
ťʲ collections
þʲ passes
ďʲ alcoholic drinks borders
ðʲ units of money ultrafriends
tˣ straits & harbours
č synthetics domesticated animals (m)
tᵘ ABS 1 produce mongers (nm)
dᵘ collections non-food salesmen (nm)
c metals and security servicemen (m)
ć abstract objects cities (as a point)
cⁱ divine & mythical
cᶜ ABS 2
k geographical country
h the sky and space
g writing & speaking seas
kʲ motion
gʲ mountains
kᵏ badlands
gᵍ social interactions in-groups
kⁱ *fish2 nobility
ħ designated invalid designated invalid
rⁱ recreations & games rough beaches
l locations rivers & canals
lˡ linears roads
Continued on next page
182 CHAPTER 16. NOUNS

Continued from previous page

Letter CAT CAT*
ł MISC and ABS 4 atomics
r war & debate vehicles (except boats)
lⁱ fluids and gases streams

16.4.3 CAT MAJ

The Major Category type, abbreviated to CAT MAJ, is a type of category
that is significantly less structured than CAT SMAJ. Unlike CAT SMAJ, it
is a bottom-up taxomonic system, with category names typically denoting
small groups of highly related objects than wide swathes of loosely related
objects.
Up to two letters can be used to name a CAT MAJ, although a vast
majority of them are unused or redundant. Enumerating them in a table
however is highly impractical as each of the classes have their own table and
there’s no obvious pattern to them either.

16.4.4 CAT MIN

The Minor Category type, abbreviated CAT MIN, is a type of category that
adjust or adds existing properties to the combination of CAT MAJ and CAT
SMAJ. Like CAT MAJ, a CAT MIN’s name can be up to two letters long.
Unlike other genders, CAT MIN can be stacked – more than one can
apply to a single word, and their overall meaning is similarly combined.
A particular reason why CAT MIN exists is to cover the fact that things
in real life generally don’t fall into neat taxonomic categories nesting into
each other in a tree structure. Instead, a loose set of attributes, that can
at first approximation be considered as binary have/have not types, can
be used to handle things that have multiple independent properties. The
resulting structure – a rubric – can still be modelled as a tree but will require
far too many category labels to keep track of. The have/have not pattern
also reappears too much to not consider separating them out as a separate
thing.
For instance, since there’s no particular reason why something’s flamma-
bility and friendliness are absolutely related, and there’s no obvious way to
determine which of the two is more important in all situations, these two are
modelled by CAT MIN words. As a result of their unrankedness, CAT MIN
words are always commutative (but certain combinations might be ruled out
for other, pragmatic reasons.)
Finally, certain attributes are common enough to repeat amongst most
or all nouns, and they occupy a slice of the CAT MIN letter combinations.
Examples include [±flammable], [±floats in water] or [±expensive/valuable],
16.4. GENDERS 183

which are relevant enough to most daily conversations to warrant such treat-
ment. Others are specific to a particular CAT SMAJ instead, e.g. the gen-
der food (f) will have the CAT SMAJ-universal [±cooked], [±allergenic],
[±shared] and so on. Such attributes typically are scattered throughout the
grid, and their positions and existences have to be learnt separately.
The below is an example of how CAT MIN is defined for a particularly
small field: chemical elements, which are divided amongst the higher cate-
gorisations of ds and cls. (35 | 34)

Period CAT MIN Group CAT MIN

1 kⁱe 1 rl
2 mᵘt 2 rvᵉ
3 pvᵉ 3 (Ac) rk
4 łgᵍ 3.04 (Np) zkⁱ
5 þz 3.05 (Pu) skᵏ
6 pnᵘ 7 sv
7 pz 11 nᵘb
12 ts
13 tᵘs
14 nᵘtᵗ
15 ðz
16 mᵘvᵉ
17 ðz
18 hlˡ

So for instance, we can say:

(19) Caesium clsþzrlłpyktᵗtʲ

c- ls- þz- rl- łpy-ktᵗ- tʲ
:[metal] :[element] :[period 5] :[group 1] [Link]-water CITE

Note that we can either use the empty root rule (section 16.7.2) or omit
some of the genders (section 16.7.1) to shorten the word. In the specific case
of elements, the shortening is fairly systematic and is standardised by some
authority or another.

16.4.5 Categories and adjectives

Quite a lot of category types here seem to duplicate the functions that other
languages will do with adjectives, and this is broadly true – in general,
“adjectives” in Drsk are either categories or quasinouns.
184 CHAPTER 16. NOUNS

16.4.6 Generic genders

As reälity cannot be divided evenly into parts with even size, several genders
are used for this purpose. They belong to two types: a catch-all type and
an ad-hoc type.
Drsk handles these aberrant cases by putting aside four CAT SMAJ in
this case: z, ł, tᵘ and cᶜ. They nominally have no specific meaning, but in
practise, they have some specific use:

• z is used when discussing words as themselves, or words that describe

other words, and is typically designated :[grammar].

• ł is used as the catch-all category, which means that it is the word that
words when nothing else seems convenient. It is typically designated
:[misc].

• tᵘ and cᶜ are typically used as ad-hoc categories to arbitrarily com-

bine items together and contrast against each other. As such they
are designated :[contraster] and :[contrastee], depending on individual
usage

Nevertheless, there is a constant and consistent habit of using all four in

a non-systematic way that makes them genuinely useful as “generic genders”.
The generic genders have no defined CAT MAJ or CAT MIN (except
when z isused as the grammatical category); these are also assigned ad-hoc,
per user per conversation, usually to allow for more generic genders to be
used or to further subdivide things that are in the generic genders.

16.5 Transformation of Genders

The difference between a root and a noun is that a single root can carry many
meanings whereas a noun can carry only a one. Differentiating between the
multiple meanings of a root and the single meaning of a noun (a single root
with no other internal words) is performed by using genders.
The origins of this system is cyclic: there used to be more roots than
the current iteration of Drsk, and at some point there was the addition of
genders in the language via an array of proto-articles sorted by a nascent
gender system. This obviates a handful of roots that would be redundant
with the new system, and those were discarded; more genders were formed to
create a hierarchy of such, and then more roots are discarded for redundancy,
eventually leading to the system we see today.
In practise, a single root can host four or five nouns. This is especially
prevalent with animals and animal products, as we can see in with the root
-sˢr-, √sheep1 .
16.5. TRANSFORMATION OF GENDERS 185

(20) a. -sˢr-
√sheep1
b. p-mᵘ-ďl-sˢr-tʲ
:[animal.F]-:[domesticated]-:[edible]-sheep1 -CITE
A *sheep
c. d-ħkⁱ-mdᵘč-sˢr-tʲ
:[material]-:[nometallic,degradeable]-:[+soft +flammable]-sheep1 -CITE
*wool
d. f-þť-sˢr-tʲ
:[food]-:[cooked, rare]-sheep1 -CITE
*mutton
e. sˢ-sˢr-tʲ
:[farmland]-sheep1 -CITE
*sheep farm

Notice that in (e), the category type of the sole object is silently changed
into CAT STAR instead of the expected CAT SMAJ. This means that it
might also be interpreted as :[crops]-sheep1 -CITE – “sheep grown to be sold”
– and this is an ambiguity that has to be resolved by context.
On occasion a single root might have two distinct classes of meaning
that are otherwise unrelated, or are only tenuously related. (29 | 7) Though
rare in absolute numbers, a very common example of one such root is drm
“√friend, cellar”, whose double identity nature is revealed in this list of ex-
amples:

(21) a. n-drm-tʲ
:[unit]-friend-CITE
friend
b. gᵍ-drm-tʲ
:[social interaction]-friend-CITE
friendship
c. tʲ-drm-tʲ
:[royalty]-friend-CITE
allyship (of a nation)
d. f-vp-drm-tʲ
:[food]-:[drink]-friend-CITE
food/wine cellar
186 CHAPTER 16. NOUNS

e. l-drm-tʲ
:[location]-friend-CITE
cellar
f. f-cⁱd–drm-tʲ
:[food]-:[preservation]-friend-CITE
walk-in freezer

Diachronically this is understood as having two roots collapse in a little

bit more enthusiastically than what is exactly warranted if the system was
simply reducing redundancy.

16.6 The noun suffix

The final mandatory part of a noun is its suffix. This suffix is morpho-
logically the same as the suffix seen at the end of internal words, and are
therefore given the same colour.
Regardless of its usage, the suffix can only be 1 letter long and can take
on the following values if it is a noun suffix. Additional values can be taken
if the suffix attaches to an internal word.

16.6.1 Full noun suffices

Full noun suffices are easily divided into several “series”, depending on roughly
where the consonant is articulated, which is faithfully replicated in the ro-
manisation as:

• The “t” series, typically dealing l Lative LAT

with morphosyntactic align-
lˡ Ablative ABL
ment:
ł Apudessive APU
tʲ Quotative QUOT; dictionary
form CITE; transitive case • The “s” series, dealing with lo-
for the verbs “is”, “have”, cation:
“is contained by” and
“and” TCOP; vocative VOC s Locative/Near LON
t Transitive case for all other
verbs T; instrumental IN- sˢ Locative/In LOI
STR; comitative COM
• The “ð” series, which is an odd
ť Intransitive IT
one out:
• The “l” series, typically dealing
with motion: ð Topic, trigger TOP.
16.6. THE NOUN SUFFIX 187

It should come to no surprise then that ð is the oldest attested suffix

of the system. In fact its application is somewhat unusual compared to the
remainder, in that it doesn’t have a “fixed” grammatical function; instead, it
carries a greater, pragmatic function of marking, quite literally, the topic(s)
of discussion.
Consider the following example sentences, which uses the topic marker
in each of the two arguments, as well as neither.

(22) a. P kʲvpsˢ-ð vbl

1 boat-TOP sell
I sell a boat.
b. P kʲvpsˢ-t vbl
1 boat-T sell
I sell a boat.
c. P-ð kʲvpsˢ-t vbl
1-TOP boat-T sell
I sell a boat.

All three of these share the same translation, and their primary difference
is to set up the expectation of what comes next. Specifically, (a) sets up the
expectation that the next sentence should discuss the boat, perhaps where it
goes to or who has received the boat. On the contrary, (c) considers the self
P to be the topic, and sets up the expectation that the next sentence would
be what P would do next. Finally, (b) sets up no such expectation, but in
general means the same thing as (c) due to the expectation that objects near
the beginning of the sentence are more likely to be topics.

16.6.2 Internal word suffices

The additional suffices available for internal words are not organised into
series, and are roughly evenly distributed throughout the consonant space.
They are as follows: (29 | 10)

mᵘ Possessive/Inalienable POI

gᵍ Possessive/Semi-inalienable POS; hierarchical HIE

þ Possessive/Alienable POA

cᶜ Comitative-fusional/copulative COMC; appositional APP

y Comitative-hybrid COMH; exocentric EXOC

ħ Endocentric ENDC
188 CHAPTER 16. NOUNS

These act as boundaries to internal words, though as usual it is not

possible to parse a word into internal words on these alone as roots also
use these letters. For internal words, only the l and s series of suffices are
available, for semantic reasons.
The three possessive suffices mᵘ, gᵍ and þ follow the principles of sepa-
rability, an ideal first pioneered by civilisations further south from there and
later on found its way into the global civilisation via the Vohalyosun. In
essence, in the language of separability as seen in (29 | 3), mᵘ binds aspects
to form units; gᵍ binds colours to bases; and þ binds fragments to com-
pounds. However, despite being called “colours” and “bases”, in Drsk colours
are typically handled by ħ, as colours are considered endocentric in nature,
that is to say, the bulk of the semantic nature of a phrase such as “green
table” is on the “table”, with “green” being the secondary part. In general,
only “concrete objects” get to use the first three, whereas the rest have to
use the other three, though there are some exceptions.

(23) ť- ml- mkⁱd- gᵍ- dsk- tʲ

:[fire]- :[strength]- emission- PSEP- generator- CITE
The power of a generator

(24) ť- ml- lⁱłd- ħ- dsk- tʲ

:[fire]- :[strength]- status- ENDC- generator- CITE
The status of a generator

“Power” is considered a colour because it is a quality that can be com-

pared against disparate objects with relative ease. Since “status” is a quality
that differs depending on the specific object it attaches to, it is therefore not
a colour, and instead is an endocentric property.

16.6.3 The non-suffix

Finally, there is the special suffix p. This suffix is the nil suffix; it is inserted
where one is grammatically required but none of the ones above apply given
the correct context. This does not occur when the noun is part of the main
sentence but does occur when nouns are required in certain constructions,
such as in situations where nouns find themselves in dredge.

(25) rskⁱ sˢcrlⁱ-p čmčm PRNᵘDLⁱ psˢvþkⁱdt

[Link] yesterday-NSC capture <name> <[Link]>
Yesterday PRNᵘDLi caught fish
16.7. OMISSION OF NOUN COMPONENTS 189

Here, the phrase rskⁱ sˢcrlⁱp is a very common phrase meaning “yester-
day” or “tomorrow” (depending on context); this is a type of dredge, which
we will deal with in chapter 19. Regardless of its form, the noun embedded
in such a dredge does not participate in the inflection of the main sentence
and therefore gains the “non-suffix”.

16.7 Omission of noun components

Unusually, it is not the root that is the only obligatory part of a noun;
the other two parts can omitted, though not both at the same time. The
omission behaviour and motivations is somewhat different for both of them,
though a common cause is that without the omission a noun can be too
long.
The primary reason why one would wish to omit parts of a noun, in
particular the genders, is that with everything together – gender, root and
suffix – gets very long and unwieldy. To alleviate this redundant parts are
removed. This also has a historical reasoning for this in the form of bare
root writing.

16.7.1 Omission of Genders

Genders can be omitted in both directions, but the motivation between
the directions are different. Omitting genders from a particular direction
is named from that direction, so omitting genders from the left is called
“left omission”. (28 | 14) If all genders are omitted, then this is called “total
gender omission”.
Right omission is by far the more common of the three, where CAT
MIN alone or with CAT MAJ is deleted. There are several reasons why one
might wish to omit from the right:

1. The noun is actually the prototype of the object, so any other subcat-
egorisation is not required and not desired. This especially combines
well with root omission, and is described in section 16.7.2.
2. The object is not known to have any of the properties that lower
categorisations is given, so none would be, to not be misleading.
3. The full form (CAT SMAJ, CAT MAJ, CAT MIN) of the word is
given already, and so fully specifying the noun at a later utterance is
not necessary.
4. Some methods of combining words require non-head nouns to perform
right omission.

Left omission is somewhat less common as most of the common cases

are already performed using right omission, but there are some uses for it:
190 CHAPTER 16. NOUNS

1. If a series of nouns, particularly in a list not denoted by f, has the same

common higher categorisations, any such categorisation are omitted
after first mention.
2. If it is desired to put more emphasis on the finer genders the coarser
ones may be omitted to “reveal” them.
Total omission is a special case of both directions of omission where
the whole gender is omitted. This is also known as bare root writing. It
is largely a artefact and explained in more detail in Section 14.2.
It is otherwise used when creating certain other types of compound
words, particularly ones that tend to have a head word subsume all the
other words. For instance, in (24) above, the word þbgᵍvlⁱłdtʲ “status” has
all of its genders omitted. This is not so much a type of omission but the
usage of a root to create a new compound, but is counted as an omission for
historical reasons.

16.7.2 Omission of Roots

Omission of roots is governed by the Empty Root rule, which is not a single
rule but a handful of vaguely related rules that all contribute in different
ways to interpreting an empty root.
The first and most obvious way the root can be omitted is if the meaning
of the gender itself already encapsulates the meaning and therefore any
additional root is pointless. For instance:

(26) n- tʲ
:[unit]- CITE
unit

It is possible for a CAT SMAJ to be paired with a CAT MAJ when

omitting the root for this reason, but it is not possible for a CAT MIN to
be included.

(27) s- d- tʲ
:[reptiles]- :[sliding]- CITE
*snake

Another reason why the root might be omitted is to create pronouns.

The examples above can be reïnterpreted as the following:

(28) n- Ø- tʲ
:[unit]- PRN- CITE
he (a unit)
16.7. OMISSION OF NOUN COMPONENTS 191

(29) s- d- Ø- tʲ
:[reptiles]- :[sliding]- PRN- CITE
it (a *snake)

In this case CAT MIN can also be included:

(30) ť- ml- č- Ø- tʲ
:[fire]- :[strength]- :[+flammable]- PRN- CITE
it (a strong and flammable object)

Note that these two cases are ambiguous against each other, though as
it turns out the semantic difference between the two can easily be irrelevant
or resolved in context.
The third way the empty root rule is used is to mark an object as an
unknown, as part of a question. This can be distinguished from the above
examples by taking into account that questions always end in kdᵘ sm.

(31) V gᵍ-Ø-tʲ zr, kdᵘ sm.

2 ::[social interaction]-Q-T have, yes no.
What kind of friends do you have?

Another way that an empty root is used is when a list of objects would
otherwise all use the same root. In this case, the first one would get the full
root, and the remainder get no root.

(32) n-Ø-tʲ f gᵍ-drm-tʲ tʲ-Ø-tʲ

:[unit]-PRN-T and :[social interaction]-friend-T :[unit]-REP-T
He is a good friend and ally.

Finally, there is a small and irregular usage of an empty root rule. This
is something that happens fairly uncommonly, but the words are culturally
relevant. (23 | 11) If the category is CAT STAR, then the root with zero
letters in it37 changes the meaning of the gender a little to refer to a special
example of the category which is very relevant to the speakers, as in the
following table. “Very relevant” is of course a little bit subjective, and in the
end largely assumes that the speaker has allegiance to Klzdmk.
37
Technically, the root with zero letters in it is not exactly the same as the a root omitted
via the empty root rule, even if it’s using the prototype rule. However, the end result doesn’t
make a meaningful difference, so we conflate the two here.
192 CHAPTER 16. NOUNS

CAT* Meaning of CAT* Meaning of CAT*-PROT-tʲ

dd Planet Pasaru
k Country Klzdmk
y Nobility The King of Klzdmk
mᵘ Generic area Apurhagat

This can be highly ambiguous:

(33) a. k-Ø-tʲ
F[country]-PROT-CITE
Klzdmk
b. k-tʲ
:[geography]-CITE
Geography

Despite this, it’s not considered to be a big enough problem to need

solving. In any case, these highly abbreviated forms are only used sparingly,
such as casual references to the relevant objects, e.g.

(34) k-Ø-sˢ k-plⁱt-tʲ

F[country]-PROT-LOI :[geography]-water-CITE
Klzdmk’s waters

With regard to ktʲ specifically, the name of the country klzdmk can be
understood as having the root .lzdmk., with the CAT STAR here providing
the missing k. This pattern holds for many country names that surround
Klzdmk (e.g. the J.-Senlis’s name is krⁱtᵗsˢ, which can similarly be under-
stood as the root .rⁱtᵗsˢ. with the CAT STAR playing the same role) as well
as any place names that coincidentally have a k as the first consonant but
this is not easily generalised to all place names in the world.

16.7.3 Omission of suffices

Although technically considered required, the suffix can be 0 or 1 long when
the gender is a CAT* rather than the usual trio, and the case where the
suffix is 0 characters long is essentially an omission.
Suffices can be omitted in that case primarily for visibility reasons: in
order to make sure that names can be identified from both the front and the
back, things that are common to all nouns are omitted.
This means that in maps, where the suffix tʲ would normally mark the
end of every word in the citation form, no longer have it, saving precious ink
and space. Indeed, this is one of the primary motivation of the omission:
reduction of redundancy in certain contexts.
16.8. MULTIPLE-ROOT NOUNS 193

16.8 Multiple-Root nouns

It’s not very common, but a single noun can sometimes contain multiple
roots. This is a particular result of several possible mechanisms, which we
will outline below.
The first and most important thing to note is that there can only be
one root that can have its genders; all other roots must have its genders
stripped. In a sense, this is equivalent to saying that the gender goes with
the word, not the root.
To get around this one can have it so that every word has the same
gender so it makes no odds if all but one of them are stripped. In this case
the gender is shared amongst all of them, and this is generally a desirable
thing to happen.

(35) a. [Link]ⁱt.tʲ water (for drinking or cooking)

b. [Link]ʲ.tʲ oil
c. ďʲ.svᵉ.tʲ wine
d. [Link]ʲ.[Link]ᵉ.[Link]ⁱt.tʲ cooking fluids

Otherwise, the internal word that would have its genders kept for the
word as a whole is the word’s “head” (of “-marking” type). For example, in
(35), we have combined three roots into a single word, with the third one
becoming the head word.
And this brings us to the various reasons why one would join roots like
that: to create a single word that has the combined meaning of all three. In
(35), the combination is comitative, where the word just means what any
of the components mean put together. This is very common but it is rarely
exhaustive due to the trickiness of defining a category by enumerating its
members. Hence, the usage of y.
Another reason why multiple-root nouns might be required is to indicate
an overlap of classes of objects that the words describe:

(36) a. kʲ.mᵘď.tʲ car

b. n.mᵘcd.tʲ driver
c. kʲ.mᵘď.cᶜ.mᵘcd.tʲ driving

To do this, the individual words are suffixed using cᶜ. The example
above actually show a common usage of cᶜ: combining a profession and an
inanimate object that is associated with it to create a word for the action
that someone with that profession would do.
These combinations are commutative – the only reason why one order of
words is used over another order revolves on historical or idiomatic reasons,
194 CHAPTER 16. NOUNS

and has to be learnt separately as an individual rule. However, there are

other affices that don’t behave the same way.
One of the most common examples of things in this type is the en-
docentric suffix, ħ. Here the two roots are combined together in an “of”
relationship, like so:

(37) a. t.rⁱ[Link].tʲ joining

(Note that this word is derived from a verb.)
b. ðʲ.bkᵏd.tʲ corporation
c. [Link].tʲ atom
d. [Link]ᵏd.ħ.vdtlk.tʲ corporate merger
e. [Link].ħ.vdtlk.tʲ nuclear fusion

(38) a. n.tʲ man

b. tʲ.tʲ king
c.
n.tʲ.ħ.tʲ parliamentarian
n- tʲ- ħ- Ø- tʲ
:[unit]- king- ENDC- unit- CITE
d. *n.ħ.tʲ.tʲ
n- Ø- ħ- tʲ- tʲ
:[unit]- unit- ENDC- king- CITE

Although the situation in (37) is the normal situation, the slightly rarer
situation in (38) show that the two arguments cannot be swapped with each
other. In the case of (38), we show that this is the case even if the two
roots are the same – and even if they are empty! This is also exhibited by
any general pair of two nouns: when they both are independent entities,
swapping them can create two different words, or in the case of (d) an
accidental gap.38
There’s a point to be made about these words being ambiguous with
the internal words seen in section 16.6.2. The ambiguity is not considered
important enough to resolve explicitly other than by context.

16.9 Unit names – the full story

Names for individual units are very complicated, with large amounts of
structure and ability to discard that structure depending on various factors.
38
Notice also that the empty roots become visible again – and are in fact just the gender’s
form – when the root no longer matches the gender.
16.9. UNIT NAMES – THE FULL STORY 195

(23 | 30) This highly structured format makes it very similar to nouns as

well as the complicated name structure of most Western names, but the
Drsk unit name is a single “word” at least phonologically, and it’s legally
understood to be a single word as well.
One thing to note here is that names form a significant structure, and
it’s reused in some other places related to names (see section 16.11), so it is
considered part of the language, though at some point it may be considered
syntax. Names for units are generally given a Drsk treatment even if they
are foreign, which then justifies its inclusion in this grammar.39

16.9.1 The Four Components of Unit Names

Recall figure 16.2. There are four individual components in a name: (23 | 29)

Status One of 44 two-letter combinations that represent the status – as in

social capital – of the individual relative to that of the speaker. This
list is fixed and no more can be admitted. It is allowed to be unstated
however.

Origin Two letters that quickly describe where the named individual comes
from. With just two letters and large number of possible places, no
single place can lay exclusive claim to a pair of letters; instead it must
share with a large number of other places. It is generally desired that
places close to each other don’t share the same letter pair though. As
new place names are made all the time, the pairs are constantly being
reused for them.

Commune A generalised version of a family name that’s three letters long.

Instead of just blood relationship though, the commune largely consists
of a circle of friends that all know each other as well. The result is that
a bunch of families are all put together into the same commune and
therefore share the same name. There is a large number of names and
more are made all the time, but there are also a number of commonly
used names as well as

Unit Five letters are concatenated together. There is no overall structure

in the language, but individual subcultures and language communities
may impose additional structure to what each position means.
39
Natives and nationals are normally given by parents and are mostly unregulated – they
know most of the rules already, even if only subconsciously – but foreigners that learn Drsk
must seek the expertise of a name-giver in order to receive their Drsk names, which is socially
mandated and individuals must have one to interact with Drsk society in general. Name-
givers are a legitimate if somewhat sparce profession and they do client work as well as
convene to produce native Drsk names for famous and dead foreigners that gain relevance
in Klzdmk.
196 CHAPTER 16. NOUNS

These are concatenated together as in figure 16.2, which in particular

state that the individual’s name (the unit) is escaped using delimited es-
capes. Only the unit is necessary; the rest are optional and are usually
omitted after the first use.

16.9.2 The status

While the Origin and Commune parts are more or less straightforward,
Statuses are much more complex and involved solely by the fact that they
are so limited. The fixed list of 44 statuses are various combinations of three
factors:

Fitness An abstract value that represents the individual’s health and abil-
ity to do actions in general. There are seven possible values for fitness,
which are numbered from 0 to 6.

0, Destroyed the named individual is dead or otherwise permanently

incapacitated.
1, Disabled the named individual is heavily wounded and cannot
perform much physical action but is still alive and capable of
social interactions.
Incapacitated fitness values 2 and 3 work in combination. The two
both mean that the named individual is wounded in some manner
that’s visible and debilitating, but not yet lethal or permanent.
The fitness value 2 is used when the individual is newly wounded
(“deteriorating”), whereas 3 is used for when the individual is
being healed from a lower fitness value (“recovering”).
4, Busy or Unavailable the user is in good health and mind but
is preoccupied with some event or action that makes them tem-
porarily unavailable.
5, Healthy the named individual is in “nominal status”. There are
no special conditions attached to this fitness.
6, “Shining” the named individual is doing better than normal, and
is considered to be “shining”, or in his prime. It is normally used
to denote the individual is in a state of heroism or momentary
greatness which does not last for much longer than a day.

Engagement The ability for the individual to meaningfully interact with

society at large, loosely defined. This is a binary option: engaged or
disengaged.

Caste Caste is an expression of relative power between the speaker and the
named individual. There are four primary values, plus a fifth neophyte
value that is seldom used.
16.9. UNIT NAMES – THE FULL STORY 197

This level exists regardless of the actual system in use in locations with
different or nonexistent caste systems, and is fairly stable through
time. In particular this means that this caste system doesn’t fully
reflect caste systems in Klssrdmk at any one point in time, only an
anachronistic combination of all of them.
AS, Absolute Superior Status used for kings, queens and individu-
als representing the Sun (Setura), the two moons, and the rest of
the planets orbiting the Sun. The list of individuals in this caste
is definite and fixed, with a list promulgated with the language
description every so often.
RS, Relative Superior Those in position of authority and demands
respect, as well as those that are mentioned first in a conversation
or other utterance, individuals named with the CAT STAR of m.
E, Equal This is a more radical caste that is usable for anything. It
is used to hide the social relationship between the speaker and
the named subject, but using it is tricky because it is also used
to the exclusion of other castes.
This caste is unique amongst the rest as obviously being from a
particular time period close to the present day.
RI, Relative Inferior The general purpose caste, used if no other
category applies. They are also used for those more junior than
others, as well as those that are not relevant to the current con-
versation, outsiders of any type, names for animals and other
inanimate units that receive unit names, and those that are men-
tioned later.
AI, Absolute Inferior Non-units that receive unit names, any dead
and formerly alive units, intangible objects (except those that are
deemed deities), and inanimate objects that receive unit names.
There is a fixed list of those that are in the AI caste but no such
list can be created physically because the number changes greatly
in a short period of time, and more are always added quickly.

Fitness and engagement are combined to form eleven possible states,

where certain combinations are judged impossible:

Fitness
6 5 4 3 2 1 0
Engagement
Disengaged IV V X XI
Engaged I II III VI VII VIII IX

And the eleven possible states are combined with the five castes to form
the 44 possible statuses, again with the rest judged to be impossible:
198 CHAPTER 16. NOUNS

Caste
AS RS E RI AI
State
I mᵘm sp ks ds
II mᵘr sł ył ðł
III sd kd dsˢ
IV pď sˢď yď ðď hs
V pb sˢb kb db hb
VI pf svᵉ yvᵉ ðvᵉ
VII sþ kþ ðþ
VIII mᵘł stˣ ytˣ ðtˣ
IX sŋ kŋ ðŋ
X sˢč yč ðč hč
XI pč sˢcᶜ kcᶜ dcᶜ hcᶜ

The “impossible” combinations are not literal, but they are real in a sense
that they have justifications for their existence. For example, a shining fit-
ness always implies engagement in society, as being shining means that the
individual is filled with life and heroism, which necessarily requires engage-
ment with society, so fitness 6 cannot be paired with disengaged. On the
other hand, those that are dead can still be a significant legacy to the society
around them, which makes them engaged – hence the existence of a fitness
0 engaged.

16.9.3 The Origin and the Commune

These two name components are fairly straightforward pieces that broadly
describe which geographical area and societal group the unit belongs to or
comes from respectively.
The geographical component is the “origin” component. Despite its
grandiose name its length, being only two letters long, means that it’s not
so much an indication of geographical origin as just a code to that origin.
Many places can share the same letter pair, and it is essentially a matter
of convention that places that are “not too close together” don’t share the
same letter pair. For example, the origin for the centre of Trokier is tg, but
so is a district in Nffek named tknᵘg. This part is very similar to the system
of telegraph codes40 used for Japanese train stations.
Unlike telegraph codes however, there is no requirement that the geo-
graphical area has to be a particular size or take a specific form – anything
that has any sort of relevance whatsoever can form the basis for getting
an origin letter pair. There are also places that have codes that seem to
40
[Link]
89%84%E9%81%93) (link in Japanese)
16.9. UNIT NAMES – THE FULL STORY 199

have little to no relation with any names or even previous names. These are
generally used because all more appropriate letter pairs have already been
taken by nearby places.
Furthermore, these letter names can change, appear and disappear as
their corresponding geographical locations do the same, with place names
appearing, moving, splitting, absorbing and disappearing under various cir-
cumstances. In general the origin is not a particularly useful guide to deter-
mining the actual ancestry of any unit.
This brings us to the other component, the commune. The word “com-
mune” (nsktʲ) is a translation of a collection of individuals that outranks a
family but is in turn outranked by a neighbourhood, or perhaps a village
district (the three are similar enough that they can outrank each other in
accordance to the principles of unorderable hierarchies (see section 3.8). In
any case, three characters is normally enough to make a whole root, and
that root is generally something meaningful, kind of like a surname of sorts.

16.9.4 Backslashes and the Unit Name

[In this section, “name” refers only to the unit portion of the full name.]
Conventionally in the Romanisation, names are written in all-caps, e.g.
PRNÐLˡ, SLDᵈSʲŁ.41 However, this is an artefact of rules with the backslash
character in the Rattssaw orthography.
In that orthography, the entire name is surrounded using the multiple
escapes – \[prnðlˡ\], \[sldᵈsʲł\]. This is because there is no single letter here
that is read as its name, and nor does it mean that it is treated specially
(with some exceptions below), which is what single escapes imply.
Sometimes individuals are named after a single letter or two, and as
in this case there are two possible pronunciations, here displayed with two
examples, and compared with the conventional, multiply-escaped version:

(39) a. \M = my /m/ [mə]

b. \M = md /md/ [mɨd]
c. \[M\] = my /m/ [mə]

(40) a. \M\R = mr /mɹ/ [maɹa]

b. \M\R = mdrl /mdrl/ [mɨdrɨl]
c. \[MR\] = mr /mɹ/ [mɨɹ]

41
Strictly speaking the superscript letters should also be uppercase but because of technical
limitations this cannot be done.
200 CHAPTER 16. NOUNS

This is deliberately ambiguous, as names are wont to be in places influ-

enced by the J.-Senlis.42 Note that there are additional differences arising
from the multiple escapes, which is only visible with names.

16.10 Place names

Place names are treated differently from unit names, which is pretty stan-
dard across most languages of Pasaru. As with most Earth names, they
are short and mostly unstructured. In Drsk, they are sometimes likened
to verbs in an informal manner,43 but keep in mind that unlike Drsk verbs
place names do have some small amount of structure. The structure is usu-
ally hidden away.
Place names are divided into two types: native, and foreign. The two
types are distinguished mostly using lexical clues, and aren’t just an ortho-
graphic quirk. The difference largely revolves around what is or isn’t allowed
in a particular name.

16.10.1 Features common to all place names

Place names are nouns, grammatically speaking. This means that they have
a gender and may take on a suffix. Unlike common nouns and like other
names however, this suffix is allowed to be omitted as per section 16.7.3.
Place names are free to exceed the 5 character limit that most nouns
have, but this is rarely needed. Additionally, they are normally not escaped
at all aside from where ordinary words are.
Place names are in writing always marked with a place name marker (see
section 13.6.2), which as explained in figure 13.8 is in the shape of a super-
script X, which one can remember as “X marks the spot” for a mnemonic.
This place quote is occasionally pronounced as an exaggerated [nː] which
is phonologically its own word; if used it normally does have any vowels
inserted adjacent to it. This extra marking is more common in the more
southern language regions because nearby languages there tend to also use
it.
Finally, place names are always marked using CAT STAR, and there are
some dedicated CAT STARS for place names as well. Suffices are omitted
if it would otherwise be t or tʲ, unless otherwise required to due to, e.g.
emphasis.
42
Each client state manages that ambiguity in different ways, subject to the individual
quirks and parameters of the substrate culture.
43
In section 23.1, the rationale of this categorisation is revealed. Essentially, sometimes
the words “noun” and “verb” get a bit mixed up with individual properties nouns and verbs
have. For instance, “noun” is associated with fixed, structured and composability, whereas
“verb” is associated with fluid, unstructured and atomic. This is not necessarily a linguistic
thing, just a cultural thing.
16.10. PLACE NAMES 201

16.10.2 Features exclusive to native names

“Native” names, due to historical reasons hinted at earlier, include in some
part place names in the J.-Senlis as well as names in Klzdmk. Such place
names are well-worn and tend to be fairly familiar to most Drsk speakers, as
well as conforming to extra conventions that foreign names need not follow.
The primary thing is that place names that break the character limit
tend to do so in a specific and limited way, such as those that also represent
dates. Native place names, with a few exceptions, draw from a stock of
vocabulary that is a strict superset of the normal stock. This vocabulary is
actually older and mostly historical in nature, and contains many roots that
have been lost in the root collapse events in historical times.
For those place names that do follow this particular pattern, they gener-
ally take one or two words from this stock which are relevant to the place in
question. The mean number is about 1.2 words taken, with more than half
of these taking only one word. For example, the old name for the capital,
trgr, is understood as:

(41) tr gr
[Link] palace

as it is in fact the seat of the government. (The new name, as mentioned

earlier, is a back-formation of the Rattssaw interpretation of this name.)
Other examples include:

(42) a. n fᵘk
end cape
The cape at the end of the land
b. smᵘ sˢv
granite marble
(The place where there is) granite and marble
c. ðvrkⁱ
wind
The windy place
The modern word for “wind” is:
sˢ- b- nt- tʲ
:[crops] :[weather] flow CITE

The prevalence of these names form a rough area from which a place that
can be labelled “origin of the Drsk language” can be roughly determined.
This area is roughly…
The [nː] that precedes place names are especially more prevalent amongst
native place names, due to how it better fits the Drsk phonology.
202 CHAPTER 16. NOUNS

16.10.3 Features exclusive to foreign names

In comparison, foreign names are much less restricted. Many of the place
names outside the native lands, Klzdmk in particular, are more than 6 char-
acters long, even after removal of vowels required by the language’s orthog-
raphy (they remain observed in the phonetics, though some of them are
stripped.)
One of the quirks that foreign names have that native names do not is
that they can and sometimes do include special escaped punctuation marks
as part of the name, which is a feature that is shared with some unit names.
We can have, for instance: tnᵘp\ srp, or consider the name of the Esħ Pez:
sh\ pz. This happens quite frequently when a foreign name has multiple
words. As it’s required that all place names are only one word long, if a
foreign name has a word separator it has to be escaped.
Fictional place names and place names in more “exotic” places (mainly
referring to the areas in equatorial/eastern/mainland Siťhanor) tend to take
this to the next level, often escaping not just punctuation but also digits
and even non-letter characters, such as:

• \9fkⁱl

• nt\!lⁱł

• zvᵉl\■

Such place names are typically pronounced in a haphazard way – \9fkⁱl

for instance usually has the escaped 9 pronounced gᵍħmᵘtᵗ, i.e. the pro-
nunciation for the number “nine”, even though technically it represents the
phoneme /χʷ/.

16.11 Other names

Other objects are occasionally given names, which may look like either place
names, unit names, or something else entirely. They may or may not have
internal structure, and they may or may not have some external regulation.
Foreign names are sometimes included here, but also sometimes included as
unit names.

16.12 Name-based pronouns

Most names have the peculiar property of being their own pronouns – not in
the literal sense where they cannot be replaced by pronouns, but in that their
morphology defines what pronouns they use. (34 | 17) This is independent
of the method normal nouns create their pronouns as described in section
16.7.2 and has an equally independent origin.
16.13. COLOURS 203

This form of making pronouns is simple: simply take the last letter of
the name, and that’s the pronoun. Taking the last letter of the name avoids
nasty problems with omitted sections of the full name. For example:

Name Pronoun
\[ŤSRGÐ\] \Ð
\[MTRMᵘL\] \L

These pronouns are pronounced normally: just as the name of the letter.
There are some exceptions to this relatively simple rule, mostly relating
to “anonymous names”. Where there is no particular name that can be used
anywhere, we will be forced to give our own, and normally that name is
usually just “unit” ntʲ, as a name: \[NTʲ\] (35 | 77) (usually case changes
still apply, so in transitive sentences this becomes \[NT\] and so on). This
particular pronoun abbreviates to \N, taking the first letter instead.
Since these pronouns only have one letter, they qualify as particles as
defined in section 21.4. This means that they are treated somewhat specially
compared to their names.

16.13 Colours
16.13.1 Single-Parameter colour naming
Unlike many languages in Pasaru colour terms, where the colour-space is
two-,44 five-45 or even non-dimensional,46 the terms in Drsk are one-dimensional,
and two words exploit this particular quality – essentially, they act as “to-
ward one colour” or “toward another colour”. Because the fundamental struc-
ture of the colour space is different relative to many languages, these terms
are not very easily translated.
The one-dimensional nature of the of the Drsk colour model is, however,
a little bit of a lie; while all the words lie conceptually in a line parameterised
by some number 0 ≤ t ≤ 1, the line itself snakes itself in a specific three-
dimensional space, which are hue (understood as wavelength and therefore
symbolised λ), transparency (symbolised α c.f. “alpha channel”) and in-
tensity (symbolised I).
44
Most languages in the Edrensano Ocean explicitly mark colours as part of a two-
dimensional grid.
45
Egonyota Pasaru in particular uses a five-dimensional colour space. Notice that for these
high numbers of dimensions, the colour terms are sometimes marked off as special points in
that space, and the only sense of the number of dimensions is that there are certain words
that mean that it is changing in that number of dimensions. For Egonyota Pasaru’s case this
isn’t what happens; there are genuinely ten colour terms, two for each dimension. (4 | 28)
46
Non-dimensional space is the same as n-dimensional space, where n is the number of
fundamental colour terms; the colour space is left unspecified in terms of dimension or
metric and specific points are marked out and named – “primitive names”. Other colours are
normally formed by merging two primitive names together.
204 CHAPTER 16. NOUNS

Using these three dimensions, t and suitable positive scaling factors A,

B, C, D, E, F , G and H, the line snakes across the colour space as approx-
imately:

α = −At + B (16.1)
(
C−F
λ= 3 −D u≤0
(16.2)
(Cu − D) cos (Eu) − F u u≥0
I = G(t − H)2 (16.3)
where
1
u = t − , and so (16.4)
 3 
1 2
u∈ − , (16.5)
3 3
The large amount of parameters is an issue, but a good deal of these
(e.g. A, C − F , G) can be eliminated by the correct choice of units.47 Such
a correct choice is difficult to make however so they are left undefined here.
More practically, note that this line is not space-filling; there is no way
that moving along the line can express every colour, which is why we char-
acterise the one-dimensional nature as “a little bit of a lie”. Nevertheless,
colours are still named and understood as being at or near particular values
of t. An example of using the parameter is expressed in Table 16.4.

Table 16.4: An association of the parameter t to colour words. The param-

eter is multiplied by 6 to avoid fractions.

6t Drsk English
0.0 ŋᵘd Transparent
1.0 sf Translucent
2.0 rⁱč Black
3.0 gh Red
4.0 vⁿh Sky blue
5.0 dkⁱ Bright green
6.0 sł White

Equations 16.1 to 16.3 only represent one of a small family of possible

ways the line can “snake” through the colour space; there is no actual def-
inite line, as the colour terms strictly define a volume in the 3-dimensional
47
What “units” are the three axes in is never really clear; they are normally considered
non-dimensional, thus we can set A = G = 1 without loss of information. However, this can
be a little tricky because there’s no guarantee that any λ, I and α are in [0, 1]. In particular,
λ can be denominated in units of length which will then be the wavelength of the number.
16.13. COLOURS 205

space, not single points (see section 16.13.3). A point of contention is what
happens at 0 ≤ t ≤ 13 , i.e. the word rⁱč – where the line would normally hit
black – in terms of λ, as the typical diagram usually terminates things here.
The extra-complicated definition of λ(t) can be off-putting, especially
the fact that it is defined piece-wise in equation 16.2. An alternate formula
is:

λ = (C |u| − D) cos (E |u|) − F |u| (16.6)

where all variables are as defined previously.

16.13.2 Transparency and Background

Introducing transparency means introducing what goes “behind” the trans-
parency. This additional factor is not specified by the language and is free
to vary across different contexts. (23 | 32) See Table 16.5.

Table 16.5: A list of background colours (what goes under “transparent”)

compared to various contexts.

Context Background
Painting White, none
Describing objects None
Shining light Black

Note that transparent is considered a colour in Drsk, so there’s no con-

flict with something not having a default background.

16.13.3 Outside the line

The three-dimensional nature of the ambient colour space in Drsk (ambient,
as in the environment that the one-dimensional line snakes its way around)
can be accessed in a variety of ways.
But first, here’s a way that we cannot escape the line: the two words
pnd and vᵉnᵘs are the modifiers that are used to navigate along the line,
essentially making them essentially δt and −δt respectively. Because of that,
these two words, when suitably fit with genders and suffices, cannot name
colours that are outside the line. However, they can be used to refer to other
colours that are on or nearby the line.
There’s actually no special grammar needed to create new colours. In
this case the whole thing is just simple noun-noun compounding, (that is to
say, the system described under section 16.8) using the COMC (i.e. cᶜ) fit
again with a gender and a suffix. In this case, then, the new colour is just a
simple mean of the two, at least to a simple approximation:
206 CHAPTER 16. NOUNS

 
λ1 + λ2 α1 + α2 I1 + I2
COMC : (λ1 , α1 , I1 ) × (λ2 , α2 , I2 ) 7→ , , (16.7)
2 2 2

There are also two other roots that can be used to change the colour:
you can directly change the value of two of the three parameters directly,
using the quasinouns below, joining using POI:

bsˢ POI : (λ, α, I) 7→ (λ, α + δ, I) (16.8)

rlˡ POI : (λ, α, I) 7→ (λ, α − δ, I) (16.9)
npnᵘ POI : (λ, α, I) 7→ (λ, α, I + δ) (16.10)
nzvᵉ POI : (λ, α, I) 7→ (λ, α, I + δ) (16.11)

where δ is a small increment appropriate to add to or subtract from the

other quantity.
Note that although we have said that a particular colour word represents
a point (λ, α, I), this is not exactly correct as, as with all natural languages,
there is a natural spread in how words are defined and there is no good
precise definition that can be cross-dialectal. Strictly speaking then, a given
point is actually a sphere around that point and that propagates further
through and compounded by other morphemes. This is normal, and is a
part of ordinary languages throughout the planet.
There is no way to adjust λ directly. It’s possible to use circumlocutions,
such as:

(43) ťg- rⁱč- mᵘ- vᵉh- t

::[colour] black POI [Link] CITE
Yellow (lit. “blue ‘infused with’ black”)

The actual meaning of the usage of POI here is a little bit obscure and
is mostly idiomatic. In any case, it is not generaliseable and not very pre-
dictable either.
The general method is actually to modify t, either by using a different
word or the words pnd and vᵉnᵘs which are actually separate words:

(44) ćtnpndl ťgghtʲ

ćtn- pnd- l ťg- gh-
:[abstracts]:[moving in abstract spaces] more.t LAT ::[colour] red
tʲ
CITE
Ochre (slightly into the near IR)
16.14. THE CALENDAR 207

and then the you can use the other morphemes to change the colour to
what is desired. In the end t ends up being the proxy to λ.

16.13.4 Regions as colours

Recall that in figure 3.4 we have used various colours to represent different
language regions. In the modern day, where the concept of colour is more
normalised towards those of Grade (3), there is now an additional require-
ment for the description of hues alone, not just hues mixed with intensity
and transparency. For those, the language regions themselves are used.
Let’s examine the regions and the colour that they correspond to:

Region Colour Colour (°)

Woróli Black/White –
Ntpeasdr Red 0
Trokier Orange 60
Iefiól Green 120
Bssreoml Blue-Green 180
[Link] Blue 270
Eamttel Purple 300
N. Nffek Pink 330

Keep in mind that the English translations are rough, and in particular
are compressed as compared to their actual meaning. With the eyes of a
kilis being able to see further into both the infrared and the ultraviolet, this
is inevitable. This also changes what the definition of *pink is.
As with any other situation involving gender transformation, simply pair
the root word – here the name of the language region – with the gender for
colour, ťg:

(45) ťgntpsdtʲ
ťg- ntpsd- tʲ
::[colour] ntpeasdr CITE
Red

An interesting note about this is that we are putting a word that would
normally be fitted with CAT STAR categories with those that are CAT
SMAJ.

16.14 The calendar

Drsk’s calendar is somewhat strongly tied to the language, but not exces-
sively. Nevertheless, there are some interesting grammatical features that
208 CHAPTER 16. NOUNS

are only used to describe time and time intervals. In this section, we will see
these special constructs here, as well as a simple demonstration of how the
general noun system works on a simplified domain of knowledge. (28 | 43)
A brief description of how the calendar works, independent of the lan-
guage, is available in section 2.1.4 and 2.2.4.
Like most order-sensitive constructs, calendar and time terms are es-
sentially large chunks of nouns in a specified order. Syntactically they are
considered noun phrases, and they can be expressed like so:
NP

DateP

DayP TimeP

Year Month Triad Sector Hour Degree

Num N Num Num N Num

where every instance of a final “P” means “phrase”. Note that we have
simplified the tree a little bit; it does not currently show the ways in that
each of the nodes can be omitted.
As the syntax tree shows, a date is essentially just a list of numbers, with
the exception of the traditional hours and months which are typically named.
Using the gender system described earlier, we can transform a bare number,
which normally uses the gender ćmᵘ (see section 17.1), into a number specif-
ically used for years, triads, sectors and degrees:

Thing CAT SMAJ CAT MAJ

Year sˢ dᵘc
Triad sˢ st
Sector sˢ ćp
Degree sˢ bl

Here, the CAT MAJs display their usage as bottom-up classifications,

bringing together small disparate things together without claiming totality.
Even though together they describe the entirety of their domain, the domain
is clearly smaller than :[crops] (which due to semantic drift also includes
time and time-related words) and we can also use the empty root rule to
determine that Drsk for “year” is sˢdᵘctʲ and so on.
Dates and times are the prototypical example of a noun compound –
when individual nouns or noun phrases aren’t enough to express a concept,
this is what the language resorts to. In a sense, such compounds are not
really all that different from other ways that nouns can be put together, such
as using quasinouns and CAT changes, and there is some overlap between
16.14. THE CALENDAR 209

this noun compound and those other methods. However, noun compounds
are normally considered their own thing. Sometimes they are considered
part of syntax rather than nouns themselves.
210 CHAPTER 16. NOUNS
Chapter 17

Numbers

Though they are nouns, numbers have such an unusually rich structure to
them that they deserve some description of their own. The primary features
that numbers have that most other words don’t include a limited number of
roots to describe a large number of digits, and therefore substructure inside
a root; one of the largest hotbeds of compounding and noun phrases in the
language; and a tantalising flirtation to some of the deeper maths that is
just barely out of reach with the syntax.

17.1 The digit state machine

The digit state machine is a concept that is used to name the digits in Drsk
using fewer root words that would otherwise be needed to name them. Using
only four roots, Drsk can name all twelve digits from 0 to 11. (23 | 12)
While in practise the system is ignored and all digits are learnt separately,
there is a system that is obvious and widely understood, and to an extent
is already part of the culture that speaks the language in the form of the
game of Digitex (see section 4.7.1). It revolves around the “state machine”,
which can be conceptualised as a machine that can hold one number and
processes “commands” one at a time.
In Drsk, this machine starts with the number 2, and understands the
following commands:

Double (D) on this root, double the number that is held.

Triple (T) on this root, triple the number that is held.

Decrement (E) on this root, subtract 1 from the number that is held.

Two (2) on this root, mark the digit as being complete, if there is one
already. The state machine starts anew, with the number 2 filled in
already.

211
212 CHAPTER 17. NUMBERS

These roots are chained together, right to left, to form a new state ma-
chine.
As an example of how they work, consider the root combination ETD2.
We can work it out explicitly:

Commands remaining State Notes

ETD 2 (initial state)
ET 4 D2 = 2 × 2 = 4
E 12 T4 = 4 × 3 = 12
 11 E12 = 12 − 1 = 11

So the string ETD2 represents the digit (or number) 11. It should be noted
here that at this point, the only root that represents an actual number rather
than a process is 2.
In Drsk, the commands form “roots” that are actually substructures of
real, lexical roots. So, these command roots are only one letter long:

E T D 2
ħ mᵘ gᵍ tᵗ

And they combine to the lexical roots, which are assembled as such:

0 EE2 ħħtᵗ 6 T2 gᵍtᵗ

1 E2 ħtᵗ 7 EDD2 ħmᵘmᵘtᵗ
2 2 tᵗ 8 DD2 mᵘmᵘtᵗ
3 ED2 ħmᵘtᵗ 9 TED2 gᵍħmᵘtᵗ
4 D2 mᵘtᵗ 10 DET2 mᵘħgᵍtᵗ
5 ET2 ħgᵍtᵗ 11 ETD2 ħgᵍmᵘtᵗ

To form an integer, one merely needs to concatenate the digits in almost

the usual manner. The only difference to watch out for is that now tᵗ is
considered a “special suffix” which separates digits, and it disappears when
a real suffix occurs at the end of the word.
Numbers have the gender ćmᵘ, i.e. :[abstracts]:[number]. To create the
number word, one must then combine this gender with the command se-
quence above, and add on the required suffix as grammatically appropriate.
This is already enough for arbitrary integers, as in these examples:

(46) ćmᵘħmᵘtʲ 3
ćmᵘ ħmᵘ(tᵗ) tʲ
::[number] three CITE
17.2. ORIGINS AND COGNATES OF THE DIGIT STATE MACHINE213

(47) ćmᵘħtᵗħgᵍtᵗħmᵘtᵗmᵘtʲ 153412 = 2488

ćmᵘ ħtᵗ ħgᵍtᵗ ħmᵘtᵗ mᵘ(tᵗ) tʲ
::[number] one five three four CITE

17.2 Origins and Cognates of the Digit State Ma-

chine
The digit state machine is a feature that can seem rather bizarre to those
unfamiliar with its comings and goings. For some it might come off as an
anachronism, as if one has learnt of how numbers work in a way that isn’t
simply learning of them one after another – maths before counting.
Nevertheless, no such anachronism is necessary, and the existence of the
state machine can be explained by simple if somewhat unlikely generational
change over the previous millenniums. The change is somewhat regular and
actually a little bit interesting, so we will detail how this digit state machine
is formed from a simple one-root-per-digit system, and how we can track
how alternate versions of this system has cropped up in nearby languages.
In this instance, the creation of the digit state machine requires two
things: the destruction of the old digit system, as well as destruction of
a more elaborate number system beyond the number 12. (34 | 24) The
historical versions of the morphemes are unclear in this instance and are in
any case variable over such long periods of time, so instead of referring to
them by their surface value we will represent them by symbols.
The main thing that is needed to create such a system is to replace the
raw root with a combination of other roots, some of which may be new. In
the case of Drsk, what happened is roughly as such:

1. A new morpheme E is created and replaces 5 and B with E6 and E(10)

(then still extant).

2. Even numbers are altered using “double-and-half”, e.g. the number 8

is replaced with 2-4.

3. The two created in the previous step is then replaced with another
word, represented by D, in all cases except 6 (which remains 2-3).
This is a slightly unusual exception, but it is needed to create the
digit state machine.

4. 9 becomes 33.

5. The common factor between 6 and 9 is 3. The morpheme 3 is then

altered to the tripling morpheme T, and this is transmitted into other
numbers that involve 6 (i.e. 5).
214 CHAPTER 17. NUMBERS

6. At this point, the digit state machine “comes to be” – speakers are
aware of the pattern created by the words, and that pattern has been
successfully formalised into education. In light of that, the remaining
morphemes that are independent are folded (“regularised”) into the
Machine.

7. The number 0 is now recognised and the digit 0 is added to the Ma-
chine. The old system of making large numbers is discarded in favour
of simple digit concatenation.

8. And finally, all the remaining numerical commands, such as 3 or 4, are

replaced with commands that only use 2.

In summary, we can chart the evolution of the digit state machine using
Table 17.1.
Regardless, the formation of the digit state machine is one of the less
likely events in the history of Pasaru languages, and there are some sister
systems that fail to produce a Digit State Machine or has one but is signif-
icantly less spartan (especially those that have more starting values that
were removed by step 8), which may show up elsewhere. This may be seen
in other languages that are close by geographically, or in specialised contexts.
One very critical evidence for this particular formation of the state ma-
chine is the existence of alternate but clearly related state machines that
are related but different from the one in standard Drsk. In particular, in
some more obscure dialects some of the older forms do show up in. Take,
for instance, the first six counting numbers of the Dsnptt dialect:

Number Std. Cmd. Std. Word Dsnptt Cmd. Dsnptt Word

1 … … … …
2 … … … …
3 … … … …
4 … … … …
5 … … … …
6 … … … …

Which is very clearly a variation of stage 5.

17.3 Different bases

Because of the distinction between a digit and a number, a change in base
can be a little bit more obvious than it would be otherwise. To do this, we
need to use the state machine to generate larger numbers. For example, to
express digits up to hexadecimal, then the digits 12 to 15 need to be defined
in the state machine. Here’s the standard way this is typically done.
17.3. DIFFERENT BASES 215

0 1 2 3 4 5 6 7 8 9 10 11 12

1 2 3 4 5 6 7 8 9 A B 10

E6 E10

22 23 24 2E6

D2 D4 DE6

33

ET2 T2 T3

E2 ED2 ED4

EE2 E2, EE2

EDD2 DD2 TED2 DET2 EDT2

EE E 2 ED D ET T EDD DD TED DET EDT

Figure 17.1: Evolution of the Digit State Machine.

216 CHAPTER 17. NUMBERS

12 TD2 mᵘgᵍtᵗ 14 DEDD2 mᵘħmᵘmᵘtᵗ

13 EDEDD2 ħmᵘħmᵘmᵘtᵗ 15 EDDD2 ħmᵘmᵘmᵘtᵗ

This can easily be seen to be grow very unwieldy as the digits go by, as
the commands get clumsy after a while – in the case of 13 for instance it tech-
nically exceeds the (soft) limit of 5 characters per root.48 As a result there
are some unofficial extensions that include a number of other commands,
such as tˣ “+2” (as a separate command from “2”). It is not common to in-
clude commands that replaces the state with a stack, so general commands
like addition or multiplication remains impossible.
Numbers in other bases still need to be marked for what base they are
in using a number with the alternate gender ::[base number].

(48) ćmᵘvħmᵘtᵗħħtʲ ćmᵘdmᵘtʲ 304 = 12 “three zero base four”

ćmᵘv ħmᵘtᵗ ħħ(tᵗ) tʲ -
:[abstracts]:[number]:[+[Link]] three zero CITE -
ćmᵘd mᵘ(tᵗ) tʲ
:[abstracts]:[number]:[+base] four CITE

However, some common bases do not require this suffix, as they have
separate genders. These are ::[binary], ::[senary], ::[octal] and more recently
::[hexadecimal]. Binary in particular has a special syntax that uses z and
þ to represent 0 and 1, respectively, and bypasses the digit state machine
entirely.
(In these glosses, the digit suffix no longer receives colour.)

(49) ćmᵘrⁱsþþzþzþtʲ 1101012

ćmᵘrⁱs þ þ z þ z þ tʲ
:[abstracts]:[number]:[+binary] one one zero one zero one CITE

(50) ćmᵘlfmᵘtᵗmᵘtᵗħtᵗħmᵘtʲ 44136

ćmᵘlf mᵘtᵗ mᵘtᵗ ħtᵗ ħmᵘ(tᵗ) tʲ
::[senary] four four one three CITE

48
In this case, it is fortunate that the final tᵗ is omitted to keep the actual number of
letters belonging to root below five, but it turns out this is not actually necessary as the
limit is a little bit of a soft cap.
17.4. LARGER NUMBERS 217

(51) ćmᵘlmħmᵘmᵘtᵗħmᵘmᵘtᵗħmᵘtʲ 7738

ćmᵘlm ħmᵘmᵘtᵗ ħmᵘmᵘtᵗ ħmᵘ(tᵗ) tʲ
::[octal] seven seven three CITE
It should be noted that the octal shortcut is the oldest one of the lot,
and is one of the reasonings behind the octal nature of the tap code.

(52) ćmᵘdᵈsħtᵗħgᵍtᵗmᵘtᵗħmᵘħmᵘmᵘtʲ 154D16

ćmᵘdᵈs ħtᵗ ħgᵍtᵗ mᵘtᵗ ħmᵘħmᵘmᵘ(tᵗ) tʲ
::[hexadecimal] one five four thirteen CITE

17.4 Larger numbers

It’s tempting to go off into the distance and discuss the fast-growing hierar-
chy and uncomputable functions, but we shall resist the urge to do so, and
only name numbers that use novel structure to express themselves.
The first thing to notice is that while initially just numbering the digits,
things get unwieldy quickly after the numbers reach the millions. A couple
of helper numbers, each with gender ðs. ::[large], represent large numbers,
such as the ones below.

126 -tml- 1224 -vᵉpl- 1242 -vtl-

1212 -vmᵘl- 1230 -vmᵘl- 1248 -kᵏdᵘl-
1218 -kⁱŋl- 1236 -kⁱŋl- 1254 -ŋᵘþl-

Alone, all these words mean various types of “very large number or quan-
tity”, or describe some mythical creatures of immense size (hence their classi-
fication into gender ð :[celestial]). However, if they are combined with other
numbers, they would take on the values as indicated. In some dialects, par-
ticularly around the Trokier and Urban language regions, the genders are
changed according to existing conversion rules to the normal number gender
ćmᵘ.

(53) 3150A4 1371B6 391200 = 6 918 859 249 166 771 424
3150A4 ðsvmᵘltʲ 1371B6 ðstmltʲ 391200
3150A4 *billion 1371B6 *million 391200
Here we have kept the numbers less than 1 000000 as numbers to
keep the gloss compact. The glosses for the larger numbers are *mil-
lion, *billion, &c., because they work in a similar way to what the
English word “million”, “billion” &c. would behave, even if they have
quantitatively different values.
218 CHAPTER 17. NUMBERS

This can work with conjunction with the indeterminate number to form
rounded-off numbers as well, if their values are above 126 .

(54) f 3000 ðstmltʲ ćmᵘtʲ.

and 3000 *million some
3 × 129 and change.

Keep in mind that as with all words with “and”, this is a complete sen-
tence, which results in some consequences regarding how quantitative ques-
tions are answered.
Finally, certain very large numbers are often given names, and these
names are genuine names in the unit sense of having five characters which
are all escaped. (Though, in the case of numbers, five characters is all they
have; there is no other part to their names.)

(55) ćmᵘRBMᵘSTˣtʲ 12144 (i.e. *googolplex)

(56) ćmᵘLTNDVᵉtʲ 121728

These names are not systematic, and they can also name very small
numbers or any other number as well. They are also not guaranteed to have
currency in all locales; some regions name some numbers differently.

17.5 Non-integers
When it comes to specific syntactic forms for non-integers, there are two
other types of “non-integers”: the real numbers, which are the ones with
decimal points or fraction numbers, and indeterminate numbers, which have
no definite value, but are considered to be real.
Indeterminate values are named in a straightforward manner. As for
non-integers that are definite, (which are typically restricted to the reals),
three conceptual systems handle them. These are the Egyptian, Vulgar and
Decimal systems. (23 | 33).

17.5.1 Indeterminate quantities

The empty root rule is used to name numbers that are completely indeter-
minate, i.e. have no approximate value at all. In this case, a word with
no root, only the gender and the suffix, is understood as an indeterminate
amount, unless context dictates otherwise. The word would then look like
this:
17.5. NON-INTEGERS 219

(57) ćmᵘtʲ “an indeterminate amount”

ćmᵘ tʲ
:[abstracts]:[number] CITE

The “indeterminate amount” definition conflicts with the general case

where a word with no root is typically used as a pronoun representing a
word with the same gender. In this case, it is up to context to distinguish
what it is:

(58) a. nt ćmᵘ-s ln dʲtt.

PRN:[unit] number-LON eat *fish1 .
He eats some *fish1 .
b. ćmᵘħmᵘtᵗmᵘt kmk rⁱmt. ćmᵘ-tʲ Ø ðlkᵏdtʲ.
forty stand square. PRN:[number]-TCOP COP behemoth
There are forty squares. That’s a large amount.
c. nkrstʲ kmk. ćmᵘ-tʲ Ø ðlkᵏdtʲ.
riot form. PRN:[number]-TCOP COP behemoth
A large number of riots occur.

Here, (58a) has a ćmᵘtʲ that represents an indeterminate amount, be-

cause the context (which is nil, in this case) does not have a number for
which to provide a potential pronoun to refer back do. In contrast, (58b) is
a pronoun, because a number was previously mentioned, which a pronoun
can then refer to. But the exact same sentence can be used in another
context to indicate that an unknown amount is nevertheless large, which is
demonstrated with (58c).
Although ćmᵘtʲ has no value, one can “suggest” a value to it by placing
it after a suitably-rounded word:

(59) A0.tʲ ćmᵘtʲ “about 120”

17.5.2 Egyptian fractions

Conceptually, a number N ∈ R can be derived by summing a list of integers
a0 , a1 , a2 , a3 , ..., an , ... such that
X 1
N = a0 + (17.1)
ai
0<i≤n

This is called an “Egyptian fraction notation”, as it is reminiscent of the

way that the ancient Egyptians notate them. It is not a mathematically
220 CHAPTER 17. NUMBERS

useful notation because the same number can created in many ways, even
discounting variations regarding the order of the sequence. However, it is
a fairly easy way to represent fractions, and it is the system that is most
readily used in Drsk.
The restriction that there all an are different (i.e. ai = aj ⇔ i = j)
is sometimes relaxed for certain fractions, especially those of the form n2 ,
where n is an integer. The fractions two-thirds and three-quarters are not
used however, with the former being represented by 3 3 and the latter 2 4.
In the pronunciation, we use a dredge pď EF, to be mixed into a sentence:

(60) 7 pď 2 3 17 f
7 + 2−1 + 3−1 + 19−1 = 7 101
114 ≈ 7.885

(61) pď 0 5 9 f
0 + 5−1 + 9−1 = 14
45 ≈ 0.311

In the case where there are an infinite number of integers, a finite amount
is listed and then the word ðnⁱ.dtdᵈ.tʲ “unlimited, infinite, and so on, endless
continuation” is used:

(62) 1 2 pď 4 8 14 ðnⁱdtdᵈtʲ f
1 + 2−1 + 4−1 + 8−1 + 16−1 + · · · = 2

(63) 0 1 4 9 pď 14 21 ðnⁱdtdᵈtʲ f
π2
0 + 1−1 + 4−1 + 9−1 + 16−1 + 25−1 · · · = 6

This is a more recent invention, as infinite lists were not discovered until
then.
As with the usual sentence behaviour, the nouns are fixed in place,
whereas the dredge is free to move as it pleases. The verb too is has no
fixed location, but as this is typically embedded into another sentence, the
single verb, f “and”, must occur at the ends of the sentence.

17.5.3 Vulgar fractions

We take the same principle with Egyptian fractions, but this time we use the
dredge sy VF. This time we can only accept three numbers, the integer part,
the numerator, and the denominator. The former uses the regular number,
but the latter uses a number with the modified genders for this purpose, i.e.
ćmᵘpð. ::[numerator] and ćmᵘph. ::[denominator].
17.5. NON-INTEGERS 221

(64) 6 sy ćmᵘpð.1.tʲ ćmᵘph.8.tʲ f

six VF one::[numerator] eight::[denominator] and
6 18

All three positions can be omitted, though not all positions are valid
and with some complex defaulting behaviour. If the integer part is omitted,
it is assumed to be 0; if the numerator is omitted, it is assumed to be 1;
and if the denominator is omitted, it is assumed to be 1728. This allows for
convenient *percentages to be written:

(65) sy ćmᵘpð.3.tʲ zvt f

VF three::[numerator] Ø and
3 per *mille

(66) sy ćmᵘpð.488.tʲ ćmᵘtʲ gᵍlmᵘŋstʲ f ttk

VF [Link]::[numerator] some game and win
The chance of winning this game is about 680 per *mille.

Because f absolutely requires two nouns, a dummy noun zvt is included

instead, which has some irregular behaviour, chief of which here is that its
suffix is always t.
There is some questionable behaviour with some combinations of omitted
values. Here are some ways that one can read the number 6 1728
1
, in increasing
order of acceptability.

(67) a. * 6 sy zvt f
b. ? 6 sy ćmᵘpð.1.tʲ f
c. 6 sy ćmᵘph.1000.tʲ f
d. 6 sy ćmᵘpð.1.tʲ ćmᵘph.1000.tʲ f

Here, (d) is obviously uncontroversial, if a little bit long-winded and

exacting. (a) is completely unacceptable, not only because it uses zvt to
satisfy f’s requirement of requiring two or more arguments but also because
there’s an unacceptable ambiguity – the phrase parses like it just means
“six”, without any fraction, but a careful reading shows that it doesn’t.
(c) is more or less uncontroversial as well, as it is modelled on other
completely acceptable phrases such as “two and a half” 2 sy ćmᵘph.2.tʲ f
or “five point one” 5 sy ćmᵘph.10.tʲ f, but can be a little bit jarring with
a very large denominator that is not a power of 12. (b) is understandable
but awkward, and is used in some limited contexts, such as economics:
222 CHAPTER 17. NUMBERS

(68) V dn f 6 sy ćmᵘpð.100.tʲ þtʲ

listener accept and six VF [Link] [Link]
You get six and a twelfth of a share.

(69) P dn f 6 sy ćmᵘpð.5.tʲ dʲtt

speaker accept and six VF five *fish1
I get six *fish1 and five mini-*fish1 .

and also in counting:

(70) sy pð.BBA zvt f, sy pð.BBB zvt f, 1 sy pð.0 f, 1 sy pð.1 f, 1 sy

pð.2 f

The latter of these shows an alternate way that the syntax displayed in
(67b) can be used: as a compact way to describe having a number of one
thing and another number of another, much smaller or diminutive, thing.

17.5.4 Decimal fractions

Decimal fractions are the most straightforward of the lot, as they can only
have two parts: the integer part and the non-integer part. These two are
treated as two separate numbers and joined together using “and” and the
dredge kł.

(71) kł 16 9B36 f
DF eighteen [Link] and
9 11 3 6
18 + 12 + 144 + 1728 + 20736 ≈ 18.828

Both numbers are in the standard gender, with no CAT MIN.

It is common for the decimal part to be broken up into groups of 4, which
distinguishes them from the integer part (usually broken into groups of 6),
along with the fact that they are not separated by words like *million and
*billion.
Two additional words are used to mean “continues with a pattern” kʲ[Link].tʲ
and “continues without a pattern” gᵍ[Link].tʲ. “Pattern” here means that
there is a simple description of what every digit will be, and not neces-
sarily just that the digits repeat after an initial period of no pattern. This
means that while irrational numbers will all have the word gᵍstvtmltʲ, some
irrational numbers will have the word kʲvtmltʲ instead, such as C12 , the
Champernowne number in base 12.
17.6. NUMBERS AND AFFICES 223

17.6 Numbers and affices

As can be seen from previous sections, when numbers are written as a se-
quence of digits, there can be some complications with the orthography if
when genders and suffices have to be included.
In normal writing, digits are separated from suffices and genders by the
dot, as mentioned in Section 16.4. For numbers, this remains true, but
because the digits are a different script, the dots remain in all times until
the number is spelt out.
Left omission of the gender ćmᵘ. is particularly common, but not en-
tirely universal when numbers are indicated using digits. Right omission is
generally not done unless the syntax makes individual genders clear. Total
omission is possible in as well, resulting in a bare root.
The omission of suffices is a phenomenon unique to numbers, although
it should be noted that this is normally only done in orthography and not in
speech (where it remains regular throughout). To do this one simply omits
the following suffices where one would normally appear:

• t

• ť

• tʲ

Combined with total omission of genders, this can produce the desired
effect of only having the digits appear with no letters to to indicate gram-
matical function.

17.7 Ordinal numbers

In order to create ordinal numbers, one uses inclusion. In other words, order
is created by making the number a “modifier” of sorts. For instance:

(72) 3 člmᵘďtʲ
three car
three cars

(73) čl-3.tᵗ-mᵘď-tʲ
::[traffic]-three-car-CITE
the third car

Notice in (73), the “suffix” tᵗ is restored because it no longer coïncides

with the genuine suffix.
224 CHAPTER 17. NUMBERS

Creation of standalone ordinal numbers is not straightforward, as the

natural way to form it – using ::[number] and the empty root rule – is not
used due to the uncomfortable coëxistence of the tᵗ and tʲ which would
usually be collapsed away. We in this case require the usage of the gender z
:[grammar] and the coälescence of the two ts can then occur naturally:

(74) z-3-tʲ
:[grammar]-three-CITE
third

Where this may conflict with other usages of z, the CAT MAJ kħ can
be used to disambiguate.

17.8 Quantities
There are three different ways that quantities are expressed in Drsk. The
exact mechanism used would depend on the nature of the quantity.

17.8.1 Ordinary quantities

For simple, dimensionless quantities, such as most count objects, angles and
proportions, the method used is to include it in an “and” sentence if already
present:

(75) 7 pď 2 3 17 ftʲ f
seven EF two three nineteen food and

114 units of some food

7 101

Or, if it is not available, put the quantity directly behind the number,
as in (72).
A number with dimension is treated very differently. First, we need to
figure out what a dimension is. Dimensions in Drsk are similar to dimen-
sions in physics – they are measurements that correspond to some physical
quantity, where ratios are easily measured but an absolute and universal
reference scale is hard to come by, if at all existent. In the case of Drsk,
the physical quantities are joined with monetary and financial quantities to
become dimensions as well.
For most of these objects, the correct syntax requires the “and” sentence
to be present, and certain “non-standard” units require the dredge þsn.
Invariably, non-standard means that it does not use the international units.
This system also applies to non-monetary financial quantities.
17.8. QUANTITIES 225

(76) 4 lˡ.fr.tʲ f
four ferā and
four ferā

(77) 4 þsn lˡ.pln.tʲ f

four QNT pppelen and
four *miles

Here, the standard unit is ferā, in the sense that “standard” means the
rough equivalent to the SI on Earth. In the other sentence, the unit “pp-
pelen” is used, which is a legacy (think “imperial”, hence the gloss) unit of
length common in the area – it originated from the Esħ Pez, further to the
northwest.

17.8.2 Monetary quantities

Monetary amounts are treated in yet another manner. This can be further
subdivided into three types: decimal type, non-decimal type, and national
type. The simplest type is the decimal type, which is treated as a dimen-
sionless quantity of a number that is customarily understood as a decimal
fraction. Keep in mind that a decimal-type currency typically does not have
a particular name for its subunit, which might be expected for, e.g. “cent”
to a “dollar”.

(78) 4 kł 151 ðʲ.Ø.tʲ f

four DF [Link] :[currency]-[Link] and
¤4.119

Non-decimal currencies are usually treated as a list of integers, and a list

of integers is fairly easily expressed as an “and” sentence. Because of this, it
is not generally necessary to specify that such currencies have a dimension,
as f is required for other reasons anyway.

(79) 9 7 5 3 ðʲ.Ø.tʲ f
nine seven five three :[currency]-[Link] and
¤9-7-5-3

Finally, the national currency is given an idiosyncratic system of its

own. This is because it is the only currency that is grammatically different;
instead of it being separate, the number is cast into :[currency], and it is not
a complete sentence as f is not involved. However, it does require escaped
letters, specifically \M and \Ħ.
226 CHAPTER 17. NUMBERS

Recall from section 3.4 that the structure of the native currency in Drsk
contains a large number of sub- and super-units. All of these have specific
abbreviations that use the backslash, which essentially indicate their status:

Name Symbol
ry ry
dʲgs \Mf
dʲħy \Mv
ŋrh \Ms
ħt \Ħc
tᵗħt \Ħtᵗ

Since all of the words except ry are at least partially escaped, those words
are given special syntax rules as escaped objects are typically uninflected (see
section 21.2). As such, the spoken versions of these words look very similar
to the written, symbolic version.
For instance, consider the sequence 1 Xt 5 - 72M2. This is a near-
maximal sequence of currency values, so it neatly demonstrates how the
notation transforms straightforwardly into spoken words:

(80) ðʲ1\Ħc5\Ms60\Mv2tʲ
ðʲ ħtᵗ \Ħc ħgᵍtᵗ \Ms ħgᵍtᵗħħtᵗ \Mv tʲ
1 ·Xt 5 Xt· 60 ·M 2
:[currency] one ħt five ŋrh [Link] dʲħy two CITE
1 Xt 5 - 72M2

To increase legibility two versions of the word are provided: one where
figures are used to replace the numbers, and one where they are spelt out
but spaces are used to separate individual parts of the word. In either case
it’s the pattern is still fairly obvious and simple.
The only thing to watch out for is that all numbers are mandatory, and
a denomination with no number next to it refers to a coin or note with its
value:

(81) ðʲ\Ħctʲ
ðʲ- \Ħc- tʲ
:[currency] ħt CITE
A 1 Xt note

As also remarked up on in that section, there is a number of local curren-

cies that complement the national one. All of them use the same structure
in terms of numismatics, so to no surprise they also use the same linguistic
17.8. QUANTITIES 227

system as well, with the only difference being the fact that they have a mod-
ifier. For most of the local currencies this comes as a noun compound, but
for larger or historical places a quasinoun is available and therefore used:

(82) a. The national currency:

ðʲ1\Ħctʲ
1 Klzdmk Xt
b. The Eamttel currency, using a quasinoun: ðʲðvᵉsy1\Ħctʲ
ðʲ- ðvᵉs- y- ħtᵗ- \Ħc- tʲ
:[currency]- Eamttel QN- one- ·Xt- CITE
1 Eamttel Xt
c. The Mskre currency, using an adjective: ðʲ1\Ħctʲ cⁱmskrsˢ
ðʲyħtᵗ\Ħctʲ cⁱ-mskr-sˢ
[Link] F[[Link]]-Mskre-LOI
1 Mskre Xt

In any case this is the same as the written form of the notation. (19 | 51)
228 CHAPTER 17. NUMBERS
Chapter 18

Verbs

18.1 Verb delocalisation

18.2 Affix localisation

18.3 Noun-Verb coupling

18.4 Crutches

18.5 Canes

18.6 Process, Product, Facilitator

18.7 Sporadics

18.8 Motion and Position

18.9 Conjunctions as verbs

18.10 Copula

229
230 CHAPTER 18. VERBS
Chapter 19

Dredge

19.1 Dredge delocalisation

19.2 Owned and ownerless dredge

19.3 Relational dredge

19.4 Purpose of dredge

19.5 Examples of dredge

19.6 Inflection of dredge

231
232 CHAPTER 19. DREDGE
Chapter 20

Tense, Aspect & Mood

20.1 Noun-like, verb-like and phrase-like markers

20.2 Tense

20.3 Aspect

20.4 Mood

233
234 CHAPTER 20. TENSE, ASPECT & MOOD
Chapter 21

All other words

Apart from nouns, verbs and dredge, there are a number of other words that
don’t fit into this categorisation. These are primarily words that consist of a
single, indivisible morpheme that have no particular affinity regarding fixed
positions within a sentence – thus they can be either “verb-like” or “noun-
like”.
Individually, most of these words are part of the “sundry” noun com-
ponent, which is a bit of a cop-out – that’s just saying that there are no
good categories to otherwise fit them to. Categorisation into this “sundry”
component however makes them converge to contain more or less the same
qualities as each other, which allows us to discuss them as a whole here.
We’ll consider these types of words individually in each section.

21.1 Prowords
The proword is ŋ.
The single, unique word that represents an arbitrary other word is used in
Drsk. This is very different from nearby languages, which typically restricts
such references to just nouns or just verbs, and also have multiple such words
that do the same thing.
This word, unlike other particles, can gain additional genders. If it does
so, it may become ŋr unpredictably. In this sense, it behaves similarly to a
hypothetical root, -ŋ-, which may have an alternate form -ŋr-, and doesn’t
otherwise exist. However, it is allowed to, and is used to handle indexing:

(83) (Foo ate Bar while Baz. Baz is unrelated to Foo.)

The proword usually zeros out common suffices as in section 21.4, even
in its alternate form of ŋr.
One thing of note is that prowords only weakly correspond to pronouns in
other languages. In particular, while the proword may replace any individual
word, there are already mechanisms in Drsk for turning any noun into a

235
236 CHAPTER 21. ALL OTHER WORDS

corresponding pronoun (see section 16.7.2). This means that the usage of
ŋ for such items is not necessarily needed (though it normally works as a
redundant pronoun in this case).
Even in the case of the first and second person pronoun, ŋ is still not
necessarily used for these because their purposes are also fulfilled by #p
and #v. Note that neither #p and #v are exactly first and second person
pronouns either, for reasons explained in section 21.4.
The proword normally cannot replace more than one word, but another
word, the pro-sentence (34 | 52) can. The status of the pro-sentence as a
lexical item is questionable, as its main usage is to join two independent sen-
tences together in a particular coördination strategy with the causal dredge:

(84) a. (Sentence 1).

b. (Sentence 2).
c. (Sentence 1 therefore sentence 2 using pro-sentence and connec-
tives).

But sometimes, the represented sentence can in turn have a representa-

tive word, which can be replaced by the normal proword:

(85) a. (Sentence 1).

b. (Sentence 2).
c. (Combined using pro-sentence).
d. (“Combined” using a representative word).
e. (Combined using a representative word that has been replaced
with the proword).

21.2 Escapes
The general idea of escaping terms in Drsk is when letters are interpreted
differently from what they are normally interpreted. This is very similar
to how certain programming languages deal with special interpretation of
certain glyphs, and we therefore use similar terminology and notation to
describe it. (23 | 55)
Escaping characters, on the notation level, has two different manifesta-
tions. The first one, called the simple or single escape, is simply putting
the backslash character before the character in question. The other type,
called the complex or multiple escape, requires a start and end delimiter and
surrounds a bunch of characters. The results look like this:
21.2. ESCAPES 237

(86) a. s
b. \S
c. \[S\]

(87) a. st
b. \S\T
c. \[ST\]

In the Romanisation, escaped letters are expressed in capitals, and cap-

ital letters may or may not subsume the backslash. This is primarily for
æsthetic purposes, and is not part of the B.-Rattssaw orthography. In this
book, the backslash is never subsumed by capital letters.
The notation is by far one of the easiest part of escaping to explain, with
how they are used and how they are pronounced being much more difficult
and nuanced. Let’s examine them individually and how they are considered
together.

21.2.1 Single escape

The single escape, as its name suggests, escapes a single letter or other glyph.
It is primarily used for named pronouns, which are caused by converting
a full name to an abbreviated version:

(88) a. \[STRⁱFN\] a name (prefices omitted)

b. \N

This single escaped letter is pronounced as its name, so in this example

the pronoun is pronounced /st͡ɹn/, just like the letter n would by itself. Such
an escaped letter is also used to name the letters, as such:

(89) \mᵘ għtᵗbsdtʲ gmᵘzrsˢ.

\mᵘ-Ø g-ħtᵗ-bsd-tʲ g-mᵘzr-sˢ
“Mᵘ”-[Link] ::[writing]-[Link] ::[writing]-alphabet-LOI
Mᵘ is the first letter in the alphabet.

Not just letters are available to be escaped. One of the more interesting
usages of the single escape is when the space is escaped, like so:

(90) dprⁱ\ bsˢt (a place name)

In this case, the escaped space removes the space’s original purpose to
separate words, and has become a letter of its own right. This is frequently
238 CHAPTER 21. ALL OTHER WORDS

used in place names, where foreign places that have spaces in their names
have their spaces escaped when imported to Drsk.
Non-letters that are singly escaped behave slightly differently when es-
caped. The escaped space, for instance, can be pronounced as the following,
depending on context:

1. /Ø/ (i.e. not at all, e.g. in place names that have spaces)

2. /gn̰θstʲ/ (i.e. the word “space” gnᵘþstʲ)

3. /e/ (e.g. when attempting to annotate a word with vowel markings)

4. /ʼ/ (i.e. mark the previous phoneme as ejective)

Escaping non-letter characters is easily, and frequently, generalised to

other characters, especially in fantasy and other fictions, where punctuation
marks and other similar items are escaped and turned into normal letters.
These examples are drawn from some stories:

(91) a. mᵘvᵉl\’dn
b. \ kkᵏddᵈd\,f\*

However, this is not entirely frivolous, because this treatment is also used
in academic texts where foreign words with phonemes not used in Drsk are
shown using escaped punctuation marks.49 In particular, the escaped , is
frequently used to represent /Ɂ/, such as in

(92) \[,sdc\] ”Jesdic” – E.-Pasaru ”Empire”

In this case, a singly-escaped punctuation mark behaves the same if

multiply escaped.
The backslash itself can also be escaped, which makes it a letter available
for assignment of meaning in some form:

(93) \\

21.2.2 Multiple escape

The multiple escape surrounds a list of letters and numbers, which are all
considered “escaped”. This is most commonly seen in names, where the main
portion is always escaped as such:

49
There are other ways to indicate such phonemes, including a provision in B.-Rattssaw
where several “spare letters” are used for just this purpose.
21.2. ESCAPES 239

(94) \[ŤSRGÐ\]

Multiple escapes are usually used to include foreign words and names for
units in ordinary Drsk text. They generally indicate that the word has an
idiosyncratic vowel insertion rule, which is fairly common for foreign words
whose vowels need to be preserved. Notice that the actual vowels are not
indicated; the reader has just have to know them. This can result in dialectal
variations, which we can demonstrate with the name of the planet Pasaru
in Drsk, \[PSR\]:

Language community Pronunciation

Nffek [peseɹu]
Woróli [pesəɹu]
Trokier [pɐsɐɹu]

It is expected that speakers will more or less be able to reproduce the

vowel quality from the original word (here, /pasaɹu/), so there’s not as much
dialectal variation as the non-indication might imply.
This indication of an idiosyncratic vowel insertion is also available for
certain names, even native ones.
Non-letters that have a special meaning when escaped preserve that
meaning when multiple escaped, so these two are the same:

(95) a. \[,sdc psr\]

b. \[,sdc\]\ \[psr\]

Although (b) is more technically correct, it is visually very noisy and so

the other option is used instead.

21.2.3 Escape type equivalence

Nominally, and normally, there are only two types of letters in this context:
unescaped and escaped. That is to say, it doesn’t matter whether a letter
is escaped using the single or the multiple escape; an escaped letter is an
escaped letter. So these two are, unless otherwise indicated, the same:

(96) a. \[KKT\]
b. \K\K\T

However, this is not always the case (if it is, this section would not
exist). There are some hairy points that make the two escapes not exactly
equivalent for certain glyphs. Let’s have a look at some particular cases.
First, the differences, which are easier to enumerate. The most obvious
difference between the two escape methods is a matter of pronunciation.
These two are usually pronounced differently, at least in nominal terms:
240 CHAPTER 21. ALL OTHER WORDS

Word Phonemes Realised

\d /dzgl/ [dzgl]
\[d\] /d/ [da]

Furthermore, one never indicates that a word has fixed vowel insertion
rules via single escapes; they must always be indicated using multiple es-
capes, as it is the word as a whole that has the specified vowel insertion
method.

21.3 Answer words

21.4 Particles and Exclamations

These are mostly single characters, such as P, V and f. Being a particle is
not a particular classification like noun, verb and dredge, but instead is an
attribute which can apply to any number of words.
The distinguishing factor that noun particles have is their alternative
inflections regarding suffices – some of them are replaced with zeroes, e.g. -t
or -tʲ. Their status as nouns is largely predicated upon their fixed position
in a sentence.
The most visible and common particles are the two special “pronouns”,
P and V. These two look like ordinary pronouns like a “I” or “you”, and in
some cases they are, but they aren’t exactly, as was explored in (34 | 17, 30,
48, 61).

21.5 Camera morphemes

21.6 Letter words

21.7 Comparisons

21.8 Incomparisons

21.9 Loanwords

21.10 Quasinouns
Chapter 22

Conversions by Morphology

22.1 Typing words

22.2 Tags

22.3 Gender patriation and depatriation

As will be explained in section 23.1, the root used for nouns and verbs
are .mᵘ. and .p. respectively, which although originally derived from geo-
graphical or otherwise physical artefacts, has also now generalised it to mean
“patriated” – that is to say, belonging to a particular set of large organisation.
In this sense, being a noun is considered one of these organisations.
“Patriation”, in the linguistic sense, is a method of turning a verb into a
noun by giving it genders, a suffix and a fixed location in the sentence.
(34 | 62) For instance:

(97) a. nᵘdᵘsfg eat (a meal, formal)

b. f- nᵘdᵘsfg- tʲ
:[food] eat (a meal) CITE
A formal meal

In contrast, depatriation is exactly the opposite: dropping the suffix and

gender from the nouns so they become bare roots. Words transformed via
depatriation are generally verbs.
Both methods are considered relatively new compared to the rest of the
language. As previously explained, conversions from other parts of speech to
nouns are mostly suppressed because of the closed-class nature of the noun.
This is likely derived from something like so:

(98) f-Ø-t nᵘdᵘsfg-t dpŋ

:[food]-PRN-T eat-T bind

241
242 CHAPTER 22. CONVERSIONS BY MORPHOLOGY

Which is in fact a sentence with both nouns and verbs, meaning that
one needs to have additional care over placement of words for delimiting
sentences correctly.
In a similar vein, the conversion from noun to verb is not particularly
hard but is suppressed because of the inherent ambiguity of each individual
root.
Part VII

Syntax

243
Chapter 23

Word order

23.1 Nouns and verbs

As previously emphasised in sections 16.1 and 18.1 respectively, in Drsk the
word order is completely defined for nouns and completely free for verbs.
In a sense, the relation between these two facts is not at all directed; one
can easily say that in Drsk, nouns are the parts of a sentence whose order
is fixed, and verbs are the parts of a sentence whose order is free.
This is normally expressed using an analogy of a “sentence tank”, and in
fact many of the grammatical roots for things are based on these aquatic
analogies (23 | 18):

Root Grammatical Aquatic

.mᵘ. z.mᵘ.tʲ “noun” ls.mᵘ.tʲ “island”
.p. z.p.tʲ “verb” lⁱ.p.tʲ “water/fluid”
.vᵉkᵏ. zr.vᵉkᵏ.tʲ “sentence” ðʲmᵘ.vᵉkᵏ.tʲ “tank”

As the order of the conventional nouns and verbs in a sentence are al-
ready treated elsewhere, we’ll restrict the discussion here to how they are
contextualised in a particular sentence.
The sentence tank looks a little like Figure 23.1. The tank is filled with
a liquid, representing the freedom that verbs have in their placement, which
surrounds solid islands which represent nouns. As befits their name, dredge
tend to show up near the edges of the tank, though they sometimes they
show up a little bit further away from the edge.

Figure 23.1: The sentence tank.

245
246 CHAPTER 23. WORD ORDER

TAMP NTS

NTAM VTAM SS ADD VP

NP VP ERG ACC LOC OBL

NP NP NP NP
S

TAMP NTS

NTAM VTAM SS ADD VP

NP VP ITRS LOC OBL

NP NP NP

Figure 23.2: The two ordinary sentence trees that all Drsk sentences can fit
in.

Sometimes those solid islands are tanks of their own, and this is the
central basis to recursion and relative clauses.
There is little point in describing the syntax of verbs as they are subject
to very few rules and therefore can go basically anywhere to their liking.
The few rules that do apply to them have been described in the chapter
dedicated to them already, so we will not duplicate them here. On the
other hand, nouns are very much the part of the sentence that gives it its
shape; their inability to move anywhere relative to each other with very few
exceptions means that they can be relied on to parse a sentence relatively
easily. Fortunately for learners, the rules are strict enough that there aren’t
actually all that many of them; with limited exceptions and the overarching
metaphor of the sentence tank, the noun syntax is largely trivial.
The general syntax behind a sentence would therefore be something like
one of the two trees in Figure 23.2, and verbs are allowed to roam free
within. Because of their relative freedom, they are not indicated normally
in a generic tree like this. In these trees we have included them at the latest
(rightmost) position they can appear in, and they’re joined to the part they
belong to with a dotted line to indicate their vague position.
23.2. INTERNAL WORDS 247

23.2 Internal words

The fact that nouns have a very rich structure and that they are localised
results in some interesting consequences. One of which is that there is some
level of freedom as to how constituent parts of nouns can be ordered; thus,
there is a syntax of internal words in a noun.
An example of such syntax is noted in section 23.3, where certain in-
ternal words – quasinouns – can be moved around subject to certain con-
straints. This can be so extreme that they can move out of a phonological
word entirely, but because of localisation it’s extremely difficult to add an
intervening lexical word between them.
In slightly less extreme forms of freedom, internal words do have the
ability to move amongst themselves, but this freedom is not absolute and
therefore we can discuss their restrictions.
The first and most important restriction is that internal words must not
move further to the right than the root. This means that internal words
generally exist between the gender of a word (or where it would be if it was
entirely omitted) and the suffix.
The exact order of internal words thereafter is determined via a system
of binary questions that test whether or not the internal words in question
have a particular property or not. These questions are binary – specifically,
they test whether or not the word possesses a certain binary property – and
they test properties that are semantic in nature, i.e. they depend on the
meaning of the word. These questions are called “discriminators”.
Drsk has the following discriminators:

1. Generalisability

2. Prototypicality

3. Inherence

4. Engineership

5. Direction

6. Subjectivity

Generalisability (symbol [±generaliseable]) refers to the semantic prop-

erty of those modifiers that have a qualia to them which allows them to
easily “transfer” themselves from root to root. For instance, the concepts
they refer to can be tangible in some way (e.g. “hard”, “wet”, “bright”, “grid-
ded”), or that a majority of those that understand the meaning of both
those words can apply the modifier to a novel object and arrive at the same
result (e.g. “five”, “green”, “educated-past-university”). This in particular is
in part encoded in the difference between the suffix ħ and the suffix y in
248 CHAPTER 23. WORD ORDER

that [−generaliseable] usually implies suffix ħ and [+generaliseable] usually

implies suffix y, (28 | 59) and this is true for most novel words but aren’t
true for the more commonly used words because of historical reasons.
Prototypicality (symbol [±prototypical]) refers to whether or not the
internal word has a prototype – a known object in the world which pro-
vides a reference value to which all other objects are compared to in order
to determine whether or not the modifier applies. Examples of prototypi-
cal words include modifiers indicating location of origin, temperatures and
specialisations. Non-prototypical modifiers include colours, numbers, and
sensory responses of any kind. (28 | 63).
Inherence (symbol [±inherent]) refers to whether the modifier has a
quality that is given externally (“bestowed” upon by an external enforcer) or
internally (“inherent” to the object itself, though it’s more accurate to say
that it is evaluated as such based only upon the behaviour and appearance
of the modified object, and that it cannot be removed from the head without
destroying it in some sense). Examples include:

Bestowed Inherent

• Colours
• Titles
• Size
• Usefulness
• Composition

• Interactivity • Purpose

Engineership (symbol [±engineered]) describes whether or not the in-

ternal word represents a property that is set in the way specified by the
internal word with some express intention of being that property. For ex-
ample, if a “dark room” is being contrasted with a “bright room”, or that it
is deliberately made dark to inconvenience looking for something, then both
“dark” and “bright” would be engineered in this case; whereas if a “dark room”
merely causes the inconvenience without someone deliberately darkening it,
then it is natural. (28 | 63) Note that this means that the combination of
the modifier and modified does not uniquely determine the result of this
discriminator – this is evaluated on a use-by-use basis.
Direction (symbol [±direct]) represents the state of whether or not the
property can be determined via an intuitive or everyday manner (“direct”), or
if it must be inferred by other means (“inferred”). This is not an evidentiality
marker; this refers to the (in)ability to make this measurement in principle.
For example, in the phrase “rough stone”, the internal word corresponding
to the word “rough” is always considered to be [+direct] even though the
roughness of the exact stone turns out to have never been evaluated by
the speaker. For numbers, the integers from zero to eight are considered
23.3. QUASINOUN DETACHMENT 249

[+direct], with all others considered [−direct], absent context cues from the
root.
Subjectivity (symbol [±subjective]) is exactly as its name suggests:
if it’s conceivable that two reasonable observers can disagree in honestly
attributing the modifier to the root object then it’s [+subjective]. This is
frequently seen in certain quasinoun constructions that uses the morpheme
-kt- “like”, e.g.

(99) mlnᵘ- kt- y

[Link]- like- QN
graceful, as if (but not) flying

These discriminators are applied in this order to provide the final order
on any particular collection of internal words. If a noun provides a positive
response to some discriminator, it is placed to the left of all words that pro-
vide a negative response to the same discriminator. The next discriminator
is tried to sort all remaining modifiers that produced the same response for
the previous modifiers. (28 | 66)
Discriminators take into account the word to be modified and sometimes
even the exact context in which the pair is to be used, so the same internal
word can give a different response to the same discriminator if it turns out to
modify a different word. For instance, the word “massive” (ltᵘħ) is [−direct]
relative to the root “star” (ð.pns.tʲ), but is [+direct] relative to the root
“train” (č[Link].tʲ). (28 | 65) This is because of the nature of the objects
these words refer to: it is not practicable to measure a star’s mass directly,
whereas it’s entirely possible to do so for a car’s mass. This pattern repeats
for a number of different discriminators, though it should be noted that some
discriminators are constant given just the word itself.
In the case where no discriminator can distinguish between two internal
words, no fixed order exists between them. Instead, it’s up to dialect and
individual choice, though in some cases this may be constrained by prosody.
In particular, the language prefers internal words with 2, 3 and 2 phonoruns
in that order than in any other order.

23.3 Quasinoun detachment

A typical noun would have the following structure, which we have repeated
here using a more conventional notation from Figure 16.1:
250 CHAPTER 23. WORD ORDER

G NN SUF

SM MA MI IW ROOT

IROOT ISUF
But note that a noun can have any number of IWs, and this can become
unwieldy as ROOT moves too far away from G. To resolve this, a mechanism
called quasinoun detachment is used.
To do this all we need to do is to move out some of the NNs, so they
now look like this:
N

ω ω

IW IW G NN SUF

• • IW ROOT

Here we have specifically included an additional layer that shows that

the noun, N, contains two phonological words ω. What we have done here is
to literally detach a number of the internal words from the word, and then
move them over to another (phonological) word – and as these are usually
quasinouns, hence “quasinoun detachment”. Despite there being more than
one phonological word in this word, it’s still only one grammatical word;
there’s no real obvious way to move the detached words far away from the
phonological word containing the ROOT, without a large amount of poëtic
license.
Additionally, there is a restriction on which internal words can be moved
out and which cannot. Aside from the one immediately before the root,
quasinouns can be moved out, but included nouns (e.g. in compounds)
cannot. This is the primary way to distinguish the two from each other. It’s
questionably permissible for a quasinoun to move out if it means moving
over an internal word that cannot be left inside; this depends on whether or
not they depend on each other.

23.4 Modifiers and the Modified

On a higher level, there is the idea of head components and modifier com-
ponents, which we use in multiple areas to describe how to order certain
components.
23.4. MODIFIERS AND THE MODIFIED 251

The principle here is that there is a specific order which must be obeyed
for any group of objects that follows a modifier–modified relationship. This
relationship exists if the two parts combine in the following manner:

1. The first, called the head, is a more concrete object, which is to say
that it has an independent existence in the context of the pair.

2. The second, called the tail, is a more abstract object, which does not
have an independent existence in the context of the pair. Frequently it
does not have an independent existence in many other contexts either,
but that has no particular bearing to this particular relationship.

3. The two parts are connected to signify a single object which is a blend
of the two objects.

What an “independent existence” implies is slightly more complicated to

describe, but it can be understood as whether or not the object can have a
prototype – that is, a platonic ideal that can take a tangible form. Most of
the time, for the second item, this is not possible, as they consist mostly of
things that behave semantically like adjectives and adverbs – their existence
comes from generalising many things that do not have similar tangible forms
and therefore cannot be “averaged out”.
The main reason why we refer to generic objects here is because there
are multiple instances where this relation applies. These rules applicable
to certain nouns when participating in a noun complex as well as entire
sentences. Additionally, combinations of heads and tails can themselves be
heads and tails.
With that in mind, the rules for ordering heads and tails are as such:

1. The head is always placed at the end.

2. If there is one tail, then it is placed immediately before the head,

excluding any free elements that might appear there.

3. If there is more than one tail and they apply to the head equally, then
given that they are ordered 1, 2, 3, · · ·, they are placed in the text as

2, 3, · · · , 1, h,

where h is the head. Note that this gives the impression that the
modifiers are right-branching all the way out to the second tail, which
is the overwhelming majority of all such head-tail combinations.

Item 3 raises the question of how the tails are ordered. As mentioned, it’s
not common to see more than one tail for any given head; even if there’s more
than one modifier, they sometimes occur in a way that causes one to take
252 CHAPTER 23. WORD ORDER

priority over another, meaning that recursive head-tail relations exist. The
primary usage of multiple tails is that of the case where multiple adjective-
like quasinouns modify a root with equal priority, which easily maps into
adjective-noun groups that appear in other languages.
However, we have actually encountered one such ordering: section 23.2
provides a way to assign a particular order to all internal words, and as most
internal words are also modifiers, i.e. tails, the principle is carried forward
to all head-tail sets in general. The one thing to note is that with internal
words, the final twisting in item 3 is not conducted.
Apparent exceptions to this ordering are not actually exceptions: they
are instead instances of other relationships that aren’t modifier and modified,
and therefore aren’t heads and tails, and are their own category of object
pairs.

23.5 Noun phrases and complexes

There are a handful of different ways that one can combine nouns together
into one syntactic unit. We’ve previously encountered noun compounds in
section 16.14, which is fairly simple in terms of being a noun “complex”
(actually a compound). There are others that are also available for use
in various ways, and they are characterised by one of their prototypical
examples. In this section we’ll go through them, and introduce them through
their prototypes.

23.5.1 Time and range intervals

Time can be expressed in other ways, and fortunately for us some of these
ways use some of those other ways one can combine nouns together back
into nouns, but bigger.
One of those ways is the time and range interval, which for all intents
and purposes work as a single noun. It is not to be confused with specifying
an absolute time, which we have previously encountered. The syntax is as
such:
S

TAMP NTS

VTAM NTAM V NP N

Yesterday then Go fishing N N

PRNᵘDLⁱ fish
23.6. VERB COMPLEXES 253

23.5.2 Family
Terms for family are used for describing familial relationships (obviously),
but they are also used for some other functions as well. In this case, the
nouns together work as a bit of a state machine, with some aid of CAT
transformation to make operations explicit.

23.6 Verb complexes

23.7 Noun-verb complexes

23.8 Idiomatic verb placement

254 CHAPTER 23. WORD ORDER
Chapter 24

Sentence structure

24.1 Additional arguments

24.2 Additional verbs

24.3 TAM words

24.4 Questions and responses

24.5 Narratives

24.6 Small contexts

24.7 Incomplete sentences

24.8 Dates, times and intervals

255
256 CHAPTER 24. SENTENCE STRUCTURE
Chapter 25

Repetition and Recursion

25.1 Relative clauses

25.2 Compound sentences

25.3 Set phrases

25.4 Large-scale structure

25.5 Conversions by Syntax

25.6 Definition lists

25.7 Sentence to Verbs

257
258 CHAPTER 25. REPETITION AND RECURSION
 Part VIII

Pragmatics

259
Chapter 26

Conversations

26.1 A status game

26.2 Casual conversation

26.3 Formal conversation

26.4 Multiple conversants

26.5 Announcements

261
262 CHAPTER 26. CONVERSATIONS
Chapter 27

Large-scale communication

The definition of large-scale communication is that it is a communication

that persists temporally beyond the time it is expressed. In other words, if
you can refer back to it after the communication has been completed, it is
considered large-scale.
Naturally this contains a lot of examples, but they can be subdivided
into several examples, which we will discuss in detail below.
Broadly speaking however the formats of large-scale and written com-
munication is highly compatible with the traditions of the J.-Senlis, which
is no coïncidence, considering the historical context. When it comes to this
very high-level part of the language, the actual language itself starts not
becoming very important, as broad compatibility across the international
norms becomes more important.
Nevertheless, a lot of the details here are identical to the details required
in Rattssaw, so we will draw upon resources originally described of Rattssaw
in order to complete the description – with suitable modifications for Drsk,
of course. The references to Rattssaw will remain as with the rest of this
book.

27.1 Formalising communications

Before we actually discuss how these communications are formatted, one
might wonder why anyone would want the language to include such formats
as a formal part of the language, and furthermore who is responsible for that.
To do so, we will have to discuss the identity of the language, who maintains
that identity, and how that identity is expressed through the natures of the
highest level of a language (according to the language skeleton, see Section
1.3) – its passages.
…
Another fact that one must keep in mind is that the identity of the lan-
guage is tied to its visual appearance independent of what it is writing, and

263
264 CHAPTER 27. LARGE-SCALE COMMUNICATION

nothing is more visually apparent in any given piece of writing in any lan-
guage than the way that the words are physically arranged in a page. This
forms a kind of “language-wide inscribing tool set” (think national identity,
but specifically for typesetting) that is, at least in the traditional50 consen-
sus. Thus, every part of how you format a particular letter is “the language”,
because it’s literally “where you put the words on the page”, which is very
similar to “where you put the words in a sentence”. There are differences
between the two but the traditions have decided to ignore it, and the final
result is that how you format a letter is as important to those who speak
any given language as whether you put nouns before or after adjectives.

27.2 Brief messages

Brief messages are a smaller kind of letter which are considered to be more
casual and less of a status game than the other types of messages.
A typical brief message would typically only consist of the message, with
minimal decoration or status game phrases thrown in. However, they are
still subject to some level of formatting, which is as such:

1. An optional title on its own,

2. Followed by a body that must be of limited length.

What that limited length is would vary based on the amount of writing
space available but generally does not exceed about half a minute of reading
time or five sentences.

27.3 Letters
There is a specific and very strict format for letters, which broadly speaking
is similar to the Rattssaw format. (25 | 39) The general format of a letter
is as in Figure 27.1 with each component being described later. (32 | 18)
Of particular note is that the letter looks “backwards” compared to con-
temporary Earth letters – the sender is placed at the top, while the recipient
is placed at the end. Below are details regarding each component.
We will compare some pertinent differences between this Drsk standard
format with the Rattssaw standard format, which comes of particular note
when one wants to send “international” letters that are of particular impor-
tance in diplomacy.
50
We’ve been a bit naughty here and conflated “traditional” – things that have been done
for a long time across multiple cultures and settings – with “local” – things that have been
done for a long time but across a specific group of cultures and settings. But when we are
discussing a system of cultures that is completely disparate from ours, the difference between
the two broadly disappears (though we might discuss “planetary local” systems, which if there
was only one planet available we would have called “international” systems).
27.3. LETTERS 265

[Sender]
[Send date] → [valid at]
[Salutation, typically lasting several lines. This is
padding text that helps show the extra indent.]
[The body text. Again, this is usually very long, but is re-
placed with padding text here to show the normal indenta-
tion.] [Standard final]. [Valediction].
[Bridge] [Title or summary]

[Pretitle]
[Recipient]

Figure 27.1: The general structure and indentation of a letter.

27.3.1 Sender
The sender’s name is placed at the very top of the letter, which is written
“fully” – that could mean in terms of having all components of the name
written out, but here it specifically means that the name’s genders and suf-
fices all have to be explicitly indicated (though sometimes they are in small
text). In Rattssaw, the title (which is the closest equivalent to the gender in
Drsk) is not necessary, and there’s no such thing as a suffix so it’s completely
omitted.
Because of the way that names work in Drsk, there is a little complication
with actually “naming” the sender, and the status game starts even here.
Recall that names in Drsk are given in the Northern style, so names are
literally given, probably by the sender. In this case, the sender is giving a
name to himself, which can either be previously agreed upon, or he can be
more assertive and give himself a name. If the latter however, it rarely lasts
beyond the lifetime of the conversation conducted by way of these letters.
The sender can also choose to leave this part empty or replace it with a
nonspecific “hello”, which signals the recipient to give a new name to him.

27.3.2 The two dates

There are usually two dates on the letter, which is in contrast to the general
norm, which is either one (in the case of Rattssaw) or none (most everywhere
else). These two dates are the nominal date and the “valid at” date.
The former of the two is fairly conventional – it simply indicates when
the letter is completed. Specifically, not when the letter is written, as it is
expected that the letter would take more than one day to complete.
266 CHAPTER 27. LARGE-SCALE COMMUNICATION

The other date is slightly more interesting in that it indicates the time
when the letter is supposed to be received, or a best estimation thereof.
This date is called the “activation date” or the “valid at” date (sˢrlsydmᵘtʲ,
:[time]-dealing-stamp-CITE), and is the “present moment” of the letter, with
any TAM information in the message centred at this time unless otherwise
required.
In more modern times indicates when the message in the letter is to
be “executed” (pplmᵘl) by, e.g. if the letter is a message that requests
an action to be done, then it should be done by the valid-at date; if the
letter instead provides information, then the valid at date is the date which
the information is, well, valid at. This can sometimes be at odds with the
traditional meaning, which is not as relevant anymore as postal delivery has
become more and more reliable as the years roll by.

27.3.3 Salutation

The salutation is the textual equivalent of the status game that conversations
have: at this point the writer of the letter is to set the scene regarding who’s
the superior and who’s the inferior. There is no particular set formula to
this, and as with conversations this relationship can change over the course of
the conversation, or even within a single letter, but it is absolutely required.
The salutation is also not without meaning; a new one must be written
every time appropriate to the tone of the previous letter (if any) and the
following text. The topics discussed in the salutation may however be di-
vergent from the main body, either partially or entirely, but usually does
not stray too far from the usual easy topics: weather, things that happened
or are done recently, “I saw a funny thing on the way to somewhere today”,
even some random numbers that may or may not be related to something
else.

27.3.4 Standard final

The body contains mostly free text, so there’s nothing much to discuss here.
However, the end of body contains something called a standard final, which
is a specific sentence or group of sentences that are grouped together with
the main body text typographically, but not structurally.
The standard final, where one exists in a letter, is considered to be the
written equivalent of the exit stage of a conversation. So it would normally
contain sentences that are described in that stage of a conversation, though
because of the lack of response the exact words are adjusted a little.
27.3. LETTERS 267

27.3.5 Valediction
The valediction is a formulaic ending to the letter. It is picked from a list
of about 200 or so, which are categorised using a particular code, (25 | 44 –
46) which is in turn published as part of a dictionary.
This system is entirely imported from Rattssaw, and the valedictions are
similarly translated from the language, though there is some adaptations and
local modifications to that end. Both languages however do share roughly
the same valediction codes, so they be placed in an omnibus volume.
To summarise the code, it is composed of a list of components, which
consider the following:

1. The difference in the roles of each of the conversants

2. The difference in the caste of each of the conversants

3. What day of the year it is

4. Whether the message is replying to something, or starting a new topic,

or both

5. How the sender thinks the recipient will take the message (positively,
negatively or neutrally)

6. Whether either party is an individual or on behalf of a group of indi-

viduals

7. Whether either party is an outsider to some specific group

8. Whether one particular party is part of a distinguished profession, such

as teachers and doctors

27.3.6 Title or summary of data

The valediction is then followed by the title, and the two may be separated
by a filler phrase, such as kdsˢtyl, which is derived from the Rattssaw phrase

(**) Kdl earosstael

kdl earosst-ael Ø
above summarise-eat COP
“This has been...”

The word here is atomic, i.e. does not change in any circumstance, and
occupies a special part of a sentence that otherwise would not be occupied.
It is considered a prefix word to the rest of the sentence. Any prefix word
that is appropriate to the rest of the sentence is usable as the filler phrase.
268 CHAPTER 27. LARGE-SCALE COMMUNICATION

The title itself is, as its name suggests, a brief sentence that recaps the
entire letter and gives it a summary. This is required even for relatively
brief letters, and those that are too brief to require a summary would be a
brief message instead.

27.3.7 Recipient complex

The recipient complex is the name of the receiver plus an additional pre-title
that adds the correct amount of respect to the recipient in question. The
pre-title is something that can uniquely identify the recipient given a specific
context, such as a family member position or a corporate role. Individuals
also have a pre-title that have been published for use in the postal system
like this.
The reason why the pre-title is particularly important is again due to the
naming tradition that we have previously discussed, so the pre-title is used
to allow outsiders to actually be able to pinpoint the individual to receive
the letter without reference to their own names. This is in some sense
just names in another disguise, with the key difference that such names still
don’t stick with the individual permanently – they are instead one particular
combination of an individual and its roles.

27.3.8 International letters

Writing to people outside of the Drsk- and Rattssaw-speaking sphere means
that there is no guarantee that the elaborate format would be understood by
the foreigner in question. This can be an issue as certain status-determining
elements can become impossible to complete without the other party coöper-
ating – within the culture of the J.-Senlis, this is not a problem as not com-
peting in the status game is not simply not an option having been made
culturally impossible to do so.

27.4 Notices

27.5 Articles

27.6 Books

27.7 Poëtry

27.8 Dictionaries
For reasons relating to the partly taxomonic nature of the vocabulary, the
alphabetical order of dictionaries across the world is not entirely repealed
27.8. DICTIONARIES 269

with Drsk and a small handful of other languages. In a fairly unique arrange-
ment, the Drsk dictionary is customarily divided into two sections: the first
section for nouns, and the second section for everything else.
270 CHAPTER 27. LARGE-SCALE COMMUNICATION
Chapter 28

Registers

28.1 Simple registers

28.2 Diagrams as register

28.3 Tables as register

271
272 CHAPTER 28. REGISTERS
Chapter 29

Storytelling

One of the most important parts of a language is what you can do with it.
And one of the most important things you can do with a language is to mke
up big old lies about it and call it a story.
Although it doesn’t work exactly that way in Drsk, and indeed in the
rest of Pasaru, the general idea of using language to make up stuff remains
more or less the same. In order to more accurately express this idea though,
we will have to delve into what makes a story a story in Pasaru, and how
this has not remained constant throughout the years, as one might expect
it to be on Earth.

29.1 The birth of fiction

The idea of actually outright saying things that are not true in order to
entertain and inspire is a surprisingly recent invention in Pasaru, as opposed
to the idea of saying things that are not true in order to deceive, manipulate
and punish (which is genuinely an ancient act). However, precursors to
fiction as the former do exist for a long time, particularly in the form of
religion.
We have previously examined some religions in sections 2.1.5 and 2.2.5,
which are somewhat important to setting the stage on the motivation of
certain words and grammatical formations. However, to examine their ef-
fects on fiction, we will have to go further and look at them in general.
(17 | 54)

29.1.1 From Gods to the Star Race

The oldest religions posit that there exists some single force that is every-
where and always. The normal next step here is to personify this force
into a form more units understand, and this is done in, for example, earlier
iterations of Pnfy.

273
274 CHAPTER 29. STORYTELLING

Up to about now things are still relatively recognisable as Earthlike reli-

gions. They were the first to start employing an embryonic form of fiction in
order to encourage and discourage certain behaviours. The only difference
is that the stories are always made plainly clear that the objects inside it are
ancillary, and it is always morals that are considered to be the important
part. This emphasis stifled the creation of fiction for a long while.
At this point the evolution starts becoming a bit unhinged compared to
Earth. The number of gods started to inflate over time, in a basically expo-
nential manner. Religions that started off polytheïstically of course have no
problem adding more gods quicker, whereas religions that started off with
a smaller number of gods had a little more difficulty. The general pathway
is to either take the step to duotheïsm directly, (a slower and slightly more
difficult process) or to lose focus on the single or few gods and start focusing
on the minions that these gods inevitably entail (a much faster and more
common process). As so it turns out though, monotheïsm is quite rare, so
it is not a particularly large obstacle to growth of religions.
Eventually, one way or another, a small number of immutabble gods gave
way to what is termed the “Star Race”, a large (105 < n < 109 ) list of names
and relations that is largely subordinate to the world that they produce.
Newer religions that spring off of the older ones tend to even ignore the
identities of the individual and start focusing on what kind of world would
be able to spawn theis “Star Race” in such prodigious amounts, and this
line of thought generally dominated theologies in many places beyond some
point.

29.1.2 Finance ruins everything

Meanwhile, in parallel, the ideas of economics started to take root amongst
the now intensifying long-distance trade routes. The first corporations formed
along these trade routes as couriers and inventory management, and are
looking to expand their businesses.
As so it turns out, one of their most favoured and needy clients are the
big religions and churches, who have not yet forgotten the face that religions
are built to be spread, not to ruminate endlessly in an empty building. Since
a message is considerably easy to spread as opposed to perishable goods or
physical items that are subject to theivery, religions find themselves a whole
lot of new believers along the trade routes and they pay a relatively small
amount for this vast amount of believers.
However, there is only a finite amount of followers for this world, and
soon all the easy converts have dried up. This is of course a fairly obvious
consequence when stated this way, but when you are one who only looks at
the numbers from day to day, it is quite worrying. The couriers blame the re-
ligions for not making good enough stories to sell to others; the clergy blame
29.1. THE BIRTH OF FICTION 275

the couriers for not being diligent enough to push the stories to potential
converts. Overall, things are not looking too good.
Eventually, by the luck of several factors that include bankrupting the
clergy and a prototypical form of “executive meddling”, most of the major
religions have fallen into the hands of the courier companies. To this day,
Drsk “mail”, “sermon” and “information” all share the same root bm.

29.1.3 The water runs out

Things get a little bit worse for the existing religions as the new converts
turn out to not be as faithful as previously expected. after certain incidents,
including the one outlined above, a new wave of religions came to be, which
called the Descriptivist Religions. These religions are special in that they do
not immediately claim that what they said is right, but instead the goal is to
only say correct things (the goal does not include saying all correct things
however.) As one can deduce, this is basically just science, and indeed from
these descriptivist religions, which later converged into Naturalism, slowly
became a protoscience, and then science itself was born.
A few thousand years later yet another branch of religion was developed.
This branch takes a slightly different approach to the increasingly obvious
mismatch between reälity and what their predecessors say. Instead of making
up more explanations as to how this alterante world came to be the ones
they live in, they simply gave up any claim that what they said was true
and so the Irrealis Religions are born.
These religions are different from Naturalism in that they have not con-
verged but have instead become the most dominant form of all religions.
Many classical religions (that is, any religion that is not Irrealis or Descrip-
tivist) started to gain Irrealis counterparts, and they started to overwhelm
them, though this did take some time. This signals the beginning of religions
in general, which is then compounded by the classical religions becoming no
longer mutually exclusive; it is no longer impossible to “believe” in more than
one religion.
From here we can detect the beginnings of fiction as we might know it:
after all, this is a type of fiction, as there is now no claim that what was
said was true, and it was built to entertain and inspire. For this reason, at
this point, Irrealis Religions became protofiction.

29.1.4 The tretise of fiction

The jump from protofiction to fiction required one final step, and that is the
Tretise of Fiction, published shortly after the years turned into the positives.
From here, fiction was well and truly “born”, and the idea flourished quite
happily as if it was there all along.
276 CHAPTER 29. STORYTELLING

Most of the remaining classical religions, now mostly branches of the

courier companies that have consumed them many years ago found this new
idea of fiction quite attractive and profitable and have therefore banded
together to form the World Organisation for Parahistories. This essentially
signals the end of religions as we know of them, as parahistories only claim
to be some kind of “real life fanfiction” – a genuine fiction that doesn’t make
any claims as to the answers the “big questions” that religions are wont to
answer.
And with that, fiction as we (more or less) know it was established.
There are a few wrinkles regarding how fiction works in the details, but
in general the big ideas have been established and the idea of Worldstate
Literature was born, which will become the dominant storytelling paradigm
across the planet.

29.2 Relations to reälity

When it comes to describing reälity, things are as usual a little bit hazy. All
fiction, by definition, include a little bit of an element that is false, but what
is and what isn’t is a little bit different across all the literary traditions.
For the most part, the idea of what is “acceptably false” and what must
remain true is expressed as the “detail” and “core”, respectively. It is generally
considered bad form to change the core so much that it becomes too difficult
to follow, even with considerable effort. For example, the fact that cause
precedes effect is considered “core” and should generally remain true for the
most part. In this respect, the idea is not too dissimilar to Earth ideas, only
that here the distinction is more explicit.
Using this idea we can explicitly state how much a literary tradition
“sticks to reälity” in the general sense.
As so it happens, for the most part the “core” ideas tend to be a lot more
far-reaching, with such ideas as the following being considered “core”:

• Physics remains the same – this means that things like “magic” gener-
ally remains different.

• Maths remains the same – unless one pulls out a list of new axioms
that define a different maths.

• Locations are always different – a story generally will never involve

actual places.

• Things are never done alone – any major event must always be a group
effort.

Both of these consequences are more or less predictable if we consider

the fact that fiction arose from irrealis religion, which had earlier disclaimed
29.3. WORLDSTATE (“INTERNATIONAL”) LITERATURE 277

any relation to reälity, except for those that seem to be “obvious, universal
truths”. It’s fair to say that what counts as obvious would be different
amongst the kilis.
But on the other hand, there are other things that aren’t as important
and can be kept as detail. This of course includes things like names, places
and events, but strangely also:

• Language, which is generally different for every world;

• Economics in some cases, which is somewhat interesting considering

how far removed economics is from religion in the Tree of Knowledge;

• Some biological structures change from story to story for seemingly no

reason. This is somewhat less often seen however and it is not entirely
surprising that some stories just have strange creatures for seemingly
no reason.

Overall, the idea of adherence to reälity is a little bit more strictly ob-
served in Pasaru than it is on Earth, but it is not strictly so and there are
overlaps and incomparabilities.

29.3 Worldstate (“International”) literature

Worldstate literature grⁱ.tmtʲ is the main type of literature that has domi-
nated the landscape since the publication of the Tretise. This is a method
of storytelling that everyone knows about, and it is the type that so well
embodies literature that in general it is simply referred to as “literature”, full
stop. However, since it is mostly based in the southern hemisphere, its grip
is slightly less strong here in Klzdmk, even though here it is quite strong.
Stories that use this type of framework all have that singular feature that
is common across a lot of things in Pasaru: structure, and rigorous adhesion
to it. In this case, literature has the structure seen in Figure 29.1. (14 | 9)
In worldstate literature, the plot (here indicated by the dotted box, and
the node plot) takes in the initial state of the world, iworld, and out-
puts a final state, fworld. iworld usually contains an “undesirable compo-
nent” udcomp, and plot will transform it to a “desirable component” dcomp.
This change is caused by the plot synthesising (here, the hollow arrow-
heads) two components: an animate factor animate and an inanimate factor
inanimate.
To help with the idea of the world changing itself, animate is usually
drawn from a part of the world, typically called the inhabitant, inhab. This
inhabitant is not removed from the world, nor is he treated as a separate
entity from it. inhab typically remains the same through plot and through-
out the story and remains as part of fworld after plot is complete. In the
diagram, this is indicated by the pointy arrowhead and the dotted line.
278 CHAPTER 29. STORYTELLING

iworld inanimate plot fworld

udcomp dcomp
inhab animate inhab

Figure 29.1: The structure of worldstate literature. Abbreviations are de-

fined in the text.

The idea of change then is entirely incumbent on udcomp and dcomp.

These two components together give an idea of “good” and “evil”, and it is
up to the author to decide which corresponds to which.

29.3.1 Genres
The idea of genres comes from the fact that some features of stories tend
to cluster together in otherwise inexplicable manners. These do exist quite
strongly in Worldstate literature, as it is considered essential to have some
kind of structure in determining which values of iworld, fworld &c. can
go with each other.
For the most part, genres are more or less as what we might expect on
Earth, with “set dressing” being the bulk of the idea of a genre. But because
of the differing emphasis on what is considered important, the relationship
between genre and story is very different, and this difference, in effect, makes
genre and story much more intertwined.
This can be seen by looking at some of the more “standard” genres, which
is generally indicated by the settings they inhabit:

Space The setting is in space, which vaguely says “anywhere not near a
familiar planet which is inhabited in the contemporary civilisation”.
iworld and fworld usually have the size of several planets and includes
the space between them. animate is more variable than most others
here, as they are not guaranteed to be even civilisations but can be
anything from abstract structures to simple dots on a board.
Wild Here iworld and fworld are usually some small area which contain
a large amount of certain “wilderness” features. inanimate therefore
would involve lots of natural processes, and animate would be those
that are significantly less predictable, like storms, population clusters
and the water supply.
29.3. WORLDSTATE (“INTERNATIONAL”) LITERATURE 279

Sea A similar item to the previous entry, but somewhat wetter. It is some-
what distinct from Wild because here iworld and fworld are three
dimensional rather than two.

Home An even smaller version where iworld and fworld is restricted to

a small set of houses or similar environment Strangely enough this is
sometimes considered a subtype of Space because one of its defining
feature is that dcomp and udcomp are usually some decidedly value-
neutral things that nevertheless are assigned very unusual values. The
prototypical example is “ants infesting a house”, which involve a group
of small animals hoping to use the house as an infinite food source.
Given that, it’s clear to see that animate and inanimate would contain
strange things that would be unfamiliar to all except Space.

City A genre which has iworld and fworld be an entire planet of some
kind, and can range from political thriller to warfare simulation to
even urban exploration (in which case the world would be limited to a
city). This is one of the more popular genres if only because it contains
worlds most familiar to readers.

In these genres iworld and fworld are typically of the same scale, but
there are some fictions that cross genres by changing the scale of the world
through plot.
A somewhat interesting feature is that in many situations animate re-
mains more or less the same, but other things can be fairly different. There
are exceptions of course – the space genre is very animated when it comes
to (quite literally) alien ideals – but in general if animate does not include
factions or nations it usually means “literally anything that moves unpre-
dictably”.
Note that while this nicely gives us relations between the components,
they aren’t really “genres” like romance, exploration, action, &c., but those
aren’t really applicable here as worldstate literature does not feature the
individual character as strongly as one might expect. The reason behind
this is bizarre, but ultimately comprehensible.

29.3.2 Non-characters
What is most surprising about worldstate literature is the significant absence
of characters.
To clear up what is meant by a lack of characters, let’s take a look at
some of the things that we have discussed earlier. Notice that amongst
all the existing genres, animate never really involves an individual that is
particularly standing out against the background. There’s no protagonist,
no antagonist, no major interunitary conflict, no traits, no development, not
even a catchphrase. Everything is treated in generalities, and no single unit
280 CHAPTER 29. STORYTELLING

runs it all and brings the story forward. That is what we mean when we say
that worldstate literature lacks characters.
With that in consideration, the name “worldstate literature” becomes
entirely clear: this is a literary tradition in that heroes do not exist and the
real object of interest is the “big picture” of a changing world and how its
components interact with each other. It would be similar to how a biologist
would examine the changes of and to an ecosystem, as compared to someone
who likes animals would identify and name individuals within the ecosystem.
It is, quite literally, literature that describes the state of a world and the
change thereof.
As with the lack of character-oriented genres, this in turn is not en-
tirely unexpected. Recall that in Section 29.1.1 that one of developments of
religions in Pasaru that was a global experience – as religions are here pre-
cursors to literary fiction – the explosion of pantheons from a manageable,
mundane number to a vast number that is considered functionally unend-
ing and practically innumerable and unenumerated, at least at the time.
This blasé attitude towards individuals in religion, perhaps motivated by
the fact that in reälity no “individual god” can be found, carried on over to
the literary tradition today.
It might be interesting to compare this to other literary traditions, es-
pecially those which exist in poëtry (section 27.7). In particular, when dis-
cussing certain kinds of poëtry, individuals are clearly recognisable, and in
fact are entirely identifiable; these are the authors of the poëm, or the units
that he interacts with. They are indeed partially “characterised”, but these
are very much real units that have real expectations, real values, real unitar-
ities and real qualities that ultimately defy any traditional character-based
analysis of them being mere protagonists, deuteragonists, anti-heroes, &c.
In short, what we have is a somewhat inconceivably persistent case where
the line between poëms and long-form prose has a strange wall that for some
reason has never been violated by artists, who are, as we have previously
established, wont to destroy any boundary possible with the exceptions of
those put up by moderately difficult mathematics.51
So for whatever reason, “serious” literature simply do not admit charac-
ters in their discussion, instead replacing them with groups of objects that
behave in an unpredictable, complex and ultimately more reälistic manner.
After all, unpredictability is what is required to be part of animate, which
brings as back to the fact that while individuals might not be so prominent,
they can still appear, just as a more minor and less developed part.

51
Genuinely, for whatever reasons artists on Pasaru at least have a pathological fear of
mathematics that seems to be much more sparingly inflicted on others. While there are
occasional painters, sculptors, paper and that can definitely solve a 2nd-order differential
equation correctly without much thinking, their general rarity is a well-studied effect that to
this date has no generally-accepted explanation.
29.3. WORLDSTATE (“INTERNATIONAL”) LITERATURE 281

But wait, what about inhab? As previously mentioned these are consid-
ered to be parts of animate which are drawn from fworld, so they need not
be characters, groups, or even sentient (a particularly rowdy weather system
can be considered to be part of inhab, despite it being the poster child for
being a member of inanimate). So there’s no problem with inhab, even if
here its name is somewhat misleading.

29.3.3 Deletions

Far from being impoverished by the lack of characters, one might find the
structure of literature to be far too rich, and some seek to reduce it into
something more manageable. The result is “reduced literature”. (14 | 10)
Despite its name these literary traditions are not derided as being lesser,
but they do attract a lot of beginners, which may bring some indirect scorn
as a result. It is not generally considered a good idea to actually transfer
this distaste to the actual tradition however.
The first thing that one might be tempted to remove, as seen in the
notation, is one of iworld and fworld, resulting in world. This deletion
makes it so that the world does not change at all through plot, which is why
it is called “static literature”. Because the world does not change appreciably
in static literature, but it is possible that the world might change a little bit
during the literature, static literature is also called “cyclic literature”.
An interesting sub-type of static literature is the “yearly literature”,
which is a type of static literature that describes the changes a world oc-
curs over a fixed time cycle, which is usually one year. This type of story
generally involves following the seasonal changes of a world, such as the
growth and shrinkage of snow formations, migrations of animals, transport
patterns, and so on and so forth. As the world does not change a lot and is
fairly predictable, it is natural that animate is also muted or absent from
this genre. A pictorial representation of such a story can be seen in figure
29.2.
A further reduction can be made in the form of “fully static literature”,
whose structure is seen in Figure 29.3. Here, the tradition further removes
inanimate, with the remaining plot and world being completely static.
Such severe is the staticness that often time is omitted, and instead replaced
with space.
In effect what is being written here is a guided tour of a particular
world, with plot bringing us to each loc[n] and giving us a description of
it. While it might not sound anything like a story at all, it is still given
extreme prestige amongst the kilis and is an entirely legitimate form of
literature there.
282 CHAPTER 29. STORYTELLING

world1 world2

inanimate

world0 ...

Figure 29.2: The structure of partially-static literature. The passage of time

is indicated by the red arrows; inanimate drives the change in the world.

world

loc1 loc2 loc3 locn

Figure 29.3: The structure of fully static literature.

29.4. GRADED LITERATURE 283

29.3.4 Augmentations
(14 | 11) (14 | 12)

29.3.5 Tropes
(14 | 13)

29.3.6 Worldstate Literature and Drsk

Because of the dominant nature of this particular literature tradition, there’s
an interesting dynamic between the native tradition, worldstate literature,
and Drsk. This is because of the competing nature of the identities between
all three entities, and those that identify with those entities.
First, the obvious: the domination of worldstate literature means that
there is a sizeable amount of stories written in that tradition and in Drsk.
Some of these did get a lot of recognition and fame, comparable to those
that are imported from abroad. And still some more of these also enjoyed
international success as well.
The second thing is fairly predictable in response to the first: A native
backlash in terms of territorial defence. In terms of that, there is nothing
more obviously emblematic of this backlash is Važllr’s World in Yamtoll
(Eamttel). We will describe this particular contradiction in section 29.9.

29.4 Graded literature

In contrast to the international style of literature, Graded literature is some-
what less structured and is more likely to include factors that other things
do not, such as a stock list of factors (as defined in worldstate literature)
and a fixed format for telling stories.
Originally, there was only Grade (1) literataure. When Grade (2) arrived
it basically wiped out the previous storytelling schema, and replaced it with
its own. Ironically by the time the Restitution came along Grade (2) liter-
ature is no longer practised by the actual Grade (2) inhabitants because of
other concerns they have at the time, but this does not apply to the original
inhabitants of Klzdmk, which has by the time more or less appropriated the
whole system as their own. So what is considered “indigenous literature” is
actually a colonial import that has been sufficiently naturalised.
This somewhat complicates the terminology that is used to describe the
two types of graded literature. Instead of using subscripted numbers cor-
responding to the grades, we will instead refer to pre- and post-Contact
literature, where the contact in question is between Grade (1) and Grade
(2).
284 CHAPTER 29. STORYTELLING

In an extension to the meaning of grades the vast diversity in literary

traditions in Pasaru are also understood in Drsk as separate “grades” of
literature. This can get a little bit disturbing to Earth eyes as familiar
concepts such as “plot”, “characters” and “setting” get used in different ways
from story to story and from tradition to tradition, and few of them are ways
that one would consider “storytelling” in the contemporary Earth sense.
Generally speaking, the storytelling and writing scene in Pasaru is not
only to create new stories that have the usual items in fiction, but also to
create new traditions of storytelling by going one level beyond and playing
around with the structure of the story itself. In a sense, it is not stories
that are disseminated, but the framework that the story builds on which is
the important part. This explains the somewhat extreme diversity of story
structures in Pasaru, and how English translations tend not be too accurate.
Regardless, there remain concepts that should be familiar for all Earth-
ling readers; in particular, the sense that time generally runs forward from
moment to moment enforcing a “plot” of some kind – where events follow
a causal pattern and generally outline something interesting happening –
is broadly present across most literary traditions and can be considered
universal enough that it’d show up in the names of the months in the old
Drsk-Rattssaw calendar (see section 8.2.2 for more information) and remain
legible.
Since this is a Drsk grammar however, we will be examining each of these
tradition as they appear in the Drsk corpus, and we will comment on their
popularity as such.

29.4.1 Pre-Contact Literature

As previously mentioned any pre-Contact literary traditions are mostly
wiped out by the contact, so details are somewhat spotty in this regard.
Nevertheless, there are some details that are still recognisable today, as
there has never been a dedicated culture eradication effort upon the Grade
(1) civilisation.
In broad strokes, pre-contact literature grⁱ.stˣtʲ is more or less the same
as any other literature. Instead of discussing details of the world, the focus
on the story is largely based on the idea of network reshuffling, tying in to
the religion described in section 2.1.5.
A typical pre-Contact literature schema would involve an initial network
i, which is then perturbed by a change P in one of its nodes’ properties. An
interative process c is applied to the network such that the final network,
f , satisfies some rules as established by the genre of the fiction. In effect, it
would be filling in objects that satisfy the equation:

f = (cn ◦ · · · ◦ c3 ◦ c2 ◦ c1 ◦ P )(i) (29.1)

29.4. GRADED LITERATURE 285

where each c and P are considered functions that act on the network i,
leaving ◦ as the usual function composition. The equation does not specify
the exact nature or properties that i, f must satisfy, but those are part of
the literary tradition as well.
Note that although we have described the structure of the literature,
in practise the structure is always embellished with the religious overtones,
which is of course where the structure derives from in the first place. In
this case, the networks have nodes that represent individual gods, with the
connections between nodes being whatever relationship system the story
wishes to explore, and P would be an addition of a new node, or a removal
of an existing mode. c would then be to modify the network by changing
one edge at a time until the network satisfies properties that i did in the
first place, giving us f .
This general structure is all that survives in the written and spoken
record, but it has been revived as it is and is now considered a literary
tradition with an associated territory (which is the entirety of Apurhagat).
This revived structure, which may or may not bear any resemblance to the
actual pre-Contact literature, which is retroäctively given the name “Classi-
cal Network Reshuffling” (CNR), is called “Neo-Network Reshuffling” (NNR)
to distinguish between the two.

29.4.2 Post-Contact Literature

Post-contact literature is much more definite, as there is a large corpus of
available text. This type of story is called an “allegory description”, which
in Drsk is called grⁱ.ðhtʲ.
Common to all contemporary types of literature, this does not involve
any sort of storytelling as we know it, but instead revolves around a concept
known as “scenario testing”. In it, an equilibrium eqm is proposed, which
is then repeatedly affected by some perturbations pertn[n]. After decid-
ing the consequences of such perturbations consq[n], we reset back to eqm
and start describing a second perturbation again. This cycle repeats until
the perturbations are exhausted, at which point we conclude concl that
eqm is one of stable (does not change greatly from any of the perturba-
tions), unstable (changes greatly from all perturbations), or saddlestable
(changes greatly from some perturbations, but not all). Diagrammatically,
this process is as in figure 29.4.
As with other literary traditions allegory-descriptions are fairly heavily-
bonded to the reälity of things: they describe reälistic situations that one
might encounter on an occasional basis, and serve as an aid to remember
certain (moral, social) values that are treasured in the culture.
The reason why allegory-descriptions stuck around in Drsk-speaking re-
gions is because of a particular feature that Grade (1) has which allows it
to mesh near-perfectly with the tradition. More details of the system is
286 CHAPTER 29. STORYTELLING

eqm

pertn[n]

consqn[n] n?

concl

stable unstable saddlestable

Figure 29.4: The structure and main loop of allegory-descriptions. Abbre-

viations are defined in the text.
29.4. GRADED LITERATURE 287

described in section 29.6, but in short, the looped structure of pert[n] and
consq[n] and how you can make n arbitrarily big makes it very easy for
storytellers to just tack on more and more perturbations onto the system, as
many as is required to make up another eqm. This synergy and ease of for-
mation makes it so that post-Contact literature will earn its keep for as long
as series recital is practised, and it’s not likely that’s going away anytime
soon.
An example of an allegory description in full is listed in Chapter 31, which
gives a play-by-play of how each of the components above are involved.

29.4.3 Other literary traditions – “foreign grades”

These are well-established literary traditions that have territories, but their
territory does not extend very far into Klzdmk so Drsk literature in these
traditions are scarce. For this reason, they are typically termed “foreign
grades” when one wishes to distinguish them from other graded literature
that are not the (classical) Network Reshuffling from pre-Contact, or the
Allegory Description from post-Contact. We’ll run through them briefly
here. (27 | 45)

Hammering a Function (HAF) Take a single known function f , and

then repeatedly apply it to inputs {an } to create a new set {f (an )}.
Constructing the latter from the former results in a large amount of
tension if a well-chosen f can produce unpredictable results given sim-
ple {an }.

Number and Itinerary Recital (NRE, IRE) These two are “recitals”,
which are a group of traditions that are similar in form but inde-
pendently invented many thousands of kilometres apart. In a recital,
there is a list of objects, which may be preässembled or built on the go
with the listeners. For NRE, the objects are numbers that addition-
ally satisfy some condition (such as a recurrence relation or a common
predicate that all numbers must satisfy). For IRE, the objects are
place names, which usually follow a trip across some existing road or
track or a proposed one.
Recitals, especially IRE, has significant overlap with other media, such
as art and music.

Network Reshuffling (NTR) Though previously discussed, the idea of

Network Reshuffling is also a tradition that extends beyond the home
region of Grade (1), again mostly by coïncidence. Section 29.4.1 dis-
cusses network reshuffling in general, but it equally applies to the mod-
ern implementation thereof named “Neo-Network Reshuffling” (NNR).
288 CHAPTER 29. STORYTELLING

For so-called “Chaotic Network Reshuffling” (CNR), the primary per-

turbation P is easily identified as a protagonist, which is the only
occurrence of a character in any of the traditions.

Manufacture (MFR) Creating things is a perennial favourite topic amongst

many, and so the literary tradition of Manufacture is also fairly pop-
ular. Here, we discuss how a single object – which can be a common
household item, a grand infrastructure project, or similar – and dis-
cuss its creation from its raw materials and the processes need to turn
them into the final product. It is, quite literally, a story of the creation
process.

(Virtual) Annealing (ANL) In a similar manner to real annealing, vir-

tual annealing starts off with a very unstable situation and a high
“temperature”. As time goes on, this “temperature” decreases, which
forces the state of the world to fall to some minimum, which may or
may not be a desireable one. The potential of this tradition should be
fairly obvious.

Heartbeat of a Network (HBN) Networks themselves may be the tar-

get of literature, and their evolution traced by prose. The network in
question is more literal than usual – sometimes it’s transport, other
times it’s money – and an additional “fluid” is introduced, which flows
around the network through the edges in between nodes, providing the
required time evolution.

Axiomatised literature (AXL) This is the most pathological of the tra-

ditional literature traditions, and more resembles certain types of math-
ematics than a genuine “fiction”. The first part of an Axiomatised
Literature is to discuss the eponymous axioms, ground rules that set
the scene story, usually in the form of assertions. Then the remainder
is the theorem stage, where conclusions are created from the axioms.
The mathematical nature of this literature tradition is strong enough
that it stands alone as a category of traditions.

Out of all these the most common traditions that are seen in Drsk can
be attributed to geographical proximity, and are Neo-Network Reshuffling
and Itinerary Recital. NNR is not particularly common due to its accidental
suppression, so the most common foreign grade written in Drsk is IRE.

29.4.4 Modern grades

More modern traditions don’t have any associated territory, them being
relatively young and have known origins, but they still are fairly popular
nonetheless and compete in the same arena as the other literary traditions.
29.5. FACT-TELLING 289

We’ll also run through them briefly here, though as with the foreign grades
literature in Drsk in these grades remain sparse.
Duelling Plotlines (DUP) (27 | 48) DUP has two plot that both inter-
act with the same initial world iworld to create fworld1 and fworld2.
Then, both plots then affect iworld at the same time, and that means
that the world would be pushed toward the two. and if the proportions
are selected right that means that the real fworld would not be the
same as fworld1 or fworld2.
Phase Space and Meta-time (PSM) Consider the amount of space a
plotline can follow in an abstract phase space, and then consider what
happens if the phase space has some restrictions on entry, which pre-
vents some obvious complications and resolutions. If we now introduce
the idea of a meta-time, which runs independently to the time in a plot,
and you will begin to appreciate the key idea of PSM.
Modern grades of fiction are often influenced by the formalism that un-
derpins Worldstate Literature, and seeks to modify it or add extra elements
to it in a radical manner that does not harmonise very well with the whole
system. In particular they violate some parameters in the system that re-
quire there to be a fixed number of objects in the story subjects. For in-
stance, animate can usually be unlimited in standard WSL, but in reduced
literature (RWL) animate is prohibited, so it is nil. But consider the situ-
ation where there can only be exactly one object that makes up animate,
and you have a novel literary tradition known as “Unitary Manipulation”
(UNM).
Modern grades are not only sparse in absolute terms, they are also rare
compared to those literary traditions with a historical territory. Close to
no original books are written in these traditions in Drsk, and are instead
translations from places where they are more popular. As with the rest
of the Northern Hemisphere, creation of modern grades is comparatively
lacking to the Southern Hemisphere.

29.5 Fact-telling
One consequence of the behaviour of such literary traditions is that there’s
no requirement that anything that was said has to be false, and in this case
that implies that there are some stories that are, in some sense, true. This
is of course entirely predictable if we consider the fact that descriptivist
religions initially started with this mode of storytelling, but in fact it goes
much further beyond and before them, and it is a well-respected tradition
for stories to be an accurate representation of what has genuinely occurred
in the past, beyond some warrior’s yarn about his exploits in the battlefield
or an explorer’s account of what he found in the sea last week.
290 CHAPTER 29. STORYTELLING

Fact-telling in this case refers to the act of using storytelling techniques

to describe an aspect of life that is purported to be true by the storyteller.
Uncontroversially this involves the kind of story that one “tells” by simply
relaying what happened to the speaker last night on the bus or similar, but
the distinguishing factor that makes fact-telling truly different from mere “a
funny thing happened to me on the way to the bar” is not only that such
stories can be elevated to long-from literature, but also that non-experiential
ideas can be relayed, such as an mathematical proof or a scientific experi-
ment.
This last part also gives a clue as to the usage of stories in formal writing.
Specifically, when it comes to relaying scientific findings, one common format
is to use fact-telling.

29.6 Series recital

This is not to be confused with the two recital literary traditions, as de-
scribed in 29.4.3.
Series recital is a method of storytelling that involves the story being
given over a long period of time. This should not be entirely unfamiliar, as
it is almost exactly identical to the idea of “story rounds”.
Series recital comes almost as old as the history of civilisation itself, as it
is easy to create entertainment in a way that doesn’t require any knowledge
of reading and writing; all you need is a working pair of ears and a healthy
imagination, both of which are readily available. Because of its extreme age,
however, there is a heavily formalism to the whole tradition, which we will
outline later in this section.
Generally speaking, series recital is done orally, with relatively little writ-
ten record in its earlier years (later attempts at literacy did eventually put
the practice into writing). Each section is roughly large enough to just fit in
one’s memory. As a small cultural note, this is typically done in a particular
time in the day.
In Klzdmk, this time of day changes over time. In the northern schedule
(see section 4.1.1), this particular schedule normally allocates time for series
recital during the night of day 1, a rest period, or the normal recreation pe-
riod of the afternoon of day 3. Otherwise, in the southern schedule described
in 4.1.2, the time is normally in the night of day 3. In the more modern
times, where triads give way to days, it is normally done at night, and has
been increasingly formalised to become a theatre performance of sorts.

Figure 29.5: The typical seating arrangement in doing series recital.

29.6. SERIES RECITAL 291

Series recital is performed in a small to medium group setting, with a

single orator sitting in a prominent position and one to about 20 listeners
(typically about 7) who all sit in a rough arc of a circle (see figure 29.5).
The orator, who normally either has a short cue sheet or none at all, starts
the story with the phrase:

(100) a a-p a
/aa/ /bb/ /cc/
DIST_PAST unfold-NSC happen
A long time ago, this happened.

Such is the age of the tradition that as the example shows, it is gener-
ally expected that the words involved have fixed vowels. This is commonly
obeyed across all communities, but with different pressures: more conser-
vative communities generally are resistant to language change in this area,
whereas more innovative communities are pressured to keep the old vowels
by other languages nearby that do have vowels. There are some communities
however where the vowel insertion is more variable.
If we’re picking up from the middle of a story, the phrase is then:

(101) a a-p a
/aa/ /bbp/ /cc/
DIST_PAST unfold-NSC proceed
From previously.

which may or may not include a recap of the previous rounds. As with
the other sentence, vowels are essentially static for this phrase.
The story is then told using the local vernacular, as it must be as a com-
mon writing system was not invented when this tradition is most popular.
The typical length of the session is usually about 30 min to 60 min, which is
good fraction of the day. Certain orators can go all the way up to half the
night (about 3 h).
Once the end of a recital session has been reached, the end of the story
is signalled with this sentence:

(102) a a-a a
/aa/ /bbp/ /cc/
PAST unfold-NSC walk
And so it continues.

This is used even if the end of the story has been reached. Once again
the vowels are included, though their values are more variable than the other
two.
292 CHAPTER 29. STORYTELLING

Over time as triads went out of fashion and days take over so have the
nature of series recital. As previously mentioned a number of orators manage
to grow out of their community and create *theatre troupes, which become
the primary method of story-based entertainment in the Grade (2) era.52
As a legacy, most of these *theatres operate in the evening and night and
sometimes food is served, especially if sanitation is not a particularly pressing
concern (inevitably, that means that only high-class *theatres have them –
there’s a number of complex social events that come with the intersection
of food safety and *theatres over the years but they’re all roughly the same
and are ultimately beyond the scope of this grammar).53
We have not particularly included the role of the exact literature body in
this description, but as it turns out the native grades of literature dovetails
quite nicely into series recital. The allegory-description is very easily seg-
mented into a series recital, by making each session being one combination
of pertn and conseqn. As the value n is unbounded, the story can conceiv-
ably go on for as long as there are ideas to continue the loop, which might
come from another orator, which in turn can become a form of audience
participation.

29.7 Literature and media

In a strict sense, literature requires only the combination of text and only text
to form a coherent story or other emotional generator. However, in practise
literature in Pasaru is not limited to text in general, even in the early days,
and remains heavily intertwined with various other media, primarily sound
with visuals playing a secondary role.
In broad strokes, the history of how literature and media combine had
developed more or less intuitively, with the invention of more and more
technology being combined with the basic idea of speaking words to an
audience, and creating more and more complex nuances. Ultimately, this
developed into an entire industry, with media franchises just like how it
turned out in contemporary Earth.
Nevertheless, there are a number of divergences in the development of
literature and its integration with media. Primarily, notice that there used
to be a division between the two; there can be such a thing as media with-
out much literature, which is still fairly intuitive – we’ve in fact encountered
them before in section 4.6, where we have mentioned music without words.
In deep historical times, there used to be a divide between the two, however
this boundary has largely been erased during the early to mid Grade (1)
52
For other types of entertainment, refer to section 4.7.
53
Unlike other asides, this will not be elaborated upon in an additional article. This is
primarily because the requisite background material has not been assembled at the time of
writing this grammar, and that it’s genuinely just hard to write in the first place.
29.8. LITERATURE DISTRIBUTION 293

period (i.e. not that much longer than the start of history for Klzdmk).
Additionally, the existence of multiple literature traditions happened to cre-
ate “territories” where the traditions tend to dominate mutually exclusive
media.
In the following sections we will briefly introduce some of the more impor-
tant eras of how literature takes over media, in a more or less chronological
order. A timeline is provided in figure … for easier reference, though it lacks
exact timing information due to the nature of the subject matter.

29.7.1 Recitals to books and theatre

The two recital traditions require almost from the outset an aural tradition.
This is the oldest form, where an orator just says words to an audience. The
situation then evolves into the series recital tradition that we have described
in section 29.6. Eventually, writing was invented and books with text were
used to extend the longevity of any individual story and to codify them into
a library.
History later developed the practise of series recital to theatre, which
was again earlier described in more detail in section 29.6. This

29.7.2 Song
29.7.3 Modern media
29.7.4 Franchising

29.8 Literature distribution

Inception Construction

Keep private Publish

At least 4 generations 30 years

Loss by forgetting Family mythology The Open Season

Figure 29.6: The general lifecycle of a story, from inception (top) to any of
the fates at the bottom.

(36 | 6)
294 CHAPTER 29. STORYTELLING

29.9 The Battle of the Bodies

Generally speaking, there is a strong correlation between these three things:

• The language something is written in;

• The general literary traditions (“literature body”, “literary body”) that

a story is written in; and

• The location where the story is written in.

Thus, in Drsk, the vast majority of its body of literature is not written in
International (i.e. Worldstate) Literature but instead in its own native lit-
erary tradition, i.e. allegory description (section 29.4.2). The tradition has,
like all graded literature, an associated territory and an associated civilisa-
tion that allows it to remain existent and stand its ground amongst the rest,
in something of a diverse patchwork of different storytelling frameworks that
mostly do not interact with each other.
However, with the advent of international travel and long-distance com-
munication, this geographical standing is no longer as firm as it used to be.
Individual authors and guilds of authors are increasingly attempting to write
using different traditions, including newly invented ones (see section 29.4.4),
and of course the big winner in this instance is the international literature
tradition. So much so, in fact, that it has a physical manifestation geo-
graphically: Vallzr’s54 Park in Yamtoll (Eamttel), (34 | 29) which not only
has a large amount of foreign influence in native territory in terms of brand
identity and other identifying features but is also the symbol of Worldstate
Literature in the land of the allegory-description – so much so that it has
been described as an outpost, or a beachhead. This is a subject of much
debate in Klzdmk, and many satirical posters have been drawn showing the
greatest figures in native Drsk literature “fight”55 against the figures shown
in Vallzr’s Park.
Nevertheless, a large proportion of the total amount of storytelling in
Drsk, and virtually every single one of the most iconic and representative of
54
The original and proper name for the company is Važllr. This is occasionally “misnamed”
“Vallzr” in some cases. Originally, this was a simple typographical mistake in one of the many
transport maps that bring transit options to and from the parks, but even the mighty Važllr
Collective, when protesting their own name, cannot contradict the iron grip on toponymy
that transit companies have. Hence, Vallzr.
55
In truth, such posters are more of a map-type deal; without the use of individuals,
what happens is that individual figures are represented as entire countries, and Vallzr’s
Park’s figures are also entire countries. Bonus points for actually using the real geography
of Klzdmk, but this is rare (it’s considered more ethical to use a completely fictional map
instead).
There are other ways one can approach the theme while still using individuals, however.
These can be, and sometimes are since approximately 700 to 1000 PDN, identified as char-
acters.
29.9. THE BATTLE OF THE BODIES 295

the storytelling corpus in Klzdmk, remains using the ADL tradition. The
existence of the park is symbolic of a threat towards that tradition, but at
least for now it would appear that the tradition remains safe in its well-
entrenched geography.
296 CHAPTER 29. STORYTELLING
 Part IX

Drsk in use

297
Chapter 30

The train is approaching

– The 2-79 A-66A2 train to … in nᵘfk is approaching. Please

stand back from the sloped area and be aware of exiting passen-
gers.

299
300 CHAPTER 30. THE TRAIN IS APPROACHING
Chapter 31

The *Horse1 and the Stone

This is a typical example of an allegory-description. The full text of the

story is reproduced below, along with a line-by-line gloss. (23 | 47–51)
pmᵘglˡtʲ klˡptʲ f nkⁱhsˢsˢrtʲ nᵘs. klˡptʲ pmᵘglˡtʲ z ksmᵘlˡptʲ
pmᵘzkglˡtʲ. dᵘrf vᵉm kt pmᵘt cstrzł. ft cť zkⁱrⁱstp pť kshl.
dᵈd pť lzftᵗmᵘvᵉrⁱlˡ pþr. ft kⁱrⁱstp pt nt vpþr cᶜckʲ rcᶜrⁱ cᶜv
rv. ħŋ. ft vkfr nt. ħn s. bcfp þʲcft nt þʲrðr pt kt fs.
dᵘdčyðtᵗkᵏtᵗt f pt kt zm rⁱgᵍ.
A farmer has a horse1 and a stone. The stone is heavy and
the horse is wild. The stone is tied to the horse. If the stone
is detached the horse bolts off to unknown locations. Possibly,
off the edge of the cliff. If the horse is lost, the farmer would
be unable to travel. He starves in this instance. So the farmer
must have both the horse1 and the stone. They are more useful
together than apart.

(103) pmᵘ·glˡ·tʲ k·lˡp·tʲ f nkⁱ·h·sˢsˢr·tʲ nᵘs.

G1 ·horse1 ·s1 G2 ·stone·s1 and G3 ·g3 ·farm·s1 own

(104) klˡptʲ pmᵘglˡtʲ z-Ø k·s·mᵘ·lˡp·tʲ

stone horse1 E-COP G4 ·weight·POSINAL·stone·s1
pmᵘ·zk·glˡ·tʲ.
G1 ·:[uncontrollable]·horse1 ·s1

(105) dᵘrf vᵉm kt pmᵘt

[Link] bind stone horse1
c·st·rz·Ø·ł.
:[security]·:[binding]·:[+flexibility −memory]·rope·s2

301
302 CHAPTER 31. THE *HORSE1 AND THE STONE

(106) ft cť z-kⁱrⁱstp pť k·sh·Ø·l.

PAR rope E-bind horse1 :[geography]·:[g4 ]·::[unknown]·s3 .
dᵈd pť lz·ftᵗ·mᵘvᵉrⁱ·lˡ pþ-r.
[Link] horse1 G5 ·:[+tall +high]·shelf·s6 go-ORD.2

(107) ft kⁱrⁱstp pt nt vpþr cᶜckʲ rcᶜrⁱ cᶜv rv.

a

(108) ħŋ. ft vkfr nt.

a

(109) ħn s. bcfp þʲcft nt þʲrðr pt kt fs.

a

(110) dᵘdčyðtᵗkᵏtᵗt f pt kt zm rⁱgᵍ.

a
Chapter 32

A conversation at the
market

303
304 CHAPTER 32. A CONVERSATION AT THE MARKET
Chapter 33

Regarding a lost *sheep3

305
306 CHAPTER 33. REGARDING A LOST *SHEEP3
 Part X

Additional articles

307
Appendix A

Bugscript

Preamble

This is an introduction to Bugscript specifically written for this

book but is not dependent on it and may be republished at a later
date.

[Insert E.-Pasaru text here]

We hope that with this new script, linguists from across the
world writing in any script and speaking in any language can
freely exchange the exact form of sounds, unencumbered by any
historical baggage and opaque shapes.
— Working Committee on Bugscript, 491 PDN 287°Y

Figure A.1: A sample of bugscript, showing its name in English,

/ˈbʌ[Link]ɹɪpt/.

309
310 APPENDIX A. BUGSCRIPT

Bugscript (E.-Pasaru: nens) is a system of phonetic notation that is used in

the J.-Pasaru to show phonetic information in an extensible, modular and
flexible manner. It was designed to be script- and direction-independent, as
well as highly featural and to some extent even beautiful.
The primary intended usage is to indicate phonemic information in a
manner that largely resembles how mathematical equations are in contem-
porary Earth books: on a line of its own, using a number of symbols with a
significant amount of vertical space which makes it less suitable to use inline
(though it is possible and done). Over time, bugscript has also expanded in
scope to be used to indicate phonetic information, and later to handle al-
most everything phonetics, including syllable structure, phonetic inventory
and phonotactics. It has even been used artistically to depict words in a
way that does not explicitly write them.
Its English name is attributed to its general aesthetics, which tend to
create glyphs that resemble small insects and fish. However, its E.-Pasaru
name makes no such metaphor.

A.1 History and rationale

How bugscript came about is a combination of a number of historical factors
that reflect how the kilis as a species handles languages. Therefore, we shall
examine this from a historical perspective.
From almost the first day of civilisation the species has a particular
affinity to language. We have earlier in section 1.5 shown how this affinity
ultimately makes it so that the language appears to become sentient and
able to make decisions of its own, though this is ultimately illusory and
not a useful model to predict future changes in languages. Regardless, such
an approach makes it so that language as a whole is generally treated with
slightly more reverence than they strictly deserve, though happily this means
that they are also examined more thoroughly than they otherwise would.
This means that the field of linguistics is developed much faster than on
Earth.
Additionally, the species also has a habit for inventing writing. Civilisa-
tions have invented writing many times over in different eras and times, and
there had been some level of writing in even the most isolated of cultures.
Because there is a large number of different environments such cultures in-
habit, there’s no guarantee about which direction the writing goes, what
type of orthography it is using, or even what medium the script is written
in.
The prodigious number of scripts and even writing media plus the fre-
quent contact that the major civilisations have with each other makes it so
that no single script that has risen to preeminence in the same way that the
Latin script has in contemporary Earth, at least at the time when Bugscript
A.2. BASIC STRUCTURE 311

was invented. That means that when it comes to notating language in a way
that does not use the native script – which is useful to maintain neutrality
and also to avoid nasty orthographic interference that script might have – a
completely new script would have to be invented.
And so bugscript is born, out of a design set by the International So-
ciety of Linguists, in the year −745 PDN. (20 | 55) The idea is that now
phonetic sequences are treated much like mathematical equations when it
comes to typesetting, and additionally be featural, so that the structure of
each phoneme can be laid bare. It was largely accepted by the general pub-
lic, though its bulkiness does leave some wanting. This is later addressed by
evolving its design into a cursive variant, which we will describe in section
A.11.3.
Technically, though, bugscript was born a few decades to a century ear-
lier in earlier forms, informally in academic circles where a need for pho-
netic notation was recognised, particularly in areas around the academic
heavyweights around the time, named the Edrensano Rim because it is the
collection of countries that tightly surround the western Edrensano Ocean.
This origin strongly informs the construction of the script, in particular how
bodies are allocated to which sections, but over time and through the stan-
dardisation the bias toward languages spoken in the Edrensano Rim was
lessened greatly and only a few remnants remain, though they are no less
annoying to explain for having been fewer in number.
There does remain some outstanding issues with the compatibility of
bugscript with every script of the J.-Pasaru, which will never be resolved
due to the dizzying diversity of them and how such diversity can only ever
grow, never shrink. For example, bugscript is not easily embedded with
the knitted orthography of Cindri, though it is largely compatible with its
“written” (carved) orthography. Ultimately however, the writing medium
is still constrained to be ink on paper or something that can emulate ink
on paper, as the International Society have decided that at this point the
flexibility of bugscript ends.

A.2 Basic structure

Bugscript is largely composed by closed figures called “bodies”, which are
simple geometrical shapes such as semicircles and rectangles. These are
strung together in a linear fashion along with spacers. Each body represents
a single segment which is considered to be distinguishable from other seg-
ments that are next to each other. There are also other figures that perform
the function of bodies but are more items that connect bodies to each other;
finally bodies may also have accessories to them, which may be placed above,
below or inside itself.
312 APPENDIX A. BUGSCRIPT

These segments are placed onto a grid, which is composed of squares of

a particular abstract size. This abstract size is known as the “semibody”,
with symbol sb. The height of all bodies is 2 sb, hence its name. This unit of
measure is further divided into csb, which is due to the base-12 nature of the
1
kilis civilisation equal to 144 sb. The semibody is used as the intermediate
unit to describe the size of a typeset string, e.g. 1 sb = 15 mm.
There exists a centre line for all bodies to lie on, which is called the
centre line and is given the coordinate y = 0 sb. Each body has its own
origin point, which is normally on one side of the figure or the other, but
which side is not predictable and it might even be dependent on whether or
not the figure is inverted or not, so there’s no single rule that can be stated
to where the corresponding x = 0 is.
Although we have stated that the grid is that of squares, that can be
changed to conform the figure to the surrounding context. In that case, the
definition of 1 sb is altered to mean one lattice point in the altered grid, and
the conversion rate to concrete measures is nominal.
The direction that this string of bodies are listed is in the order of speak-
ing; however the actual direction of writing is not fixed by the script but
rather by the context that surrounds it. To make the direction explicit,
a small mark can be added to indicate the start point and the direction
of reading. (30 | 63) Note that some glyphs change slightly based on that
direction.
For the remainder of this article, bugscript will be written left to right.

A.3 Measures
To further define the typographical properties of a particular iteration of
bugscript, some additional heights and widths are defined. These are in-
dicated in context in figure A.2, and we will describe them in this section.
(30 | 73)
First, let’s have a look at the horizontal points. These are as such:

Origin The origin is the anchor point from which the body is defined. It is
usually notated O. In some bodies, this is the same as point B. For
most bodies the part of the body that intersects this point is normally
flat and vertical.

Flat front The flat front is the front of a body that has a flat surface in
the front. It is distinguished from a round front by the fact that the
body at and around this point is normally parallel to the back of the
body. It is notated B.

Round front The round front is the front of the body that has a round
surface. It is notated C.
A.3. MEASURES 313

A2 A1
A
G2 G1
G
M2 M1
M
S2 S1
S
B O E
C O, B
−S2 −S1
−S
−M2 −M1
B
−M
−G2 −G1
−G
−A2 −A1
−A

−N
(a) Round body (b) Vowel body

Figure A.2: The grid that bugscript measures are based on, with some key
heights marked. Grid lines represent 0.5 sb.
314 APPENDIX A. BUGSCRIPT

End This is a point with relatively little importance except when belts are
involved. It represents the minimum amount of space that a belt or a
space requires for the body to be attached. It is notated E.

Then we have a look at the key vertical points:

Stripe height The height where most stripes attach to. This is given the
name S.

Middle height The point where some stripes and other accessories attach
to. It is normally the midpoint of the line OA. It is named M .

Gill height This is the height where gills stop at. It is named G.

Semibody height This is the height of the body above and below and is
the primary reference distance that determines the writing’s size. The
point is named A, and the length of the line segment OA is the script’s
size, named s.

Antenna height This is the height to which antennae extend up to. It is

similar to the ascender and descender height of the Latin script. It is
named N .

For bodies where B 6= O, each of the above points are split in two, where
the ones above B are given the subscript 2, and the ones above O are given
the subscript 1.
Finally, there are some constraints that exist with regard to the distances
of each of these points:

• Because of the circular nature of the most common type of body, we

have OA = BC.

• For bodies with B 6= O, convention dictates OB = OE. This also

dictates the length of OE for bodies where B = O.

• The ratio OBBC

, though not given a name, is constrained and should
exist in the interval (2, 3].

• The antenna height is normally constrained by ON1 = 2OA. This

can be varied, especially in cramped environments, to be as small as
2 , depending on whether or not bubbles exist. If they do, then the
3

minimum size of the bubbles would constrain this value from below.

These limitations are obeyed in figure A.2.

On “bugscript grid paper” – squared paper that is used to write bugscript
in – each square has the size 0.5 sb (i.e. a body would span four squares)
and there is also a ½-square gap between every 8 rows of squares. Therefore,
A.4. TYPOGRAPHICAL TERMS 315

bugscript grid paper fixes the vertical points O, M , A and N and so the
horizontal points B, C and E. The grid squares are normally very fine,
defining 1 sb ≈ 4 mm and so having grid squares of approximately 2 mm.
Figure A.2 uses the form of such grid paper, but of course uses a much
larger grid to allow for annotations to be indicated.

A.4 Typographical terms

The typography of bugscript is highly defined and was set along with the
script’s promulgation. The reason for this is historical; by the time the
script was invented typesetting has already been invented and as one of the
purposes of the script is to be able to write it in a printed book, a precise
specification is necessary. It is also important to distinguish the graphical
component of bugscript with the semantics that these components represent.
Therefore, we’ll define the typographical terms here, and mark it with
some examples here. (30 | 49) (30 | 53) (30 | 54)

A.4.1 Bodies
As previously mentioned, the body is the fundamental component of bugscript,
and is any closed figure declared to be as such. It is the fundamental object
of bugscript from which every other object is ultimately derived through re-
lationships from. As such, it’s not particularly easy or meaningful to define
what a body is, other than enumerating the common ones and defining new
figures as bodies.
Nevertheless, a body usually has the following properties which we can
refer to easily:

Front Also called “head”. This is the part of the body that is normally
disallowed to touch any other bodies. It is also the part of the body
that is nominally allowed to touch accessories. Some bodies have an
inside and an outside, an example of which is seen in figure A.3. There
may also be a space where the body may touch another body’s back
only; this is the front’s bonding point.

Back This is the part of the body where only another body or a belt can
attach – accessories cannot attach here. Note that for some bodies,
such as the vowel body as seen in Figure A.6b, there is no front but
instead two backs, one of which is where one might intuitively surmise
a “front” might be. This is a slightly confusing twist of terminology for
some.

Append A body-like object that attaches to a body. It looks like a body

in the sense that it is a closed figure, that at times it cannot easily
be said to depend on a single parent body, and that it can attach
316 APPENDIX A. BUGSCRIPT

to backs of bodies, but are distinguished from them by the fact that
they’re generally smaller than full-size bodies, are subordinate to some
body (or combinations of bodies), and that they are not belts. In a
sense they are the opposite of a free body (see …). Appends normally
have backs but no fronts, much like how the rectangle is. They always
attach to bodies on the outside or back.

Halo A body-like object that does not attach to (touch) the body. Only
appears next to fronts, and by definition is separated from it by some
whitespace. They cannot support stripes, gills, feelers, antennae or
bubbles, but can accept pedestals.

These are called the body’s parts. A visual

reference is available in figure A.5. (3)
Some bodies and accessories have an orien-
tation. (30 | 63) If they do, then that means
that they would have a slightly different ap-
pearance when the script is written right-to-left
(2) (4) (1)
than they would be if it was written left-to-right
(which is the default). The round body, in fig-
ure A.6a, is a typical example of a body with
an orientation (i.e. a chiral body). For this
case, because we are working in a left-to-right (3)
context, the body with the round part facing
left is the ortho form, whereas the body with Figure A.3: The inside
the round part facing right is the para form. In front (4), outside front (2
a right-to-left environment, these would be re- and 3) and back (1) of
versed. Figures that merely mirror themselves a body. (2) represents
between the ortho and para form are called cis- an outside front that has
morphic, whereas those that do anything else “bonding capability” – it
are called transmorphic. As a special edge may touch another body’s
case, figures that are vertically symmetric and back. The body itself is
therefore don’t change under ortho-para trans- coloured in red.
form are also cismorphic, since they are consid-
ered to have undergone a mirroring operation
that just so happens to do nothing.
Finally, there’s also two extra umbrella terms:

Body-like A body, or something that encloses a space and is derived from

them. They also have fronts, backs, or both.

Marking Any stroke that does not form part of a body-like.

Figure Any discrete object relevant to bugscript, be it body, accessory or

space.
A.4. TYPOGRAPHICAL TERMS 317

A.4.2 Special bodies

There are exceptions to the fact that bodies have all the parts described
in section A.4.1. Three particular exceptions that are especially common
include sausages, blackballs and free bodies. These are special because
they have neither a front nor a back, and additionally cannot touch or abut
other bodies (see section A.4.3). They are depicted in figure A.4.

Sausage A body with a variable width that normally does not accept acces-
sories. It normally represents a number of other typographical objects,
which may be left unspecified. It has most of the properties of a body,
except it cannot have an append. Sausages may accept antennae or
bubbles and similar antennae but the semantics associated with such
a typographical combination may not always be well-defined.

Blackball A special body that is severely limited. It is frequently described

as a “final atom”, in the sense that it is an indivisible object with no
further substructure which also cannot be modified or altered in any
semantically meaningful way. The blackball has no front, no back, can-
not have appends or halos, and cannot accept any kind of accessories
for attachments. It even is coloured in its entirety to emphasise this.
However, it is allowed to have pedestals attached to it for semantically
meaningful combinations of what a blackball represents and the usage
of orbscript.

Free body A free body does not have a back, and cannot touch any other
body either directly or through a belt. Semantically it may not even
represent a phonetic segment, but may instead represent metadata.

Table A.1 summarises the differences between these bodies and normal
bodies.

Figure A.4: Drawings of examples of the special bodies.

A.4.3 Body interactions

Touching Two objects touch when they share a stroke along their backs,
in whole or in part. This includes when they don’t line up “properly”,
with the ends of the backs aligning end-to-end. This also includes two
bodies being connected with a belt between them, by definition of a
belt.
318 APPENDIX A. BUGSCRIPT

Table A.1: Properties of the special bodies.

Name Sausage Blackball Free body

Accepts stripes and gills X – X
Accepts feelers – – X
Has backs – – –
Accepts antennae See text – X
Accepts bubbles See text – X
Accepts pedestals X X X

Abutment Two objects abut each other if they would otherwise touch if
a small space between them are removed. Strokes between the two
bodies and their accessories must not touch the other for abutment to
be occur. The space between them cannot be not wider than 48 csb,
otherwise it’s just a normal space.

A.4.4 Accessories
Bodies may have accessories, which are small markings or other figures
that belong to the body only, (and not, say, connecting two bodies together
– those would be belts, and belts are not accessories). These are visually
indicated by having the body either include the accessory or be the closest
one to it. Note that the belonging of an accessory to a body does not
necessarily correspond to the fact that the body’s corresponding phoneme
has total ownership of the accessory’s corresponding modification. (30 | 53)

Figure A.5: A colour-coded version of a body and some accessories it has.

Below is a list of accessories and their names, which are also illustrated
in figure A.5.

Stripe Any marking appearing inside a body and attaches to its front with
a bond.

Gill Any marking appearing inside a body and attaches to its front outside
of the bond area.

Feelers Any marking appearing outside a body and attaches to its front,
usually in the bond area.
A.5. SIMPLE PHONEMES 319

Antenna A marking attaching to a front outside the bond area. They

can extend beyond the normal bounds of y = ±1 sb, unlike other
accessories.
Bubbles Small blackball-like objects that appear in the same place that an
antenna can. Unlike antennae, they can be asymmetrical with regard
to the centre line.
Belt An object that is drawn to connect backs of bodies together. Any
bodies that the belt touches on its back are also connected to each
other.
Pedestals and Orbs Two objects that work with each other. Orbs are
independent objects that form part of orbscript and isn’t strictly part
of bugscript itself, for which see section A.11.1. Pedestals link up orbs
with a body, however unlike accessories there can normally only be
one body out of a sequence of bodies that are touching that can have
a pedestal, and the orb that the pedestal supports is associated with
all bodies that touch the body that in turn touches the pedestal.
Pedestals are also, due to their associated semantic value, asymmetric.

A.5 Simple phonemes

A.5.1 Basic bodies
For most languages that have conventional consonants, vowels and the usage
thereof, all bodies in use are shown in figure A.6.

(a) Pulmonic conso- (b) (c) Nonpulmonic oral (d) Approxi-

nant Vowel consonant mant

Figure A.6: Basic bodies of conventional bugscript.

For most of these, the meaning of these bodies are transparent; however
figure A.6d is a little bit tricky. The approximant body, as it is named
when used semantically, is used for phonemes that are either liquids (e.g.
/l/, /r/) or approximants or semivowels (e.g. /j/). This includes phonemes
that are pulmonic, so actually the round body does not actually include
all pulmonic consonants. The combination of the round and approximant
bodies do together include all pulmonic consonants however.
320 APPENDIX A. BUGSCRIPT

The existence of the approximant body is a historical detail, back when

they were considered to be halfway between “real consonants” like /k/ and
/s/ and “real vowels” like /i/ and /a/. The shape of the approximant body
even alludes to this, and because of this its accessories are subject to separate
rules not related to those that attach to the round body.
Additionally, in the strictest sense these bodies do not actually indicate
the consonant type but rather the role in a syllable, with the bodies pointing
left being onsets, bodies pointing right being codas, and bodies with no
direction being nuclei. This is particularly important when discussing the
phonology of languages that do not possess syllables, which we will discuss
in section A.8.

A.5.2 Consonants – articulation accessories

Speaking of accessories, we shall introduce the common accessories that are
used for these bodies. First, figure A.7 shows the common accessories for
the round body that represent the place of articulation.

(a) Labial (b) Labiodental (c) Dental (d) Sibilant

(e) Retroflex (f) Palatal (g) Velar (h) Uvular (i) Glottal

Figure A.7: Round body accessories for place of articulation.

The place of articulation for the round bodies are generally placed in
order, where the front of the body represents the front of the mouth, and
the back of the body represents the back of the mouth, with the figure of
the accessory itself further reïnforcing it – notice that the labial accessory
protrudes even further front of the body, whereas the velar, uvular and
A.5. SIMPLE PHONEMES 321

epiglottal accessories start in the same place but the ends move further
and further back in the order of the corresponding mouth position. Note
however that the specific design of each accessory does not have any direct
relationship with the place of articulation.

(a) Plosive (b) Fricative (c) Nasal (d) Tap

(e) (f) Trill (g)

Semivowel Rhotic

Figure A.8: Accessories for manner of articulation, placed on appropriate

bodies.

Figure A.8 shows the round body with accessories that represent the
manner of articulation. Here the relationship is more direct and visible:

• The plosive accessory is a dot, reflecting how they are a short inter-
ruption in the airflow;

• The fricative accessory is two parallel lines, reflecting how the tongue
(or lips) close against each other;

• The tap accessory is a short line, reflecting the short amount of time
the airflow is closed;

• The nasal accessory represents the shape of the tongue as it closes the
entire airway in the mouth, making air go through the nose;56

and so on.
Due to the approximant body taking over for a number of duties of repre-
senting liquid consonants, there are fewer of these than one would normally
56
Earlier drafts of bugscript do include a nasal body, meaning that the round body can
more accurately describe a “normal kind” of consonant – an oral, non-liquid pulmonic conso-
nant. This was abandoned eventually on the rationale that nasalisation is better described
as an accessory and bodies are not easily transformed into one.
322 APPENDIX A. BUGSCRIPT

expect from consulting an IPA chart, but where appropriate these manner
of articulation accessories can also go on the approximant body.
There are some additional dark clouds to this otherwise intuitive arrange-
ment of figures which were largely brought about by historical considerations
– as described earlier in A.1, the script unfortunately already has a small
history behind it when it was being standardised, which made some com-
promises to the featural properties of the script which would otherwise not
be made. The compromises are most visible at this point, where not only
is the manner of articulation shared between accessory and body shape but
also the fact that the “lateral”, as a specific class of phonemes, is indicated
by the approximant body having no manner-of-articulation accessories.
Now, we have enough to create a small subset of the IPA. To create
a phoneme, all we have to do is to compose the body with the correct
accessories. For instance, /ç/ is a palatal fricative, so it is notated as a
round body, with the fricative and palatal accessories:

A.5.3 Consonants – other details

Figure A.9 shows the round body with accessories that represent other fre-
quently used articulation modifiers.

Figure A.9: Round body accessories for other articulation details.

The most important of these is the voicing mark, which now allows us to
write the remaining phonemes that are notated as single characters in the
IPA. In bugscript, voicing is treated as on the same footing as certain other
things that are indicated as diacritics or modifiers in IPA, such as aspiration.
Our first foray outside of the standard IPA is the ability of these partic-
ular accessories to vary for indicating “half” or “double” versions of certain
articulations. This is used for certain languages in eastern Sitħanor, which
does include half-voicing and double-aspiration. Gradations like these are
demonstrated on /k/ in figure A.10. Further levels of aspiration, and indeed
any other kind of gradation, can be indicated by writing a number indicating
how intense the gradation is.
Note that if gradations are used, features that are indicated by pres-
ence or absence of a particular accessory are instead changed so that the
A.5. SIMPLE PHONEMES 323

Figure A.10: The phoneme /k/ being modified with various levels of aspi-
ration. In order from left to right: explicitly unaspirated, half-aspirated,
aspirated, double-aspirated, and n-aspirated. See text for further details.

accessory is always mandatory, and the absence of that accessory is dealt as

“unspecified”. For more details, see section A.9.

A.5.4 Vowel accessories

In contrast to consonant accessories, vowel accessories are fairly simple and
to some extent more systematic. There is no particular need to describe
them individually; instead, the general idea can be expressed in a single
sentence:

A vowel accessory that indicates the vowel is a notch placed in a

vowel body, with the notch position is determined by where the
vowel is in an IPA vowel chart.

Recall that the IPA chart arranges vowels so that the front vowels are
closer to the left, the back vowels are closer to the right, the closed vowels
are further up, and the open vowels are further down, so no explicit refer-
ence to Earthling constructs are strictly necessary to describe how it works.
Additionally, in right-to-left writing, the back and front vowels are reflected,
just like the positioning of the place of articulation markers.
The vowel accessories are arranged in a vowel chart in figure A.11.

Figure A.11: Bugscript vowels in a vowel chart.

The one other detail that needs to be mentioned is how rounding is

indicated. Simply put, if a vowel is rounded, then the notch is hollow. If it
is unrounded, then it is filled. Gradations of rounding can be achieved by
using hatching in between filled and unfilled.

Figure A.12: The vowel /i y/, from unrounded to rounded.

324 APPENDIX A. BUGSCRIPT

A.5.5 Non-pulmonic phoneme accessories

As accessories handle most things that are not manner of articulation, this
means that there are a whole bunch of accessories that indicate almost lit-
erally everything else.
For example, here are some of the things that are needed to indicate
certain articulation details in clicks and implosive consonants.

A.5.6 Nonspecific accessories

Nonspecific accessories are those that attach to a body which does not ac-
tually represent a modification of that phoneme. This is obscure, and it’s
more illuminating to have a look at an example instead.
The prototypical example of a nonspecific accessory is the stress feeler.
Consider the word /ˈbʌg/, which is depicted in the left of figure A.1. The
stress feeler, which are the two vertical lines stretching out from the vowel
body, attach to the vowel body, and typographically belong to it, but con-
ceptually, the stress actually belongs to the entire syllable, not just the vowel
(note that in the IPA, the stress mark goes at the front of the syllable, and
not, e.g. /bˈʌg/).
There’s no indication of when an accessory is nonspecific because when
that happens is not phonologically inherent – taking the example of the stress
feelers, the conceit is that whether stress belongs to the syllable or the vowel
is entirely up to the language, and in some cases even up to the analysis of
the language, with no real consensus of what is “actually happening”, so
to speak. Bugscript does not handle that kind of controversy and instead
requires the writer to pick a side for it in its stead by explicitly requiring a
particular accessory to be declared nonspecific outside in the meta-language.
Other accessories that are generally considered to be nonspecific include:

• Tone marks
• Volume marks
• Chireme accessories, given that they are notated as accessories and
not as full bodies in their own right
• Special marks that are designated as nonspecific, and can be user-
defined
• Marks that designated an unknown feature, which may be nonspecific
by virtue of the fact that whether or not they can be attributed to a
modification of a single phoneme is unknown, unexplored, unattested
or unexplained.

This is a semantically valid concept defined by the script specification,

but no particular glyph is designated as implementing the concept.
A.6. COMPOUND PHONEMES AND BELTS 325

A.5.7 Sub-accessories
Accessories can occasionally sprout modifiers of their own, which represent
slight changes to how the underlying feature is expressed. They are usually
independent symbols of themselves that are not attached to the thing that
they are modifying, except perhaps my indicating proximity.
The most common example of a sub-accessory is the idea of gradation,
where an accessory is modified slightly to indicate that the idea being ex-
pressed is being strengthened or weakened somehow. A classic example of
that is aspiration, which is indicated by a circle with a dot in it:

1. For strong aspiration, where this is distinct from ordinary aspiration,

the symbol is duplicated.

2. For additional levels of aspiration, the symbol is duplicated, with a

number placed in between the duplicates to indicate strength;

3. For weak aspiration, the symbol has a vertical line through it;

4. For additional weak levels of aspiration, the weak symbol is duplicated,

with an optional number placed between to indicate the level.

Such pattern repeats for all such gradation-based sub-accessories. Note

the usage of the script-specific digits; this can be avoided if this is not
desirable by using “international digits” if needed (usually flag-based digits)
or simply repeating the relevant symbol that many times.
There are some accessories that are not very amenable to this particular
style, and for those a different style is used. …
And then there are “qualitative sub-accessories”, sub-accessories that add
a new discrete category that is not necessarily comparable in absolute nu-
meric terms to the unmodified accessories. This is most frequently seen in
stress markers, …

A.6 Compound phonemes and belts

Some phonemes are actually two other things that look like phonemes that
are put together, and there are also some concepts that are dependent on the
connection between two phonemes. For these objects, compound phonemes
and belts come in handy.
Interestingly, for things like coärticulation, Bugscript solves the problem
very elegantly by simply placing all symbols of articulation in the same body,
so this is not a problem there. If there is however some need to separate
the two phonemes that form a coärticulated phoneme, then one can use a
compound phoneme in the form of a “ribbed” body, seen in figure A.13, or
the affricate.
326 APPENDIX A. BUGSCRIPT

Figure A.13: A ribbed phoneme

Bugscript also contains explicit symbols for affricates and coärticulated

consonants that are meant to be indicated separately, which is figure A.14.
These two concepts are denoted using the same symbol, coincidentally the
same way as in IPA.

Figure A.14: The coärticulation belt between two phonemes.

As previously discussed such items are called “belts”. Such belts can act
as internal components in a syllable, or they can also act as spacing between
syllables, in which case they would indicate something like liaison. A few
examples of these are indicated in figure A.15.

Figure A.15: Further examples of belts.

Belts are also used in other circumstances which we will cover later.
A.7. MODIFIER-BASED PHONEMES 327

A.7 Modifier-based phonemes

A.8 Non-oral phonemes and syllabic denial

A.9 Archiphonemes and indistict phonemes

A.10 Beyond phonetic notation

A.10.1 Notating phonetic inventory
A.10.2 Notating syllable structure
A.10.3 Timing

A.11 Bugscript derivatives

A.11.1 Orbscript
A.11.2 Picscript
A.11.3 Little bugscript

A.12 Glossary of bugscript terms

A.13 Programmatic generation of bugscript

328 APPENDIX A. BUGSCRIPT
Appendix B

Octovexillology

Preamble

This is a re-typeset version of the article “Octovexillology”, available

in https:// [Link]/ Pseudo/ [Link]. It
is reprinted here to allow the book to remain self-contained.
This iteration of the article is as it was in 2016-06-04, with some
minor alterations to allow for successful typesetting and to fix figure
references that won’t make sense in the new environment.
Because it was not written specifically for Drsk, the representative
language assumed in this article is E.-Pasaru. Therefore, a new
section, section B.5, is appended to this article for a glossary.

With groups come identification of groups, and with that come a class of
objects termed graphical group identification device must be created. In the
lands and countries of the J.-Pasaru, this device is called an ovecs /oʊ.vɛks/
(plural: ovexes /oʊ.vɛks.əs/, E.-Pasaru: žtlĭm /ʒt.lɨm/). This document
describes what an ovecs is, how it is used, how one is made, and how it is
different from an Earthling flag (and how similar the two can be sometimes).

B.1 Components of an ovecs

An ovecs is a graphical group identification device that is primarily distin-
guished by the following parts, ordered by distinctiveness:

1. The primary board, noticeably octagonal, called a banner (E.-Pasaru:

sen). Apart from its interesting eight-sided shape, it is also the largest
part of the ovecs. The word “sen” in colloquial contexts can also refer
to the entire flag, but in specialist speech is only ever referring to this
board.

329
330 APPENDIX B. OCTOVEXILLOLOGY

2. The legend (E.-Pasaru: üjebraot), a rectangular board with some let-

ters written on it. There are some rules regarding what can be written
on the board, but while it is separate from the banner, it is an es-
sential and integral part of the ovecs and cannot be removed without
justification or context.

3. Zero or more plates (E.-Pasaru dulĭť), each of which represent a num-

ber or a numeral. These hexagonal plates, despite appearing between
the banner and the legend when used, are ancillary data to the origi-
nal flag, and addition or removal of a plate assembly does not change
the meaning of the flag appreciably, politically speaking. However, it
does contain important information with regards to succession.

4. Last but not least is the pole (E.-Pasaru lëdæn), which is a vertical
bar that supports the entire assembly at its centre. While it might
seem to be a trifling detail to indicate that the pole is at its centre,
this is in fact a very important part of what makes an ovecs an ovecs.

Three of those four items are indicated

in the example ovecs as seen in figure B.1.
(There are actually real example ovexes as there
are provisions to reserve some designs specif-
ically for documentation purposes, but that’s
for later.) But first, let’s go through each com-
ponent in turn.

B.1.1 The Banner

The banner is the most important part of an
ovecs, so most would be forgiven for saying that
it is the only part that matters. While it has
many things in common with the face of an or-
dinary Earthling flag, it fulfils a slightly differ-
ent corner of identification phase space. Figure B.1: An example of
an ovecs, that of Vohalyo.
Names
The banner is always an octagon with internal
angles of 135°, and each corner and side has dif-
ferent names. A corner is called a pœ, and the
edges are called naon.
̲
The long names are all descriptive – naonüjbŕeyaot is on the side of the
legend, hence its name. The short names on the other hand are all named
after units: the corners are the names of the Eight Leaders that have rescued
the nation of Vohalyo from its near death in the year −40 PD, and the eight
B.1. COMPONENTS OF AN OVECS 331

I Px S
Ķ E

L Tx

Pxx N

Ģ P

Æ LZ
T Y B

Figure B.2: Two banners with the symbols for the names of the corners
(left) and sides (right) marked.

Table B.1: List of names of all corners and sides

Type Symbol Short Name Long name Translation

Corner T Troťnu Üjbŕeul̲ First legendary
Y Yœn Üjbŕegi̲ Second legendary
Ģ Ģłyar Pedul̲ First lower
P Puunžol Pedgi̲ Second lower
L Ltģra Azul̲ First upper
Tx Tyaontë Azgi̲ Second upper
I Igléďs Lëdænul̲ First polar
Px Pešvo Lëdængi̲ Second polar
Side B Blťu Üjbŕeyaot Legendary
Æ Ænteħn Pedul̲ First lower
LZ LZaon·gerd Pedgi̲ Second lower
Pxx Pllŭiwën Cŕednul̲ First central side
N Nbřurag Cŕedngi̲ Second lower
Ķ Ķťülzéj Azul̲ First upper
E Eďdorae Azgi̲ Second upper
Š Šbílsa Lëdæn Polar
332 APPENDIX B. OCTOVEXILLOLOGY

sides are named after their first eight descendants. Additionally, the central
point also gets a name, cŕeďnzatť, and the names of the sides can also refer
to the midpoints of those sides, if one explicitly prefixes either name with
the word pœ.

Contents
So what does one do with the banner? Draw on it, of course.
Specifically, to make a flag there are three main parts: lines, colour and
devices. Lines and colour complement each other, and devices sit atop of
them.
Lines (E.-Pasaru: tšop, or pallo in informal contexts) cut up the banner’s
surface into a finite number of regions. Although lines in most flags are
straight, they are allowed to be curved, as long as areas cut up by the lines
are all simply connected and there is an area δa > 0 that is less than all
the areas of the region (the Rule of Non-Infinitesimals: no cutting up the
name surface into any number of infinitely small areas.) The value of δa is
sometimes set by flag manufacturers explicitly.
Lines are defined by their origin and destination points, which start off
from the seventeen origin points defined for every flag (the eight corners,
the eight midpoints and the centre). New points can be declared as the
intersection of two lines, or from sliding an existing point along a previously
defined line. Although curved lines are allowed, they are defined largely
informally, usually by saying that they go across the point “in the wrong
direction by some distance”.
Once all the lines are defined, the colours are placed in. They are usually
only defined as basic colour terms: “red”, “green”, &c. They also obey a much
looser version of the rule of tincture: “metals” now include such colours
as sky and lime, and merely indicate that the colour is “high luminosity”.
Additionally there are some colours that have medium luminosity, which can
coëxist with both metals and colours (low intensity colours). Furthermore,
because of the way the flags are created, gradients are possible in limited
circumstances.
Then, on top of it, come the devices, which is just an arbitrary figure.
They do not obey even the relaxed rule of tincture, and can be any shape,
size and number, but still obey the Rule of Non-Infinitesimals. They all have
one anchor point or two bonding points that is used to define the device’s
position and orientation.

B.1.2 The Legend

Below the banner is a rectangular board. The board has to be white, with a
black border, and has black text. The text must have two or more characters
in it, but it must not have an odd glyph length.
B.1. COMPONENTS OF AN OVECS 333

Figure B.3: The twelve plates in the Standard Set. They represent the digits
1 to 12 (0), in ascending order from top to bottom, left to right.

The origins of the legend is straightforward. Too many groups have the
same banner design, so to differentiate a piece of paper with a short code
was stuck on. Eventually this code was formalised and standardised, and
the thing in general is accepted to become part of the flag. The Rule of
Evens was created to ensure an æsthetical balance.
An ovecs can have more than one legend. This can happen for countries
because it is officially bilingual, or writes in two different scripts in the same
language. Otherwise, it might be because the group it represents has many
names or that the combination of banner and first legend still has a collision.
Legends closer to the banner are considered more important.

B.1.3 Plates
Between the legend or the plates are zero or more hexagonal plates, known
simply as plates (E.-Pasaru: dulĭť). They are ciphers, in the sense that each
one encodes a digit, though they historically encoded an entire integer. They
are nonessential, which means that they can be discarded if space requires
it.
The number that they encode is the flag’s succession index. When a
country has a change in leadership, either by hereditary succession or by
election, the flag is updated to reflect this change in leadership in a process
named succession. This is done by simply incrementing the number written
on the plate by 1.
Because of their nonessentiality, they are also largely unstandardised
on an international level, which means that different vexillological unions,
named pacts (AOďwāg), can have different digits. The picture indicated
above – that of the Standard – is adapted as the international standards for
any new pacts, but existing ones can continue to use their own set if need
334 APPENDIX B. OCTOVEXILLOLOGY

Figure B.4: The standard plate for the relative marker, to be placed just
below the banner. Repeat this plate twice to indicate a negative relative
number.

be. This is also a factor in explaining why the standard set has irregular
behaviour for the digits 9 through 0 (12) – it was adapted as a combination
of the most popular ones, and the combination just so happens to make sense
in its historical content. Specifically, the figures are assigned by adding more
to the previous one, but their position is based on defining a final symbol
and working backward.
The existence of pacts also help in keeping the number small; this is
done in two ways. First, unless there is only one country in a pact, there
must always be a country in a pact that has the smallest succession index
(either 0 or 1). If there is no such country, then 1 is subtracted from all
members’ succession index until one of them hits the minimum. This is
called snapback (nŕaoj) and happens automatically. The other more drastic
measure is to simply declare unanimously that the successor index is to be
reset to 0 at the beginning of some year for all members that did not receive
an exemption.
Plates are useful in indicating the change in leadership of a particular
country and are used extensively in historical maps where an increase in the
succession index is a handy representation of change of leadership. However
in the wild plates are much rarer and usually only appear in the homes
or graves of current and former leaders of any type. They are, however,
permitted to be attached to any flag provided that the correct succession
index is used.
The frequency of snapback largely depends on the values of the pact:
they might desire a large cardinal number to appear on their flags, in which
they will be very rare if present at all; but other pacts, that prefer the exact
opposite, might snapback a lot, to the point of it being a natural occurrence.
However, no nation desires a very large number to show on the flag, if only
for the simple reason that the reader of the flag will complain about having
to process such a large number, or that too many digits would make the
flagpole unnecessarily tall (a taller flag is harder to see from down below
and is therefore considered less important than a shorter flag.) Nations deal
with this in several ways:
B.1. COMPONENTS OF AN OVECS 335

1. They can simply declare snapback to occur for their entire pact. This
is easy if they are in their own pact, but is rarely done because resetting
the succession index to zero makes it look like the country has recently
underwent a revolution.

2. They can limit succession to be only allowed once a sufficiently long

period, such as annually or biannually. Merely restricting succession
to a maximum of twice per year brings the number of digits required
for the foreseeable future to be a reasonable if slightly uncomfortable
five digits, and the lower the succession rates, the slower digit creep
occurs.

3. They can use a special relative plate, as pictured in figure B.4, and
continually change the number that the relative plate adds. This is all
but resetting the succession index, but with with a special disclaimer
about it being “just a trick”, as it were.

B.1.4 The Pole

The pole is a nonessential part of the flag, and in fact it is so nonessential

that for the most part the public does not know about it. Nevertheless, it
is still a topic worthy of a brief discussion.
For the most part, flags do not define pole colours, and they are usually
left as the default colour of whatever material the pole is made of (e.g. brown
wood or silver metal.) When they do, only three solid colours can be part
of a pole, and if there’s more than one they must be striped. The directions
of the stripes are usually horizontal or vertical, but can also be in any other
direction if it is so warranted.
As mentioned earlier, the pole is not an essential part of the flag. Like
the plate, it can be removed from a depiction of the flag without changing
what it represents, but unlike the plate it does not have any special meaning
that can be easily generalised to all flags. This makes them ignored even
in places where plates would generally be included, explaining their general
obscurity among the public.
All boards must be centred on the pole. Furthermore, they must have
the same separation, usually 144
9
of the height of the banners. The top of the
pole must stay between the bottom and the top of the banner. but where
exactly cannot be defined by the flag and can be wherever is convenient to
the flag maker. They can be fastened in any inconspicuous way as well,
although the majority are done by gluing a handle to the back that slots
into the hole.
336 APPENDIX B. OCTOVEXILLOLOGY

5m

Figure B.5: A schematic of a common but loose field with enough room for
four flags. Each flag pole has a field of about 5 m across and there is little
overlap of fields of different flags. The flag poles are in black, the field in
dark green, and the field’s ring in light green. Most other types of fields,
if not bespoke, are much tighter than this, but this design emphasises the
circular nature of the field.

B.2 Mounting a flag

B.2.1 Simple mounting
With a flag being created, all that is left is to put it up somewhere. The
practise of putting up a flag is called mounting. There are several con-
ventions for mounting flags, and those are, with minor variations, uniform
across the world, by chance or by force.

Fields and their Rings

Flags are mounted in a specific area, known as the field (senparť).
̲ The field
may be, but sometimes is not, clearly delimited, and may be defined in one
or both of these two methods:-

1. The field is the area around the centre of the pole that is less than
a certain distance away from it. This distance is approximately two
metres, but different countries can have different values ranging from
one to ten metres.
2. The field is the area of the smallest rectangle that can cover the area
defined by the first definition.

Additionally, there is an additional area known as the field’s ring (nař

ek senparť)
̲ that can be any shape or size as long as it covers all of the field.
B.2. MOUNTING A FLAG 337

The field’s ring is invented because the strict definition of the field makes it
hard for architects to plan ahead for them.
There are norms with regards what should and should not be done in
the field. For instance, one should not leave rubbish or other loose objects
in it, and one should not run or unnecessarily disturb things inside it unless
it is to remove rubbish or similar. Additionally, no object may be taller than
the flag in the field, and there should be an open line of sight from the flag
in some direction. though this is a loose, optional requirement and in any
case the definition of an open line of sight is a very regional matter.
On the other hand, the field’s ring is much more lax and there are not
many rules that are common to even a majority of all flag protocols. For
the most part, they come in two flavours: the first is that they are just a
random patch of grass, reserved for the expansion of the field but as of of
yet still open for general use, or they can be extensions of the field and its
rules to an arbitrary shape and size. The latter is often seen in places where
the field is nearly or completely enclosed, like a “hole” in a building, and the
former is seen almost everywhere else, though as always, exceptions exist.

Poles and Distance

The centre of a pole is the nominal location of the flag, that is, if someone
asks for a millimetre-precise location of a flag, one points to the centre of its
(primary) pole. It is therefore important to keep in track where poles are.
In general, the heights of all poles should be the same in the same field,
although exception may be granted to exactly one pole, which can be shorter
than all the others. That pole, which if present must be at some extreme
position of the field, is the position of honour in which a particularly impor-
tant flag may be placed, but which flag this might be is context-dependent
and may not necessarily be the national flag. In any case, the poles must be
at least six times taller than the size of all the banners, which are regulated
for each flagpole such that each one has the same height. A popular industry
standard is to have them be 533 cm tall – 1 ferā – and the special pole be
12 ferā tall.
11

The minimum separation of each pole must be at least half the length
of the poles, but cannot exceed three-quarters that length. For a common
layout, where all the poles are lined up straight, popular standards include
8 , 12 or 3 of the length of the pole. However, it is not necessary for flagpoles
5 7 2

to be lined up like this. They can be in any configuration as desired, as long

as the overall shape has some subjective level of symmetry or repetition
or follow the contours of some natural feature. For those layouts, which for
large fields can be very peculiar indeed, the separation rule is relaxed slightly,
removing the minimum and trebling the maximum. Flags still cannot touch
each other physically, however.
338 APPENDIX B. OCTOVEXILLOLOGY

Finally, a completely unofficial but very common practise is to put a

piece of hard ground in the area directly behind a flagpole. The reason is
a practical one: a hard ground makes it easy for putting up and setting
down a flag by allowing a steady platform for one to put a ladder or similar
implement on. This is sometimes conspicuous and sometimes disguised, and
there is a public perception that it is an important part of a flagpole, even
though it isn’t.

Raising or lowering a flag

Putting up and setting down a flag is a straightforward manner, and is

usually not a very ceremonious one. There are two types, and it depends on
whether or not the pole is a permanent installation or not.
The first type, mobile flags, can be short (three metres) and usually
have pole markings. They are mounted simply by ramming the pole into
the ground. These poles have a sharpened bottom so as to ease ramming,
and fields that have holes for mobile flags are usually tapered at the bot-
tom to accommodate this spike. Their shortness also indicates that they
have prestige over the permanent flags, which is useful as guests are usually
treated as such. Removing them is also simple; simply take the pole out of
its original place.
Permanent flags and flagpoles are slightly different. Permanent poles
are rarely coloured, and those that are are usually reserved for a single flag.
They have holes drilled onto the back for the mounting brackets that are
glued on every sign’s back. More recently sophisticated backs have been
created with slots that can slide open and shut so that almost all heights
can be accommodated.
With permanent poles one mounts a flag by bringing it over the top of the
pole, then sliding it down to the desired height. Then, using either ordinary
nails or special screws with twist knobs, the flag is screwed on, from bottom
to top. This is why there are hard surfaces at the back of the pole; one would
mount it by going that high up and fixing the flag in place. Dismounting
the flag is trivial, then; simply mount the flag, but time-reversed.

B.2.2 Manymounting
Manymounting is the practise of mounting more than one flag on the same
pole. It is different from merely putting one flag next to another in a field,
and there are two, not just one, way to mount more than one flag on one pole.
The most obvious way that this can be done is to simply put one flag on top
of another. This is called serial manymounting (E.-Pasaru: želuuedulb ̲ ̲
lsa)
One can repeat this for up to 6 flags at a time. The hard cap is to ensure
that the highest flags are not so high that its details cannot be seen without
B.2. MOUNTING A FLAG 339

Figure B.6: An example of two flags mounted serially. In this example, two
documentation flags are used. Even so, the tallness of this flag combination
justifies the hard limitation of six flags per pole mounted serially.

using some kind of periscope or ladder. Still, the hard cap is rarely reached;
the maximum that most will ever see is three, and two is far more common.
If for whatever reason more than six is needed, the typical solution is to
mount the next set in another pole, slightly above and to the viewer’s right
of the mounting point of the first pole with a beam angled to that effect. The
tilted beam is to ensure that the situation is not to be confused with parallel
manymounting. However, with each new extension beam one fewer flag can
be mounted, leaving a new maximum of 21 flags; and of course all of those
extra flags have to be supported somehow, usually using counterweights on
the left-hand side or lots of trusses. Still, the new 21-flag limit is considered
high enough for virtually all circumstances.
This kind of mounting is fairly simple and is used to indicate hierarchal
relationships, such as city/state/country or individual/boss. It should be
noted that the flags closer to the bottom are the ones that are considered
more important, not the ones at the top; the reasoning here is that reading
things is done starting at normal eye level and moving away from that,
which in the case of flags, one starts at the one closest to eye level, which
is the bottom flag. This also explains why the hard cap of six is not very
restrictive; often one does not sign every hierarchical level on a flag, leaving
only two or three important ones above one’s own.
Since flags are allowed to specify their own pole colours, what happens
when two flags specify different colours are put together? There are several
solutions: the most obvious one is to select one flag that dominates over
the other, in which case we choose the bottom flag, as it is in the position
of honour. But one can also elect to use the neutral colour, favouring no
one, or paint different sections of the pole according to the flag just above
it. Different societies have different solutions, and all three mentioned here
are commonly understood.
340 APPENDIX B. OCTOVEXILLOLOGY

Figure B.7: An example of parallel manymounting. Note that as these

are documentation flags, these two configurations do not conflict with each
other. Also, the manymounting represents a union with code 04.
B.3. MAKING A FLAG 341

The other way to do it is to mount many flags, one next to each other,
but on the same (modified) pole. This is called parallel manymounting (E.-
Pasaru: želuuedulp̲ llŭinbřu)
̲ and is much more complex than serial many-
mounting. In parallel mounting, the pole splits into two or more at some
specified point, and from there on each one can can sport any flag or flag
combination.
Parallel manymounting has several features that distinguishes it from its
serial cousin. For instance, unlike serial manymounting there is no upper
bound to the number of flags that can be parallel-manymounted in the same
pole; one can choose to place as many flags in parallel on the same pole as
one wishes, though of course limitations can occur by other means, such as
available space.
These flags can be very large – extreme cases can have more than thirty
or forty flags mounted on the same pole, so extra support is needed for
these flag assemblies. All flags are fed through a truss (E.-Pasaru: aķdžœ)
at the lower end of their individual poles, which then joins to the master
pole (E.-Pasaru: lelëdæn), which is the pole that finally attaches the whole
lot to the ground. In order to allow further support, diagonal rods known as
̲
struts (E.-Pasaru: unzledæn) and vertical rods known as auxiliary poles can
be used to reïnforce the structure. Struts do not touch the ground, whereas
auxiliary poles do.
Secondly, flags mounted in parallel have equal standing with each other.
This can be overridden by having one flag be shorter than all the rest, but
this is rarely done and in the end the cultural norm is that when flags are
mounted in parallel, no object referred to in that assembly is more important
than another. If however one is required, one can mount a flag in series,
but make the upper “flag” a parallel manymounting.
Finally, parallel mounting results in a consequence – there is a space for
a legend to be placed at the base of the manymounting. This legend, which
even if present is still just called a legend, is used to label the manymounting
in its entirety, and usually draws its symbols from the name of the union of
the objects that take part in the manymounting. This means that in some
cases, unions don’t have flags of their own, just the parallel manymounting
of its members.

B.3 Making a flag

Unlike Earthling (Western) flags, which are typically made out of coloured
fabric, ovexes are made out of hard, easily painted material such as wood or
stone. Naturally, this means that paint is used to draw the picture in.
Those who make flags are called flagmen (yessen),̲ unsurprisingly. As a
group they make sure everything from forest to finish goes smoothly. They
342 APPENDIX B. OCTOVEXILLOLOGY

source the building material, cut it down to size, and draw the lines on the
banner. We will now describe how the flag production process works.

B.3.1 Creation process

It begins with the order to make a flag. All flags have a specification, a
code in the cypher sense that specifies exactly where the lines are drawn
and what the colours are, without actually drawing a diagram, which is at
the time very space-consuming. It is through this medium that a flagman
creates the flag. The code itself is made out of a list of shorthands and
names, names for all of the points and most of the lines. More recently, a
more public-facing version was created that replaces the code with a bunch
of drawings, but this is only used in very formal contexts and not as the
actual plan to make flags.
The flagman then scores the lines using a special set of rulers, both for
lines and special curves, a couple of dividers, compasses, neusis rulers, and
other geometrical implements. When all the lines are scored, (which include
the lines that are thin bands in the final flag as well as lines that only act
as helpers) the areas scored out are then painted, though not necessarily in
the order the specification asks for. Next, if the flag calls for it, devices are
attached to the specified anchor points, completing the banner.
The banner and any plates are partially prefabricated; the plates are
completely so, but the banner’s prefab is just a white rectangle with a thin
black border. The flagman has to write the banner’s characters in himself,
but this can also be done using a prepared stencil.
Finally all the display parts have a bracket glued onto their backs, which
are the part that allows them to be attached to any flagpoles. Interestingly,
though flagmen make some poles, they are not responsible for all of them;
architects and other home planners take care of all the static flagpoles, and
the ones that the flagmen are responsible for are mobile.
With all the parts done, all that’s left is to assemble the flag. This is
not done by the flagmen; all they do is to put all the flags into a box with
a couple of nails and screws which they then ship off to the recipient. The
screws are not conventional screws; as mentioned earlier they have a knob
so they can be screwed or unscrewed without a screwdriver. The exact form
of the screw used to be a trade secret, but nowadays is so strongly tied with
flags no one ever uses them for anything other than flags. The whole box
is then sealed, decorated and shipped off to the customer, which can be as
humble as a single individual or it can be an entire country(’s government).

B.3.2 Materials
At first the flags are made out of wood and stone, and these are still how
traditional flags are made today. However, in more recent times flagmen
B.4. REPRESENTATIVE POWER OF A FLAG 343

have experimented with more modern materials, such as aluminium and

steel with retroreflective material.
With the change in material, change in production technique is required
as well. By the year 19400 PDN, flags, especially when mass-produced,
are computer-designed and also built by robots using plotters and other
such flag-making devices. There’s a minor demand for hand-made flags, so
flagmen are still in employment, but their numbers have never grown since
their peak at 12500 PDN.

B.4 Representative power of a flag

Flags are typically associated with countries, but that’s not the only thing
they can do. They can, and often do, represent private entities such as
companies, individuals, informal groups, events, ideas, groups of ideas, and
even otherwise random assortments of unrelated objects. This is slightly dif-
ferent from the flags that Earthlings are used to because they are generally
used only to represent a political subdivision, though of course they can and
do get used to represent other things.
One usage case of the pole colour is to indicate the class of object that
the flag represents. This is not standardised and is usually just an ad-hoc
convention, but such conventions have been used and exploited over the
years in different countries and different situations. For instance, in early
9th century Vohalyo, yellow poles generally indicate guilds and unions, green
poles generally indicate the rural contingent, and red poles are for couriers
and healthcare agents. “Official” (governmental) flags remain neutral in pole
colour.
One surprising class of flags are flags that don’t represent anything at all,
and are universally regulated to not represent anything at all. These are the
documentation flags, which are a special class of extremely simple flags that
are reserved for describing how they work in documents like this one. These
flags are extremely simple, usually just a plain colour with a corresponding
figure drawn inside, and the legend are two duodecimal digits. As a result,
144 flags can be used at a time, but there are actually about 800,000 of
them. As a consequence, no legend in the wild can solely consist of digits.

B.5 Glossary of ovecs terms

Generally speaking, the technical terms of an ovecs are directly borrowed
from E.-Pasaru to Drsk, so the translations are actually pretty straightfor-
ward.
344 APPENDIX B. OCTOVEXILLOLOGY
 Part XI

Appendices

345
Appendix C

Lexicon

This is the lexicon for all Drsk words that are in this book and a few more
besides.
The lexicon is formatted according to the traditional layout, which we
describe in section 27.8. However, as mentioned the actual entries are not
fully specified by that convention, so we will specify how this is done for
the purposes of this chapter. There is no language-wide standard for the
formatting of an entry in a dictionary, only how to arrange them.

C.1 Etymological information

This lexicon provides a brief summary of etymological information. That
information does not contain much historical information; unless otherwise
indicated, only words in the modern day are displayed in each entry, with
the historical forms implied.
For the purposes of this lexicon, the modern day has these era letters
active. Section 1.7 and 6.1 describe era letters in more detail.

Language Era letter

Drsk M57
Rattssaw V
Nnn Heeel Ee
E.-Pasaru 1
Proto-Remmsp S58
xRron –59
Egako O

57
Drsk itself is also the implied language if not indicated, as it is the subject language.
58
This represents the state of the language as it was in the year −11 900 PDN.
59
xRron does not use era letters; its name changes with every change of era.

347
348 APPENDIX C. LEXICON

This roughly corresponds to the year 10 000 PDN. Languages not in this

list will have their era letters indicated explicitly; that letter can be assumed
to be in the modern day unless otherwise noted.

C.2 Entry samples

Sample of a noun entry
The following is a sample of a noun, with some annotations:

.zz(1) .(2) .lk(3) < R zlk /sslk/ “Name”(4) Unit quote.(5) Marks
the name of the unit.(6)

1. CAT MAJ. The CAT SMAJ is omitted for nouns.

2. CAT MIN

3. Root; note that the suffix is missing.

4. Etymology information; is restricted to the immediate origin, which

could either be R, for Rattssaw; N, for Nnn Heeel; I, for words inherited
from proto-Remmsp; or the language name for any other language.
This section may sometimes be missing for complicated, unknown or
uncertain origins. Modern cognates are given rather than forms in
earlier eras, unless specified otherwise.

5. Short gloss. This is the basis for the translation in the interlinear gloss.

6. Longer description, if needed.

Sample of a verb entry

The following is a sample of a verb, with some annotations:

bvᵉlł < I bvell /bvɛɮ/ “Halt”(1) Interrupt

1. The etymology section, as described earlier in part 4 in section C.2.

Sample of an entry for other types of word

The following are samples for words that is neither a verb or a noun:

kdᵘ s.  (1) ‹ I gedy /gɛdy/ “Yes” Yes

ls.y(2) < R oelss /oeltˢ+ / uncertain Untimely. In a way that
does seems out of time order.
C.3. KEY TO ABBREVIATIONS 349

1. Where the role of a word is not indicated explicitly, its part of speech
is indicated here.

2. If they are indicated explicitly, then they are separated out using a
dot, and the part of speech is not indicated. Here, the suffix .y means
a quasinoun, for which see section 21.10.

C.3 Key to abbreviations

Table C.1: Table of language abbreviations

Abbreviation Language
 R Rattssaw
 RU Rattssaw U
 N Nnn Heeel
 I Proto-Remmsp
 X xRron
 G Egako
 E E.-Pasaru

Table C.2: Table of word type abbreviations

Abbreviation Word type

Sundry s.
Dredge, owned d.
Dredge, ownerless D.
Dredge, relational r.

Inflectional morphemes are not shown in this lexicon.

350 APPENDIX C. LEXICON
Appendix D

Reference

The primary reference of this book, as previously mentioned, is the Book

of Conworlds. This chapter gives all pages that are relevant to Drsk, and
associates them with a section number. However, some details do not relate
directly to Drsk but are cited here anyway because they are relevant in other
ways.
Unless otherwise indicated, the index heading is the page number, and
this is followed by either a section number that the page is relevant for, or
a single dash if it is relevant to Drsk but otherwise did not receive a section
number.
For sections that contain multiple books, the index heading contains the
book number, a hyphen, then the page number instead of just being the
page number.

Book 23

01a 2, 3 13a 16.4.1, 16.4.2 22a 21.1

02a 9 14a 21.10 23a 16.7.1

03a 12 15a 16.5 24a 21.1

04a 10, 10.2 16a 16.6 25a 24.2

05a 13.2 17a 18.9 26a 19.5

06a 12.5, 11.1 18a 24 26b 19.5

10a 16.2 19a 25.1 27a 19.5, 19.3

11a 16.4.2 20a 20 28a 19.5

12a 17.1 21a 24.1 29a 16.9

351
352 APPENDIX D. REFERENCE

30a 16.9 41a 13.6.2 53a 21.4

31a 21.5 42a 16.4.6 54a 1.7

31b 21.5 43a 27.8 55a 13.6.3, 21.2, 16.9

32a 16.13 44a 3.3 56a 21.3

33a 17.5 45a 12.5 57a 24.4

34a 21.7 46a 10.3 58a –

35a 21.7 47a 29.4, 29.5 59a –

36a – 48a 31 61a 12.9

37a – 49a 31 62b 12.9

38a 19.5 50a 31 62a 3

39a 19.6 51a 31 63a –

40a 19.5 52a 16.4.4 64a 13.5

Book 24

01a 18.3, 18.4 10a – 28a –

02a 18.5, 18.6, 18.7 12a 16.9 29a –

03a 18.2 13a 17.8 30a –

04a 19.5 14a 25.6 31a –

05a 16.9 15a 3.3 58a 18.9

06a 15.1 16a 25.7 67a 24.7

07a 16.9 17a 22.1 69a 32

08a – 18a 19.5 70a 3.3

09a – 18b 19.5 77a 3

Books 25 to 27
BOOK 28 353

25-08a 28.1 26-52a 16.10, 16.14 27-31a ?

26-16a 25.5 27-12a 16.10, 16.14 27-32a 14.1

26-33a 18.8 27-17a 16.10 27-33a 14.2

26-34a 21.8 27-28a 13.4 27-39a –

26-35a 18.8 27-29a 13.3, 13.6.2 27-40a 14.2

26-51a 16.10 27-30a 14.3 27-42a 10.5

Book 28

14a 16.7.1 49a 23.5.1 66a 23.2

21a 17.8 51a 23.5.1 67a 23.2

22a 24.7 52a 23.8
68a 23.3
40a 26.2 54a –
69a 23.2
40b 26.2 55a –
73a 23.2, 26.1
42a 3.3.2 59a 23.2
78a 23.5.2
43a 23.5.1 60a 13.7

44a 23.5.1 63a 23.2 79a 23.5.2

48a 23.5.1 65a 23.2 80a 23.5.2

Book 29

01a 23.5.2 10a 16.3, 16.6 17a –

02a 23.5.2 12a – 19a –

06a 23.5.2 13a – 23a 23.5.2

07a 16.5 14a 23.5.2 26a –

09a 16.3 15a – 55a 26.5, 27.2

Books 30 and 31
There are no pages about Drsk on these books.
354 APPENDIX D. REFERENCE

Books 32 to 34

32-15a 16.3 32-70a – 34-17a ?

32-16a 27.7 32-71a – 34-18a 12.7

32-17a 25 32-77a ? 34-24a 17.2

32-18a 27.3 33-07a – 34-25a 12.10

32-49a ? 33-51a – 34-29a 29.9

Appendix E

The language skeleton

This is a sketch of the language skeleton of Drsk. We’ll structure this as

a list, flattening everything from figure 1.1 into a list: one where the top
level is each of the items in the list, and the second level is a one-sentence
summary of a feature, with a corresponding section link. Top level items
that are emphasised represent object-level items in the skeleton (i.e. they
describe actual objects in the language).
Features that have blanks to be filled in have the blanks indicated using
underline.

• Phonemes (phonology)

• Phonotactics

• Morphemes

• Morphology

• Lexicon

• Semantics

• Grammar

• Syntax

• Script (orthography)

• Sentences

• Pragmatics

• Passages

355
356 APPENDIX E. THE LANGUAGE SKELETON
Appendix F

Useful lists

This chapter provides some useful lists that have been referred to in the main
text. They generally refer to cultural concepts that are unique to Pasaru,
and mostly do not correspond to any particular word in English or any other
language on Earth.
Apart from listing each of entry in Drsk, we will also list them in E.-
Pasaru, Rattssaw and Nnn Heeel to allow readers to compare the languages
and how these words adapt to other languages.

Month names
This section lists the month names in the modern calendar in the various
languages mentioned in this grammar. For more on the calendar, please
see section 2.5.4. Note that these months are not the same as the months
discussed in section 8.2.2, as that uses a different calendar system (specif-
ically, a hybrid between 2.1.4 and 2.2.4).
Each month is listed twice: once in the language’s orthography, and the
other time in IPA. (1 | 20)

# E.-Pasaru Drsk Rattssaw Nnn Heeel

1 Yoni nv
/joni/ /nv/
2 Mayonaķa
/majonaqa/
3 Ďoste
/ðoste/
4 Kalo
/kalo/
5 Irda
Continued on next page

357
358 APPENDIX F. USEFUL LISTS

Continued from previous page

# E.-Pasaru Drsk Rattssaw Nnn Heeel
/iɹda/
6 Pol
/pol/
7 Leba
/leba/
8 Defaran
/defaɹan/
9 Rodigey
/ɹodigej/
10 Strabot
/stɹabot/
11 Aďlagu
/aðlagu/
12 Üstab
/ystab/
13 Sedaorui
/sedɐrʉ/
14 Etesžabu
/etesʒabu/
15 Haģdozese
/haɢdozese/
16 Yistẽsolťu
/jistẽsolθu/
17 Ȝelyȝaħu
/ɣeljɣaxu/
18 Atdeg
/atdeg/
19 Jelpon
/ʔelpon/
20 Asiyoniks
/asijoniks/
21 Ostabolga
/ostabolga/
Appendix G

Colour list

Below is a colour test. It is a listing of all colours that are used in this book.
Any colour used in this text other than the default text colour should always
be included in this list.
First, we list the colours, each one with saying the same thing:

Sample text Sample text Sample text Sample text Sample text
Sample text Sample text Sample text Sample text Sample text
Sample text Sample text Sample text Sample text Sample text
Sample text Sample text Sample text Sample text Sample text
Sample text Sample text Sample text Sample text

Then we list the colours again, this time giving their names. We have
named them in the same order. Bullets have additionally been placed to
allow easier segmentation.
Bronze • Brown • Red • Orange • Mandarin • Gold • Yellow • Lime
• Green • Teal • Sky • Blue • Navy • Purple • Violet • Lavender
• Amaranth • Magenta • Pink • Salmon • Black • Grey • Silver •
Notebook default text

359
360 APPENDIX G. COLOUR LIST
Part XII

Postface

361
History

363
364 HISTORY
Acknowledgements

General

Attributions
Figure 3.7 uses drawings from the Noun Project, which is available under
the Creative Commons Attribution Licence 4.0.

365
366 ACKNOWLEDGEMENTS
Afterword

367
368 AFTERWORD
GNU Free Documentation
License

Version 1.3, 3 November 2008

Copyright © 2000, 2001, 2002, 2007, 2008 Free Software Foundation, Inc.

<[Link]

Everyone is permitted to copy and distribute verbatim copies of this

license document, but changing it is not allowed.

Preamble

The purpose of this License is to make a manual, textbook, or other func-

tional and useful document “free” in the sense of freedom: to assure everyone
the effective freedom to copy and redistribute it, with or without modifying
it, either commercially or noncommercially. Secondarily, this License pre-
serves for the author and publisher a way to get credit for their work, while
not being considered responsible for modifications made by others.
This License is a kind of “copyleft”, which means that derivative works
of the document must themselves be free in the same sense. It complements
the GNU General Public License, which is a copyleft license designed for
free software.
We have designed this License in order to use it for manuals for free
software, because free software needs free documentation: a free program
should come with manuals providing the same freedoms that the software
does. But this License is not limited to software manuals; it can be used
for any textual work, regardless of subject matter or whether it is published
as a printed book. We recommend this License principally for works whose
purpose is instruction or reference.

1. APPLICABILITY AND DEFINITIONS

This License applies to any manual or other work, in any medium, that
contains a notice placed by the copyright holder saying it can be distributed

369
370 GNU FREE DOCUMENTATION LICENSE

under the terms of this License. Such a notice grants a world-wide, royalty-
free license, unlimited in duration, to use that work under the conditions
stated herein. The “Document”, below, refers to any such manual or work.
Any member of the public is a licensee, and is addressed as “you”. You accept
the license if you copy, modify or distribute the work in a way requiring
permission under copyright law.
A “Modified Version” of the Document means any work containing the
Document or a portion of it, either copied verbatim, or with modifications
and/or translated into another language.
A “Secondary Section” is a named appendix or a front-matter section
of the Document that deals exclusively with the relationship of the publishers
or authors of the Document to the Document’s overall subject (or to related
matters) and contains nothing that could fall directly within that overall
subject. (Thus, if the Document is in part a textbook of mathematics, a
Secondary Section may not explain any mathematics.) The relationship
could be a matter of historical connection with the subject or with related
matters, or of legal, commercial, philosophical, ethical or political position
regarding them.
The “Invariant Sections” are certain Secondary Sections whose titles
are designated, as being those of Invariant Sections, in the notice that says
that the Document is released under this License. If a section does not fit
the above definition of Secondary then it is not allowed to be designated
as Invariant. The Document may contain zero Invariant Sections. If the
Document does not identify any Invariant Sections then there are none.
The “Cover Texts” are certain short passages of text that are listed,
as Front-Cover Texts or Back-Cover Texts, in the notice that says that the
Document is released under this License. A Front-Cover Text may be at
most 5 words, and a Back-Cover Text may be at most 25 words.
A “Transparent” copy of the Document means a machine-readable copy,
represented in a format whose specification is available to the general public,
that is suitable for revising the document straightforwardly with generic
text editors or (for images composed of pixels) generic paint programs or
(for drawings) some widely available drawing editor, and that is suitable
for input to text formatters or for automatic translation to a variety of
formats suitable for input to text formatters. A copy made in an otherwise
Transparent file format whose markup, or absence of markup, has been
arranged to thwart or discourage subsequent modification by readers is not
Transparent. An image format is not Transparent if used for any substantial
amount of text. A copy that is not “Transparent” is called “Opaque”.
Examples of suitable formats for Transparent copies include plain ASCII
without markup, Texinfo input format, LaTeX input format, SGML or XML
using a publicly available DTD, and standard-conforming simple HTML,
PostScript or PDF designed for human modification. Examples of trans-
parent image formats include PNG, XCF and JPG. Opaque formats include
 371

proprietary formats that can be read and edited only by proprietary word
processors, SGML or XML for which the DTD and/or processing tools are
not generally available, and the machine-generated HTML, PostScript or
PDF produced by some word processors for output purposes only.
The “Title Page” means, for a printed book, the title page itself, plus
such following pages as are needed to hold, legibly, the material this License
requires to appear in the title page. For works in formats which do not have
any title page as such, “Title Page” means the text near the most prominent
appearance of the work’s title, preceding the beginning of the body of the
text.
The “publisher” means any person or entity that distributes copies of
the Document to the public.
A section “Entitled XYZ” means a named subunit of the Document
whose title either is precisely XYZ or contains XYZ in parentheses follow-
ing text that translates XYZ in another language. (Here XYZ stands for
a specific section name mentioned below, such as “Acknowledgements”,
“Dedications”, “Endorsements”, or “History”.) To “Preserve the Ti-
tle” of such a section when you modify the Document means that it remains
a section “Entitled XYZ” according to this definition.
The Document may include Warranty Disclaimers next to the notice
which states that this License applies to the Document. These Warranty
Disclaimers are considered to be included by reference in this License, but
only as regards disclaiming warranties: any other implication that these
Warranty Disclaimers may have is void and has no effect on the meaning of
this License.

2. VERBATIM COPYING
You may copy and distribute the Document in any medium, either com-
mercially or noncommercially, provided that this License, the copyright no-
tices, and the license notice saying this License applies to the Document are
reproduced in all copies, and that you add no other conditions whatsoever
to those of this License. You may not use technical measures to obstruct or
control the reading or further copying of the copies you make or distribute.
However, you may accept compensation in exchange for copies. If you dis-
tribute a large enough number of copies you must also follow the conditions
in section 3.
You may also lend copies, under the same conditions stated above, and
you may publicly display copies.

3. COPYING IN QUANTITY
If you publish printed copies (or copies in media that commonly have
printed covers) of the Document, numbering more than 100, and the Doc-
ument’s license notice requires Cover Texts, you must enclose the copies in
372 GNU FREE DOCUMENTATION LICENSE

covers that carry, clearly and legibly, all these Cover Texts: Front-Cover
Texts on the front cover, and Back-Cover Texts on the back cover. Both
covers must also clearly and legibly identify you as the publisher of these
copies. The front cover must present the full title with all words of the title
equally prominent and visible. You may add other material on the covers in
addition. Copying with changes limited to the covers, as long as they pre-
serve the title of the Document and satisfy these conditions, can be treated
as verbatim copying in other respects.
If the required texts for either cover are too voluminous to fit legibly,
you should put the first ones listed (as many as fit reasonably) on the actual
cover, and continue the rest onto adjacent pages.
If you publish or distribute Opaque copies of the Document numbering
more than 100, you must either include a machine-readable Transparent
copy along with each Opaque copy, or state in or with each Opaque copy
a computer-network location from which the general network-using public
has access to download using public-standard network protocols a complete
Transparent copy of the Document, free of added material. If you use the
latter option, you must take reasonably prudent steps, when you begin dis-
tribution of Opaque copies in quantity, to ensure that this Transparent copy
will remain thus accessible at the stated location until at least one year after
the last time you distribute an Opaque copy (directly or through your agents
or retailers) of that edition to the public.
It is requested, but not required, that you contact the authors of the
Document well before redistributing any large number of copies, to give
them a chance to provide you with an updated version of the Document.

4. MODIFICATIONS
You may copy and distribute a Modified Version of the Document under
the conditions of sections 2 and 3 above, provided that you release the
Modified Version under precisely this License, with the Modified Version
filling the role of the Document, thus licensing distribution and modification
of the Modified Version to whoever possesses a copy of it. In addition, you
must do these things in the Modified Version:

A. Use in the Title Page (and on the covers, if any) a title distinct from
that of the Document, and from those of previous versions (which
should, if there were any, be listed in the History section of the Docu-
ment). You may use the same title as a previous version if the original
publisher of that version gives permission.

B. List on the Title Page, as authors, one or more persons or entities

responsible for authorship of the modifications in the Modified Version,
together with at least five of the principal authors of the Document (all
 373

of its principal authors, if it has fewer than five), unless they release
you from this requirement.

C. State on the Title page the name of the publisher of the Modified
Version, as the publisher.

D. Preserve all the copyright notices of the Document.

E. Add an appropriate copyright notice for your modifications adjacent

to the other copyright notices.

F. Include, immediately after the copyright notices, a license notice giving

the public permission to use the Modified Version under the terms of
this License, in the form shown in the Addendum below.

G. Preserve in that license notice the full lists of Invariant Sections and
required Cover Texts given in the Document’s license notice.

H. Include an unaltered copy of this License.

I. Preserve the section Entitled “History”, Preserve its Title, and add to
it an item stating at least the title, year, new authors, and publisher of
the Modified Version as given on the Title Page. If there is no section
Entitled “History” in the Document, create one stating the title, year,
authors, and publisher of the Document as given on its Title Page,
then add an item describing the Modified Version as stated in the
previous sentence.

J. Preserve the network location, if any, given in the Document for pub-
lic access to a Transparent copy of the Document, and likewise the
network locations given in the Document for previous versions it was
based on. These may be placed in the “History” section. You may omit
a network location for a work that was published at least four years
before the Document itself, or if the original publisher of the version
it refers to gives permission.

K. For any section Entitled “Acknowledgements” or “Dedications”, Pre-

serve the Title of the section, and preserve in the section all the sub-
stance and tone of each of the contributor acknowledgements and/or
dedications given therein.

L. Preserve all the Invariant Sections of the Document, unaltered in their

text and in their titles. Section numbers or the equivalent are not
considered part of the section titles.

M. Delete any section Entitled “Endorsements”. Such a section may not

be included in the Modified Version.
374 GNU FREE DOCUMENTATION LICENSE

N. Do not retitle any existing section to be Entitled “Endorsements” or

to conflict in title with any Invariant Section.

O. Preserve any Warranty Disclaimers.

If the Modified Version includes new front-matter sections or appendices

that qualify as Secondary Sections and contain no material copied from the
Document, you may at your option designate some or all of these sections
as invariant. To do this, add their titles to the list of Invariant Sections in
the Modified Version’s license notice. These titles must be distinct from any
other section titles.
You may add a section Entitled “Endorsements”, provided it contains
nothing but endorsements of your Modified Version by various parties—for
example, statements of peer review or that the text has been approved by
an organization as the authoritative definition of a standard.
You may add a passage of up to five words as a Front-Cover Text, and
a passage of up to 25 words as a Back-Cover Text, to the end of the list of
Cover Texts in the Modified Version. Only one passage of Front-Cover Text
and one of Back-Cover Text may be added by (or through arrangements
made by) any one entity. If the Document already includes a cover text for
the same cover, previously added by you or by arrangement made by the
same entity you are acting on behalf of, you may not add another; but you
may replace the old one, on explicit permission from the previous publisher
that added the old one.
The author(s) and publisher(s) of the Document do not by this License
give permission to use their names for publicity for or to assert or imply
endorsement of any Modified Version.

5. COMBINING DOCUMENTS
You may combine the Document with other documents released under
this License, under the terms defined in section 4 above for modified versions,
provided that you include in the combination all of the Invariant Sections
of all of the original documents, unmodified, and list them all as Invariant
Sections of your combined work in its license notice, and that you preserve
all their Warranty Disclaimers.
The combined work need only contain one copy of this License, and
multiple identical Invariant Sections may be replaced with a single copy.
If there are multiple Invariant Sections with the same name but different
contents, make the title of each such section unique by adding at the end
of it, in parentheses, the name of the original author or publisher of that
section if known, or else a unique number. Make the same adjustment to
the section titles in the list of Invariant Sections in the license notice of the
combined work.
 375

In the combination, you must combine any sections Entitled “History”

in the various original documents, forming one section Entitled “History”;
likewise combine any sections Entitled “Acknowledgements”, and any sec-
tions Entitled “Dedications”. You must delete all sections Entitled “En-
dorsements”.

6. COLLECTIONS OF DOCUMENTS
You may make a collection consisting of the Document and other docu-
ments released under this License, and replace the individual copies of this
License in the various documents with a single copy that is included in the
collection, provided that you follow the rules of this License for verbatim
copying of each of the documents in all other respects.
You may extract a single document from such a collection, and distribute
it individually under this License, provided you insert a copy of this License
into the extracted document, and follow this License in all other respects
regarding verbatim copying of that document.

7. AGGREGATION WITH INDEPENDENT

WORKS
A compilation of the Document or its derivatives with other separate
and independent documents or works, in or on a volume of a storage or
distribution medium, is called an “aggregate” if the copyright resulting from
the compilation is not used to limit the legal rights of the compilation’s
users beyond what the individual works permit. When the Document is
included in an aggregate, this License does not apply to the other works in
the aggregate which are not themselves derivative works of the Document.
If the Cover Text requirement of section 3 is applicable to these copies
of the Document, then if the Document is less than one half of the entire
aggregate, the Document’s Cover Texts may be placed on covers that bracket
the Document within the aggregate, or the electronic equivalent of covers if
the Document is in electronic form. Otherwise they must appear on printed
covers that bracket the whole aggregate.

8. TRANSLATION
Translation is considered a kind of modification, so you may distribute
translations of the Document under the terms of section 4. Replacing Invari-
ant Sections with translations requires special permission from their copy-
right holders, but you may include translations of some or all Invariant
Sections in addition to the original versions of these Invariant Sections. You
may include a translation of this License, and all the license notices in the
376 GNU FREE DOCUMENTATION LICENSE

Document, and any Warranty Disclaimers, provided that you also include
the original English version of this License and the original versions of those
notices and disclaimers. In case of a disagreement between the translation
and the original version of this License or a notice or disclaimer, the original
version will prevail.
If a section in the Document is Entitled “Acknowledgements”, “Dedica-
tions”, or “History”, the requirement (section 4) to Preserve its Title (sec-
tion 1) will typically require changing the actual title.

9. TERMINATION
You may not copy, modify, sublicense, or distribute the Document except
as expressly provided under this License. Any attempt otherwise to copy,
modify, sublicense, or distribute it is void, and will automatically terminate
your rights under this License.
However, if you cease all violation of this License, then your license from
a particular copyright holder is reinstated (a) provisionally, unless and until
the copyright holder explicitly and finally terminates your license, and (b)
permanently, if the copyright holder fails to notify you of the violation by
some reasonable means prior to 60 days after the cessation.
Moreover, your license from a particular copyright holder is reinstated
permanently if the copyright holder notifies you of the violation by some
reasonable means, this is the first time you have received notice of violation
of this License (for any work) from that copyright holder, and you cure the
violation prior to 30 days after your receipt of the notice.
Termination of your rights under this section does not terminate the
licenses of parties who have received copies or rights from you under this
License. If your rights have been terminated and not permanently reinstated,
receipt of a copy of some or all of the same material does not give you any
rights to use it.

10. FUTURE REVISIONS OF THIS LICENSE

The Free Software Foundation may publish new, revised versions of the
GNU Free Documentation License from time to time. Such new versions
will be similar in spirit to the present version, but may differ in detail to
address new problems or concerns. See [Link]
Each version of the License is given a distinguishing version number. If
the Document specifies that a particular numbered version of this License
“or any later version” applies to it, you have the option of following the terms
and conditions either of that specified version or of any later version that
has been published (not as a draft) by the Free Software Foundation. If the
Document does not specify a version number of this License, you may choose
any version ever published (not as a draft) by the Free Software Foundation.
 377

If the Document specifies that a proxy can decide which future versions of
this License can be used, that proxy’s public statement of acceptance of a
version permanently authorizes you to choose that version for the Document.

11. RELICENSING
“Massive Multiauthor Collaboration Site” (or “MMC Site”) means any
World Wide Web server that publishes copyrightable works and also pro-
vides prominent facilities for anybody to edit those works. A public wiki
that anybody can edit is an example of such a server. A “Massive Multi-
author Collaboration” (or “MMC”) contained in the site means any set of
copyrightable works thus published on the MMC site.
“CC-BY-SA” means the Creative Commons Attribution-Share Alike 3.0
license published by Creative Commons Corporation, a not-for-profit corpo-
ration with a principal place of business in San Francisco, California, as well
as future copyleft versions of that license published by that same organiza-
tion.
“Incorporate” means to publish or republish a Document, in whole or in
part, as part of another Document.
An MMC is “eligible for relicensing” if it is licensed under this License,
and if all works that were first published under this License somewhere
other than this MMC, and subsequently incorporated in whole or in part
into the MMC, (1) had no cover texts or invariant sections, and (2) were
thus incorporated prior to November 1, 2008.
The operator of an MMC Site may republish an MMC contained in the
site under CC-BY-SA on the same site at any time before August 1, 2009,
provided the MMC is eligible for relicensing.

ADDENDUM: How to use this License for your

documents
To use this License in a document you have written, include a copy of the
License in the document and put the following copyright and license notices
just after the title page:

Copyright © YEAR YOUR NAME. Permission is granted to

copy, distribute and/or modify this document under the terms
of the GNU Free Documentation License, Version 1.3 or any
later version published by the Free Software Foundation; with
no Invariant Sections, no Front-Cover Texts, and no Back-Cover
Texts. A copy of the license is included in the section entitled
“GNU Free Documentation License”.
378 GNU FREE DOCUMENTATION LICENSE

If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts,

replace the “with … Texts.” line with this:

with the Invariant Sections being LIST THEIR TITLES, with

the Front-Cover Texts being LIST, and with the Back-Cover
Texts being LIST.

If you have Invariant Sections without Cover Texts, or some other com-
bination of the three, merge those two alternatives to suit the situation.
If your document contains nontrivial examples of program code, we rec-
ommend releasing these examples in parallel under your choice of free soft-
ware license, such as the GNU General Public License, to permit their use
in free software.