Principles of English spelling formation

Studies of English spelling have primarily focussed on correspondences between spelling and sound among core, standard spellings. Segmental-level correspondences have been examined in detail (Venezky 1970, Cummings 1988, Carney 1994, Rollings 2004), while recent work shows that English spellings also encode supra-segmental information (Evertz and Primus 2013, Evertz 2014). An outstanding problem is the degree to which morphemic spelling is applied across the system (c.f. Venezky 1970: 120; Carney 1994: 18). Berg et al. (2014) observe that the spelling of affixes is more stable than the spelling of bases, hence <profane> alternates with <profanity>, not *<profanety> or *<profaneity>. Yet none of these theories address in detail why certain spellings are chosen over others. This thesis examines how English spellings are formed, particularly where compromises are made in the representation of both morphological and phonological information. The primary focus is on the spelling of inflections, but also derivations, names and non-standard spellings. The study also examines how we know if a spelling is a good phonographic match for a word. Five principles of spelling formation are proposed. 1. Any-Spelling principle: All words must have some spelling. 2. Distinctiveness principle (DSTNCT): Different words should have different spellings. 3. Identity Preservation principle (IdP): Related words should have related spellings. Hence complex words take their spelling from the spellings of their subcomponents. 4. Phonographic Matching principle (PhM): A word's spelling should represent its phonological form. 5. Invariance principle (INVRNC): Established spellings cannot be changed. The principles are manifested differently according to the category of word being spelt. Compound words adhere strongly to IdP. Regular inflections follow IdP by default, <jump, jumped, jumping>, but may involve PhM amendments, hence <dope, doped, doping> not <dope, *dopeed, *dopeing>. Affixed derivatives show similar patterns, <mode, modal>, regardless of stress-shifting, <origin, original>, while etymological influences complicate the picture, <possible, probable>. DSTNCT sometimes differentiates lexical homophones. <flour>, <flower>, but it is common among certain names, <Webb>, <Blu Tac>, <OutKast>. Non-standard spellings violate INVRNC by definition, and the conditions for spelling variation arise in part due to phonological changes. Abbreviations frequently obey IdP, e.g. <a.k.a.>, but clippings may not, <telly>, <fridge>. Etymological spelling is deemed to be an example of IdP, either among unchanged base forms, <macchiato>, or adapted polymorphemic words, <philosophy>. This re-assessment helps to solve the long-standing problem of how to integrate etymological spelling into synchronic theories of spelling. A flexible unit of spelling, the complex pleremic unit, is identified and it accounts for IdP's various manifestations. Native, monomorphemic base forms are not examined in detail as the priority is how new and complex words are spelt. This method is taken from word-formation studies (Marchand 1969, Bauer 1983, Plag 2003). The initial model of spelling formation assumes that polymorphemic words obey IdP by default, but that PhM amendments can be made, where necessary, if possible. The second iteration builds on Evertz's (2014) graphematic hierarchy to show how unsatisfactory spellings can be identified by comparing spelling and phonology at all hierarchical levels, including phonemes, syllables, and feet. The third iteration shows how Optimality Theory can be used to explain how one imperfect spelling, e.g. <doped>, can be chosen over an even less perfect spelling, *<dopeed>. Two further applications arise from the model: the pronunciation of many polysyllabic words can be predicted from the spelling with greater accuracy; and a path is provided by which spelling pronunciation can be predicted from potential ambiguities in decoding.

This study provides a method for studying a wide range of English spellings. I argue that there are five basic principles of English spelling and that different kinds of words may be spelt in different ways, subject to different applications of these principles, which state: 1. Identity Preservation (IdP) A spelling takes its form from the spellings of its subcomponents. Hence related words have related spellings. 2. Phonographic Matching (PhM) The spelling must match the known phonological form and changes may be made, where necessary, if possible, by amending the spelling-to-sound correspondences 3. Distinctiveness (DSTNCT) Words with different meanings should have different spellings, if possible. 4. Invariant spelling Once a spelling has been settled upon, it is not changed. 5. Any spelling All words must have some spelling. It has long been known that polymorphemic words tend to keep the spelling of their constituent morphemes ‘as much as possible’ (Venezky 1970: 120), and I examine how much this is possible, by looking at how amendments are made, firstly in inflected forms, and then in affixed derivation, where the complicating factor of etymology often plays a part. It is argued here that morphemic spelling is fundamentally the same as etymological spelling, as both are different manifestations of IdP, one connecting the meaning of related English words, in spite of differences (electric, electricity, electrician) the other connecting English to other languages (e.g. psychology, courgette). English spellings are thus formed from ‘lengthwise’ units of spelling, which may be reduced but never broken (psychotherapist, psychedelic, psy trance). English spelling formation is modelled visually by constructing orthographic trees and comparing orthographic structure against phonological structure. Where conflicts arise in the application of IdP, draft spellings can amended where appropriate, hence <sin>+<ing>  <sinning> not *<sining>. A simple notation is introduced to show the main details, and this may be applicable in the classroom. The model also provides a viable path for the under-explained phenomenon of spelling pronunciation. It is also shown that new and creative spellings may be subject to users’ awareness of higher-level structures in the formation of spellings, and this explains why ‘constructed homophones’ are possible, where distinctive spellings are formed by altering units of spelling, at any phonological level, hence the band names OutKast, Altern-8, Xzibit, INXS, etc. These spellings require readers to reconstruct a known phonological form, from a new spelling. This logic can be extended to all spelling, and the argument here is that readers must map from spelling to sound at several phonological levels (phonemes, rhymes, syllables, feet), an extension of Venezky’s (1970) model. Another central argument of the thesis is that new spelling units arise due to changes in phonology that are not matched by changes in spelling. Hence old spelling units can be redeployed for new purposes in new spellings. The English writing system is thus characterised as being in permanent flux, as it accumulates new spellings over the centuries.

Skase Journal of Theoretical Linguistics, 2011

This study of standard and non-standard English spelling patterns makes steps towards developing a theory of spelling which can explain both within the one model. The study explores the formal differences between standard English spelling and ‘constructed homophony’, a sub-branch of non-standard spelling, and some existing theories of standard spelling are updated to cover both. In the latter section, the goal is to provide the outline of a visual model which can be used to predict all possible well-formed English spellings of all possible well-formed English words, even allowing for the interchanging of phonograms, morphograms and syllabograms.

The analysis of the distribution of English compounds extracted from the LDCE enables us to note some important constraints governing the frequency of English compound types and to note some definite tendencies in their spelling. It is not difficult to notice that the number of compound types progressively decreases as the complexity of their components increases. There are 3456 compounds whose components are both simplex nouns; the number of compounds in which one component is a suffixal derivative is much smaller –1374, and even lower is the number of compounds in which both components are suffixal derivatives, only 116. A prefixal derivative in one component appears in 55 compounds, but both components are prefixal derivatives in only one compound. There are altogether 210 compounds in which one component is also a compound, but there is only one compound in which both components are compounds. 57 compounds include a phrase in one of their components, but none have phrases in both components. Simultaneously with the increased complexity of the constituents, fewer compounds are written solid, and must increasingly be written open or hyphenated. This analysis suggests that the human mental ability to process more complicated compound structures in the course of normal social intercourse is quite limited. Recent psycholinguistic research has confirmed that compounds written with a space are easier to parse and understand. As a visual cue – a space – cannot be supplied in speech, compound 16 types with complex constituent structures, although quite frequent in some types of written English, are not used so often in spoken English (Miller 2006). For example, in our corpus there is just one compound with a recursive structure – daylight saving time. Without exception, compounds of this type are always spelled open – writing so the speller shows his/her willingness to make the reader’s task easier.31 This compound, the only recursive compound in the LDCE, cannot be found in the spoken component of the BNC. We should note, however, that compounds do not seem to be used frequently in English according to the frequency lists in Francis and Kučera (1982) and Leech et al. (2001). The reason for such ranking of compounds is that frequency lists do not count open compounds which are very common in written English, especially in the press, and solid compounds do not seem to be particularly productive in English (Inhoff et al. 2000, Juhasz 2003). Hence, we can conclude that solid compounds seem to differ from open compounds in four important features: spelling, morphological structure, type frequency and productivity. In spoken English, on the other hand, the counterparts of open compounds tend to realize only those compound types which have simpler constituent structure, although one must presume some variability depending on the genre. The behaviour of compounds in spoken English is still an area in which much more research is needed. I argue, therefore, that the ease of morphological processing influences the distribution of complex words in English compounds and the way in which they are written. Compound types which are easier to process are more frequent, and are also more often spelled solid. In recent linguistics, there have already appeared analyses which proposed accounting for morphological facts by the way language is usually processed. For example, Cutler et al. (1985) argued that the preference of language users to process stems before affixes has as a result the wider use of suffixes than prefixes in world languages. Similarly, Hay argued that the ordering of affixes can be explained by the ease of parsability of particular affixes (s. also Aarts 2004). The case of spelling is, however, more complicated – as a part of communicative acts, the account of it must involve pragmatic factors. The spelling can be explained only if we consider the options which the speller has in the given communicative situation. Following Grice’s communicative postulates, the speller has to try to make the message most easily available to the reader, and normally adopts the way of spelling suitable for the message he/she wants to communicate. The variability of the spelling that we encounter with some compounds is connected with the fact that spellers do not necessarily make the same decisions. Our prediction, the verification of which is left for further research, is that compounds with complex constituents should not in a greater number take part in the variability of spelling. It is clear that my observations on type frequencies, structure and spelling of English compounds need to be confirmed in a much larger corpus. Objections could be raised that these observations merely reflect the specificities of spelling of compounds in the LDCE, but I believe that the correlation with the structure of compounds points to their more general holding.

Yearbook of Morphology 1998, 1999

2000

The influence of spelling conventions on phonological knowledge was investigated. In Experiment 1 five-year old preliterate children and eight-year-old literate children were compared on their intuitive syllabification (word fragmentation) of disyllabic Dutch words with a single intervocalic consonant (e.g. /åp\l/, 'apple'). The larger number of ambisyllabic responses in the older age group could either be a reflection of the eight-year-olds' more mature phonology or an interaction between phonological knowledge and spelling conventions. Experiments 2 and 3, using literate and illiterate adults respectively, were designed to disentangle these alternative accounts.

1972

Although approximate) one-half of the English 1 'xicon can be spelled according to pho me-grapheme correspondences, many words in the remaining half of the le con can also be spelled systematically on the basis of their morpheme-properties rather than on the bases, V of their pronunciations. This paper discusses the bases, for assuming that English orthography is organized morphologically as well as phonologically and examines a number of spelling problems Which can be solved on the basis of morphemic informatioh. A list of references is included. (Author/JM). a * * 'via the ERIC Document Reproduction Service.(EDRS). EDRS is not * responsible for the quality of the original document. Reproductions * * supplied by EDRS are the best that can be made from the original.

An overview of the goals of English orthography counters the misconception that its spelling is chaotic and unprincipled. Direct representation of the speaker's phonemes is not its only goal. But even the sound-to-letter correspondences are not as inconsistent as widely believed. A survey of first-grade text vocabulary shows that spelling consistency is increased significantly if one takes into account the position of the phoneme within the syllable and the identity of the phonemes in the environment. Environmental influences within the rime are especially important. Understanding these patterns may reduce the complexity of spelling for learners and those with spelling problems.

Word Structure, 2015

This reference work on English morphology can be qualified as the (for a long time needed) successor to Marchand's famous handbook The categories and types of present-day English word formation, of which the second and last edition was published in 1969 . The book to be reviewed here, however, has a larger scope, as it does not only deal with word formation but also with inflection. Hence, it is a comprehensive book on English morphology. The authors of this book are all senior researchers in the domain of English morphology, with an individual track record of important publications on English morphology. So it was a good idea of these authors to work together to produce an authoritative volume on English morphology. What are the main features of this book compared to Marchand's book? First of all, it incorporates the results of decades on research on English morphology since the 1960's. Second, it is based on huge corpora, of a size that was unthinkable in the time that Marchand wrote his book. The main corpus used are COCA (the Corpus of Contemporary American English), the British National corpus, CELEX, and the Google Book Corpus. In addition, various dictionaries and reverse dictionaries were used. Many examples of complex words are

Linguistica, 2005

It is a well-known fact that in English, syllabification of derived words differs according to the attaching affix, Chomsky and Halle (1968). In words such as hinder, meter, burgle the final sonorant of the roots /hindr/, /mitr/, /burgl/ is syllabic in word final position, following the rule of schwa insertion that makes a final sonorant preceded by a consonant syllabic. However, in related forms where these roots are followed by a vowel-initial affix, such as hindrance, metric, burglar, the sonorants in question are not syllabic, butare syllabified as onsets of the following syllable. Not all affixes beginning in a vowel have the same effect on syllabification. The participle forming affixing triggers the schwa-insertion regardless of its vowel-initial status, e.g. (hinder /hindgr/: hindrance /hindrans/, but hindering /hindgril]/, */hindril]/). Chomsky and Halle (1968) treat this property as inherent to the attaching affix; i.e.-ance in hindrance differs from-ing in hindering with respect to the triggering of the schwa-insertion rule. Using a finer-grained syntax of words, this paper derives the differences in pronunciation of the above mentioned words as following not exclusively from a diacritic on the affix, as in Chomsky and Halle (1968), but rather from the attachment position of the affix in the syntactic structure of the word.