Lexical Grammar

In this paper wc examine a subset of polyscmous elements, the logical structure of nominals, and argue that maw cases of polysemy have well-defined calculi, which interact with the grmnmar in predictable and determinate ways for disambiguation. These calculi constitute part of the lexicai organization of the grammar and contribute to the lexical semantics of a word. The lexieal system of the grammar is distinct from the conceptual representation associated with a lcxieal item, where polysemy is less constrained by grarmnar. We propose a structured' semantic representation, the Lexical Conceptual Paradigm (LCP) which groups nouns into paradigmatic classes exhibiting like behavior.

This paper describes the basic architectural concepts that underlie the formal theory of Lexical-Functional Grammar. The LFG formalism, which h a s e v olved from previous computational, linguistic, and psycholinguistic research, provides a simple set of devices for describing the common properties of all human languages and the particular properties of individual languages. It postulates two l e v els of syntactic representation for a sentence, a constituent structure and a functional structure. These are related by a piece-wise correspondence that permits the properties of the abstract functional structure to be de ned in terms of con gurations of constituent structure phrases. The basic architecture crucially separates the three notions of structure, structural description, and structural correspondence. This paper also outlines some recent extensions to the original LFG theory that enhance its ability t o express certain kinds of linguistic generalizations while remaining compatible with the underlying architecture. These include formal variations in the elementary linguistic structures, in descriptive notation, and in the arrangement of correspondences.

Two major religions, Judaism and Christianity, use the ancient Hebrew Bible as Holy Scripture. These
books were translated in the last three centuries before the common era. The oldest of these
translations is the Septuagint, a Greek translation. Not only are the Hebrew and Greek texts that were
involved in the original translation process missing, but precious little is known about the doctrine and
translation methods of the translators of the Septuagint. Much can be learned about these crucial
issues, however, if the translation technique followed by those ancient translators is studied by
comparing the present Hebrew and Greek texts.

A new method to determine and describe the translation technique of the Septuagint was proposed
and tested in this dissertation. This method is based on the use of the Computer Assisted Tools for
Septuagint Studies (CATSS) data base and statistical methods. The translation technique of the book
Deuteronomy was described using different criteria, all of which measure the frequency of non-literal
renderings. Three different groups of criteria were utilized, viz. the Tov criteria as proposed by E. Tov,
criteria defined using the markers in the CATSS data base called the CATSS criteria, and grammatical
criteria using the person of the verb. Each criterion was applied to the data base individually. The
translation units were determined first, after which the translation technique found within the
translation unit was described. The methodology implemented discriminates between significant and
insignificant trends in translation technique.

It became clear that the results of the different criteria indicate different translation units and different
translation techniques for each of the criteria. Except for some criteria using the person of the verb,
very little indication was found that the traditional translation units are supported by the data used in
this study. In fact, it seems as if translation units should be determined before the translation
technique is described. The translation technique should then be described according to the indicated
units.

Not all the Tov criteria could be utilized, but their results are in agreement to some extent. The CATSS
criteria proved to be more difficult to implement than expected, but some of the criteria rendered
excellent results. The person of the verb was discussed in detail using 12 different criteria. The results
of the criteria utilizing the person of the verb are disappointing, and provide some scope for future
research.

The results rendered by this new approach are firm and easy to interpret. In addition, it is possible to
utilize these results when dealing with specific text-critical problems.

Our goal is to develop a semantic theory that is equally suitable for the lexical material (words) and for the larger constructions (sentences) put together from these. In 2.1 we begin with the system of lexical categories that are in generative grammar routinely used as preterminals mediating between syntax and the lexicon. Morphology is discussed in 2.2, where subdirect composition is introduced. This notion is further developed in 2.3, where the geometric view is expanded from the standard word vectors and the voronoids introduced in Chapter 1 to include non-vectorial elements that express binary relations. These eigenspace techniques receive further use in 2.4, where some crucial relational devices of syntactic theory, thematic relations, deep cases, and kārakas are addressed. How much of syntax can be reconstructed with these is discussed in 2.5. 2.1 Lexical categories and subcategories Whether a universal system of lexical categories exists is still a widely debated question. Bloomfield, 1933, and more recently Kaufman, 2009 argued that certain languages like Tagalog have only one category. But the notion that there are at least three major categories that are universal, nouns, verbs, and adjectives, has been broadly defended (Baker, 2003; Chung, 2012; Haspelmath, 2021). 4lang subdivides verbs into two categories: intransitive U and transitive V; retaining the standard N for noun; A for adjective; and also uses D for aDverb; and G for Grammatical formative. While this rough categorization has proven useful for seeking bindings in the original 4 and in other languages, there is no theoretical claim associated to these categories, nei

In a collocation, the choice of one lexical item depends on the choice made for another. This poses a problem for simple approaches to lexicalisation in natural language generation systems. In the Meaning-Text framework, recurrent patterns of collocations have been characterised by lexical functions, which offer an elegant way of describing these relationships. Previous work has shown that using lexical functions in the context of multilingual natural language generation allows for a more efficient development of linguistic resources. We propose a way to encode lexical functions in the Lexical Functional Grammar framework.

Different from most previous studies of language transfer phenomena in second language learning which remain at an observational and descriptive level, this study proposes that the major source of language transfer is the interference of first language abstract lexical structure. It assumes that any interlanguage system, like other linguistic systems, has an abstract lexical structure containing several discrete but interacting subsystems: lexical-conceptual structure, predicate-argument structure, and morphological realization patterns. Unlike abstract lexical structures in other linguistic systems, the abstract lexical structure in second language learning has different sources. This study claims that any abstract lexical structure in second language learning contains more abstract elements than surface configurations of language, that is, language-specific lemmas underlie lexical entries, and such language-specific lemmas are in contact in second language learning, which can be s...

Wasow (1977) argues that linguistic theory should recognize two qualitatively distinct types of rules: syntactic rules, which can affect more "superficial" grammatical function properties; and lexical rules, which affect deeper lexical semantic properties of lexical items. However, lexicalist theories of grammar have replaced syntactic rules with lexical rules leaving Wasow's dichotomy potentially unexplained. Our goal in this paper is to recapture Wasow's insight within a lexicalist framework such as HPSG. Building on Sag & Wasow's (1999) distinction between lexeme and word, we claim that there is a contrast between lexical rules that relate lexemes to lexemes (L-to-L rules) and lexical rules that relate words to words (W-to-W rules) and that these differences follow from the architecture of the grammar. In particular, we argue that syntactic function features (ARGST, VALENCE, etc.) are not defined for lexemes, while lexical semantic features (CONTENT) are. Fr...

Our goal is to develop a semantic theory that is equally suitable for the lexical material (words) and for the larger constructions (sentences) put together from these. In 2.1 we begin with the system of lexical categories that are in generative grammar routinely used as preterminals mediating between syntax and the lexicon. Morphology is discussed in 2.2, where subdirect composition is introduced. This notion is further developed in 2.3, where the geometric view is expanded from the standard word vectors and the voronoids introduced in Chapter 1 to include non-vectorial elements that express binary relations. These eigenspace techniques receive further use in 2.4, where some crucial relational devices of syntactic theory, thematic relations, deep cases, and kārakas are addressed. How much of syntax can be reconstructed with these is discussed in 2.5. 2.1 Lexical categories and subcategories Whether a universal system of lexical categories exists is still a widely debated question. Bloomfield, 1933, and more recently Kaufman, 2009 argued that certain languages like Tagalog have only one category. But the notion that there are at least three major categories that are universal, nouns, verbs, and adjectives, has been broadly defended (Baker, 2003; Chung, 2012; Haspelmath, 2021). 4lang subdivides verbs into two categories: intransitive U and transitive V; retaining the standard N for noun; A for adjective; and also uses D for aDverb; and G for Grammatical formative. While this rough categorization has proven useful for seeking bindings in the original 4 and in other languages, there is no theoretical claim associated to these categories, nei

This paper describes the central role played by default inheritance in Word Grammar, a theory of language knowledge and processing. A single formalism is used to represent knowledge at the levels of morphology, syntax, and semantics. A single rule of inference is used to inherit knowledge at all of these levels. This rule is distinctive in that it requires defaults to be explicitly overridden in the case of exceptions. The explicit overriding rule is used in syntax to achieve what other theories achieve by means of transformations, metarules, or lexical rules.

We probably all share an interest in syntax, so we would dearly love a clear and certain answer to the question: what is syntactic structure like? Is it based on dependencies between words, or on phrases? What kinds of relation are there? And so on. But before we can answer relatively spe­ cific questions like these, we must first answer a much more general question: What kind of thing do we think language is? Or maybe: Where do we think language is – nowhere, in society, in our minds? Our answer will decide what basic assumptions we make, and how our discipline, linguistics, relates to other disciplines. Is language a set of abstract patterns like those of mathematics, without any particular location? This is a popular answer, and makes a good deal of sense. After all, what is language if not ab­ stract patterning? The patterns made by words in a sentence, or by segments in a syllable, are cer­ tainly abstract and regular, and can be studied as a branch of mathematics – as indeed t...

English gerunds such as (We weretalking about) John having a sabbaticalcombine the internal characteristics ofa clause with the external characteristicsof a noun phrase. Previous analyses havetried to recognise the mixed characterof gerunds by assigning them two separatenodes, one verbal and the other nominal.However, dependency analyses such as WordGrammar allow only one node perword, so they do not allow analyses of thiskind. Two-node analyses are strongevidence against dependency analysis, so itis important to be sure that they areneeded. The present paper presents ananalysis similar to the one recently proposedby Malouf in which the verbal and nominalclassifications are combined on asingle node which inherits both verbaland nominal characteristics; butunlike Malouf's analysis, it does notassume phrase structure. Like his, it exploitsthe logic of multiple default inheritance whichallows a single node to inherit fromtwo supercategories, in this case from both`verb' and `no...

A recent trend in psycholinguistic research has been to posit prediction as an essential function of language processing. The present paper develops a linguistic perspective on viewing prediction in terms of pre-activation. We describe what predictions are and how they are produced. Our basic premises are that (a) no prediction can be made without knowledge to support it; and (b) it is therefore necessary to characterize the precise form of that knowledge, as revealed by a suitable theory of linguistic representations. We describe the Parallel Architecture (PA: Jackendoff, 2002; Jackendoff & Audring, 2020), which makes explicit our commitments about linguistic representations, and we develop an account of processing based on these representations. Crucial to our account is that what have been traditionally treated as derivational rules of grammar are formalized by the PA as lexical items, encoded in the same format as words. We then present a theory of prediction in these terms: linguistic input activates lexical items whose beginning (or incipit) corresponds to the input encountered so far; and prediction amounts to pre-activation of the as yet unheard parts of those lexical items (the remainder). Thus the generation of predictions is a natural byproduct of processing linguistic representations. We conclude that the PA perspective on pre-activation provides a plausible account of prediction in language processing that bridges linguistic and psycholinguistic theorizing.

The analysis of counterfactuality needs a good shake-up. Wolfgang Klein's article does an excellent job in pointing out what is stale and unproductive in the literature, often inherited from philosophical discussions with limited linguistic relevance. And he posits a fresh alternative that is closer to the linguistic facts, especially as we know them from English and other Germanic languages. In this response we discuss three points that, we believe, are compatible with Klein's proposal but may bring it even closer to the linguistic facts, including for languages not discussed in Klein's paper. Apart from English, the data we use mainly come from our work on Australian languages (Verstraete 2005a, 2006, Luk in preparation), but also from a study of a broader cross-linguistic sample (Van linden & Verstraete 2008). 1. Polarity reversal rather than truth or falsity Much of the discussion of counterfactuality is framed in terms of truth or falsity (like the term itself, in fact), as inherited from work in logic and adjacent areas of philosophy. Klein's target article nicely points out that there are problems with this approach: neither counterfactual protases nor apodoses have to be 'contrary to the facts', as demonstrated by his examples (4) and (5). Instead, Klein proposes an analysis in terms of the relation between actual and nonactual worlds: a counterfactual protasis introduces a world that is not the actual one but otherwise warranted as the actual one. Under this analysis, the default interpretation of this type of protasis as 'contrary to the facts' comes from the fact that not talking about the actual world typically has a reason, the most obvious one being "that nothing can be said about the actual world" (p. 26). But this does not preclude that the situation in the protasis could also apply to the actual world, as in Klein's example (4). The last part of this argument is somewhat similar in spirit to Gricean approaches to counterfactuals (Ziegeler 2000, Verstraete 2005a, Van linden & Verstraete 2008), which state that in a 'knowable' domain like the past, using a modalized expression (e.g. X would have Yed) is less informative than using its bare counterpart (X Yed), and thus triggers the quantity-based implicature that 'X Yed' does not apply, i.e. 'X did not Y'. As an implicature, however, this can be cancelled and is thus not an inherent part of the semantics of counterfactuals, just as Klein shows for his (4).

Penelitian ini dilaksanakan untuk mengetahui apakah online reading dan discussion boardspada facebook sebagai kegiatan prewriting dapat meningkatkan kualitas writinglebih baik daripada kegiatan pre-writing melalui printed text reading, aspek writing apa yang meningkat secara signifikan dan persepsi siswa terhadap ORDB. Pada desain quantitatif, uji tes awal dan tes akhir kelompok kontrol dilakukan dan pada desain qualitatif, kuesioner dilaksankan.Pada pengumpulan data, teswriting, analisa dokumen, dan kuesioner digunakan. Subyek dibagi dua kelompok, eksperimen dan kontrol, masing-masing terdiri dari 17 siswa. Hasil penelitian menunjukan nilai signifikan dua arah 0.000. Berdasarkan aspek writing yang meningkat secara signifikan, ORDB meningkatkan kualitas writing terutama pada vocabulary. Siswa meningkat pada vocabulary dengan penambahan 1.62 (8.1%) dengan pre-writing melalui printed-text reading dan 3.06 (15.3%) dengan ORDB. Hasil penelitian menunjukkan perbedaan pada kedua kegiatan ...

Korean has a complex inflectional system, showing agglutinative morphology and using affixation as the major mechanism for word formation. A prerequisite to the successful development of any syntactic/semantic parsers for the language thus hinges on the efficient lexicon that can syntactically expand its lexical entries and map into syntax and semantics with robust parsing performance. This paper reports the system of the Korean Resource Grammar developed as an extension of HPSG (Head-driven Phrase Structure Grammar) and the results of implementing it into the Linguistic Knowledge Building (LKB) system (cf. Copestake 2002). The paper shows that the present grammar proves to be theoretically as well as computationally efficient enough in parsing Korean sentences.

The task of translating between programming languages differs from the challenge of translating natural languages in that programming languages are designed with a far more rigid set of structural and grammatical rules. Previous work has used a tree-to-tree encoder/decoder model to take advantage of the inherent tree structure of programs during translation. Neural decoders, however, by default do not exploit known grammar rules of the target language. In this paper, we describe a tree decoder that leverages knowledge of a language's grammar rules to exclusively generate syntactically correct programs. We find that this grammar-based tree-to-tree model outperforms the state of the art tree-to-tree model in translating between two programming languages on a previously used synthetic task.

The task of translating between programming languages differs from the challenge of translating natural languages in that programming languages are designed with a far more rigid set of structural and grammatical rules. Previous work has used a tree-to-tree encoder/decoder model to take advantage of the inherent tree structure of programs during translation. Neural decoders, however, by default do not exploit known grammar rules of the target language. In this paper, we describe a tree decoder that leverages knowledge of a language's grammar rules to exclusively generate syntactically correct programs. We find that this grammar-based tree-to-tree model outperforms the state of the art tree-to-tree model in translating between two programming languages on a previously used synthetic task.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

Lexical rules are used in constraintbased grammar formalisms such as llead-Driven l)hrase Structure Grammar (IIPSG) (Pollard and Sag 1994) to express generalizations atnong lexical entries. '['his paper discusses a number of lexical rules from recent I[PSG analyses of German (tlinri<;hs and Nakazawa 1994) and shows that the grammar in some cases vastly overgenerates and in other cases introduces massive spurious structural ambiguity, if lexical rules ap: ply under unification. Such l)rot)lems of overgeneration or spurious ambiguity do not arise, if a lexical rule al)plies to a given lexical ent;ry iff the lexical entry is subsumed by the left:hand side of the lexical rule. I,'inally, the paper discusses computational consequcnce~s of at)plying lexical rules under subsuml)tion.

This paper aims to examine how to analyse characteristics of lexical collocations of individual language from the typological point-of-view. For the purpose, this paper suggest an method, or Lexical Function, that was created and used mainly by Igor Mel'čuk. According to his approach, all lexical collocation can be explained by lexical function which grasps the lexical relation between the base and collocate. In this paper I showed some examples of lexical function-standard lexical function and non-standard lexical function-which can be used to analysing the characteristics of Korean collocations by comparing the Korean collocation to Japanese and Chinese ones. For instance, Oper1 and Open/Close are very useful lexical functions for showing the common point and the difference among those languages from the perspective of linguistic typology.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

The present study tries to explore the knowledge and the use of verb-noun collocations in spoken and written discourse of English as foreign language (EFL) learners in Iran. In addition, the current study aims at examining and describing possible correlation and the differences in knowledge and use of verb-noun collocations among four groups of Iranian EFL learners representing four academic years (freshman, sophomore, junior, and senior). The study also qualitatively examines the factors that influence Iranian EFL students’ use of verb-noun collocations. The participants in this study comprised of four groups of 212 male and female English students according to their academic years in the Department of English Translation, at the Islamic Azad University, Tehran North Branch (IAUTNB) in Tehran, Iran. Data was collected by two tests: a Cloze Test, and a collocation test, and a series of six pictures for writing and speaking. The Test of Collocation was a C-Test in which there were 50...

In this paper wc examine a subset of polyscmous elements, the logical structure of nominals, and argue that maw cases of polysemy have well-defined calculi, which interact with the grmnmar in predictable and determinate ways for disambiguation. These calculi constitute part of the lexicai organization of the grammar and contribute to the lexical semantics of a word. The lexieal system of the grammar is distinct from the conceptual representation associated with a lcxieal item, where polysemy is less constrained by grarmnar. We propose a structured' semantic representation, the Lexical Conceptual Paradigm (LCP) which groups nouns into paradigmatic classes exhibiting like behavior.

Introduction 2 An overview of the most relevant approaches to lexical representation 3 The format of a lexical template 4 Lexical templates in the primary lexicon 4.1 Contact-by-impact verbs 4.2 Break verbs 4.3. Consumption verbs 4.4 Sound verbs 4.5 Cessative phase of verbs of existence 5 Constraining the generative power of a lexical template 6 Final remarks References

Large-scale knowledge-based machine translation requires significant amounts of lexical knowledge in order to map syntactic structures to conceptual structures. Tfiis paper presents a framework in which lexical knowledge is separated into different levels of representation, which are arranged in a hierarchical model based on principles of knowledge representation and lexical semantics. The proposed methodology is language-independent, and has been used to organize lexical knowledge for both English and Japanese.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

* We wish to thank the audience and organizers of the first NetWordS workshop. P. Acquaviva's research was supported by an Alexander von Humboldt fellowhip, which is gratefully acknowledged. Faults and omissions are the authors' responsibility.

Penelitian ini bertujuan untuk meningkatkan kemampuan menulis recount text siswa pada kelas VIII SMP Tiga Hati Kepenuhan Hulu, menggunakan media surat pribadi.Penelitian ini adalah Penelitian Tindakan Kelas (PTK) yang bertujuan untuk memecahkan masalah pada kemampuan siswa dalam menulis dalam bahasa inggris. Peneliti melakukan dua siklus dimana setiap siklus terdiri dari perencanaan, tindakan, observasi dan refleksi. Data kualitatif dikumpulkan melalui observasi, catatan harian dan wawancara, sedangkan kuantitatif dikumpulkan melalui tes. Hasil dari penelitian ini menunjukan bahwa penerapan media surat telah sukses sejak ada peningkatan pada kemampuan menulis siswa. Hasil dari penelitian ini menunjukkan adanya peningkatan kemampuan menulis siswa dari rata-rata siklus 1 (60,85) ke (73,50) rata-rata di siklus 2.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

The purpose of this paper is to provide a description of negative word formation processes in Spanish and English. The methodology for the description is based upon the conception of word formation in the Lexical Grammar Model, which considers it as a phenomenon with two fundamental perspectives: (i) as a lexicological process of creation of a new lexical unit out of existing lexical material; that is, the derived word is the result of combining predicates, which can be free or bound morphemes; and (ii) as a grammatical process that involves the interplay of different types of conditions that range from the phonological level to the pragmatic one, via the syntactic and the morphological stages.

The objectives of this study were to investigate how the lexical chunk approach can be used in teaching writing to KPTM students, improve their performance in writing skill and determine students’ perception of the lexical chunk approach in writing. This is a descriptive study that combined both qualitative and quantitative designs. A total of 48 students taking diploma courses at Kolej Poly-Tech Mara, Bangi were selected as sample using purposive sampling method. The field work carried out for twelve weeks. The instruments used in this study include written test taken from question bank, interview and demographic questionnaire which had been validated by two experienced lecturers. Prior to the actual research work, a pilot study was carried out using 15 KPTM students. In answering the first research question, thorough explanation on how the lexical chunk approach was taught in writing class was given, including the teaching materials used in accomplishing the mission. For the secon...

Considering the seq2seq architecture of Yin and Neubig (2018) for natural language to code translation, we identify four key components of importance: grammatical constraints, lexical preprocessing, input representations, and copy mechanisms. To study the impact of these components, we use a state-of-the-art architecture that relies on BERT encoder and a grammar-based decoder for which a formalization is provided. The paper highlights the importance of the lexical substitution component in the current natural language to code systems.

English gerunds such as (We were talking about) John having a sabbatical combine the internal characteristics of a clause with the external characteristics of a noun phrase. Previous analyses have tried to recognise the mixed character of gerunds by assigning them two separate nodes, one verbal and the other nominal. However dependency analyses such as Word Grammar allow only one node per word, so they do not allow analyses of this kind. Two-node analyses are strong evidence against dependency analysis so it is important to be sure that they are needed. The present paper presents an analysis similar to the one recently proposed by Malouf in which the verbal and nominal classifications are combined on a single node which inherits both verbal and nominal characteristics; but unlike Malouf's analysis it does not assume phrase structure. Like his, it exploits the logic of multiple default inheritance which allows a single node to inherit from two supercategories-in this case from both 'verb' and 'noun'. As Malouf points out, multiple inheritance works because English grammar is organised in such a way that the characteristics of these categories are orthogonal. In short, a gerund is both a verb and a noun, as in traditional analyses. Simple stipulations are needed to allow for 'possessive' subjects (e.g. about John's having a sabbatical) and a number of very specific constructions peculiar to gerunds: no in prohibitions or existentials (e.g. No playing loud music! There's no mistaking that voice), and a very few constructions which demand a gerund rather than a noun phrase (e.g. It's no use ..., They prevented us from ...). (2) a. We were talking about John not having a sabbatical. b. We were talking about John soon having a sabbatical. These are characteristics which not only distinguish verbs from nouns but also distinguish them, at least in combination, from other word classes. On the other hand, it must also be a noun because the phrase that it heads is used as the object of a preposition (about), and could be used in any other position where plain noun phrases are possible: (3) a. John having a sabbatical upset Bill.

The categorization and usage of conditional clauses has drawn the interest of generations of linguists working in various linguistic frameworks. Introducing the students to the academic studies on this topic is a compulsory prerequisite, but it should be properly balanced with their own research experience and that is where language resources fit in. The paper focuses on conditional clauses presented through lexical-semantic databases such as FrameNet and a number of freely available English corpora-The British National Corpus (BNC), The Corpus of Contemporary American (COCA), The News on the Web Corpus (NOW), etc. Practical applications The paper presents an innovative approach to teaching conditional sentences in an English Practice Class. The main objective is to implement language resources such as lexical-semantic databases and corpora in order to enhance the students&#39; analytical abilities and improve their language competency. In addition, students can apply this approach ...

The categorization and usage of conditional clauses has drawn the interest of generations of linguists working in various linguistic frameworks. Introducing the students to the academic studies on this topic is a compulsory prerequisite, but it should be properly balanced with their own research experience and that is where language resources fit in. The paper focuses on conditional clauses presented through lexical-semantic databases such as FrameNet and a number of freely available English corpora-The British National Corpus (BNC), The Corpus of Contemporary American (COCA), The News on the Web Corpus (NOW), etc. Practical applications The paper presents an innovative approach to teaching conditional sentences in an English Practice Class. The main objective is to implement language resources such as lexical-semantic databases and corpora in order to enhance the students&#39; analytical abilities and improve their language competency. In addition, students can apply this approach ...

Cet article propose une nouvelle facon d&#39;analyser des unites phraseologiques. Il s&#39;agit d&#39;aller au dela de la simple classification en collocations et en idiomes. Dans la premiere partie de cet article, nous considerons les unites phraseologiques en tant qu&#39;objets du discours et non en tant qu&#39;unites syntaxiques ou semantiques. Dans la partie principale, nous appliquons cette methode au cas d&#39;une nouvelle unite phraseologique emergeant dans «Trainspotting» de l&#39;ecrivain ecossais, Irvine Welsh.

Cet article propose une nouvelle façon d'analyser des unités phraséologiques. Il s'agit d'aller au delà de la simple classification en collocations et en idiomes. Dans la première partie de cet article, nous considérons les unités phraséologiques en tant qu'objets du discours et non en tant qu'unités syntaxiques ou sémantiques. Dans la partie principale, nous appliquons cette méthode au cas d'une nouvelle unité phraséologie émergeant dans Trainspotting de l'écrivain écossais, Irvine Welsh.

An automatic method for annotating the Penn-II Treebank (Marcus et al., 1994) with high-level Lexical Functional Grammar (Kaplan and Bresnan, 1982; Bresnan, 2001; Dalrymple, 2001) f-structure representations is presented by Burke et al. (2004b). The annotation algorithm is the basis for the automatic acquisition of wide-coverage and robust probabilistic approximations of LFG grammars (Cahill et al., 2004) and for the induction of subcategorisation frames (O'Donovan et al., 2004; O'Donovan et al., 2005). Annotation quality is, therefore, extremely important and to date has been measured against the DCU 105 and the PARC 700 Dependency Bank (King et al., 2003). The annotation algorithm achieves f-scores of 96.73% for complete f-structures and 94.28% for preds-only f-structures against the DCU 105 and 87.07% against the PARC 700 using the feature set of Kaplan et al. (2004). Burke et al. (2004a) provides detailed analysis of these results. This paper presents an evaluation of the annotation algorithm against PropBank (Kingsbury and Palmer, 2002). PropBank identifies the semantic arguments of each predicate in the Penn-II treebank and annotates their semantic roles. As PropBank was developed independently of any grammar formalism it provides a platform for making more meaningful comparisons between parsing technologies than was previously possible. PropBank also allows a much larger scale evaluation than the smaller DCU 105 and PARC 700 gold standards. In order to perform the evaluation, first, we automatically converted the PropBank annotations into a dependency format. Second, we developed conversion software to produce PropBank-style semantic annotations in dependency format from the f-structures automatically acquired by the annotation algorithm from Penn-II. The evaluation was performed using the evaluation software of Crouch et al. (2002) and Riezler et al. (2002). Using the Penn-II Wall Street Journal Section 24 as the development set, currently we achieve an f-score of 76.58% against PropBank for the Section 23 test set.

In this paper we present a number of experiments to test the portability of existing treebankinduced LFG resources. We test the LFG parsing resources of Cahill et al. (2004) on the ATIS corpus which represents a considerabley different domain to the Penn-II Treebank Wall Street Journal sections, from which the resources were induced. This testing shows an underperformance at both c-and f-structure level as a result of the domain variation. We show that in order to adapt the LFG resources of Cahill et al. (2004) to this new domain, all that is necessary is to retrain the c-structure parser on data from the new domain. This section describes our baseline experiments to determine the portability of the resources of Cahill et al. (2004) to a new domain, the ATIS corpus.

We investigate Arabic Context Free Grammar parsing with dependency annotation comparing lexicalised and unlexicalised parsers. We study how morphosyntactic as well as function tag information percolation in the form of grammar transforms (Johnson, 1998, Kulick et al., 2006) affects the performance of a parser and helps dependency assignment. We focus on the three most frequent functional tags in the Arabic Penn Treebank: subjects, direct objects and predicates. We merge these functional tags with their phrasal categories and (where appropriate) percolate case information to the non-terminal (POS) category to train the parsers. We then automatically enrich the output of these parsers with full dependency information in order to annotate trees with Lexical Functional Grammar (LFG) f-structure equations with produce f-structures, i.e. attribute-value matrices approximating to basic predicate-argument-adjunct structure representations. We present a series of experiments evaluating how well lexicalized, history-based, generative (Bikel) as well as latent variable PCFG (Berkeley) parsers cope with the enriched Arabic data. We measure quality and coverage of both the output trees and the generated LFG f-structures. We show that joint functional and morphological information percolation improves both the recovery of trees as well as dependency results in the form of LFG f-structures.

We present data-driven methods for the acquisition of LFG resources from two German treebanks. We discuss problems specific to semi-free word order languages as well as problems arising from the data structures determined by the design of the different treebanks. We compare two ways of encoding semi-free word order, as done in the two German treebanks, and argue that the design of the TiGer treebank is more adequate for the acquisition of LFG resources. Furthermore, we describe an architecture for LFG grammar acquisition for German, based on the two German treebanks, and compare our results with a hand-crafted German LFG grammar.

An automatic method for annotating the Penn-II Treebank (Marcus et al., 1994) with high-level Lexical Functional Grammar (Kaplan and Bresnan, 1982; Bresnan, 2001; Dalrymple, 2001) f-structure representations is presented by Burke et al. (2004b). The annotation algorithm is the basis for the automatic acquisition of wide-coverage and robust probabilistic approximations of LFG grammars (Cahill et al., 2004) and for the induction of subcategorisation frames (O'Donovan et al., 2004; O'Donovan et al., 2005). Annotation quality is, therefore, extremely important and to date has been measured against the DCU 105 and the PARC 700 Dependency Bank (King et al., 2003). The annotation algorithm achieves f-scores of 96.73% for complete f-structures and 94.28% for preds-only f-structures against the DCU 105 and 87.07% against the PARC 700 using the feature set of Kaplan et al. (2004). Burke et al. (2004a) provides detailed analysis of these results. This paper presents an evaluation of the annotation algorithm against PropBank (Kingsbury and Palmer, 2002). PropBank identifies the semantic arguments of each predicate in the Penn-II treebank and annotates their semantic roles. As PropBank was developed independently of any grammar formalism it provides a platform for making more meaningful comparisons between parsing technologies than was previously possible. PropBank also allows a much larger scale evaluation than the smaller DCU 105 and PARC 700 gold standards. In order to perform the evaluation, first, we automatically converted the PropBank annotations into a dependency format. Second, we developed conversion software to produce PropBank-style semantic annotations in dependency format from the f-structures automatically acquired by the annotation algorithm from Penn-II. The evaluation was performed using the evaluation software of Crouch et al. (2002) and Riezler et al. (2002). Using the Penn-II Wall Street Journal Section 24 as the development set, currently we achieve an f-score of 76.58% against PropBank for the Section 23 test set.

The distinction between raising and subject-control verbs, although crucial for the construction of semantics, is not easy to make given access to only the local syntactic configuration of the sentence. In most contexts raising verbs and control verbs display identical superficial syntactic structure. Linguists apply grammaticality tests to distinguish these verb classes. Our idea is to learn to predict the raising-control distinction by simulating such grammaticality judgments by means of pattern searches. Experiments with regression tree models show that using pattern counts from large unannotated corpora can be used to assess how likely a verb form is to appear in raising vs. control constructions. For this task it is beneficial to use the much larger but also noisier Web corpus rather than the smaller and cleaner Gigaword corpus. A similar methodology can be useful for detecting other lexical semantic distinctions: it could be used whenever a test employed to make linguistically interesting distinctions can be reduced to a pattern search in an unannotated corpus.