File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/98/w98-0501_metho.xml
Size: 26,642 bytes
Last Modified: 2025-10-06 14:15:06
<?xml version="1.0" standalone="yes"?> <Paper uid="W98-0501"> <Title>Towards an implementable dependency grammar</Title> <Section position="5" start_page="3" end_page="3" type="metho"> <SectionTitle> 3 Parallelism between the syntactic </SectionTitle> <Paragraph position="0"> and semantic structures Obviously, distributional descriptions that do not contribute to their semantic analysis can be given to linguistic strings. Nevertheless, the minimal descriptive requirement should be that a syntactic description is compatible with the semantic structure. The question which arises is that if the correspondence between syntactic and semantic structures exists, why should these linguistic levels be separated. For example, Sgall (1992, p. 278) has questioned the necessity of the syntactic level altogether. His main argument for dispensing with the whole surface syntactic level is that there are no strictly synonymous syntactic constructions, and he therefore suggests that the surface word order belongs more properly to the level of morphemics. This issue is rather complicated. We agree that surface word order does not belong to syntactic structure, but for different reasons. In contradistinction to Sgall's claim, Mel'~uk (1987, p. 33) has provided some evidence where the morphological marker appears either* in the head or the dependent element in different languages, as in the Russian &quot;kniga professor+a&quot; (professor's book) and its Hungarian equivalent '~professzor kSnyv/e'. Consequently, Mel'~uk (1987, p. 108) distinguishes the morphological dependency as a distinct type of dependency.</Paragraph> <Paragraph position="1"> Thus morphology does not determine the syntactic dependency, as Tesni~re (1959, Ch. 15) also argues.</Paragraph> <Paragraph position="2"> For Tesni~re (1959, Ch. 20:17) meaning (Fr. sens) and structure are, in principle, independent. This is backed by the intuition that one recognises the existence of the linguistic structures which are semantically absurd, as illustrated by the structural similarity between the nonsensical sentence &quot;Le silence vertebral indispose la voie licite&quot; and the meaningful sentence &quot;Le signal vert indique la voie libre&quot;. The independence of syntactic and semantic levels is crucial for understanding Tesni~re's thesis that the syntactic structure follows from</Paragraph> <Paragraph position="4"> the semantic structure, but not vice versa. This means that whenever there is a syntactic relation, there is a semantic relation (e.g. complementation or determination) going in the opposite direction. In this view, the syntactic head requires semantic complementation from its dependents. Only because the syntactic and semantic structures belong to different levels is there no interdependency or mutual dependency, though the issue is sometimes raised in the literature.</Paragraph> <Paragraph position="5"> There is no full correspondence between the syntactic and semantic structures because some semantic relations are not marked in the functional structure. In Tesni~re (1959, p. 85), for example, there are anaphoric relations, semantic relations without correspondent syntactic relations. null</Paragraph> </Section> <Section position="6" start_page="3" end_page="3" type="metho"> <SectionTitle> 4 Surface representation and </SectionTitle> <Paragraph position="0"> syntactic structure</Paragraph> <Section position="1" start_page="3" end_page="3" type="sub_section"> <SectionTitle> 4.1 The nucleus as a syntactic primitive </SectionTitle> <Paragraph position="0"> The dependency syntactic models are inherently more &quot;word oriented&quot; than constituentstructure models, which use abstract phrase categories. The notion of word, understood as an orthographic unit in languages similar to English, is not the correct choice as a syntactic primitive. However, many dependency theories assume that the orthographic words directly correspond r to syntactic primitives (nodes in the trees). Although the correspondence could be very close in languages like English, there are languages where the word-like units are much longer (i.e. incorporating languages).</Paragraph> <Paragraph position="1"> TesniSre observed that because the syntactic connexion implies a parallel semantic connexion, each node has to contain a syntactic and a semantic centre. The node element, or nucleus, is the genuine syntactic primitive. There is no one-to-one correspondence between nuclei and orthographic words, but the nucleus consists of one or more, possibly discontiguous, words or parts of words. The segmentation belongs to the linearisation, which obeys language-specific rules. Tesni~re (1959, Ch 23:17) argued that the notion word, a linear unit in a speech-chain, does not belong to syntactic description at all.</Paragraph> <Paragraph position="2"> A word is nothing but a segment in the speech chain (1959, Ch 10:3).</Paragraph> <Paragraph position="3"> 7See Kunze (1975, p. 491) and Hudson (1991).</Paragraph> <Paragraph position="4"> The basic element in syntactic description is the nucleus. It corresponds to a node in a dependency tree. When the sentence is represented as a dependency tree, the main node contains the whole verb chain.</Paragraph> <Paragraph position="5"> There are at least two reasons why the concept of the nucleus is needed. In the first place, there are no cross-linguistically valid criteria to determine the head in, say, a prepositional phrase. One may decide, arbitrarily, that either the preposition or the noun is the head of the construction. Second, because the nucleus is also the basic semantic unit, it is the minimal unit in a lexicographical description.</Paragraph> </Section> <Section position="2" start_page="3" end_page="3" type="sub_section"> <SectionTitle> 4.2 Linearisation </SectionTitle> <Paragraph position="0"> Tesni~re makes a distinction between the linear order, which is a one-dimensional property of the physical manifestations of the language, and the structural order, which is two-dimensional.</Paragraph> <Paragraph position="1"> According to his conception, constructing the structural description is converting the linear order into the structural order. Restricting himself to syntactic description, Tesni~re does not formalise this conversion though he gives two main principles: (1) usually dependents either immediately follow or precede their heads (projectivity) and when they do not, (2) additional devices such as morphological agreement can indicate the connexion.</Paragraph> <Paragraph position="2"> Although Tesni~re:s distinction between the linear and structural order corresponds to some extent with the distinction between the linear precedence (LP) and the immediate dominance, there is a crucial difference in emphasis with respect to those modern syntactic theories, such as GPSG, that have distinct ID and LP components. Tesni~re excludes word order phenomena from his structural syntax and therefore does not formalise the LP component at all.</Paragraph> <Paragraph position="3"> Tesni~re's solution is adequate, considering that in many languages word order is considerably free. This kind of &quot;free&quot; word order means that the alternations in the word order do not necessarily change the meaning of the sentence, and therefore the structural description implies several linear sequences of the words. This does not mean that there are no restrictions in the linear word order but these restrictions do not emerge in the structural analysis.</Paragraph> <Paragraph position="4"> In fact, Tesni~re assumes that a restriction that is later formalised as an adjacency princi-</Paragraph> <Paragraph position="6"> ple characterizes the neutral word order when he says that there are no syntactic reasons for violating adjacency in any language, but the principle can be violated, as he says, for stylistic reasons or to save the metric structure in poetics. If we replace the stylistic reasons with the more broader notion which comprises the discourse functions, his analysis seems quite consistent with our view. Rather than seeing that there are syntactic restrictions concerning word order, one should think that some languages due to their rich morphology have more freedom in using word order to express different discourse functions. Thus, linearisation rules are not formal restrictions, but language-specific and functional. null There is no need for constituents. Tesni~re's theory has two mechanisms to refer to linearisation. First, there are static functional categories with dynamic potential to change the initial category. Thus, it is plausible to separately define the combinatorial and linearisation properties of each category. Second, the categories are hierarchical so that, for instance, a verb in a sentence governs a noun, an adverb or an adjective. The lexical properties, inherent to each lexical element, determine what the governing elements are and what elements are governed.</Paragraph> <Paragraph position="7"> There are no simple rules or principles for linearisation. Consider, for example, the treatment of adjectives in English. The basic rule is that attributive adjectives precede their heads.</Paragraph> <Paragraph position="8"> However, there are notable exceptions, including the postmodified adjectives s, which follow their heads, and some lexical exceptions 9, which usually or always are postmodifying.</Paragraph> </Section> </Section> <Section position="7" start_page="3" end_page="3" type="metho"> <SectionTitle> 5 Historical formulations </SectionTitle> <Paragraph position="0"> In this section, the early formalisations of the dependency grammar and their relation to Tesni~re's theory are discussed. The dependency notion was a target of extensive formal studies already in the first half of the 1960's ldeg. 8Example: &quot;It is a phenomenon consistent with ...&quot; 9Example: &quot;president el.ect&quot; ldegA considerable number of the earlier studies were listed by Marcus (1967, p. 263), who also claimed that &quot;Tesni~re was one ol the first who used (dependency) graphs in syntax. His ideas n,ere repeated, developed and precised by Y. Lecer\] ~ P. l&quot;hm (1960), L. Hirschbe~ and L Lynch, particularly by studying syntactic projectivity and linguistic subtrees.&quot;</Paragraph> <Section position="1" start_page="3" end_page="3" type="sub_section"> <SectionTitle> 5.1 Gaifman's formulation </SectionTitle> <Paragraph position="0"> The classical studies of the formal properties of dependency grammar are Gaifman (1965) and Hays (1964) 11, which demonstrate that dependency grammar of the given type is weakly equivalent to the class of context-free phrase structure grammars. The formalisation of dependency grammars is given in Gaifman (1965, p. 305): For each category X, there will be a finite number of rules of the type X (YI, Y2&quot;'&quot; Y! * Fi+t-. &quot;Yn), which means that YI&quot;&quot; &quot;Y~ can depend on X in this given order, where X is to occupy the position of ,.</Paragraph> <Paragraph position="1"> Hays, referring to Gaifman's formulation above, too strongly claims that &quot;\[d\]ependency theory is weakly omnipotent to IC theory. The proof is due to Gaifman, and is too lengthy to present here. The consequence of Gaifman's theorem is that the class of sets of utterances \[...\] is Chomsky's class of context-free languages.&quot; This claim was later taken as granted to apply to any dependency grammar, and the first, often cited, attestation of this apparently false claim appeared in Robinson (1970). She presented four axioms of the theory and claimed they were advocated by Tesni~re and formalised by Hays and Gaifman.</Paragraph> <Paragraph position="2"> Thus, the over-all result of the Gaifman-Hays proof was that there is a weak equivalence of dependency theory and context-free phrase-structure grammars. This weak equivalence means only that both grammars charac-S lTesni~re is not mentioned in these papers. Gaifman's paper describes the results &quot;... obtained while the author was a consultant for the RAND Corporation in the summer of 1960. ~ Whereas phrase-structure systems were defined by referring to Chomsky's Syntactic Structures, the corresponding definition for the dependency systems reads as follows: &quot;By dependency system we mean a system, containing a finite number of rules, by which dependency analysis for certain language is done, as described in certain RAND publications (Hays: February 1960; Hays and Ziehe, April 1960). ~ Speaking of the dependency theory, Hays (1960) refers to the Soviet work on machine translation using the dependency theory of Kulagina et al. In Hays (1964), the only linguistic reference is to the 1961 edition of Hjelmslev's Prolegomena: &quot;Some of Hjelmslev's empirical principles are closely related to the insight behind dependency theory, but empirical dependency in his sense cannot be identified with abstract dependency in the sense of the present paper, since he explicitly differentiates dependencies 1ram other kinds at relations, whereas the present theorlt intends to be complete, i.e. to account lor all relations among units</Paragraph> <Paragraph position="4"> terize the same sets of strings. Unfortunately, this formulation had little to do with TesniSre's dependency theory, but as this result met the requirements of a characterisation theory, interest in the formal properties of dependency grammar diminished considerably.</Paragraph> </Section> <Section position="2" start_page="3" end_page="3" type="sub_section"> <SectionTitle> 5.2 Linguistic hypotheses </SectionTitle> <Paragraph position="0"> Tesni~re's Hypothesis, as Marcus (1967) calls it, assumes that each element has exactly one head.</Paragraph> <Paragraph position="1"> Marcus also formulates a stronger hypothesis, the Projectivity hypothesis, which connects the linear order of the elements of a sentence to the structural order of the sentence. The hypothesis is applied in the following formulation: let X &quot;- ala 2 ...ai ...an be a sentence, where ai and aj are terms in the sentence. If the term ai is subordinate to the term aj, and there is an index k which holds min(i,j) < k < max(i,j), then the term ak is subordinate to the term aj.</Paragraph> <Paragraph position="2"> This is the formal definition of projectivity, also known as adjacency or planarity. The intuitive content of adjacency is that modifiers are placed adjacent to their heads. The intuitive content behind this comes from Behaghel's First Law 12 (Siewierska, 1988, p. 143).</Paragraph> <Paragraph position="3"> The adjacency principle is applicable only if the linear order of strings is concerned. However, the target of Tesni~re's syntax is structural description and, in fact, Tesnibre discusses linear order, a property attributable to strings, only to exclude linearisation from his conception of syntax. This means that a formalisation which characterises sets of strings can not even be a partial formalisation of Tesni~re's theory because his syntax is not concerned with strings, but structures. Recently, Neuhaus and BrSker (1997) have studied some formal properties of dependency grammar, observing that Gaifman's conception is not compatible either with Tesnibre's original formulation or with the &quot;current&quot; variants of DG.</Paragraph> <Paragraph position="4"> There are several equivalent formalisations for this intuition. In effect they say that in a syntactic tree, where words are printed in linear order, the arcs between the words must not cross. For example, in our work, as the arc between the node &quot;what&quot; and the node &quot;do&quot; in l~ &quot;The most important law is that what belongs to* gether mentally (semantically) is placed close together</Paragraph> </Section> <Section position="3" start_page="3" end_page="3" type="sub_section"> <SectionTitle> 5.3 Formal properties of a Tesni~re-type DG </SectionTitle> <Paragraph position="0"> Our current work argues for a dependency grammar that is conformant with the original formulation in Tesni~re (1959) and contains the following axioms: * The primitive element of syntactic description is a nucleus.</Paragraph> <Paragraph position="1"> * Syntactic structure consists of connexions between nuclei.</Paragraph> <Paragraph position="2"> * Connexion (Tesnikre, 1959, Ch. 1:11) is a binary functional relation between a superior term (regent.) and inferior term (dependent). null * Each nucleus is a node in the syntactic tree and it has exactly one regent (Tesnibre, 1959, Ch. 3:1).</Paragraph> <Paragraph position="3"> * A regent, which has zero or more dependents, represents the whole subtree. * The uppermost regent is the central node of the sentence.</Paragraph> <Paragraph position="4"> These axioms define a structure graph which is acyclic and directed, i.e. the result is a tree. These strong empirical claims restrict the theory. For example, multiple dependencies and all kinds of cyclic dependencies, including mutual dependency, are excluded. In addition, there can be no isolated nodes.</Paragraph> <Paragraph position="5"> However, it is not required that the structure be projective, a property usually required in many formalised dependency theories that do</Paragraph> <Paragraph position="7"> not take into account the empirical fact that non-projective constructions occur in natural languages.</Paragraph> </Section> </Section> <Section position="8" start_page="3" end_page="3" type="metho"> <SectionTitle> 6 The Functional Dependency </SectionTitle> <Paragraph position="0"/> <Section position="1" start_page="3" end_page="3" type="sub_section"> <SectionTitle> 6.1 On the formalism and output </SectionTitle> <Paragraph position="0"> It has been necessary to develop an expressive formalism to represent the linguistic rules that build up the dependency structure. The descriptive formalism developed by Tapanainen can be used to write effective recognition grammars and has been used to write a comprehensive parsing grammar of English.</Paragraph> <Paragraph position="1"> When doing fully automatic parsing it is necessary to address word-order phenomena.</Paragraph> <Paragraph position="2"> Therefore, it is necessary that the grammar forrealism be capable of referring simultaneously both to syntactic order and linear order. Obviously, this feature is an extension of Tesni~re's theory, which does not formalise linearisation.</Paragraph> <Paragraph position="3"> Our solution, to preserve the linear order while presenting the structural order requires that functional information is no longer coded to the canonical order of the dependents Is.</Paragraph> <Paragraph position="4"> In the FDG output, the functional information is represented explicitly using arcs with labels of syntactic functions. Currently, some 30 syntactic functions are applied.</Paragraph> <Paragraph position="5"> To obtain a closer correspondence with the semantic structure, the nucleus format corresponding to Tesni~re's stemmas is applied. It lSCompare this solution with the Prague approach, which uses horizontal ordering as a formal device to express the topic-focus articulation at their tectogrammatical level. The mapping from the tectogrammatical level to the linear order requires separate rules, called 8hallow rules (Petkevi~, 1987). Before such a description exists, one can not make predictions concerning the complexity of the grammar.</Paragraph> </Section> </Section> <Section position="9" start_page="3" end_page="3" type="metho"> <SectionTitle> WAS RUNNING DOG IN HOUSE THE THE </SectionTitle> <Paragraph position="0"> is useful for many practical purposes. Consider, for example, collecting arguments for a given verb &quot;RUN&quot;. Having the analysis such as those illustrated in Figure 2, it is easy to excerpt all sentences where the governing node is verbal having a main element that has &quot;run&quot; as the base form, e.g. ran, &quot;was running&quot; (Figure 2), &quot;did run&quot; {Figure 3). The contraction form &quot;won't run&quot; obtains the same analysis (the same tree although the word nuclei can contain extra information which makes the distinction) as a contraction of the words &quot;will not run&quot;.</Paragraph> <Paragraph position="1"> As the example shows, orthographic words were segmented whenever required by the syntactic analysis.</Paragraph> <Paragraph position="2"> This solution did not exist prior the FDG and generally is not possible in a monostratal dependency description, which takes the (orthographic) words as primitives. The problem is that the non-contiguous elements in a verbchain are assigned into a single node while the subject in between belongs to its own node.</Paragraph> <Paragraph position="3"> For historical reasons, the representation contains a lexico-functional level closely similar to the syntactic analysis of the earlier English Constraint Grammar (ENGCG) (Karlsson et al. , 1995) parsing system. The current FDG formalism overcomes several shortcomings 14 of the earlier approaches: (1) the FDG does not rely on the detection of clause boundaries, (2) parsing is no longer sequential, (3) ambiguity is represented at the clause level rather than word level, (4) due to explicit representation of dependency structure, there is no need to refer to phrase-like units. Because the FDG rule formalism is more expressive, linguistic generalisation can be formalised in a more transparent way, which makes the rules more readable.</Paragraph> </Section> <Section position="10" start_page="3" end_page="8" type="metho"> <SectionTitle> 7 Descriptive solutions </SectionTitle> <Paragraph position="0"/> <Section position="1" start_page="3" end_page="8" type="sub_section"> <SectionTitle> 7.1 Coordination </SectionTitle> <Paragraph position="0"> We now tackle the problem of how coordination can be represented in the framework of dependency model. For example, Hudson (1991) has argued that coordination is a phenomenon that requires resorting to a phrase-structure model.</Paragraph> <Paragraph position="1"> Coordination should not be seen as a directed functional relation, but instead as a special connexion between two functionally equal elements.</Paragraph> <Paragraph position="2"> The coordination connexions are called junctions in Tesnibre (1959, Chs. 134-150). Tesni~re considered junctions primarily as a mechanism to pack multiple sentences economically into one. Unfortunately, his solution, which represents all coordinative connexions in stemmas, is not adequate, because due to cyclic arcs the result is no longer a tree.</Paragraph> <Paragraph position="3"> Our solution is to pay due respect to the formal properties of the dependency model, which requires that each element should have one and only one head} 5 This means that coordinated elements are chained (Figure 4) using a specific arc for coordination (labeled as cc). The coordinators are mostly redundant markers (Tesnibre, 1959, Ch. 39:5) 16, especially, they do not have ~4Listed in Voutilainen (1994).</Paragraph> <Paragraph position="4"> lSThe treatment of coordination and gapping in Kahane (1997) resembles ours in simple cases. However, this model maintains projectivity, and consequently, both multiple heads and extended nuclei, which are essentially phrase-level units, are used in complex cases, making the model broadly similar to Hudson (1991).</Paragraph> <Paragraph position="5"> ~The redundancy is shown in the existence of asyndetic coordination. As syntactic markers, coordinators are not completely void of semantic content, which is</Paragraph> </Section> </Section> <Section position="11" start_page="8" end_page="8" type="metho"> <SectionTitle> LOVE JOHN MARY BILL AND AND JOAN </SectionTitle> <Paragraph position="0"> as they do in many word-based forms of dependency theory (e.g, Kunze (1975) and Mel'/:uk (1987)).</Paragraph> <Paragraph position="1"> Unlike the other arcs in the tree, the arc marking coordination does not imply a dependency relation but rather a functional equivalence. If we assume that the coordinated elements have exactly the same syntactic functions, the information available is similar to that provided in Tesnibre:s representation. If needed, we can simply print all the possible combinations of the coordinated elements: &quot;Bill loves Mary&quot;, &quot;John loves Mary ~, etc.</Paragraph> <Section position="1" start_page="8" end_page="8" type="sub_section"> <SectionTitle> 7.2 Gapping </SectionTitle> <Paragraph position="0"> It is claimed that gapping is even a more serious problem for dependency theories, a phenomenon which requires the presence of non-terminal nodes. The treatment of gapping, where the main verb of a clause is missing, follows from the treatment of simple coordination.</Paragraph> <Paragraph position="1"> In simple coordination, the coordinator has an auxiliary role without any specific function in the syntactic tree. In gapping, only the coordinator is present while the verb is missing. One can think that as the coordinator represents all missing elements in the clause, it inherits all properties of the missing (verbal) elements (Figure 6). This solution is also computationally effective because we do not need to postulate empty nodes in the actual parsing system.</Paragraph> <Paragraph position="2"> From a descriptive point of view there is no problem if we think that the coordinator obtains syntactic properties from the nucleus that demonstrated !)5' the existence of contrasting set of coordinators; 'and', 'or', 'but' etc.</Paragraph> <Paragraph position="3"> it is connected to. Thus, in a sentence with verbal ellipsis, e.g. in the sentence &quot;Jack painted the kitchen white and the living room blue&quot;, the coordinator obtains the subcategorisation properties of a verb. A corresponding graph is seen in Figure 6.</Paragraph> <Paragraph position="4"> Due to 'flatness' of dependency model, there is no problem to describe gapping where a sub-ject rather than complements are involved, as the Figure 5 shows. Note that gapping provides clear evidence that the syntactic element is a nucleus rather than a word. For example, in the sentence &quot;Jack has been lazy and Jill angry', the elliptic element is the verbal nucleus has been.</Paragraph> </Section> </Section> class="xml-element"></Paper>