File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/96/w96-0310_metho.xml
Size: 29,261 bytes
Last Modified: 2025-10-06 14:14:24
<?xml version="1.0" standalone="yes"?> <Paper uid="W96-0310"> <Title>Lexical Rules for Deverbal Adjectives</Title> <Section position="3" start_page="90" end_page="91" type="metho"> <SectionTitle> 2. The Ontology-Based Approach to Adjectival Semantics and Lexicology </SectionTitle> <Paragraph position="0"> In this section, we briefly review the basis of our approach to adjectival meaning and illustrate it on three examples of adjectival lexicon entries, i.e., one each for the three major classes of adjectives. null</Paragraph> <Section position="1" start_page="90" end_page="90" type="sub_section"> <SectionTitle> 2.1 The Ontological Approach </SectionTitle> <Paragraph position="0"> Our work on adjectives forms a microtheory used by the MikroKosmos semantic analyzer. The architecture of MikroKosmos is described in Onyshkevych and Nirenburg 1994 and Beale et al.</Paragraph> <Paragraph position="1"> 1995. The MikroKosmos project is a component of a knowledge-based machine translation system (see Nirenburg et al. 1992). The purpose and result of the MikroKosmos analysis process is a rendering of the source language text into an interlingua text. The interlingua language is called the &quot;text meaning representation&quot; (TMR) language, and the TMR of a text is its representation in this particular type of interlingua. TMRs are realized in a frame-based language, where frame names typically refer to instances of ontological concepts and slots are usually filled with values of properties of those concepts. An ontology is, thus, a necessary prerequisite for building a TMR language. null &quot;An ontology for NLP purposes is a body of knowledge about the world (or a domain) that a) is a repository of primitive symbols used in meaning representation; b) organizes these symbols in a tangled subsumption hierarchy; and c) further interconnects these symbols using a rich system of semantic and discourse-pragmatic relations defined among the concepts&quot; (Mahesh and Nirenburg 1995: 1). The function of the ontology is to supply &quot;world knowledge to lexical, syntactic, and semantic processes&quot; (ibid).</Paragraph> <Paragraph position="2"> The lexicon in MikroKosmos &quot;mediates between the TMR and ontology&quot; (Onyshkevych and Nirenburg 1994: 2). Lexicon entries for most open-class lexical items represent word and phrase senses, which can be either directly mapped into ontological concepts or derived by locally (that is, in the lexicon entry itself) modifying constraints on property values of concepts used to specify the meaning of the given lexical item. In the following section, we briefly illustrate the structure of those parts of the lexicon entry in MikroKosmos which bear on the description of the three types of adjectival meaning, scalar, denominal, and deverbal.</Paragraph> </Section> <Section position="2" start_page="90" end_page="91" type="sub_section"> <SectionTitle> 2.2 The Ontological Approach to the Meaning of Adjective Types </SectionTitle> <Paragraph position="0"> Our microtheory associates the meaning of a typical truly scalar adjective with a region on a scale which is defined as the range of an ontological property. The contribution that the adjective makes to the construction of a semantic dependency structure (TMR) typically consists of inserting its meaning (a property-value pair) as a slot in a frame representing the meaning of the noun which this adjective syntactically modifies.</Paragraph> <Paragraph position="1"> Thus, in big house, big will assign a high value as the filler of the property slot SIZE of the frame for the meaning of house (see also Raskin and Nirenburg 1996). (1) is a partial lexical entry for big, with just two of the 13 lexical zones represented: (1) (big (big-Adj ~CAT adj)</Paragraph> <Paragraph position="3"> SIZE is an ontological concept of the SCALAR-PHYSICAL-OBJECT-A'FrRIBUTE-PROPERTY type, with the term 'scalar' used here, as it is customarily, primarily in the sense of 'gradable.' Big is, of course, a typical gradable adjective, and as such, has a numerical scale associated with it the microtheory. Each numerical scale can be measured in actual measuring units, such as linear-size in feet, yards, or millimeters, or time in seconds. But often natural language expressions do not refer to absolute magnitudes but rather to abstract relative ones, as in the case of big. We assume a 0 to</Paragraph> <Paragraph position="5"> Big will, then, get something like a '> 0.75' value on the SIZE scale (1). These values are a crucial part of the lexical mapping (LEX-MAP) included in the semantics (SEM-STRUC) &quot;zone&quot; of a scalaradjective lexical entry. 2.</Paragraph> <Paragraph position="6"> Equally crucial is the syntactic-semantic mapping between the syntactic-structure (SYN-STRUC) and SEM-STRUC zones with the help of special variables. In the former, there are two subcategorization patterns, marked 1 and 2, listed in SYN-STRUC of (1). The former pattern corresponds to the attributive use of the adjective: the noun it modifies is assigned the variable $varl, and the adjective itself the variable $var0 in the modifier position. The latter pattern presents the noun, bound to $varl, in the subject position and the adjective in the predicative position. In the SEM-STRUC zone, instead of variables which are bound to syntactic elements, the meanings of the elements re-</Paragraph> </Section> </Section> <Section position="4" start_page="91" end_page="93" type="metho"> <SectionTitle> 2 Similar values are assigned to any scale based on a property concept. It should be noted, however, that nu- </SectionTitle> <Paragraph position="0"> merical values like these correspond to the feature of gradability, which extends beyond scalarity: all scalar adjectives are gradable but not all gradable adjectives are true scalars. The two features are typically confused in the literature. This is why we distinguish here between true scalars, i.e., those adjectives whose meanings are based on a scale that is a property concept in the ontology, and all the other adjectives which are gradable and may be also loosely referred to as &quot;scalars.&quot; Thus, many denominal and deverbal adjectives, in fact, most adjectives, can form a comparative or superlative degree form with the help of more __ or most .... respectively, e.g., more medical (3ii), more aeronautical (4), more employable (6), more abusive (9-10). All of these adjectives are, therefore, gradable. Now, none of them is anchored in a scale-type property concept, however, which makes them all, technically, non-scalars but, in more practical and customary terms, non-true scalars.</Paragraph> <Paragraph position="1"> ferred to by these variables (and marked by a caret, '^') are used. That is, ^$varl reads as &quot;the meaning of the element to which the variable $varl is bound.&quot; Among the constraints listed in the SEM-STRUC zone of an entry, are selectional restrictions (the noun must be a physical object) and relaxation information, which is used for treatment of unexpected (&quot;ill-formed&quot;) input during processing. null Thus, an entry like (1) should be read as follows: the second subcategorization pattern describes the N-Copula-Adj construction; * the LEX-MAP part of the SEM-STRUC zone defines the lexical semantics of the adjective by assigning it to the class of SIZE adjectives, that is, anchoring it in the ontological propertytype concept SEE and stating that it is applicable to physical objects and that its meaning is a high-value range on the SIZE scale/property.</Paragraph> <Paragraph position="2"> In the case of continuous scales, like SIZE, the acquisition of all the adjectives served by this scale is greatly facilitated and expedited after the first one of them gets a lexical entry: each new adjective needs only an appropriate range assigned to it, and the rest of the information in the semantic zone of the entry, as indeed in all the other zones as well, remains the same. We will recall this methodological circumstance shortly, in Section 3.</Paragraph> <Paragraph position="3"> The most general case of a denominal adjective-entry and its connection to that of the corresponding noun is demonstrated in (3):</Paragraph> <Paragraph position="5"> The formula of transition from the noun entry to that of the adjective is simple and transparent, and it remains constant for this type of adjective. In Raskin and Nirenburg (1995: 39-41), we discuss the desirability, feasibility, and usefulness of replacing the catch-all &quot;pertain-to&quot; dependency with a more specific and informative one, and while we demonstrate several of those, we argue for staying at this generic level of granularity in a significant number of cases.</Paragraph> <Paragraph position="6"> Denominal adjectives are relative, and, as such, are expected to be non-scalar and non-predicating. A true relative adjective cannot indeed be used predicatively and or comparatively, but, practically, it is hard to come up with an example which is guaranteed against that. After all, (4i-ii) are not re- null ally ill-formed, and it is hard to imagine a more truly-relative adjective than aeronautical 'related to aeronautics.' (4) (i) (ii) His approach to the problem was aeronautical.</Paragraph> <Paragraph position="7"> His approach to the problem was much more aeronautical than mine.</Paragraph> <Paragraph position="8"> Clearly, a productive semantic process, a shift, takes place here, probably along the lines of (5), and, therefore, a dynamic rule exists which creates adjective entries for these predicating, pseudoscalar, pseudo-qualitative senses of the seemingly perfectly relative adjectives. Their one telling difference from the truly qualitative, predicating, scalar adjectives is that the relative adjectives cannot make the qualitative shift in the attributive position.</Paragraph> <Paragraph position="9"> (5) Pertaining to \[noun meaning\] --->Pertaining more to \[noun meaning\] The difficulty of finding a relative adjective which would be absolutely resistant to a predicative shift of meaning is similar to the difficulties Katz (1972: 753) had, looking for a noun which could never be used with good.</Paragraph> <Paragraph position="10"> A large class of adjectives whose meanings are derived from those of verbs also straddle the scalar / non-scalar divide. The event-related scalars (actually, gradables--see fn. 2) do not really differ from true scalars in terms of their gradability (6); the event-related non-scalars can acquire grad- null ability at the cost of a meaning shift or marginal acceptability (7).</Paragraph> <Paragraph position="11"> (6) (7) (i) Jake is employable (ii) Jake is very employable (iii) Jake is more employable than Bob (iv) Jake is most employable of all (v) Bob is barely employable (i) ?Jake's initiative was abortive (ii) ?Jake's initiative was very abortive (iii) ?Jake's initiative was more abortive than Bob's (iv) ?Jake's initiative was most abortive of all (v) ?Bob's initiative was barely abortive Examples (8-10) illustrate a typical transition from the lexical entry of a verb to those of the deverbal adjectives. Both the verb entry and the adjective entries are much more complex than those for nouns and denominal entries.</Paragraph> <Paragraph position="12"> This sense of the verb, &quot;insult verbally,&quot; is anchored in the process concept COMMUNICATIVE-EVENT, modified by two evaluative attitudes, involving two unmapped (i.e., internal for the zone) variables, refseml and refsem2. Attitudel means that the beneficiary or. more appropriately, the victim, of the process, i.e., the abused person and/or the speaker evaluate(s) the theme of the abuse, i.e., the contents of what is said, as bad; attitude2 means that the agent of abuse-V1 thinks of the beneficiary pretty poorly.</Paragraph> <Paragraph position="13"> To derive the semantics zone of an adjectival entry from that of the corresponding verbal entry, one must first identify the case, or thematic role (such as agent, theme, beneficiary, etc.) filled by the noun modified by the adjective in question. What is abusive is either the event (E) itself, as in abusive speech or abusive behavior, or the agent (A) of the event, as in abusive man or abusive neighbor. AbusivelE (9) is then the eventive sense of the adjective formed from abuse-V1 (8), and abusive lA (10) is the agentive sense of the adjective in the same sense of abuse. There may, in fact, be at least 4 more senses, and entries, for the adjective, but we will come back to them shortly, in Section 4.1.</Paragraph> </Section> <Section position="5" start_page="93" end_page="96" type="metho"> <SectionTitle> 3. Lexical Rules </SectionTitle> <Paragraph position="0"> In this section, we introduce our notion of lexical rules (LRs), place their origin in the descriptive methodology and heuristics of large-scale lexical acquisition, and establish some usability parameters for them.</Paragraph> <Paragraph position="1"> The &quot;transition formulae&quot; from a noun lexical entry (3i) to that of a denominal adjective (3ii) or from a verb lexical entry (8) to those of deverbal adjectives (9-10) are examples of LRs in which we are interested here. Each LR takes a ready entry (El) and creates another entry (E2) out of it automatically:</Paragraph> <Paragraph position="3"> available heuristics for the lexical acquisition process (see Raskin andNirenburg 1995: 41-57).</Paragraph> <Paragraph position="4"> These include methods of reducing unnecessary polysemy, gauging the optimal grain size, and determining the meaning, not the least of which is selecting the ontological concept in which the entry is anchored. These methods are, of course, captured in an array of semi-automatic tools of ontological and lexical acquisition developed for the MikroKosmos project (see, for instance, Mahesh 1996 and Viegas and Raskin 1996).</Paragraph> <Paragraph position="5"> These tools stemmed from a pilot project, in which templates for types of lexical entries were de- null veloped truly manually and rapidly propagated on the basis of such obvious paradigmatic relations as synonymy and partial synonymy, antonymy, and hyper-/hyponymy, and much less obvious relations of partial overlap, such as exemplified above in connection with the lexical entry for big (1) and its easy adaptation to all the other SIZE adjectives by modifying the numerical value on the SIZE range (2), if necessary. The corpus of 6,000 English adjectives (an intersection of LDOCE and a corpus of 1987-99 Wall Street Journal articles) yielded some 250 S/ZE adjectives, all of them propagated from the same template that underlies (1). The least trivial relationship to establish involved constrains on the modified noun, thus recognizing an adjective like well-endowed as characterizing the size of a certain part of human male anatomy, a different part of human female anatomy, or the amount of money available to an institution.</Paragraph> <Paragraph position="6"> The discovery and use of such relationships and rapid propagation methods based on them lay the foundation for LRs. In fact, each such relationship and related method is a LR. Because we are engaged in massive lexical acquisition, we are obviously interested in LRs, which are:</Paragraph> <Paragraph position="8"> Perhaps the most easily discoverable LRs involve E l/E2 pairs which are morphologically related, especially if their share the roots. Some of these rules are transcategorial, i.e., E 1 and E2 belong to different lexical categories. Obviously, the denominal adjective LR and the deverbal adjective LR (12i-ii, respectively), exemplified in (3) and (8-10), are precisely such LRs.</Paragraph> <Paragraph position="10"> The additional attraction for using (12i-ii) in massive acquisition is the facilitation of the least trivial element of lexical semantic heuristics, namely, the discovery of the ontological concept on which the lexical entry should be based. While a relatively small class of true scalars is more or less easily associated with certain property concepts, there are many more process and object concepts, and it is much easier to relate the meaning of a verb or a noun to one of those. Besides, it is precisely the denominal and deverbal adjectives which are very hard to relate to a property concept directly, so the LRs come in quite handy.</Paragraph> <Paragraph position="11"> Several of the LRs discovered manually in the pilot project turned out to be &quot;large,&quot; in the sense of their massive productiveness (see, for instance, Viegas et al. 1996), such as the comparativedegree LR, applicable to every one of the 4,000-plus gradable adjectives. Given the paucity of relative adjectives:in English (because of the adjectival use of nouns), the denominal adjective LR (12i) produces around 300 entries in our English corpus. The deverbal adjective LR falls in between, with an output of almost 1,000 entries.</Paragraph> <Paragraph position="12"> We have foundlvirtually no LR to be exception-free, and that reduces, of course, the degree to which a large LR can be used fully automatically, thus raising the cost of their application.</Paragraph> </Section> <Section position="6" start_page="96" end_page="101" type="metho"> <SectionTitle> 4. Deverbal Adjectives: A Full(er) View </SectionTitle> <Paragraph position="0"> In this section, we offer a bird's eye view of large-scale acquisition of deverbal adjectives with the help of LRvA, both in the general, common case, with its non-trivial complications, deviations, and exceptions, and in the particular case of the single largest and seemingly most regular subclass of deverbal adjectives, namely, those ending in -able/-ible. The review is conducted from the perspective of highly desirable full automation vs. the reality and unavoidability of exceptions requiring manual treatment.</Paragraph> <Section position="1" start_page="97" end_page="99" type="sub_section"> <SectionTitle> 4.1 LRvA: Common Cases and Complications </SectionTitle> <Paragraph position="0"> The common case of deverbal adjective acquisition includes most deverbal adjectives, and it is discussed first. The most statistically and conceptually significant complications, deviations, and exceptions from the common case are briefly sketched next.</Paragraph> <Paragraph position="1"> The 950 or so deverbal adjectives in our English corpus have received their entries as a result of the application of the deverbal adjective LR (12ii). The LR exists in at least these 6 forms, corresponding to the event or its semantic cases/thematic roles: * event-itself (E), e.g., abusive in (9): abusive behavior; * agent-of-event (A), e.g., abusive in (10): abusive husband; * beneficiary-of-event (B), e.g.,free in free bird; * theme-of-event (T), e.g., automatic in automatic elevator; * instrument-of-event (I), e.g., poisonous in poisonous food; * location-of-event (L), e.g., international in international company. As exemplified in (9-10), each case-related sub-LR operates in a well-defined way, placing ^$varl in the SEM-STRUC zone in the value slot of the appropriate case of the event frame. Similarly, LR E, the event-itself sub-LR places ^$varl in the event position. Obviously, then: (13) (i) (ii)</Paragraph> <Paragraph position="3"> Abuse has two more senses, abuse-V2 &quot;violate a law or a privilege&quot; and abuse-V3 &quot;assault physically.&quot; Abusive can inherit all of those senses, both in the eventive and agentive varieties, thus adding 4 more senses to its superentry. Abusive is indeed a pretty typical and &quot;easy&quot; example of LRvA.</Paragraph> <Paragraph position="4"> Even so, there are several aspects of its application which require human judgment.</Paragraph> <Paragraph position="5"> The first obvious complication is that not all of its semantic cases of a verb are usable for the LR-in fact, only one of the three semantic cases of abuse, namely, agent is but beneficiary and theme are not. This throws the first monkey's wrench into making the LR fully automatic. Here, as in many similar cases in side and outside of this LR, we face a dilemma. We can either try and discover a rule which marks some semantic cases as participating in the LR and others as not participating in it. Such a principle is not immediately clear and may: there are other factors which push this latter decision forward.</Paragraph> <Paragraph position="6"> It is interesting to note that the sense distinctions in abusive seem to be less significant than in abuse. Typical of many cases of polysemy, this may be an indication that the distinctions in the dictionaries may be disposable (see the Spanish verb dejar reduced from 52 dictionary meanings to 7 in Nirenburg et al. 1995). Perhaps less so in the case of abuse than in the case of abandon directly below, it is reasonable to suggest that, in many cases of dictionary polysemy, it is the single sense of the verb modified by different types of nouns that can fill its case slots. Such cases are candidates for reducing polysemy as part of the heuristics procedures mentioned above (see also Raskin and Nirenburg 1995: 41-45).</Paragraph> <Paragraph position="7"> A human must decide whether the adjective inherits all the senses of a polysemous verb because it is not always the case that it does--unlike abusive. Thus, impregnable has only the figurative sense of impregnate, Out of the 5 senses of abandon: * abandon-V1: &quot;stop trying&quot; (aspectual); * abandon-V2: &quot;leave somebody or something at location&quot;; * abandon, V3: &quot;withdraw support&quot;; * abandon-V4: &quot;give up claim&quot;; * abandon-V5: &quot;yield to emotion&quot;-abandoned has 6 senses corresponding to just three of the senses (some examples below may have other meanings along with the one exemplified):</Paragraph> <Paragraph position="9"> * abandoned-4B, e.g., abandoned child; deg abandoned-4T, e.g., abandonedproperty-and 1, possibly no longer existing sense corresponding to another sense of the verb: * abandoned-5T, e.g., John was abandoned to his grief (theme is used here in the experiencer mode).</Paragraph> <Paragraph position="10"> Abandoned can, of course, be a participle as well, but we have discovered no reason whatsoever to treat participles differently from deverbal adjectives. Surely, participles may have syntactic dependents and be used postpositively in English--more so than adjectives, but this is easily accounted for in the appropriate patterns (not demonstrated) in the adjective SYN-STRUC zone. An adjective, just as a participle, is a device to raise a proposition into a higher one. In fact, we suggest that all clearly participial adjectives must be treated exactly as we are proposing here, that is, with the help of LRvA: the only alternative is unfolding them into full-fledged clauses, and this is something MT should, and usually can, avoid dealing with--on the basis of the practical effability principle (Nirenburg and Raskin 1995: 45-47). That the difference between participles and adjectives may be a grammatical artifact--or that, at least, it is of no significance for computational semantics--can be indirectly but rather convincingly supported by the case of Russian, where the grammarians have been trying to prop up the distinction by insisting on the doubling of the spelling of n in the suffix of the past participle but not of the related adjective, as in ranennyy v ruku boetz &quot;fighter wounded in the arm&quot; and ranenyy boetz &quot;wounded fighter&quot;: the speakers reject this distinction both in pronunciation, where no consonant doubling is detectable, and in spelling, where the two forms are universally confused.</Paragraph> <Paragraph position="11"> Cases like impregnable above raise another significant issue. The treatment of negation in deverbal adjectives is not very difficult: the negative quantifier can, in most cases, be simply added to the non-negative adjective, and this applies to all of the negative prefices, such as un-, in- (ira-, etc.), dis-. But impregnable, like quite a few other negative deverbal adjectives, does not have a non-negative counterpart, which looks like an accidental gap. In fact, the non-existence of a deverbal adjective, positive or negative, from the physical meaning of impregnate &quot;make pregnant&quot; is also an accidental gap. And, of course, a further complication is that im- in the verb is not negative while in the adjective it is: this is, apparently, the reason for the non-existence of the positive adjective, but this reason is itself an accidental gap.</Paragraph> <Paragraph position="12"> Gaps of this kind abound throughout the corpus. Why, for instance, there is perishable but not rottable. As we will see below, the former is a somewhat rare bird among the deverbal adjectives ending in -ble, but, on the face of it, there is no difference between food products rotting and food products perishing--in fact, the former sounds more natural than the latter. Nevertheless, rottable sounds less acceptable than perishable.</Paragraph> <Paragraph position="13"> Many deverbal adjectives in the corpus derive their lexical entries from verbs which are not morphologically related to them (14). The LR, however, works with the same ease as in typical, morphologically derivative cases. An even stronger case of suppletivism involves processes which do not have single-word verbs denoting them. Nevertheless, it often makes sense--and this is a human judgment--to construct such verbs out of the appropriate ontological concept with necessary constraints, because this is still the easiest way to construct lexical entries for some adjectives (15). Obviously, this kind of supersuppletivism would be impossible and arbitrary if implemented outside a justified situated ontology.</Paragraph> <Paragraph position="14"> Related to suppletivism, we discover quite a few deverbal adjectives &quot;shared&quot; by several verbs on a synonymous sense. This, abandoned-2 is a deverbal of both abandon-V2 (see Section 4.1.3 above) and leave in the same sense of &quot;leave somebody or something at location&quot; because the latter's &quot;own&quot; deverbal, left, is not comfortable in adjectival use, especially attributively:</Paragraph> </Section> <Section position="2" start_page="99" end_page="101" type="sub_section"> <SectionTitle> 4.2 Adjectives Ending in .ble </SectionTitle> <Paragraph position="0"> A large subclass of deverbal adjectives are adjectives that end in able/ible (cf. Kjellmer 1986 and also Hall 1877, Jespersen 1942, Marchand 1960, Abraham 1970, Meus 1975, for a discussion of such adjectives in English, even though much of the discussion sheds little light on the semantic and lexicographic issues in hand). Just about all of them mean 'something than can be \[verb\]-ed': thus, readable means 'something that can be read.' In other words, a typical -able entry is derived from the lexical entry of the appropriate verb, with the positive potential attitude added, and in either the beneficiary or theme role, depending on the animateness/inanimateness of $varl, respectively (17i-ii).</Paragraph> <Paragraph position="1"> The modality in the entries above corresponds, of course, to 'can' in the informal formula above. On the surface of it, these adjectives seem to be the best candidates for a fully automatic LR, especially if the beneficiary/theme distinction can be taken care of on the basis of the animateness or inanimateness of the modified--and this can be done. The problem is, however, that the beneficiary/theme combination of semantic cases for these deverbal adjectives, while accounting for a huge majority of such adjectives in the corpus (close to 270 out of the 300 or so), is only one of the 17 possible semantic-case combinations: (ii) A E culpable < do-bad (iii) E disreputable <feel-shame (iv) T E comprehensible < understand-1 (v) can+A perishable < perish-2 (vi) can+A can+E audible < hear (vii) can+B acceptable-2 < accept-1 (viii) can+B can+T definable < define-l-3 (ix) can+B can+T can+A adjustable < adjust-l-2 (x) can+L livable < live-1 (xi) can+T actionable < sue (xii) not+A incapable-2 < be-able (xiii) not+can+A inviable < live-1 (xiv) not+can+B unacceptable-2 < accept-1 (xv) not+can+B not+can+T undefinable < define-l-3 (xvi) not+can+L uninhabitable < live-1 (xvii) not+can+T impassable < pass-l-4 Stripped of 'can' and 'not,' there are still 9 different types, and because 'can' and 'not' do not &quot;apply&quot; universally, their presence or absence are not trivial. Except for types (viii) and (xv), most of the types have few examples in the corpus, and perishable is alone in its type (v). What it means, again, is that human judgment is necessary in deciding which, if any, case form of LRvA to apply to each verb.</Paragraph> </Section> </Section> class="xml-element"></Paper>