ADJECTIVES IN DICTIONARIES: HOW WELL DO MODERN DICTIONARIES OF ENGLISH DEAL WITH POLYSEMOUS ADJECTIVES?

by

Jonathan Stammers

A dissertation submitted to the School of Humanities of the University of Birmingham for the degree of MA in Language and Lexicography

Centre for English Language Studies The University of Birmingham 2004

Abstract

The polysemy displayed by adjectives is of a strongly context-dependent type, which many might not recognise as true polysemy at all. A great deal of the literature concerning polysemy has little bearing upon adjectives. Adjectives are also a fairly varied word-class, thus posing a range of challenges to the lexicographer. This dissertation gives a detailed evaluation of six recently published monolingual English dictionaries, and in particular the way they deal with polysemous adjectives. A random sample of such adjectives is chosen from one of the dictionaries. Sample concordances from the Bank of English corpus for each adjective are analysed before their respective dictionary entries are evaluated.

2

Contents 1. Introduction...................................................................................................................5 2.Theoretical and lexicographical issues: How are polysemous adjectives best analysed, defined and presented?......................................................................................................7 2.1 How are they best analysed?....................................................................................7 2.1.1 Criteria for sense division.................................................................................7 2.1.2 Adjectives and polysemy .................................................................................9 2.2 How are they best defined?....................................................................................10 2.3 How are they best presented?................................................................................12 2.3.1 The role of example sentences........................................................................12 2.3.2 The structure of the dictionary entry..............................................................13 2.3.3 A word about the users of dictionaries...........................................................14 3. Methodology................................................................................................................15 3.1 The dictionaries chosen.........................................................................................15 3.2 A random sample of polysemous adjectives.........................................................17 3.3 Analysis of corpus data..........................................................................................17 3.4 Evaluation of dictionary entries.............................................................................18 4. Results: An evaluation of dictionary entries for polysemous adjectives.....................19 4.1.1 Dictionary evaluation: canonical....................................................................19 4.1.3 Summary of sense distinctions: canonical......................................................20 4.1.3 Results of corpus data: canonical...................................................................21 4.2.1 Dictionary evaluation: deep............................................................................22 4.2.2 Summary of sense distinctions: deep..............................................................24 4.2.3 Results of corpus data: deep...........................................................................24 4.3.1 Dictionary evaluation: floating.......................................................................26 4.3.2 Summary of sense distinctions: floating.........................................................27 4.3.3 Results of corpus data: floating......................................................................28 4.4.1 Dictionary evaluation: idle.............................................................................29 4.4.2 Summary of sense distinctions: idle...............................................................30 4.4.3 Results of corpus data: idle.............................................................................30 4.5.1 Dictionary evaluation: marginal.....................................................................32 4.5.2 Summary of sense distinctions: marginal.......................................................32 4.5.3 Results of corpus data: marginal.....................................................................32 4.6.1 Dictionary evaluation: particular....................................................................34 4.6.2 Summary of sense distinctions: particular......................................................35 4.6.3 Results of corpus data: particular...................................................................35 4.7.1 Dictionary evaluation: remote........................................................................36 4.7.2 Summary of sense distinctions: remote..........................................................37 4.7.3 Results of corpus data: remote........................................................................37 4.8.1 Dictionary evaluation: stable..........................................................................39 4.8.2 Summary of sense distinctions: stable............................................................39 4.8.3 Results of corpus data: stable.........................................................................39 4.9.1 Dictionary evaluation: unbalanced.................................................................41 4.9.2 Summary of sense distinctions: unbalanced...................................................41 4.9.3 Results of corpus data: unbalanced.................................................................41 5. Conclusion...................................................................................................................43 References........................................................................................................................46 Appendix: Sample concordance lines: marginal.............................................................48 3

1

Introduction

There has been tremendous change for dictionary makers in recent times. With considerable developments within the discipline of linguistics, and the rapid development of Information Technology, computational lexicography, electronic corpora and the Internet, lexicographers are hard pressed to keep pace with the many innovations whilst continuing their demanding and highly detailed work at a sufficient pace to keep up with competitors, and working practices may have to be changed frequently with the huge amount of linguistic data suddenly available to them. There is an immense amount of information they could include, but for the great limitations of time, money, and of space in (typically) a single volume. In the midst of this change the aim remains broadly the same: to present the vocabulary of the language clearly and comprehensively enough to enable their intended readers to grasp its meaning. So how well are they doing? It is easy for many to quibble and make unjustified criticisms of dictionaries without a due appreciation of these pressures. Linguists too can have unreasonable expectations of consistency and precision, and in their preoccupation with theoretical issues can overlook practical ones. Sue Atkins puts it well when she speaks of a dichotomy between, on the one hand, “theorists to whom the abandonment of rigor means the abandonment of intellectual standards and hence the debasing of the work”, and on the other, “practitioners to whom a dictionary is something people open in an attempt to find out what a word means and how to use it, and for whom ‘rigor’ often collocates with ‘mortis’” (Atkins 1993: 25). Wishing to avoid these pitfalls, a detailed evaluation is here attempted of some of the leading British monolingual dictionaries of English. The analysis is restricted to a specific type of word presented in dictionaries: adjectives with multiple senses. Henri Béjoint has noted that “the comparison of how a certain number of dictionaries distinguish multiple meanings is potentially interesting” (Béjoint 2000: 228), and this will be a primary focus, adjectives being a particularly difficult class to divide into senses. Adjectives have other particular problems. They are strongly dependent on what it is they modify, and other contextual factors. Two particular problems pointed out by Philip Gove are: (1) that adjectives do not correspond to any ‘visualizable objects’ in the real world, being states or qualities that are difficult to conceive of ‘as distinct from specific things having that state or quality’; and (2) that they do not occur in the typical hierarchies of sense, so resist the traditional analytical style of definition (Gove 1968: 5). For a third problem pointed out by Gove, see below (4.1.1). After considering the theoretical issues involved, and their relation to lexicographical practice (Chapter 2), an analysis is undertaken of a random sample of 4

polysemous adjectives: firstly analysing citations for them drawn from the 450-million word Bank of English corpus, and then their treatment in six recently published English dictionaries. The dictionaries, as well as the procedures involved in the investigation, are described in more detail in Chapter 3, and the results given in Chapter 4, and summarised in Chapter 5, with some conclusions and more general observations.

5

2

Theoretical and lexicographical issues: How are polysemous adjectives best analysed, defined and presented?

In studying polysemous adjectives in particular, this paper focuses on a narrow division of the lexicon which is only rarely dealt with in linguistic or lexicographical literature. Adjectives are rarely singled out for specialist treatment except in systematic grammatical descriptions. Polysemy is discussed very widely, but the relevance of much of this discussion to adjectives is often far from clear, especially in the more theoretical and abstract writings on semantics. Before going on to analyse some specific adjectives and their treatments in dictionaries, it is right to here focus attention on what has been written concerning the theoretical issues involved with polysemy and adjectives, as well as some more practical issues concerning the process of making a dictionary entry. The best ways of treating such words in a dictionary are considered, along with the relevant theory, in the various key stages of preparation: analysis, defining, and overall presentation and clarification.

2.1 How are they best analysed? Once suitable and sufficient citations have been collected for a certain keyword, the first major task for the lexicographer is analysis. The usages of the keyword represented in the relevant data must somehow be put into groups that can be defined, and otherwise described, together. In his Manual of Lexicography, Zgusta writes: “…though it is an error if the lexicographer tries to make the single senses more sharply distinguished from each other than is indicated by the data, the establishment of the single senses and their organization are, on the other hand, among the lexicographer’s cardinal tasks.” (Zgusta 1971: 64)

2.1.1 Criteria for sense division But on what basis should these ‘single senses’ be ‘established’? Moon (1987b: 89-101) gives an account of many relevant criteria, based on the practices developed by the Cobuild project. She discusses syntax, collocation, derivative forms, etymology, phonology and intonation, the influence of real-world knowledge and uniqueness of referent, lexical relations, connotation and allusion, and pragmatics (cf. Svensén 1993: 112-114). Amongst these, the formal criteria, particularly syntax and collocation, are

6

found to be the clearest ones for distinction. In terms of adjectives and this investigation it is also worth considering lexical relations in addition to these two. Lexical relations, in particular the paradigmatic variety, include synonymy, hyponymy, meronymy, and antonymy (see, e.g. Cruse 2000: 150-76; Cruse 1986), and these can each on occasion be good criteria for distinguishing senses (Svensén 1993: 114; Cruse 2000: 107). If two readings of a word have different lexical relations, this is an indication that they are separate. This is of course fine if the networks of lexical relations have already been worked out, but this is unlikely to be the case at the analysis stage of dictionary production. Penelope Stock gives a critique of a working method described by John Ayto where “firstly the lexicographer should consider the superordinates of each of the meanings of the lexical item” (Stock 1984:131), and says: “…it seems to me that as a theoretical model it is insufficiently detached. For in order to obtain the appropriate superordinates for related senses of a word the lexicographer must, first, have already distinguished the senses in his or her mind, and second, have decided at what level of superordination the genus word will be chosen. Further, although the model seems to work very satisfactorily with respect to concrete nouns referring to fairly common objects in the real world, it is not at all clear that it would be so satisfactory with words which are more abstract… or with words of other word classes than noun or verb” (Stock 1984: 132)

Indeed, adjectives do not generally have anything like a superordinate, and they are unusual in that they appear to be arranged in the mental lexicon chiefly on the basis of antonymy, with synonymy playing an important secondary role (see Gross et al. 1989: 92, 95; Miller 1998: 48-52). An antonym relation appears to be one between words rather than concepts (Gross et al. 1989; cf. Murphy & Andrew 1993), so that while an adjective like strong can be grouped together with a whole host of synonyms (sturdy, powerful, tough etc.), it has a particular relation with weak as its antonym that it does not share with synonyms of weak such as puny or frail1. But antonymy could be used to distinguish senses on occasion: light with antonym dark is separate from light with antonym heavy, for example. Stock (1984: 132ff.) goes on to propose another method involving the criteria of syntax and collocation, and again based upon working practices developed during the Cobuild project. First one looks at the syntactic structure in which the keyword appears. What is the wordclass? Is the verb transitive or intransitive? Is the noun count or uncount (cf. Svensén 1993: 112-4). Next, one may consider any collocational patterns. What effect do the words in the immediate environment have? These considerations can help to initially group meanings based on ‘minimal contextual evidence’. Then if the keyword within a citation could have more than one reading, it should be defined for each of these readings. So syntax and collocation are clearly important. For an adjective, an essential criterion in distinguishing its senses would be the (type of) noun that it collocates with. There are also a huge range of syntactic clues that might be crucial in certain cases. A possible distinction is between senses where the adjective is in attributive (‘my old girlfriend’) or predicative (‘my girlfriend is old’) position2. Or there may be both ungraded (‘these numbers are odd’) or graded (‘these numbers are very odd’) senses3. Particular grammatical formulations can often signify particular senses. During a brief overview of prototype theory, Jean Aitchison discusses old. She claims it has a default 1

This may be an effect of frequency. The Longman Grammar of Spoken and Written English reports: “Interestingly, several of the most common attributive adjectives in both conversation and fiction comprise contrasting pairs” (Biber et al. 1999: 514). 2 Although most of the time an adjective with a certain sense can appear in either position. 3 Such minimal pairs are uncommon though: in real language there are usually several syntactic or collocational differences between distinct senses.

7

meaning, ‘aged’, and that in some cases like this, “a basic meaning can be detected by a lack of restrictions on the surrounding syntax”(2003: 64). The less basic meaning ‘former’ is often indicated by a particular syntactic feature: possession (‘our old house’). Likewise, other syntactic features can indicate distinct senses. 2.1.2 Adjectives and polysemy Cruse (2000: 110ff.) identifies and discusses various different varieties of polysemy4, including autohyponymy, automeronymy, autosuperordination, autoholonymy, systematic polysemy, facets, and perspectives, but it is far from clear that any of these types of polysemy can be applied to adjectives. Certainly, none of the examples given in these sections are adjectival. Some varieties of polysemy involve the lexical relations of superordinate-hyponym, or meronym-holonym (part-whole), relations that do not readily apply to adjectives. Others involve particular features or properties of, or emphases on, a subject, and adjectives obviously do not denote subjects. Adjectives do not have features; most of them5 are features. The one type of polysemy that does seem to relate well to adjectives is what Cruse calls ‘sense spectra’ (ibid: 119-20). Stock also discusses the concept, discussing the type of blurring between senses “where a word seems to operate on a cline between two or more meanings” (Stock 1984:137). She gives the example of culture, of which a large proportion of the citations seem to indicate a number of rather vague areas of meaning, which cannot be clearly defined: relating to the arts or sophistication, relating very generally to a society or civilization or something broader or narrower, relating to heritage or tradition, or relating to some identified section of society, e.g. “the throw-away culture”. The traditional dictionary format is hard pressed to deal with this phenomenon. Many discussions of polysemy are highly theoretical6, but this investigation is more a practical one into how well dictionaries deal with adjectives that they present as polysemous. The primary concern is with sense discrimination in a dictionary rather than the mental lexicon. Whether polysemy, however defined, exists or not as an abstract concept, and whether or not it can apply to adjectives, the ‘polysemous adjectives’ analysed are a practical category determined by the dictionary, even if it might be more accurately described by other terminology. A more relevant and practical discussion is conducted by Moon of “whether all words that are treated as polysemous in dictionaries are actually polysemous” (Moon 1987a: 173). She discusses what she calls “quasi-monosemous” words, i.e. words that “have a single meaning or semantic core underlying their various uses, and yet are difficult to explain in a single dictionary sense and definition” (ibid: 174). Amongst these she includes ‘delexical’ verbs such as take, give, or keep (in such diverse readings as ‘keep warm’, ‘keep thinking’, ‘keep 4

Cruse uses the term in an abstract sense, but for the purposes of our investigation, ‘polysemy’ refers in more practical terms to there being more than one sense in a dictionary entry. Similarly one may speak of ‘senses’ in more practical terms to divisions of a dictionary entry and the meanings they represent. 5 These are generally the descriptive adjectives, which constitute the majority of the wordclass, and are to be distinguished from relative adjectives which refer to a noun, as well as some other classes such as colour adjectives and postdeterminers. The distinction between descriptive and relative adjectives (e.g. in Miller 1998) roughly corresponds to that which the Cobuild grammar (Sinclair (Ed.) 1990) makes between qualitative and classifying adjectives, based upon gradability (cf. Cruse on gradable and nongradable adjectives). The Longman Grammar (Biber et al. 1999) uses the terms descriptors and classifiers (or relational adjectives). 6 Some attempts to refute polysemy amount to little more than unhelpful quibbling over terminology. Catford rejects the term polysemy, but only in order to create new terms of his own: eurysemy, stenosemy and oligosemy. He claims that it “seems probable that all polysemes can be discovered to be, in fact, either (1) homonyms, i.e. two quite distinct words… or else (2) eurysemes, that is, words with a single meaning, but which covers [sic] a wide range” (Catford 1983: 25).

8

away’ etc.), various nouns which can be conceived of in terms of various features or perspectives (she discusses mouth and top), and notably, adjectives. Adjectives, she says, are “notoriously hard to divide into senses” since they are “highly context-dependent and flexible, taking on as many meanings as you like or have space for”. The example she gives is of light. This, she claims, has only “two main strands of meaning” (which nevertheless “intertwine”): “not heavy in weight” and “not intense or great in amount, degree, etc.” But she goes on to list ten context groupings which each require different wordings to explain their meanings: “a light rain”; “a light blue shirt”; “the light breeze”; “a light sleep(er)”; “her light voice”; “a light lunch”; “ a light white wine”; “light injuries”; “light housework” and “her light graceful step” (ibid: 179). Thus although she claims only two true ‘senses’, at least these ten usages would apparently need to be treated separately in a dictionary. The highly context-dependent nature of adjectives should be stressed here. Seppänen (1984) argues against a proposal by Carole Hines7 that good and great are actually monosemous (having only one sense), with differences arising from combination and thus not belonging to the adjective. Hines’ position would seem somewhat extreme taken to such lengths, yet one can well understand it. Katharine Miller makes the following point in the course of her account of how adjectives and other modifiers were dealt with in the WordNet lexical database: “Adjectives – whether polysemous or not – are selective about the nouns they modify. The general rule is that if the referent denoted by a noun does not have the attribute whose value is expressed by the adjective, then that adjective-noun combination requires a figurative or idiomatic interpretation[8]. For example, a building or a person can be tall because buildings and persons have HEIGHT as an attribute, but streets and stories do not have HEIGHT, so tall street and tall story do not admit literal readings… It is really a comment on the semantics of nouns, therefore, when it is said that adjectives vary widely in their breadth of application… The semantic contribution of adjectives is secondary to, and dependent on, the nouns that they modify” (Miller 1998: 55).

It may therefore be a problem for the lexicographer deciding to what extent it is possible or helpful to treat an adjective in isolation, if they are so strongly dependent upon their noun collocates. But it can at least be observed that noun collocates, or at least types of noun collocate, seem a promising place to start in distinguishing meaning. The methodology used in the production of the Cobuild dictionary involved classifying senses of polysemous adjectives “according to the semantic class of the noun which the adjective modifies” (Hanks 1987: 122); also, in his discussion of syntactic criteria for the demarcation of meaning, Svensén suggests that a suitable demarcation criterion for adjectives may be “the way in which the meanings depend on the words qualified” (Svensén 1993: 113).

2.2 How are they best defined? Defining words is a complex art deeply enshrined in tradition. Patrick Hanks (1987) gives a good overview of the issues as he explains the new style of definition used in the Cobuild dictionary. He discusses the many conventions that have developed 7

Originally made in Hines, C.P. (1971) “Lexical Integrity: Good, Great, and Well”, in The Fifth LACUS Forum 1978, Columbia, S.C. 8 As a working rule this does not seem very helpful, since how is one to establish in the first place which attribute the adjective is supposed to have before determining whether the noun it relates to has it? It would seem to be a rather back-to-front approach, and it seems questionable whether it would work for more complex adjectives than the very simple one given. The general point she goes on to make, though, is helpful.

9

over the centuries: formulae such as of or pertaining to, or any of various, ambiguous use of etc., and the widespread use of parentheses for obscure functions. These have tended to make definitions difficult to interpret, and possibly to impose ‘bogus precision on word meaning’ (Hanks 1987: 116). He contends that the root of the problem lies back in the 17th and 18th centuries, when dictionary entries were just becoming formalised, and when developments in philosophy and logic led to an unhealthy obsession with substitutability of definitions. Later came similar preoccupations with reductionism (isolating minimal units), and with making definitions into precise inventories of sufficient conditions. There arose an unjustified snobbery about more discursive styles of explanation which may be clearer and more representative. The traditional style of definition is concerned with giving an exhaustive and precise account of word meaning, to the exclusion of syntax, collocation, and usage restrictions or preferences. By contrast, the new Cobuild definitions are in ordinary English prose and where possible contain the word being defined in its natural syntactic environment. The emphasis is not on defining extensively, but rather indicating typical usage. Different strategies and considerations are discussed: for the definition (or rather, the explanation) of various types of lexical item; for indicating narrow selection restrictions as well as very general usage; for indicating classes of participant roles; and for dealing with items with particular connotations. He notes particular problems phrasing the explanations of some adjectives (ibid: 130). In terms of syntax, most adjectives can occur in either predicative or attributive position, and the predicative position allows an explanation sentence such as: “Something that is capacious has a lot of room or space to put things in.” However, the fact that some adjectives can appear only in attributive position caused a particular problem in the phrasing of a Cobuild explanation. The absence of a sufficiently general head noun in English means that, while they can be defined in restricted senses (“A commercial product is made to be sold to the public”), for general senses they cannot. One cannot say “*A commercial something or someone…,” so in this case there is no alternative but to fall back on a slightly more traditional style: “Commercial means involving or relating to commerce or business.” Consistency is very important to lexicographers, and when it comes to the phraseology of definitions, the conventions are often set out quite rigorously. Gove (1968), for example, set out at considerable length the formulae to be used for defining adjectives in Webster’s Third New International Dictionary. But as illustrated above, careful consideration of phraseology is no less necessary for new styles of definition than for the traditional styles. According to Landau (1984: 140) “every dictionary has its own recommended style for defining adjectives.” He gives a list of 37 introductory words or phrases (traditionally) used by dictionaries in defining adjectives (e.g. associated with, being, having, indicating, like, made of, relating to, etc.) and notes the sloppy practice of stringing such phrases together in order to combine senses, as in “of, for, characterized by, or resembling…”. This is one way of dealing with the kind of “quasi-monosemous” adjectives that Moon discussed (Moon 1987a: 181). Moon also suggests – as a more radical possibility – that the notion of syntactic substitutability of the definition could be dropped for these words, and short accounts written of meaning and function, as is generally the practice with grammatical words. However, assuming that substitutability is somehow desirable, the more important issue behind this is that of precision of definition. A definition should say what a word means. Svensén (1993: 116-24) makes the distinction between a paraphrase and a true definition. Whereas a paraphrase rewrites the name of the sign, a true definition ‘explicitly represents the content aspect of the sign.’ The problem with adjectives is that it is often not possible to do any more than paraphrase them fairly 10

approximately within a substitutable definition. In many cases the difficulty arises because the properties the adjectives designate are so basic that they are virtually impossible to define (see, e.g. Wierzbicka 1993: 65ff. on the so-called ‘indefinables’ good and bad). They invariably defy analytical definition since they cannot usually be broken down into component features, and do not form hierarchical systems as nouns do9. The best one can hope for with simple descriptive adjectives is a simple descriptive modifying clause or phrase. Otherwise one may have to rely on the particularly inexact practice of defining adjectives using other adjectives. Defining an adjective simply using a synonym – although a fairly common practice – is obviously limited, and perhaps useful only as a kind of gloss, where the more specific information about the adjective is given in some other way. Defining an adjective with a wide range of usage using another adjective with a wide range of usage merely asserts that there is an area where the two ranges overlap, without specifying the boundaries of that area. The two ranges will not be identical; language naturally rejects absolute synonyms. A somewhat more satisfactory solution may be to use a string of near-synonyms: a thesaurus style. Stock (1984: 139) suggests this as a way of treating senses which merge into one another, with the support of examples. Taken together, the synonyms can cover the full usage range of the word they seek to define. Alternatively, if it can be made clear that they are all referring to the same particular usage, they could highlight the area on the sense spectrum where they all overlap as the sense of the headword, although this may be less practical. Hanks comments regarding this practice: “It is short and succinct, and wastes little space. However, it too has its problems: chief among them is the fact that it carries to an extreme the general failure of English dictionaries to specify the syntactic conditions under which any one of the defining synonymies is applicable” (Hanks 1993: 104).

2.3 How are they best presented? There are many other crucial elements to a good dictionary entry apart from clear sense distinction and good definitions. Depending on the particular headword and the aims and target audience of the particular dictionary, there are all kinds of different information that may be important, be it concerning grammar, syntax, collocation, register, frequency or etymology. These can all add to overall presentation and clarification. 2.3.1 The role of example sentences One important element of the dictionary entry worth focusing upon is the illustrative sentence or phrase, since it potentially plays a crucial role. Example sentences “perform a useful backup to the explicit grammatical designation, in clarifying in real language data what is stated abstractly and generally” (Jackson 1985: 58). They thus appeal much more directly to the intelligence and intuition of the reader, helping them to grasp meaning without having to rely on the metalanguage and stylistic conventions of the lexicographers. One of the most impressive hallmarks of Johnson’s dictionary is his selection of high-quality and interesting literary quotations, and similarly the Oxford English Dictionary is notable for the sheer volume of its diachronic citations. There is certainly much to be said for using examples from real texts. Fox (1987) argues this case, with particular emphasis on the benefits to learners of English seeing the word in its characteristic environment, illustrating typical collocation and syntax much more reliably than could be represented by an example invented by the 9

With the possible exception of some very technical adjectives.

11

lexicographer based on his or her intuition. One risks sacrificing a great deal of naturalness in abridging examples, but Fox notes the fact that citations may sometimes – albeit reluctantly – have to be edited. This could be for a variety of reasons: usually because the example is too long (space restrictions being so pressing) or complicated, or perhaps in the interests of good taste or political correctness. But certainly with advances in text processing and corpus linguistics, lexicographers now have little difficulty in finding citations, for English dictionaries at least. On the contrary, they risk being swamped by far too many citations, not knowing where to begin analysing them (Simpson 2002: 13). In the case of adjectives, a typical short noun phrase is often sufficient to clarify a sense (for example, “raw materials” for the ‘unrefined’ sense of raw), although this requires great sensitivity on the part of the lexicographer; often a longer example will be required. It depends, amongst other things, on how simple or complicated the adjective is in the relevant sense, how much it depends on the wider context for its internal meaning, and also its typical syntactic behaviour. Is it usually in attributive or predicative position? Is it gradable? Is it usually followed by a prepositional phrase? What kind of noun does it collocate with? Here a tricky problem is indicating variability or range. A single example does not show whether the adjective in the relevant sense is restricted to a certain noun or whether it can be used very widely, nor whether its syntax may vary. Of course, such information may then have to be presented in some other way, perhaps more formally. 2.3.2 The structure of the dictionary entry How can the structure of a dictionary entry most effectively present an adjective’s different senses, as well as – perhaps – its basic meaning? There are definite limitations to the conventional format, that is, a list of numbered senses. It imposes an often misleading sense of hierarchy onto the various senses, and struggles to show the precise relationship between senses, using the same system of presentation for specialist meanings, metaphorical meanings, and contextual meanings (Moon 1987a: 182). It cannot show where there is a fuzzy rather than a clear distinction between senses (Stock 1984: 139; Hanks 1993: 103). This may suggest a need for more flexible entries: a more thesaurus-like structure to one entry and a strongly hierarchical one for another; new systems of labelling to denote different kinds of sense distinctions; or it may be an area where hypertext and the possibilities afforded by electronic dictionaries will make a significant difference. Dictionaries for now tend to have uniform structures throughout, with the part of speech often forming a major division, but the structure of the senses may be flat (where all senses have equal status) or hierarchical (Atkins 1993: 19). A hierarchical structure, in view of the problems noted above, obviously allows more scope for arranging and grouping senses. This generally involves numbered sub-senses (1a, 1b etc.), and these can be useful in representing usages on a semantic cline (Stock 1984: 140), as adjectives often are. Alternatively there may be no sense numbers at all in an entry, as in earlier editions of The Chambers Dictionary. This may lead to more ambiguity, but at least presents no unrealistic pretension of hierarchy or autonomy of senses. The ordering of the senses may be historical (as in the OED), by frequency (as in Cobuild and other dictionaries for advanced learners) or logical. Ordering by frequency precludes a hierarchical structure, but at least ensures the major senses can be immediately found. These are often buried lower down the entry if the ordering is historical. A logical ordering gives some freedom to the lexicographer in arranging senses in a way that portrays a clear overall description of a word (See Svensén 1993: 12

212-3), but is far more subjective than other methods of ordering, and if a logical ordering is employed, the lexicographer must work hard to ensure logical consistency at least within each entry. Higher levels of distinctness of senses can otherwise be indicated by separate entries, usually marked with numbers in superscript or subscript after the headword. This practice is usually restricted to homographs – words which are unrelated but merely happen to be spelt the same. This is rarely the case with multiple adjectival senses, although the example of light is one exception. 2.3.3 A word about the users of dictionaries One final note: as was beginning to become apparent above in the discussion of different structures, the target audience of the dictionary is a major factor. This investigation evaluates monolingual dictionaries both for native speakers and for learners of English, but the characteristics and requirements for each group are quite different. Learners’ dictionaries, largely since they are a more recent phenomenon, tend to be a good deal more innovative and experimental than their counterparts for native speakers. They are also less concerned with minority vocabulary or senses, and have to be a good deal more explicit in the information they give. Nevertheless, since these are all dictionaries for human beings rather than machines, the intelligence and common sense of either audience should not be underestimated. They are, in Schelbert’s words, “poetic beings”, “quite capable of interpreting and integrating meanings” (Schelbert 1988: 67). There will always be new or unique senses of words that will have to be deduced given a certain context, so dictionaries neither can nor should try to cover absolutely every eventuality. No dictionary user likes to be patronised. Native speakers especially are more likely to be interested in the rarer senses of a word where their intuition cannot reveal the meaning, than in hair-splitting distinctions between common meanings they know well. Bearing this in mind, unreasonable expectations should not be placed upon the dictionaries analysed.

13

3

Methodology

The above discussion of how dictionaries can best deal with polysemous adjectives gives some context for an investigation into how well modern dictionaries actually do this. The results of this investigation are given in chapter 4, but in this chapter a brief description is given of how those results and evaluations were made. First, six dictionaries were chosen; second, a random sample of polysemous adjectives was chosen; next, corpus analysis was undertaken on those adjectives (before studying the dictionary entries), and finally, the entries for those adjectives in each of the six dictionaries were evaluated.

3.1 The dictionaries chosen Six of the most comprehensive monolingual English dictionaries were chosen10; each of them recently published in Britain, and in one single volume. Four (Collins English Dictionary (CED), Oxford Dictionary of English (ODE), Chambers 21st Century Dictionary (CTD) and New Oxford Thesaurus of English (NOTE)) are aimed at native speakers of English, so that this study concentrates on such dictionaries. But for comparison, two EFL (English for Foreign Learners) dictionaries were also chosen: the Macmillan English Dictionary (MED) and Collins Cobuild Advanced Learner’s English Dictionary (COBUILD). Here is a brief summary of each dictionary: CED, 6th Ed. (2003) The first edition of CED was published in 1979, based on the American model of the ‘college dictionary’ (see Béjoint 2000: 75-7). As such it was one of the first British dictionaries structured on an encyclopaedic word-list rather than the strongly analytical macrostructure that was previously the norm, and it includes a great deal of encyclopaedic information and much specialized vocabulary. This sixth edition claims to have greatly updated its coverage of regional and specialist vocabulary, and to have drawn on the Bank of English corpus in its revision. ODE, 2nd Ed. (2003) ODE’s first edition was published in 1998 as the New Oxford Dictionary of English (NODE). It was written on new principles, separating it somewhat from the lineage of OED-based dictionaries such as the Concise Oxford Dictionary. Chief 10

Although one is called a thesaurus, we can still refer to it as a dictionary, since a thesaurus is, broadly speaking, a dictionary of synonyms.

14

amongst these principles is an emphasis on ‘central and typical meaning’ rather than ‘necessary conditions’ in definition (see ODE: Preface to the first edition). Its entries consist of numbered core meanings to which a number of unnumbered subsenses may be attached. It is based largely on the 100-million-word British National Corpus, as well as data from the Oxford Reading Programme upon which the OED is based, and a new reading programme targeting specialist fields. It also claims emphases on ‘explaining and describing as well as defining’, and on international varieties of English. CTD, Revised Ed. (1999) Although this is marked as only a revision of the first edition of Chambers’ ‘21 st Century Dictionary’, this dictionary actually goes right back to 1901 as Chambers Twentieth Century Dictionary. This edition has shed some of the more unhelpful tendencies it has inherited, yet remains very traditional11. It has adopted sense numbers rather than separating senses with a semicolon, has dropped its strongly analytical macrostructure, has adopted IPA for pronunciations, and has begun to include example phrases and sentences where they are deemed necessary. However, definitions remain highly compact and traditional. This is the only one of the six dictionaries not to be formally associated with a particular corpus. MED, 1st Ed. (2002) MED is a new EFL dictionary borrowing many features from its competitors (such as the Oxford Advanced Learners Dictionary and COBUILD), and generally adopts a very flexible approach to the vocabulary based on pedagogic considerations. It concentrates on treating a core of the 7,500 words deemed most likely to be needed by students in great depth and with great clarity, giving a much more brief treatment of less common words. It is based upon analysis of its own corpus, in collaboration with the University of Brighton’s I.T. Research Institute. COBUILD, 4th Ed. (2003) COBUILD’s first edition was published in 1987, based almost exclusively on a thorough analysis of what later came to be called the Bank of English corpus, in collaboration with Birmingham University. Its innovative features are many (see Sinclair (Ed.) 1987). We have already noted its radical defining style. It was also one of the first dictionaries to sort senses by frequency, include only real examples from an electronic corpus, and include a separate column giving highly detailed grammatical information in code form. Its focus is on the commoner words in the language. NOTE, 1st Ed. (2000) NOTE is a highly innovative new thesaurus, based on various existing references and lists of words grouped together along with analysis of the British National Corpus. As well as long lists of synonyms it includes antonyms, and where appropriate gives terms related in other ways. It contains many tables of hyponyms (e.g. a list of names of deer species under deer. In many ways it takes a very dictionary-like approach: listing alphabetically, separating by sense as well as by part of speech, and including illustrative sentences for each sense. Its headword list is necessarily restricted to the 11

With hindsight, the new 2003 edition of The Chambers Dictionary would have been a better and more interesting dictionary to study. This retains the traditional features mentioned here (notably the usage of semicolons) and is highly respected, particularly amongst word-game enthusiasts. The fact I was unaware of at this selection stage was that CTD is a slightly abridged version for a more contemporary audience. In any case, CTD seems to be less prestigious than its more traditional and unabridged relative The Chambers Dictionary.

15

commoner words of English, but those it treats it treats very thoroughly, and this includes all the words in our sample.

3.2 A random sample of polysemous adjectives Since it is the dictionaries that are being evaluated, the random sample is based upon those words presented as polysemous adjectives in one of the dictionaries, CED. Its flat structure makes it simple to calculate based on the number of senses. In early attempts to seek a sample I found that some adjectives that did not seem to be particularly polysemous, such as lax or leaden, were given as many as five senses. Six senses therefore seemed a reasonable cut-off point, so I aimed to find a random sample of ten adjectives with at least six adjectival senses. CED 6 has 1872 pages, which, divided by ten, rounds to 187. Using a computer program I generated a random integer between 1 and 187 and the first result was 61. I then went through CED and, starting from page 61 I looked through until I found a word marked as an adjective with at least six adjectival senses. In some cases this involved searching for several pages. The adjectives chosen by this method were apart, canonical, deep, floating, idle, marginal, particular, remote, stable and unbalanced. Apart is problematical. According to three out of the six dictionaries studied, it is not an adjective at all, rather an adverb, whilst MED and COBUILD give it one adjectival sense and the rest adverbial. However, CED has made what seems fair to assume is an error on the part of its compilers by marking all seven of its senses under the word-class label “ADJECTIVE OR ADVERB”. For this reason, apart is not analysed as part of this investigation, so that nine rather than ten adjectives are evaluated. Otherwise, this seems a suitably varied sample, the analysis of which should serve to highlight many key issues.

3.3 Analysis of corpus data A fairly crude method of analysing corpus data was employed. A sample of concordance lines was extracted from the Bank of English corpus for each adjective12, and from a printout, each example was manually grouped into different senses on scrap paper. The sample was generally of 200 lines, which proved sufficient in all cases except particular (see 4.6.3), but in the cases of canonical and unbalanced, 100 lines was deemed sufficient due to the comparatively low overall frequency of these words. The grouping was chiefly done on the basis of noun collocate (early experiments grouping idle on more syntactic grounds proved less than useful), although other factors inevitably came into play, be it only on an intuitive level. Most obviously, the effect of my existing knowledge and experience (or lack of it) of the adjectives in their various usages will have been a major factor. The results were then put into the ‘results of corpus data’ tables shown in Chapter 4, together with frequency and collocational data gleaned from the corpus. The aim in presenting these results was not to produce some exemplary treatment of each adjective against which to criticise each dictionary entry. On the contrary, the expectation was always that these state-of-the-art professional dictionaries would treat the words better, more comprehensively and elegantly than I could based only on a restricted sample and my own background knowledge and intuitions. It should 12

The concordance lines (with a limited context of 80 characters) for one of the adjectives, marginal, is given in the appendix as a sample, whilst the complete set of corpus data analysed can be found on the enclosed CD-ROM with either 80 or 512 characters of context for each line.

16

also be stressed that although these results tables appear in chapter 4 after the discussions of the dictionaries and the sense discrimination tables, the results tables were produced before looking at the dictionary entries in any detail. I was therefore content on occasions to leave my results tables relatively ‘messy’ and dotted with question marks, and resisted the temptation to tidy these tables up after later studying the dictionary entries, in order to present a truer picture of my analysis with all its limitations. The tables aim to give a representative sample of all differing usages found in the concordances, and the overall aim of the analysis was to build up a sufficiently detailed picture of the complexities of each word in real usage, so as to be in a position to understand and judge the quality and comprehensiveness of the dictionary entries.

3.4 Evaluation of dictionary entries Having studied each dictionary entry for a certain adjective in detail, I was able to write a short appraisal of the strengths and weaknesses of each. It quickly became apparent what the particular aims and strengths of each dictionary were, and by reading through an entry under the assumption that it excelled in relation to those aims and strengths, particular errors or weaknesses frequently became apparent. Also, sense discrimination tables were made as a complement to the written evaluations, in order to illustrate the coverage of each dictionary and to compare the way they divided the senses.

17

4

Results: An evaluation of dictionary entries for polysemous adjectives

4.1.1 Dictionary evaluation: canonical This analysis begins with a somewhat atypical adjective. Canonical is a relative adjective; a complex adjective based on the complex noun canon. It functions in the same way modifying nouns can in English, as in ‘canon law’, for example. It is the type of adjective Gove probably had in mind when he pointed to one particular difficulty with adjectives: “Many adjectives – especially those which are noun-derivatives with suffixes of multiple meaning – present the paradox of having numerous meanings but low over-all frequency. Practical considerations call for treating the various meanings of such adjectives in minimum space, using defining formulas which will cover several related meanings simultaneously; yet the acceptable practice of writing definitions which have breadth of sense-coverage too easily degenerates into an unacceptable practice of writing definitions which are merely vague or actually inexact.” (Gove 1968: 5).

As is seen from 4.1.3, its meaning is so technical as to be difficult to ascertain from even a relatively wide context. A glance at its strongest collocates shows that it is found in technical environments. Having some background in Biblical Theology, the only sense of this word I was truly familiar with was the one I have numbered 1f. Even there one can see that the adjective can have various semantic relationships with the noun it modifies. The analysis of this data was undertaken before taking more than a brief glimpse at the dictionary entries, so as to avoid pre-empting the dictionary analysis. I was therefore content to leave my results table in this case looking rather messy: it will thus be a truer test of the dictionaries’ ability to clarify this difficult word. The two learners’ dictionaries offer little help. The COBUILD explanation applies directly only to the one noun-collocate status (as well as to writer in the example); only part of my sense 1e is represented. As far as this goes the explanation is clear13. MED is very brief, including no examples. Sense 1 is especially vague, but is also an admirably concise summary of some of the senses, albeit to the exclusion of the more technical ones. This is in line with MED’s stated policy (in its introduction) of giving ‘a simple and clear explanation of meaning’ for ‘very infrequent (receptive) words’, the brevity enabling them ‘to include much more vocabulary of this type.’

13

And perhaps indicates that these examples belong with 1b or 1c rather than 1e.

18

The conciseness of CTD here is even more to be marvelled at. Senses 2 to 4 consist only of synonyms, in defiance of the apparent complexity of the word, while s.1 amounts to little more than a cross-reference to canon, with some elaborate hedging. In terms of explanation it is of little help; presumably the user is expected to refer to the entry for canon, where more explanation is given. NOTE concentrates on what equates to my 1a-c in its s.1, with a long list of adjectives that combine to give a good sense of the meaning. The one it represents via capitals as the most important is recognised, perhaps a key element of meaning I missed in my analysis. There is also flexibility of format, seen in its presentation of the canonical hours in a table, representing a hyponym rather than a synonym relation. To really begin to grasp the complicated meanings represented in 4.1.3, though, one must turn to CED and ODE. There one sees, for instance, that my s.2 citations actually refer to church law rather than authority more generally, a point difficult to deduce from the examples, but this may be because some of my examples there belong with ODE s.3/ CED s.5 and s.6. This fine distinction of CED’s between the clergyman sense and the cathedral chapter sense is a surprising one. Here the limitations of CED’s flat structure is obvious: this slight distinction gives these senses equal status and autonomy with the other very different senses. CED includes canonic as a variant form, unlike ODE which treats it separately, but with its s.2 being a variant form of canonical. This seemingly explains why CED includes a musical sense unattested in my sample: ODE’s s.1 of canonic. ODE’s structuring of s.2 (of canonical) is good, relating several senses together coherently: the scientific sense comes under the same ‘umbrella’ as the religious one, and the common theme is made clear. Clearly my s.3 (not as technical as it at first seemed) belongs with 1d. The meanings explained are admittedly less specific than those seen in the examples, but this is only right for a general dictionary. Canonical is used very technically within a number of academic disciplines with particular nuances, but the full explanation of such senses belong in the literature specific to that discipline rather than in a work of general reference.

4.1.3 Summary of sense distinctions: canonical (sense) §4.1.3 CE D according to recognized laws (1a), 3 3 accepted as authoritative 1b,(1d) typical, regular 1c belonging to literary canon 1e 1 belonging to sacred canon 1f, 1g based on church (canon) law 2 2 relating to cathedral chapter 5 relating to canon (clergyman) 6 in form of musical canon 4 ‘canonical hour(s)’ (for prayers) under 2 s.e:1 ‘canonical hour(s)’(for marriage) s.e:2

ODE 2:sub3 2:sub1 2:sub2 2 1 3 s.e:1 s.e:2

CTD (3) 2 (1) 4 s.e:1 s.e:2

MED

COB.

1 2 -

1 -

Key to Abbreviations used: s.1, s.2, etc. =sense 1, sence 2, etc. 1: sub2 =2nd (unnumbered) subsense of sense 1 s.e. =separate (compound) entry s.e:1 =1st sense of separate entry ex. =example 19

NOTE 1 2 table -

4.1.3 Results of corpus data: canonical Total frequency in corpus: 226 Tagged as adjectives: 226 Sample size: 100 lines (all tagged as adjectives) Strongest noun collocates14: status, works, prayers, gospels, hours, variate, texts, Jesus, variates, readings, writers, writings. Excluded from sample before analysis: None. Sense/ subsense discovered 1a: (?)16 Traditionally accepted to be true or right 1b:(?) Especially authoritative

1c: (?) Held to be fundamental and indicative 1d: (?) acknowledged by zoology 1e:(?) Related to a special corpus of authoritative literature

(TYPE(s) OF) nominal collocate15 OPINION

PRACTICE DOCUMENT

forms cool reading ?performance EXAMPLE

study analogy life forms feature space TEXT / WORK OF ART

(works)

PAINTER writers status (x5) place (various)

?

Examples …Wilson takes what was by his time the canonical Southern view: Reconstruction was an unmitigated evil…/ …these by now canonical critical arguments…/ …The canonical response to this greenhouse effect is to cut the amount of co2…/Reliance on espoused practice (canonical practice) can blind an organization…/ …the canonical NSC-68, the actual blueprint for the cold war,…/ …the self-styled canonical list of weird band names…/ …the canonical left-wing version, formulated by the Comintern…/ …upward intellectual mobility is no longer associated with the personal mastery of canonical forms./ …clarity and wit that transcend the wasted cliches of canonical cool./ It's the individual response and the canonical reading./ … whether there can be such a thing as a canonical performance…/ The comically naive routines with a looking glass `are almost a canonical description of Freud's mirror-phase and its perils"./ …the canonical example of why this is so./ my canonical example of this type of activity…/ The latter study, which is often cited as `canonical" in the field of feminist scholarship…/ …the Copycat domain and a few canonical analogy problems in it…/ …dissociating itself from the rigor mortis of canonical life forms…/ …a secondary stabilisation rather than a canonical feature of terrestrial vertebrates./ The otus of C. jemtlandica do not occupy a distinct portion of the canonical space…/ The canonical works on these themes are Scorsese's The King of Comedy, and in the theatre Trevor Griffiths's Comedians…/ …through the analysis of canonical texts…/ …paganism's canonical texts, such as J.G. Frazer's The Golden Bough…/ …great, canonical poetry… say, Shakespeare and… Milton…/ …Lawrence, Maupassant, Chekhov, Hemingway.20 From these "canonical" sources she forged her own prose…/ …canonical works preserved in books and museums…/ …any number of "canonical" works were far from that in their own time./ …how a text becomes canonical./ The constellation drawings associated with his catalogue soon became canonical,…/ critics concerned with the recovery of a buried past often bring out the repressive aspects of "canonical"

works./ …whether Murdoch makes it on to the canonical lists…/ Caravaggio has attained canonical status: his portrait now appears on Italy's 100,000-lire banknote./ …pocket-sized selections from the works of canonical writers from Montaigne to Saki…/ The novel achieved a wide currency, gaining a close to canonical status…/ …keeps… Shakespeare in canonical place./ …between the canonical and the demotic…/ `the African body emerges as a canonical announcement of a promised or covenanted body."/

14

The noun collocates with highest t-score within a 4:4 span of the node adjective; figures based on the entire Bank of English corpus rather than the sample alone. 15 In this column, CAPITALISED words give types of noun collocate rather than the collocates themselves, which appear in lower case, generally where they are one-off cases. Items in bold type are collocates that appear several times in the sample, and their frequency is shown where it is 5 or more. 16 Question marks are used in the ‘sense’ column of these results tables to indicate that the glosses of meaning were merely guesses at the time of my original analysis, or that the meaning was unclear from context. This applies to nearly all senses of canonical, since it is a particularly difficult and technical word. Where question marks appear in the noun collocate column, this generally indicates that there was a problem with the particular example; either that it was unclear, or that I was unsure of its meaning.

20

1f: Included in or relating to the Bible

SCRIPTURE (Gospels)

writers Jesus 1g:(?)Selected as Authoritative within Islam 2:(?) Related to Church authority or hierarchy

prayers sayings ACTION

AUTHORITY

norms existence ?monstrosity ?

hours

3: ? (Technical sense in maths or physics)

equation

? Of some responsibility

age

variable correlation axis ?function

The major sources for our knowledge of Jesus are the canonical Gospels./ …occurring in the NT in reference to the canonical OT literature…/ …association with Solomon seems to have justified canonical status for both Proverbs and the questionable Ecclesiastes./ …various extra-canonical sources…/ It is hard to read the NT and gain any secure feeling at the present that canonical writers were attacking the Gnostic devotees or mythologizers. / The canonical Jesus is the figure that fills the four official gospels…/ …the canonical prayers can be performed anywhere and are absolutely obligatory upon every Muslim./ …the different phases of the daily canonical prayers,…/ …the canonical sayings of the Prophet…/ …received canonical approval by Pope Paul VI in 1977…/ On at least two occasions the 1170 church received canonical visits from the Dean and Chapter of St Paul's./ go to the Court Christian to make `canonical compurgation" if he could prove he was clergy./ …the practical, ascetical and canonical renewal the Church needs…/ …through the proper canonical channels…/ …forms of liturgy that are not official and do not have canonical permission…/ Rome announced an emergency relaxation of canonical law,…/ Bloom's book is not really about the canon at all, but about canonical power./ …liturgical and canonical norms…/ … a seminary that already had no canonical existence./ …this would have been a canonical monstrosity since the Mission de France was for a given country, for a precise purpose…/ …singing the eight canonical hours./ …the prayers were arranged according to the canonical hours of the day./ …the canonical equations for Newton's laws…/ …clear trends within the discriminant space, as defined by the canonical variables./ …along the canonical variate…/ Using canonical correlation and regression analyses…/ The first canonical axis accounted for 77% of the shell morphology variation…/ …the canonical XYZ axes…/ Canonical Chi2 Function Eigenvalue Percent of variance correlation…/ …youths of canonical age (at least twelve for girls and fourteen for boys)…/ …the canonical age to become the incumbent,…/

4.2.1 Dictionary evaluation: deep Deep is a very basic descriptive adjective, and as such is not possible to define using simpler language. It is a tricky concept to describe since it is bound up in ‘orientational metaphors’ (see Lakoff & Johnson 1980: 14-21) which profoundly affect the way we think, so that we cannot easily tell what is a figurative extension and what is not. My s.1 is particularly difficult to define since it is a relative concept17. This is the most polysemous adjective in our sample, with all but NOTE showing many additional entries for common compounds such as ‘deep sea’, ‘deep vein thrombosis’, etc., as well as some rarer ones. ODE even has as a separate entry for ‘deep mourning’, which one would have thought was not a fixed expression meriting its own entry, but merely a strong collocate which belongs with ODE s.2. Compared with the extreme conciseness of its definitions for canonical, CTD here seems positively verbose. NOTE’s style seems to come into its own here: although a thesaurus, it does the job of a dictionary well, using explanatory phrases and idioms as well as synonymous adjectives. It effectively distinguishes the senses by the example phrases, with each subsequent synonymous term substitutable with deep in decreasing order. Although its flat structure does not enable it to show the relationship between meanings, nevertheless it has good logical ordering. COBUILD’s defining style shows the semantic restrictions well (compare very general s.1: “If something is deep…” with 17

Note the repeated use of the word relatively in CED; also the difficulty of my sense 2b.

21

s.10: “A deep breath or sigh…”, and the learners’ dictionaries give the clearest explanations. It is very puzzling in MED, though, to see a variety of defining styles: s.2, s.3, s.4, s.6 and s.7 using Cobuild-style explanations, but s.1 and s.5 using more traditional definitions. It is difficult to say whether this is deliberate or an editorial inconsistency. It is certainly somewhat confusing. S.1 and s.5 are not quite substitutable either: one cannot speak of ‘taking a breathing a lot of air into or out of your body breath’18. COBUILD rather surprisingly (considering my results) has the ‘deep container’ sense as the second most frequent. Structurally, the way it puts adjectival and adverbial usages together under one sense number is noteworthy, as is its grammar column, which includes a great deal more explicit syntactic information – if one can only decode it – than other dictionaries, and this without cluttering the explanation. For example, “v-link ADJ in n” as the syntactic pattern for the ‘deep in thought’ usage (s.9), and “amount ADJ”, “n ADJ”, “how ADJ”, “as ADJ as” and “ADJ-compar than” as the possible syntactic patterns for s.3 (measuring); compare ODE’s slightly ambiguous note, “after a measurement and in questions”, in parentheses before its definition. There are numerous usages not represented in my sample, e.g. “that Thomas is a deep one” in ODE s.2 (cf. MED s.7; COBUILD s.13; CTD s.11)19. But this is unsurprising with so many different senses. Each entry is very long and therefore quite demanding on the user wishing to refer quickly to it. CED and COBUILD are especially difficult to scan through. Here MED’s introductory key to senses is very useful. ODE’s hierarchical structure also makes it easier to read, although in this case the logical ordering is spoilt by the ‘breath’ sense being out of place, doing violence to s.1’s logical order20. The example phrases play a key role in such a complex entry as this. NOTE, MED, COBUILD and ODE are very clear in this respect, having examples for every sense. Even ODE gives more than one example for senses with variable syntactic form (e.g. “a deep alcove | deep in the woods”). In their absence, a usage restriction note is of some help (e.g. “said of a person:” in CTD s.11) is of some help, though more clumsy. Otherwise a definition on its own may be unhelpful and ambiguous, leaving the user guessing as to which usage it refers. Consider CED’s s.6, for instance. Not only does it use a very complex word ‘abstruse’ to define a very simple one; without either example or note, “difficult to understand or penetrate” is precisely what it is!

4.2.2 Summary of sense distinctions: deep (sense) §4.2.3 extending far down extending far back extending far in positioned far down / in

1a,(b) 1d 1e

CE D 1 2 1

18

ODE

CTD

MED

COB.

NOTE

1

1 2 1,2

1 1c 1b (1b)

1 (2) 4 -

1

1:sub1

Moreover, this is an analysis of the adjective overly affected by the noun. Compare my sense 3, which none of these dictionaries agree with: from the ‘carpet’ example I may have allowed my image of a deeppile carpet to reflect too much on the internal meaning of deep, deciding that it meant dense, whereas it probably belongs with sense 1a. The ‘knowledge’, ‘recollection’ and ‘body of work’ examples may indeed mean ‘learned or intellectually demanding’ (as CED sense 7) rather than detailed or varied. This illustrates how subjective this type of analysis can be. People can all make slightly different mental maps of how senses interrelate and function. 19 Also CTD’s sense 4, as in “lined up four deep”, which is adverbial according to COBUILD sense 5. 20 It surely belongs on its own, with sense 2, or possibly at the very end of sense 1.

22

2 -

in depth (measuring) as far as a point (‘waist deep’) in no. of ranks (‘three-deep’) far from batsmen (cricket) far down/across field (sport) of great intensity (emphasis) strongly felt (emotion) sound, heavy (sleep) high volume, intense (breath) obscure or difficult (thought) profound or complex(thought) clever or complex (person) secretive, mysterious (person) adsorbed, rapt (‘deep in…’) darkly shaded (colour) low-pitched (sound) ‘Deep South’ ‘deep sea’ (/‘deep-sea’) ‘deep freeze’ ‘deep space’ ‘deep vein thrombosis’ ‘deep kiss’ ‘deep structure’ (grammar) ‘deep litter’ ‘deep throat’ ‘deep-dish pie’ ‘deep green’ (political) ‘deep ecology’ ‘deep-discount bond’ ‘deep therapy’ ‘deep mourning’

1c (ex. in 1c) 2a 2b (ex. in 2a) 2c 4, (3b) (1:ext) 6 5 (ex. in 1e) (ex. in 1a) (ex. in 1e) ex. (ex.) -

4a,b 3 8 5 11,6 7 (10,7) (11,6)

9 12 13 s.e. s.e. **

s.e. s.e. s.e. s.e. s.e. s.e. s.e. s.e. s.e. s.e. -

1:sub2 1:sub3 1:sub4 1:sub6 1:sub7 2 2:sub1 (ex. in 2)

1:sub5 2:sub3 2:sub2 2:sub5 2:sub4 4 3 s.e. s.e. s.e. s.e. s.e. s.e. s.e. ex. under deep-dish

s.e. s.e.

3 4 12 13 5 9 5 10 11 sub 7 8 s.e. (s.e.) s.e. s.e. s.e. s.e. s.e. s.e. -

1a 2 6 5 7a 7 sub 4 3 s.e. (s.e.) s.e. s.e. -

3 5 6 (6) 8 10 14 13 9 11 12* s.e. (s.e.) s.e. s.e. -

4.2.3 Results of corpus data: deep Total frequency in corpus: 46,266 Tagged as adjectives: 44,060 Sample size: 200 lines (all tagged as adjectives) Strongest noun collocates: breath, water, sea, blue, red, [feet]21, voice, trouble, pockets, purple, South, end. Excluded from sample before analysis: 36 lines (34 adverbial usages where deep modifies a verb (“they defended deep”; “deep-rooted”) or a preposition (“Deep down I know…”); 1 noun (“Spirits of the Deep”) and 1 proper noun (as name of a song). Sense/ subsense discovered

(TYPE(s) OF) nominal collocate

Examples

*

Note the typographical error here, with the example for s.11 placed within the definition for 12. s.e. for deepfreeze (one word). 21 Where the noun collocate is not generally the one directly modified by the adjective it is placed in square brackets, as here. **

23

3 4 5 8 6 7 10 9 -

1: Descriptive/measuring function …drilling oil wells offshore in deep water…/…in the deep ocean…/ 1a: which BODY OF …in medium to deep water./ …the deep channel…/ …a deep fiord…/ … goes down WATER the deep springs of creativity for the Carry On films…/ a long way …in deep waters…/ …the deep-ocean floor…/ …the deep-sea vent…/ GEOGRAPHIC- …the deep gorge spanned by Brunel’s suspension bridge…/ … icebergs with deep keels…/ …deep valleys…/ …a deep chasm./ AL FEATURE …a deep trench…/ …bumps and deep ruts./ …deep trough…/ If your soil is not as deep or as fertile as you would like…/ soil Perennials with taproots tend to re-sprout and have very deep roots roots…/ …a sector with such deep community roots./ …it is being made a scapegoat because it has deep pockets./ CONTAINER …and several deep stowage wells let in to the cockpit sole./ A beaten metal working process for for forming deep hollow shape shapes from a single flat round piece of sheet metal./ Do they intend to throw you in at the deep end or will there be in at the a full induction programme?/ …thrown into the deep-end of a deep end complex operation./ I was in at the deep end,…/ …Deep Pan…/ …a deep plate./ 1b: thick THIN THING …20mm (12 plus 4 plus 4) is rather deep for a fretsaw./ (relatively) …5mm-deep dowel holes…/ …three-metre-deep steel cylinders…/ 1c: which THING / …10” Deep Pie…/ …a strip of fabric 13 cm deep…/ …the River measures a DISTANCE Platte—a mile wide and an inch deep…/ …in Atlantic water one specified mile deep./ …it is a mere 20ft deep…/ …the earth is a foot deep distance in water…/ The tubers will be… put 2ft apart and 8in deep,…/ … (following leaf shoots set 2cm deep…/ …the olive trees are knee deep in measure) meadow flowers…/ Skin Deep…/ …a wad of paper this deep…/ …scarred by a deep wound./ …the young elephants had deep cuts as 1d: greatly WOUND a result./ …after suffering a deep liver cut in a car crash,…/ indenting Nepal,…has long been deep in the valleys of world statistics./ … 1e: PLACE he hooked Curran to deep backward square./ …the Deep South…/ … positioned Deep Space Nine…/ …a sleepy backwater deep in the southlands./ far in / low …the big, deep galleries that have good ceilings…/ ( in cave) down gallery He entered stage left, already deep in a song,…/ extension22 POINT IN TIME 2: Intensifying function A deep, respectful silence…/ …invisible in deep dusk…/ …a deep 2a: of STATE sleep…/ He recommends deep relaxation…/ …in deep trouble…/ …deep great problems…/ …a deep conflict…/ …any question of its NOT going intensity ahead shows just how deep the divisions within the party's ranks have become./ …the corruption of politics and the bureaucracy is so deep…/ …of deep and profound significance…/ significance There were changes, real and deep, in his girlfriend./ …deep and CHANGE expedient change./ …deep assimilation…/ …further deep cuts in farm subsidies./ …a catastrophic deep-stall from which no recovery was possible…/ ACTION He took a deep gulp of the dreadful wine./ Cue deep, dark plotting in the court./ …a deep mutual interaction…/ interaction …at a very deep level…/ …of deep and profound significance…/ 2b: EMOTION …a deep but inarticulate sense that there is something strongly dangerously bogus about science,…/ …expressions of deep felt (concern) concern…/ …a deep affection…/ …a deep fondness…/ They have an (love) active sense of humor, a deep love of life./ …deep, deep pride…/ (depression) …deep political convictions…/ …a deep depression…/ …deep sadness…/ …a deep loathing…/ …deep resentment…/ …deep misgivings…/ …a deep embarrassment…/ TRANSATLANTIC diplomacy is in deep gloom about Europe…/ She took a deep breath./ …took several deep breaths…/ 2c: of high breath Deep abdominal breathing is encouraged./ volume (?) 3: Dense …a plush, deep-pile carpet…/ …foot-deep shag pile carpeting, 3a: dense carpet pile which may have cost the earth…/ physically …a clear and deep knowledge of a person…/ 3b: knowledge …I didn’t have a real deep recollection at the time…/ detailed recollection Then there is his body of work, as deep and lucid as a blue 3c: having body of work sky./ But it takes decade upon decade of development to make a variety & sport truly deep with talent./ quality sport 22

With hindsight this analysis is false, and this usage really deserves a separate sense.

24

4: significant or meaningful 5: lowpitched

MENTAL PROCESS

6: of a dark shade

COLOUR (blue x5)

level / interest SOUND

COLOUR OF THING (red x5) (?)

suntan vein

? ? ? ? ? (poetic)

bramble Throat bay (?) door nymph -

…with tremendous interest and deep attention…/ …deep thought…/ … deep processing, or the processing of meaning…/ …learning that is not deep because the method does not require the learners to think about the meaning of the association./ …at a very deep level…/ …deep interests in literature…/ …the deep bayou musical rumblings of Tom Waits./ …a deep, full growl…/ I think people are always amazed by your kind of voice— that it could—that—that a voice can be that deep./ …nearer white than the deep blue…/ …Deep Dynasty blue with pale blue walls./ …Deep Purple…/ It is a brilliant, deep yellow./ … light gold summer leaves turning glowing, deep gold in winter./ …It’s a dense, matt, deep reddish-pink…/ …deep shades…/ …the deep blue topcoat Daddy brought her…/ …DEEP BLUE SEA…/ …a deep blue lake…/ …17 deep purple sculptures…/ …deep-grey eyes…/ …deep pink Susan or Betty…/ ( apparently some kind of flower) …deep red – almost purple skin./ …deep red lipstick…/ deep red, white-eyed flowers…/ …a deep red or rust-coloured eclipse…/ …a deep suntan…/ …that deep tan,…/ …and thereby avoiding the deep-vein thrombosis known as economyclass syndrome./ …Deep Bramble…/ (actually here the name of a racehorse, but = sense 3a or 6?) …Carter, creator of the X-files, will keep Deep Throat in mind…/ …cliffs alternating with beaches, deep bays, dunes and marshes…/ Flexible ‘deep door" storage with removable dairy compartment…/ “Deep nymph techniques on rivers"…/ ( in fly fishing) …a world I know well in my deep dark heart./

4.3.1 Dictionary evaluation: floating This is a participial adjective (or, as it would be called in Sinclair (Ed.) 1990, an ‘-ing’ adjective23), based on the polysemous verb ‘to float’. It thus has somewhat borderline status as an adjective, and can only be called one in attributive position; in predicative position it is analysed as a verb. Because any verb can similarly function as an adjective in attributive position (as in, e.g., ‘a panting dog’), it is a very subtle distinction to make, and this explains why floating has not been tagged as an adjective in any of its instances in the Bank of English. It also goes some way to explaining why COBUILD does not treat it at all, except in the one phrase floating voter. Much of the polysemy of floating derives directly from the polysemy of float, but not all. It is this ‘not all’, i.e. the unpredictable usages, that generally leads to it having a headword in its own right. Only ODE and NOTE here include the more predictable main sense corresponding to my s.124, giving the entries completeness, but CED begins with less obvious senses, and yet still gives it far more senses than the others, especially if one includes the compound entries. By comparison, ODE deals with it very sparingly: just two senses including the grammatically predictable sense. My results table in many ways bears little relation to the dictionary entries. Not only is my s.1 not represented except in ODE and NOTE (with the result that the distinctions I make within it are of little consequence); my s.3 and s.4 are defined together in CTD s.1 and MED. With hindsight I see that my not making this distinction would have allowed me to classify many of my ‘rogue’ (question-marked) senses. Perhaps in my analysis I was here too concerned with the precise meaning accrued by the adjective, to the neglect of other relevant factors. Certainly there were some examples, e.g. the ‘circular disc’ example that I was inclined to gloss simply as ‘which is floating’, since it is effectively a predictable attributive use of the verb float. This is 23

Floating is a classifying –ing adjective, describing a process or state, and as such is not gradable: one cannot speak of a *very floating thing, for instance (see Sinclair (Ed.) 1990: 76-8). 24 NOTE includes an additional predictable sense (2) not represented in our sample.

25

also relevant to my ‘impressionistic usages’, where I felt unable to pin down the exact meaning. As for the technical economic usage, of which I was uncertain (s.2), only CED (s.4) makes the key technical distinctions here, with uncharacteristically wide use of subsenses. In fact s.4b seems to be doubled by a separate compound entry25. Here CTD is particularly odd: only its s.1 is a general sense; s.2-4 are all strongly restricted. It is the only dictionary not to give a separate entry to floating voter, the only phrasal use deemed worthy of one by MED and COBUILD. The different dictionaries can also vary in the precise formulation of a headword to define a specific sense: ODE has an entry for ‘floating-point’, CED for ‘floating-point representation’, and CTD for ‘floatingpoint notation’. This illustrates the lexicographer’s dilemma: how fixed does an expression have to be to have its own entry? As for the citations I failed to understand: ‘semi-floating ENV axle’ is probably explained by CED s.5; ‘floating island’ by its second sense in CTD (I would have been most confused had I had only CED to rely upon to explain this!); and ‘floating point’ by ODE, and less directly by CTD and CED. Generally speaking, CED deals with a larger number of specific senses than we might reasonably have expected, and ODE (and to a lesser extent CTD) with fewer. ODE fails to explain even the relatively basic ‘not fixed’ sense. As for the learner dictionaries, it is surprising that floating has no entry in COBUILD26, especially when we compare its very full treatment of a grammatically similar item in its entry for dying. It is understandable, though, that MED deals with it only briefly in view of the single star it awards it for frequency. 4.3.2 Summary of sense distinctions: floating (sense) §4.3.3 CED on water’s surface, buoyant in mid-air, hovering of no fixed loyalty (voter) not fixed, not settled, itinerant variable (currency) displaced (kidney) free-moving (machinery) not connected to voltage ‘floating dock’ ‘floating rib’ ‘floating charge’ ‘floating debt’ ‘floating island’ (vegetation) ‘floating island’ (dessert) ‘floating-point’ ‘floating capital’ ‘floating heart’ (plant) ‘floating policy’

1a,b,c,d s.e. 4, 3

3; s.e. 1

2

4c 2 5 6 s.e. s.e. s.e. 4b; s.e. s.e. -

(‘bone’ex.) (‘axle’ex.) (ex. in 2) ex. ex. -

25

s.e.: representation

4a s.e. s.e.

ODE

CTD

MED

1 s.e.

2 1 3 4 s.e. s.e. s.e:1 s.e:2

2 s.e. s.e. s.e. s.e. s.e. s.e. -

s.e.: notation

s.e. -

NOTE

s.e. 1

COB . s.e. -

-

-

5 -

All the compound entries are shown on the previous interleaved page along with the main entries for floating. With other entries such as deep, space did not allow this, but these other entries are particularly relevant here. There are many of them for specific meanings bound to a single noun collocate. 26 In fact, earlier editions of Cobuild did have an entry. Hanks (1987: 130) refers to it (“If a city or area has a floating population…”).

26

1 2 3 4

‘floating cloche’

-

-

s.e.

-

-

-

-

4.3.3 Results of corpus data: floating Total frequency in corpus: 6,345 Tagged as adjectives: 0 Sample size: 200 lines (all tagged as verbs) Strongest noun collocates: rate, lines, rates, exchange, line, water, [air], voters, notes, city, [sea], body. Excluded from sample before analysis: 96 lines (93 verbal usages: “floating around”; “floating in the pond”; “considering floating the business”

etc. ; 2 gerund nouns; 1 in programming language). Sense/ subsense discovered 1a: which is situated on the surface of the water

(TYPE(s) OF) nominal collocate OBJECT

1b: which is actually a watercraft of some kind

BUILDING / INSTITUTION

ORGANISM

coffin 1c: which is erected next to and over the water 1d: which will float in water

STRUCTURE MACHINE market (City) fishing line (line) fishing tackle ?shoes

2: dependent upon the economy (?)

exchange rates (x5)

(rates) rate (x5) currency ?LIMIT ?charge

3: not settled in one place

PROFESSIONAL

4: not tucked away or bound (?) detached (?) freefloating

HAIR

PEOPLE

CLOTHES bone mood anxiety GROUP

of no fixed loyalty

voter voters (x8)

Examples …the obstacles and floating debris confronting him./ There were bits of floating wood, but they were small and broken,…/ …the mean relief Of a floating log./ …an array of floating buoys each anchored by a rope…/ The presence of floating organisms in an area far from their usual habitat suggests that they were transported there…/ …casting off from our floating home for short periods…/ …he began trading from the barge, using it as a floating restaurant…/ …working on the floating research centre…/ …floating gin palaces of the sort favoured by the late Robert Maxwell./ …St John's Floating Hospital…/ Naiad became a floating coffin with the bodies of skipper Bruce Guy, 51, and sailor Phil Skeggs, 34,… still on board./ They also are discussing erecting a series of floating bridges./ Floating pontoons designed to ride out waves…/ They aim to use a 44-tonne floating derrick to drill into the sandbank and recover timber samples…/ …a visit to Thailand's largest floating market…/ …Floating City…/ …using floating lines baited with pilchards…/ This month try floating and slow sink lines using Orange Shrimp…/ …This line is the ideal stillwater floating line and sells for £39.95…/ Carp to 16 lb 8 oz on floating baits./ Bad idea 1.THE Dymaxion Car …2. Floating shoes./ …the uncertainties that attend floating exchange rates…/ …freely floating exchange rates…/ variable or floating interest rates…/…variable or floating rate note…/… floating-rate debt…/ …it was a normal development in a world of floating currencies./ …and places a `floating cap" at 95% of the median home price in higher-cost areas./ …a fixed and floating `charge" over the group's assets…/ …being cared for by six nurses and two `floating nurses"./ …the so-called `floating" population of migrants who have moved from the countryside to the cities to seek work,…/ …just a floating menagerie of lotus-eaters, looking for the next committee-approved hang-out./ …red lipstick, long, floating tresses, the lot. / …floating tendrils of hair,…/ Tops and floating petticoats…/ …she saw a number of floating black capes./ ( of nuns) …an operation to remove a piece of floating bone from his shoulder,…/ …the amorphous, free-floating moods that were more common among the older students…/ …free-floating anxiety…/ (as symptom. =independent of circumstance?) …these men, driven by Islamist zeal, formed a free-floating pool of Afghan Arab mujaheddin in Afghanistan…/ (as sense 3?) I'm something of a floating voter./ The drive to convince floating voters…/ Duncan Smith is illplaced to win the support of floating voters or Tory deserters./

27

? ? ?

missile…facility bed island

?

signifier

?

ocean riches

?

circular disc

?

marker

?

point

?

axle

(poetic/ impression -istic usages)

MUSIC

peacocks

…the site of a floating missile-tracking facility…/ Floating beds come with crisp cotton sheets…/ (=water bed?) In fact, it is more like one large floating island than several small ones - as you might normally see in restaurants./ …lesbian music emerged in the mainstream as a series of floating signifiers, linked to feminist/lesbian sensibilities…/ …this spa offers three detox packages and my floating ocean riches" took place in the authentic Moorish suite…/ The Earth was some sort of floating circular disc, and the Sun and Moon were

ring-shaped bodies,…/ …the average movement between

floating and permanent markers is between 0.2 and 1.0 cm…/ …AMD's 3dnow! technology, launched with the K6-2 microprocessor, adds 21 floating point SIMD instructions…/ The hypoid, semi-floating ENV axle on Barrie's car has the standard 3.64:1 ratio…/ The first song, `Helmet On", is one of those lazy, floating melodies that will linger with you all summer…/ …with its airy textures and floating trumpet…/ …his floating sound and buoyant, lyrical style…/ …we paraded ourselves like floating peacocks down Boylston St…/

4.4.1 Dictionary evaluation: idle In my analysis of idle I was highly conscious of two main strands of meaning (‘not working’ and ‘without content’), sub-divisible into further senses. The analysis of the dictionaries is broadly quite similar to mine, though this may be slightly obscured by the structure of all except ODE27. It is interesting to see how both my analysis and CED’s, for example, group separable senses together under ‘x or y’ formulations, but differently divided: compare my s.2a and s.2c with CED’s s.5 and s.6. Similarly there are differences in priority: whilst I saw the ‘unemployed’ sense as the general sense for the ‘not working’ meaning, ODE has the ‘lazy’ sense instead, which it illustrates with the highly appropriate collocate ‘students’! MED makes an interesting distinction in its s.2a, which it specifies as occurring ‘often in negatives’ – I simply included these senses with my s.1a, and it seems the other dictionaries did something similar. The other three dictionaries are true to form: CTD is concise yet comprehensive; aside from a lack of examples for s.1 and s.2; COBUILD once again gives very clear and explicit explanations of typical usages for each sense; and NOTE gives an excellent overview, with an impressive array of suitable synonyms, especially for s.1 and s.5. The division by illustrative example works very well, although it is not perfect: perhaps only one or two of the synonyms in its s.5 would substitute well in ‘idle curiosity’, for example. In terms of coverage, only ODE and NOTE explicitly include the ‘time’ sense (my s.1c)28. MED stands out as the only one to include the fixed expressions “the idle rich” and “it would be idle to do sth29”. None explicitly acknowledge the unique status of bone as an emphasizing submodifier in the ‘lazy’ sense, although MED has it in an example, and NOTE as an “informal” register synonym. In my analysis I found several borderline adverbial uses (see 4.4.3) which I, on this occasion, did not exclude from my sample since they fitted so well with the adjectival senses. Interestingly, none of these dictionaries acknowledge an adverbial 27

Except that ODE gives the ‘money’ sense a separate entry. This is understandable, since it is somewhat different, though clearly most closely related to the ‘not working’ sense. 28 One could speculate that this sense is implicitly covered by the word ‘unoccupied’ in sense 1 of CTD and CED, but that is getting into very ambiguous territory, and one would hope that these dictionaries would not intentionally be so vague. 29 Its use of abbreviations such as ‘sth’ in syntactic formulae could, however, be criticised.

28

sense. ODE, MED, COBUILD and NOTE include examples with lie, sit, stand and leave under adjectival senses, and admittedly verbs like these “can be used as link verbs with an adjective as the complement” (Sinclair (Ed.) 1990: 174), but it not quite clear whether all of the verbs in my sample, notably live, should be included in this set, and I note this mainly to draw attention to the ambiguity of adjectival against adverbial usage.

4.4.2 Summary of sense distinctions: idle (sense) §4.4.3 lazy, work-shy (person) not doing anything (person) unemployed (person) not in use (thing / place) spare, unfilled (time) ‘idle time’ not accruing interest (money) without purpose, trivial empty (threat) ‘idle wheel’ ‘idle pulley’

1a 1 1b (ex. in 1b) 1c 2c, (2a) 2b, (2d) -

CE D 4

ODE

CTD

MED

1

2

1

1:sub1

2 s.e. 3 5,7 7 s.e. s.e.

1:sub2 1:sub3 3 2 2:sub s.e. -

2 2a 1a 1 3 4 -

1 s.e. 3,4 5 -

COB. 3 1 2 4 5 -

NOTE 1 2 3 4 5 6 -

4.4.3 Results of corpus data: idle Total frequency in corpus: 2,538 Tagged as adjectives: 2,360 Sample size: 200 lines (all tagged as adjectives) Strongest noun collocates: rich, hands, speculation, curiosity, [bone], threat, chatter, workers, time, gossip, boast, plant. Excluded from sample before analysis: 18 lines (Eric Idle, River Idle; 4 incorrectly tagged verbal usages: e.g. “idle their time away”; 3 uses with nominal function: e.g. “the idle and the profligate”; 1 typo: “He was his idle”). Sense/ subsense discovered 1a30: not doing work (or: not being used)

(TYPE(s) OF) nominal collocate PERSON GROUP DEVICE VEHICLE WORKPLACE LAND MONEY

1b: unwilling to do work bone idle 1c: during which little 30

rich PERSON GROUP PERSON/GROUP

TIME PERIOD (time)

Examples …hating to be idle, he took part-time caretaking work…/ …Wolfe and I had been idle for a long time./ …the Government should not be lying idle…/ …while staff at a unit sat idle…/ …the contract-hire machines had been idle for more than half a day…/ …batteries which are left idle for long periods. At other times buses and trains are idle./ …an idle building site… / …to

run the idle plant…/ …much farm land is lying idle under the EC's set- aside scheme./ …the OPEC countries found themselves with huge amounts of idle funds due to increased revenue from oil./ The world divided into the idle rich and the labouring poor./ Sometimes, when I'm feeling really idle…/ You idle bastard," Hector told him./ Or are British clubs too idle to look in their own backyard…/ …already quite enough of the bone-idle shysters for my liking./ …they don't have idle time on their hands…/ …the micro-processor's idle time…/ ( more technical sense) …other

General sense, of which 1b-d may be seen as specializations.

29

is done 1d: not accruing interest 2a: irresponsible or illconsidered

(extension)

(hours) MONEY SPEECH ACT

speculation (x5) patronising / ex. (curiosity) PERSON

It [would be] idle to [pretend] that… 2b: unreliable or sloppy 2c: trivial, not important, or futile

-

(function as manner adverb) 2d: not realistic

patterns

2e: (function as intensifier) (poetic)

SPEECH ACT (threat) matter ACTIVITY ACTION

IMAGINATION

cause (various)

-

proles spend their idle hours tanking up at a grubby downtown bar…/ …the idle and easy lives of the clergy./ …there is a cost to holding idle cash balances…/ …probably never intended as a serious challenge, only an idle remark…/ …to suggest that claim is more than idle gossip./ …no longer just the idle chatter of Westminster minorities…/ He had said it before, and it was no idle compliment;/ …to protect the group from `idle speculation and harmful innuendo"./ No idle patronising going on here./ …this was no idle exercise./ Call it idle curiosity on my part./ But all week idle gossips suggested the President would be at her right hand, so to speak./ It would be idle to pretend the system is perfect./ It is idle to imagine that, in a world devoid of tariffs and trade barriers, disagreements, and indeed wars, would not arise./ This is no idle boast, for I am something of a connoisseur…/ … they considered this was no idle threat./ My reference to Ovid wasn't idle, but…/ These things are not idle matters…/ …the idle pleasures of these days./ It's not just an idle thing that we do on a Saturday afternoon…/ …merely to catalogue the host satellites would be as idle as the cataloguing of telescopes in, say, Herschel's day./ Fancy made idle patterns in the water with her hand; she looked young and ingenuous to Jewel as she spoke,…/ …peace is not an idle daydream,…/ to believe that what you wish and expect to happen is not just an idle dream./ Mine’s not an idle cause./ ( quote from Shakespeare) But Swans are no idle playthings./ ( here Swan = type of boat) The equation of Eden with maternal nurturance was not an idle conceit or a slip of the pen;…/ …an idle luxury, a waste of valuable time and energy./ Wars are not begun out of casual caprice or idle fancy,…/ In ether rolls the idle ball, Its firmness we doubt not at all,…

============================================================== (Borderline adverbial usages (by verbal collocate))

lie sit stand leave stay / remain make / feel go / spend live

…much farm land is lying idle under the EC's set- aside scheme./ …Jan Eriksson said the Government should not be lying idle./ Professionals with no time frame to complete the transaction keep sitting idle and draw handsome salaries./Tractors sit idle. /Why do NHS operating theatres stand idle while patients wait…/ …if you leave it idle today yes you can use it tomorrow but…/ Investors Stay Idle to Avoid Mistakes…/ …his army remained idle in Boston./ …fretted that the ever-growing numbers of foreigners made Kuwaitis incorrigibly idle…/ Sometimes, when I'm feeling really idle…/ …to let a field go idle…/ …Florida Pearl had spent a full month idle in his box…/ …those who thought it of more advantage to the world to live idle than by labor.”/

4.5.1 Dictionary evaluation: marginal Marginal presents a complex picture of multiple senses. Some relate in a variety of ways to various senses of margin, and many are technical. It is thus similar to canonical in many ways. ODE, CED and CTD each begin or end with a kind of ‘catchall’ sense to cover all possibilities of its use as a relative adjective (CED s.1 is the most general, whilst ODE s.1, “relating to or at the edge or margin” is syntactically flawed). However, it has many descriptive uses too, and these are significantly more frequent. CTD and ODE give good coverage, lacking only an adequate representation of COBUILD’s s.2. NOTE’s entry is very short, perhaps a reflection of the technical nature 30

of the other senses, giving them few or no synonyms. In particular note the difficulties of the technical financial usages: there is some disagreement between ODE and CED over their essential meaning. The results of analysis indicate many very frequent collocates of a financial nature, which seem likely to be technical terms. But of these, only ‘marginal cost’ is directly dealt with in separate entries, and that only by ODE and CTD.

4.5.2 Summary of sense distinctions: marginal (sense) §4.5.3 CED slight, very small not important or relevant powerless (people) won by few votes (constituency) borderline, near lower limit at margin of profitability of small unit changes (cost) of increases in income (tax) at margin of cultivation (land) on water’s edge written in margin of page otherwise related to ‘margin’ ‘marginal cost’ ‘marginal costing’

2a 3a 3b 2b 2c, (3e) 1

ODE

CTD

3

2

1

5 2 4 6 -

3:sub1 3 2:sub1 2:sub2 3:sub2 1:sub

1

1

s.e.

s.e. -

3 2 4 5 s.e. -

7

3d (5) 4 (3c), (6?) (under 1) -

MED 1 2 3 4 -

COB .

NOTE

1

1

2 3 -

2 -

4 -

4.5.3 Results of corpus data: marginal Total frequency in corpus: 5,392 Tagged as adjectives: 5,392 Sample size: 200 lines (all tagged as adjectives) Strongest noun collocates: cost, rate, tax, revenue, seats, product, curve, rates, seat, utility, costs, labor. Excluded from sample before analysis: 5 lines (4 noun usages (“his four-way marginal at the last election”); 1 line with crucial text missing). Sense/ subsense discovered 1: (technical financial usages) ? small-scale / ?relating to profit margin

(TYPE(s) OF) nominal collocate cost (x24) costs (x5) rate (x7) tax rate (x7) product (x10) revenue (x8) utility (x5) benefit propensity means investment

Examples Price is equal to the firm's marginal cost at an output rate of 30,000 units per month./ …the marginal cost of supplying water is small…/ We usually show a single marginal cost curve for a firm…/ (5 occurrences of ‘cost curve’) …implies that average costs are constant and equal to marginal costs./ …their relative marginal operating costs…/ …the marginal rate of substitution…/ That will leave Britons paying the lowest marginal rate…/ …a marginal tax rate of 28%…/ (5 x ‘tax rate’) …equating the value of the input's marginal product to the input's marginal factor cost…/ …the marginal product of labor (MPL)…/ (6 x ‘product of labor’) …its marginal revenue is $180…/ …the MARGINAL UTILITY theory of value…/ …a point at which marginal benefit equals marginal cost…/ …the marginal propensity to consume all goods, including imports, is 0.8…/ Plot the cell means on a graph and calculate marginal means/ …the marginal investment in the second machine…/

31

2a: on a small scale, slight, or incremental

CHANGE

lead STATE

2b: closerun, almost equal

2c: difficult to judge, borderline 3a: less important, sidelined (= ‘of marginal (2a) importance)

possibility scale profits problem villain seat (x5) seats (x9) ward constituency electorate [tennis match] decision category point voter PERSON/ GROUP STATE

ACTIVITY

3b: minority

3c: barely surviving (=on the margins) 3d: barely fertile

place PERSON/ GROUP perspective position living operations mines people ENVIRONMENT

3e: exceptional 4: written in margin 5: located on the edges

case

6: (zoological term) (?)

series

? (= not reliable?)

clutch

WRITING plants weed shelf

cell

In contrast to these abrupt and fundamental changes, the German experience has been one of continual marginal adjustment./ …a marginal improvement of 1.16 points…/ …a marginal recovery, adding 4p to 620p…/ …marginal improvements matter./ …with a marginal increase for those in favour./ …marginal increases in class sizes harms standards./ …the decline was marginal…/ …enough to give the American a marginal lead going into today's free programme./ …this proved to be of increasingly marginal relevance…/…Wigan's marginal superiority in the first half./ … religious movements have only occasional and marginal importance in developed nations…/ …a marginal possibility that Labour could lose the seat…/ …on a small, even marginal scale./ The profits in this business are too marginal…/ …the problem is marginal but does need to be taken seriously./ …with muse and marginal East End villain, George Dyer./ …the marginal seat of Lowe in NSW…/ …the party's most marginal target seat…/ Labour also did badly in marginal seats…/ … marginal Tory-held seats…/ …marginal wards…/ …voters in marginal constituencies…/ …the Tory faithful in marginal Labour and Liberal Democrat constituencies…/ Queensland now has 17 marginal electorates…/ …it will be marginal -a point here, a piece of luck there…/ This alone made the decision to contract out more marginal."/ These were marginal decisions in a game that ebbed and flowed…/ Both of us fall into that marginal category./ …one good point and one very marginal one./ The marginal voter is clearly the person that we're trying to reach with this outreach program…/ …Alen Boksic came good until the Europe final, when he was strangely marginal…/ …with other parties such as the United Nations and the European Community remaining marginal players…/ …private governments reduce the official ones to a largely marginal and ceremonial existence./ …only a marginal interest in the development and strengthening of equal opportunity…/ `What is perceived as marginal at any given time depends on the position one occupies."/ …a `marginal and very small-scale operation'…/ …the marginal, part-time jobs generally done by women…/ …feminist therapy continues to be marginal to most of the mental health disciplines…/ …their resentment at the marginal place they have been allotted in the scheme of things…/ These marginal groups…/ …really only a marginal candidate…/ …a marginal party without the geographical concentration to elect more than the odd candidate…/ …poets whose cultures were for a long time regarded as 'marginal…/ …story from the marginal perspective of a eunuch lover./ The marginal position of women, especially black or working-class women,/ …many are unemployed or must eke out a marginal living as street vendors or in the `underground"…/ They are extremely marginal operations and simply cannot tolerate the slightest cut to their resource allocation/ …some of Anglo's marginal mines will become loss-making…/ It is not that weak, marginal people are witches…/ …contemporary hunter-gatherers exist in marginal environments…/ …upland areas with thin soils and harsh climates may be considered marginal…/ …in many of the marginal farming areas./ One enemy of simplicity is the marginal case. A case may be marginal because the draftsman failed to foresee it./ The stamps will be printed against a beige background and the marginal inscriptions will be black./ …marginal notes…/ Remove dead foliage from marginal (poolside) plants and cut down the remaining growth…/ …float fishing alongside marginal weed in summer./ …retrieve it back up a steep marginal shelf./(underwater) …the marginal continental shelves…/ The marginal series is dominated by the sixth marginal which is ventrally expanded as in Basilemys and Nanhsiungchelys./ …a narrow costal cell that is not divided by a longitudinal intercostal vein, 4 submarginal cells, and 2 marginal cells./ I stalled at the start, just like Goodwood, because the clutch in those cars was very marginal./

32

?

owner

?

worker

? ?

revolution differentiation

…particular sections of the population (the elderly, marginal owners, the homeless)…/ …the rational marginal worker will withdraw labor with the expected higher price level cutting into his real wage…/ …a marginal workers strategy…/ The question of the marginal revolution of the 1870s…/ They, in turn, are associated with rational strategies for increasing the marginal differentiation of religious `products"/

4.6.1 Dictionary evaluation: particular Particular is a tricky and awkward word to analyse. There is more to it than meets the eye, and more differences between the dictionaries’ treatments of it than are immediately apparent. One problem in analysing its different senses is frequency. In the corpus its usage as postdeterminer is by far the most frequent, followed by phrasal ‘in particular’, then the ‘significant’ sense and only a sprinkling of other senses. My rather basic methodology was insufficient here, and even an additional sample was of limited help31. The problem with the dominant sense here is that, although universally classed as an adjective, it is really a kind of function word. In the example ‘a particular historical generation’, particular perhaps has more in common functionally with a than with historical. As seen in my s.1, the type of nominal collocate really has no bearing on its sense here. It is treated as a function word by COBUILD, MED, and rather refreshingly, even by ODE, with an explanation of its function rather than an attempt at anything like a substitutable definition. As is seen in CTD, and especially in CED, that attempt can end very clumsily. CED is alone here in having three very technical senses (s.5, s.6, and s.7), none attested in our sample. Only its sense in the field of logic is included by ODE. Neither gives an example of usage here, but CED interestingly gives an example to illustrate the concept. It seems odd that ODE has placed this sense alone under the main functional sense in its hierarchy, especially in view of its unusually restricted coverage otherwise in this entry. CTD, here typically concise and implicit, is also notable for some slightly bizarre examples. ‘He’s very particular about the washing-up’ is a little strange out of context, and ‘took particular care’ looks very odd without a subject. If short examples are needed, they should be better chosen. MED has a very clear and full explanation for s.3 especially, and makes an interesting distinction (3a) of ‘I’m not particular’ in spoken English. Most interesting, though, is its inclusion of a list of strong collocates at the end. These are presumably the most significant nominal collocates, but the precise reason for their inclusion here is not quite clear, especially since the concordance sample for particular was the only one of the nine samples to have no noun collocate with a frequency of 5 or more. 4.6.2 Summary of sense distinctions: particular (sense) §4.6.3 CED (postdeterminer function) distinct, unique especially intense, marked

1 4

+ examples in ‘borderline cases’

2

31

ODE

CTD

1

1

1

2

2

2

MED 1 4 2

COB . 1 2 3

There may be other methods that would be of use here. For example, noting that many minority uses are graded, we could search for particular immediately preceded by an adverb. This yields 490 results, mostly of ‘very particular’, but many of these seem relevant. More examples of the ‘fussy’ sense appear.

33

NOTE 1 2

fussy, difficult to satisfy detailed, meticulous affirming for class (logic) (differential equations sense) ‘particular average’ ‘particular solution’ ‘Particular Baptist’

3 -

4 3 6 5 s.e. -

3 1:sub s.e. s.e.

3 4 -

3, 3a -

4 -

3 4 -

4.6.3 Results of corpus data: particular Total frequency in corpus: 61,939 Tagged as adjectives: 61,286 Sample size: 200 lines (all tagged as adjectives) (+an additional 200 lines32) Strongest noun collocates: interest, attention, area, case, type, problem, circumstances, concern, problems, group, groups, needs. Excluded from sample before analysis: 66 lines (64 instances of in particular (fixed expression with particular as noun: e.g. “Japan in particular and Asia in general”); 1 noun (“the most exciting particular was that she had been re-engined in 1973”). Sense/ subsense discovered 1: (usage as postdeterminer: function to make reference specific rather than general)

(TYPE(s) OF)

nominal collocate THING (software) TIME PLACE number PERSON GROUP EVENT CAUSE/ EFFECT SUBJECT ASSET OPINION sound/ shape/ disease soil/ method relations/ form

industries case FEATURE GROUPING VARIETY way choice role phraseology word shortage/ use 32

Examples …how a particular boat will behave in rough weather…/ Any particular software…/ …a particular program…/ …living through a particular time…/ a particular historical generation…/ …in this particular instance…/ …your particular school…/ … in a particular pub or hotel…/ …the particular localities' problems…/ …to live at a particular address…/ …that particular part of the province…/ …their own particular area of interest…/ …a particular number of slots…/ This particular bloke…/ …a particular science teacher…/ …support by particular interest groups…/ …a particular interpretive community…/ …a particular organization…/ …this particular `crime"…/ …my own particular roadshow…/ …this particular incident…/ …that particular game…/ …one particular reason…/ …a particular offensive economic objective or effect…/ One particular implication…/ …their own particular art…/ …a particular discipline…/ …in any particular field…/ …particular things that interest me…/ …a particular account or fund…/ …any particular stock index option…/…essential in the production of a particular commodity./ …whether marketing management should adopt particular stances…/ …the particular beliefs we hold…/ …create a particular sound…/ … a particular navel shape…/ …a particular autoimmune disease…/ … suitable for your particular soil…/ …a particular method…/ … clearly tied to particular power relations…/ Should my writing take a particular form?/ …special knowledge of particular industries,…/ …in this particular case…/ …relating to your particular case…/ …human breeders selecting for particular traits…/ particular aspects of the Library and its collections…/ …this particular batch of currency…/ …one particular set of institutions…/ …its particular variety…/ a particular kind of renovation in our cities…/ Their observations have a particular flavor…/ …played a particular style…/ …approaching your project in a particular way./ …to respond in particular ways to objects. / …a particular choice of map coordinates…/ …perform well in a particular role…/ …the particular phraseology of the Constitution…/ I do feel the inclusion of one particular word in letters and articles to be somewhat offensive./ …compensate for their particular shortages./ …furniture that had no particular use…/ …no particular casualty…/

A second sample to search for minority senses, analysed in less detail. Results are those in italics.

34

(borderline cases) 1 or 2

2: marked, significant

casualty (various)

QUALITY following influence success favourite interest/weight

3: fussy (?) 4: unique or distinctive

PERSON (bonds) identity history &c.

? (incoherent; spoken)

…rarely treated with any particular prurience./ …no particular signature of any nation…/ …he was in no particular hurry…/…no particular rationale…/ …things I made particular notes about…/ … particular citations in the text to be looked up later./ …the particular disabilities of the British labour movement…/ …do you have any other particular dislikes…/ There is one particular reason why it was tough…/ …which have particular charm of their own./ This is of particular importance during the breeding season…/ This strategy has particular relevance…/ …society has a particular interest in punishing hate crimes…/ …she has a particular following among boys…/ They had a particular influence on the fourteenth-century mystical movements…/ …a particular success for the right bank of the Gironde…/ My particular favourite…/ …I read with particular interest your editorial…/…His words carried particular weight…/ Power and capacity for the most particular baker./ …differing levels of supply and demand for municipal bonds particular to the State of New Jersey./ …Scotland acquired a more distinctive, more particular identity for her…/ Only France has a history, and a sense of independence, as particular as Britain's./ erm your problem customers. Yes. I mean I have my particular does. Yeah./

4.7.1 Dictionary evaluation: remote The basic notion of (long) distance is common to all the senses of remote, whether it is a spatial, temporal, emotional or a metaphorical distance designated (this last category accounting for a high proportion of the usages, as seen in 4.7.3). One could well argue that this is actually a monosemous word, with different readings arising purely from differences of context and collocation; Moon’s idea of ‘quasi-monosemous’ words (Moon 1987a) would apply particularly well to remote. In CED, as in 4.7.3, the sense ‘located far away’ is put first, perhaps usefully, for logical coherence. However there was only one example of this in a non-metaphorical sense in the sample, and interestingly neither ODE nor COBUILD include it. ODE has also, once again, sullied its logical order by putting the ‘operated from a distance’ sense as a subsense of the (dominant) ‘far from civilization’ sense. CED once again has the widest coverage of senses, omitting only the ‘accessed via network’ sense (ODE s.5; CTD s.4), and the ‘unconnected with’ sense (although this usage may be hinted at by its s.4, a sense not found in our sample). CTD is again logically ordered and not unsuitably concise, although it gives no examples at all except for s.6, where the sense is absolutely dependent on collocation with chance or a related word. MED again has a useful guide at the top of its entry, a list of senses, and gives clear explanations of all the important senses. COBUILD has very clear and specific definitions, especially for s.1, s.4 and s.5, but in s.3 the definition does not seem to fit with the example, which would go better with s.5. NOTE again gives an impressive range of synonyms, especially for s.2, where many regional variants from Australian, North American and South African English are also included, and it does not balk at including the humorous formation ‘unget-at-able’ as an informal synonym. Its “in the depths of —” is slightly ambiguous, but presumably refers to expressions such as ‘in the depths of the jungle’ or ‘in the depths of nowhere’.

35

4.7.2 Summary of sense distinctions: remote (sense) §4.7.3 CE D far from civilization 1b 2 distant (in space) 1a 1 distant (in time) (ex. in 1a) 3 unconnected with (‘remote from’) (ex. in 1a) (4?) unlikely 3 6 aloof 4 7 operated from a distance 2b 8 distantly related 4 accessed via a network 2c removed from point of action 2a 5 ‘remote control’ under 2b s.e. ‘remote sensing’ under 2b s.e. ‘remote access’ under 2a s.e. ‘remote sensor’ s.e. ‘remote job entry’ ‘remote handling equipment’ -

ODE

CTD

MED

COB.

NOTE

1 1:sub2 2 3 4 1:sub1 2:sub 5 s.e. s.e. -

2

1 1a 1b 5 2 3 4 s.e. s.e. -

1 2 3 4 5 s.e. s.e. s.e. -

2 1 3 4 5 -

1 (5?) 6 7 3 5 4 s.e. s.e. s.e. s.e. s.e.

4.7.3 Results of corpus data: remote Total frequency in corpus: 12,116 Tagged as adjectives: 12,116 Sample size: 200 lines (all tagged as adjectives) Strongest noun collocates: control, areas, island, area, village, [TV], communities, sensing, parts, possibility, region, villages. Excluded from sample before analysis: 11 lines (6 noun usages: 5 where ‘remote’ = ‘remote control handset/device’ and 1 other = placement: (“Caracas was to be its last remote…”); 5 instances of “remote-controlled” (adverbial). Sense/ subsense discovered 1a: far off metaphorical extensions

(TYPE(s) OF) nominal collocate table goal time language corner (WEBSITE) commissions

1b: far from (highly populated) civilization / (difficult to get to)

PLACE (areas x5) (area) / (part)

GEOGRAPHICAL FEATURE

Examples Gerhardt found a seat at a remote table…/ …donating… to some goal as altruistic as it was remote./ Proconsul,… was too different from ourselves and lived at too remote a time to really capture the public's imagination./ …the language of the Opposition in eighteenth-century Britain strikes many historians as especially remote from reality./ …what dark, remote corners of the Net are young teenagers able to explore?/ …commissions often lacked the necessary funds to perform their tasks well or were too remote from the operating problems of line agencies to provide the positive assistance they required./ …remote places such as the Scottish highlands…/ Nowhere so remote that WACH didn't want a piece of it./ …one of the truly remote and wild parts of Africa./ …remote upland areas of Morocco…/ …remote areas of the developing world./ …a remote mountainous area…/ …the remote Mustang region…/ …a remote part of the Amazon region…/ …this unbelievably remote setting…/ … remote or developing countries./ Using someone's else's continent, however remote or unpopulated, betrays an arrogance of colonial character./ …a more remote territory…/ …the country's remote north-east,…/ …a remote Scots hillside./ …the remote Panjshir Valley,…/ …the remote mountain scrublands of Mali,…/ …a remote habitat of dry forest and coastal wetland./ …a remote wilderness…/ …this remote

36

SETTLEMENT (village) BUILDING

=remote part(s) of who are in remote places

1c: far apart (metaphorical extensions) ?far from what is familiar 2a: not onsite / at another location 2b: done / controlled from a distance 2c: over the Web / telephone ? 3: negligible or small

lane phone box residency Australia PERSON GROUP (communities) categories idea language ELECTRONICDEVICE ELECTRONICSYSTEM access DEVICE (control x21) PROCEDURE SOFTWARESYSTEM (?) tarot reader tank socket POSSIBILITY

considerations need 4a: showing no emotion / withdrawn

PERSON

smile wedding 4b33: not close emotionally 33

PERSON

archipelago…/ … a remote gorge in north-eastern Portugal…/ …a remote, rural village near Kongoussi…/ …the remote Andean village of Cabana…/ …the remote town of Katima Mulilo…/…remote farms and even more remote Indian settlements./ …living in a remote farmhouse cellar…/ their remote home in Furnace, Argyll…/ But they are now insisting the refugees are held in a remote camp site./ …isolated in his remote castle…/ …solar power was installed at a remote health clinic…/ …even into the remote, unpaved country lanes./ A REMOTE phone box in the middle of a scorching wasteland…/ …any adverse effect of remote residency would be identified…/ …throughout remote Australia./ `And we really want to give people in regional and remote Australia equal opportunity."/ …the rural and remote nurses incentive scheme…/ …undernourished remote others of a minimally human existence./ …museum explorations have drawn previously `remote" societies into the international system…/ …a remote settler society…/ …to ensure liquor licensing laws are taken more seriously in remote communities./ …remote Aboriginal communities…/ Sometimes these categories are within easy reach of our own,… Sometimes they are considerably more remote./ …creativity implies an associative gradient that slopes gently to give access to large numbers of remote ideas,/ Psyche's fourth task is the most important and most difficult of all. Few women reach this stage of development and its language may seem strange and remote./ …their remote camera… searching for the remains of the Titanic./ …to link the camera to the remote monitor./ …a remote command system located on another barge…/ …the `remote monitoring system" Unscom used to relay video pictures…/ …remote access to medical and social care…/ …remote central locking…/ …network TV remote relays…/ …remote control…/ (=handset: 18 occurrences) …remote control unit…/ …a remote control telephone…/ …remote control racer…/ …remote reprogramming and maintenance…/ …the Center for Remote Sensing…/ …remote viewing…/ …remote-control microsurgery…/ …The client provides a remote interface for the job manager…/ … the small office that uses the X.25 protocol to communicate with remote sites…/ … Intellivoice Remote Agent software…/ …Remote and synchroniser interface,…/ …`remote tarot readers" - tarot readers over the telephone…/ …the absence of a remote-tank socket…/ …they have only a remote chance of becoming infected with HIV…/ …the prospect of system change resulting from war remains remote./ there always remains the possibility, however remote, of mistaken identity…/ His golf was always a spectacle of the possible, however remote, rather than the probable./ Such considerations, however, still seemed remote, if not irrelevant…/ …the need for political and military conflict seems so unnecessary and remote./ …she is very beautiful but very remote./ Outwardly, she has always seemed as remote and forbidding as that apartment building,…/ Some say that Andy Roberts is a bit remote, but he loves to talk cricket,…/ …a remote, undemonstrative family…/ …the coldly remote… senior officers of the old school…/ …both Federico and Battista are purposefully stylized and iconlike in order to appear as commanding, remote images…/ On his lips there was the fixed, remote smile to be seen on all Gothic carvings…/ She remembers watching the wedding of the Prince and Princess of Wales and thinking it was all very remote and unreal./ …he could hardly have been more remote from the Forte minions and was not greatly mourned./ Remote adults could be irked at little emotional cost, but parents were (usually) the two most beloved and powerful people in a boy's life./

Probably belongs with 1a metaphorical extensions.

37

4.8.1 Dictionary evaluation: stable At the analysis stage, grouping senses was rather difficult here. The gloss ‘not subject to change’ covers all readings, with the distinctions being subtle. Perhaps the problem is the metaphorical basis of most readings. The meaning is based on a concrete concept, but there are very few concrete usages in the sample, and the emphases of the metaphoric extensions vary: strength, unlikelihood of toppling, unlikelihood of collapsing, durablility, or safety; but these can easily overlap. Also, the change something is not subject to may be of an economic, political, chemical, medical, or relational kind (or otherwise), and this at first seems an obvious way of delineating senses, but is an unhelpful one. This situation is best represented by ODE’s single sense with four subsenses. CED is very precise in its technical distinctions (s.4, s.5 and s.6), but oddly omits two important senses. Also, NOTE s.3 is based on two examples rather than one, but the usages have so much in common that the long list of synonyms is not easily divided.

4.8.2 Summary of sense distinctions: stable (sense) §4.8.3

CED

ODE

CTD

MED

not likely to fall over, steady mentally sound, sensible lasting, safe (e.g. relationship) established, safe (e.g. government) unwavering (e.g. prices) not deteriorating (health condition) not undergoing decay (particle) unchanging chemically (compound) not self-oscillating (electronics) ‘stable equilibrium’

1 3(?) 2 (2) 4 5 6 -

1 1:sub2

1 3b 2b 2a

1c 1h 1b, 1d 1d, 1g 1a 1f, 1a 1e -

1:sub3 1:sub1 1:sub4 s.e.

3a -

NOTE

2 3

COB . 5 2

1

1

3

1a

3

4

4

-

-

-

1 2

4.8.3 Results of corpus data: stable Total frequency in corpus: 14,125 Tagged as adjectives: 12,095 Sample size: 200 lines (all tagged as adjectives) Strongest noun collocates: condition, prices, relationship, environment, value, [hospital], form, companion, prospects, government, relationships, inflation. Excluded from sample before analysis: 40 lines (all (homonymous) noun usages). Sense/ subsense discovered

(TYPE(s) OF) nominal collocate 1: Not subject to change 1a: invariable, VALUE not fluctuating (level) FORM STATE

Examples

…customers would prefer stable prices./ …a stable net asset value./ …a stable exchange rate…/ …keen to keep brand equity values stable…/ …a high stable level of income…/ …demand remained stable./ If IQ tests do not produce very stable scores…/ …the infection among injectors is… about 7 per cent and stable."/ …a stable temperature…/ …stable formats…/ …his search for a stable identity…/ His condition was listed as stable./ …in a serious but stable condition in a hospital

38

line-up way ACTIVITY / ACT JOB PEOPLE 1b: free from upheaval, safe

RELATIONSHIP

relp (x8) STATE GROUP / PERSON [soldiers] INSTITUTION / CIVILIZATION

PLACE ENVIRONMENT

TIME/PERIOD

1c: physically safe from external forces 1d: which will not disappear or end

car SUPPLY influence career

1e: which will not be changed in form

SUBSTANCE

[CFCs] [Sun] creature

1f: out of danger, in stable condition

area PERSON

1g: wellordered and reliable

SYSTEM /

1h: selfcontrolled or sane ?

PERSON

INSTITUTION

kidney unit…/ …on a stable financial footing…/ …this can never represent a very stable state of affairs./ …the situation in Panama is stable enough…/ …the line-up as it is today is very stable./ …some stable, enduring way in which we can characterize an individual…/ …stable activity, above 50 growth…/ …stable economic growth…/ …reach a stable decision./ … to boost the satellite so that it regained a stable orbit./ … workers in central firms receive high wages and stable employment…/ …extremely stable work histories…/ …are we sufficiently stable as subjects to engage in public discourse…/ …business woman, looking for a stable relationship…/ …the one stable relationship in her life…/ For years this relationship seemed stable enough to rely on for policy purposes,…/ …a stable marriage…/ …stable adulthood…/ …their top squad has remained so stable, with few injuries and suspensions./ …to make sure our community stays stable./ …`strong and stable" families…/ …or at least a stable parent substitute./ …indeed, you are stable if you're in bankruptcy…/ …whether they're stable in that--in that sort of mission./ The Ohlone culture remained stable until the Spaniards came in 1769./ …how to turn poor, undemocratic multi-ethnic states into stable, democratic ones, without bloodshed or a tyranny of the ethnic majority,…/ …a strong, stable and prosperous China./ …a stable Europe…/ Lhasa is stable."/ …the north-west is largely stable…/ … somewhere that was stable but still entrepreneurial./ …a stable place for a young child./ …the more stable environment provided by a traditional two- parent family./ …children from stable homes…/ …the more stable world of The Brady Bunch."/ …stable times…/ …the relatively stable 1970s./ …the brief and unusually stable postwar period…/ It felt very stable, even in a gale, although it did feel very bumpy on rough country roads./ …the Volvo 480 is spirited yet assured, sophisticated yet stable, sporty yet safe./ Britain has had stable food supplies for centuries./ …its light output remains stable over long periods of time./ …with a group of stable clients,…/ …we look for a stable support from the United Kingdom./ He is a stable influence…/ …what Billy gained from giving up his stable career was anyone's guess./ …if the substrate is stable and hasn't been coated with any other type of varnish…/ Non-toxic, stable, they came to be used in dozens of industrial products…/ …the stable, nonradioactive isotope carbon-13./ …the water remains clear and stable for extended periods./ …it has remained stable for billions of years…/ …the common loon, this stable creature with 60 million years of evolutionary fine-tuning…/ …in geological terms, the area is as stable as you're likely to get…/ `He's stable now," mccracken said of his son. `He's out of any danger…/ A girl aged nine in the same hospital is stable./ They were all `stable" in hospital last night./ New mothers are not 100 per cent stable healthwise for about three to five months after giving birth…/ Nor is the system likely to be financially stable./ …a stable economy…/ `It's faster and more stable…( said of I.T. Operating System)/ …perceived by the City as a stable business./ …a more stable and sensible political system./ …some form of stable political constitution…/ …Italy had a stable government…/ …the stable regime instituted by Frank Clark…/ Let us hope that the new management of the Academy is stable enough to award her her Doctorate./ …establish a stable framework for financial planning…/ …the region's most stable equity market…/ …society has become more stable, with more people owning more property…/ The war that's been set up has given us space for some notvery-stable people to do whatever they want…/ …helped me to become the stable and well-adjusted creature I am today./ …despite John's aim of designing-in stable understeer,…/

39

4.9.1 Dictionary evaluation: unbalanced It is an interesting quirk of this random sample that the last two adjectives should be antonyms, albeit indirect antonyms. As a result the general picture of sense distinctions is similar to that for stable, but due to the effect of collocation, the frequencies of corresponding opposite senses vary. Notably, the ‘deranged’ sense of unbalanced is very common, and is the first sense in all except CED and CTD, whereas the corresponding usage of stable to mean ‘sane’ is only a minor sense. NOTE here has a particularly long list of synonyms for this sense, mental disorder being a productive area for the formation of euphemisms and slang. Otherwise the dictionaries deal very briefly with this word, perhaps partly due to its low frequency, but also because it is based on a verb (cf. 4.3.1 on floating) and an affix. Essential information can thus be drawn by comparison with the entries for unbalance, balanced, and balance, with only unpredictable usages needing to be described. However this does mean that common usages such as “unbalanced diet” are not covered by any of these entries, except vaguely by CED’s s.1, “lacking balance”.

4.9.2 Summary of sense distinctions: unbalanced (sense) §4.9.3 CE ODE D deranged, disturbed 1g 3 1 one-sided, biased 1a, 4 2 irrational, erratic 1g, 1e 2 lacking balance 1f, 1c, (1b) 1 (bookkeeping sense) 5 (electronics sense) 1d 6 -

CTD

MED

COB.

NOTE

2 3 1 4 -

1 2 -

1 2 -

1 2 -

4.9.3 Results of corpus data: unbalanced Total frequency in corpus: 859 Tagged as adjectives: 724 Sample size: 100 lines (all tagged as adjectives) Strongest noun collocates: diet, growth, path, economy, force, view, mind, relationship, split, system, analogue, stereo. Excluded from sample before analysis: 1 line (verbal usage). Sense/ subsense discovered 1a: onesided, not fair

(TYPE(s) OF) nominal collocate DISCOURSE/ IDEOLOGY

view POLICY RELATIONSHIP

1b: of uneven composition

way situation SYSTEM ratio ?split

Examples

The FDA promptly censured it for unbalanced information./ …the inaccurate and unbalanced article written by your journalist./ … facts which the judge had chosen to ignore in his unbalanced summing-up./ …an inordinate and unbalanced anti-Americanism…/ … presenting an unbalanced picture of public life…/ …just how unbalanced our view of animals is,…/ …a similarly unbalanced policy…/ …a `purely Nato" alternative to the peace deal would be `unbalanced"./ …put pressure on France to sign an 'unbalanced' deal./ …the relationship between France and West Germany was unbalanced…/ …if… a sexual relationship is narrow and/or unbalanced…/ …if he lived his life in an unbalanced way…/ …the unbalanced military situation in Bosnia…/ …an unbalanced economy searching for equilibrium./ …an unbalanced budget…/ Societies could become dangerously unbalanced./ …an unbalanced mix…/ Some unbalanced social profiles are not the result of prejudice./ …studies that correlate unbalanced sex ratios with active warfare./ …Unbalanced split…/ …combinations of

40

colours [sculpture] dance diet (x7) teaching INSTITUTION/

TEAM 1c: not countered by its opposite, unchecked 1d: ? (technical) 1e: (?) variable 1f: unable to stay up 1g: not selfcontrolled, deranged

force pressure liquid crystal boom growth path output circuits beans tree rod PERSON position PERSON

ASPECT OF PERSON letter role

unbalanced colours…/ …there is no such thing as am unaesthetic or unbalanced Anish Kapoor piece./ …so that the dance does not become unbalanced by either the yin or yang energies./ …an unbalanced diet, excessively high in sugars./ 'The teaching is unbalanced,' says Smithers. 'Forty per cent of those teaching combined science are biologists, but only 10 per cent are physicists.'/ …a biased and unbalanced House of Lords…/ …an unbalanced midfield…/ …Ingesson's lack of match-fitness left them unbalanced on the left./ …the back row is still unbalanced./ …an unbalanced force acting on a body will accelerate that body./ …the problem of unbalanced pressures./ The liquid crystals used in displays are electrically unbalanced…/ …the unbalanced and unsustainable Lawson boom of the late 1980s…/ …the unbalanced growth path of a closed economy,…/ …with internal battery power, and unbalanced outputs./ …balanced and unbalanced analogue i/o,/ …unbalanced to balanced circuits…/ …all the beans and grains contain all the amino acids but, from the body's point of view, they are rather unbalanced…/ …any size or shape of tree, however unbalanced…/ Always position the reel at the top, otherwise the rod will feel unwieldy and unbalanced./ He did get a bit unbalanced that day, on the brow of the Dip…/ …when the head is held in an unbalanced position…/ She was unbalanced and this fear became an obsession./ He had a very unhealthy interest in young girls -- he was definitely unbalanced."/ I wasn't mentally unbalanced, I knew what I was doing./ …suggesting that she was mentally unbalanced…/ …the work of a mentally unbalanced individual…/ I always believed there was something profoundly unbalanced in Martin./ …the result of an unbalanced mind./ When your life is unbalanced, as mine was…/ …it was a violent, unbalanced letter… Its manner,… made me wonder whether the man were not off his head./ Those eyes seemed to have led him to play unbalanced roles…/

41

5

Conclusion

This analysis has been aimed at singling out a very specific type of word in order to give a detailed review of how well dictionaries deal with it, and doing so on the basis of the dictionaries’ own classification. Yet adjectives as a word class form a fairly heterogeneous set. Singling out polysemous adjectives is in some ways helpful, but then again dealing with adjectival senses of a word together with non-adjectival senses may also be desirable, possibly even where the word form varies slightly, where those senses are very similar and inter-dependent. COBUILD’s treatment of adverbs under many sense numbers dedicated to adjectives is a good illustration. For a certain word used as an adjective, its non-adjectival usages form a part of the word’s whole polysemy, where word class is only one step in a syntactic separation of senses. An investigation into a more tightly defined linguistic class, such as purely predicative adjectives (cf. Gross et al. 1989: 93), or purely descriptive adjectives, may have yielded clearer results. This highlights the vagueness of dictionary classification, and more precise word classification, or a re-evaluation of word classes, may be desirable within at least some general dictionaries. And yet mixed as our sample was, there are whole grammatical classes34 of (possibly polysemous) adjectives it did not cover: evaluative adjectives (e.g. good, nice, poor), time-related adjectives (new, old, early), colour adjectives (red, black), or topical adjectives (environmental, sexual, linguistic). The type of polysemy displayed by the nine adjectives analysed is somewhat varied. The picture is rather complicated for adjectives such as canonical and marginal, which have many very technical usages, low overall frequency, and – more importantly – both relative and descriptive senses. Furthermore, their polysemy is partly based on the polysemy of their related nouns (canon and margin). Similarly the polysemy of floating and unbalanced is based partly on their related verbs, so that the polysemy here is not a purely adjectival phenomenon. Particular is again a difficult case in having one very frequent use as a function word, some borderline usages, and a few other senses restricted by gradability, predicate position, or by use in certain technical contexts. So whatever one says about the polysemy of adjectives, it is difficult to generalise. Looking to the remaining four adjectives (deep, idle, remote and stable), one may hope for a more ‘pure’ picture of the type of polysemy that is distinctive to adjectives; unsullied by such complications of senses that are not wholly adjectival or are related to non-adjectival words. Of these, remote and stable have one clear basic meaning common to all usages; idle has two main strands of meaning, giving a clear hierarchy of meaning; and deep is a little more complicated, but has a basic meaning 34

See Biber et al. 1999: 508-9ff. on these classes.

42

common to most of its senses, plus a few other relatively distinct senses restricted either by the type of noun collocate (e.g. ‘sleep’, ‘breath’, or ‘colour’) or by syntax (‘threedeep’, ‘deep in [ACTION]’, etc.). The contextual differences between some senses are very subtle. For example, a person described as ‘idle’ may be permanently lazy, may be failing to do given work, or may have no work to do, and specifying such distinctions is difficult. There is often disagreement between dictionaries in how many different senses they show when there is this type of vagueness, where a continuum of senses is to be divided up. For certain adjectives like stable or remote, one can well question whether there is true polysemy occurring at all. These are clear cases of what Moon (1987a) called “quasi-monosemous” words, which have a single meaning, but which cannot be defined in isolation because of the effect context has upon them. However, polysemy is a somewhat relative concept. Some adjectives have closely clustered and similar usages, and others more widely differing and discrete senses. Each adjective has to be presented on its own merits in whatever way seems appropriate. Sometimes there will be a clear hierarchy of senses, a basic common strand of meaning, or a semantic cline of meaning. Sometimes the picture will be more scattered. Whilst much has been written on adjectives from a purely grammatical perspective (e.g. Cruse 1986; 2000), and also to some extent from the perspective of cognitive linguistics and the mental lexicon (e.g. Gross et al. 1989; Murphy and Andrew 1993), it would seem that they are not taken into account sufficiently in the field of semantics. Works on lexical semantics often generalise on word meaning in terms of a word’s features or properties, but adjectives do not have features; they are features. They are difficult to analyse in isolation because they are essentially an aspect of the noun which they modify or characterise. Unbalanced as in ‘unbalanced diet’ is very different from unbalanced as in ‘unbalanced mind’, and the common meaning they share – the essential contribution of unbalanced in each case – is very difficult to establish. The semantic role of adjectives probably needs further investigation. The above analysis has also given many valuable insights into the strengths and weaknesses of the dictionaries chosen, at a very detailed level. Overall, no one dictionary stands out as significantly superior or inferior to the others. Each has its weaknesses as well as its strengths, but none is without anything to commend it. Perhaps CTD and CED, in view of their older more traditional editorial foundations, are a little weaker than the others overall, although CED is to be praised for the range of specialized meanings it covers, and CTD for its conciseness. ODE’s structural format lends itself best to representing polysemous adjectives, although there are recurring errors with the logical ordering within that format, and important senses are sometimes left out. CED’s and CTD’s structures are less suitable, atomising meanings too much without showing the relationship between them; and for clarity’s sake they might do well to include more examples, although space is clearly a great limitation. It is unsurprising that those dictionaries based on a more recent all-over analysis of the language – rather than following in a long tradition – should tend to reflect the common usages more accurately. Advances in corpus linguistics enable them to analyse the vocabulary more easily and thoroughly. They also tend to be clearer, and take into account much of the theory discussed in Chapter 2 regarding how best to define and present adjectives. Even the newer dictionaries for native speakers are able to shake off many of the more unhelpful features of lexicographical tradition35. In ODE, for example, this is seen in its clear hierarchy of meaning, its use of examples, and a rather more flexible approach to defining. COBUILD’s innovation is well-known, and represents the collocational and syntactic preferences particular to each adjective admirably. As well as the implicit 35

For example, those criticised by Patrick Hanks (1987), who also happens to be the editor of NOTE.

43

information it gives through explanations and examples, it also gives very detailed syntactic patterning information, in coded form, possibly of more interest to a dedicated linguist than to a learner. MED treats the high-frequency adjectives very clearly, adopting a flexible approach. It gives useful guides to the senses at the head of some entries, and has good coverage of fixed expressions. The less frequent and more technical words and senses are treated less well (or often not at all) by these EFL dictionaries, for reasons of space and emphasis on core vocabulary. NOTE is an intriguing and highly innovative thesaurus which distinguishes senses very effectively, using illustrative examples and synonymous phrases, and regional and idiomatic expressions, as well as adjectival synonyms. It gives a comprehensive list of synonyms for each sense of an adjective, and one or more antonyms, helping the native speaker understand the nuances of meaning very clearly. It thus does the job of a dictionary well, at least for polysemous adjectives. The native speakers’ dictionaries generally contrast markedly with the EFL dictionaries: they include more obscure senses; and are – if not less clear – certainly less explicit in the distinctions they make between senses: they give fewer examples, do not show so many usage restrictions or preferences, give no indication of frequency, and have less collocational or syntactic information. This is probably because native speakers are better at disambiguating and have less need of explicit information, but this is difficult to test: whereas the format of EFL dictionaries is based to a large extent upon the actual needs of the users, – and as a result tends to be innovative and flexible – dictionaries for native speakers are more bound by tradition. It is always a fine balance between showing too many senses of a word, thus over-complicating an entry, or too few, over-simplifying it. These dictionaries in most cases strike this balance well. Common fixed adjective-noun compounds are frequently given separate and full entries. CTD’s entries for these, perhaps because are more recently written, do not tend to suffer from the extreme brevity of the central entries. Perhaps one can dimly perceive a tendency in CED (quite apart from greater inclusion of obscure technical senses) to split meanings where CTD might tend to lump them together, with ODE somewhere in-between and more variable in coverage. For COBUILD and MED, the issue of frequency also comes into play, but for core meanings, the sense distinctions seem to be based on solid collocational and syntactic foundations. For NOTE it is a similar ‘solid foundation’ of paradigmatic relations that makes it unlikely to make over-fine distinctions in meaning, since this would result in too many overlapping synonyms. This study has highlighted many of the strengths and weaknesses of different dictionaries’ approaches to defining and presenting multiple senses of adjectives, and the varying patterns of adjectival polysemy which make flexibility such an asset in a dictionary’s treatment of such words. Innovations in the analysis, defining style and microstructure of the newer dictionaries allow polysemous adjectives to be dealt with better than ever before. Further studies along similar lines may usefully be undertaken on the treatment of other specific grammatical classes in dictionaries: most obviously adverbs, which are an even more problematic and fragmented class. Whilst on one level the analysis shows how varied the polysemy of words can be, and therefore how important the lexicographer’s intuition, flexibility and autonomy is in treating them well, such studies as this – which analyse the language in detail, as well as evaluating how well it is reflected in dictionaries – may contribute to a theory of lexicographical best practice for the various different word classes and types of vocabulary.

44

References Aitchison, J. (2003) Words in the Mind (3rd Edition), Oxford: Blackwell Publishing. Atkins, B.T.S. (1993) “Theoretical Lexicography and Its Relation to Dictionary Making” in Dictionaries, Number 14. pp. 4-43. Béjoint, H. (2000) Modern Lexicography, Oxford: Oxford University Press. Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. (1999) Longman grammar of spoken and written English, London: Longman. Carter, R. (1998) Vocabulary: Applied Linguistics Perspectives (2nd Ed.), London: Routledge. Catford, J.C. (1983) “Insects are Free: Reflections on Meaning in Linguistics” in Language Learning 33: pp13-32. Cruse, D.A. (2000) Meaning in Language, Oxford: Oxford University Press. Cruse, D.A. (1993) “On Polylexy” in Dictionaries, Number 14. pp. 88-96. Cruse, D.A. (1986) Lexical Semantics. Cambridge: Cambridge University Press. Fox, G. (1987) “The Case for examples” in John Sinclair (Ed.) Looking Up, pp.137-149. Gove, P.B. (1968) “On Defining Adjectives: Part I” in American Speech 43, 5-32. Gross, D., Fischer, U. and Miller, G.A. (1989) “The Organization of Adjectival Meanings” in Journal of Memory and Language 28, pp.92-106. Hanks, P. (1993) “Lexicography: Theory and Practice” in Dictionaries, Number 14. pp. 97-112. Hanks, P. (1987) “Definitions and Explanations” in John Sinclair (Ed.) Looking Up, pp116-136. Jackson, H. (1985) “Grammar in the Dictionary” in R.F. Ilson (Ed.) Dictionaries, Lexicography and Language Learning, Oxford: Pergamon Press. Lakoff, G. and Johnson, M. (1980) Metaphors We Live By, 1st edn, Chicago: University of Chicago Press. Miller, K.J. (1998) “Modifiers in WordNet” in C. Fellbaum (Ed.) WordNet: An Electronic Lexical Database, Cambridge, Massachussetts: MIT Press. Moon, R. (1987a) “Monosemous Words and the Dictionary” in A.P. Cowie (Ed.) The Dictionary and the Language Learner. Lexicographica Series Maior, Tübingen: Max Niemeyer Verlag. pp.173-82. Moon, R. (1987b) “The Analysis of Meaning” in John Sinclair (Ed.) Looking Up, pp.86-103. Murphy, G.L. and Andrew, J.M. (1993) “The conceptual basis of antonymy and synonymy in adjectives” in Journal of Memory and Language, 32, pp.301-19. Quirk, R., Greenbaum, S., Leech, G. and Svartvik, J. (1972) A Grammar of Contemporary English, Harlow: Longman. Schelbert, T. (1988) ‘Dictionaries – too many words?’ in M. Snell-Hornby (Ed.) Zürilex’86 Proceedings, Francke, pp.63-70. Seppänen, A. (1984) “Lexical Integrity or Semantic Diversity: good, great and well” in English Studies 65/6. pp.534-49.

45

Simpson, J. (2002) “The Revolution of English Lexicography” in Dictionaries, Number 23, pp.1-15. Sinclair, J. (Ed.) (1998) Collins COBUILD Grammar Patterns 2: Nouns and Adjectives, Glasgow: HarperCollins. Sinclair, J. (Ed.) (1990) Collins COBUILD English Grammar, Glasgow: HarperCollins. Sinclair, J. (Ed.) (1987) Looking Up: An account of the COBUILD Project in lexical computing, London: HarperCollins. Stock, P. (1984) “Polysemy” in R.R.K. Hartmann (Ed.) LEXeter ’83 Proceedings, Tübingen: Max Niemeyer Verlag. Svensén, B. (1993) Practical Lexicography: Principles and Methods of Dictionary-Making, Oxford: Oxford University Press. Wierzbicka, A. (1993) “What Are the Uses of Theoretical Lexicography” in Dictionaries, Number 14, pp.44-78. Zgusta, L. (1993) “Lexicography, Its Theory, and Linguistics” in Dictionaries, Number 14. pp.130-38. Zgusta, L. (1971) Manual of Lexicography, The Hague: Mouton.

Dictionaries: CED: Collins English Dictionary, 6th Edition (2003). Jeremy Butterfield et al. (Eds.), Glasgow: HarperCollins Publishers. ODE: Oxford Dictionary of English, 2nd Edition (2003). Catherine Soanes & Angus Stevenson (Eds.), Oxford: Oxford University Press. CTD: Chambers 21st Century Dictionary, Revised Edition (1999). George Davidson (Ed.), Edinburgh: Chambers Harrap Publishers Ltd. MED: Macmillan English Dictionary, 1st Edition (2002). Michael Rundell (Ed.), Oxford: Macmillan Publishers Ltd. COBUILD: Collins Cobuild Advanced Learner’s English Dictionary, 4th Edition (2003). John Sinclair (Ed.), Glasgow: HarperCollins Publishers. NOTE: New Oxford Thesaurus of English, 1st Edition (2000). Patrick Hanks (Ed.), Oxford: Oxford University Press.

46

Appendix: Sample concordance lines: marginal36 brmags/UK al, when he was strangely marginal

Top soccer in the League, a wbe/UK asons of economic efficiency marginal

In total, infrastructure times/UK Greg? Both say it will be marginal -a point here, a piece of luck usbooks/US as been one of continual marginal adjustment. What is most strik times/UK ously in the loose for the marginal advantage that might turn a cl usacad/US ombined firm? Suppose the marginal and average variable cost curv sunnow/UK spot-kick decisions were marginal and the Tykes suffered after guard/UK oks like. If it's slightly marginal and in walks an absolute stunn strathy/CA th Rankin nickel mine,a `marginal and very small-scale operation' usbooks/US ficial ones to a largely marginal and ceremonial existence.

times/UK ped again, the decline was marginal and easily eclipsed by increas brspok/UK mprovements are only very marginal and most of the improvement < brbooks/UK imates may be considered marginal, as may remote land at a great strathy/CA nd though it is a rather marginal aspect of the standard, it is usacad/US en. `What is perceived as marginal at any given time depends on econ/UK Inverness, won his four-way marginal at the last election with just usacad/US ou reach a point at which marginal benefit equals marginal cost. usacad/US ty: DTB/DQ. By analogy to marginal benefit, marginal cost is defi usacad/US nother way of saying that marginal benefit(P) equals marginal cost usacad/US rate is the one at which marginal benefit (as measured by the mar npr/US Shuda says the problem is marginal but does need to be taken brbooks/UK # This is really only a marginal candidate for rotation as most strathy/CA emy of simplicity is the marginal case. A case may be marginal brmags/UK Both of us fall into that marginal category. Joe's not old enough strathy/CA submarginal cells, and 2 marginal cells. This venation is typical usbooks/US questions than answers. marginal: Close to a lower limit, on the bbc/UK ctorate in this contest is a marginal constituencies: INSERT ACT times/UK time talking to voters in marginal constituencies. Many had voted newsci/UK ocean area-seaward of the marginal continental shelves and often wbe/UK iminate the benefit of lower marginal corporate income tax rates, usacad/US f each unit produced. The marginal cost of producing the five usacad/US t marginal revenue equals marginal cost, specified in equation usacad/US ses by 1 unit. We present marginal cost for our example in column usacad/US cost that is captured by marginal cost. So if we add up the ma usacad/US We usually show a single marginal cost curve for a firm because usacad/US curves are U-shaped. The marginal cost curve intersects the aver usacad/US Exercise 10.3 Suppose the marginal cost curve for a perfectly usacad/US ce is equal to the firm's marginal cost at an output rate of 30, usacad/US f the individual long-run marginal cost curves for prices above usacad/US stomer. With the tax, the marginal cost curve shifts upward to usacad/US s supply curve equals its marginal cost curve for all points above usacad/US different from short-run marginal cost. Moreover, price must refe usacad/US market, with each firm's marginal cost given by: MC&symbequal # 2 usacad/US lock because price equals marginal cost for all electricity consum usacad/US er demand causes a higher marginal cost. What is the profitusacad/US er do not equate price to marginal cost. Typically, P&symbmult;MC usacad/US e tax is set equal to the marginal cost of the externality at the brbooks/UK iring only to recoup the marginal cost of an extra film print or brbooks/UK wage rate, and hence the marginal cost of labour (MFC), are const brbooks/UK

is, the ratio of the MARGINAL COST of good X to the marginal econ/UK gain much either since the marginal cost of supplying water is smal econ/UK re not necessarily equal to marginal cost. The average electricity econ/UK ugh it would not affect the marginal cost in the long run.

Where strathy/CA esult, price is equal to marginal cost. In equilibria (2 # 5), on usacad/US are constant and equal to marginal costs. Thus, the total costs of brbooks/UK make the assumption that marginal costs are rising over the relev brbooks/UK es will typically exceed marginal costs. This means that by sett econ/UK an society, bear all of the marginal costs of motorised travel-a mos oznews/OZ r west into the more than marginal country.

Everyone knows tha usbooks/US n seemed apt: except for marginal decisions--such as whether to u times/UK try for Lewsey. These were marginal decisions in a game that ebbed strathy/CA can avoid pricing below marginal `delivered' cost. Note that the 36

A complete set of all concordance data used in analysis can be found on the enclosed CD-ROM.

47

times/UK ares are available -- with brbooks/UK egies for increasing the indy/UK c love affair with muse and oznews/OZ de. Queensland now has 17 usacad/US hunter-gatherers exist in usbooks/US moralistic in tone. With usacad/US ame manner. Note that the bbc/UK e the staples in many of the brmags/UK in this business are too usnews/US nt on R&D. This means the npr/US kei average seesawed between brbooks/UK ment, instead of against brbooks/UK appeared menaced. These strathy/CA ry net of adjustment for brbooks/UK re marginal - `nasty but usacad/US have only occasional and wbe/UK 101 share index registered a times/UK the leasing division for a indy/UK ar against the worst, where econ/UK economic case for trimming usbooks/US iberals, it's in Sweden. usnews/US ly lead to an increase in wbe/UK posal might result only in a indy/UK e original tally but with a guard/UK s country which shows that oznews/OZ ons centres have recorded usephem/US beige background and the brmags/UK nal ambition, with only a brbooks/UK net present value of the times/UK very badly - discussion of oznews/OZ ters would campaign inthe times/UK rt of the Tory faithful in guard/UK ugh to give the American a oznews/OZ 28 were ALP seats, 5 were usbooks/US ployed or must eke out a usacad/US on a graph and calculate econ/UK six months, some of Anglo's brbooks/UK only the drag of private newsci/UK eeping aids. Without such brbooks/UK h in those cars was very usbooks/US good point and one very usacad/US largely on their relative oznews/OZ take. `They are extremely brbooks/UK population (the elderly, econ/UK ent in the boom came in the strathy/CA l Front. This was then a usbooks/US e, `It is not that weak, indy/UK nder's great story from the guard/UK ss their resentment at the bbc/UK European Community remaining times/UK 1 Remove dead foliage from brbooks/UK ties more generally. The guard/UK His projections point to a usacad/US 8.2 show the average and usacad/US . Beyond 6 employees, the usacad/US put fixed. Similarly, the usacad/US on. Similarly, there is a usacad/US oportion to the change in usacad/US erm F/ L is the usacad/US to a certain value of the usacad/US pital used increases, the usacad/US the value of the input's usacad/US ) equals the ratio of the usbooks/US was the germ from which brbooks/UK d not, particularly with usbooks/US ions such as eliminating brbooks/UK PROPENSITY TO IMPORT and brbooks/UK Y TO IMPORT, MULTIPLIER. brbooks/UK hange in consumption

marginal differences in prices -- from marginal differentiation of religious ` marginal East End villain, George Dyer. marginal electorates -- those where a marginal environments, these environmen marginal exceptions and to varying degr marginal factor cost exceeds the wage marginal farming areas. Even the C3 marginal for anybody to absorb stock los marginal, from-this-day-forward cost of marginal gains and losses, finally endi marginal groups. In the Irangate affair, marginal groups include the unemployed marginal heterogeneity. The model is th marginal," I said - and the Superintend marginal importance in developed nations marginal improvement of 1.16 points to marginal improvement on the prices that marginal improvements matter.

Back marginal income-tax rates was - and sti Marginal income taxes for the rich are marginal income-tax rates in 1990, he marginal increase in the liability expos marginal increase for those in favour. marginal increases in class sizes harms marginal increases or decreases in marginal inscriptions will be black. marginal interest in the development and marginal investment in the second machi marginal issues, feeling of uncertainty, marginal Labor seats of Richmond and Pag marginal Labour and Liberal Democrat marginal lead going into today's free marginal Liberal and 1 marginal National marginal living as street vendors or in marginal means, then answer the followi marginal mines will become loss-making marginal net product, which was determin marginal notes, subheadings, page number marginal. Once I got going I found I was marginal one. There is one completely marginal operating costs. But multipart marginal operations and simply cannot marginal owners, the homeless) and the marginal, part-time jobs generally done marginal party without the geographical marginal people are witches but only th marginal perspective of a eunuch lover. marginal place they have been allotted marginal players, the United States is marginal (poolside) plants and cut down marginal position of women, especially marginal possibility that Labour could marginal product of labor at the differe marginal product of labor curve is below marginal product of capital MPK is defi marginal product of labor for the second marginal product. Learning Exercise 9.5: marginal product of labor and the term marginal product, which equals the wage marginal product of labor, MPL, increas marginal product to the input's marginal marginal product of labor (MPL) to the marginal productivity analysis of labor marginal products. In the adverse econom marginal products, simplifying product MARGINAL PROPENSITY TO TAXATION. The lar marginal propensity to save (MPS # the marginal propensity to consume (MPC #

48

brbooks/UK tic goods is 0.6 and the marginal propensity to consume all good usacad/US e, under competition, the marginal rate of transformation in usacad/US onal drugs. An increasing marginal rate of substitution is a prope usacad/US marginal utility and the marginal rate of substitution. As noted usacad/US es on the Budget Line The marginal rate of substitution is equal econ/UK e Britons paying the lowest marginal rate of the big seven industria econ/UK raged mining. The effective marginal rate in British Columbia has usnews/US igher-income people pay a marginal rate of 28%. Richard Gephardt guard/UK y trap caused by very high marginal rates of tax". Britain was unus oznews/OZ free threshold and lower marginal rates. According to the Institu times/UK nd leisure group, staged a marginal recovery, adding 4p to 620p, af brbooks/UK ed to be of increasingly marginal relevance as the i980s progress usacad/US arrings, for example, its marginal revenue is $180 (i.e., 200& usacad/US marginal cost as Setting marginal revenue equal to marginal cost, usacad/US sects the dominant firm's marginal revenue curve, mrdominant, at usacad/US s), demand Dplayers), and marginal revenue (mrplayers) curves for brbooks/UK ng a demand for tea. See MARGINAL REVENUE PRODUCT, FACTOR MARKET brbooks/UK e is a sharp step in the marginal revenue curve corresponding to brbooks/UK output. MC together with MARGINAL REVENUE determines the level of brbooks/UK F DEMAND, TOTAL REVENUE, MARGINAL REVENUE, MARGINAL REVENUE PRODU brbooks/UK ism. The question of the marginal revolution of the 1870s raises guard/UK poverished and essentially marginal. Rightly or wrongly, we don't npr/US s --operate on a small, even marginal scale. That doesn't keep new brmags/UK himself and represents a marginal seat, Kensington, where leaseho oznews/OZ > Other seats include the marginal seat of Lowe in NSW held by Ma oznews/OZ tourage as it entered the marginal seat of Hervey Bay more than a oznews/OZ e. Under-achievers! LABOR marginal seat aspirant Cheryl Kernot has oznews/OZ r; I am a regional-based, marginal seat backbench MP. `While I adm guard/UK sts in half their winnable marginal seats and half the seats where guard/UK y might lose over 50 fewer marginal seats than under any alternati econ/UK r, Labour also did badly in marginal seats, which will decide the oznews/OZ e level is the battle for marginal seats in Sydney, one of which oznews/OZ aged to hold on to enough marginal seats to win. Votes gained by oznews/OZ t which assuredly cost it marginal seats last Saturday, especially indy/UK ion to help Tory mps in the marginal seats of Monmouth and Brecon strathy/CA neimongolensis.

The marginal series is dominated by the sixt brmags/UK trieve it back up a steep marginal shelf.

These are just a few econ/UK or a long time regarded as 'marginal # Slovenes, for example, and, strathy/CA as considered deviant or marginal, suggesting that Winnipeg's new guard/UK st responsible for Wigan's marginal superiority in the first half. times/UK Eastwood, the party's most marginal target seat.

Sir Malcolm brbooks/UK ng rate and a 30 percent marginal tax rate.)

The present val econ/UK because although the top marginal tax rate is low, the threshold econ/UK come tax faced an effective marginal tax rate of 75% on the real ret oznews/OZ e taxed at the investor's marginal tax rate. So 65% more money oznews/OZ 00 and $75,000 a year the marginal tax rate will be cut from 50 to oznews/OZ ter tax. If you were on a marginal tax rate of 44.5 percent, the usnews/US th a $100,000 house and a marginal tax rate of 28% that faces an oznews/OZ ation joint venture mines marginal. The four mines presently pay indy/UK cision to contract out more marginal." Then came the abolition of guard/UK New fields have become so marginal they are being left in the seab econ/UK or the economy at large was marginal. This view has changed, especi usacad/US t therapy continues to be marginal to most of the mental health guard/UK chools are concentrated in marginal Tory-held seats. Interest in Mr guard/UK ficial, who adjudicates on marginal try decisions, is well establis brbooks/UK of reasoning led to the MARGINAL UTILITY theory of value whereby usspok/US in this way, in kind of a marginal utility sense. Is the credibil wbe/UK consumption decision, the marginal utility of present consumption usbooks/US and `functions." In the marginal-utility world of the usbooks/US posed a constancy on the marginal utility of money. The other brbooks/UK it was still remote and marginal viewed from Santa Barbara and npr/US beauvoir (County Clerk): The marginal voter is clearly the person indy/UK ionalists gain on Labour in marginal: Voters' Panel: Llandovery indy/UK policy of selling flats in marginal wards to potential Tory voters brmags/UK d float fishing alongside marginal weed in summer.

Pole angl times/UK field well-known faces in marginal Westminster seats such as Ayr. times/UK rne (Lab 5.0%) A three-way marginal which Labour's Candy Atherton usbooks/US ). However, the rational marginal worker will withdraw labor with oznews/OZ trial and trade issues; a marginal workers strategy that secured

49