19 LANGUAGE Dušan Gabrovšek University of Ljubljana Faculty of Arts, Slovenia “Micro” Phraseology in Action: A Look at Fixed Binomials Summary Multiword items in English are a motley crew, as they are not only numerous but also structurally, semantically, and functionally diverse. &e paper o'ers a fresh look at %xed binomials, an intriguing and unexpectedly heterogeneous phraseological type prototypically consisting of two lexical components with the coordinating conjunction and – less commonly but, or, (n)either/ (n)or – acting as the connecting element, as e.g. in body and soul, slowly but surely, sooner or later, neither #sh nor fowl. In particular, their idiomaticity and lexicographical signi%cance are highlighted, while the cross-linguistic perspective is only outlined. Key words: phraseology, fixed binomial, collocation, idiom, compound, advanced learners’ dictionary, bilingual dictionary Stalne dvočlenske zveze kot primer delovanja “mikro” frazeologije Povzetek Večbesedne enote v angleščini so ne le številne, ampak tudi strukturno, pomensko in funkcijsko izredno raznolike. Članek obravnava stalne dvočlenske zveze, zanimivo in heterogeno vrsto frazeoloških enot, ki jo tipično tvorita dva leksikalna elementa, ki ju povezuje priredni veznik and, redkeje pa tudi but, or ali (n)either/(n)or (n.pr. body and soul, slowly but surely, sooner or later, neither %sh nor fowl). Prispevek se ukvarja predvsem z idiomatiko in leksikografskimi vidiki tovrstnih zvez, le na kratko pa z njihovim medjezikovnimi  vidiki. Ključne besede: frazeologija, stalna dvočlenska zveza, kolokacija, idiom, zloženka, dvojezični slovar UDK 81’373.74’374=111=163.6 DOI: 10.4312/elope.8.1.19-29 20 Dušan Gabrovšek “Micro” Phraseology in Action: A Look at Fixed Binomials “Micro” Phraseology in Action: A Look at Fixed Binomials 1. Introduction: The ABCs of Fixed Binomials In linguistics, a %xed binomial 1 is a structurally frozen and often irreversible conventionalized sequence of two content words – occasionally including proper names – used together as an idiomatic expression or collocation, belonging to the same grammatical category, and having some semantic relationship. Fixed binomials can ful%ll a variety of communicative functions, often showing emphasis or gradation, and indicating emotional involvement, informativeness, or precision. &ey are conjoined by some syntactic device such as and or or, with and clearly predominating: aches and pains, bed and breakfast, before and after, business or pleasure, cause and e$ect, clear and concise, deaf and dumb, drink and drive, each and every, food and water, give and take, good or bad, heaven and hell, here and now, here and there, hook and eye, knife and fork, life and death (also life or death), north and south, older and wiser, once or twice, pots and pans, pure and simple, research and development, salt and vinegar, sadder but wiser, soap and water, sooner or later ; Adam and Eve, Jekyll and Hyde, David and Goliath ; [to be] neither here nor there, in every shape and size, [in the] dim and distant past, to win (something) / to beat somebody fair and square. &e prototypical binomial, it will have been noticed, can be extended in a variety of ways, the additional constituents being either (almost) obligatory or more or less optional. &ese %xed strings are also known by a number of other designations, including %xed order coordinates, irreversible binomials, irreversible coordinates, binomials, binomial pairs, freezes, twin formulas, paired parallel phrases, co-occurrences, or (roughly) Siamese twins. While not being exactly overwhelming in number, thus representing a micro topic within phraseology in quantitative terms, this type of multiword unit is surprisingly varied, though structurally simple, and hence not really micro (that is the reason for quotes in the title), straddling as it does both compounds and idioms on the one hand, and collocations on the other. 2. Phraseological Status and Meaning of Fixed Binomials As a phraseological category, %xed binomials are diverse in that they can be semantically either transparent or opaque (or somewhere on the cline between the two), just as they can be either frozen or only “loosely” %xed, the latter meaning that the order can be reversed (e.g. day and night – night and day). In some cases, the key distinction between collocation-type and compound-/idiom-type binomials is blurred at best. In most cases, however, a basic contrast can be made between the former, which are semantically transparent (e.g. ladies and gentlemen, aims and objectives) and the latter, which are opaque (e.g. bread and butter, hit and run, up and about, divide and rule/conquer, under lock and key). Furthermore, a number of %xed binomials are also characterized by a substantial amount of conventionality and stereotyping, resulting in (heavily) clichéd and sometimes pragmatically complex strings such as #rst and foremost, give and take, prim and proper, slowly but surely, in every shape and size, [not] in any shape or form, when all’s said and done, or in this day and 1 Judging from Google-derived evidence, fixed binomial is largely a mathematical and statistical term. 21 LANGUAGE age. Some of them are semantically and pragmatically complex while comprising only the “easy” words; thus the spoken expression when all’s said and done is far more than what it says on the surface: it is ‘used to remind someone about an important point that needs to be considered’ (Mayor 2009, 42). Likewise, their conventionality and/or %xedness can be observed whenever the sequence involved is “illogical”; for instance, while lightning logically must precede thunder, the %xed binomial – thunder and lightning – de%es this logical order. 2 Some binomials are pragmatically restricted and often “extended” by other items, thus making up larger %xed sequences, as in our thoughts and prayers are with the [e.g. family of the deceased person]. Yet others are restricted in other ways, for example stylistically, as in the old-fashioned “emphatic” binomial (to be full of) vim and vigor. Binomials can be (heavily) institutionalized and thus compound-like: bed and breakfast, black and blue, cause and e$ect, fame and fortune, hide-and-seek, rock and roll. Moreover, some are so heavily conventionalized that they are recorded as main entries in dictionaries in an unconventional or abbreviated form, e.g. rock’n’roll, R & B (=rhythm and blues), R & D (=research and development), R & R (=rest and relaxation [also an AmE term for a holiday given to army people after a long period of hard work or during a war]. &e reasons for their unconventionality can be entirely commercial, as observed e.g. in snap N slice, the name of a kitchen cutter extensively advertised on TV in 2009. Indeed, binomials used as (brand-)names of consumer products are not di@cult to %nd: Fresh & Clean (tissues), Relax & Tone (body massager), Head and Shoulders (shampoo), Speak & Spell (an American children’s educational toy), for example. Finally, some binomials resemble patterns, in that they comprise both “real” words 3 and “slot- like” spaces to be %lled by any out of a speci%able set of lexical items, as in to up and [do something], for instance he upped and left. Even though it is possible, in principle, to reverse the customary order of %xed binomials “if special e'ect is meant to be expressed”, violating the %xed order is often regarded as a source of unacceptability (James 1998, 72). I would add that if anything, wrong ordering also a'ects idiomaticity in the broader sense of restricted nativelike textual selection and nativelike sequencing. Let us note, merely as an aside, that it can be quite di@cult to answer the very basic question of what exactly counts as a binomial as contrasted with, or distinct from, a mere grammar-based and -derived combination of noun + conjunction + noun. Take simple and direct, for instance: It is a %xed binomial and, as such, recorded in dictionaries? What criteria were applied? Was it 2 Of course, the “logical” ordering is not too difficult to find, as in (stories of) survival and recovery, spotted on CNN in February 2010. Overall, such sequences may be considered somewhat different from “genuine” binomials, because they only seem to follow logical ordering and are thus less likely to cause any interlingual difficulty, not to mention their phras(eologic)al status, but then quite a few of the “logical” ones too appear to be heavily conventionalized too (e.g. [to be] born and bred, hit and run, wear and tear, seek and find, dead and buried). 3 Note that a handful of items are likely to lend themselves to forming a variety of binomials, say the adjectives nice (nice and warm, nice and clean, nice and easy), and bright (bright and breezy, bright and early, bright-eyed and bushy-tailed), possibly due to their ability to function as intensifiers. 22 Dušan Gabrovšek “Micro” Phraseology in Action: A Look at Fixed Binomials (only) semantic (non-)compositionality? Syntagmatic features? What about frequency of co- occurrence? Can the available criteria be applied successfully to all binomial sequences? 3. Significant Features of Fixed Binomials Binomials can be semantically more or less transparent (come and go, friends and acquaintances, loud and clear, peace and quiet), opaque (cloak and dagger, pins and needles, part and parcel, ‘a necessary feature’; ins and outs, ‘all the facts and details’), 4 or on the cline somewhere between the two end-points (rough and ready, ‘not perfect but good enough for a particular purpose’; bits and pieces informal ‘any small things of various kinds’; once or twice, ‘a few times’; facts and #gures, ‘the basic details, numbers etc concerning a particular situation or subject’). Secondly, %xed binomials can be polysemous, whether with two senses each, typically with a literal and a %gurative reading (e.g. wear and tear), 5 or semantically more complex (e.g. [be] neither here nor there; cat and mouse; black and white; bread and butter). &irdly, some of them are also grammatically restricted as they can only be used in the plural (e.g. twists and turns, swings and roundabouts, by leaps and bounds). Quite a few other features will be referred to later due to the fact that they are di@cult and/ or problematic. Speci%cally, %xed binomials can consist of phrases joined by prepositions, they can be extended, some are reversible, comprise the same item used twice, are used as di'erent word classes, can be open-ended and rather elusive, all of which contributes to their heterogeneity and to making them something of a lexicographer’s headache. 4. Brief Review of Literature Following the two pioneering studies, the rather obscure Abraham (1950) and especially Malkiel (1959), the topic of %xed binomials has received its share of attention, for instance in Norrick (1988), including a handful of cross-linguistic studies such as Klégr (1991) and Ernestova (2007). &ere also exists a dissertation-type study of binomiality in the %eld of law (Dámová 2007), where %xed binomials are particularly common and have therefore been studied fully as part of specialized communication (Gustafsson 1984, Bhatia 1994). 6 &ey include aid and abet, assault and battery, cease and desist, law and order, null and void, breaking and entering, without let or hindrance, and health and safety. 7 4 Cruse (1986, 39-40) discusses their fossilization in terms of degrees of semantic opacity, where the constituent elements begin to lose their independent semantic value: “[a]s degree of opacity diminishes, we approach the somewhat indeterminate transitional zone between opacity and transparency.” 5 Thus one can figuratively speak about, for instance, somebody’s success story starting to show a little wear and tear. 6 Piirainen (2008, 222) observes, while citing evidence from German, that binomials can be traced back to gestures once performed in court, together with ancient wordings of laws, which may well be the reason why they seem to be so typical of the field. In legal English binomials are 4-5 times more common than in other prose texts (Gustafsson 1984, 123). The magisterial Fowler, however, suggested another reason: “Their abundance in English is perhaps partly attributable to legal language, where the multiplication of near-synonyms is a normal precaution against too narrow an interpretation, and also contributes a pompous sonority to ceremonial occasions” (Fowler 1965, 554). 7 They can be found in business English too, witness e.g. terms and conditions and profit and loss. 23 LANGUAGE Contemporary accounts of %xed binomials are not too frequent; they are largely corpus-based (Hatzidaki 2000). &e most thorough analysis to date is Benor and Levy (2006), where it is pointed out, based on an analysis of some 700 binomial tokens retrieved from online corpora, that there are a number of semantic, metrical, and frequency-based constraints that contribute signi%cantly to binomials’ ordering preferences, overshadowing the phonological factors that have traditionally been given priority. 8 &e %xedness of binomials often re_ects what has been labeled a syntax of preference, e.g. the positive concept preceding the negative one (good and bad). &us Adam and Eve iconizes hierarchy or preference, whereas Cain and Abel re_ects precedence of the %rstborn rather than preference. By contrast, here and there and this and that iconize a preference for proximity as against distance (Anderson 1998, 267). &ere are very few linguistics textbooks and vocabulary books that include a section on binomials (Gramley and Pätzold [2004, 58, passim] 9 ; McCarthy and O’Dell [1994, Section #77, pp. 154-55] and McCarthy and [O’Dell 2001, Section #72, p. 81]). Similarly, most dictionaries of linguistics terms give the concept short shrift. Even the World Wide Web is weak in its coverage of %xed binomials: A Google search does turn up quite a few hits for the concept, but they refer chie_y to the mathematical/statistical notion. &ose that are related to linguistics are few, aside from being mostly bloggers’ random comments. Almost none of the works referred to above are devoted to the lexicographical aspects and implications of %xed binomials. Hence this paper. Its orientation is quite broad; %rst it provides a concise description of the phenomenon and then looks at the basics of its lexicographical treatment, especially in a bilingual framework. &at is why the term dictionary has been used here merely as a convenient abstraction indicating the standard alphabetized general-purpose language-reference source, no attempt having been made to discuss speci%c (types of) dictionaries, phraseological or other, and their salient features. Likewise, cross-linguistic issues have been raised only in general terms, which is certainly not to suggest that they are uninteresting or even irrelevant – quite the reverse, in fact. 5. Problems in the Treatment and Handling of Fixed Binomials First of all, %xed binomials can be, one, not single-word items but phrases conjoined by and/ or/but (e.g. once and for all, last but not least, take it or leave it), and two, connected not by conjunctions but rather by prepositions (one after another, from head to foot, tit for tat). Being joined by a preposition rather than a conjunction creates something of a categorial problem, because prepositions express various relationships, say temporal and spatial ones, while conjunctions “only” join. However, some authors make no distinction between “conjunctive” and “prepositional” binomials (e.g. Norrick 1988). Also, some binomials comprise not only phrases but also a comma instead of a conjunction (as e.g. in easy come, easy go, ‘something 8 Yet the ordering of binomials not infrequently exhibits a considerable amount of variation (Benor and Levy 2006). Variation can indeed be extensive, as in art(s) and science(s) / science(s) and art(s). 9 Of course, phraseology textbooks – not that there are that many in existence – are likely to be more generous in this respect, witness especially Fiedler (2007, 40–3, 63–4, passim). 24 Dušan Gabrovšek “Micro” Phraseology in Action: A Look at Fixed Binomials - especially money - obtained easily was quickly used or spent’). Are these any di'erent? Not likely – the comma, after all, is a kind of and-type conjunction. Second, as has already been pointed out, there exist moderately or heavily extended binomials (e.g. the rank and #le, [we’re] ready and waiting, every now and then, somebody’s likes and dislikes, by leaps and bounds, to go to rack and ruin, to be few and far between, live and let live, [the] nuts and bolts [of something], to look/search high and low, in this day and age, over and done with, come hell or high water). Are they to be given the same status and treatment as their binary, “basic” counterparts? Again, some are transparent and often emphatic ([we’re] ready and waiting), while others are opaque (come hell or high water) while also including instances of the emphatic type (to be few and far between). Does this fact call for a di'erent treatment? A single policy is di@cult to apply because the extensions may vary a lot in terms of length as well as their nature (some are largely optional/additional, whereas others are virtually impossible to leave out, thus rendering the basic binomial somewhat suspect as to its very existence). &ird, some binomials are reversible, meaning that they exist in two synonymous forms/strings 10 (such as pleasures and problems, o$ and on, clear and speci#c, night and day), thus displaying a “looser” %xedness while %rmly remaining %xed multiword units. A few synonymous binomials can also be found that are not identical (as in errors and omissions and errors or omissions [McIntosh 2009, 564]). Very rarely does the reversal result in a di'erent meaning, as in salt and pepper (‘condiments’) vs. pepper and salt (‘color’) (Cruse 1986, 47). Reversal may only exceptionally occur within a phrase, or even within two parallel phrases, usually without a'ecting the semantics of the phrase, as in easy come, easy go vs. come easy, go easy. Note that unless there are semantic consequences, the possibility of reversing the sequence is, overall, not that important, given that in each such case the string still keeps Pawley and Syder’s (1983) widely cited “nativelike selection” criterion of idiomaticity. Fourth, some (extended) binomials comprise the same item used twice (e.g. less and less, through and through, to be on the up and up 11 ). &ese are simple in structure, as there can be only one ordering; however, they too are either idiom-like or collocation-like. Fifth, binomials used as di!erent word classes (e.g. to and fro [noun, adjective, or adverb] or crash and burn [verb, noun, or adjective]) are likely to be an encoding problem, speci%cally with those L1 languages that lack conversion as a word-formation process. Generally, this feature is easier to show e@ciently and systematically in monolingual than in bilingual dictionaries. Sixth, the most intriguing – and di@cult – feature of binomials may be their “open-endedness”, witness e.g. the pattern nice and [adjective]: clean/easy/warm/slow/quiet... Many binomials appear to be elusive; (new) coinages often go unrecorded in dictionaries (e.g. gently and e$ectively, 10 Very rarely does a fixed binomial exist in two non-synonymous forms connected with different conjunctions but consisting of the same lexical components, an example being life and death and life or death. 1 1 This particular binomial, let us note, exhibits both “varietal” polysemy and obligatory expansion. Its polysemy stems from the fact that it has a different meaning in AmE (‘honest and doing things legally’) than in BrE (‘becoming more successful’). 25 LANGUAGE in%ation and unemployment, [to battle] boos and bruises 12 , love and a$ection, money and business [newspaper section title], plain and simple, attractive and appealing, up and coming, (to be) up and running, words and phrases [also title of a book], and many more). &e idea that a %xed binomial is “used to convey a single meaning”, proposed by H.W . Fowler in his Dictionary of Modern English Usage of 1926 (Fowler 1965, 554), may well be somewhat elusive! Indeed, the latest revision (Burch%eld 1996, 712) expands the earlier de%nition, recognizing the issue: binomials “often have the same meaning as each unit in the pair (or a slightly strengthened one), or are related in other formulaic ways”. Four types are recognized (ibid.): a) those used mostly for emphasis (e.g. bag and baggage, bits and pieces/bobs, rant and rave, in any shape or form), b) "xed collocations, with one of the components being used in an archaic sense or the combination having acquired a meaning di'erent from that of either component alone (e.g. at someone’s beck and call, odds and ends, part and parcel, spick and span), c) those which consist of associated ideas (e.g. hu$ and pu$, nuts and bolts, thick and fast, ways and means), d) those which consist of opposites or alternatives (e.g. hit and miss, through thick and thin, to and fro). Anyway, even though binomials can be quite common, some of them hardly ever get listed in dictionaries (e.g. theory and practice, to shoot and kill (someone), with millions of Google hits and an extensive record in today’s corpora 13 but scant lexicographic evidence). Also, the back- cover blurb of a recent book on slang (Adams 2009) says, inter alia, “Adams shows it [slang] is much more than just _ash and trash.” Is there an English dictionary that has managed to record the string? Moreover, even culture-bound binomials can get short shrift; thus very few English dictionaries record publish or perish, a phrase used for describing the harsh realities of (originally American) academic competition, for which there are currently (June 2010) 166,000 Google hits, 65 occurrences in the WebCorp Live corpus, and 31 occurrences in the Corpus of Contemporary American English, for example. Finally, binomials are quite frequent in English, which makes it di@cult to work out a single lexicographical policy. T o put it simply, are all binomials the same? Should they all be listed in dictionaries in bold? 14 In terms of what criteria? &eir commonness is easy to show; for example, a New York Times obituary 15 of horticulturist David Murbach contains, inter alia, the following: the world of trees and gardens ; holiday crowds oohed and ahhed ; a height and width [of trees] that we need ; I was more or less the one who was there to say ... . See what I’m driving at? 12 Taken from a Yahoo article on a famous NBA basketball player’s performance in a late 2009 game. The binomial may well have been patterned on an existing binomial, viz. cuts and bruises. 13 Thus there are 470 occurrences of theory and practice and 49 occurrences of shoot and kill in the 400-million-word Corpus of Contemporary American English. 14 In this respect, dictionaries are not reliable, perhaps due to the (occasional?) application of an “intuitive” approach; thus a binomial may get listed, even several times, but not in bold (e.g. arts and crafts in the Longman [Mayor ed. 2009, 80, 393]). 15 Published on 5 January 2010 and written by James Barron. Retrieved from http://www.nytimes.com/2010/01/06/ nyregion/06murbach.html 26 Dušan Gabrovšek “Micro” Phraseology in Action: A Look at Fixed Binomials 6. Dictionary Treatment of Fixed Binomials 6.1 Monolingual Learners’ Dictionary &e treatment of %xed binomials in most reputable English dictionaries is largely unsystematic, some being included and others being left out, with inconsistencies being quite common, not to mention the possibility of errors occasionally creeping in. Signi%cantly, too, one and the same binomial may be given a very di'erent treatment even in comparable dictionaries; for instance, the latest revised editions of three of the leading advanced learners’ dictionaries of English record the %xed binomial each and every 16 in the following manner: as each and every one in the Macmillan English Dictionary (Rundell ed. 2007, 464) as both each and every (subentry) and each and every one of (boldfaced part of an illustrative example) in the Longman Dictionary of Contemporary English (Mayor ed. 2009, 530) it is ignored altogether in the Oxford Advanced Learner’s Dictionary (T urnbull and Lea eds. 2010). Generally speaking, in such dictionaries most binomials are included as – whether boldfaced or not – examples of use, sometimes with explanations of their meaning in parentheses; quite a few are not entered at all. Learners’ dictionaries of English include many more binomials than their native-speaker-oriented relatives, which is both quite logical and quite appropriate, given the more diverse reference needs of their users coming from a variety of L1 backgrounds. As to the general lexicographical policy, what should be done at all costs is preserving the distinction between compound-like and idiom-like binomials on the one hand, and those which are merely %xed sequences: the former, being as they are semantically non-transparent, should all be duly listed and de%ned. By contrast, the latter are clearly less crucial, and indeed do not, for the most part, represent a decoding problem, even though they clearly contribute to better language production in the encoding process, as their role in contributing to textual idiomaticity is quite signi%cant. 6.2 Bilingual Dictionary As to the general cross-linguistic perspective of binomials as observed in bilingual lexicography, the treatment of %xed binomials in general-purpose bilingual dictionaries should re_ect the standard practice of handling %xed binomials primarily in terms of their semantic (non-)opacity and %xity, adding in each case a speci%c cross-linguistic dimension adopted for the bene%t of the reference needs of the primary target group of their users, notably with respect to their L1 background. What this basically means is that bilingual dictionaries should not hesitate to show %xed binomials in bold especially for encoding tasks. &e (partly) opaque ones should all be duly listed and translated, on a par with idioms. 16 There are 1285 occurrences of this binomial in the Corpus of Contemporary American English. 27 LANGUAGE Furthermore, special care should be devoted to recording those binomials which are transparent and employ di'erently ordered %xed sequences in the two languages (e.g. SI z dušo in telesom [‘with soul and body’] vs. EN (with one’s) body and soul, or SI jasno in glasno [‘clearly and loudly’] vs. EN loud and clear). Such cases are not infrequent, let alone exceptional: the %xed-word- order conventionality of %xed binomials can be simply (and often largely unpredictably at that) implemented di'erently in di'erent languages, in that the %xed sequences in question can easily be reversed, as in the two cases above. Moving – however brie_y – beyond the English and Slovene languages, for example, it has been pointed out that “in German and Italian you go ‘forth and back’ (hin und her, avanti e indietro),” and that “in Malay you address ‘gentlemen and ladies’ (tuan-tuan dan puan-puan). Neither way is more logical than the other, and while some may see cultural pointers determining the order of the items, they are probably best seen simply as %xed, arbitrary strings that combine two opposing items from the same lexical %eld.” (Carter and McCarthy 1988, 25) Finally, there are in principle four main options available to the lexicographer for the treatment of %xed binomials: (1) main entry (recommended for opaque/institutionalized compound-like binomials [rock’n’roll, hammer and sickle, down-and-out, #sh and chips]), (2) de"ned subentry (recommended for opaque idiom-like binomials [in the here and now, ‘at the present time’; be neither here nor there, ‘be irrelevant’; fetch and carry, ‘do simple and boring jobs for someone as if you were their servant’; hue and cry, ‘angry protests’]), (3) highlighted part of an example of use (recommended for collocation-like binomials [war and peace, theory and practice, %ora and fauna, loud and clear]), (4) ignored altogether (not really recommended as this policy is likely to result in the violation of acceptability/idiomaticity, if not more. However, most transparent %xed binomials can be left out of dictionaries catering only to their users’ decoding needs). &e four options should be selected judiciously and re_ect consistency in the application of sound criteria. &is may not always be easy, particularly in distinguishing between (1) and (2) and deciding on either the one or the other. Moreover, there is more to the lexicographical treatment of binomials than this – for example, should they, as a matter of principle, (always) be contextualized or not? In all (kinds of) dictionaries? Should extended binomials be given a di'erent treatment than the basic binary ones? Finally, faced with the issue of including or excluding %xed binomials for the sake of reassurance (this kind of lexicographical service is likely to be needed, in most cases, only for encoding tasks), the practicing bilingual lexicographer should, by and large, do well to follow intuition coupled with available teaching experience, and to consult extensively both learner corpora and _uent speakers of both languages. 28 Dušan Gabrovšek “Micro” Phraseology in Action: A Look at Fixed Binomials 7. Conclusion Fixed binomials represent an intriguing and surprisingly diverse if minor category within the phraseology of English. &is, however, should not deter the bilingual lexicographer from treating them as carefully and consistently as any major type of multiword items in English. Again, the basic element of lexicographical policy is that opaque binomials, being as they are idiom- like or compound-like vocabulary units, must always be listed and de%ned; on the other hand, whenever a binomial is transparent in meaning, it should merely get listed (not de%ned!) only in dictionaries designed also with an encoding component in mind. What must be the priority, in addition to the cross-linguistic orientation applied to the two languages under consideration, is the inclusion of all semantically opaque binomials, to be followed by the representation of their idiomaticity in the broader sense. &e former desideratum is essential in decoding L2 texts. On the other hand, the representation of their idiomaticity is to be shown via a judicious inclusion/exclusion policy for transparent binomials; this desideratum is signi%cant in encoding, as it is likely to make for decidedly better – more idiomatic – L2 text generation. Here, in tabular form, are the prototypical cases of %xed binomials in terms of their lexicographical treatment: EXAMPLE TYPE DICTIONARY STATUS sh and chips compound main entry in the here and now idiom boldfaced dened subentry war and peace collocation boldfaced (part of) example of use errors and omissions collocation (part of) example of use or le€ out &ese, to be sure, are but broad guidelines. Matters of detail will have to be dealt with in a separate paper incorporating a number of corpus-extracted examples, but on the basis of the general remarks presented in the paper. Bibliography Abraham, R.D. 1950. Fixed Order of Coordinates: A Study in Comparative Lexicography. Modern Language Journal 34, no. 4 (April): 276-87. Adams, M. 2009. Slang: The People’s Poetry. New York: Oxford University Press. Anderson, E.R. 1998. A Grammar of Iconism. Madison, NJ: Fairleigh Dickinson University Press; Cranbury, NJ and London: Associated University Presses. Benor, S.B., and R. Levy. 2006. The Chicken or the Egg? A Probabilistic Analysis of English Binomials. Language 82, no. 2 (June): 233–78. Also available online at http://idiom.ucsd.edu/~rlevy/papers/binomials-accepted.pdf 29 LANGUAGE Bhatia, V. 1994. Cognitive Structuring in Legislative Provisions. In Language and the Law, ed. J. Gibbons, 136–55. Harlow, Essex and London: Longman. Burchfield, R.W., ed. 1996. The New Fowler’s Modern English Usage. First edited by H.W. Fowler. 3 rd edn. Oxford: Clarendon Press. Carter, R.A., and M.J. McCarthy. 1988. Vocabulary and Language Teaching. Applied Linguistics and Language Study. Harlow, Essex: Longman Group UK. Corpus of Contemporary American English (COCA). At http://www.americancorpus.org/ Cruse, D.A. 1986. Lexical Semantics. Cambridge Textbooks in Linguistics. Cambridge: Cambridge University Press. Dámová, P. 2007. The Language of Law: A Stylistic Analysis with a Focus on Lexical (Binomial) Expressions. Thesis, Masaryk University in Brno, Faculty of Education, Department of English Language and Literature. Also to be retrieved from http:// is.muni.cz/th/79361/pedf_m/THESIS.doc Ernestova, M. 2007. Role of Binomial Phrases in Current English and Implications for Readers and Students of EFL (English as a Foreign Language. In Literacy without Boundaries: Proceedings of the 14 th European Conference on Reading, Zagreb, Croatia 2005, ed. G. Shiel, I. Stričević and D. Sabolović-Krajina, 273–8. Osijek: Croatian Reading Association. Also to be retrieved from http://www.hcd.hr/datoteke/Zagreb_Conference_Proceedings.pdf Fiedler, S. 2007. English Phraseology: A Coursebook. Narr Studienbücher. Tübingen: Gunter Narr. Fowler, H.W. 1965. A Dictionary of Modern English Usage. 2 nd edn. rev. by Sir E. Gowers. Oxford: Clarendon Press. [First edition published 1926] Gramley, S., and K.-M. Pätzold. 2004. A Survey of Modern English. 2 nd edn. London: Routledge. Gustafsson, M. 1984. The Syntactic Features of Binomial Expressions in Legal English. Text – Interdisciplinary Journal for the Study of Discourse 4, no. 1–3: 123–42. Hatzidaki, O. 2000. Binomials and the Computer: A Study in Corpus-Based Phraseology. Paper presented at ALLC/ACH 2000, University of Glasgow, 21–25 July 2000. To be retrieved from http://www.arts.gla.ac.uk/allcach2k/Programme/session5. html#513 James, C. 1998. Errors in Language Learning and Use: Exploring Error Analysis. Applied Linguistics and Language Study. Harlow, Essex: Addison Wesley Longman. Klégr, A. 1991. A Note on Binomials in English and Czech. Acta Universitatis Carolinae Philologica 1, Prague Studies in English 19: 83–8. Malkiel, Y. 1959. Studies in Irreversible Binomials. Lingua 8, no. 2 (May): 113–60. Rptd. in his Essays on Linguistic Themes, 311–55. Oxford: Blackwell, 1968. Mayor, M., ed. 2009. Longman Dictionary of Contemporary English. 5 th edn. Harlow, Essex: Pearson Education / Longman. McCarthy, M.J., and F. O’Dell. 1994. English Vocabulary in Use. Cambridge: Cambridge University Press. McCarthy, M.J., and F. O’Dell. 2001. Test Your English Vocabulary in Use: Upper-Intermediate. Cambridge: Cambridge University Press. McIntosh, C., ed. 2009. Oxford Collocations Dictionary for Students of English. 2 nd edn. Oxford: Oxford University Press. Norrick, N.R. 1988. Binomial Meaning in Texts. Journal of English Linguistics 21, no. 1: 72–87. Pawley, A., and F.H. Syder. 1983. Two Puzzles for Linguistic Theory: Nativelike Selection and Nativelike Fluency. In Language and Communication, ed. J.C. Richards and R.W. Schmidt, 191–226. Harlow, Essex: Longman Group. Piirainen, E. 2008. Figurative Phraseology and Culture. In Phraseology: An Interdisciplinary Perspective, ed. S. Granger and F. Meunier, 207–28. Amsterdam: John Benjamins Publishing. Rundell, M., ed. 2007. Macmillan English Dictionary for Advanced Learners. 2 nd edn. Oxford: Macmillan Education. Turnbull, J., and D. Lea, eds. 2010. Oxford Advanced Learner’s Dictionary of Current English. [First ed.] By A.S. Hornby. 8th edn. Oxford: Oxford University Press. WebCorp Live: The Web as Corpus. At http://www.webcorp.org.uk/index.html