Printer Friendly

Verbal idioms in focus - towards the continuum of idiomatic expressions. (Linguistics).

1. Introduction

Idiomaticity has been traditionally regarded as one of the most complex linguistic phenomena. The discussion on the problem how certain aberrant idioms should be represented in and explained by regular frameworks has been the theme of a number of linguistic inquiries. By now, many extensive discussions have been published in which one can find the classes of idiomatic expressions and their features (e.g., Makkai 1972; Fraser 1970; Weinreich 1969). The purpose of the present paper is to try to compare four classes of idiomatic expressions of verbal nature -- phraseological verbs, phrasal verbs, primary verb idioms and prepositional verbs -- by means of some objective grammatical tests, such as passivisation, substitution, deletion or insertion and to determine in consequence which classes are more restricted in their grammatical behaviour and which are more free. Finally, it is hoped that the differences in behaviour between the members of the discussed classes would justify the claim that idiomaticity is grad able and these classes differ in the degree of it. The analysis is conducted in agreement with the main principles of Transformational Generative Grammar, since I subject complex phrases to some transformational tests.

As regards the definition of idiom adopted in the present paper, the key principle will be the non-compositionality of meaning (after Hockett 1958 and Makkai 1972). It is possible to say which construction is more compositional (and consequently less idiomatic), as when comparing highly non-compositional bite the dust (in the sense of 'be killed, fall to the ground') with more regular let the cat out of bag (meaning 'tell a secret without intending to do so'). In the former case the verb let as well as the prepositions out of contribute its usual meanings to the sense of the compound, and it is the nouns cat and bag which denote other things than usually.

The next principle justifying the idiom label in this paper will be the integrity of the structure of the construction. As will be demonstrated later on, there are some phrases which allow some movement of their parts (passivisation, particle movement) or change of the structure (the addition of a modifier, the deletion of some part, the substitution of one element for another). Since literal phrases allow all such modifications without any restrictions, it is assumed here that if some idiom does not allow some modification which a literal expression of the same syntactic structure does, then it is the sign that it is relatively highly idiomatic. When there are two idioms of the same structure, one of which allows some movement or change while the other does not, then it seems justified to say that the latter is more frozen (and more idiomatic) while the former is less restricted (less idiomatic, more literal).

2. Historical perspective

The definition of the term idiom has been understood differently over the course of years. Hockett (1958: 171-173) claims that this is a phrase whose meaning is non-compositional, that is the meaning of the whole cannot be fully deduced from the meanings of the parts. To give the example, the sum of the usual meanings of hot ('having a high temperature') (1) and dog ('male canis') do not fully account for the sense of the idiom hot dog ('a boiled or grilled frankfurter in a bun' (Makkai 1972: 30-3 1)). Hockett also maintains that an idiomatic phrase should be any expression of variable reference, without a fixed meaning in all situations (here he mentions anaphor one, numerals, deictic demonstratives this, that, proper names and personal pronouns. He claims that there are idioms of a larger size than a single word and that idioms are not only limited to lexis (phrasal or lexical idioms), but idiomaticity may also be present in syntactic constructions. Metaphors, hyperboles or puns are equally idiomatic as phr asal idioms if their meaning is non-compositional and if, when decoded literally, they seem to be out of context.

Stratificational Grammar (Makkai 1972) adopts the Hockettian principle of non-compositionality of meaning, also labelling as idioms complex expressions whose aggregate meaning do not equals the sum of the constituent parts (just as kick the bucket meaning 'die' is in no way deducible from the usual senses of kick 'hit with the foot' and bucket 'a vessel of wood etc. for carrying water'). Since Stratificational Grammar acknowledges the existence of levels (strata) of language, Makkai (1972) proposes the classification of idioms according to the stratum they exist on. Thus, there are three idiomaticity areas in English. The first is morphology, with such words as conflate, conduct, originally Latin compounds, but which are single words in English. The next is lexology, where there are such idiom classes as tournures (kick the bucket), phrasal verbs (make away, bring up), phrasal compounds (hot dog), and binomials (to and fro, by and large). Finally, there are sememic idioms, such as idioms of institutionalised language (Could you pass me the salt), familiar quotations/proverbs (when in Rome ..., veni, vidi, vici) or highly culture-specific idioms (have two strikes against oneself, never come to the first base with something -- Makkai's 'First Base' Idioms). He also noticed that tournures such as kick the bucket are restricted in some ways, since they do not allow the change of the article or the number of the noun (he called it the "article ban" and the "singular/plural ban").

A completely different view on the definition of an idiom is expressed by Charles Ruhl, who claims in his 1976 "Idioms and data" article that an idiomatic expression is a construction whose words exist in other contexts but never with the same meaning as in this construction. Due to this more flexible definition of an idiom, Ruhl allows the constituent parts of an idiom to contribute to the total meaning of an expression without fully accounting for it (as demonstrated by his analysis of constructions with the verb hit - Ruhl 1976). Ruhl claims that there are no idioms as such, because the apparent oddity of some constructions is only due to the insufficient amount of gathered data, such as the collocation drive *drive WITH a certain speed vs. French aller AVEC avec une certaine vitesse will become perfectly regular and normal when we gather enough contrastive data. Consequently, Ruhl denies the label of an idiom to a number of Makkai's idiomatic classes, including phrasal verbs, proverbs or idioms of institu tionalised social language.

Idiomatic expressions have always triggered a lot of research of transformational grammarians, since the restrictions in their behaviour (e.g., the inability to passivise spill the beans in the sense of 'give away information' with its allowed passivisation when understood literally as 'allow the seeds of a vegetable to run) questioned some of the principles of the transformational model of language. Here I will review a few attempts to solve the problem of irregular idiomatic expressions.

Jerrold J. Katz and Paul M. Postal in their 1963 article "Semantic interpretation of idioms and sentences containing them" try to deal with the problem of restricted idiomatic constructions by claiming that sentences containing idiomatic expressions will be still generated by grammar, but the process of generation will be somewhat different from that of "ordinary" phrases. They investigated the issue why a sentence Sam kicked the bucket yesterday can only be passivised when its literal meaning is meant, and not when the sentence is to be decoded idiomatically. The authors propose the following solution: first of all, dictionary entries of the Lexicon should be of two types: 1) for "ordinary" lexical items and 2) for idioms. During the derivation, the Deep Structure for a literal and for an idiomatic meaning is the same and transformations which follow are triggered by formatives already present in the Deep Structure of sentences. These formatives disturb the relations between terminal strings and nodes domina ting them in such a way that transformations are blocked in the case of idiomatic interpretations, while still allowed when the decoding is to be literal. The semantic interpretation of a given string takes place in the Surface Structure after the derivation is over, and strings with blocked transformations are assigned idiomatic meaning while the transformed strings are interpreted literally.

For Wallace L. Chafe the problem of idioms was the key piece of evidence questioning the validity of the transformational model of language. In his 1968 article "Idiomaticity as an anomaly in the Chomskyan paradigm", Chafe's stand is still transformational, but he proposes that semantics be the starting point of sentence derivation, taking over the place of syntax and operating jointly with it (in this way he can be regarded as a potential ally of the movement of Generative Semantics developed by George Lakoff and James D. MacCawley).

Chafe claims that in the derivation (moving from meaning to sound), there must be a conversion of semantic units into utterable phonetic elements -- what he calls a "symbolization of meaning", with semantics as the starting point in this process. The change of a concept into an idiomatic expression -- idiomatization -- is a special case of this semantic conversion, producing the discrepancy between semantic and post-semantic arrangement Idiomatization, according to Chafe, is a historical process, in which certain specific constructions enter into a special kind of semantic gap, thus producing their unusual meaning. After such a gap has taken place, the original semantic arrangement is still present in language (thus allowing for the old literal arrangement to exist alongside the idiomatic construction), but in addition, a new semantic unit is formed by a shrinkage of the sum of constituent sense of an expression into a new unitary meaning. If an idiom has a literal counterpart, then both are symbolised in the same way -- that is the post-semantic idiomatic kick the bucket is literalized into its original meaning (literalization being the opposite of symbolisation, the process of conversion from post-semantic to semantic arrangement).

Bruce Fraser (1970) claims that the derivation of idiomatic expressions has to be a different from the one of literal phrases, namely that the lexical insertion of an idiom takes place as if the phrase were a single lexical unit semantically (the verb phrase kick the bucket as the verb die), but then the phrase is "dismantled" to see if its parts also conform to selectional restrictions and to allow for their regular conjugations and declensions. After this analysis, the semantic reading is associated with the lowest constituent dominating all parts of an idiom. Fraser also notices that some idioms have some transformational deficiencies, since they do not allow some modifications they should because of their structure, and establishes the "hierarchy of frozenness", with individual idioms situated on various levels of the hierarchy according to the observed abberancies. In the present paper I am following this idea, but my point will be to determine the restrictions of idiom classes, and not of individual con structions as Fraser did.

3. The transformational analysis of the restrictions of idiom classes

The method of the analysis of verbal idioms in this paper will be the following. By means of such objective tests as movability, insertion, deletion, substitution, as well as by judging the non-compositionality of the phrase from the contribution of its constituent parts, the differences in behaviour and restrictions of the classes will be observed on examples of actual sentences. This will give the idea which of them are more frozen, restricted and non-compositional, and, consequently, has the greatest degree of idiomaticity, and which are more free, regular and as such less idiomatic.

The present analysis encompasses only verbal classes of the similar V + N (P) structure, since they can be compared by means of such objective tests as pronominalisation, ellipsis or passivisation. Also, they look similar but behave differently, and as such are a strong case in favour of my theory here. Finally, it is difficult, if at all possible, to compare idioms of different structure (such as verbal with nominal) or of different size and status (e.g., phrasal verbs and proverbs/quotations), since there are no objective tests for such divergent types of constructions and the comparison would be too subjective to have any explanatory power.

The first class to be discussed here are tournures -- Makkai's (1972) phraseological idioms, of the V + N (P) structure, such as kick the bucket, bite the dust, toe the line. The next group are phrasal verbs (V + Prep + (NP)), but only those whose meaning is non-compositional and cannot be directly derived from the senses of the parts. Therefore, The food went down his throat will not be treated here, but His speech went down very well will, because the decoding in the latter case cannot be done by simple summing up the constituent senses. The third class are primary verb idioms (the term mine after Ruhl 1976). These are also similar V + NP constructions with a primary verb (the most common ones -- such as do, make, get, let, keep, give, etc.) followed by a noun or a prepositional phrase (let the cat out of bag, get the sack, pay hommage to, take advantage of etc.). The last verbal class are prepositional verbs (V + Prep constructions such as apply for, beware of, etc.), previously not regarded as idiomatic a t all. They are included here on the grounds that the choice of a preposition a verb takes is strictly restricted, sometimes idiosyncratic and may cause serious encoding problems for second language learners, who are likely to translate the combinations from their own languages. This, in my view, is the justification of the decision to label this class as idiomatic. (2) Also, prepositional verbs have the same structure as phrasal verbs and it will be demonstrated how different the two classes are in their transformational restrictions.

Tournures, or phraseological idioms, are probably the most widely recognised of all non-literal phrases. Most linguists, while investigating the properties of idioms, took this class into consideration. The name was coined by Makkai (1972), and means in French 'the turn of the phrase'. The examples of such constructions as kick the bucket, bite the dust, toe the line show that there is little connection between the meanings of the constituent parts and the sense of the whole construction, and this relatively high non-compositionality of meaning is one of the key arguments for its high degree of idiomaticity. Therefore, this class of idioms is highly difficult to learn and use properly by second language learners, since in the wording of the phrase there is little clue as to its real intended meaning. Also, the possibility of erroneous decoding (this term after Makkai 1972) is greatest in this case, since tournures can be understood both idiomatically (kick the bucket as 'to die') and literally ('to strike a p ail with one's foot'). This is an extremely serious factor concerning this class of idioms, since decoders have no clue which could signal whether the phrase is to be understood literally or idiomatically.

As far as the syntactic behaviour is concerned, it is to be noted that phraseological idioms are probably the most aberrant, frozen, unpredictable and irregular of all idiomatic constructions. Having the ordinary and perfectly regular structure V + N (P), they should permit, without the loss of idiomaticity, some modifications that verbal phrases of this structure normally do. However, the idiomatic kick the bucket is highly restricted in its behaviour, since it does not allow many movements that the same kick the bucket in literal decoding does, which is the proof of its relatively high degree of idiomaticity.
Sam kicked the bucket yesterday
* Sam kicked a bucket/buckets yesterday. MODIFICATION
* Sam kicked _ bucket yesterday. DELETION
* Sam kicked the big bucket yesterday. ADJECTIVE INSERTION
* The bucket was kicked by Sam yesterday. PASSIVISATION
* It was Sam who kicked the bucket yesterday CLEFTING
* Sam kicked the ball and the bucket CONJUNCTION/ELLIPSIS

* Sam brought the bucket and kicked it yesterday. PRONOMINALISATION

When the idiomatic meaning is in question, all the modifications shown above are not allowed, otherwise the phrase loses its idiomatic reference and denotes only the literal sense 'to strike a pail with one's foot'. Indeed, it is not right to put asterisks marking the ungrammaticality of the above sentences, since these are perfectly acceptable with the literal meaning in mind. Thus, it should be remembered that phraseological idioms such as kick the bucket can be decoded in two ways, and it is only the idiomatic decoding which is highly restricted. As demonstrated by the examples above, the idiomatic kick the bucket cannot be modified in any way, nor can the order or the composition of the phrase be changed without the loss of idiomatic meaning. This fact, together with the high degree of non-compositionality of meaning, gives this type of phrases high degree of idiomaticity.

The next class of idioms to be discussed, phrasal verbs, constitute a heterogeneous group of expressions in themselves, highly productive and frequent in English. These are phrasals, which fulfil the basic criterion of some degree of non-compositionality of meaning. So, for instance, make up in the sense of 'to use cosmetics in order to beautify one's skin' is idiomatic, since the component structures of the phrase do not fully account for the total meaning. Of course, phrasals are by their nature highly polysemic, hence the same combination of words can have both literal and idiomatic decoding. Here only the noncompositional constructions are taken into account.

Idiomatic phrasal verbs can be decoded in two ways (get down as either literally 'swallow food with difficulty' or idiomatically 'make somebody depressed'). Idiomatic phrasal verbs are not fully compositional as far as the meaning of the whole in relation to the meanings of the parts is concerned. There are again varying degrees of compositionality inside this category, and there are some phrasals which are more compositional (some elements of the total meaning can be attributed to the parts by sometimes remote associations). As Ruhl (1976) demonstrates, in some cases one can decompose idiomatic phrasals into some distinguishable atoms of meaning: up meaning completion, ascent or upward direction -- go up the hill, finish the bottle up, get up, bring up (in the upward direction from a child to an adult); off meaning detachment or disconnection of two elements -- cut off the electricity, turn off the radio, take off the coat, keep off the grass, etc. Because of this greater degree of compositionality of meanin g in comparison with the phraseological idioms, phrasal verbs are generally easier to decode and learn. Their constituent parts are usually well-known words with multiple meanings, which makes the decomposition of compounds even easier. However, the compositionality of meaning is not full in the case of phrasal verbs, since otherwise they are not regarded as idioms.

As for modifications that phrasal verbs should allow by virtue of their V + P structure, it has to be said that in comparison with prepositional verbs of the same structure they are more restricted. Thus, they do not allow insertion, deletion, co-ordination, which means that their structure has to be unchanged. Sometimes it is possible to substitute some other word for one of its elements (put/turn on/off the radio), but it seems to be the result of the polysemous relations between verbs and particles, and as such is not any piece of evidence in favour of the looser internal structure of idiomatic phrasals.

As for movability, phrasal verbs are not totally restricted, as tournures are, since some of them permit particle movement. But they do not permit pronominalisation (since there is no noun which could be substituted with a pronoun) nor ellipsis.
* The footprints on the bank safe INSERTION/COORDINATION
has given back and away the
* In the morning we ran _ of beer, DELETION
so we finally got to bed.
* I will try to give up for all you SUBSTITUTION
have suffered because of my
He turned the radio on/turned on MOVABILITY (PARTICLE
the radio. MOVEMENT)
* He first took up the job and then ELLIPSIS
off it.

Generally speaking, it is difficult to generalise about the modifications a given item from any class does not allow, since within each type there may be constructions which will have some idiosyncratic features of behaviour. However, when comparing tournures with primary verb idioms or phrasal verbs with prepositional verbs, it becomes evident which are more restricted and irregular and which are more free and less idiomatic in consequence.

To sum up, phrasal verbs are a class of idiomatic expressions whose meaning is to some extent compositional, which means that it is possible to attribute some of the total meaning to the particle or the verb. As for their syntactic behaviour, they are restricted in many ways when compared with prepositional verbs, but they do allow particle movement and some synonymous replacement and are more regular than phraseological idioms. Taking all this into account, it may be said that non-literal phrasal verbs are less idiomatic than phraseological verbs but more idiomatic than prepositional verbs. (3)

Primary verb idioms is the term adopted by me after Ruhl (1976) to denote constructions composed of a primary verb (do, make, keep, get, take, leave etc. -- the most familiar, multi-meaningful verbs) plus a noun phrase or a prepositional phrase. The examples include a number of phrases of a similar structure such as let cat out of bag, get the sack, keep up one head, make the most of I deliberately distinguish a separate class for such constructions, because they have different behaviour and features from other idioms of a similar V + NP type, namely tournures.

It should be noted that phrases belonging to this class might be split into two separate parts: the primary verb and the noun or prepositional phrase following it. The primary verb, being usually extremely frequent and by consequence possessing multiple meanings, is used in one of its senses. For instance, in let cat out of bag (meaning 'to reveal a secret'), let is used more or less in its usual sense, namely 'to allow somebody to do something'. This is also the case with get the sack ('to be fired, dismissed from a job'), where the primary verb get seems to be used in its primary sense 'to receive'. It is only the unusual combination of the primary verb with the rest of the phrase which makes the resultant construction idiomatic. In consequence, the degree of compositionality of the whole phrase is greater than in phraseological idioms, since at least one element of the construction contributes its ordinary meaning to the sense of the whole, but definitely smaller than in prepositional verbs, where all comp onent structures contribute their full meanings to the composite structure.

Since these constructions are more compositional in meaning, it is easier to decode them, and the fact that the primary verb (usually very well-known even to beginning learners) contributes one of its usual meanings makes the rest of the phrase easier to deduce from the context.

As regards syntactic behaviour, this class of idioms seems to be relatively free, which means that it allows a number of operations which non-idiomatic constructions of the same structure do but are also less frozen as compared with other idioms of the same structure (here I am referring to phraseological idioms, which have identical structure, and not really to phrasal verbs or prepositional verbs, which lack the object). One can do passivisation (The homage was paid to him at the end of the service), pronominalisation (First he broke her heart and then _ her spirit) and ellipsis (After making no headway in the morning we finally made some _ in the afternoon). The last two examples are based on Langacker (1987: 474-480).

As regards the possibility of internal modification, primary verb phrases, like most idioms, do not allow any deletion (* We let the cat _ of the bag) nor synonymous replacement (* He finally made up his brain to marry her), though it is possible to insert adjectives modifying nouns (we made some headway).
Sam kept an eye on my baby.

* Sam kept the eye/eyes on my baby. MODIFICATION
* Sam kept _ eye on my baby/kept _ on my baby. DELETION
* Sam kept a watch on my baby. SUBSTITUTION
?It was an eye that Sam kept on my baby. CLEFTING
Sam kept an alert eye on my baby. ADJECTIVE INSERTION
An eye was kept on my baby by Sam. PASSIVISATION
Sam kept a ball in his hands and an eye on my CONJUNCTION/ELLIPSIS
Sam put an eye on my baby and kept it on him PRONOMINALISATION
until the end of the film.

In general, it needs to be noted that primary verb phrases, though certainly idiomatic to some degree, are far more free and compositional than the prototypical phraseological idioms, which is proved by their greater syntactic freedom, greater compositionality and lesser internal frozenness. They allow more modifications than phrasal verbs or tournures, and their meaning is also more compositional, which may be the result of the fact that they are created from primary verbs, which have multiple meanings and one of them may be the one intended in a given compound. This seems to be the evidence in favour of the claim that primary verb idioms are less idiomatic than phrasal verbs or phraseological idioms, since the latter allow less modifications and their meaning seems to be less compositional.

Prepositional verbs are verb-particle constructions, on the surface similar to phrasal verbs, such as apply for; beware of etc. English grammar specifies that some verbs require the use of some particles, and as a result the verb and the particle become a kind of a unit, to be learnt and stored as a whole. Such constructions are fully compositional, as the meaning of the construction apply for is fully accounted for by the senses of the component parts, and the resultant phrase is the sum of the meanings of apply and for.

Because of this full compositionality, prepositional verbs do not cause any problems in decoding, since it is enough to know the meanings of the component words to get the sense of the expression, and one does not need any extralinguistic or contextual knowledge. In this sense such constructions are easy to comprehend and analyse when encountered for the first time in a piece of discourse or a text. However, prepositional verbs seem to be difficult to encode for second language learners, as the choice of the particle in such constructions is arbitrary and sometimes idiosyncratic for English. Therefore, this class may pose certain encoding difficulties, as language learners are likely to transfer the combinations from their mother tongue into English.

As for the changes in internal structure, they seem to permit some insertion or co-ordination (He is stubbornly applying/applying stubbornly for this job, He is applying and applying for it), ellipsis/deletion of a particle may be permitted only if it is recoverable and when both the speaker and the hearer will have the information to supply the missing element (as in he is applying for the job of a manager, but not for the one of a secretary). Synonymous replacement, though rather acceptable, as such is again the result of the polysemy of categories, as seems to be the case in He talked about/on this matter or You can always rely on/rely upon him.

Prepositional verbs are not "frozen" in the sense that they allow the disconnection of the two elements of the phrase without the loss of the holistic nature of the construction (high degree of movability). This can be visible in the case of wh-questions (For what job do you apply?) or clefting constructions (For this job to apply is a ridiculous idea).
Sam is applying for this job.
Sam is applying and applying for CONJUNCTION
this job.
Sam is applying for this job, but ELLIPSIS
not for that one.
For what job is Sam applying? MOVABILITY
Sam is stubbornly applying/applying ADJECTIVE INSERTION
stubbornly for this job.
For this job to apply would be a CLEFTING
stupid idea.
ou may apply and apply, but I know DELETION
you won't succeed
* Sam is applying to this job. SUBSTITUTION

To sum up, it can be said that prepositional verbs are highly free literal expressions, unrestricted and fully compositional. Their degree of idiomaticity is rather small, but it is transparent in their posing encoding problems, being arbitrary and idiosyncratic in the choice of the elements. Definitely, they are least idiomatic of all the four classes discussed in the present thesis, which is demonstrated both in the high degree of compositionality and in the syntactic freedom they possess as demonstrated by the sample sentences.

4. Implications for future research

At this point, it might be useful to try to establish the definition of the term which would reflect the differences between the classes discussed above and which would grant them all idiomatic status. Thus, we might say that an idiomatic phrase is a complex construction, whose total meaning is non-compositional to some degree (the constituent parts contribute varying amount of meaning to the sense of the whole expression), has a varying degree of frozenness as regards insertion, deletion or substitution of its elements and is restricted to a certain degree with regards to the movements the literal phrases of the same structure allow.

As can be seen, the above definition is rather wide and it uses such expressions as "to a different degree", "have a varying degree", etc. It might be criticised as not precise and too broad to explain anything, but it has to be noted that it encompasses all the classes mentioned in this paper, which have a clearly different degree of idiomaticity. Thus, perhaps it might be a good idea to try to adopt in the transformational per se discussion of idiomatic expressions some of the assumptions of the cognitive model of categorisation. Founded on the ideas by Wittgenstein (1953) and developed by Rosch (1975, 1977), the Cognitive Science rejects the objective Aristotelian classification, where an item in order to be granted a given status had to display all the sufficient and necessary conditions to a full degree. According to Rosch (1977) and Langacker (1987), it is the prototype of a category (its most typical, characteristic member) which possesses the greatest amount of characteristic features, but other membe rs may have a varying degree of some features, while totally lacking the others. The consequence is that there are more and less prototypical members of a given category, just as there are things more closely or more remotely reminding the most prototypical piece of furniture, which seems to be the chair. The varying degree of the possession of some feature (here of idiomaticity) positions the members of the category in certain places on the scale from the most to the least prototypical item. The boundaries between categories are fuzzy and overlap. The ordering of the members of a category on the scale may be subjective, since sharp objective criteria are no longer used.

Here the application of the above principles might justify the labelling of all the four classes as idioms despite their obvious differences in behaviour and status, being the consequence of the varied degree of idiomaticity. It is well known that Transformational Generative Grammar (the framework I am using in my analysis) and Cognitive Grammar introduced in the previous paragraph stand in sharp opposition and are believed to be unbridgeable. It is not my intention to bridge this gap, but I believe that it is justified to use some elements of the other model if it helps to solve the problem. Therefore, my point is to open the ground for the discussion of linguistic issues with the use of both opposing grammars.

Below I propose the scale of verbal idioms, reflecting the degree of frozenness and of compositionality of meaning.





It represents the continuum of idiomaticity as established on the basis of the criteria adopted in the present thesis. As can be seen here, both phraseological idioms and prepositional verbs are most closely related to the opposite ends of the scale. This does not mean, however, that they are the absolute prototype and periphery respectively. The scale is open, so new items might be added here, and also these two opposite cases do not possess all the required features in the full degree.

It is evident that out of the four discussed classes tournures are the most idiomatic type, which is justified by their highly restricted behaviour and the number of modifications they do not allow. Also the aggregate meaning of the construction is almost totally non-compositional, which is also the evidence for the idiomatic character of the phrase. Prepositional verbs are their opposite, as being highly non-idiomatic, demonstrated in the fact that their meaning is fully compositional and that they allow a great number of movements and modifications that other constructions of the same structure (namely phrasal verbs) do not. As for the two remaining classes, it seems that phrasal verbs are more idiomatic than primary verb idioms, since the meaning of the former is less obviously compositional and phrasals are more restricted as regards syntactic movements.

5. Conclusion

In conclusion, this paper has been an attempt to prove the existence of the varying degree of idiomaticity in different verbal expressions. It was demonstrated in sample sentences which types are more restricted than the others, which seems to be the sign of their greater degree of idiomaticity. With the help of objective transformational tests, such as deletion, substitution, passivisation or insertion, it was possible to say which of the four classes is the most restricted and irregular, and as such closest to the prototype of the category of idiom. By contrast, the least idiomatic of the four types was established, and between these two opposing poles the other two classes were positioned. In this way, the continuum of idiomaticity of verbal idioms was created. It is hoped that this method might be useful in analysing other classes of idioms, for instance nominal constructions, and that a similar continuum might be established also for them.

As for the practical applications of the proposed analysis, it is believed that the findings about the degree of idiomaticity and the restricted modifications might be utilised in lexicography. Nowadays, dictionaries do not give too much information on how to use idioms properly, failing to recognise the fact that they do not allow certain operations that literal phrases of the same structure do. If we add to the definitions some clues as to the syntactic behaviour of idioms, then second language learners might find it easier to use idioms appropriately (for the full treatment and the proposed changes in idiom entries see Krajka, forthcoming).

(1.) All meaning paraphrases are after Oxford advanced learner's dictionary of current English.

(2.) Makkai (1972) calls such constructions "idioms of encoding", since they are difficult to encode properly if one does not know them. They are distinguished from "idioms of decoding", such as tournures (kick the bucket), which maybe decoded in a wrong way, the result of which may be the disinformation of a learner.


Chafe, Wallace L.

1968 "Idiomaticity as an anomaly in the Chomskyan paradigm", Foundations of Language 4:109-125.

DiPietro, Robert J. -- Edward L. Blansitt (eds.)

1976 The Third LACUS Forum. Columbia, S.C.: Hornbeam Press Inc.

Fraser, Bruce

1970 "Idioms within a transformational grammar", Foundations of Language 6: 22-42.

Hockett, Charles F.

1958 A course in modern linguistics. New York: Macmillan.

Hornby, Albert S. -- A.P. Cowie (eds.)

1980 Oxford advanced learner 's dictionary of current English. Oxford: OUP.

Katz, Jerrold J. -- Paul M. Postal

1963 "Semantic interpretation of idioms and sentences containing them", Quarterly Progress Report 70 (MIT Research Laboratory of Electronics), 275-282.

Krajka, Jaroslaw

forthcoming "The gradability of idiomatic expressions -- Lexicographic definitions in focus".

Langacker, Ronald W.

1987 Foundations of cognitive grammar. Vol. 1. Theoretical prerequisites. Stanford: Stanford University Press.

Makkai, Adam

1972 Idiom structure in English. The Hague: Mouton.

Puhvel, Jaan (ed.)

1969 Substance and structure of language. Berkeley -- Los Angeles: University of California Press.

Rosch, Eleanor H.

1977 "Human categorization. Advances in cross-cultural psychology", in: N. Warren (ed.), 1-49.

Rosch, Eleanor H. -- C. Mervis

1975 "Family resemblances: Studies in the internal structure of categories", Cognitive Psychology 7: 573-606.

Ruhl, Charles

1976 "Idioms and data", in: Robert J. DiPietro -- Edward L. Blansitt (eds.), 456-466.

Warren, N. (ed.)

1977 Studies in cross-cultural psychology,. London: Academic Press.

Weinreich, Uriel

1969 "Problems in the analysis of idioms", in: Jaan Puhvel (ed.), 23-82.

Wittgenstein, Ludwig

1953 Philosophical investigations. Oxford: Blackwell.
COPYRIGHT 2000 Adam Mickiewicz University Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Krajka, Jaroslaw
Publication:Studia Anglica Posnaniensia: international review of English Studies
Date:Jan 1, 2000
Previous Article:Some hints on the importance of teaching idiomaticity (1). (Linguistics).
Next Article:Optimizing interaction in the second language classroom. (Linguistics).

Related Articles
The idiom experience.
Ralph Lemon.
On the idiomatic status of English compounds. (Linguistics).
Some hints on the importance of teaching idiomaticity (1). (Linguistics).
What users do with dictionaries in situations of comprehension deficit: an empirical study (1).
The Lexicon in Dickens.
Variation in multi-word units: the absent dimension.
I hear what you're saying: sometimes what we hear and what we put on paper are entirely different things.
Vocalis Ltd.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters