Printer Friendly

Towards a dynamic account of phraseological meaning: creative variation in headlines and conversational humour.


In half a century the study of fixed expressions (2) has undergone a formidable evolution both with respect to the characterisation of their formal-syntactic features as well as their description on the semantic-conceptual plane. Whereas early Russian and German paradigms claimed phraseology as a linguistic discipline in its own right, focussing on linguistic elements which can be negatively defined in terms of irregularities (stability, idiomaticity, lexicalisation), recent approaches adopt a more positive perspective vis-a-vis the linguistic behaviour of the expressions under scrutiny. Both syntactic and semantic properties of fixed expressions are increasingly looked at in terms of the regularities they share with other, non-phraseological linguistic units, thus replacing the traditionalist dichotomy of free vs. fixed expressions by a prototypical category structure. Eventually, this evolution must tackle the long-standing question to what extent 'the' semantics of phraseological expressions can be analysed from a particularist point of view, or, to put it in a more provocative way: What arguments can be found for maintaining the study of phraseology as a sub-discipline of a genuinely usage-based and dynamic account of meaning? With regard to this theoretical issue, the present paper subscribes to a non-particularist, holistic perspective on the semantic status of fixed expressions.

The present contribution pursues a threefold aim. On a first, overall theoretical level, this paper envisages a general outline for the development of a dynamic and integrated account of phraseological meaning. To achieve this goal, I adopt Cognitive Linguistics (CL) as a promising theoretical framework, which advocates the characterisation of meaning in terms of a subject-related conceptual structure as well as the integration of discourse elements as an inevitable aspect of a genuinely usage-based account of meaning. Second, I propagate a non-particularist point of view, including both fixed as well as non-fixed expressions in the analysis. In line with this view, the semantic structure of phraseological expressions is expected to be fully analysable in terms of cognitive categories of construal as they operate throughout all kinds of language use. In this respect, I expect to derive an ecological validity from the approach presented here, as it may become apparent that its range of application extends well beyond phraseological expressions alone (3). Third, finally, with regard to the kind of data investigated in this paper, I want to raise specific interest for the intensified study of creative as well as other types of less conventional language use in order to gain a better insight in the way cognitive mechanisms of construal (metaphor, metonymy, analysability, figure/ground alignment etc.) can be operated in a flexible way, thus revealing a fundamental prototype structure on the level of our cognitive abilities.

This paper is structured as follows. The first three sections focus on the theoretical concepts that are of major importance for the application of an integrated account of meaning. Section 2 introduces Cognitive Linguistics as the general framework of this contribution by describing the basic cognitive linguistic tenet that meaning is essentially a conceptual value, pivoting around the interacting subject as mediating structure. In section 3, I discuss isomorphism (especially analysability) and motivation as two long-standing, but often misrepresented basic mechanisms of phraseological semantics. Section 4, then, advocates the inevitable inclusion of discourse elements into a usage-based account of semantics and suggests, accordingly, the application of an integrated model of semantic analysis. In this respect, I introduce Langacker's (2001) 'Current Discourse Space' model as a valuable template for an encompassing analysis of our linguistic data. Section 5 describes the process of creative variation in the processing and interpretation of fixed expressions, as it becomes apparent in two different types of (mini)texts: newspaper headlines (5.1) and a specific type of conversational humour (5.2). I mainly make use of English data, but with regard to the general theoretical claims presented here, I occasionally include Dutch and German utterances as well.


In developing an adequate account of the semantic properties of fixed expressions, linguistic theories in general as well as phraseology as a specific linguistic discipline have started to move away from the traditional view according to which semantic stability features as a prominent characteristic of any kind of fixed expressions. One of the linguistic paradigms, which strongly opposes to a static account of meaning, and which I will be using as the theoretical framework for the present contribution, is Cognitive Linguistics (CL) as it is developed in the general grammatically oriented approach by Langacker, Talmy, Croft & Cruse and others (4).

One of the pillars of CL is the non-restrictive definition of semantic structure as extending well beyond the boundaries of the linguistic system as such. Cognitive Linguistics rejects the traditional view of meaning as the fixed value of a linguistic expression, determined in terms of truth conditions or interna1 semantic relations such as synonymy, hyp(er)onymy etc. Instead, as CL identifies the conceptualising subject as the mediating structure between word and world, meaning advances to a cognitive structure, which is inherently embedded in a larger context of knowledge, understandings and belief The distinction between linguistic core meaning and encyclopaedic knowledge is discarded, ultimately defining semantics essentially as conceptualisation (Langacker 1987). This view runs counter to truth-conditional and generative approaches to meaning which advocate a one to-one mapping between externa1 world and linguistic-conceptual structure. Rather, dynamic semantic theories like CL argue that "situations can be 'construed' in different ways [...] and different ways of encoding a situation constitute different conceptualizations" (Lee 2001: 2). Or as Croft and Cruse (2004: 42) put it, "in cognitive linguistics conceptualization is the fundamental semantic phenomenon; whether alternative construals give rise to differences in truth conditions or not is a derivative semantic fact".

Taken to the level of specific linguistic expressions, meaning emerges from the interaction between the so-called profile (what is being designated) and the base of an expression, which comprises all kinds of conceptual and contextual elements activated with different degrees of prominence as the conceptual background of the designated entity. Accordingly, the meaning of an utterance such as (1) can only be characterised with respect to the broader conceptual background on which this utterance is being used.

(1) You drink too much

Depending on the characterisation of relevant knowledge domains such as LAW AND ORDER or MEDICINE or WlNE TASTING etc., the sentence in (1) each time activates a different meaning. In function of the interpreting individual's personal experience but also as a reflection of deeply entrenched socio-cultural conceptual patterns, some elements in the conceptual base turn out to be more salient than others (Giora 1997, 2003). In this view, crucially, meaning is not simply 'retrieved' as an inherent semantic structure, the value of which remains stable irrespective of the interacting subject (5). From a CL viewpoint, instead, a linguistic utterance as such merely functions as a cue the processor uses as a starting point in a process of meaning construction.

On this account, a complex morphological or phraseological expression does not require a specific semantic theory. To the extent that complex expressions inherently involve several constituents as well as--in most cases--a so-called 'image component' (the composite literal meaning), an adequate characterization of meaning requires all these levels of linguistic organisation to be considered. In essence, however, the same semantic mechanisms apply to these levels and structures as well, according to which every substructure (constituents, image component) may potentially contribute to the conceptual base, on the background of which the expression profiles its overall meaning (6). Compare in this respect the German fixed expressions in (2), which all share the same referential meaning "X is stupid". However, since they differ from each other by one constituent and, consequently, in their image component as well (7), their semantic structure is not identical as each linguistic substructure, e.g. Wasser, Stroh, Scheisse, activates its proper conceptual contentas an element of the base onto which the expression's profile "X is stupid" is realised.

(2) a. Er hat Wasser im Kopf ('he has got water inside his head')

b. Sie hat Stroh im Kopf ('she has got straw inside her head')

c. Er hat Scheisse im Kopf ('he has got shit inside his head')


With regard to an adequate semantic analysis of fixed expressions, the so-called 'building block' metaphor is abandoned, according to which phraseological expressions represent secondary constructs, the meaning of which is considered (not) derivable from the stable meaning(s) of their individual components (Cuyckens et al. 2003: 14ff). Instead, a dynamic view on semantics--as adopted by CL--defines the complex (fixed) structure as "a coherent structure in its own right" (Langacker 1987, 453). At the same time, CL embraces compositionality, analysability and motivation as prominent semantic principles, which may interact and (partially) determine the successful interpretation process of fixed expressions (ib. 448ff; see also Langlotz 2001, 2006 (8) and Mena, forthcoming) (9).

3.1. Isomorphism

As a matter of fact, analysability and compositionality represent two different perspectives (top-down vs. bottom-up) on the phenomenon of isomorphism, which pertains to the one-to-one relationship between form and meaning of a complex linguistic expression. Next to the principle of compositionality, according to which an expression's overall meaning (partially) results from the components' meaning, involving different kinds of specialisation, an expression's degree of analyzability is determined by the extent to which the interpreter recognises the contribution of single components to the overall phraseological meaning ('co-activation'). Especially with regard to the creation of a motivational link between the overall phraseological meaning and individual components, analysability represents an important semantic operation (infra).

In their application of CL-theory onto Dutch phraseology, Geeraerts (1995) and Geeraerts/Bakema (1993, 196ff) describe the analysability of an expression as the result of a projection of the familiar overall meaning onto the different component parts. In this respect, an expression like (3), which contains a unique constituent (heinde), provides a good illustration of this top-down interpretation process (ib.; see also Langacker (1987, 465) who labels it "back formation").

(3) van heinde en verre (lit. 'from near and far': "from everywhere")

It appears that heinde, which is etyniologically related to dt./engl. hand, meaning 'near, at hand', is attributed a variety of etymologically 'incorrect' interpretations, which are still in remarkable accordance with the expression's well-known overall meaning. Accordingly, questioned about the meaning of this component, which does not occur outside the context of this expression anymore, native speakers of Dutch give unexpected, yet 'motivated' answers such as einder ('horizon'), van op de heide ('coming from the heath'), vati bij de heidenen ('coming from the heathens') etc. This observation demonstrates that (historical) motivations may even change on the basis of a synchronically conceived phonetic resemblance between a single component of the expression and any other word which might fit a motivated meaning description for the expression (see Feyaerts 1992).

3.2. Motivation

Because of its subjectiveness, motivation has traditionally been considered a problematic notion for an adequate semantic description. In a dynamic account of meaning, however, motivation represents a highly relevant parameter, which contributes to the overall semantic analysis of an expression. Motivation can be described as a synchronically and individually determined semantic process, in which language users attempt to make sense of a specific expression by establishing an interpretational link between the well-known overall (phraseological) meaning of that expression and any other linguistic or conceptual element. The question whether this self-construed interpretation coincides with the diachronically correct one, is irrelevant, as meanings may change constantly as the result of a process of reinterpretation (see Geeraerts 1995).

Motivational links with a fixed expression's overall meaning can be established on any level of linguistic and conceptual organization (10). Most straightforward, in this respect, are expressions such as (4), the motivation of which is achieved through its full analysability in terms of the constituent parts.

(4) to cast pearls before swine ('to waste precious things, gifts, ... on those who cannot appreciate them')

This example is motivated by the perfect transparency (top-down isomorphism) between form and meaning, according to which pearls corresponds with 'precious things, ...' and swine with 'those who cannot appreciate them'. Obviously, some phraseological expressions are only partially analyzable, as in (3) where only verre--an archaic variant of present-day Dutch ver ('far')--offers a clear connection to the overall meaning. Although analysability provides an important type of motivational link, both categories do not coincide as an expression's motivation does not necessarily involve one or more constituents. In this respect, the expression in (5) represents a good illustration, as its formal and semantic structure are not isomorphic. Instead, the motivation is provided by the highly salient sene of the image component as such (e.g. a metaphor).

(5) to be on the same wavelength as someone

(6) to take the bull by the horns

The expression in (6) demonstrates that the level on which an expression's analysability becomes apparent, is not always clearly distinguishable, as different interpreters may associate the expression with different 'meanings' or semantic descriptions. For example, depending on whether one describes the expression in (6) as 'to tackle a problem at its most difficult aspect' or as 'to grapple fearlessly with a problem', the analysability is to be localised on the level of individual constituents or the level of the image component as a whole respectively. This observation makes clear, once again, that an adequate semantic description requires the interacting subject being taken into account. The example in (3), which apparently activates some motivations on the basis of phonetic resemblances, demonstrates that motivational links (or: re-motivations) can be established on the basis of highly individual associations as well.

In order to provide an optimal understanding of motivation as an integral dimension of meaning, it is important to draw attention to the non-identity of the notions motivation and grounding as the latter is concerned with uncovering an expression's ontogenesis or, in cognitive-semantic terms, its 'experiential ground'. This ultimate ground may consist of some universal bodily experience, or--as far too often neglected in cognitive metaphor theory--a cultural-historical structure as well. It appears that both notions, grounding and motivation, are quite often confused or not identified as such, leading to erroneous assumptions about the semantic properties of complex expressions. A rather restrictive interpretation of motivation, for example, can be observed in Verstraten (1992,232), as the degree of transparency of fixed expressions--defined here as a subjective notion, closely linked to motivation--is said to depend on the synchronic awareness of the original metaphors and metonymies, associations, folk models, etc. Awareness of the historical meaning of an expression, however, is not the only factor by which the degree of its synchronic transparency is determined, as processes of re-interpretation have to be considered as well. 'The same observation can be made for Kowalska-Szubert (1996, 113), where the motivation of an expression is claimed to be dependent on the individual's awareness of the expression's genesis. Again, a property like this needs to be discussed in terms of conceptual grounding, instead of motivation or transparency.

Crucially, both grounding and motivation of a fixed expression can, but do not necessarily need to coincide, as illustrated by the following expression.

(7) Hij is niet van gisteren

[He is not from yesterday]

"He is clever, not stupid"

As the biblical grounding of this expression's neutral non-negated form van gisteren zijn ('to be from yesterday': "to be behind in time, not from today; to be not informed", Job 8:9) may not be clear anymore, the expression's well known overall meaning "to be clever, not stupid" is motivated nowadays in two different ways. On the one hand, in accordance with its biblical grounding, the expression in (7) is motivated as 'he is not too old; he is from today'. On the other hand, however, the same overall meaning apppears to be re-motivated by the opposite concept 'not too young' as well, implying that the person in question is more experienced than one might think. Reinterpretation processes like these may be due to changing sociocultural circumstances as illustrated by an expression as to keep someone at arm's length (discussed by Lakoff (1987, 447ff). As the original experiential grounding of this image is not transparent anymore, two opposite interpretations (with correspondingly different motivations) come to mind: 'keep someone at a certain distance' and 'be very close to a person' respectively. At this point, it is important to notice that in most cases the process of de-motivation does not lead all the way down to a stage of complete opacity, in which no motivating link is available whatsoever. Language users always tend to establish new and other motivations. This does not mean, however, that no opaque (or: 'idiomatic') expressions can ever occur (11).

Concluding this brief overview, it is clear that an adequate semantic analysis of fixed expressions requires an unrestricted integration of the semantic properties discussed here as they put the subject in charge of a dynamic process of meaning construction.


In section 2, we have identified the definition of meaning in terms of conceptualisation as one of the major principles of the CL-paradigm. Yet, taken by itself, this characteristic still does not provide us a fully dynamic account of meaning. In order to arrive at that point, our analytical tool needs to implement a second cornerstone of CL-theory, which characterises linguistic structure essentially as usage-based (12).

According to usage-based models of language, a speaker's linguistic system is experientially grounded as it is determined by usage events (Langacker 1987, Barlow and Kemmer 2000) (13). Accepting this observation, of course, means that the primary source of information is the actual use of language utterances in context. In contrast to most formal approaches to language, CL argues that there is no strict boundary between linguistic and contextual information and that language is grounded in discourse and social interactions (Langacker 2001: 143). This means that dimensions such as the context of speech as well as the conceptual elements which constitute the 'shared knowledge' among interlocutors all play a central role in semantics, and they cannot be separated from the "core meaning" of an utterance. Somewhat surprisingly, though, CL has only recently started to develop a fully integrated account of meaning in which discourse elements, as elements of the rich conceptual base, are systematically included into the semantic analysis as well (Langacker 2001) (14).

Langacker's notion of a current discourse space (CDS), defined as a mental space "comprising those elements and relations construed as being shared by the speaker and hearer as a basis for communication at a given moment in the flow of discourse" (2001: 144), represents a serious attempt to establish a unified treatment of semantics. This inclusive account is also highly cohesive, integrating the objective conceptual content of an utterance with every element that pertains to the interactive circumstances of the ongoing discourse. The notion of a CDS allows us to integrate utterances of both speaker (S) and hearer (H) into a coordinated discourse representation. Figure 1 represents how, in successful communication, speaker (S) and hearer (H) are joined in their coordinated ,focus on the conceptual entity that is designated by the linguistic unit (profile). Bearing in mind that meaning resides in the tension between a linguistic unit's profile and its conceptual base, all other elements in Figure 1 may be evoked as relevant structures of the base. As such, the viewing, frame represents the immediate scope of attention, which delimits those conceptual entities which are of particular relevance and immediately conceivable at any given moment in the unfolding process of discourse.

In a sentence like (8), the metaphorically used word bird profiles an airplane focus), but in the immediate background (viewing, frame) the knowledge domain BIRDS (ANIMAL), which functions as the source structure of this metaphor, is activated as well. The same dual structure determines the meaning of a fixed expression such as to give someone the green light in (9). This expression profiles the meaning "to give someone permission to do something that they were planning to do or have asked for", but at the same time the 'literal' meaning of green light in association with the specific scene expressed by the image component is activated as well, as a prominent part of the viewing frame (the base).


(8) Its tanks both fuelled up, this bird can fly more than 5000 km.

(9) Finally they gave him the green light.

Obviously, the CDS also contains a vast amount of knowledge presumed to be commonly accessible and on the basis of which speaker and hearer engage in all kinds of interaction. Since any utterance (or usage event) is embedded in a--broadly defined--context of speech, elements pertaining to both bodily, mental, social and cultural circumstances may be conceptualised as aspects of the base as well (Langacker 2001: 145). As the central element of the context of speech, finally, the ground consists of the speech event itself, the speaker and hearer, their interaction (the double-sided arrow), and the specific circumstances (time and place) of the utterance (15).

The multifaceted CDS provides the schematic conceptual setting for linguistic meaning. It is the operational playground for the various cognitive mechanisms of construal, which are constantly at our disposal to rearrange the interna1 structure among the elements of the CDS and therefore enable us to decide in what way an experience will be represented.

The representation on a time axis of successive frames, each of which depicts the scene being negotiated by the speaker and hearer at a given instant, brings in the dynamics of a discourse dimension. Langacker (2001: 151) ascribes "linguistic structures (of whatever size)" a discursive meaning as he defines them as "instructions to modify the current discourse space in particular ways. Each instruction involves the focusing of attention within a viewing frame." Accordingly, a sequence of attentional frames, each of which corresponds to one updating of the CDS, can be represented as in figure 2 where heavy lines identify the frame being acted on by speaker and hearer, labeled the zero or focus frame. This frame is preceded by a minus frame and followed by a plus frame.

It is clear that the theoretical and methodological implications of adopting a genuine cognitive linguistic approach to the semantic analysis of (fixed) linguistic expressions are not trivial. First, it needs to take into account the rich conceptual landscape in which meaning emerges through interaction with the subject, involving semantic, pragmatic, contextual, cultural, and even (inter)personal information. Second, a cognitive account has to come to terms with the dynamic aspect of meaning in discourse, along which expressions invite us to constantly update and modify the current discourse space.


In the following section, we make an attempt to meet both requirements by applying Langacker's CDS-model (2001) to the creative use of fixed expressions in two different types of context: expressive newspaper headlines and witty anecdotes involving a specific kind of conversational humor ('trumping').


Many (cognitive) linguistic studies present their empirical analysis in a rather restrictive way: examples are stripped of any context and in most studies the type of data under investigation is limited to conventionalised language use. Hence, no systematic attention is paid to the study of less objectifiable (and collectable) language use such as creative, humorous, expressive language. For the present purpose, we will focus our attention on two specific types of creative language involving phraseological expressions. We will look first at the generation of an effect of wit in economic newspaper headlines and second, we will analyse a few cases of verbal 'trumping', a specific type of conversational humour.

As to the theoretical question to what extent the analysis of non-conventionalised, occasionally modified and novel utterances can be of any use to phraseological research, the answer touches upon the creative potential as a fundamental characteristic of human cognition. Looking at creative language use allows us to draw a more adequate picture of the way in which our experience is structured by cognitive construal mechanisms such as metaphor, metonymy, figure/ground-arrangement, etc. Crucially, the identification of basically the same conceptual patterns in non-conventionalised as well as conventionalised expressions reveals that both types of linguistic utterance need to be analysed in terms of the 'standard' cognitive construal operations. It puts forward the observation that on a metalinguistic level, these construal mechanisms can be applied in a marked, non-prototypical way, yielding an effect of novelty, expressiveness, etc. (16) Accordingly, prominent instances of creative language use such as humour and wit exploit this flexibility in several respects. From the perspective of phraseology, a systematic analysis of the structural patterns of creativity will integrate variation as a categorial feature of fixed expressions. In this respect, we refer to several studies in the (mainly German) tradition of phraseology research, in which from a non-holistic point of view (see section 1), specific patterns of phraseological variation have been observed (Burger et al. 1982: 68-104, Hemmi 1994, Wotjak 1992, Sabban 1998, Balsliemke 2001, Stockl 2004).

5.1. Headlines

Quite some research has been done both by linguists and discourse analysts on the different communicative functions of headlines. Apart from the summarising function (Bell 1991), headlines frequently serve as eye-catchers to persuade the reader to continue reading the article they accompany (Alexander 1997). Dor (2003) argues that both these functions serve the same goal on a higher functional level: headlines are designed to optimise the relevance of the stories for the readers. Alexander (1997: 94) notes that headlines are "generally used to catch the attention of the reader in a witty fashion or to provide a wordplay that ties in with the subject matter of the article". In the context of this paper, we look at a specific type of newspaper headlines, which potentially generate an effect of wit next to their primary referential function". Specifically, we will focus our attention on headlines, in which the effect of wit involves the creative re-motivation of a phraseological expression. It will be considered, then, to what extent the CDS-model (section 4) can be implemented in the analysis of specific utterances as an adequate tool to account for the various conceptual and discursive aspects that constitute a meaning structure.

Consider the headlines in (10)-(12), collected on the basis of a one-week survey of the economy pages of the Financial Times, each of which contains a fixed expression (marked in italics) that contributes to the effect of wit. In the following analysis, we take (10) as a representative example. This headline comments on the tumbling Carlsberg share(s) following a negative statement concerning Russia, one of Carlsberg's growth markets.

(10) Russia takes froth off Carlsberg results (Financial Times, 21/02/03)

(11) Drug case may cause chronic pain for Bayer (Financial Times, 27/02/03)

(12) The Agnelli family is again in the driver's seat at Fiat (Financial Times, 27/02/03)

In a successful interpretation of this headline, two meanings with different degrees of salience are activated. Taking into account factors which determine a meaning's relative salience, such as experiential context, conventionality and familiarity, the conventionalised, figurative interpretation of this headline appears to be the prominent one (18). Accordingly, the verbal expression to take froth off activates its overall phraseological meaning ("cause to diminish"). Next to this primary interpretation, however, a second, less salient meaning is activated as well without dismissing the first meaning, causing a slight effect of wit (19). It appears that this secondary meaning ties up with the literal level involving the image component of the phraseological expression, according to which one can literally 'take froth off a beer'. As we do not simply want to confirm the contextual presence of and switching between two different meanings, but instead describe the operational impact of the utterance on different levels of the current discourse space, the contextual and conceptual elements of these headlines are of particular interest for our purpose. In example (10), the activation of the secondary meaning is triggered by the use of this expression in the context of a financial report on the Carlsberg brewery. What can be observed, then, as crucial components of the semantic structure, is a clustering of three construal mechanisms operating on different elements and in different dimensions of the CDS.

A first construal mechanism operates as a sort of figure/ground shift within the lexical-semantic structure of the phraseological expression. During the activation of the secondary meaning, the image component (with froth as the central constituent) is brought onstage into the focus of attention (20). With regard to the complex semantic structure of fixed expressions (section 2), this shift unfolds the analysability of the phraseological expression to take froth off something before our eyes. Indeed, the witty effect can only succeed through the activation of the (literal) image component together with the literal meaning of froth as its most relevant constituent.

Second, and a key element in this semantic shift, the conceptual structure of Carlsberg gravitates from a contextually embedded concept (emerging in an article about the financial situation of this company in the economy pages of a newspaper) to a concept in which elements belonging to our shared knowledge about this company (a brewery and its product; major characteristics of beer etc.) are highlighted. Although in the salient financial interpretation of this utterance, Carlsberg may already subsume these product-related knowledge structures as distant elements of the conceptual base, they are promoted into the viewing frame as the fixed expression to take froth of something is interpreted literally. Accordingly, one might say that for this constituent as well, the witty interpretation involves some sort of figure/ground shift within the realm of the conceptual base.

A third construal operation pertains to the inner metonymic structure of the knowledge structures, which are activated as elements of the conceptual base for the secondary, witty interpretation of Carlsberg. Crucially, only if one activates the conceptual link between the company and beer as its product (first metonymy: PRODUCER FOR PRODUCT) as well as the link between froth as a prominent pari and beer as the entire product (second metonymy: PART/WHOLE), the witty interpretation of this headline can succeed. In a dynamic model of semantic analysis, the operationalisation of these three construal mechanisms, in close interaction with each other, captures the semantic evolution in this small piece of discourse.

The activation of both interpretations, causing a certain effect of wit, may be represented as in figure 3. In line with Langacker's (2001) CDS-model, two interpretational 'stages' (or frames) in the discourse process are depicted: the most salient, contextually embedded meaning is represented as the minus (first) frame whereas the realisation of the non-salient, witty interpretation appears as the zero (second) frame.

In order to get a clear understanding of the depicted semantic elements and relationships, the visualisation requires some further clarifications. Inside the viewing frame, the complex structure of a fixed expression is represented schematically by 'Z' which stands for the overall phraseological meaning; 'XY' stands for the image component and 'X' and 'Y' for the individual components--taken in their literal sense--of the expression (2)'. In this example, 'CA' stands for Carlsberg and 'RU' for Russia.

A Comparison of both interpretation frames with each other makes clear that an adequate description of the semantic structure of this headline exceeds a mere identification of 'the' two meanings involved in the interpretation of the phraseological expression. In Langacker's terms, instead, the following "instructions" to modify, the previous frame of attention in the current discourse space are to be noted:

1. Hinging on the common element 'Carlsberg', the focus of attention (profile) shifts from the minus to the zero frame, by taking in the image component ('XY') of the fixed expression as well as froth ('Y') as its central constituent on the one hand, and dropping out the phraseological meaning ('Z') together with the contextual element Russia, on the other. Accordingly, the second, witty interpretation redirects the viewing frame, in which the utterance is profiled, from a contiguous relation 'COMPANY--FINANCE' to 'COMPANY--BEER'.


2. Next to this lexical-semantic adjustment, a clear conceptual rearrangement of the current discourse space has to be identified as well, notably with regard to the conceptual structure of Carlsberg. In the zero frame, the (economical) contextual elements are downplayed as the most prominent structures of the conceptual base in favour of a substructure pertaining to the interlocutor's common knowledge about breweries and their products. This realignment is represented by the reduced heaviness of the lines which define the dimension of context in the zero frame. The representation of the relevant substructure Y--CA (standing for froth and Carlsberg respectively) in the conceptual space of our shared knowledge indicates the enhanced prominence of this conceptual relationship in the base of the profiled meaning. The dotted lines represent a relation of conceptual identity (22).

3. The two consecutive arrows connecting both elements in this conceptual substructure (Y--CA) represent the two metonymic contiguity relationships which can be identified as relating the conceptual structures of froth and Carlsberg: part-whole and producer-product.

4. Finally, the double symbol ('><') between the two frames (or 'usage events') indicates that unlike regular puns (compare footnote 18), the first, most salient interpretation is not dismissed completely in favour of a second, less salient one. Rather, a kind of bisociation is established, in which the interpreter activates both meanings 'simultaneously' (23).

Basically the same analysis applies to the description of the examples in (11) and (12). There also, a witty, secondary interpretation is activated next to a contextually embedded salient one. And just like the Carlsberg-example, these two other cases also activate a second interpretation, in which (parts of the phraseological expression can be interpreted literally, thus demonstrating the cognitive plausibility of motivation and analysability as semantic characteristics of complex expressions. Next to the analysability on the lexical-semantic leve1 of the expressions (cause chronic pain and to be in the driver's seat), a successful, witty interpretation of the utterances also subsumes the activation of a (double) metonymic relation in the conceptual plane, linking the concept of these companies (Fiat, Bayer) to a prominent part of their products (driver's seat) or, in the case of Bayer, to a major objective (fighting pain) of their products (medicines).

To conclude this section, we briefly discuss two additional examples. The German economic headline in (13) instantiates the same semantic blueprint as the examples discussed above. It is a nice illustration, though, how merely one single constituent--and not the entire image component--may be activated in its literal meaning.

(13) Konsumflaute geht Bahlsen beim Umsatz auf den Keks (Neue Presse, 05/06/02)

('As for turnover, consumption lull gets on Bahlsen's nerves')

A crucial element in this headline is the use of the German expression jmdm. geht etwas auf den Keh, which means 'something gets on somebody's nerves'. In this expression, the constituent Keh, which literally means "cookie", is used metaphorically to refer to the human head.

As this expression is used in a headline about the financia1 situation of Bahlsen, a famous producer of cookies, both the literal meaning of Keh (lexical-semantic shift; figure/ground shift) as well as the metonymic link between Bahlsen and a cookie (Keks) as its product are activated, thus causing a slight effect of wit. Here also, the prominent conceptual background (base) of the utterance is realigned from the salient, contextually prominent domain of economics to common knowledge and experiences about the company's products. The only difference with the previous examples lies in the fact that in (13), the phraseological expression is not fully analysable to the extent that its image component as such does not represent a meaningful German expression. Therefore, unlike the Carlsberg-example in (10), this figure/ground shift only concerns the constituent Keks. Accordingly, the semantic structure of this headline may be represented as in figure 4, where 'KO' stands for Konsurnflaufe, 'BA' for Bahlsen and 'Y' for the central constituent Keks.


The last example in this section draws attention to the ecological validity of the integrated semantic representational model used in our present analyses. From a cognitive linguistic point of view, it is not unimportant to notice that an adequate description of the semantic structure of fixed expressions, in which both conceptual and discourse elements are accounted for, does not require a particularist treatment, in which phraseology-specific analytical tools are set up. It can be shown, instead, that basically the carne cognitive construal operations are active throughout the usage of (non-)fixed linguistic expressions.

(14) U.S. slowdown punctures Michelin's profits (Financia1 Times Online: www.ft.corn) reversal.

Although the headline in (14) does not contain a fixed expression, it causes a similar witty effect as in the examples (10)-(13) discussed above. Here also, both a conceptual and a lexical-semantic figure/ground shift determine the double interpretation. The conceptualisation of Michelin shifts by a double metonymic relationship from a company in its financia1 aspects to that company's products (tires) and from the product to a typical action affecting that product in a negative way (puncture). Next to its salient metaphoric interpretation ("diminish"), the verb puncture also profiles its literal meaning, thus connecting with the metonymic extension of Michelin. Compared to the examples above, the only difference lies in the absence of a complex structure, which identifies a semantic level for the single constituents, the image component as well as the overall phraseological meaning. The example in (14), on a lexical-semantic level, is restricted to the polysemy of a single verb.

5.2. Conversational Humour: trumping

As a final illustration of the way in which fixed expressions may be manipulated creatively, thus revealing the cognitive plausibility of motivation and analysability as semantic properties, we take a look at a specific type of conversational humor, labeled 'trumping' (24). In this conversational setting, adversaria1 agents exploit the linguistic-conceptual construal of each other's utterances in order to gain the upper hand in a humorous verbal duel. In doing so, an agent can trump an adversary by demonstrating a "hyper-understanding" of the lexico-conceptual structure of an opponent's utterance. This subversion of construal operations like metaphor, metonymy and salience leads to a sudden modification of the discourse space that has been set up in the previous utterance(s).

"In essence, trumping occurs when an initial utterance U by an agent S (the instigating speaker) evokes a counter-utterance U' from a second agent H (the responding hearer), where U' undermines U (and thus S) not by mere contradiction or non-acceptance, but by revealing U to be fundamentally unsuited to the communication intent of S" (Veale et al., in press).

It is crucial that the reply (U') by the second agent involves a clear parallelism with the first agent's utterance (U) in some key aspect, whether phonetic, lexical, structural or conceptual, in order to achieve the effect of neutralising U using S's own language choices. Without a substantial parallelism between U' and U, the second agent's response does not subvert U but is at best a mere refutation of U. Mere contradiction or disagreement does not constitute trumping; compare, for instance, the following example:

(15) S (consoling): There is always light at the end of the tunnel...

H (angry): Mind your own business, you idiot!

It is clear that H is adversaria1 to S, yet (15) is not an instance of trumping because H merely rebuffs S without exploiting any linguistic or conceptual element of U against S. The conversation in (16), on the other hand, provides a perfect illustration o f a trumping game (25).

(16) Emperor Charles the Bald (S): What separates an Irishman from a fool?

Irish philosopher John Scotus (H): Just this table.

The humorous effect in (16) hinges on the polysemy of the verb separates, which motivates the interplay of different construal mechanisms within the current discourse space. First, the verb itself can be used both in a metaphorical ("express contrast and difference") as well as a literal ("spatial disconnection") meaning, the situation of a philosophical discussion clearly favouring the metaphorical interpretation as the salient one. John Scotus seizes the opportunity and exploits the verbal polysemy by (hyper-)interpreting the verb in its non-salient, literal sense. As such, he still provides a valuable answer to the question. What is achieved, then, is a figure-ground reversal as the literal reading, which in the emperor's question resides in the base o f the profiled metaphoric meaning, is brought onstage as the newly designated meaning, relegating the metaphoric interpretation to the base (26).

The second construal operation that contributes to the humorous effect of this dialogue, is a figure-ground reversal in the conceptual organisation of the attentional frame, just like in the witty headlines discussed above. Through the use of the deictic demonstrative (this) referring to a specific table, the spatial situatedness of both speech act participants as well as the participants themselves enter the viewing frame (conceptual base) without being profiled (subjectification) (27). Interestingly, the introduction of this deictic element triggers another, third rearrangement within the current discourse space, as both noun phrases in the initial question (an Irishman, a fool are no longer assigned their salient generic interpretation, but an individuated, referential meaning instead. The philosopher being an Irishman, the emperor cannot be construed but as instantiating the concept FOOL.


The integrated semantic structure of this trumping game can be represented as in figure 6, where the first frame (minus frame) stands for the utterance by the speaker (S) and the second (or zero) frame for the trumping reaction (U') by the hearer (H). The polysernous structure of the verb separates ('V') is represented by 'X' and 'Y', which stand for the literal and metaphoric meaning respectively. The subject (What) in the question by S appears as 'A', the constituents Irishman and fool as 'I' and 'F'. The arrows related to elements in the ground indicate the different alignment of the two turns in this brief dialogue, whereas the bigger representation of H in the zero frame symbolises the winner of the verbal trumping game. In the zero frame, the answer by H (focus o f attention) activates crucial elements inside the viewing frame. The figure-ground reversal in the polysernous structure of the verb is achieved off-stage, but well within the viewing frame of the utterance's semantic structure. This figure-ground reversal is represented by the literal meaning ('X') appearing in bold face in the zero frame. 'The gravitation of the context as the most prominent element of the conceptual base ('philosophical question') towards the localisation of the ground inside the viewing frame, is symbolised by the heavier circle of the ground in the zero frame. Finally, the process of instantiation of the schematic concepts IRISHMAN ('I') and FOOL ('F') is indicated by the lowercase 'i' and 'f respectively in the zero frame. The dotted lines express a relation of referential identity between H and the Irishman ('i') and S and the fool ('f).

To close off our empirical survey, we take a look at two examples of a trumping constellation, in which the manipulation of a fixed expression contributes to the realisation of a humorous effect. In (17), a dissatisfied CEO complains about the presumed lack of efficiency and productivity in his company:

(17) CEO (S): I do the work of two men for this company!

(H): Yes, Laurel and Hardy.

Just like in the previous example, this brief dialogue witnesses a lexical-semantic figure-ground reversal. In this case, the fixed expression to do the work of two men is used by the first agent in its salient phraseological meaning "to work more than normally expected" which reflects positively on the agent. In his answer, H seizes the opportunity to reverse this value judgment by exploiting the same expression in its literal meaning in the conceptual base of his answer. By his positive answer to the question (Yes), H signals initial support for the CEO'S remark, but naming the names of two famous vagabonds immediately reverses this positive signal. Comparable to the example in (16), the second agent trumps the first one by providing an answer that instantiates the literal meaning of the (fixed) expression ("to do the work of two persons") in the question by the first agent. Whereas in (16), the referential answer by H (this table) triggers the literal meaning of the entire expression (separates), in the present case, Laurel and Hardy instantiates a single constituent of the fixed expression (two men). Nevertheless, the humorous effect of this dialogue also resides in the combination of both a lexical-semantic figure-ground reversal as well as a process of instantiation. With regard to the conceptual background of both utterances in (17), the salient frame of reference is the socio-economic ideal of hard work being positively evaluated. In line with the logic of this culturally determined value concept, two persons may indeed be considered to be more productive than one. In his specific reply, H undermines this inference all along with the entire logic behind it, promoting another cultural stereotype (the two incompetent, trouble-causing vagabonds) as the major frame of reference for S's initial remark. By doing so, the initial utterance by S generates a negative value judgment on the thematised productivity scale.


The semantic structure o f this trumping game can be represented as in figure 7. In the minus frame (at the left of the figure), the focus of attention centers around the profiled meaning ('Z') of the fixed expression to do the work of two men. The index '+' signals the positive value of this meaning. The symbol 'XY' stands for the image component (literal meaning) of the fixed expression with 'X' and 'Y' as a schematic representation of the single constituents. 'A' stands for the subject of the utterance and 'B' for the prepositional phrase for this company. The broken line represents the instantiation of the subject by S. The zero frame at the right, which represents the reply by H, profiles the concept of Laurel and Hardy as a contextually non-salient instantiation (broken line) of one constituent (two men: 'Y'). As such, this instantiation activates a strong inference of a negatively loaded concept of productivity, efficiency etc., hence the symbolisation as 'XY-.

In our last example, taken from Kotthoff (in press), the trumping mechanism resides in a creative exploitation of the image component of the fixed expression (to go right in one ear and out the other) used by S.

(18) S: Your nagging goes right in one ear and out the other

H: That's because there is nothing in between to stop it.

With the phraseological meaning of this expression ("not paying any attention to something") the utterance by S profiles a manifest lack of interest towards a previous utterance made by H. The repost by H, which actually profiles a concept of stupidity, hinges on the analysability at the level of its image component. As a matter of fact, the utterance by H activates a culturally determined cognitive model about the human head as the locus of intelligence. In line with this cognitive model, a prominent image used to refer to a certain lack of intelligence is the representation of the head as a (nearly) empty container (see also Feyaerts 1999). Accordingly, H elaborates this image component by adding a causal logic to it. More specifically, the utterance by H describes a direct cause which suits two different 'effects'. On the one hand, it provides an explanation for the phenomenon (literally) described by S, thus depriving S of any intentionality or control in this matter: the nagging goes in and out because of an undesirable malfunction in the head of S. Crucially, on the other hand, this reply actually profiles in a metonymic-contiguous way the concept of stupidity as the most fundamental effect of this cause. Although the reaction by H does not contain any explicit reference to the fixed expression, the literal meaning of the latter provides the essential element of parallelism (here also, a figure-ground reversal), which turns this dialogue into a case of conversational trumping. So, here again, it seems justified to consider a lexical-semantic figure-ground reversal to be the most relevant construal mechanism operating thoughout this humorous conversation.


This contribution has been an endeavor into the rich and dynamic semantics of phraseological expressions. Starting from two cornerstones of cognitive linguistics, the characterisation of meaning in terms of conceptualisation and discourse, we aimed at providing a dynamic and genuinely usage-based account of meaning. In order to get maximal exposure to the dynamicity of semantic structure we have focused our attention on two specific types of creative language use: witty newspaper headlines and trumping as a case of conversational humour. With respect to the semantic analysis of fixed expressions, this kind of data highlights the cognitive plausibility of semantic principles such as analysability and motivation as essential aspects of an integrated description of phraseological meaning. For both types of discourse investigated here, headlines and humorous trumping, it appears that the observed creative use of fixed expressions, which mostly involves the construal of a figure-ground reversal, heavily relies on the exploitation of both semantic principles. It would be interesting to conduct further systematic research on other types of creative language use in order to determine to what extent the semantic structure of phraseological expressions is as flexible as the meaning of other, non-phraseological expressions.

As a template for our empirical description, we have used Langacker's (2001) Current Discourse Space-model, which turns out to be an elegant and adequate account for representing semantic structure in a dynamic and integrated way. This adequacy derives from the observation that the CDS-model perfectly meets the two central requirements of a genuine cognitive approach to meaning analysis: 1. the representation of meaning in its inherent relationship to the rich, subject-related conceptual background, and 2. the incorporation of aspects of discourse.

Throughout the study we have run our analyses both on those utterances, which do and those that do not contain phraseological expressions. It has come out, crucially, that apparently no specific analytical tools are required to describe the semantics of phraseological units, and that, by hypothesis, the description of the conventional, non-creative use of fixed expressions does not require a phraseology-specific semantic model either. Without generalising prematurely, we may derive from this observation a certain ecological validity of this model of integrated semantic analysis. In this respect, fixed expressions occupy one point on a continuum of linguistic structures with which it shares a family resemblance to a greater or lesser degree and onto which the same cognitive construal mechanisms are applicable as onto any other linguistic structure.

By way of conclusion, it seems reasonable to claim that there is a mutual benefit in conducting usage-based analyses of phraseological expressions from a cognitive linguistic point of view. From the perspective of CL, it is interesting to pay careful attention to fixed expressions as they inevitably present a more complex and therefore more exploitable structure involving a clustering of different lexical elements as well as multiple meanings with different degrees of salience on different levels of linguistic organisation. From the perspective of phraseology research, it is interesting to adopt a holistic, non-particularist approach to the semantics of fixed expressions in order to gain maximal insight into the observation that throughout language, basically the same mechanisms of cognitive construal can be identified.


Alexander, Richard J. 1997. Aspects of Verbal Humour in English. Tubingen: Gunter Narr.

Balsliemke, Petra. 2001. ,,Da sieht die Welt schon anders aus"--Phraseologismen in der Anzeigenwerbung. Modifikation und Funktion in Text-Bild-Beziehungen. Baltmannsweiler: Schneider Verlag Hohengehren.

Barlow, Michael & Suzanne Kemmer (eds.). 2000. Usage-based Models of Language. Stanford: CSLI Publications.

Bell, Allan. 1991. The Language of News Media. Oxford/Cambridge: Basil Blackwell.

Brone, Geert. in press. Hyper- and misunderstanding in interactional humor. Journal of Pragmatics.

Brone, Geert & Kurt Feyaerts. 2003. The cognitive linguistics o f incongruity resolution: marked reference-point structures in humor. K.U.Leuven: Department o f Linguistics Preprint nr. 205. 58pp.

Brone, Geert & Kurt Feyaerts. 2005. Conceptual Integration as Discourse-Supportive Strategy in Economic Texts: A Case-Study for Headlines and Economic Cartoons. In G. Erreygers & G. Jacobs (eds.), Language, Communication and the Economy, Amsterdam/Philadelphia: John Benjamins, 73-99.

Brone, Geert; Kurt Feyaerts & Tony Veale. in press. "Humor research and cognitive linguistics: common grounds and new perspectives." Humor. International Journal of Humor Research. Special issue: Cognitive Linguistic Approaches to Humor. (Geert Brone, Kurt Feyaerts & Tony Veale, eds.), in press.

Burger, Harald. 1998. Phraseologie. Eine Einfuhrung am Beispiel des Deutschen. Berlin: Erich Schmidt Verlag.

Burger, Harald; Annelies Buhofer & Ambros Sialm. 1982. Handbuch der Phraseologie. Berlin: de Gruyter.

Croft, William & Alan D. Cruse. 2004. Cognitive Linguistics (Cambridge Textbooks in Linguistics). Cambridge: Cambridge University Press.

Cuyckens, Hubert; Rene Dirven & John Taylor (eds.) 2003. Cognitive Approaches to Lexical Semantics. Berlin: Walter de Gruyter

Dmitrij Dobrovol'skij & Elisabeth Piirainen. 2004. Figurative Language: Cross-Cultural and Cross-Linguistic Perspective.

Dor, Daniel. 2003. "On newspaper headlines as relevance optimizers." Journal of Pragmatics 35-5 (03), 695-721.

Feyaerts, Kurt. 1992. Hebt u ook iets op uw kerfstok? Fraseologisch gebonden woorden in het Duits en het Nederlands. In: Handelingen der Koninklijke Zuidnederlandse Maatschappij voor Taal- en Letterkunde en Geschiedenis. xlvi, 73-88.

Feyaerts, Kurt. 1999. Metonymic Hierarchies. The Conceptualization of Stupidity in German Idiomatic Expressions. In: G. Radden & K.-U. Panther (eds.): Metonymy in Language and Thought. Amsterdam/Philadelphia: John Benjamins, 309-332.

Feyaerts, Kurt & Geert Brone. 2005. Expressivity and metonymic inferencing: Stylistic variation in non-literary language use. Style. 38-1, 12-36.

Geeraerts, Dirk. 1995. Specialisation and reinterpretation in idioms. In: M. Everaert et al. (eds.): Idioms: Structural and Psychological Perspectives. Hillsdale, 57-73.

Geeraerts, Dirk & Peter Bakema. 1993. De prismatische semantiek van idiomen en composita. Leuvense Bijdragen 82, 185-226.

Giora, Rache1.1997. Understanding figurative and literal language: The graded salience hypothesis. Cognitive Linguistics, 8-3, 183-206.

Giora Rachel. 2003. On Our Mind. Salience, Context, and Figurative Language. Oxford: Ox ford University Press.

Hemmi, Andrea. 1994. ,,Es mu [SS] wirksam werben, wer nicht will verderben. " Kontrastive Analyse von Phraseologismen in Anzeigen-, Radio- und Fernsehwerbung. Bern: Peter Lang.

Koestler, Arthur. 1964. The Act of Creation. London: Hutchinson.

Kowalska-Szubert, Agata J. 1996. De kool en de geit. Nederlandse vaste verbindingen met een dier- of plantelement. Utrecht.

Lakoff, George. 1987. Women, Fire and Dangerous Things. What Categories Reveal about the Mind. Chicago: Chicago University Press.

Langacker, Ronald. 1987. Foundations of Cognitive Grammar. Vol. 1. Stanford: Stanford University Press.

Langacker, Ronald. 2001. Discourse in Cognitive Grammar. Cognitive Linguistics 12(2), 143-188.

Langlotz, Andreas. 2001. Cognitive principles of idiom variation: idioms as complex linguistic categories. Studii italiani di linguistica teorica e applicata 2, 289-302.

Langlotz, Andreas. 2006. Idiomatic Creativity. A cognitive-semantic model of idiom-representation and idiom-variation in English. Amsterdam/Philadelphia: John Benjamins.

Lee, David. 2001. Cognitive Linguistics. An Induction. Oxford: Oxford University Press.

Mena Martinez, Flor. (forthcoming) Cognitive Tools in Phraseological Research. In: A. Hacki-Buhofer and H. Burger (eds.) Phraseology in Motion. Hohengehren: Schneider Verlag.

Reddy, Michael. 1979. The conduit metaphor--a case of frame conflict in our language about language. In: A. Ortony (ed.): Metaphor and Thought. Cambridge: Cambridge University Press, 284-324.

Sabban, Annette. 1998. Okkasionelle Variationen sprachlicher Schematismen. Eine Analyse franzosischer und deutscher Presse- und Werbetexte. Tubingen: Gunter Narr.

Stockl, Hartmut. 2004. Die Sprache im Bild- Das Bild in der Sprache. Zur Verknupfung von Sprache und Bild im massenmedialen Texf. Berlin/New York: Walter de Gruyter.

Van Hoek, Karen et al. (eds.). 1999. Discourse Studies in Cognitive Linguistics. Amsterdam/Philadelphia: John Benjamins.

Veale, Tony; Kurt Feyaerts & Geert Brone. in press. "The cognitive mechanisms of adversaria1 humor." Humor. International Journal of Humor Research. Special issue: Cognitive Linguistic Approaches to Humor. (Geert Brone, Kurt Feyaerts & Tony Veale, eds.), in press.

Verstraten, Linda. 1992. Vaste verbindingen. Utrecht: LEd.

Wotjak, Barbara. 1992. Verbale Phraseolexeme in System und Text. Tubingen: Niemeyer.

(1.) I want to thank two anonymous reviewers and especially Flor Mena and Geert Brone for valuable comments on all earlier version of this article.

(2.) I use the notion fixed expression interchangeably with phraseological expression, considering both as cover terms for a series of other, not entirely synonymous expressions.

(3.) I borrow the notion of ecological validity from a paper by Vealc et al. (in press) on the exploitation of cognitive construal mechanisms in conversational humour.

(4.) Hence. We do not adopt a 'cognitive' approach in terms of Relevance Theory.

(5.) Compare, in this respect, Michacl Reddy's (1979) critica1 analysis of the Conduit-metaphor.

(6.) "As a rule, the image component is involved in the cognitive processing of the idiom in question. What this means for the semantic description of idioms is that relevant elements of the inner form have to be included in the structure of the senmantic explication". (Dobrovol'skij & Piirainen, 2004)

(7.) The image component depicts the 'literal' scene o f a person with water, straw or shit inside his/her head.

(8.) Unfortunately. this highly interesting book on 'Idiomatic Creativity' appeared after the final edition o f this article, so that no specific references could be included.

(9.) It would be erroneous and unjust with respect to the rich tradition of phraseology research to claim that CL has identified and articulated all of these mechanisms. It does seem to be the case though that a CL-inspired approach to phraseology has drawn systematic attention to them as inevitable components of an adequate semantic analysis.

(10.) See also Dobrovol'skij & Piirainen, 2004; Burger (1998: 660, however, restricts motivation to the occurrence of conventional linguistic relationships between the overall phraseological meaning and 'the' non-phraseological meanings of words and word groups, "die in ihrer freien Redeutung am Zustandekommen der phraseologischen Bedeutung beteiligl sind (...)" (ib.).

(11.) Compare, in this respect, Lakoff (1987: 451): "lt is important to bear in mind that we are, of course, not claiming that all speakers make complete sense of all idioms. Quite the contrary. There may be occasional idioms that are completely arbitrary for all speakers. (...) As one would expect, not all speakers make the same sense of all idioms. (...) Just as there are considerable speaker-to-speaker differences in the details of rules of grammar, and very great differences in vocabulary, so there are differences in the images associated with idioms

(12.) The outline of this section is largely inspired by Brone, Feyaerts & Veale (in press).

(13.) Compare Langacker (2001: 146) in this respect: "To think about linguistic units in isolation from usage events is at best an analytical convenience, and at worst a serious distortion."

(14.) Uncovering the tight interaction between contextual and linguistic understanding is the main research goal of the steadily growing field of cognitive discourse analysis (Langacker 2001, Van Hoek 1999). One of the advantages of such a perspective is that it can provide "a valuable corrective to the often-assumed dichotomy between cognitively-oriented studies, which often ignore the interactional aspects of discourse, vs. interaction models, which often de-emphasize cognitive processes" (Barlow and Kemmer 2000: xvii).

(15.) It should be noted that the notion 'ground' as it is used here, does not correspond either to the definition of 'ground' as the conceptual background in contrast to some foregrounded 'figure', nor to the notion of historical 'grounding' of a meaning as opposed to its synchronic motivation (section 3.2.).

(16.) See also Brone & Feyaerts (2003).

(17.) See also Brone & Feyaerts (2005).

(18.) See Giora (1997,2003) for a detailed account of the notion salience.

(19.) Unlike 'regular' puns, the cases presented here do not achieve their effect of wit by switching from one. contextually salient but erroneous interpretation to another, less salient one. Instead. a kind of 'bi-sociation' (Koestler 1964) of two habitually incompatible interpretations is established.

(20.) It may be clear that it is only legitimate to categorise this constmal operation in terms of a 'reversal' if one accepts that in the processing of the witty secondary interpretation, the phraseological meaning is briefly moved to the conceptual background (base) of the utterance.

(21.) This representation is inspired by the prismatic model of meaning as described by Geeraerts & Bakenia (1993).

(22.) I am aware of the fact that this visualisation makes abstraction of the difference between the symbolic nature of the elements of the utterance (inside the focus of attention) on the one hand, and mere conceptual elements which are not symbolised by any linguistic form.

(23.) Accordingly, we describe this semantic operation in terms of a figure/ground shift rather than a figure/ground reversal.

(24.) For an introductory discussion of this phenomenon, see Veale, Feyacrts & Brone (in press). The examples discussed here with a specific focus on the semantics of fixed expressions, also figure in this contribution. An in-depth analysis is provided by Brone (in press).

(25.) The example is taken from Veale el al. (in press).

(26.) It is justified lo speak of a true reversal as the metaphoric meaning, even after the semantic shift, still remains conceptually present in the scope of predication of the literal meaning.

(27.) It is, in fact the ground as such which turns into the viewing frame. Next to the participants and the local deixis, the speech act itself eiiters the scope of predication as well, as the unexpected answer requires a brief focus on the semantic structure of the utterance. Accordingly, the demonstrative operates as a 'grounding predication'.

Kurt Feyaerts *

University of Leuven

* Address for correspondence: Kurt Feyaerts, University of Leuven, Fac. Letteren, Blijde Inkomststraat 21, B-3000 Leuven, Tel: 32 16 32 48 05. Fa*: 32 16 32 47 67, e-mail:
COPYRIGHT 2006 Servicio de Publicaciones de la Universidad de Murcia (Murcia University Press)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Feyaerts, Kurt
Publication:International Journal of English Studies
Article Type:Report
Date:Jan 1, 2006
Previous Article:Sustainability of phraseological image in discourse.
Next Article:Occasional adnominal idiom modification--a cognitive linguistic approach.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |