Forms and Functions of Idiomatic Expressions in Conversational Interaction^*

Günter Schmale (Metz)

1 Introduction¹

A study of phraseological or formulaic expressions (FE)² in German talk shows (cf. Schmale 2012)³ revealed that participants almost systematically treat expressions which are semantically non compositional, thus by definition idiomatic, especially those containing metaphors and/or an images, in specific ways. In fact, participants auto- or hetero-rephrase, i. e. repeat, idiomatic expressions (IE) in a more or less identical form (cf. id. 2001b); they might also auto- or hetero-paraphrase them by a non-formulaic semantic equivalent or even another idiomatic expression or use an idiomatic expression to paraphrase a non-formulaic one (cf. id. 2007); participants make metalinguistic comments on idiomatic expressions (cf. id. 2009a), (word)play with them (cf. id. 2005a) or produce concomitant nonverbal activities (cf. id. 2005b) relating to elements of the idiomatic expression.

Schmale (2012) puts forward the hypothesis that conversational treatment of IEs in televised talk shows could be due to a particular effort made by participants when speaking publicly on television: producing utterances of a certain length or manifesting that they are able to create alternative formulations, indicating via metalinguistic comments that they are aware of using prefabricated – widely known and used – expressions, demonstrating creativity by (word)playing with IEs.

The present article aims to examine use and functions of idioms in non-televised conversation in order to establish whether similar or identical phenomena of what was called "conversational treatment" can be encountered in a type of speech-exchange system which is not or less marked by institutional influence on conversational organization, and, what seems most important, does not take place in public. Even if lexical choice in general and use of idiomatic expressions in particular do not seem to be imposed on talk show participants⁴, these can neither freely negotiate turn-taking⁵ nor deal with topics of their own choice. It equally seems likely that participants make particular efforts at correct and eloquent verbalisation when speaking publicly on television, for instance deliberately using idiomatic expressions so as to display their language competence. Non-televised, non-public, less institutionalised conversation, on the contrary, even if participants are aware of being recorded, is not subject to organizational and linguistic constraints of the same kind and could thus yield different results. Participants might not use metaphorical and/or imagery formulaic expressions or not treat them as they would in televised talk shows.

This hypothesis will be examined in four corpora of conversations in an "ecological" environment between participants pursuing a genuine communicative objective, that is to say that they are not talking about an imposed subject which they are supposed to produce speech about so that a linguist can record and use it for analysis.⁶ The authentic corpora studied are:

• [TEL] Deutsche Telefongespräche (Schmale (1982/83)): German telephone conversations in different types of private, commercial and institutional contexts⁷; 47 telephone conversations representing a total of 113 minutes of recordings from private, commercial, administrative contexts.

• [EiK] Erstgespräche im Krankenhausalltag (Walther 2005): admission interviews or rather informal conversations between patient and nurse in hospital; 47 conversations, most of them between 10 and 15 minutes long.

• [SSG] Sprechstundengespräche an der Hochschule (Boettcher et al. 2008): a university professor advising his students about their term papers, exams etc. during his office hours; 23 conversations transcribed of 10 minutes length on average.

• [KIOSK] Tagesgeschehen in einem Kiosk (Schmitt 2008): conversations between owner and guests of a very small sort of working-class café (a kind of refreshment kiosk); transcripts of 10 conversations from 3 to 44 minutes (average about 15 minutes).⁸

Before going into analytic detail in sections 3 and 4, the notion of idiomatic expression will be declined in section 2.

2 Idiomatic Expressions – Defining Criteria

Considering that the study of conversational treatment of FEs in German talk shows revealed that participants mainly and almost systematically treat FEs which contain a metaphor, by definition figurative, or an image which is not necessarily figurative or metaphorical, the present analysis of authentic conversations as to the possible treatment of FEs will equally concentrate on expressions containing metaphors and/or an images, i. e. on idiomatic expressions (= IEs).

The notion of idiomatic expression as employed in the present paper was adopted from Burger's (2010) classification of phraseological expressions. He first of all distinguishes referential, communicative and structural phrasemes. Structural formulaic expressions are for instance dual conjunctions such a neither – nor or with reference to, having for sole function the creation of grammatical relations (cf. id.: 36); communicative FEs⁹, playing an important role in the execution of communicative functions, such as good morning or excuse me (cf. ibd.). Referential phraseological expressions are being subdivided into nominative and propositional phrasemes. The latter possess sentence-value, either as proverbs (All that glitters…) or as common places (e. g. the truism You only live once.).¹⁰ The former are again subdivided into three different classes: idioms, partial idioms and collocations. Collocations are semantically compositional relatively fixed expressions, such as brush one's teeth or meet the demand; full idioms are semantically non-compositional and partial idioms contain at least one element having its "normal" meaning.

Following Burger's (2010) classification the present article will be dealing with fully idiomatic expressions which are semantically non compositional, i. e. no lexical element keeps its non-phraseological signification, so that there is – semantic – discrepancy between literal and phraseological meaning.¹¹

Three types of – semantic – idiomaticity can be distinguished:

(i)    IEs containing unique or archaic lexemes, "surviving" within an IE only, such as jemanden ins Bockshorn jagen ( not be irritated, discouraged, frightened, misled) or many a mickle makes a muckle, go haywire, take potluck.

(ii)  IEs including pictorial elements, images which do not correspond to a concrete model, i.e. which are not metaphorical as there is notertium comparationis between a logically reproducible source domain and an abstract target domain, thus no semantic basis whatsoever, e. g. die Radieschen von unten beschauen, manger les pissenlits par la racine, both meaning 'push the daisies', 'be dead and buried'.

(iii) IEs involving figurative elements which do represent an abstract state of affairs via a concrete model, being thus metaphorical, such as das fünfte Rad am Wagen ('be the odd one out', literally: 'be the fifth wheel of the cart'), einen Schlussstrich ziehen ('draw a line', 'wipe the slate clean'), zweigleisig fahren ('pursuing a two-fold strategy', literally: 'travel simultaneously on two tracks'). According to Burger (2010), this relation between – concrete – source and – abstract – target domain must be evident or obvious for the average speaker on the grounds of the respective language system without having recourse to the context.

So, whereas (i) and (ii) do not allow a context-free interpretation of the phraseological meaning unless the language user knows the meaning of the archaic element or the non figurative image, (iii) offers the possiblity of such a context-free interpretation to the average speaker by establishing – semantic – relations between the concrete model represented by the metaphorical expression and the abstract state of affairs designated.¹²

Burger (2010) distinguishes semantic – objective – motivation from intuitive, individual, situational motivation, i. e. idiomatic pictorial ("bildhaft") or figurative ("bildlich") IEs may be understandable, so to speak motivable¹³ by language users who (a) find the image plausible because of associations based on certain elements, (b) can interpret the IE on the basis of contextual elements, (c) possess etymological knowledge (cf. id.: 69). On the other hand, it is the average language user who achieves objective motivation on a semantic basis, but unless this average user is Chomsky "ideal speaker-hearer's" twin brother or sister, how can every single member of this family come to the same and identical – metaphorical – conclusion? What if a user does not interpret a metaphorical relation between source domain Nackenhaare hochstehn¹⁴ (cf. ex. (5) below) and target domain be very frightened because for him this image represents rather a modern hairstyle and not an analogy with certain animal's behaviour? Is it not a metaphore in this case?

Besides, Dobrovol'skij (1997) points out that native speakers do not hesitate to find explanations, that is to motivate the image of an IE, even if this does not correspond to – semantic – reality. In Farø's (2006) terms this would be – intuitive – "iconism" as opposed to semantically motivated "iconicity" and unmotivated "iconography", i. e. an image such as push the daisies which does not possess a semantic basis.

Following the analytic mentality of the ethnomethodological conversation analyst (cf. Schenkein 1978), which we are aiming to adopt in the present paper, analysing conversational phenomena in participants' categories, one might ask if manifest participants' interpretations do not have priority on analysts' or semanticists' objective judgments by means of semantic criteria. According to ethnomethodological conversation analysis the constitution of meaning is a hic et nunc interactive process, based on participants' reciprocal conversational activities. Analysing authentic conversation without applying the aforementioned perspective would seem counterproductive; we will therefore systematically take into account contextual and sequential features of IE use, in particular participants' activities subsequent to the production of an IE manifesting partners' interpretation.

3 Types of Conversational Treatment in Conversation Interaction

Our analyses of the four aforementioned conversational corpora proceeded by a qualitative approach rather than a quantitative automatic scan, i. e. reading through them page by page, line by line, word by word. Even if we had had at our disposal a genuinely complete inventory of German idiomatic expressions would we have found strongly modified forms on the one hand and idiomatic neologisms on the other hand considering that neologisms are lemmatized in official collections only years later¹⁵ – without even accounting for individual situational formulaic coinages (cf. Schmale 2011 or 2013a)? What is more, would an automatic search in a digitalized corpus be really more economical than simple reading through as one would have to check every single item on the pre-established idiom list which would take at least as much time as a thorough reading? Besides, would this method bring to light those instances of interactive conversational treatment in the centre of our study?

Conversational sequences containing idiomatic expressions, treated or untreated by participants were then transcribed according to adapted¹⁶ GAT-conventions (cf. appendix; Selting et al. 1998), taking into account preceding and following conversational activities of the idiomatic expression in question, in order to facilitate its analysis. However, English translations provided by the author of the present article do not cover entire sequences, but only the IEs concerned as well as preceding or following activities absolutely necessary to the understanding of the sequence being analysed by the non-German speaking reader.

The chosen procedure revealed the following phenomena of conversational treatment in the conversational corpora: auto- and hetero-rephrasals (pt. 3.1), as well as auto- and hetero-rephrasals of IEs (pt. 3.2), metadiscursive comments referring to IEs (pt. 3.3), and finally, distanciations as non-paraphastic reformulations of IEs (pt. 3.4). Section 5 examines cases of IEs which are not being treated by participants. Unfortunately, apart from one single example, nonverbal activities cannot be dealt with as they were not transcribed in the corpora of face-to-face conversations at our disposal. Furthermore, somewhat surprisingly considering the often joking manner in the conversations examined, no instances of obvious play on words on IEs were detected.

3.1 Auto- and hetero-rephrasals of idiomatic expressions

Both the producer of idiomatic IEs himself (3.1.1) and the addressee (3.1.2) can rephrase an IE in a subsequent turn, i. e. reuse it – apart from deictic and other minor changes – in its initial form. The analysis of several transcribed sequences of this type will reveal forms and functions of auto- and hetero-rephrasals.

3.1.1 Auto-rephrasals of idiomatic expressions

In sequences (1) and (2) the IE producer reuses his own figurative-metaphorical expression. In (1) he rephrases his own contruction in order to produce an auto-correction, self-initiated, a common method of repair.

(1)       Ulf and Hans talking about somebody called Monza. (KIOSK, Szene 5: 159-172.)¹⁷
  01      U    der monza wenn=d dem die finger gibsch musch-
                   'monza when you give him your fingers you must'
  02      ?     (               )
  03      U    wenn=d dem die hand gibsch musch die finger zähle;
                   'when you give him your hand you must count your fingers afterwards'
  04      H    die musch zähle die musch zähle;
                   'you've got to count them you have to count them'

In fact, in his first attempt to produce a phraseological metaphor¹⁸ speaker Ulf confounds Finger and Hand, he therefore interrupts his construction in order to self-correct his unfinished erroneous IE in a new – correct – construction (cf. (l).03). This type of rephrasal, in principle non-specific for IEs as any construction may be rephrased for the same reasons, constitutes the most basic form of rephrasal, its function being limited to the sole purpose of reparation.

Like H's partial hetero-rephrasal ((l).04) in (1), Ulf's auto-rephrasal in the following transcript (2) has functions which go beyond a simple correction. Talking about a letter written in a language he cannot understand and which he believes might contain information about illegal activities, U uses a metaphorical idiom (cf. (2).02) insinuating that only a trustworthy person must see or read this letter.¹⁹

(2)       Ulf (= U) has received a letter in an unknown language and does not know what to do with it. (KIOSK, Szene 2: 81-97.)
  01      U    ich weeß ned of des en linker brief is
  02      U    weesch der wo des in die händ kriegt-
                   'you know the one who reads this letter (lit.: lays his hands on it)'
  03      R    ach so
  04      U    des kann drinsteh
  05      U    weil des sin des sin zwee pole illegal vun vun drüwe rüwwerkumme […]
  12      U    un deswege muss des jemand in die händ kriege,
                   'and this is why only someone very reliable must read this letter'
  13      U    do wo ma sich druff verlosse kann

U then develops the reasons why only someone reliable must lay his hands on this strange letter, and, in order to implicitly verbalize the consequence of the wrong person reading it, reuses the same metaphor. This figurative IE conveys a rather complex state of affairs in a concise figurative form which easily and clearly links a concrete model (in die Hände kriegen) to an abstract state of affairs ( control something, have a certain power over it). To express its entire implications non-phraseologically would definitely need a greater number of constructions than the present IE. What is more, using and reusing a prefabricated expression provides cognitive relief for the producer, i. e. he does not have to create his own construction which does not imply, however, that the speaker does not actively have to integrate this IE into his turn and into his presentation of a state of affairs (cf. Gülich 2008). The general function of the IE in this thematic sequence is to support an argument in the opening and closing statement of the sequence, so to speak enframing his own statement and highlighting it via a highly illustrative or even vivid pictorial expression.

3.1.2 Hetero-rephrasals of idiomatic expressions

As in transcript (1), line 04, where the addressee H of an initial IE partially hetero-rephrases his partner's idiomatic expression, participant H in (3) rephrases the previous speaker's idiomatic expression. linke Sau is a non-metaphorical a priori non-compositional comparison²⁰ which again is highly idiomatic conveying a complex contents in a very concise form. Apart from facilitating H's formulation task, thus providing cognitive relief, it allows H to confirm U's statement and to simultaneously demonstrate reciprocity with his partner by using exactly the same formulation.²¹

(3)       Ulf and Hans talking about somebody called Monza. (KIOSK, Szene 5: 159-172.)
  12      U    is e ganz linki sau is des;
                   'he's a double-faced insidious bastard he is'
  13      H    des weeß ich dass des e linki sau is;
                   'I know that he's a double-faced bastard'

Transcript (4) shows that partial hetero-rephrasals (cf. (4).02) can also serve as an attempt to obtain further information about the nature of the Attentat ('attack') which is not immediately successful as E contents himself with a simple affirmation (cf. (4).03). Two more turns playing with the literal sense of the metaphorical idiom²² are necessary before the professor E finally details what the potentially dispreferred or harmful action consists in.

(4)       Beginning of a conversation between secretary and employer. (TEL (35) 108: 4-12)
  01      E    frau meier ich habe ein attentat auf sie vor;
                   'ms meier I'm planning an attack on you'
  02      S     ein attentat;
                   'an attack?'
  03      E    ja=a,
  04      S     tödlich oder (.) kommt man [noch mim leben davon,]
                   'fatal (lit.: deathly) or am I going to survive?'
  05      E                                                   [ich hoffe nich tödlich;]
                                                                'I hope you will (lit.: I hope not fatal)'

3.1.3 Brief summary of functions of auto- and hetero-rephrasals of IEs

Let us briefly summarize the different kinds of functions occupied by IEs in auto- and hetero-rephrasals.

Specific functions of IE-rephrasals are:

• the complex meaning/concise form function: participants rephrase a IE which has a complex meaning rendered in a concise form, a paraphrase of the same contents in a non-phraseological way could be a lot more cumbersome;

• the cognitive relief function: by (re)using a pre-fabricated construction one does not have to find one's own wording; using a IE in a specific context is nevertheless a creative act;²³

• the illustrative function: a pictorial and/or figurative IE could be particularly vivid or illustrative of the state of affairs designated, probably being more expressive²⁴ than its non-phraseological semantic equivalent;

• the enframing function: by (re)using a IE in the opening and closing statement of a sequence a participant may support his own argument.

General functions of rephrasals – not specific to IE-use – are:

• the repair function: to execute an auto- or hetero-correction or to initiate an intercomprehension sequence;

• the confirmation/questioning function: by – partially – hetero-rephrasing a participant may confirm, refute or question his partner's statement and/or IE;

• the reciprocity function: by hetero-paraphrasing and confirming partner's IE a participant demonstrates reciprocity in reusing the latter's construction.

3.2 Auto- and hetero-paraphrases of or by idiomatic expressions

Without wanting to draw a – in any way significant – conclusion from this fact, it can be observed that paraphrases are by far the most frequent type of conversational treatment in the four examined corpora, both by the producer of the IE himself and the addressee of the turn containing the IE. Three different types can be distinguished: idiomatic paraphrases of non-idiomatic expressions (3.2.1), non-idiomatic paraphrases of idiomatic expressions (3.2.2), and idiomatic paraphrases of another idiomatic expression (3.2.3).

3.2.1 Idiomatic paraphrases of non-idiomatic expressions

This category of auto- and hetero-paraphrases can again be sub-divided into three classes: metaphorical auto-paraphrases (3.2.1.1), metaphorical hetero-paraphrases (3.2.1.2), and non-metaphorical idiomatic auto-paraphrases (3.2.1.3).

3.2.1.1 Metaphorical auto-paraphrases

Metaphors – the "mapping from a source domain to a target domain" in Lakoff's (1992) terms – can be phraseological as in (5) and (6) hereafter.

(5)       Patient could not get used to a bigger car. (EiK, no. 7/Azubi: 105-109.)
  01      P     ne, und sagen wir mal, wat ich immer noch so hatte, das autofahrn,
  02      P     (hier) autobahn; da hab ich immer en bisschen panik gehabt;
                   'on the motorway I used to panic a little bit'
  03      P     lief zwar immer alles gut; kein unfall gehabt, aber-
  04      P     da (standen schon mal die nackenhaare hoch) ((lacht))
                   'it made my hair stand up on the back of my head'

Panik haben ((5).02) may be a collocation, but, owing to its compositional semantics, it is not idiomatic. The pictorial paraphrase in line 04, however, is strongly idiomatic and potentially metaphorical as the linguistic image depicted by Nackenhaare hochstehn may be interpreted as 'being panicked or frightened'.

The same possible metaphorical relation between source and target domain exists in the IE der mensch ist ein gewohnheitstier in (6), line 06,²⁵ which paraphrases the preceding construction ich hab mich dran gewöhnt which is then repeated in (6).07, following the metaphorical paraphrase, probably to strengthen the speaker's argument.

(6)       Left-handed patient P was very disturbed by an intravenous line placed on her left hand.           (EiK, no. 10/Azubi: 273-280.)
  01      P     plötzlich ham sie die eine hand nicht mehr da das is nich so einfach; […]
  05      P     doch ich hab mich inzwischen dran gewöhnt; […]
                   'but I got used to it eventually'
  06      P     der mensch ist ein gewohnheitstier;
                   (lit.:) 'we're all creatures of habit'
  07      P     ich hab mich inzwischen dran gewöhnt;
                   'I got used to it eventually'

In both (5) and (6) the metaphorical auto-paraphrase might be considered by their producer as more vivid or illustrative of the fact enunciated via the linguistic image employed. We would not go as far as to say, however, that figurative or pictorial expressions are principally more expressive than their non-phraseological equivalents (cf. Schmale 2010, 2013b), as this does not seem to be an idiom-inherent property, but rather dependent on a number of contextual factors.

3.2.1.2 Metaphorical hetero-paraphrases

Idiomatic metaphorical hetero-paraphrases can equally be produced by the addressee of a preceding non-idiomatic, though phraseological²⁶, expression.

In sequence (7), patient P supports and reinforces nurse K's affirmation, demonstrating reciprocity, by paraphrasing her construction – 'you're having a lot of fun' ((7).01) – by way of the metonymical²⁷ idiomatic expression jubel trubel heiterkeit ((7).02).

(7)       Nurse asking about patient's family and her grandchildren in particular. (EiK, no. 9/Azubi: 115-118.)
  01      K    a:h; (2.0) da ham se ja auch' da hat man auch immer viel spaß; ne,
                   'you're always having a lot of fun'
  02      P     ja; da is immer jubel trubel heiterkeit zu hause; ne,
                   'yeah your home is full of laughter and merriment/joy and good cheer'

This IE can also be considered as more vivid and illustrative, maybe stronger than the non-phraseological Spaß haben ('have fun').

kein Freund sein von ('not be keen on') ((8).05) in sequence (8) serves as a metaphorical paraphrase of K's preceding question which contains the non-idiomatic collocation eine Allergie oder Abneigung gegen etwas haben ((8).02). Rather than being strongly illustrative of the fact designated one might conclude to a euphemism which in itself implies a certain amount of expressiveness.

(8)       Nurse asking about patient's eating habits. (EiK, no. 14: 26-29.)
  01      K    so; essen und trinken können se ganz normal;
                   'so you can eat and drink normally'
  02      K    oder ham sie gegen irgendetwas eine allergie oder ne abneigung,
                   'or is there anything you are allergic to and you do not like at all'
  03            (2.0)
  04      P     ich ess norMAL ALles; (außer) weiße bohnen;
                   'I can eat normally everything apart from white beans'
  05      P     dat is das einzige, die ich' wo ich kein freund von bin;
                   'that's the only thing I'm (lit.: no friend of) not keen on really'

3.2.1.3 Non-metaphorical idiomatic auto-paraphrases of non-idiomatic expressions

Not every idiomatic paraphrase is automatically metaphorical.

(9)       Student having obtained a poor mark for his term paper is asking for explanations. (SSG, no. 16: 120-131.)
  02      P     dieses thema is son bisschen ähm is nich ganz so leicht in n griff zu kriegen,
  03      P     weil es ähm äh w' weil man einmal die verschiedenen altersgruppen hat,
                   'this topic is not easy to treat as you've got different age groups on the one'
  04      P     und verschiedene kontexte in denen die fremdsprachen lernen oder erwerben;
                   'hand and different types of contexts in which people learn foreign languages'
  05      P     ich kann jetz nich äpfel und birnen miteinander ((lächelnd)) vergleichen; ne,
                   'you cannot compare apples and oranges/you have to compare like with like'

The IE Äpfel und Birnen nicht miteinander vergleichen können (cf. (9).05) contains in fact a strong image which is not metaphorical, however, due to the fact that vehicle properties of the source domain ÄPFEL UND BIRNEN are in no way limited to the sole domain of DIFFERENCE, but also and primarily enclose FRUIT, HEALTHY, TASTY, PIPFRUIT and so forth. The tertium comparationis of source and target domain, condition sine qua non so as to be able to conclude to an obvious metaphorical relation between A and B remains in fact too vague. One could of course conclude that DIFFERENCE is being highlighted in Lakoff's acceptation of conceptual metaphor, the aforementioned properties such as FRUIT etc. being hidden.²⁸ However, in terms of a pragmatic approach (cf. Glucksberg 2001), determining metaphors via a context-sensitive linguistic analysis, the necessary relation between source and target domain is by no means obvious.

One would not even be tempted to classify the paraphrase in (10) as metaphorical. It is non-compositional because of the non-productive lexeme gemoppelt, part of the irreversible binomial doppelt gemoppelt, in which the first lexeme doppelt – typically – carries the signification of the complete expression.

(10)     Professor explaining how to draw a conclusion. (SSG, no. 5: 245-263.)
  01      P     wenn sie eine arbeit schreiben die selber nur FÜNFzehn seiten lang is; ne,
  02      P     .h dann hat es eigentlich kaum sinn am ende noch einma:l- (.)
                   'does it make sense at the end to once again'
  03      P     auf einer oder zwei seiten NOCH einmal ein fazit zu ziehen;
                   'on one or two pages once again to draw a conclusion'
  04      P     das is n bisschen doppelt gemoppelt; also m:-
                   'it'd mean doing the same thing twice (it'd be redundant)'
  05      P     nach fünfzehn seiten hat man als leser noch den überblick;

In both cases, the idiomatic paraphrase appears to express the initially verbalized contents in a more illustrative way, though probably not intensifying it, and, at the same time, in a humorous manner. As these paraphrases are produced by the professor who criticizes his student's term paper their function may reside in the domain of lessening face threat for the student via a humoristic formulation.

3.2.1.4 Brief summary of functions of auto- and hetero idiomatic paraphrases of non-             idiomatic expressions

Obviously IE-specific basic functions developed for rephrasals (cf. 3.1.3) such as the cognitive-relief function, the complex meaning/concise form function, the illustrative function also apply to idiomatic paraphrases of non-idiomatic expressions. One specific function has to be added:

• the demonstration of linguistic competence function: apart from being able to (re)produce a highly pictorial expression²⁹, the producer of an idiomatic paraphrase, especially when it is containing an image, may be considered as particularly – linguistically – competent.

Apart from the confirmation function and the reciprocity function, general functions of paraphrases – not specific to IE-use – are:

• the intensification through lengthening function: starting out from the hypothesis that a certain length adds to the force of an argument a speaker may produce an auto-paraphrase in order to add another construction to his turn;

• the downtoning of potentially face-threatening assertions through humorous final statements aiming at the constitution or preservation of reciprocity.

3.2.2 Non-idiomatic paraphrases of idiomatic expressions

A second class of auto- and hetero-paraphrases are those paraphrasing idiomatic metaphorical and/or pictorial FEs by non-idiomatic, i. e. semantically non-compositional though possibly phraseological expressions. Two types are to be distinguished: non-idiomatic auto-paraphrases of metaphorical expressions (3.2.2.1) or hetero-paraphrases of the same kind (3.2.2.2).

3.2.2.1 Non-idiomatic auto-paraphrases of metaphorical expressions

The most frequently encountered type within this class are non-idiomatic auto-paraphrases of metaphorical expressions.

(11)     Nurse explaining the aim of the admission interviews. (EiK, no. 7/Azubi: 204-207.)
  01      K    alle menschen sind verschieden; ne,
  02      P     ja;
  03      K    kann man nich alle eh eh so nehmen wie- (4.0)
  04      K    über einen kamm scheren;
                   'you can't tar everyone with the same brush'
  05      P     nee im gegenteil; (2.0) ich [sage immer, man müsste']     richtich;
  06      K                                                [man müsste alles indivi]duell'
  07      K    man muss eigentlich auch jeden individuell beTREUen;
                   'in fact one should treat everyone individually'

By paraphrasing his own potentially metaphorical image über einen Kamm scheren ('tar with the same brush') ((11).04)³⁰ via the non-phraseological construction jeden individuell betreuen ('treat everyone as an individual') ((11).06-07), nurse K, demonstrating her communicative competence, specifies the meaning of the idiom. Doing this she might strengthen her argument by making it perfectly clear – and maybe by making it longer.

Whereas the specification in (11) may be considered as redundant from a purely semantic point of view³¹, such a detailing paraphrase seems indispensable in following transcript (12). zweigleisig fahren ('pursue a two-fold strategy/lit.: travel on two tracks simultaneously') ((l2).02) does in fact not imply what this "two-fold strategy" consists in; the professor therefore has to explain to his student that he has to draw on two different sources – one for the autobiographical writings, one for the literary productions – for her term paper (cf. (12).03-07). This is why his non-phraseological paraphrase first and foremost has a necessary explanatory function in this case.

(12)     P explaining different editions of Goethe's complete works. (SSG, no. 3: 016-046.)
  01      P     wenn man sich jetzt entscheidet für ne gro:ße kommentierte goetheausgabe,
  02      P     dann muss man wohl zweigleisig fa:hrn im augenblick,
                   'in that case you'd probably have to pursue a two-fold strategy'
  03      P     .hh ehm für die: ehm autobiographischen zeugnisse […]
                   'as far as autobiographical materials are concerned'
  04      P     muss man die weimarer ausgabe zitiern; die GROsse alte sophienausgabe;
                   'you'd have to quote the weimar edition'
  05      P     aber bei den liteRArischen texten würde ich sagen, ham sie ehm die wa:hl,
                   'but concerning the literary texts I should say you've got the choice between'
  06      P     zwischen ENTweder der jetz gerade abgeschlossenen
                  'the edition of the klassikerverlag which has just been finished'
  07      P     ausgabe klassikerverlag, suhrkampverlag, […] oder der HANsaausgabe;
                  'or the hansa-edition'

To a lesser extent, the same explanatory function holds for the non-idiomatic paraphrase in next sequence (13) where patient P has to explain what her backing down consisted in. Producing this paraphrase, P might also lengthen and thus intensify her statement.

(13)     Rather than phoning the emergency doctor as she had wished the patient's neighbours   called an ambulance to take her to hospital. (EiK, no. 13/Azubi: 60-64.)
  01      P     (nicht wahr,) nachher konnt ich keinen rückzieher mehr machen; ne,
                   'afterwards I couldn't back down/cancel everything you see'
                   […]
  04      P     nee; konnt ich nich sagen, so nun fahrt wieder nach hause; ja is gut; ne,
                   'I couldn't tell them come on go back home it's alright you know'
  05      P     das geht doch nich; ne, [die] hätten mir das ja nich abgenommen; ne,
                   'I couldn't do that could I they wouldn't have accepted it right'

In sequence (14) hereafter, length certainly does not play a role considering the very brief paraphrase wie mans sacht of the metaphor³² der Ton macht die Musik ((14).12).

(14)     Nurse asking patient to tell her frankly what he did not like in hospital. (EiK, no. 7/Azubi: 244-246 + 249-252.)
  01      K    dann hab ich eigentlich nur noch den wunsch dass sie wort halten, […]
  03      K    mir noch mal sagen was ihnen alles NICHT gefallen hat; […]
  11      K    also ich persönlich find das SEHR in ordnung; und es is mit den'
  12      K    ich sach immer, da macht der ton die musik; nich, auch wie mans sacht.
                   'I always say it's not what you say you know but also how you say it'
  13      P     ja::; dat is' dat stimmt;

K might wish to explicate his metaphorical idiom by a semantically compositional paraphrase, although this does not seem very likely. It is far more probable that K produces the paraphrase because P does not take over the turn following K's turn-exit device nich (cf. (14).12), which induces P to keep the turn.

A paraphrase can also be reduced to a simple noun-phrase as in next sequence (15).

(15)     Nurse asking patient about his sleeping habits. (EiK, no. 17: 29-32.)
  01      K    und eh schlafen sie denn sehr gut? oder ham sie schlafstörungen? […]
  03      P1   ((lacht)) ich schlaf schlecht; ((lacht))
  04      P2   der? (.) der hat en ganzen urwald die nacht abgeholzt;
                   'he cut down an entire jungle during the night'
  05      P2   ja SOwas hab ich noch NI:E gehört;
  06      K    also SIE schlafen sehr gut?
  07      P2   so ein schnarcher hab ich mein Leben noch nich gehört;
                   'I've never ever heard anybody snoring like him in my whole life'

Even though the pictorial expression in question (cf. (15).04) is perfectly interpretable in a sequence dealing with sleep disorders, patient 2 delivers a paraphrase, wishing to insist on the disturbance caused by P1's snoring rather than delivering an explanation of his metaphor.

3.2.2.2 Non-idiomatic hetero-paraphrases of metaphorical expressions

As in previous types of conversational treatment, non-idiomatic paraphrases can also be pro­duced by the addressee of a preceding metaphorical expression. We shall examine three occurrences in order to find out whether their functions are identical to auto-paraphrases or not.

(16)     Nurse checking patient's blood pressure. (EiK, no. 18: 107-109.)
  01      K    ham sie schon mal gelegentlich bluthochdruck, oder-
  02      P     nee=nee; man sacht mir immer ich hätte en blutdruck wie en junger herrgott;
                                 'one tells me that my blood pressure is divine/absolutely perfect'
  03      K    ((lächelt))
  04      P     ob dat stimmt weiß ich nich;
  05      K    na dann schaun wir mal; ((K kontrolliert Ps Blutdruck, 12.0)
  06      K    wunderbar; hundertdreissich zu siebzich;
                   'wonderful yes a hundred and thirty to seventy'

In sequence (16) paraphrases patient P's idiomatic comparison wie en junger herrgott ((16).03) through a simple adjective – wunderbar ((16).06) – which he then substantiates by indicating P's ideal blood pressure. He so confirms P's statement and, at the same time, produces a "face-flattering act" (cf. Kerbrat-Orecchioni 2005).

In the following extract (17), on the other hand, nurse K's non-idiomatic paraphrase (cf. (17).05) is more than a simple confirmation, in this case she actually demonstrates or ensures understanding of patient P's fresh or idiosyncratic metaphor ((17).04).³³

(17)     Admission interview between nurse and patient. (EiK, no. 17: 52-53.)
  01      K    hatten se sonst noch schwierichkeiten, oder […]
  04      P     nö; einzige schwierichkeit is, ich hab ne zwiebel im portemonnaie;
                   'the only problem is I've got an onion in my purse'
  05      K    ah so ja; hm; [((lacht))] da fangen sie immer an zu weinen; [ne, ((lacht))]
                   'I see yeah hm you start crying do you'
  06      P                          [((lacht))]                                                             [ja; ((lacht))]

3.2.2.3 Brief summary of functions of auto- and hetero non-idiomatic paraphrases of             idiomatic expressions

Contrary to specific functions for IE-use in rephrasals (cf. 3.1.3) or idiomatic paraphrases of non-idiomatic expressions (cf. 3.2.1.4), non-idiomatic paraphrases of idiomatic expressions do not occupy IE-specific functions. As a matter of fact, the functions held by the non-idiomatic paraphrases discussed on the basis of sequences (11) to (17) are general functions which could equally be performed when non-formulaic constructions are being reformulated. Apart from the previously mentioned intensification through lengthening function and the demonstration of linguistic competence function, these general functions are:

• the specifying function, i. e. the non-idiomatic auto-paraphrase explains, defines, clarifies or specifies the state of affairs designated by the IE; this function is the basis of:

• the intercomprehension function, i. e. the paraphrase has the role to establish mutual understanding by facilitating comprehension for the hearer; or, in hetero-paraphrases, the hearer may check his own comprehension or explicitly demonstrate that he believes to have understood correctly;

• the intensifying function: by clarifying part of his previous turn a speaker may strengthen his argument;

• the confirmation function: through a hetero-paraphrase the addressee of an IE may confirm the IE-producer's statement.

3.2.3 Idiomatic paraphrases of another idiomatic expression

Far less frequent than the first two categories of paraphrases discussed in 3.2.1 and 3.2.2 are those instances where the producer of an IE himself or his addressee paraphrases the initial IE by another semantically equivalent IE. Once again three different types are represented in the corpora: metaphorical auto-paraphrases (3.2.3.1), metaphorical hetero-paraphrases (3.2.3.2), and non-metaphorical idiomatic hetero-paraphrases of another IE (3.2.3.3).

3.2.3.1 Idiomatic auto-paraphrases of another IE

In both transcripts (18) and (19), metaphorical expressions are being paraphrased by another metaphorical IE.³⁴

(18)     Student explaining the scope of her term paper to her professor. (SSG, no. 6: 018-057.)
  01      P     hm; .hh es is natürlich schon ein weiter kreis den sie da ziehn; ne,
                   'it's of course a large circle you are drawing isn't it'
  02      P     wenn ses einerseits mit de:m griechenbild eh etwa goethes, (.)
  03      P     und des e:h (.) deutschen klassizismus um achtzehnhundert vergleichen,
  04      P     andererseits jetz aber auch zurückgehn auf die (.) griechischen tragödien; […]
  06      P     n (1.0) eh sehr GROSser (.) kontext den sie da aufbaun;
                   'it's of course a very large context you're constructing'

Even if the sense of the IE (cf. (18).01) in the context of sequence (18) is not lemmatized in specialized dictionaries³⁵, a successful Google search substantiates our classification of the present form as a formulaic expression. The same observation holds true for its metaphorical idomatic paraphrase einen großen Kontext aufbauen ((18).06).

P employs both idioms in order to open and to close his statement about S's work, so to speak enframing his argument between two metaphors. One could, however, not conclude as to an intensification of his argument via the second idiom as the first metaphor seems a priori more illustrative of the state of affairs designated than the first.

In transcript (19), on the contrary, P's paraphrase by way of a fresh metaphor³⁶ (cf. (19).10) is definitely more figurative and thus more illustrative of the fact signified than the partially³⁷ metaphorical IE schleichender Verlauf ((19).08).

(19)     Patient has been hospitalised for myelophthisis. (EiK, no. 2/Azubi: 34-38.)
  01      K    wissen se was da so für folgen eintreten können (bei) knochenmarksschwund?
  08      P     […] der verlauf is SCHLEIchend;
                   'the disease is characterised by unpredictable relapsing progression'
  09      K    hm;
  10      P     es wird immer weniger; wie ne kerze immer weniger wird,
                   'it's getting less and less like a candle slowly burning down'

Rather than clarifying the signification of schleichender Verlauf, P's paraphrase aims to strengthen the expression of the fact enunciated via a highly illustrative metaphorical comparison.

3.2.3.2 Metaphorical hetero-paraphrases of another IE

Sequence (20) is an instance of an idiomatic hetero-paraphrase of an initial idiomatic expression by another speaker.

(20)     Student trying to defend a term paper having received a poor mark by her professor. (SSG, no. 16: 307-333.)
  04      S     irgendwie wars für mich reizvoller, […]
  06      S     das vielleicht aus ne:m differenzierten blickwinkel zu betrachten,
                   'it seemed more interesting to start out from as large a perspective as possible'
  07      P     nee d' also das d' das- auch in EHren; dass sie dann e:h gucken;
  08      P     wenn ich jetz eh SINNvolle methodische entscheidungen machen will; […]
  11      P     dann muss ich alles mögliche in blick nehmen; (.) nur ä:hm-
                   'in order to be able to make methodological decisions I have to integrate as many aspects
                    as possible'
  12      P     sie nehm dann alles mögliche in n blick, und es gerät ihnen aus der hand;
                   'and this is what you do but it's getting out of hand'

Both metaphors are situated in the same domain. After student S's initial IE, indicating that she tried to treat her subject from as large a perspective as possible (cf. (20).06), her professor employs a very similar metaphorical IE. However, not to confirm S's statement, but using it as a starting point to tell his student that her approach, commendable as such, is getting out of hand (cf. (20).11). P then even rephrases his own IE as a rhetorical means to develop his argument; basically S makes the right decision, i.e. she adopts a large perspective, but by adopting this perspective, she loses track: if x ((20).08), then y (IE in (20).11), but y (re-IE in (20).12) is unsuccessful.

3.2.3.3 Non-metaphorical idiomatic hetero-paraphrases of another IE

Hetero-paraphrases can of course also be non-metaphorical as in sequence (21) hereafter.

(21)     Nurse asking about patient's alcohol consumption. (EiK, no. 17: 25-29.
  01      K    trinken Sie gern irgend en wein, bier oder so, […]
  08      P     ja- wie sich dat so- [so']
  09      K                                   [ab] und zu mal?
                   'from time to time?'
  10      P     jojo; so alle jubeljahre;
                   'yeah once in a blue moon'
  10      K    ab und zu mal bier?
                   'a pint from time to time'
  11      P     so alle jubeljahre;
                   'once in a blue moon'

Nurse K's subsequently paraphrased question ab und zu mal? ((21).09) has non-compositional semantics and is therefore strongly idiomatic. Patient P answers by equally calling upon a highly idiomatic IE, containing a unique element, so alle Jubeljahre ((21).10). However, this IE not only provides an answer to K's question, it also modifies its implications. In fact, alle Jubeljahre could be far less frequent than ab und zu. This interpretation is confirmed by the following intercomprehension pair (cf. (21).10-11): whereas P could have contented himself by simply giving an affirmative answer, he insists on rephrasing his IE which, for him, is semantically more precise than K's formula.

3.2.3.4 Brief summary of functions of idiomatic paraphrases of idiomatic expressions

Evidently, most of the functions described for the first two types of paraphrases involving IEs also apply to paraphrases of an IE by another IE. This is the case for the complex meaning/concise form function, the illustrative function, which is most probably even reinforced by this type of paraphrase, the enframing function, definitely the intensification function, as well as the confirmation function for hetero-paraphrases. But the function particularly developed through idiomatic paraphrases resides in the demonstration of linguistic competence: being able to reformulate an idiom by another idiom is certainly a sign of developed language proficiency.

 3.3 Metadiscursive comments of metaphorical expressions

As opposed to the great number of different types of metadiscursive treatment of IEs in the talk show corpus (cf. Schmale 2012), participants only very rarely comment on their own idiomatic productions in the authentic conversational corpora studied. Hereafter two of the scarce occurrences encountered in the admission interviews at hospital.

(22)     Nurse asking patient to tell her frankly what he did not like in hospital. (EiK, no. 7/Azubi: 244-246 + 249-252.)
  01      K    dann hab ich eigentlich nur noch den wunsch dass sie wort halten, […]
  03      K    mir noch mal sagen was ihnen alles NICHT gefallen hat; […]
  11      K    also ich persönlich find das SEHR in ordnung; und es is mit den'
  12      K    ich sach immer, da macht der ton die musik; nich, auch wie mans sacht.
                   'I always say    it's not what you say but how you say it you know'
  13      P     ja::; dat is' dat stimmt;

In sequence (22), nurse K produces a pre-positioned metadiscursive comment ich sach immer ('I always say') on the metaphorical idiomder Ton macht die Musik (lit.: 'the sound makes the music'). Rather than indicating that the FE used is not her own production³⁸, K highlights the fact that she frequently employs it, implicitly admitting that it is formulaic.

In the following sequence (23), P's analogical and nicely illustrative construction mit musike commented subsequently by sag ich immer ((23).10), qualifying it consequently as formulaic via frequent use, the IE is effectively of his own invention.

(23)     Patient had to be taken to hospital in an ambulance. (EiK, no. 10/Azubi: 9-12.)
  01      P     ne also am samstag in der früh war ich VOLLkommen apathisch; […]
  09      P     und dann braucht ich, was mir noch nie passiert is- (.)
  10      P     ich meine dass se mich mit musike sag ich immer
                   'I mean they took me in an ambulance with music as I always say'
  11      P     im krankenwagen transportieren, das bin ich schon langsam gewohnt aber-
  12      P     eine sauerstoffmaske hab ich noch nie bekommen;

Contrary to a formula of the wie man so schön sagt type, which we find in the talk show corpus, the comment ich sach immer resp. sach ich immer does in no way indicate that the expression thus characterized is a generally known IE. It nevertheless carries the implication of frequent use which is one of the defining criteria of formulaicity.

Metadiscursive comments can consequently be pre- or post-positioned. They manifest the presence of formulaic language by indicating more or less frequent use as well as personal or common use of an expression. Participants possibly demonstrate linguistic competence in reflecting on their language use; they may equally take a precautionary communicative mea­sure by admitting that they frequently use a formula which might be considered as over-used.

3.4 Non-verbal activities referring to an idiomatic expression

As initially mentioned, only one single occurrence of a nonverbal activity referring to an IE was found in the conversational corpora.

(24)     Professor saying he would be delighted if a student suggested an exam topic by himself. (SSG, no. 1: 214-221 + 245-249.)
  01      P     wenn sie n gutbegründeten vorschlag haben; […]
  06      P     also ich würde in die luft gehen vor vor vor freude,
                   'I would jump for for for joy'
  07      P     wenn ich jemanden im seminar hätte, der mit
                   'if a student in one of my classes'
  08      P     <<auf den Tisch schlagend> so nem> ganz konkreten vorschlag ankommt;
                   '<<slapping his hand on his desk> came> with a well-defined subject for a paper'

Subsequently to having produced the idiomatic – metaphorical?³⁹ – expression in die Luft gehen vor Freude ('jump for joy') ((24).06), professor P slaps his hand on his desk in order to punctuate his turn and to make it more expressive.⁴⁰

3.5 Non-paraphrastic reformulative treatment – Distanciation or correction of an       idiomatic expression by another idiomatic expression

According to Gülich/Kotschi's (1995) classification of procedures of discourse production, the reformulative treatment of preceding utterances can be paraphrastic (rephrasals or para­phrasals) or non-paraphrastic. Apart from paraphrastic instances of discourse production, treated in the sections 3.1 to 3.4, the corpora also contain occurrences of non-paraphrastic treatment of previous constructions. Gülich/Kotschi (1995) subdivide these into distanciations of varying strength on the one hand and corrections on the other hand. In order to facilitate analysis the following three sequences will be treated as distanciations without going into analytic detail.

(25)     Ulf talking about a friend of his. (KIOSK, Szene 8: 428-437.)
  01      U    de borelli is=n guter freund vun mir;
  02      U    is zwar en linker vogel awwer des mescht nix-
                   'he's a double-faced/insidious kind of person but it doesn't matter'
  03      U    awwer en guter kern hot=er trotzdem; is=des richtisch, […]
                   'but he's not that bad after all (lit.: he's a good-hearted person at core)'
  04      U    we:sch wie=ch mä:n irgendwie hot=er=n gute kern;

U here reconsiders his own idiomatic non-metaphorical IE (cf. (25).02) by producing, following the contrastive coordinator or distanciation marker aber ('but') ((25).03), another idiomatic, this time metaphorical IE which relativizes the preceding, thoroughly negative idiom: x behaves very badly, but he is not profoundly wicked after all.

(26)     Patient talking about her asthma attacks. (EiK, no. 10/Azubi: 158-159 + 162-165.)
  01      P     ich hab noch en stückchen garten, das hab ich alles selbst besorgt;
  02      K    oh;
  03      P     und denk ich mir na ja, dann- wenn du das noch kannst,
  04      P     dann haste auch (die) asthma im griff; aber wie gesacht
                   'you think your asthma is under control but then'
  05      P     ein WETTERumschwung WIRFT mich jetzt toTAL aus der BAHN;
                   'a sudden change of the weather will totally knock you off the track'

Following the distanciation marker aber wie gesacht ((26).04) patient P produces a strong distanciation, invalidating almost totally her preceding IE im Griff haben, meaning that her asthma is under control. In fact, she has to realize that she is totally "knocked off the track" ("aus der Bahn geworfen") at every change of weather. This is why IE 1 im Griff haben and IE 2 aus der Bahn werfen are antonyms in this context.

The same observation applies to the following sequence (27) where IE 1 am Rande behandeln ((27).04) and IE 2 in den Vordergrund stellen ((27).11) are a priori antonyms.

(27)     Student of economics having problems with finding an exam topic for her subsidiary subject economic geography. (SSG, no. 1: 193-210.)
  01      S     das problem bei mir is, ich möchte das ganz gerne
  02      S     mein thema auch WIRTschaftlich (.) eh en bisschen bearbeiten;
                   'I'd like to treat my subject from an economic point of view'
  03      S     also mit wirtschaftlichem (.) HINtergrund, und es hieß zwar wunderBAR,
  04      S     aber ich sollte das eher nur am rande behandeln;
                   'but I was supposed to deal with it only marginally'
  09      P     eh aber ham sie ma den versuch gemacht,
  10      P     dass sie (.) eh n thema von sich aus vorschlagen;
                   'have you tried to suggest a topic (to your professor)'
  11      P     das diese ökonomischen aspekte stärker in den vordergrund stellt; (.)
                   'putting these economic aspects into the foreground (ie focussing on them)'

However, in this case, IE 2 does not invalidate IE 1, it rather supports student S's argument. She has developed that the professor in charge of her subsidiary subject does not accept an entirely economical subject for her final exam paper. This is why the professor in charge of her major subject suggests that she herself should propose a subject focussing on economic aspects. So even if IE 1 and IE 2 are semantic antonyms, the speaker using IE 2 does by no means invalidate S's point of view in this context, on the contrary, he supports it.

4 Conversationally untreated idiomatic expressions

As in the talk show corpus examined in Schmale (2012), not every single idiomatic expres­sion in the conversational corpora containing an image, be it metaphorical or not, is being treated by participants using the described conversational means. While the analytic mentality of the conversation analyst allows to sequentially reconstruct functions of conversational treatment of IEs containing an image, it is inconceivable for conversation analysis to elucidate why participants do not treat an IE via the described means.⁴¹ On the other hand, one can examine the procedures by which participants manage to understand IEs containing often opaque images without manifesting the slightest comprehension problem.

As a matter of fact, in the transcripts (28) to (35), reproduced hereafter, addressees do not even exert their fundamental right to initiate repair. If this fact can be interpreted as the absen­ce of an explicit manifestation of comprehension problems, it does not automatically imply addressee's – correct – understanding of the IE in question. However, given the – conver­sational – fact that participants produce without exception coherent, at least not rejected, next activities, it is likely that they globally achieve an interpretation – which does nevertheless not have to be semantically accurate – corresponding by and large to the expectations of the speaker whose turn comprises the IE concerned. Four different configurations are represented:

I. The addressee disposes of both semantic and co(n)textual clues for the "correct" interpretation of the pictorial IE.

(28)     Two students have come to discuss their excursion paper with the professor. (SSG, no. 11: 596-610.)
  01      P     wir haben am ersten standort auch eh diese verirrten (.)
  02      P     fingerhüte gefunden, die da so vor sich hinkümmerten; (.)
  03      P     eh die noch eingetragen sind vom waldrand. (1.0)
  04      P     ja, dadurch darf man sich dann nicht zu sehr (.)
  05      P     auf den holzweg führen lassen.
                   'you mustn't let yourself lead onto the wrong track'

Even though Holzweg is definitely a non-productive lexeme, surviving exclusively within this IE which moreover is synchronically speaking non-metaphorical⁴², participants may still interpret it adequately in its sequential context. As a matter of fact, one does not need the noun constituent Holz in order to interpret in a context-adequate way sich auf den (Holz)Weg führen lassen, especially as P negates his construction having as a verbal base the modal (nicht) dürfen, thus expressing a strongly dispreferred activity. So regardless of the knowledge of the term Holzweg P's addressee is able to produce a coherent next activity.

The same statement possibly applies to sequence (29) hereafter.

(29)     Shop assistant B is upset because she forgot to make a client pay for her tobacco. (KIOSK, p. 99: 22-25.)
  01      A    ich krieg jetz noch e bissel dabak [noch],
  02      B    [ou] ich hab de tawak vergesse
  03      B    dem peter abzuziehe; himmel arsch und zwirn;
                   'Christ almighty/bloody hell (lit.: heaven arse and thread)'
  04      A    du=sch des mol glei bei mir dezu;

Even if a hearer does not know that Himmel Arsch und Zwirn ((29).03) is a most vulgar curse, he can still deduce from the preceding turn – B has forgotten to make a client pay for his tobacco (cf. l. 02-03) – and the crude lexeme Arsch that B is expressively uttering her dissatisfaction, probably even that she is swearing.

II. The addressee disposes of semantic "leads", but of no co(n)textual ones.

In the following two sequences the hearer can resort to semantic clues, but not to contextual ones.

(30)     During a routine examination the patient's GP discovered an anomaly on her lung x-ray and recommended further detailed tests. (EiK, no. 23: 71-77.)
  01      P     dann hat er gesagt, (.) nein mit sicherheit is da etwas;
  02      P     muss sofort ein ct machen; […]
  04      P     dann hab ich himmel und hölle in bewegung gesetzt,
                   'then I moved heaven and earth'
  05      P     und hab direkt für den nächsten tag (2.0) e:h
  06      P     einen termin bekommen im josefs krankenhaus zur ct;

in Bewegung setzen ((30).04) is in fact a semantically largely transparent collocation, its accusative complement Himmel und Hölle can either embrace a totality of things or facts which are very different, signifying thus that P made a lot of effort. The context, though, does not provide any supplementary indications whatsoever.

In sequence (31), IE übers Knie brechen implies a violent and rash action while the context itself does not seem to supply any lead in this direction.

(31)     C is inviting his brother to tea. (TEL. (7), p. 11: 14-25.)
  01      C    woll=se rüberkommn,
  04      W    nee wei=ich noch nich wenn die als' (.) kartoffel aufgrabn;
  05      W    aber ich hänge jetz ziemlich in=er luft;
                   'I don't really know what to do right now (I'm suspended in mid-air…)'
  06      W    eigntlich hätt=ich heut namittach bei kuhlmanns arbeitn solln aber- […]
  09      W    ja::- weiß nich s=wär jetz (-) also n bisschn übers knie gebrochn;
                   'it'd be really a bit you know rash/precipitated if I'd come over right now'
  10      W    wenn ich kommn wollte,

III. The addressee has no semantic indications, but the context might provide them.

In sequence (31), discussed just before, another IE (cf. (31).05) is not metaphorical as the image in der Luft hängen does thus not correspond to the abstract state of affairs expressed, i. e. 'I don't really know what to do'. However, in this case the context provides the hints necessary to a context-adequate interpretation of the IE. In fact, the preceding – ich weiß noch nich ('I don't know yet') – and following construction – eigentlich hätte ich bei kuhlmanns arbeiten sollen ('normally I should have worked at the x's') – indicate that W is hesitating between two alternatives and is finding it difficult to make a decision. So even if partner C does not precisely decode W's IE, he can still produce a coherent next activity on the basis of contextual clues.

In following transcript (32),  the IE zwischen Tür und Angel ((32).03) is equally not a semantically motivated metaphor. Again the co(n)text could yield the necessary elements for an interpret­tation permitting the production of a conditionally relevant next turn.

(32)     Patient suffering from cancer talking about her experience with another hospital. (EiK, no. 23: 80-83.)
  01      P     das war natürlich ne ziemlich brutale [geschichte.                ja;]
  02      K                                                                 [im sankt barbara war das?]
  03      P     im sankt barbara. GANZ brutal zwischen tür und angel ham se mir gesagt,
                   'they told me very quickly/in passing (between the door and the hinge)'
  04      P     da is NICHTS, aber auch GAR NICHTS zu machen.
  05            'there is absolutely nothing we can do for you (P is suffering from cancer)'

brutale geschichte ((32).01) and GANZ brutal ((32).03) could actually trigger off an interpretation of the sequence corresponding to the IE-producer's attentions. Having said this, the understanding of the precise meaning of the IE zwischen Tür und Angel ('very rapidly, without going into any detail') is not really necessary to produce a coherent next activity. It is in fact not surprising that participants succeed in interpreting top down or on the basis of contextual clues without really understanding every single detail.

IV. The crucial case: neither the semantics of the IE nor its context of use are a priori of any help.

In all sequences (33), (34) and (35), neither contextual clues nor metaphorical relations between source domain and target domain can prima facie help to interpret the IEs employed by the respective speakers.

(33)     Patient has been taken to hospital as an emergency and to undergo surgery immediately. (EiK, no. 18/Azubi: 59-61.)
  01      P     und bin gestern abend noch ins krankenhaus gekommen;
  02      K    hm;
  03      P     in metten; und jetz ham sie ja gemerkt
  04      P     dass ich auch gerne in meiner heimat wohnen wollte,
  05      K    ja;
  06      P     da ham se denn nägel mit köpfen gemacht und ham mich-
                   'they crossed the Ts and dotted the Is (lit.: they made nails with heads)'
  07      P     ich find dat schön;
  08      K    ja das is gut (so);


(34)     Patient asking nurse to wish him good luck for his operation at the end of their exchange. (EiK, no. 28: 150-151.)
  01      K    gut herr müller; [das wars;]     das wars schon;
  02      P                               [halten=s] bitte-
                   'keep your please'
  03      P     halten se die daumen dat dä (.) dat se (richtich) (.) dat der wat findet;
                   'keep your fingers crossed that he'll find something'
  04      P     dat se mich dann nach dem (.) auf de berlitzer straße ins krankenhaus tun;
  05      K    werd ich machen;
                   'I will – promised'


(35)     Patient telling trainee nurse about her family and her son in particular. (EiK, no. 7/Azubi: 73-75.)
  01      K    is ihr sohn ihr einziges kind,
  06      P     (ja;) […] aber ich meine dat iss on hans dampf in allen gassen; ((lacht))
                   'but I mean he's a jack-of-all-trades ((laughing))'
  07      K    was macht ihr mann beruflich?

There are three basic explanations for the total absence of repair in these sequences: (a) the addressee understands the semantically opaque IEs because he knows their phraseological meaning; (b) he does not understand them completely or correctly, but still manages to produce a next activity which corresponds or at least is not opposed to the IE-producer's expectations; (c) the IE plays a marginal role in its host turn, its non-taking into account is thus negligible and not remarked by participants. Provided, this goes without saying, that the IE-producer manifestly reacted if he detected a misinterpretation on his partner's behalf. In fact, he could well just act as if everything was normal even though the addressee evidently misunderstood the intended meaning.

5 Conversational treatment of IEs – A step forward in the analysis of conversational    functions of idiomatic expressions?

In the quest to fill a gap in phraseological research as far as studies of authentic conversa­tional materials are concerned, and in particular, forms and function of referential idiomatic expressions, four corpora of authentic telephone and face-to-face conversations were studied. Following the results of a previous study of televised German talk shows, the research project presented in this paper examined different types of conversational treatment of formulaic expressions containing a metaphor and/or an image, i. e. preceding, concurrent or subsequent conversational activities referring to an idiomatic pictorial construction. Pictorial and/or figu­rative idioms were in the centre of the present investigation on account of the fact that they were undergoing almost systematically conversational treatment in the talk show corpus.

The study of authentic conversations revealed that six types of conversational treatment – by the producer of the IE himself and by the addressee – of formulaic expressions containing an image are achieved by participants: rephrasals (cf. ex. (1) to (4)), paraphrases of non-idiomatic expressions by idiomatic constructions (cf. ex. (5) to (10)), paraphrases of idiomatic expressions by non-idiomatic ones (cf. ex. (11) to (17)), paraphrases of idioms by another idiom (cf. ex. (18) to (21), metadiscursive comments of IEs (cf. ex. (22) and (23)), nonverbal activities relating to IEs (cf. ex. (24)), and finally, distanciations as non-paraphrastic reformulations of IEs in sequences (25) to (27).

Various conversational functions of rephrasals and paraphrases of IEs were elucidated. Those specific for IE-use are: the cognitive relief function, the complex meaning/concise form function, the illustrative function, the enframing function, the demonstration of linguistic competence function. Probably more general functions of conversational treatment of follo­wing, on-going or previous activities which are not necessarily specific for IE-use are: the intensification through lengthening function, the reciprocity function, the specifying function, the intensifying function, the intercomprehension function, the repair function, the confirma­tion or invalidation function, the downtoning function, and finally the distanciation function.

Subsequently to the analyses of conversational treatment of IEs, a reflexion on the absence of treatment (cf. ex. (28) to (35)) was proposed in section 4. As conversational analysis has no means to explain why a speaker does not treat an IE in the described ways, unless he metadiscursively comments on it, an attempt was made to explain via four different means of interpretation how participants succeed to understand often opaque images in IEs.

However, the question why participants do treat some pictorial IEs but not others remains crucial. Is it the nature of the image or the semantic structure of the IEs in question which is responsible for the absence of treatment? In Schmale (2012) we hypothesized that part of the idiomatic stock of a language, e.g. sich pudelwohl fühlen; die Quittung wird sich zeigen; auf jmdn. herabblicken; über Leichen gehen, is effectively memorized and used like a lexeme, possibly also due to the fact that the metaphor and/or image is so worn out that users do not perceive their pictorial nature any more. However, this might explain the absence of metadiscursive comments, aiming to demonstrate the user's linguistic awareness and competence, but it could by no means account for not producing rephrasals or paraphrases. It therefore seems far more likely that conversational treatment or its absence are motivated by speaker preferences, thematic development or contextual factors rather than the semantic nature of an idiomatic expression. Cotext, context and general knowledge of the world outrank in fact semantics as far as the interpretation of highly contextualized conversational activities is concerned. For the same reason, a context-free – thus obligatorily semantic – determination of metaphorical relations as suggested by Burger (2010) is not likely to produce reliable results. In fact, a language user might not be able to understand a metaphorical idiom such as die Nackenhaare stellen sich hoch ( the hairs stand up on the back of one's neck) when presented in isolation, but interpret its sense or function in an authentic context in a sufficiently adequate way to be able to produce a coherent or conditionally relevant next activity. Rather than confronting the "average" native speaker with tests explaining isolated linguistic metaphors, one should proceed by a corpus-based or corpus-driven approach which would yield empirical usage-based results.

References

Burger, Harald (1998/⁴2010): Phraseologie. Eine Einführung am Beispiel des Deutschen. Berlin: Erich Schmidt.

Coulmas, Florian (1981). Routine im Gespräch – Zur pragmatischen Fundierung der Idiomatik. Wiesbaden: Athenaion. (= Linguistische Forschungen 29).

Dausendschön-Gay, Ulrich/Gülich, Elisabeth/Krafft, Ulrich (2007): "Vorgeformtheit als Ressource im konversationellen Formulierungs- und Verständigungsprozess". In: Hausendorf, Heiko (ed.) (2007): Gespräch als Prozess. Linguistische Aspekte der Zeitlichkeit verbaler Interaktion. Tübingen, Narr: 181–219. (= Studien zur Deutschen Sprache 37).

Dobrovol'skij, Dimitrij (1997): Idiome im mentalen Lexikon. Ziele und Methoden der kognitiv­basierten Phraseologieforschung. Trier: WVT. (= Fokus 18).

Dudenredaktion (ed.) (1998): Duden 11. Redewendungen und sprichwörtliche Redensarten. Wörterbuch der deutschen Idiomatik. Mannheim etc.: Dudenverlag.

Farø, Ken (2006): "Ikonographie, Ikonizität und Ikonizismus: Drei Begriffe und ihre Bedeutung für die Phraseologieforschung". In: Linguistik online 27/2, 57–71. http://www.linguistik-online.de/27_06/faroe.pdf, accessed April 18, 2012.

Glucksberg, Sam (2001): Understanding Figurative Language. From Metaphors to Idioms. Oxford: OUP. (= Oxford Psychology Series 36).

Gülich, Elisabeth (2008): "Le recours au préformé: une ressource dans l'interaction conversa­tionnelle". In: Durand, Jacques/Habert, Benoît/Laks, Bernard (eds.) (2008): Congrès mondial de linguistique française. Paris, 9-12 juillet 2008. Paris, ILF/EDP Sciences: 93. http://www.linguistiquefrancaise.org/index.php?option=com_article&access= standard&Itemid=129&url=/articles/cmlf/pdf/2008/01/cmlf08315.pdf , accessed January 06, 2013.

Gülich, Elisabeth/Kotschi, Thomas (1995): "Discourse Production in Oral Communication. A Study Based on French". In: Quasthoff, Uta M. (ed.) (1995): Aspects of Oral Communication. Berlin/New York, de Gruyter: 30–66.

Kerbrat-Orecchioni, Catherine (2005): Le discours en interaction. Paris: Colin.

Lakoff, George (1992): "The Contemporary Theory of Metaphor". http://terpconnect.umd.edu/~israel/lakoff-ConTheorMetaphor.pdf, accessed November 15, 2013.

Lüger, Heinz-Helmut (1999): Satzwertige Phraseologismen. Eine pragmalinguistische Unter­suchung. Wien: Praesens.

May, Peter (2012): The Lewis Man. London: Quercus.

Mel'čuk, Igor (2011): "Phrasèmes dans le dictionnaire". In: Anscombre, Jean-Claude/ Mejri, Salah (eds.) (2011): Le figement linguistique: La parole entravée. Paris, Honoré Champion: 41–61.

Schemann, Hans (1993): Pons. Deutsche Idiomatik. Die deutschen Redewendungen im Kontext. Stuttgart/Dresden: Klett.

Schenkein, James N. (1978): "Sketch of an Analytic Mentality for the Study of Conversational Interaction". In: Schenkein, James N. (ed.) (1978): Studies in the Organization of Conversational Interaction. N.Y. etc., Academic Press: 1–6.

Redensarten-Index – Wörterbuch für Redensarten, Redewendungen, idiomatische Ausdrücke, feste Wortverbindungen. http://www.redensarten-index.de/suche.php, accessed November 12, 2013.

Sacks, Harvey/Schegloff, Emanuel A./Jefferson, Gail (1974): "A Simplest Systematics for the Organization of Turn-Taking for Conversation". Language 50/4: 696–735.

Schmale, Günter (1999): "Ich bin bei uns in der Stadt in aller Munde! So wie Bill Clinton? – Interaktive Behandlung vorgeformter Sequenzen in deutschen Talkshows." In: Fernandes-Bravo, Nicole/Behr, Irmtraud/Rozier, Claire (eds.) (1999): Phraseme und typisierte Rede. Tübingen: Stauffenburg, 159–171. (= Eurogermanistik 14).

Schmale, Günter (2001a): Le traitement conversationnel de phrasèmes dans les talk-shows de la télévision allemande. Unpublished Monography. Université de Nantes.

Schmale, Günter (2001b): "Rephrasages comme traitement conversationnel de phrasèmes dans les talk-shows de la télévision allemande". Beiträge zur Fremdsprachenvermittlung 39: 47–71.

Schmale, Günter (2001c): "Aspekte der verbalen Interaktion in deutschen Talkshows am Beispiel der interaktiven Behandlung von Phrasemen". In: Behr, Irmtraud (ed.) (2001): Télévision et Internet: le parlé et l'écrit. Asnières, P.I.A.: 29–56. (= Langue, Discours, Société).

Schmale, Günter (2005a): "Wortspiele mit phraseologischen Ausdrücken in deutschen Talk­shows". Deutsch als Fremdsprache 4: 215–219.

Schmale, Günter (2005b): "Nonverbale Aktivitäten bei der Äußerung von Phraseologismen". Studia Germanica Universitatis Vesprimiensis 9/2: 159–173.

Schmale, Günter (2007): "Paraphrases phraséologiques dans la conversation". In: Kara, Mohammed (ed.) (2007): Usages et analyses de la reformulation. Recherches Linguistiques 29: 163–175.

Schmale, Günter (2009a): "Metalinguistic Comments and Evaluations of Phraseological Expressions in German Talk Shows". Textes & Contextes 4, http://revuesshs.u-bourgogne.fr/textes&contextes/document.php?id=877, accessed November 12, 2013.

Schmale, Günter (2009b): "Phraseologische Ausdrücke als Bestandteil des Fremdsprachen­erwerbs – Überlegungen zur Phraseodidaktik auf der Grundlage einer korpusbasierten Analyse deutscher Talkshows". In: Bachmann-Stein, Andrea/Stein, Stephan (eds.) (2009): Beiträge zur Fremdsprachenvermittlung 15: 149–179.

Schmale, Günter (2010): "Ist ein idiomatischer Ausdruck immer expressiv? Korpusbasierte und fragebogengestützte Beobachtungen zu einer verbreiteten Prämisse". In: Kuiper, Koenraad (ed.) (2010): Yearbook of Phraseology I. Berlin/New York, de Gruyter Mouton: 97–124.

Schmale, Günter (2011): "Was ist in der Sprache 'vorgeformt'? Überlegungen zu einer erwei­terten Definition sprachlicher Präformiertheit". In: Schäfer, Patrick/Schowalter, Christine (eds.) (2011): In mediam linguam. Mediensprache – Redewendungen – Sprachvermittlung. Festschrift für H.-H. Lüger zum 65. Geburtstag. Landau, VEP: 177–190.

Schmale, Günter (2012): "The Conversational Treatment of Idiomatic Expressions in German Talk Shows – A corpus-based study". In: Pamies, Antonio et al. (eds.) (2012): Phraseology and Discourse: Cross-Linguistic Corpus-based Approaches. Baltmannsweiler, Schneider: 181–192.

Schmale, Günter (2013a): "Qu'est-ce qui est préfabriqué dans la langue? – Réflexions au sujet d'une définition élargie de la préformation langagière". Langages 189, 27–45.

Schmale, Günter (2013b). "Une expression idiomatique est-elle plus expressive qu'une expression non-idiomatique?". In: Chauvin, Catherine/Kauffer, Maurice (eds.) (2013): Écart et expressivité. La fonction expressive. Besançon, PUFC: 113–128.

Schmale, Günter (forthcoming): "Forme de base, variation, modification – Expressions idiomatiques en contexte conversationnel." Exposé dans le cadre du Colloque Variation, invariant, variété – Université Paul Verlaine, Metz, 22. und 23. März 2013.

Schreiber, David/Mahlow Cerstin/Juska-Bacher, Britta (2012): "Phraseologische Neologismen: Identifikation und Validierung". Yearbook of Phraseology 3: 3–30.

Selting, Margret/Auer, Peter et al. (2011): "A system for transcribing talk-in-interaction: GAT 2. Translated and adapted for English by Elizabeth Couper-Kuhlen and Dagmar Barth-Weingarten". Gesprächsforschung – Online-Zeitschrift zur verbalen Interaktion 12: 1–51. http://www.gespraechsforschung-ozs.de/fileadmin/dateien/heft 2011/px-gat2-englisch.pdf , accessed January 02, 2013.

Analyzed Conversational Corpora

Boettcher, Wolfgang et al. (2005): ' ich komm (0) weil ich wohl etwas das thema meiner hausarbeit etwas verfehlt habe' – Sprechstundengespräche an der Hochschule. Ein Transkriptband. Radolfzell: Verlag für Gesprächsforschung.

Schmale, Günter (1982/83): Telefongespräche. Ein Transkriptionskorpus. Unpublished. Bielefeld.

Schmitt, Reinhold (2008): Small Talk, Klatsch und aggressive Spiele. Ein Transkriptband zum kommunikativen Tagesgeschehen in einem Kiosk. Mannheim: Verlag für Gesprächsforschung.

Walther, Sabine (2005): Erstgespräche zwischen Pflegepersonal und Patienten im Krankenhausalltag. Ein Transkriptband. Radolfzell: Verlag für Gesprächsforschung.

Appendix – Transcription Conventions

Adapted by selecting specific elements from both fine and basic transcript following Selting/Auer et al. 2011.

Sequential structure

[ ] – overlap and simultaneous talk

Other segmental conventions

: – lengthening of sound (according to length)

' – cut-off by glottal closure

= – liaison phenomena (s=wär, nee=nee, hot=er)

n (= ein), eh/ehm/ähm (hesitation signals), ne (turn-exit device) – phenomena of spoken language

Final pitch movements of intonation phrases

? – rising to high

, – rising to mid

– – level

; – falling to mid

. – falling to low

Accentuation

SYLlable focus accent

Pauses

(.) – micro pause, estimated, up to 0.2 sec. duration appr.

(-) – short estimated pause of appr. 0.2-0.5 sec. duration

(--) – intermediary estimated pause of appr. 0.5-0.8 sec. duration

(---) – longer estimated pause of appr. 0.8-1.0 sec. duration

(0.5) (2.0) – measured pause of appr. 0.5 / 2.0 sec. duration

Laughing

((lacht)) ((lächelnd)) – laughing or smiling

In- and outbreaths

.h .hh – inbreaths (according to length)

h hh – outbreaths (id.)

Other conventions

((kontrolliert Blutdruck)) – comments on nonverbal activities or events

<<surprised> > – interpretive comment with indication of scope

(……) – unintelligible passage (according to length)

(may I) – assumed wording

[...] – omission in transcript