8.6 Preference

The conflict that arises from 53 gives rise to a sequence of talk which confounds the usual literature on preference organisation, particularly that aspect of it to do with accounting. To investigate this the section begins with a reminder of the role of humour in preferred and dispreferred turns. Earlier it was added to the two separate tables of Levinson and Heritage, as follows:


Preferred Response

Dispreferred Response




Table 5. Preference organisation of the action of humour.

If we consider RB’s 53 as a humorous action we have seen that there is a mixed response - amused laughter from many and a serious criticism from BM. It is this latter response that shapes much of the rest of this stretch of talk and it is this response which will be focused on here. First we will consider possible reasons for a non-amused dispreferred response and then look at how, given such a conflict, accounting is done.

Labov and Fanshel (1977), when discussing the dispreferred second pair part of requests, that is, refusals of or putting off requests (in this case requests to do some dusting), provide the following possibilities.

Existential status:

Isn’t it dusted already? I did dust it.


Is it 3.00? It’s not the time I usually dust the house.

Need for the action:

It looks clean to me. Doesn’t it look clean to you?

Need for the request:

Don’t worry, I’ll do it. Don’t you think I’ll do it?


Do you know where the dust rag is? I don’t have time to
do it today.


Is it my turn to dust the room? It’s not my job.


I may if I’m in the mood. I don’t feel like it right now.


Who are you asking to dust the room? You’re not my boss.

Table. 7. Some dispreferred second pair parts to requests (p.87)

A similar taxonomy can be offered for the dispreferred second pair part of humour, that is, for non-amusement. (This does not aim to be exhaustive. We may like to recall the list of differing factors affecting humour support given by Carrell in 5.1 above.)


I don’t understand.

Actor’s incompetence:

The joke is badly/incorrectly told.


The joke offends my morals/beliefs.


I can’t express amusement here now (my boss is here/this
is a funeral.)


I hate the actor.


The joke is at my/my group’s expense.


I am not presently receptive to humour (I have a headache/
I have just received bad news.)


The actor is my inferior.

Table 8. Some dispreferred second pair parts to humour

As the focus here is on the dispreferred response of non-amusement, we will start from BM’s critical 59. This is an ideological response, condemning RB’s 53 as sexist. JK’s 61 (‘You tell him’) lends support to this ideological critique. (Her initial response of amusement will be considered below in the discussion on politeness phenomena.) Her ideological objections could well be twofold: doubly ideological – she objects in terms of her beliefs about women’s role in society and also because she is a Thatcherite Conservative; doubly as butt – she is a woman and also a Conservative MP, both characteristics she shares with the target of 53. EM’s dispreferred response could well be for some of the same reasons as JK; ideological (her views on women’s social role), and also as butt (she is a woman). (LL is seen laughing in response to 53 and so provides a preferred turn. She is silent for the rest of the extract.)

It needs to be remembered that in the textbook treatment of dispreferred responses the dispreferred turn is (1) delayed (2) may have delay components and (3) is accounted for. Recall this example in which all three features occur, as numbered.

C: Um I wonder if there’s any chance of seeing you tomorrow sometime (0.5)
morning or before the seminar
(1) (1.0)
(2) R: Ah um (.) I doubt it
C: Uhm huh
(3) R: The reason is I’m seeing Elizabeth
(Levinson 1983:309)

What happens in PI, however, is not that the respondents’ utterances have these key features, but, on the contrary, it is the actor’s speech which actually exhibits them. And at this juncture two points concerning accounts can be added. Scott and Lyman in their seminal work on accounts define an account as ‘a linguistic device employed when an action is subjected to valuative enquiry’ (1968:46), and Buttny sees the canonical accounts sequence as:

problematic event – blame – accounts – evaluation

Though this is not always the case with accounts – a cursory look at the example given immediately above shows that culpability is not always a factor – this model is applicable to the situation under review here. Certainly RB’s action is subjected to valuative enquiry (59), and certainly blame is apportioned (69):


[Straight-faced] Now see that to me is a sexist comment because you’re=


=Now that to

me was a joke



[To BM]

You tell him!






Yeah but you’re saying because she was strong>



Now wait a minute


>she had to be a man


No I think (.)


[Knowingly] Aaah!


Not because she was strong (1.5) because she was mean


[unclear] the same!




[unclear] the same!

[smiling] Shame!






But here it is RB’s 63 (‘Now wait a minute’) and 65 (‘No I think (.)’ ) which can be seen as delay components and his 60 (‘Now that to me was a joke’) and 67 (‘Not because she was strong (1.5) because she was mean’) as accounts for his action, the latter of which, note, also contains a significant delay. Although this confounds the usual conversation analytic findings on dispreferred turns, there is an explanation and this will be provided shortly. Of further interest at the moment is the way that BM, JK, and EM each in their own way reject RB’s various attempts to account which leads him, the actor, to yet further accounts.

Their challenges produce from RB what Davidson (1984) calls ‘subsequent versions’. Davidson discusses such actions as invitations, offers, requests and proposals, to which we here add ‘humour’. One of the situations she describes is ‘subsequent versions after actual rejection’ (p.107), something of obvious relevance for this discussion. This can lead to a sequence such as:

Initial version


Subsequent version



Naturally, subsequent versions can also be rejected, as in this example.

1. A:

Gee I feel like a real nerd you c’n ahl come up









4. B:

Nah that’s alright wil stay down he


5. A:


We’ve gotten


color TV



7. B:

tch hh I know but u- we’re watching the Ascent


’v Man, ’hh en then the phhreview: so: y’know wil


miss something if we come over


Thus, A’s initial version in 1 and 2 (an invitation) is rejected in 4 and the subsequent version in 5 and 6 is also rejected in 7-9. (Note also that this exchange follows the usual dispreferred pattern concerning delay (3), delay components (7), and account (7-9), all such features coming from the recipient of the action.)

In PI we get the following:

Initial version:

RB’s 53 Joke - ‘Thatcher was a man’


BM’s 59 Criticism - ‘That is sexist’


JK’s 61 Criticism - ‘You tell him’

Subsequent version:

RB’s 60 Account - ‘That was a joke’

Subsequent rejection:

BM’s 62 Criticism – ‘…because she was strong’

Subsequent version:

RB’s 67 Account – ‘No…because she was mean’

Subsequent rejection:

EM’s 68 Criticism – ‘…the same!’


JK’s 69 Criticism – ‘…the same!’ and Admonishment –


It is at this point that RB, angry that he has to account to interlocutors for his action to which they have responded with dispreferred turns, seeks a rational justification for his joke. We shall shortly return to this sequence with help from Pomerantz but first we note that the above sequence of versions and rejections fits neatly with Pomerantz’s view on the sequential consequences of pursuing responses.

If a speaker suddenly realises that what he or she had suddenly asserted is insulting or offensive to the recipient, he or she might modify the assertion in the direction of being less insulting and offensive. Part of the job would be to be convincing, to present the different position as a credible one.

It has been established that RB’s ‘different position’ is not seen as credible by the pursuers of an account – BM, JK, and EM.

As for RB’s move away from accounting to attacking JK (70), Pomerantz has another point which is relevant here. (It is recognised that it is also possible to see RB’s 70> as a continuation of his accounting via justification, and not as a move away from accounting.) She comments that a speaker, when meeting with a dispreferred response, can go over presumed common knowledge to see ‘what, if anything, is not established and accepted as fact’ (p.159). This is precisely what RB does with his questions about ‘Milk-Snatcher Thatcher’ through which he turns his defence into attack. That is, rather than having to continue his accounting himself, he tries to get some kind of account from one of those who have responded with a dispreferred turn. However, he has no support and his joke and accounts remain rejected. (Note that this episode is not the same type as that discussed earlier in 8.2.2 ‘assessing competence in the details of talk’. There is a significant sequential difference. The former was a kind of pre-sequence to clear the way for further talk; this latter comes after a problem has been encountered and so is a kind of repair.)

There are two further features of this part of the exchange that need to be discussed - its self-reflexive nature and what Thomas (1986) would call a ‘hierarchy of obligatingness’. First, self-reflexivity.

Cicourel notes the ability of talk to ‘fold back’ on itself (in Watson, R., 1992:7). Such activity Garfinkel and Sacks call formulating:

A member may treat some part of the conversation as an occasion to describe that conversation, to explain it, or characterise it, or explicate. Or translate, or summarise, or furnish the gist of it, or take note of its accordance with rules, or remark on its departure from rules. That is to say, a member may use some part of the conversation as an occasion to formulate the conversation…
(1970:350,original emphasis)

This is precisely what happens with BM’s 59 – ‘that to me is a sexist comment’. His talk is explicitly about the talk.

Heritage and Watson (1979) see formulations as ‘deeply implicative for subsequent talk’ (p.142), and the details of the above discussion would confirm that this is indeed the case for the stretch of talk under review. BM’s metacomment in 59 explicitly marks the end of the collaborative nature of the group’s talk. Heritage and Watson further see formulations as part of an adjacency pair: formulation – decision, where the decision about the formulation can be a confirmation or a disconfirmation. A disconfirmation ‘may minimally terminate an ongoing stream of topical talk and initiate a search for a fresh basis on which concerted comprehension can be established – thereby bringing some stretch of talk “back to square one”’ (p.144). Thus RB’s disconfirmation (60) of BM’s formulation (59) does lead to talk that tries (unsuccessfully) to explicitly establish a new framework, and this in turn eventually leads to a change of microtopic (92). This all tallies perfectly with the claim made in 8.4.4 that it is at the point in the talk when RB’s 53 is challenged that the first explicit definition of what is and what is not a fit subject for humour within the group is made. That is, it becomes necessary for the participants to lay down guidelines and this is initiated through formulations; until a ‘square one’ is explicitly established, the subject of the talk at this point is the talk itself.

Concerning the second point, ‘obligatingness’, Thomas remarks that it is not the case that in any given interaction illocution A is automatically followed by perlocution B. She suggests a hierarchy stretching from the minimally obligating, for example, phatic utterances, to the highly obligating, for example, summonses which name the addressee (1986:249). (She notes that such obligatingness differs between languages – in Russian and German, for instance, ‘Thank you’ is more obligating than in English.) All of this is made stronger if you include features concerning the relationship between the interlocutors, S and H: ‘The size of discoursal imposition is…determined by the degree of “obligatingness” + the power of S over H’ (p.250). (Once again, social relations have a strong bearing on matters.)

This leads us to ask just how obligating is a joke, and what of the power relations in this interaction? It is not easy to treat these two factors separately, for the degree of obligatingness to laugh at a joke is heavily context-dependent. Not only is power an issue, but such a simple factor as others being present is also a necessary consideration. If there are only two interlocutors, the requirement for the recipient to laugh when told a joke is greater than if there are many present, as in the latter case responsibility to be amused (a politeness consideration) is distributed in proportion to the size of the audience. Here the immediate audience layer is quite small (the three other panellists and the host) and they are all physically close to one another, so for the discussants, given their collaborative indulgences up that point, there would seem to be a reasonably strong obligation to display amusement.

Power is indeed also a factor, and we have seen that in, for example, Coser’s (1960) study of hospital staff meetings (1.2, 8.1), the degree of obligatingness is usually greater for those lower down the hierarchy. As for relations of our interlocutors, it was stated in 8.1 that the relations among the discussants were formally symmetrical, apart from the fact that BM, as host, has a certain institutional power over the guests. It is of note that what we get in response to RB’s 53 is an amused response from the majority of the studio audience and his equals on the panel but an offended response from BM, the person with institutionally more power. JK and EM do support BM’s objection once it is made, but their initial response is one of (perhaps reluctant) amusement. Such equivocal responses are heavily bound up with politeness and it is to this area we now turn.

8 Humour in Context: The Thatcher Joke>

8.1 The Performers, the Audience and Space>

8.2 The Discussants>

8.3 The Collaborative Floor vs The Single Floor>

8.4 Negotiating The Serious Import of Humour>

8.5 Two Psychological Perspectives on Joking>

8.6 Preference

8.7 Politeness>