See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/257198811 Telling people what to do (and, sometimes, why): Contingency, entitlement and explanation in staff requests to adults with... Article in Journal of Pragmatics · May 2012 DOI: 10.1016/j.pragma.2012.03.014 CITATIONS READS 25 159 2 authors: Charles Antaki Alexandra Kent 117 PUBLICATIONS 2,471 CITATIONS 10 PUBLICATIONS 44 CITATIONS Loughborough University SEE PROFILE Keele University SEE PROFILE All content following this page was uploaded by Charles Antaki on 09 May 2014. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Available online at www.sciencedirect.com Journal of Pragmatics 44 (2012) 876–889 www.elsevier.com/locate/pragma Telling people what to do (and, sometimes, why): Contingency, entitlement and explanation in staff requests to adults with intellectual impairments Charles Antaki *, Alexandra Kent School of Social, Political and Geographical Sciences, Loughborough University, Loughborough LE1 3TU, UK Received 17 January 2012; received in revised form 28 March 2012; accepted 30 March 2012 Abstract How do support staff resolve the interactional dilemma of getting their clients to do things, while respecting their independence? In a corpus of over 200 everyday requests made by residential home staff to adults with an intellectual impairment, the staff tended to use formats which claimed high entitlement to be obeyed, and made little acknowledgement of the contingencies facing their interlocutors. Bald imperatives were overwhelmingly the most common format used. The findings suggest that staff resolve the dilemma of care and control mostly in favour of getting jobs done, at the expense of residents’ potential trouble in fulfilling their requests. In the rare cases where requests were accompanied by explanations, these legitimised the staff member's entitlement, or showed their awareness of the contingencies that could affect the resident's response: this provides useful evidence of the reality of these categories to the participants. We discuss three factors that might influence the degree of directiveness in the request: the physical immediacy of the action; a prior fault; and an obligation of the requester to instruct and socialise. © 2012 Elsevier B.V. All rights reserved. Keywords: Requests; Directives; Imperatives; Entitlement; Contingency; Intellectual impairment; Institutional talk; Conversation analysis 1. Introduction People can be requested to do things in a variety of ways. Two highly significant elements to the request, systematised by Curl and Drew (2008), are the degree of entitlement that the requester claims for themselves (a factor prefigured in Heinemann, 2006), and the degree to which the requester seems to appreciate the difficulties, or other contingencies, that might hinder the recipient performing the requested action. To take two invented examples, the imperative do X! claims greater entitlement, and less concern with the recipient's contingencies, than a formulation such as I’d be grateful if you could do X. These elements are interactionally useful, as they allow the requester to adjust the exact wording of their request according to how much entitlement and allowance for contingency that they want to project. Curl and Drew (2008) show that the adjustment will be made in the light of the circumstances that the requester finds themselves in, vis-àvis the request recipient. This tells us a great deal about the interactional status that both the requester and the request-recipient orient to. As we shall try to show in this article, it also gives us a sharp profile of the institutional world in which they live. * Corresponding author. E-mail address: c.antaki@Lboro.ac.uk (C. Antaki). 0378-2166/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.pragma.2012.03.014 Author's personal copy C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 877 1.1. Entitlement and contingency As an illustration of the way the dimensions work, Curl and Drew (2008) show that, for example, if you make an out of hours call to the doctor, you may mark your lack of medical expertise, and the likelihood that your request will cause the doctor inconvenience by using an I wonder if. . . formulation (such as I’m wonderin’ if a doctor could call and see [name] please). That shows your low entitlement in making the demand, and your lively awareness of the doctor's contingencies. Whether or not patients do or do not (really) have such a low entitlement – however that ‘really’ might be determined – the case is that, as the authors put it, people can ‘‘construct themselves as potentially lacking entitlement’’ (Curl and Drew, 2008:148). On the other hand, if one wants to project having a very good case for calling the doctor, one may use a request which hinges on a modal verb like could, may or will (for example: can the doctor come out as soon as possible?). This is less common in calls to the doctor than in calls to family member, where presumably one may have more grounds for signalling a greater expectation of one's wishes being satisfied, and less concern for the recipient's contingencies (to use an example from Curl and Drew: can you come over in the morning? Curl and Drew, 2008:137). The great benefit of Curl and Drew's empirical, Conversation-Analytical approach is that it frees us from a priori estimates of face, or of speakers’ pre-existing social roles, which can vitiate more traditional pragmatic treatments of requests (see Curl and Drew:130–135 for a fuller discussion of such problems). What the authors identify are displays, under the speaker's control, designed to meet the demands of the moment. Formats which display the requester's tentativeness, uncertainty and so on, or its opposite, mark the degree of the speaker's entitlement; formats which display their appreciation of the recipient's needs and wants, the troubles they might face, and so on, or fail to do so, mark the degree of their awareness of contingency. By finding that such displays could be laid out on these two dimensions, Curl and Drew have given researchers a useful framework to consider what happens in various parts of the social world. Craven and Potter (2010) apply them to requests at family mealtimes; Aronsson and Cekaite (2011) to requests in family life more generally; and, in a different environment, Keisanen and Rauniomaa (forthcoming) to exchanges between car drivers and passengers. Attention is also being paid to entitlement and contingency in requests in the institutional world (see, for example, Kuroshima, 2009, on ordering in a restaurant; Lee, 2011 on airline service encounters, and Harris et al., 2011, on counselling). In this article we want to stay in the institutional world, and consider the case of requests made by staff members to adults with intellectual impairments. We shall see that the status of staff and residents – and the presumptions about the rights of one and the other – is sharply captured by the way the staff use entitlement and contingency in their requests. 1.2. The status of request recipients We shall be looking at adults with intellectual disabilities, but is there any evidence already that it might matter to whom the request is made? Indeed there is. In studying parents’ requests to their children, Craven and Potter (2010) reaffirm the utility of the Curl and Drew dimensions, but extend them to show that parents at mealtimes preferentially use more directives than requests, claiming still more entitlement and showing even less concern for their children's contingencies when compliance is not forthcoming. There were a great deal more examples of bald imperatives in the parents’ talk – for example, hold it with two hands or don’t play – which did not feature in Curl and Drew's adult data. Intuitively, it seems likely that adults and children are treated differently when it comes to getting them to do something, and no doubt for multiple reasons. Strongly entitled directives fully restrict the response options to just compliance (sometimes through forced physical manipulation). This provides a very strong indicator to the recipient of what constitutes acceptable behaviour in that specific situation. In adult–adult conversation, entitled directives appear to be invasive and face threatening social actions (Brown and Levinson, 1987). However, the same might not necessarily be true for children. Among many other such sentiments, we might quote Shakespeare (1998) who points out that ‘‘because children are not effectively full members, much of their lives is spent in social interactions that offer them directives concerning how to achieve full membership’’ (1998:25). Having a clear steer towards an appropriate and acceptable response action may in fact facilitate successful participation in the interaction, and offer a scaffold around which the child can build their set of discursive resources. This gives us the background against which to ask what happens when the adults are people with intellectual impairments, who occupy an interactionally marginal social role, with some of the elements of childhood disentitlement and underprivilege. We should be careful in making any sort of equation between the abilities and experiences of children and adults with intellectual impairments – for one thing, the terms are too imprecise to allow any sensible tally – but in terms of their (attributed or allowed) rights and obligations, at least, some useful comparisons can be drawn (even so, this is still a large topic; for a view from an interactional perspective, see Williams, 2011). This article, then, is an empirical investigation of a situation where the recipient of the request does not have the full range of cognitive competences of the typical adult. The question is: how do staff members treat them, when issuing requests? What entitlements to obedience, and what awareness of their clients’ difficulties, do the staff display? Author's personal copy 878 C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 1.3. Requests and directives The modal and I wonder if. . . forms of request, between adults in ordinary conversation, show that the speaker is displaying at least some awareness of their lack of entitlement and some appreciation of the contingencies under which the recipient might have to perform that action. The directives that Craven and Potter discover parents using at mealtimes (hold it with two hands and so on), however, show neither of these. Because we shall soon see that directives are very common in staff members’ talk with the intellectually disabled people they support, we need just to consider then whether directives are a separate kind of interactional phenomenon, which act outside the framework empirically established by Curl and Drew, and which we shall use in the analytical body of this article. For Craven and Potter, directives’ lack of display of a concern for the speaker's entitlement, or of the contingencies constraining the recipient, means that they are qualitatively different from the kind of requests studied by Curl and Drew. Their case is that ‘‘Requests have to be accepted before they can be performed; directives just need to be complied with. In this respect requests and directives are performing two different actions.’’ (Craven and Potter:426) This seems, on reflection, to be a grammatical distinction without a practical difference: although the recipient of could the doctor come. . . or I wonder if the doctor could come. . . is obliged to produce a response in the appropriate format (yes or no, in various forms) the real force is of course pragmatic, and a yes without the despatch of the doctor would be a noticeable lapse. Indeed Craven and Potter then go on to document that directives, when they initially fail to get a response, get progressively more demanding; and they get more demanding along exactly the two dimensions identified by Curl and Drew – the parent presuming greater entitlement to be obeyed, and more indifference to the child's contingencies. Although, as Craven and Potter show empirically, ‘‘The talk of the speaker who issues the directive is not oriented to acceptance; their talk orients entirely to compliance’’ (p. 438), that is also true of requests; if there is a difference, it is that the request has a (purely formal) middle step of requiring verbal agreement before the ‘real’ response of physical compliance. The upshot of this discussion of Craven and Potter's argument is that we may proceed with a consideration of directives without dwelling on their supposed qualitative difference from requests; the difference is well catered for by the difference in degree of speaker entitlement and indifference to recipient contingency. What we do get from Craven and Potter is that it is these dimensions which are seem to be the crucial factor in the initial sense of the utterance as a directive to action. As Craven and Potter observe: ‘‘What is striking about the directives we have discussed . . . is that they orient to neither capacity nor desire. In the extreme case this does not just involve verbally directing, but issuing threats, or physically moving the recipient to the required position.’’ (Craven and Potter, 2010:438). We have a platform, then, for seeing how they work in a new, but potentially analogous situation: the interactions between staff and adults with intellectual impairment. There are two elements to this scene that pull the requester in different directions. On the one hand, the people who will be asked to do things are adults, not children; and, ostensibly, they have adults’ social rights. Moreover, the staff who work with these adults are employed precisely to facilitate their lives, help them make choices, and carry out their wishes; on paper, at least, they are there to ‘support’ and not lead (see Clement and Bigby, 2010, for an account of care-home staff responsibilities). On the other hand, adults with intellectual impairments can find everyday tasks difficult, and can be slow to understand and comply with help and advice; and staff are mandated by their job descriptions to make sure that institutional tasks be done efficiently. The first set of considerations would encourage requesting – that is to say, following Curl and Drew, they would encourage the staff to show some lack of less than full entitlement to make the request, and to make some acknowledgement of the contingencies facing the resident. The second set of concerns, on the contrary, would point towards showing greater entitlement, and less concern for the resident's situation compared to the institutional demands of the moment. There is, then, a potential difficulty here. How do support staff solve the dilemma of direction (which gets the job done) and laissez-faire (which respects the recipient's autonomy)? 2. Data The data extracts presented here come from videos shot in an ethnographic study1 of residential services for adults with intellectual disabilities provided by a National Health Service Trust in the South of England. These data2 come from one site, where five male adults lived together in an ordinary suburban house, supported by a rotating set of care-workers, two of whom were on duty at any time. The men had a range of communicative abilities. All could apparently follow simple conversations, but they varied in how articulate they were in producing speech. The staff's duties were to supervise and 1 The project was funded by the Economic and Social Research Council of the UK, grant no. RES-148-25-0002. I’m grateful to Chris Walton who made the video recordings for the project, after gaining the trust of residents and staff. 2 Ethical approval for the research was sought and gained from the author's institution, which held the grants which funded the research, and from the relevant Regional Ethical Committee. Written permission to record and publish data was granted by all participants. All names of speakers, and of people and places mentioned, have been changed. Author's personal copy C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 879 help plan daily activities like cooking, shopping, social excursions and so on, and to oversee the good functioning of the residents’ general living arrangements. To find requests in their various formats, we chose five episodes, representative of everyday staff-resident interaction in the home: food-preparation in the kitchen; laying the table for a meal; a meeting to plan holidays; a meal; and an evening in a social club. Those five episodes together yielded 3 h 27 min of recording, and about 234 tokens of requests of various kinds. 2.1. Caveat No great store ought to be set by the exact numbers of requests, for two main reasons. One is that there are alternative ways to count repeated requests of the same kind within the same turn or across subsequent (but unresponded-to) turns; we chose to count them as one, so as to be conservative with the frequencies, but it would have been legitimate to choose otherwise. (Thus where the staff member says turn it on. (1.5) on, (1.0) turn it on, we conservatively count that as one directive, not three.) The other is that there is a reasonable case for either counting or not counting subsequent different request formats in the same turn as discrete events; we chose to do so, but one could also have categorised them as hybrid or compound requests. (Thus we treated you need to speak, you can speak as expressing separately countable requests.) So the absolute numbers are approximations, and it is safer to take them just as indices of general proportions. The more interesting issue is what these different formats imply as to the relative statuses of requester and recipient. 3. Analysis 3.1. Analysis of the most-used formats The most common format used by staff (accounting for about two-thirds of the requests for action) was the imperative. After a consideration of their entitlement and contingency implications, we will turn to the less uncommon of the remaining formats, but the very rare formats we will leave out of consideration entirely (though we give examples in Appendix A). The imperative was the dominant form in each of the five scenes we sampled. It was most prominent in the kitchen, where the staff member was supervising the residents in preparing food (and where direction in handling equipment and operating appliances might be expected). But it was also the most-used format even when staff and residents were having evening drinks at a social club. Here are illustrative examples, one from each setting. 3.1.1. Bald imperatives (about two-thirds of the requests were in this format) Example 1. VD21 kitchen min 11.58 1 2 ((Dom is peeling potatoes with his back to camera)) Kev: [TD$INLE]3 thassit, (.3) °right° turn it round 'en, (1.0) turn it round, thassit, (3.0) gorrit? 4 5 (1.0) Kev: thassit. Example 2. VD19 supper min 38.103 1 2 3 4 [TD$INLE]5 6 7 8 Oona: ((to Oliver)) go on Oliver, serve yerself, [((Oliver gets up from his chair and moved round the table to get closer to the bowls))] [ (2.5) ] Oliver, what're y' having fruit or chocolate, (2.0) which one y' havin. Fruit. = w'll wa-wait my- ah wait till ( ) gets the spoon. 3 Note that at line 1, Oliver is not near enough the bowls to make a (visible) choice, so this is unlikely to be heard as an offer; Oonagh's imperative mobilises him to get up and approach them. Author's personal copy 880 C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 Example 3. VD10 laying table min. 6.30 1 2 [TD$INLE] Tim: Tim: 3 ( ) ((demonstrating pouring from carton into jug)) tip it (up), (.) and pu' some (orange) in ih. Example 4. VD16 Social club, min 1.52 1 2 ((Dave hands Dom an empty crisp packet) Dave: h'y're, Dom, (.) h'y'are. (.) Dom, [TD$INLE]3 4 5 [ (6.0) ] [((Dom blows into it)) ] Kath: do the business, 6 [ (3.0) ] 7 [TD$INLE]8 [((Dom continues to inflate the bag)) ] make sure it makes a good pop now please. 9 ((Dom eventually pops it)) Example 5. VD 17 planning meeting min. 14.00 1 2 [TD$INLE]3 Kath: ((describing photo)) that's me and you at missiz [Name]'s house. >now< - listen carefully now, where do you want 4 to go on holidays. A close look at one example of an imperative will be enough to show its presumptions. Consider Example 5, which happens at a meeting meant to solicit the residents’ preferences for their holidays. Kath orients to Dom's interest in a slide-show of images showing on a computer by offering an identification of its participants. To this she latches a new topic preface (>now<) and issues an imperative (listen carefully now) designed to bring Dom back to the business at hand. There is no negotiation as to her entitlement to issue such an order, or of what contingencies might prevent Dom from carrying it out (he might, for example, prefer to dwell longer on the images in the slideshow). Listen carefully now and its fellows made up about two thirds of the requests. By embodying very high entitlement and very little consideration of the recipient's contingencies, they are at the opposite pole from the I wonder if. . . formulation that Curl and Drew discovered among their corpus of calls to doctor's surgeries. Nor was there much use of the modal forms (can you. . ., would you. . . etc.) that Curl and Drew found among adults in everyday conversation. There were very few of such requests: there were only one or two examples of each of will you or could you, and even the most frequent, can you do X, made only about 15 appearances among the c. 234 cases. 3.1.2. Telling, not asking The imperative formulation that forms the bulk of the requests in our residents’ home are, as Craven and Potter put it, ‘‘built as a telling, rather than asking.’’ (p. 423). One specific prediction from Craven and Potter's work is that sequences of directives, if not complied with, would become progressively more like tellings and less like askings. Entitlement would grow, and appreciation of the recipient's contingencies would fall away. Here is an example of such a progression from their meal-time parent/child data: Example 6. (From Craven and Potter:427, requests put in bold) 1 Mum [kath’rine] >c’you move< [along] a little bit please.] 2 Mum [((starts to push chair next to Kath)) ] Anna [((moves out of the way of the chair)) ] Anna .hhu: [TD$INLE]3 4 Author's personal copy C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 5 Kath [((swings legs round to side))] 6 Kath [nng ] (.) I wanna sit 7 [<on> th- ] 8 → Mum [KATh’rine], [katherine don’t] be:- (.) do:n’ be= 9 Mum 881 [((shakes head))] 10→ Mum =horrible. [↑come on, mo:ve back ple:ase. ] 11 [((restarts pushing chair towards Kath))] 12 Kath aah 13 Mum [((pushes Kath and her chair backwards)) ] 15 Kath [Aaa:::how:::::::↑:::::: ((dur 3.1)) ] 16 Mum [((moves other chair into position))] 17 Mum [((picks Anna up and sits her on the chair))] 18 Kath [↑↑aaoo[ww::::: ((dur 2.8)) ] 14 [(2.0) ] 19→ Mum [y’need t’be ki::nd to yo:ur ] 20→ si:ste:r. (0.2) [now mo:ve your le:g] round the=] 21 Mum 22 Kath [((moves Kath’s leg round)) ] [↑A:::::h! ] 23→ Mum =front. 24 (0.4) Mum's request progresses from c’n you to don’t be to you need to and ultimately to the bald now move your leg round the front. In the institutional data, there are still more graphic escalations, as in this case: Example 7. VD21 kitchen 14.07. Resident Oliver is being supervised in operating the oven. 01 a Kev [TD$INLE]02 d'y wanna [turn'ee oven on, Oliver:¿ [((Oliver looks away)) 03 b Kev >Oliver< (.) can 04 [you turn the oven on for me, [Kev points to the oven] 05 [ 06 [((Oliver walks past the oven))] 07 Kev 3.0 ] Oliver. 08 [ 09 [((Oliver is out of shot, but apparently 1.0 ] 10 11 c Kev turns towards Kevin))] Turn the oven on, ((pointing at oven)) 12 [TD$INLE] [ 13 14 c Kev [((Oliver seems to have turned away again)) ] OLIver¿ (.5) Turn the oven on ((jabbing towards oven)) 1.0 ] 15 this one, 16 [ 17 [((Oliver is out of shot but presumably operates 18 19 d Kev a control on the oven)) ] no, turn it on. 20 21 2.5 ] (1.5) Kev thassit, right round, (1.0) >thassit<. The staff member's requests move from (a) d’y wanna through (b) can you to the bald (c) turn the oven on (twice) and ultimately the elliptical (d) turn it on. The initial display of concern for Oliver's contingencies (whether or not he wants to turn Author's personal copy 882 C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 the oven on, or indeed can do so) disappears after non-compliance.4 This is a still more uniformly progressive version of the sequence that Craven and Potter found around the family dinner table, exemplified in Example 6. We should contrast this move to greater assertiveness with cases reported by Aronsson and Cekaite (2011) in their study of parent–child negotiations about chores and activities around the house. There, on at least some occasion, parents would downgrade a directive in the face of non-compliance, and instead issue a concession or a qualified request (thus come down now! would be re-issued as yeah after the song you’ll come down – Aronsson and Cekaite, 2011:146). In our staff-resident data, there were no cases in which the requests went in this direction of travel, possibly because of the difference in assumed ability between children (who can be expected to understand, but won’t comply) and the residents (who might not understand in the first place) (and see also Zinken and Ogiermann, 2011). In sum, then, the heavy use of imperatives by staff makes these data looks much less like what Curl and Drew found among adults (institutionally or conversationally), or among Aronsson and Cekaite's (2011) families when there was space for negotiation. The pattern is much more like what Craven and Potter found among their sample of parents and children at mealtimes. The tone around their family tables, and in this residents’ home, is of orders rather than requests. 3.2. Other formats which the staff used The other formats which attracted anything approaching frequent usage were do you want to do X (used about once in every ten requests) and, to a lesser degree, you need to do X and can you do X (used about once in every twenty). Here are examples of all three: Do you want to do X Example 8. VD19 supper min 36.40 1 Oona Oliver, d'you want to take these 2 [TD$INLE]3 ((points at, then picks up, two jars)) out to the fridge, (2.0) (there's a-) and 4 grab a pudding, You need to do X Example 9. VD19 supper min 37.18 1 ((Kath pours out something from a tin into a 2 bowl; Alec stands by)) 3 Kath: d'y want a spoon, (.5) (>get a<) spoon: (.) to serve it, 4 [TD$INLE]5 (2.0) 6 ((Alec attends to something else on the table)) 7 Kath: 8 Alec: 9 Alec. [(.) you need to get a spoon, °don't yer°, [(whuh-) ((Alec moves off)) 4 To emphasise that the progression is a matter of dealing with the non-compliance of a request, and not simple repeating: it appears even such exchanges as the one below, where three minutes elapse between the staff member issuing the request (already in you need to X format) and dealing with non-compliance with the still more direct imperative give it a proper wash. VD21 kitchen min. 46.08. Resident Dom is being supervised by staff worker Kevin. 1 Kev when you done tha::t, (.3) listen, Dom (.) 2 DOM. (.) when y' done that y'need to 3 wash y'lunch box out (.3) 'kay? 4 (.5) 5 >g'man< 6 [ (3 minutes) ] 7 [((Kevin goes out of the kitchen; Dominic ] 8 [washes up at the sink, then takes lunch ] 9 [box off screen to another part of the ] 10 [kitchen. Kevin returns, goes off screen ] 11 [to where Dominic is. The lines below are ] 12 [off screen)) ] 13Kev >look isstill go' food in'th bottom<. 14 (3.0) 15Kev look, (.) isstill go' food (>all o'er it<) 16 y'need to wash it out properly Dom.=give 17 it a proper wash wi' th'scrubber, (.8) with 18 the scrubber, with the::- (.) er- (.) with 19 the washin'up °thingie° Author's personal copy C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 883 Can you do X Example 10. VD21 kitchen min 6.19 1 Kev c'mon then, get on w'y' potatoes, (1.0) Kev Kev this one? can you show me which one you use, 2 3 4 [TD$INLE] use that one. ((indicates chopping board)) 5 by the light ((moves to window)) 6 which chop-((coughs)) which one- 7 what colour While you need to do X is a format which takes no account of the recipient's contingencies, both do you want to do X and can you do X do mark some recognition of the recipient's situation. However, it is these formats which get trumped by more directive ones when they fail – as in the case of example 9 above, in which the staff member has recourse to you need to get a spoon, don’t yer when her d’you want a spoon results in no compliance. Among the rest of the formats, there were occasional usages such as the ‘‘could you do X’’ format (could you bring those- the- (er) the vegetables over, VD21). However, most implied high entitlement and low concern for contingency. Among these, the most directive was ‘‘you’ve got to’’ (as in you’ve got to get a move on, VD21). 3.3. Fault and correction One strong feature of all the requests in this corpus is that they tend to be issued when the resident had done, or was doing, something that the staff considered to need correction. This is not universal (see, for example, Example 4 where, on a social outing, the support worker jokingly ‘instructs’ a resident to pop a crisp-bag even though he seemed to be managing perfectly well on his own). But it was very common, as all the other examples attest. This is another significant difference not only with Curl and Drew's corpus of calls to the doctor, but also of their telephone corpus among friends and family, where the medium of communication doesn’t easily afford such joint orientation to an ongoing physical activity. There are two aspects to it: on the one hand, the possibility open to the requester to treat non-compliance as a temporary and corrigible matter, attributable to the recipient's effort and will (as opposed to something permanent and outside their control); and the fact that what is requested is an action in the here-and-now, the performance of which is easily and immediately visible. Both of these are true (or can plausibly be treated as being true by the requester) in the setting of the residential home. But in calls to the doctor, at least, the latter condition doesn’t quite obtain in the same way – what is requested is a decision, with the performance (the doctor actually arriving at the requester's home) some distance away. While it might be in principle possible for the call-taker's decision-making process to be drawn out and public enough for the caller to monitor it, and upgrade their request forms when it seems not to be working in their favour, there seems to be nothing like that in the Curl and Drew examples. And, presumably, the call-taker's reluctance to grant the request is not a matter of faulty effort or lack of will. Craven and Potter's site of the family mealtime, however, offers just those possibilities for the requester to treat noncompliance as something corrigible. It is notable that, although their account makes little of it, the mealtimes directives tend strongly to be issued as correctives to some piece of inappropriate or unsatisfactory behaviour on the child's part. Thus, to list some of the most self-evident example:: hold it with two hands (Extract 1, p. 423); don’t play; (Extract 2, p. 423); no no put that down (Extract 3, p. 424); sit and ask (Extract 4, p. 424); don’t be horrible, come on, move back please (Extract 6, p. 427). 3.4. Explanations: orienting to the staff member's entitlement and the recipient's contingencies Although the dominant request formats strongly implied high entitlement and low contingency, the staff did on occasion moderate this assertiveness by providing either the politeness token please (which we shall not analyse here), or explanations as to why the action needed to be performed. There were about a dozen such explanations among the c. 234 requests. Given that, according to Schegloff (2007:83) requests are regularly accompanied by explanations, just over in Author's personal copy 884 C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 twenty is a low frequency. When they did occur, however, the staff's explanations were given both to address entitlement and contingency. 3.4.1. Entitlement When the explanation addressed the staff member's entitlement to make the request, it was usually by dint of revealing something the staff member knew, and the resident seemed not to (or seemed not to have thought of at the moment). In other words, we can see it as the staff drawing on an epistemic authority to assert a deontic authority – the authority to get someone to do something (for recent work on deontic authority in conversation, see Stevanovic and Peräkylä, 2012). In the extract below Dominic seems to decline two invitations to make some white sauce, and Kevin issues an imperative for him to peel potatoes instead. Example 11. Kitchen min 15.40 1 Kev y'gonna help make some white sauce? 2 listen, Dom wanna (help) make some 3 white sauce, 4 [ 5 [((Dom seems to decline)) ] 6 → Kev well have a go with that one cos we've 7 → only got one of them peelers [TD$INLE] (2.0) ] Dominic may or may not know that the kitchen possesses only one peeler, but in any case it is Kevin who articulates that state of affairs, at lines 6 and 7 as entitlement to issue his instruction. Peeling potatoes is a very distinct cooking activity to making white sauce. Kevin is changing the action he is attempting to get Dominic to perform. However, both activities are part of the broader cooking project that Kevin is attempting to engage Dominic in. Therefore there is some potential ambiguity about whether this is the third attempt to get Dominic to contribute to the cooking process, or a first attempt at getting him to peel potatoes. As a first request (for a new action), the imperative form is particularly strong. Craven and Potter (2010) show it is often deferred until less entitled request formats have failed. The presence of the explanation may help to manage the potentially invasive and forceful nature of an imperative form being used for a first request. Kevin pairs his imperative with a report that they have only got one of them peelers through the use of cos. Cos positions the ‘only one peeler’ report as an account for issuing the ‘potato peeling’ directive. As an account, only having one peeler refers to a physical state of affairs, external to the participants’ motives or desires. Kevin reports a state of being outside his control at exactly the moment he is accounting for using a highly entitled first request. In their discussion of parental threats at mealtimes, Hepburn and Potter (2011) show how parents can vary the format of a threat to either ‘‘expose or embed’’ their control over the negative consequences for disobedience (p. 115). We suggest that the act of reporting a state of affairs offers an account for the directive that positions it's origin as external to the speaker and therefore less entitled than might otherwise be the case. It is not necessary for the reported state of affairs to relate to information previously unknown to the recipient. The explanation could even assert a state of affairs that the staff member treated as already known to the resident, as in example 12 below: Example 12. VD19 supper min. 1.20 1 2 ((Dom has brought in a jar of something)) Kath the chili belongs in the fridge Dom'nic [TD$INLE]3 you know it yerself (.) now that's wrong, 4 that's fer cooking. 5 ((Dom goes off toward kitchen with it)) Although there is no explicit request – the move works as what Schegloff calls a ‘‘noticing’’ (Schegloff, 1988) – Dominic understands the explanation, and the announcement that his choice was wrong, to be a call to correct a fault. Accordingly, he goes off to the kitchen, returning later with the correct jar. Author's personal copy C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 885 3.4.2. Contingency With regard to contingency, an explanation was sometimes a sign that the staff member had noticed something that was hindering the recipient's performance of the action, as in this case: Example 13. V21 Kitchen min. 37.20 ((Kevin is directing Henry in bringing something out of the oven, which he does tentatively)) 1 Kev [TD$INLE] awright, bring one out, thassit round 2 → the top=won't be hot if you've got 3 → the gloves on, you'll be all right 4 gorrit? Kevin's imperatives in lines 1 and 2 seem to meet some hesitation from Henry. The contingencies that face Henry (i.e., the danger of a hot oven) are public, and his lack of immediate compliance can’t be attributed to something internal such as lack of effort or attention. It may be that this is what occasions the use of an explanation or justification of the request (arrowed lines). What seems to be common to these explanations, whether addressing entitlement or contingency, is that they come after a failed request, or in parallel with the action being inadequately carried out (as in Example 13 above, where Henry is tentatively bringing something out of the oven). There were few examples in the data of an explanation that preceded, and set the ground for, a subsequent request.5 Again, this seems to show the staff's orientation to efficiency (although we might speculate that, in fact, some requests might have been more quickly complied with had their rationale been given first). 3.5. Deviant cases: staff orientation to low entitlement and high contingency The staff were perfectly capable of using low-entitlement and high-contingency requests. But they only did so systematically when prefiguring their own wants or actions, as the following examples illustrate. Example 14. VD21 kitchen min 20.06 [TD$INLE] 1 → Kev sorry Henry can I just pop in there a minute 2 ((gently moves H out of way to sink)) Example 15. VD21 kitchen min 32.20 Henry, can you just move over that bin 2 → so I c'n get rid of these potato [TD$INLE] 1 → Kev 3 peelin's please 4 ((H moves rubbish bin and K deposits peelings)) 5 Here is one example of an explanation not being subsequent to faulty action: in the extract below, Kath raises with Henry the topic of the evening's leisure activity. 1 Kath tonight's club, now. 2 Hen (urh, yuh) 3 Kath >(oh- if-) if< you're going then, 4 you need to get your money out, c'z you need 5 to pay fifty pee. 6 (1.0) 7 Hen (huh yerh, yuh) 8 9 10 11 Kath so when you've finished (g- put) (.) yer: (1.0) box an' your money, (2.0) get your fifty pee out of your wallet, yeh? (.) for: (.) club. (.) >°yeh?°< At lines 3–4, Kath issues an if-then formatted observation which explains the reason for her subsequent imperative in line 9: get your fifty pee out of your wallet, yeh?. Author's personal copy 886 C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 Example 16. VD19 supper min 4.30 1 2 ((Vic is pouring from a jug)) Kath: Victor is that salt in front of you, [TD$INLE]3 4 [(2.0) Vic: [((nods while pouring from jug, puts jug down)) 5→ Kath: may (I/ we) have it please, (.3) Kath at line 2 issues a grammatical interrogative, which pragmatically implies that she wants Vic to pass her the salt that is within his reach. She follows it up with an explicit request which, by being cast as asking for permission, displays an orientation to her ostensibly low entitlement. There were, to be sure, very few of such requests that marked low entitlement and an appreciation of the contingencies facing the recipient; but they were always used when the staff member wanted something for themselves. This demonstrates (should it have been in doubt) that the staff were quite aware of the dimensions along which requests for action operate; after all, they form part of our everyday conversational resources. But it brings into sharp relief the institutional motivations for the staff's use of imperatives when it came to requesting actions that were, ostensibly, for the benefit of the good running of the home. 4. Discussion The study reported here set out to ask how staff who support adults with intellectual impairments resolved the institutional dilemma of getting their clients to do things, while observing the requirement to respect their independence. Following inspection of about three and a half hours of staff-service user interactions in a residential setting, the answer is that the staff tend (overwhelmingly) to issue requests in formats that assume their complete entitlement to do so, and which make no provision for contingencies that might hinder the client. Bald imperatives (such as turn it around ˈen, cˈm ˈere a minute, go ˈn wash yer ˈands, have a go with that on, say it again slo::wly,) accounted for about two thirds of the approximately 234 requests identified. Among the other formats that attracted much usage, only two (do you want to do X and can you do X?) marked an awareness of the service-user's contingencies. Where staff offered explanations for their requests (as they did in only about a dozen cases in the corpus), these tended to be post hoc accounts of the staff member's epistemic entitlements to make the request, or designed to remove contingencies standing in the resident's way – a useful source of evidence of the participants’ own orientation to these categories. Two main conclusions can be drawn from the findings. The first is that the framework of entitlement and contingency established by Curl and Drew (2008) continues to prove its usefulness as a fix on the interactional dynamics of requests, and seems to be a members’ category as much as an analysts’ one. However, we can see that the pattern among adults that Curl and Drew uncovered is vastly different when it comes to interactions between a sample of support workers and people with intellectual impairments. The second conclusion takes us into the realm of institutional talk, and what this pattern of high entitlement and low contingency tells us about the staff's relationship with their clients. Here the comparison with parents’ directives to children is revealing. Craven and Potter's (2010) account of a sample of British family mealtimes shows that parents are prone not only to issue high-entitlement and low-contingency directives, but also to ramp them up to bald imperatives when the children do not comply. If anything, the patterns among the staff members in our data is still starker, with bald imperatives being the main weaponry, not just the heavy guns brought in when lesser arms fail. Our pattern is even less like Aronsson and Cekaite's (2011) finding that parents do sometimes downgrade their requests by conceding some acknowledgement of their children's contingencies; in our sample this did not happen once in over two hundred times. The closest we came to seeing such acknowledgement was in staff's occasional explanations of the reasons for their claimed entitlement, or of the contingencies affecting their client. But these were few and far between. We should make it clear that although the format of the staff's requests for action are heavily skewed towards the imperative, it is not the case that the tone of their interactions with the residents was oppressive or unfriendly, let alone bullying. Admittedly, such impressions are intangible, but the videos show (and the ethnographic notes confirm) that the general atmosphere of the house was benign. Nevertheless the institutional force of ‘getting things done’ was clearly at work when there were specific tasks to be carried out. This forms another, and very generally applicable, contrast with the kinds of requests surveyed in Curl and Drew's original work. In asking for the doctor to come out, the crux of what is requested is the decision to come out, not the physical actions that the doctor would use to realise it. We can see three distinct dimensions along which the kind of requests that feature in Curl and Drew's work differs from the ones we report here. One is the relation between the requester and recipient; in calls to the doctor, and in calls between family Author's personal copy C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 887 members or friends, no party is acting as morally, culturally or contractually responsible for the developing behavioural and social competency of the other party. Of course, they could do so, but our prediction is that interactional trouble would ensue. (There may be rare examples of such misalignment among adults, and we will keep an eye out for it in further trawls of such corpora as the Holt data of telephone calls between friends and family). So one factor that moves the request along the gradient towards directives could be the degree to which the requester is playing out their obligation to instruct or socialise the other (as would indeed be the case among parents and their children, or between teachers and pupils). A second factor is the degree to which the request follows a fault. Again, it is within any speaker's power to treat noncompliance with their request as being a corrigible fault, attributable to some failure on the part of the recipient. No doubt this does happen (with potential subsequent turbulence) in some service encounters (one example that comes to mind is the case of a very badly misaligned request in a call to the emergency services, reported and analysed in Zimmerman, 1998:88–89). It doesn’t seem to happen in Curl and Drew's calls to the doctor, nor in calls to family and friends; but it is routine in our data, and that speaks to its specific institutional character – or rather, how the participants embody and realise the institution of ‘care’ and ‘support’, at least on the occasions surveyed. The last factor we might note is the degree of immediacy and physical embodiment of the action requested. In our setting, the resident is being asked to do some physical action, and it is exactly his realisation of it that is at issue (turn it on; bring one out, thassit, round the top; make sure it makes a good pop now please; tip it up and put some orange in it; and so on). Looking back at the Conversation-Analytic literature on requests and directives, there does seem to be a sharp distinction between requests for abstract or future events (permission to do something, a commitment to a future action, and so on) on the one hand, and very concrete actions in the here and now (turning on an oven, passing a plate, and so on). It may be that the degree to which the request is seeking an immediate physical change in the world is a factor which helps determine the format that the requester uses; our intuition (prompted by what we see in our data, and that of the family mealtimes in Craven and Potter) is that more concrete and immediate actions afford speakers more leeway for claiming greater entitlement and showing less awareness of contingency. But this is speculation and, like the possible role of socialisation, and the way that request formats instantiate institutional mores, would need to be confirmed by further research. Appendix A. Little-used request formats (once or twice each in c. 234 cases) not otherwise exemplified in the text You've got to do X VD21 kitchen min.13.28 Kev: 'ave y'sel-, (.3) 'ave a tart then, then you've got to peel the potatoes, you've got to get a move on cos y- y- you've got (your / an) awards ceremony tonight, What you're going to do is X VD21 kitchen min.14.40 [TD$INLE] Kev: d'y wanna get a tray out, thassit. bring it over here then. >what y' g' do's< put it in the oven. wanna put it in the oven? You want / don't want X orto do X VD10 laying table min.7.56 Tim you need to fill it up a bit more, ((Tim then inspects the left-hand jug, Tim then [the right hand jug)) [yeh- wan' a bit more, ((points towards something not visible)) Author's personal copy 888 C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 You can do X VD10 laying table min. 8.25 ((Dom holds out a large bottle of juice to Tim)) Tim: no you can open it Will you do X VD 17 planning meeting min. 23.40 Kath: will you look at me a moment please If you do X then Y will happen VD19 supper min. 40.50 ((Oliver is trying to pour something out of a carton)) yes yes >Ol'ver< if you keep tipping it over Kath it will come out (.) believe me, that's it, good lad, all round, tha:t's it, I suggest you do X VD19 supper min. 6.20 ((Dominic is standing, ladle in one hand and plate in the other, apparently intending to serve himself from a dish in the middle of the table; a large jug is in the way)) Kath: Dominic, (.7) I suggest you put the plate down an' mo:ve the jug off it. [TD$INLE] ((Dominic makes no move, but Chris (researcher) has already started to reach for the jug; completes its removal)) (Are) you going to do X VD21 kitchen min. 37.33 Kev: right, (1.0) I'll turn it over, (.3) you gonna pop it back in f'me, (.5) Hen: ( ) Could you do X Vd21 kitchen min. 20.18 Kev: Oliver, (.3) pop those in the >bin (mate)<, (1.5) ((cough)) ((points)) c'ld y'bring those- the- >er< the vegetables over, Shall we do X VD21 kitchen min. 15.15 Kev: sh' we pop this ( ) to open the oven up Hints VD19 supper min 0.50 Oona ((to Alec)) the erm- (.) dish is very hot Author's personal copy C. Antaki, A. Kent / Journal of Pragmatics 44 (2012) 876–889 889 References Aronsson, K., Cekaite, A., 2011. Activity contracts and directives in everyday family politics. Discourse & Society 22, 137–154. Brown, P., Levinson, S.C., 1987. Politeness: Some Universals in Language Usage. Cambridge University Press, Cambridge. Clement, T., Bigby, C., 2010. Group Homes for People with Intellectual Disabilities: Encouraging Inclusion and Participation. Jessica Kingsley Publishers, London. Craven, A., Potter, J., 2010. Directives: entitlement and contingency in action. Discourse Studies 12, 419–442. Curl, T.S., Drew, P., 2008. Contingency and action: a comparison of two forms of requesting. Research on Language and Social Interaction 41, 129–153. Harris, J., Danby, S.J., Butler, C.W., Emmison, M., 2011. Extending client-centered support: counselors’ proposals to shift from email to telephone counseling. Text and Talk 32 (1), 21–37. Heinemann, T., 2006. ‘‘Will you or can’t you?’’ Displaying entitlement in interrogative requests. Journal of Pragmatics 38, 1081–1104. Hepburn, A., Potter, J., 2011. Threats: power, family mealtimes and social influence. British Journal of Social Psychology 50, 99–120. Keisanen, T., Rauniomaa, M. Organization of participation and contingency in pre-beginnings of request sequences. Research on Language and Social Interaction 45(4), forthcoming. Kuroshima, S., 2009. Another look at the service encounter: progressivity, intersubjectivity, and trust in a Japanese sushi restaurant. Journal of Pragmatics 42, 856–869. Lee, S.-H., 2011. Managing non-granting of customers’ requests in commercial service encounters. Research on Language and Social Interaction 44, 109–134. Schegloff, E.A., 1988. Goffman and the analysis of conversation. In: Drew, P., Wootton, T. (Eds.), Erving Goffman: Exploring the Interaction Order. Polity Press, Cambridge, pp. 9–135. Schegloff, E.A., 2007. Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge University Press, Cambridge, England. Shakespeare, P., 1998. Aspects of Confused Speech: A Study of Verbal Interactions Between Confused and Normal Speakers. Lawrence Erlbaum Associates, New Jersey. Stevanovic, M., Peräkylä, A., 2012. Deontic authority in interaction. Research on Language and Social Interaction 45 (3). Williams, V., 2011. Disability and Discourse. Wiley-Blackwell, Chichester. Zimmerman, D., 1998. Identity, context and interaction. In: Antaki, C., Widdicombe, S. (Eds.), Identities in Talk. Sage, London, pp. 87–106. Zinken, J., Ogiermann, E., 2011. How to propose an action as objectively necessary: the case of Polish Trzeba x (‘‘One Needs to x’’). Research on Language and Social Interaction 44, 263–287. Charles Antaki is Professor of Language and Social Psychology at Loughborough University's School of Social, Political and Geographical Sciences, where he is a member of the Discourse and Rhetoric Group. His research interests are in Conversation Analysis. Among his recent publications is the edited collection Applied Conversation Analysis (Palgrave-Macmillan). Alexandra Kent (née Craven) is a Lecturer in Social Psychology at Loughborough University's School of Social, Political and Geographical Sciences. She is a member of the Discourse and Rhetoric Group. Her research interests include family mealtime interaction, the negotiation of power and authority in interaction, children's talk and socialisation, and parenting practices.