Whitehead and Lerner: "When Are Persons 'White'?

When are persons ‘white’?:
on some practical asymmetries of racial reference in talk-in-interaction1

Discourse & Society
Copyright © 2009
SAGE Publications (Los Angeles, London, New Delhi, Singapore and Washington DC) www.sagepublications.com Vol 20(5): 613–641 10.1177/0957926509106413

A B S T R A C T This report contributes to the study of racial discourse by examining some of the practical asymmetries that obtain between different categories of racial membership as they are actually employed in talk-in- interaction. In particular, we identify three interactional environments in which the ordinarily ‘invisible’ racial category ‘white’ is employed overtly, and we describe the mechanisms through which this can occur. These mechanisms include: (1) ‘white’ surfacing ‘just in time’ as an account for action; (2) the occurrence of referential ambiguities with respect to race occasioning repairs that result in overt references to ‘white’; and (3) the operation of a recipient design consideration that we term ‘descriptive adequacy’. These findings demonstrate some ways in which the mundane invisibility of whiteness – or indeed, other locally invisible racial categories – can be both exposed and disturbed as a result of ordinary interactional processes, revealing the importance of the generic machinery of talk-in- interaction for understanding both the reproduction of and resistance to the racial dynamics of everyday life.
K E Y W O R D S : conversation analysis, membership categorization devices, race, racial categories, whiteness


Much of the early work on racial discourse took the form of largely quantitative content analytical studies of the portrayals of particular racial or ethnic groups in news media, television shows, and other forms of public discourse. This research demonstrated some of the ways in which people of color tend to be portrayed in negative or stereotypical ways, while white people are generally portrayed more sympathetically (Van Dijk et al., 1997). In comparison to these mainly quantitative studies of topic content, as Van Dijk et al. (1997: 166) point out, ‘discourse analytical studies of more detailed properties of text and talk about
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
614 Discourse & Society 20(5)
ethnic events and ethnic relations were rare until the 1980s’. One noteworthy exception to this trend can be found in the early work of Harvey Sacks (begun in the mid-1960s) on the everyday use of membership categories and membership categorization devices, or MCDs (Sacks, 1972a, 1972b, 1986, 1995), in which the explication of racial membership categories played an important part.

More recently, a substantial and growing body of literature has provided detailed qualitative examination of many aspects of racial discourse (see, for example, Bonilla-Silva, 2002; Bucholtz and Trechter, 2001; Buttny, 1997; Condor, 2006; Condor et al., 2006; Stokoe and Edwards, 2007; Van den Berg et al., 2004; Van Dijk, 1987, 1992, 1993; Verkuyten et al., 1994; Wetherell and Potter, 1992). Much of this work has focused on the ideological character of ‘race talk’, examining expressions of racial prejudice, and the ways in which such expressions are connected to broader patterns of racism and inequality. While racist discourse is certainly a crucial object of study, such discourse depends upon the availability of the racial categorization of persons as a resource; this is how racism is implemented as a basic form of social organization at the point of its production in talk-in-interaction. Moreover, this organization underpins not just racist discourse, but also any other form of discourse in which race is used, including anti-racist discourse. It is thus important to examine the mechan- isms through which racial structures are reproduced in individual episodes of interaction, regardless of whether or not the discourse being produced therein is judged as ‘racist’. In light of this, it is significant to note that few studies beyond Sacks’ early work have examined the ways in which racial categories are themselves employed in the course of talk-in-interaction, and hence the ways in which race as a social institution is reproduced in action (but see West and Fenstermaker, 2002; Whitehead, in press). Moreover, few studies have examined the organization of racial categories in interactions in which race is invoked seemingly ‘incidentally’ in the course of whatever actions speakers are performing, rather than being elicited for research purposes, as in the case of research interviews or focus groups in which researchers prompt partici- pants to discuss matters of race (cf. Wilkinson and Kitzinger, 2003).2

In this report, we contribute to the investigation of racial discourse in which racial membership categories are mentioned overtly. We demonstrate some of the ways in which these racial categories are made relevant and employed by speakers in talk-in-interaction, and show how this in situ organization of membership categorization enables and thus reproduces the social organ- ization of race. The overt use of racial categories is one way participants make race manifest in their daily lives and it is the ‘inference-rich’ character of these categories that underwrites how members make racial sense of the social life of their society (Sacks, 1995: 40–1). We begin by reviewing and amplifying Sacks’ work on the use of racial categories as accounts for action, before focusing more specifically on how Sacks’ findings can be brought to bear on the investigation of the practical asymmetry between different racial categories in talk-an-interaction, with a taken-for-granted category (e.g. ‘white’) routinely remaining unexpressed (even when relevant), while other race categories are overtly mentioned.
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 615

Using race to account for action

Sacks’ work on membership categorization devices (Sacks, 1972b, 1995) demonstrates the way in which social categories provide a means for storing and organizing common-sense cultural knowledge. As a consequence, the sheer mentioning of a category can marshal common-sense knowledge about that category, and can thereby stand as an adequate account for social action (Kitzinger, 2005a, 2005b; Sacks, 1972b; Whitehead, in press). Thus, in the following excerpt, a request of a (male) stranger needs no overt account.3 This request occurs on an airplane after the passengers have boarded and taken their seats.
Excerpt 1 [GL:FN:UAL 3/27]
  1. 1  Passenger A: I wonder if you would mind trading seats with my wife?
  2. 2  Passenger B: Sure
  3. 3  PassengerA: Thankyou.
Here, the categorical reference to ‘my wife’ supplies a tacit account for the request by Passenger A – thus making the passenger’s reason for producing it a matter of taken-for-granted category-bound cultural knowledge, rather than requiring a local, situation-specific or person-specific account.4 In a similar way, Sacks’ (1984, 1986) pioneering work on racial membership categories demonstrates how racial person references can be used to make sense of actions. Thus, Sacks shows how simply employing a racial category in referring to a person can serve to tacitly account for how that person could be seen to be (legitimately or illegitimately) performing a particular action.
In addition to its use in person references, this practice can be used in the production of racial place references. In this way, the formulation of a place reference (cf. Schegloff, 1972) as a racialized location can be employed as a tacit account for action, as shown in Excerpt 2. In this case, Ron is talking about a town that his family used to frequent while on vacation. At line 1, he begins to describe their hangout (‘a hotel’), but cuts it off to give some background in- formation. Ron then explains that there were two liquor licenses in the town: one in a hotel, and the other ‘in the black section’ (line 2).

Excerpt 2 [Bonelli]
1 Ron: 2
4 Ron: 5

6 Ron: 7
So there’s a hotel there’s two liquor licenses in the town one is at a hotel and one is (down) like in the black section (0.4)

they got tuh hang out in the hotel and they got tuh know the owners an:: they know ev’rybody .hhhh ev- they spent a ton a’ money down there.
Ron’s repair at lines 1 and 2 adds an element to the telling that is presented as factual background. His characterization of one of the liquor licenses as being ‘in the black section’ serves as an account for why the people in his story spent a lot of time in the hotel (where the other liquor license was located) with the
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
616 Discourse & Society 20(5)
‘so’ at line 4 serving to make overt that what follows is the consequence of the circumstance he has just described.
Note that the racial membership category ‘black’ is employed overtly in the formulation of the ‘black section’, as a contrast to the non-racial place formulation ‘hotel’. Although the hotel is not identified racially, by contrasting it with a race-specific place, Ron tacitly establishes race as a relevant defining feature of the hotel,5 producing it as ‘not black’. Thus, by formulating one location in an asymmetrical contrastive pair in terms of a racial membership category, Ron tacitly racializes the other location in the pair, and this racial cat- egorization is designed to account (without a second thought) for why his family spent their time and money in one place rather than the other. The racial place reference is sufficient to disqualify that location, thus establishing the hotel as the only reasonable place to patronize, and thereby furnishing a tacit account for why it became their hangout.6 It was, in effect, the only place in town, but its whiteness remains invisible – for the participants – as a result of being made available only tacitly, rather than being employed overtly (cf. Wilkinson and Kitzinger, 2003). This asymmetry between ‘white’ and other racial categories, such that whiteness is produced as ‘invisible’, taken-for-granted, or neutral, and forms the normative backdrop against which other racial categories are viewed, is a central phenomenon in the scholarly literature on race, and one that we take up in the remainder of this article. We turn first to a brief review of the literature on this topic.

Categorical asymmetry and the invisibility of whiteness

The status of whiteness as invisible or taken-for-granted has been recognized by social scientists going at least as far back as W.E.B. Du Bois’ classic writings on the concept of ‘double consciousness’ (Du Bois, 2003 [1903]). Du Bois describes the way in which black people (and, arguably, people of color more generally), living in a white-dominated society such as the United States, ex- perience a sense of viewing themselves through the eyes of others; of having ‘double selves’ as a result of having their own sense of self, while also being aware of how they are viewed by members of the dominant group (Du Bois, 2003 [1903]). A consequence of the double consciousness experienced by people of color, but not by white people, is that white people do not have to recognize the role that their racial category membership, and the privileges associated with it, plays in their lives. In this way, white people can avoid viewing race as a significant aspect of their identity, or as a factor that shapes their everyday experiences, making race, for them, effectively invisible. People of color, on the other hand, are constantly reminded of their ‘otherness’, and of the subordinate position in which their racial category membership places them, making race a constantly visible feature of their lived experiences (Du Bois, 2003 [1903]).
A number of race scholars have more recently elaborated on the conse- quences of the asymmetries between whiteness and other racial categories first illuminated in Du Bois’ early writings. These scholars have critically investigated
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 617
the ways in which a ‘color-blind’ ideology, enabled and underpinned by the invisibility of whiteness, serves to support a system of white privilege, and the accompanying disadvantages for people of color. The color-blind ideology described in this literature is characterized by members of the dominant (white) racial group viewing themselves in non-racial terms, as ‘just people’, rather than identifying as members of a racial category (McKinney, 2003). Consequently, the framework of norms and values associated with the dominant group comes to be unquestioningly, and hence invisibly, treated as equally applicable to members of other groups (McIntosh, 1988). Thus, although discourses of color- blindness may arise from well-intentioned attempts to ‘move beyond race’, such positions begin from a predominantly white experience of the world, where race is perceived as unimportant, thereby negating the life world of people of color, whose experiences are still very much shaped by race (Simpson, 2008).
The lack of recognition in color-blind ideology of the ways in which race shapes people’s experiences and life chances can result in racial inequalities being blamed on the deficiencies of individuals, while the role played by systems of racial privilege is ignored, and hence is reproduced unchallenged (Frankenburg, 1993; Lipsitz, 1995, 1998). Moreover, such a perspective results in anyone who is not white being defined as ‘other’ against the backdrop of normative whiteness, resulting in people of color being treated as deviant in a range of ways (Collins, 2004; Lipsitz, 1995, 1998; McIntosh, 1988). In this way, ‘[a]s the unmarked category against which difference is constructed, whiteness never has to speak its name, never has to acknowledge its role as an organizing principle in social and cultural relations’ (Lipsitz, 1995: 369; emphasis added).7
The literature on whiteness and color-blind ideology thus illustrates the importance of the invisibility of whiteness as a factor underpinning systems of racial privilege and disadvantage, and points to the importance of investi- gating the mechanisms through which invisible whiteness is actually produced and reproduced, as well as ways in which it can be resisted or subverted (cf. Frankenburg, 1993). One way of developing an empirically grounded account of such mechanisms is to examine the practical asymmetries of racial categories at the point of their actual social production in talk-in-interaction. Although no research that we are aware of has employed such an approach to studying whiteness, the work of Kitzinger and her colleagues demonstrates its utility for exploring a different kind of invisibility, namely the production and reproduction of taken-for-granted, invisible heterosexuality (e.g. Kitzinger, 2005a, 2005b; Wilkinson and Kitzinger, 2003), and the ways in which such invisibility can be exposed or subverted (e.g. Land and Kitzinger, 2005; Wilkinson and Kitzinger, 2003), in talk-in-interaction.8
Against the backdrop of the ordinary ‘seen but unnoticed’ character of whiteness among ‘white’ participants, it seems worthwhile to investigate those circumstances in which the asymmetrical invisibility of ‘white’ as a membership category is disturbed, and the whiteness of persons surfaces in talk-in-interaction. In what follows, we pursue this matter, turning our attention to the question posed in our title (‘When are persons “white”?’) by considering how and under what circumstances the membership category ‘white’ is employed overtly in
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
618 Discourse & Society 20(5)
talk-in-interaction. As we will show, in some circumstances, the taken-for- granted character of whiteness is directly disturbed when it intersects with aspects of the organization of talk-in-interaction. However, we will also show that in other circumstances, ‘white’ is not the asymmetrically taken-for-granted racial category. In these cases, the overt use of the category ‘white’ does not disturb the ordinary asymmetry of racial categories, but instead reveals a circumstantial inversion of the asymmetrical alternatives, such that a category other than ‘white’ is treated as asymmetrically taken-for-granted.
The data excerpts we examine below were collected from a number of dif- ferent audio- and video-taped data corpora recorded over the course of several decades, ranging from the 1960s to the early 2000s. While it will be apparent to readers that some of our data excerpts exemplify features of racial references that are no longer as commonly employed as they once were (e.g. the use of the membership category term ‘colored’), we focus our analysis on aspects of racial reference that do not appear to hinge on the specific category term employed.
Drawing on Sacks’ work on MCDs, and the use of racial categories to account for action (Sacks, 1984, 1986, 1995), we begin by examining some ways in which the category ‘white’ is made explicit ‘just in time’, in order to account for victimization or prejudice by reference to other racial categories. We then examine a particular sequential environment in which ‘white’ can be made visible, namely when it is used to repair referential ambiguities resulting from race having previously been made relevant. Finally, we suggest another basis for the deployment of racial categories in talk-in-interaction, which we term descriptive adequacy. That is, when formulating descriptions of persons in which observable features of the referent are prominent, speakers employ a racial category in cases where a recipient would otherwise envisage a person of the taken-for-granted racial membership category. The use of a racial category for descriptive adequacy thus constitutes an interactional mechanism through which ‘white’ can be made explicit in circumstances in which a category other than ‘white’ is treated as the invisible category – that is, when the categorical asymmetry is inverted.

When ‘white’ surfaces ‘just in time’ to provide an account

In this section, we examine cases in which the overt mentioning of a person as ‘white’ is employed as an adequate account for actions and attitudes associated with that person. As Sacks (1972b) has demonstrated, there is always more than one correct membership category available to a speaker when referring to a person categorically (e.g. sex and stage-of-life MCDs). From this demonstration, Sacks concludes that the use of any particular category can be understood as a selection from among equally correct possibilities, and thus can be inspected for why one was chosen over another, and what the selection of this particular category accomplishes (cf. Schegloff, 1997). In this way, members’ culturally based categorical knowledge can be marshaled to fit a referred-to person to an
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 619
action, so as to account for how the referred-to person could be seen to properly and naturally act in a particular way.9 Here, we examine how and when this can operate for the otherwise asymmetrically invisible category ‘white’.
The first two cases we examine come from a setting – a ‘race training’ workshop – in which race has an occasion-based omni-relevance as a topic. It is a setting in which speakers talk about race, and display knowledge and sens- itivity about its importance, and participants seem to be held accountable for doing so (cf. Whitehead, in press). In these cases, we can observe the intersection of the use of racial categories to serve as accounts for action with the ordinarily taken-for-granted status of whiteness as irrelevant to action. In both cases, the relevance of the membership category ‘white’ is initially left tacit, and ‘white’ is only introduced overtly ‘just in time’ to serve as an explanation.
In Excerpt 3, Megan initially refers to her father non-racially, as ‘my father’ (line 2), thereby treating him as invisibly white, and only subsequently does she expand this relationship categorical reference to ‘my white father’ (at lines 9–10). Here, Megan employs ‘white’ overtly in the service of explaining her own racial prejudices, while mitigating her personal responsibility for those prejudices.
Excerpt 3 [TDC WG, 22]
1 MEG: 2

5 6 7 8 9
10 11 12 13 14
I would say that I- I actually had a similar experience
in terms of growing up that my-
my father had (0.6)
a: a difficult situation or experience with (.)
↑in this
case particularly an
African American person. .hh And I will say that one of the things that- that it impacted
me on is that I too I think was raised with sort of
this (.) in one breath, (0.3) everyone’s treated equal,
.hh yet (.) at the s- by the same token, (0.2) the
examples that I was ever given about (0.2) when
my white father was impacted by someone it was by a person of color and I think that that very much played .hh for me growing up about what my (.) ideas are a- ideas are and have been about .hh u:m (0.4) people of color: (.) sort
of victimizing my family.
Megan’s use of a racial category in referring to her father serves to under- score the causal link she is proposing between her family history and the racial prejudice she is now admitting to, and accounting for. By making her father’s racial membership category (‘white’) explicit in contrast to the racial member- ship category (‘person of color’, lines 10–11) of those who she describes as having mistreated him, she seems to indicate that as a child she understood that he was victimized as a white person. His race can be seen to explain her per- ceptions as a child of his treatment at the hands of others – it was not just that people of color ‘impacted’ her father, but that they did so as people of color through their treatment of her father as a white person. In this way, Megan’s overt use of the category ‘white’ aids in explaining the development of her prejudice as a white person. She proposes that it was her recognition (as a child) of the contrast between the racial categories of her father and the people with whom
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
620 Discourse & Society 20(5)
he had ‘difficulties’ that resulted in her developing the racial prejudices that she is now admitting to.
In the next case, Sammy tells a story about the experience of a friend to sup- port his claim that being white does not always guarantee ‘racial privileges’. He alludes to race, but does not overtly mention it either in referring to his friend, the neighborhood in which the events took place, or the residents of the neigh- borhood, through most of the setup of the story.10 As he continues the story, he never makes explicit the racial character of the neighborhood, but does overtly identify his friend as ‘a white guy’ just after describing the neighborhood as ‘tough’ (lines 10–11).
Excerpt 4 [TDC WG, 21]
1 SAM: 2

5 6
  1. 7  TR2:
  2. 8  SAM:
  3. 9  TR2:
  4. 10  SAM:
11 12 13
here, I’d been out here for about a year, (0.5) u:m
(0.3) and a guy I went to high school with came out here from Wisconsin an’ .hh a:n’ u:m (0.4) pt .hh he was a:
(.) motorcycle mechanic an’ got a jo:b down in oh jeez where was it? (0.6) U::m way down off of Imperial.
Mm hm.
.hh u:m (.) y- you know where that is [an: an that is a

[Mm hm.
tough neighborhood. .hh u::m, a::nd (.) I mean
he was a
white guy, (0.3) and was told by: people that he worked with, .hh to not get caught here after dark. ((continues))
By identifying his friend’s race only just before launching into the events of the story, rather than when he first introduced the friend, Sammy underscores the relevance of his friend’s race for understanding the racialized character of the ‘tough neighborhood’ and for explaining the subsequent treatment he received there. By setting up the scene in this way, Sammy is able to show that his friend’s race was consequential for what happened to him in this neighbor- hood, while never making explicit its racial composition. Thus, we see ‘white’ surface in this case as an explanation (for the attack) while leaving inexplicit the racial membership of the attackers.11 As in Excerpt 3, the referent’s race (white) is only made explicit at the point at which it is employed contrastively to account for the actions of other (racialized) persons – i.e. to formulate the basis of the victimization as relevantly racial victimization of a white person.
In Excerpt 5, ‘white’ surfaces in a similar way, albeit in this case to explain the mistreatment of black people at the hands of others. However, unlike the two preceding cases, in Excerpt 5, the routine practice of leaving ‘white’ unexpressed – even when the race MCD has been made relevant – becomes a source of trouble that results in repair. In this excerpt, Ken is reporting what he heard someone say about the way black people are mistreated in the South. Here the speaker can be seen to be explicitly oriented to the explanatory power of ‘white’ as an overt form of racial person reference: the original non-racial reference form at line 24
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 621
(‘these guys’) is replaced by a racial reference form (‘white’) at just the point at which the speaker is about to formulate an action as racist.
Excerpt 5 [GTS I, 88–9]
1 Ken: 2

  1. 4  ():
  2. 5  Ken:
8 Louise: 9

  1. 10  Ken:
  2. 11  ():
  3. 12  Ken:
  4. 13  Louise:
  5. 14  Ken:
  1. 16  Ken:
  2. 17  Roger:
  1. 19  ():
  2. 20  ( ):
  3. 21  Ken:
23 Ken: 24

  1. 27  Louise:
  2. 28  Ken:
In Mammoth I heard a- all these men fr’m the Forest Rangers departmin talkin about, (0.9)

[In uh, certain places in-in the United States in uh, where izzit. Where they pick cotton all the time, uh, (1.0)

Yeah. In the- [the

[hhhh [deep south,
where all the
He said thet-
Wake him up.
He’s not asleep.
thet all the
colored people,
uh, walk- walk down the street ’n
they may be all dressed up er sump’n en these guys ’n white- uh white guys’ll come by with .hh
Eh-uh mu:d, ink er ennything en throw it at ’em er throw bricks at ’em er ennything.
This repair reveals Ken’s treatment of the non-racial reference form (‘these guys’) as insufficient to serve as an account for the action about to be described. This leads Ken to overtly specify the race of the referents so as to make explicit that the conduct he is about to formulate is not just the action of ‘these guys’ but of ‘white guys’.12 Thus, as in Excerpts 3 and 4, this excerpt provides a case in which ‘white’ is made an overt part of a person reference so as to call attention to race ‘just in time’ to provide an account, and further shows that a speaker may disrupt the progressive realization of the story in order to do so.
In this section, we have shown that ‘white’ can remain an unexpressed, but relevant, racial membership category in juxtaposition to other racial categories – until it is employed as an account for action.13 Next, we turn to an examination of cases in which the invisibility of whiteness (as a tacit racial reference) results in referential ambiguity that is then repaired by employing a racial category overtly.
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
622 Discourse & Society 20(5)
Taken-for-granted racial categories as a source of referential ambiguity
In this section, we show how the invisibility of whiteness, when juxtaposed to other overtly mentioned racial categories, can itself be a source of referential ambiguity that requires repair – a repair whose production results in racial cat- egories (including ‘white’) being overtly mentioned. Once racial membership has become relevant through a locally initial overt racial reference, locally sub- sequent non-racial references can be treated as vulnerable to race-based sources of trouble that engender repair.14
We begin by describing a case, shown in Excerpt 6, in which a locally sub- sequent non-racial reference – produced after the race MCD has been made relevant, and possibly hearable as a tacit reference to white people – is replaced with an overt (racist) reference to black people. Here, Jim is bragging about what an acquaintance did during a high school track meet. In doing so, he refers racially to the students from one of the visiting schools (line 5) and describes his acquaintance’s conduct in racial terms (line 8), thereby making race relevant.
Excerpt 6 [GTS IV, 23–4 (simplified)]
1 Jim: 2
  1. 3  Roger:
  2. 4  Ken:
  3. 5  Jim:
  4. 6  Ken:
  5. 7  Jim:
8 9
10 11
Like yesterday there was a track meet at Pallisades. Rees was there. Isn’t that a reform school? Rees? Yeah

[Buncha niggers an’ everythin’
He went right down on that field an’ he was just sittin there talking
like a nigger,
n’ all the guys, an’ y’know all these niggers are all up [there an’-
[You mean Negro, doncha.
As a non-racial reference form employed after race has been made locally relevant, ‘all the guys’ (line 9) is potentially ambiguous: it could be understood either as a tacit reference to a previously unmentioned group of (invisibly) white people, or as another reference to the previously referred-to black people. Jim then treats it as ambiguous by disambiguating the race of the referents, repairing to reformulate them explicitly as ‘all these niggers’ (line 9). It is thus the asymmetrical structure of racial reference – with ‘white’ ordinarily left tacit – that appears to be the source of the ambiguity here, resulting in another racial category being overtly formulated. We now turn to two cases in which this asymmetry results in repairs through which ‘white’ itself surfaces as an overt form of person reference (and, unlike in the previous section, these repairs occur in environments in which ‘white’ does not appear to serve as an account for action).
In Excerpt 7, drawn from the ‘white group’ component of the race training workshop, a trainer is responding to a discussion about racially motivated violence against both people of color and white people, making the ‘who’s in
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
  1. 1  ( ):
  2. 2  TR1:
  1. 4  ():
  2. 5  TR1:
  3. 6  ():
  4. 7  SAM:
  5. 8  TR1:
[w- one of the costs you’re talking about is that

there’s this big confusion about (.) ↑who’s hurting who, [hh [Mm
how did it start [how does it e:nd. [(Mm hm.)
Mm hm.
Um,andsolongasanyof usinthisroomseeacostin
>terms of racism< .hh we have a reason to fight racism. Daily. We have a reason to get uncomfortable to go to workshops like th(h)is. tuh .hh talk with
people of color, to talk
each other. to sart- s- white support groups that- antiracist groups, .hh to keep working on this because there is a cost, whether or not we can (.) pinpoint the start, beginning, end, middle, there’s a- there’s a reason.
Whitehead and Lerner: When are persons ‘white’? 623
her reflection on the ‘confusion about who’s hurting who’ (line 3) hearable as tacitly referring to ‘white people’ and ‘people of color’. She goes on to make three references to collectivities (‘any of us in this room’, line 8; ‘we’, lines 9 and 10) that, in this sequential context, and given the designed racial composition of the group, are hearable as tacitly referring to group members as ‘white people’. She then produces a somewhat complex list of items that formulate anti-racist actions that participants in the ‘white’ group can take. In offering one item (‘talk with people of color’, line 11), she overtly refers to ‘people of color’ as a collectivity. After offering the next item (‘talk with each other’, lines 11–12), which tacitly refers to ‘each other’ as white people exclusively, she begins another item on the list (‘to sart- s-’ which is apparently headed towards ‘start support groups’).15 However, before completing the item, she cuts it off and inserts ‘white’ (at line 12), yielding ‘white support groups’. She thereby repairs a formulation-in-progress that could be heard as a non-race-specific reference (and therefore a suggestion about forming a multiracial support group) into an explicitly ‘white’ racial reference.
Excerpt 7 [TDC WG, 26]
By converting what could be heard as a non-racial formulation-in-progress (‘support groups’) to a racial formulation (‘white support groups’), the trainer treats the non-racial version as possibly ambiguous and in need of further specification. In this case, then, as in Excerpt 6, the inclusion of the category ‘white’ appears to be a result of the ambiguity produced by the composition of the speaker’s turn-so-far, rather than being warranted by the use of ‘white’ as an account. The ambiguity in this case thus appears to have been produced by the speaker making the race MCD locally relevant through her overt reference to ‘people of color’ at line 11 and then, more proximately, through the just preceding suggestion that drew together white people and people of color (at lines 11–12).
While the self-initiated repairs in the above excerpts demonstrate speakers’ orientation to the potential ambiguity of their own non-racial references after the race MCD has been made relevant, Excerpt 8 shows that recipients can also be oriented to such referential ambiguity. In this case, taken from a ‘group
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
624 Discourse & Society 20(5)
therapy session’ for teenagers, Roger formulates a group of students using a racial category (‘colored guys’, line 1) and then connects them explicitly to ‘fighting’ (‘So there was many fights needless to say’, line 8). Subsequently, Roger connects himself to ‘fighting’ (‘Most of the time I was kicked out for fighting’, line 15), without specifying who he fought with – but perhaps, given the sequential position of his assertion, intimating that the fights may have been with black students. The therapist (Dan) then pursues an explicit formulation (at line 22) of who Roger had fought with.
Excerpt 8 [GTS II, 36–7]
  1. 1  Roger:
  2. 2  Ken:
  3. 3  Roger:
  4. 4  Ken:
  1. 6  Roger:
  2. 7  Al:
  3. 8  Roger:
  1. 10  Roger:
  2. 11  ():
13 Al:

15 Roger: 16
17 Ken: 18

  1. 19  Roger:
  2. 20  Al:
  1. 22  Dan:
  2. 23  Roger:
25 Roger:
So these colored guys,
hhh heh
They thought they were bad shit.
Probally were. (Matter of fact.)
hh heh aie
So there was many fights needless to say.
And I wasn’t what [you’d call an easy kid to get along with.

Mosta the time I was
kicked out fer fightin. (2.5)
Yer going, (1.2)
I’m gone. Hah

Fighting with whom?
White kids.
I only once hadda fight with a
colored guy.
Here, Dan is not simply targeting the fact that who Roger fought with was left unsaid, but rather ‘Fighting with whom?’ can be understood as being asked as a consequence of the local relevance of race for categorizing students. Dan thus orients to the ambiguity over whether Roger’s non-specification of the race of the student(s) he fought with could be taken to mean that either they were of the same racial category as the last explicitly referred-to category (i.e. ‘colored’), or members of the invisible or taken-for-granted category for this speaker and these recipients (i.e. ‘white’) which would not ordinarily require overt formulation. Roger then supplies the clarification sought by Dan, identifying the previously unspecified racial category as being the taken-for-granted category (at line 23), and thus treating Dan’s question as having been a request for the racial cat- egory of the people involved. In this way, as in earlier excerpts in this section, a locally initial racial reference is followed by a subsequent non-racial reference
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 625
being treated as ambiguous and in need of clarification, although this time by a recipient rather than the original speaker.
These cases expose one sequential environment (i.e. after the race MCD has been made relevant by the production of a locally initial racial reference) in which the race of ordinarily invisible white persons may surface overtly. It is in environments such as this – where race is topicalized or overt racial refer- ences are employed – that the ordinary asymmetry of racial categories can be countervailed, and (following the consistency rule) the taken-for-granted cat- egory associated with non-racial references can no longer be unambiguously assumed to apply. An orientation to this can be seen when a formulation is repaired by a speaker, or when repair is initiated by a recipient, resulting in the overt formulation of the otherwise taken-for-granted category (in these cases, ‘white’), thus disturbing its ordinary invisibility.
By contrast, in the next section, we examine the surfacing of racial for- mulations in environments in which racial membership categories do not seem to be relevant at all. We offer one possible systematic basis for otherwise seemingly gratuitous uses of a racial membership category by speakers. First, we present two cases in which a racial formulation is supplied where not doing so is treated as giving the impression – for this formulation, this recipient, this speaker and this sequential environment – that the person so-formulated will be understood to be (invisibly) white. Next, we show that this practice is also employed on occas- ions where the local taken-for-granted racial category is something other than ‘white’. We then conclude the section by describing one (apparently unintended) consequence of this practice.
Descriptive adequacy: a systematic basis for racial categorization
While we have shown that speakers employ racial formulations that operate to provide adequate accounts for actions, not all racial formulations are employed in this fashion. Some appear to be used gratuitously, insofar as the referred-to person’s race is not topically relevant, and the category-bound inferences and actions that can accrue to the category do not seem to be relevant (cf. Kitzinger, 2007). In this section, we show that some seemingly gratuitous instances of racial categories in fact appear to have a systematic basis in the organization of talk-in-interaction. When describing a person in terms of their visible physical features, speakers can employ a racial category in cases where a recipient may well otherwise envisage a person of another membership category. This exposes the categorical asymmetry between racial categories: the use of one racial category for descriptive adequacy constitutes an interactional mechanism through which a second racial category is revealed as the locally taken-for-granted racial category.
In demonstrating this phenomenon, we begin with cases in which a person is overtly identified as black, thereby revealing ‘white’ as the locally taken- for-granted category. We then show that, in some circumstances, ‘white’ can be employed for descriptive adequacy, revealing another membership category as
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
626 Discourse & Society 20(5)
taken-for-granted – and thereby establishing another environment where persons can be overtly formulated as ‘white’. This surfacing of ‘white’ thus rests on the ordinary operation of the categorical asymmetry of race.
At the beginning of Excerpt 9, Louise is telling a story about how surprisingly tall some 12-year-olds are. In receipt of this story, Ken tells a second story (Sacks, 1995: 764) that supports her observation. He tells about how tall some (formerly) short students at the military school he once attended had become. In the course of referring to one of the students as an example, Ken elaborates the reference with a seemingly gratuitous racial identification at lines 18–19 (‘Fred Thomas who’s a colored guy, at school?’).
Excerpt 9 [GTS I, 104–5]
1 Louise: 2
  1. 3  Roger:
  2. 4  Louise:
  3. 5  Roger:
  4. 6  Louise:
  5. 7  Ken:
  6. 8  Louise:
Y’know some of ’em are damn tall and good looking they could pass for (t)-nineteen. [A twelve year old guy comes over
[But they don’t-
I say who’s y- older brother is he? He’s not he’s in the A7.

But they don’t have a brain to go with it hehhh
These kids I don’t believe it they’re six foot,
They’re five eight y’know, you’re looking up, he’s twelve and a
  1. 9  half !
  2. 10  Ken: There’s a-
  3. 11  Roger: ( )
  4. 12  Louise: They are.
13 Ken: 14

17 18 19 20
  1. 21  (Ther:)
  2. 22  Louise:
There’s a lotta people- lotta kids now who seem to be getting a
lot taller. Cause I useta- I useta have a complex I really did
for being so dog-g-doggone tall. In military school?
Everybody’d go ‘There’s Storky’ They called me Storky for
years! And now I-I have to look up to some of these guys you
know, it’s amazing!
There- there’s Fred Thomas who’s a colored guy, at school? He’s- he’s two and a half inches taller than I
And he used to look up to you.
Although height, rather than race, is the physical feature of relevance to the second story that ‘Fred Thomas’ is used to exemplify, Ken includes a racial iden- tification as part of a parenthetical expansion of the reference. The main work of the parenthetical expansion is to connect the referent to the military school and thus to the place Ken himself had been singled out as especially tall (and therefore given the nickname ‘Storky’). The racial identification does not appear to be doing any additional referential or explanatory work here, since Ken goes on to state that the referent is now ‘two and a half inches taller than I am’ (lines 19–20), thus continuing to treat height as the primary feature of relevance. Furthermore, Ken has worked to produce the racial identification in a way that minimizes its interference with his point about height, placing it in a subsidiary or secondary position in the reference (cf. Kitzinger, 2000: 181–8), between the referent’s name and how the speaker knows him (i.e. as a classmate). In addition,
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 627
the racial identification does not seem to be produced as derogatory in any way and thus appears to be completely gratuitous. In light of this, we might ask why Ken adds a racial formulation at just this point in the telling, if that formulation is not doing any relevant referential or explanatory work.
It seems to us that the solution to this puzzle can be found in the operation of ‘recipient design’ in talk-in-interaction (Sacks and Schegloff, 1979; Sacks et al., 1974). Recipient design concerns the myriad ways that speakers adjust what they are saying and the way they say it to accommodate the particular recipients they are now addressing. This may include an orientation to a need to overtly specify a racial category when introducing a character in a story who is not a member of the category recipients would otherwise take for granted. Ken thus adds a racial identification in the service of a physical description where a failure to do so would likely result in his recipients presuming that the referent was a member of the taken-for-granted category; that is, they would presume, unless otherwise informed, that he was racially ‘just like us’ (i.e. ‘white’). In a way, this parallels Land and Kitzinger’s (2005) analysis of telephone calls involving lesbian speakers. They demonstrate that not identifying a referent’s membership in a non-normative sexual orientation category at the first available opportunity may result in trouble later in the interaction, as a result of recipients incorrectly assuming that the referent is heterosexual.
In the case of lesbian partners, there is a systematic possibility that the sex category of the speaker’s partner will surface incidentally in subsequent talk, given the gender-identifying feature of the (singular) pronouns ‘he’ and ‘she’ used for locally subsequent person reference in English. This same systematic possibility does not accrue to membership in a racial category, but there is always the possibility that racial identity can nonetheless surface in subsequent talk. It could thus be the case that some seemingly gratuitous racial references serve as a method for pre-empting the kind of social–interactional difficulties that may arise if speakers do not identify a referent’s not-taken-for-granted racial category membership at the first opportunity, even when race is not topically relevant and does not seem to be specifically relevant for understanding the referent’s conduct.16
Ken’s introduction of the referent to exemplify a visible feature (his height) may prove to reveal a more systematic recipient design basis for seemingly gratuitous racial identification. There is evidence that speakers can be oriented to what their descriptions may educe recipients to envisage, and therefore they may subsequently add an increment to their turn so as to modify the image they are portraying.17 It is thus worth entertaining the possibility that here Ken is oriented to the likelihood that his formulation of the referent’s height will lead his recipients to envisage the referent’s appearance, such that neglecting to identify his racial category – when he is a member of a racial category that cannot be taken-for-granted – may result in recipients mistakenly envisaging an individual of the taken-for-granted category. This recipient design consideration, occasioned by the relevance of a description of other observable features of a referent, reveals the importance of the taken-for-granted (e.g. whiteness) in
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
628 Discourse & Society 20(5)
its absence. Here, we can see one way that a taken-for-granted racial category can be consequential for action in talk-in-interaction, even as it remains a tacit ingredient of the action.
A strikingly similar case of the apparent operation of this mechanism can be found in Excerpt 10, in which the participants of a dinner conversation are discussing the topic of age. In this case, a speaker inserts a racial identification into her description of a news story she has recently read about ‘very old people’, and about one such person in particular. Here, age, especially old age, is explicitly treated as a visible feature at lines 1–2 (‘That’s really frightening when your daughter looks as old as you do’). Subsequently, in telling about the ‘oldest American’, Beth identifies him as black (at line 9).
Excerpt 10 [Chinese Dinner, 24–5]
1 DON: 2
3 BET: 4

  1. 5  ANN:
  2. 6  BET:
  3. 7  JOH:
  4. 8  BET:
  5. 9  BET:
  6. 10  JOH:
  7. 11  DON:
  8. 12  DON:
  9. 13  BET:
  10. 14  DON:
  11. 15  JOH:
  12. 16  BET:
=That’s really frightening when your daughter looks ez old ez you do or something.
Y’know I’ve been reading about very old people lately,

[Like th’had en article in the Roll[ing Stone with this guy=
[( ).
=a hundred’n[thirdy.
The oldest] Ameri[can. He’s] a [black guy=]
=who’s suppose t’be=
[Wuh-what is thi:s.]

=who [lives in] Florida ’n th[ey- [Peking. ] [
[ ] [That is- ]
[ Beef, ]
[Wo:[:w. [innerviewed im.
While surprising tallness was the newsworthy visible feature in Excerpt 9, surprising longevity is of topical concern in Excerpt 10. Despite this, after initially referring to the referent simply as ‘this guy’ (at line 6), and then going on to specify his age and nationality (which are both topically relevant), Beth identifies him with a racial category (‘black’), before going on to specify the state in which he lives (‘Florida’, line 13).18 Again, as in Excerpt 9, the racial iden- tification is placed in a subsidiary or secondary position in the reference, between the identification of the referent’s age and his location. Furthermore (and again as in Excerpt 9), this racial identification was not required for the purposes of the point the speaker was making about the age of the referent or as an ex- planation of it, and the racial category does not seem to have been produced as derogatory, or pursued by the speaker after its production. Moreover, John’s uptake in line 15 (‘Wo::w’) displays his orientation to the story as being pri- marily concerned with the remarkable age of the referent, rather than being about his racial category. However, as in Excerpt 9, this referent was connected to a noteworthy observable feature, such that not supplying a not-taken-for- granted racial category could result in recipients incorrectly envisaging a member of the taken-for-granted racial category. This may be particularly relevant here
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 629
in light of Beth’s identification of the referent as ‘American’. That is, her sub- sequent racial identification of the referent may be evidence of an orientation on her part to a presumption that, unless otherwise specified, ‘American’ will be taken-for-granted by her recipients to mean ‘white’. Thus, in both Excerpts 9 and 10, a recipient design consideration, bound up with the description of a visible feature of a person, seems to provide a basis for the use of a racial refer- ence.19 Having described the operation of descriptive adequacy in the organ- ization of recipient design, we now turn to a case in which the operation of descriptive adequacy yields another way in which persons can be ‘white’.
In Excerpts 9 and 10, the presumptive taken-for-granted category oriented to by the speaker is ‘white’. However, ‘white’ is not always the taken-for-granted racial category. Racial categories that are not ordinarily taken-for-granted in conversation can serve as the presumptive taken-for-granted category in some circumstances, such as in interactional contexts in which the ratified partici- pants are all of the same other-than-white racial membership category (see, for example, Bucholtz, 2001; Cutler, 2007; Hartigan, 1999), and the otherwise taken-for-granted category ‘white’ can then be employed overtly to achieve de- scriptive adequacy. In such cases, the same type of asymmetry obtains, and the ame practice is involved, but the positions of the particular categories are different. In this way, taken-for-granted categories can be understood as taken- for-granted for this speaker when mentioning this category of referent among these co-participants in this sequential environment.20
In Excerpt 11, ‘Hispanic’ is revealed as a taken-for-granted category when ‘white’ is employed to achieve descriptive adequacy. In this excerpt, the par- ticipants in a barbershop conversation have been discussing the nephew of one of the participants and his prospects as a football player.
Excerpt 11 [Barbershop, 2–3]
1 Antonio: 2
  1. 3  Nick:
  2. 4  Barber:
  3. 5  Antonio:
  4. 6  Barber:
  5. 7  Johnny:
  1. 9  Barber:
  2. 10  Antonio:
  3. 11  Barber:
13 Barber: 14
15 Antonio: 16
17 Barber:
<Yeah he’s still growin like a weed too ’ez about six I think ’ez about six three:?
[right now?=
=’ez about two eighty five. How taller you Johnny, I’m six two:.

(0.8) Aright.
Yeah he’s very comprable d- t- [ta (him)
[Yeah:: I would think so.

I wouldn’ want eny uh these guys
runnin at me (.) with their equipmen on [( )
[Well my nephew his mother’s white so he looks- he looks more white th’n Hispanic.
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
630 Discourse & Society 20(5)
18 19 20 21
Wi- the b- wi the< s- body size come from (.) her side er or the [dad’s side.

[Got it from my side my brother yeah
Boasting (at lines 1–2), Antonio gives a visual approximation of his nephew’s height and then, after two of his recipients register this, he adds an approximation of the nephew’s weight (at line 5).21 At line 6, the barber, apparently for the sake of comparison, asks Johnny how tall he is. Following Johnny’s response (line 7), and the barber’s receipt (line 9), Antonio explicitly confirms the comparison the barber was making between his nephew and Johnny (line 10). The barber then agrees with this assessment (line 11), and begins to produce an appreciation of players of this stature (‘I wouldn’t want any of these guys running at me with their equipment on’). It is at this point, and just after the barber has remarked on envisaging players ‘running at me’, that Antonio produces a racial identification that explicitly treats the taken-for-granted category as ‘Hispanic’. That is, after giving a reason for his appearance (‘his mother’s white so’, line 15), he reveals that his nephew ‘looks more white than Hispanic’ (line 16).
As in Excerpts 9 and 10, race is not the feature of primary topical relevance (instead, height, and more generally physical build, are topically relevant), so it was not topically necessary for Antonio to head-off any possible misconception about his nephew’s (apparent) racial category for assessment of his build as a football player.22 Furthermore, the racial identification does not seem to be em- ployed to explain action. However, as in the earlier excerpts, the speaker reveals an orientation to recipient design concerns with respect to race, in ensuring that recipients are aware of the referent’s non-membership (or in this case, perhaps, marginal membership) in what is treated in this circumstance as the taken- for-granted racial category. Moreover, as in Excerpts 9 and 10, the feature for which the referent is being invoked is one involving visible features, such that Antonio may be oriented to the possibility that, if he does not specify the racial appearance of the referent, his recipients (especially given the kinship relation- ship of the referent) will very likely envisage someone who ‘looks Hispanic’. This case adds further evidence for speakers’ orientation to descriptive adequacy in formulating references to non-present persons, and constitutes another way in which persons can be ‘white’.
While the above excerpts provide evidence for a systematic recipient design basis for the production of some seemingly gratuitous racial references, Excerpt 12 demonstrates one potentially systematic consequence of practices associated with an orientation to descriptive adequacy: namely that such overt mentioning of race, once produced, can subsequently be taken up by recipients as an account for action, even when it was not employed for that purpose initially. In this excerpt, Roger draws on a racial membership category, first employed by another speaker to satisfy descriptive adequacy, in order to provide a sardonic account for the lack of attention by authorities to the shooting of a ‘little colored kid’ in Dallas at
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
1 Louise: 2

5 Louise: 6
7 Louise: 8

9 10
11 Roger: 12
  1. 13  Roger:
  2. 14  Ken:
  3. 15  Roger:
  4. 16  Louise:
  5. 17  Roger:
  6. 18  ():
  7. 19  Roger:
  8. 20  Louise:
  9. 21  Roger:
  10. 22  Ken:
  11. 23  Roger:
  12. 24  Louise:
  13. 25  ( ):
Whitehead and Lerner: When are persons ‘white’? 631
the time of the Kennedy assassination. Here, the link between the membership category ‘colored’ and what can be inferred about members of that category (and specifically relational inferences about how members are treated by others and why they would be so treated) is made explicit as a method for defeating Louise’s accusation of a cover-up.
Excerpt 12 [GTS I, 86–7]
Y’know they’re not even sure.
You know there was another bullet?
A liddle
colored kid w’z brought in,
Yihknow when this happened
I w’z watchin it. .hhh A liddle colored kid was brought into the hospit’l with a bullet wound. ’n they never said anything after that. I mean,
Why should it it’s in Dallas.
heh .hhhh
ehh [heh!

[Bullets are intended fer [liddle [mhhhh
c(h)olored [k(h)ids. [heh heh
ehh .hh! .hhh .hhh heh heh [heh
[Keep ’em from growin’ up into big colored me:n.
heh! hhhh
At line 3, Louise first asserts the claim that there was a second bullet fired at the scene of the Kennedy assassination and then, following a 1.4-second interval, she proceeds to describe the evidence for her claim (‘A little colored kid was brought in’, line 5). Although this is produced with continuing intonation, she does not immediately continue. After an interval during which there is apparently no uptake, Louise seems to restart her turn in a manner that makes clear she is going to produce her evidence as a story. She does this by halting the progressive realization of her story and repairing its delivery to describe what she was doing at the time (‘when this happened I was watching it’, line 7), thereby also giving an epistemic basis for her claim: seeing it on television. Her explicitly visual for- mulation of her description provides evidence for the operation in this case, as in the previous ones we have examined, of an orientation on the speaker’s part to descriptive adequacy. At lines 7–8, Louise reiterates her claim and expands it tomaketheconnectionexplicit(‘Alittlecoloredkidwasbroughtintothehospital with a bullet wound’), thus reproducing the overt racial reference as ‘what she saw’. However, unlike the earlier cases, the overt racial reference produced by
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
632 Discourse & Society 20(5)
the speaker is taken up by a recipient, thus demonstrating that, once a racial reference surfaces, it can be exploited for its ‘inference rich’ features.
Following Louise’s suggestion of a cover-up of additional bullets fired during the assassination (‘they never said anything after that’, line 9), Roger counters her suggestion that this absence should be understood as an active suppres- sion of evidence. He dismisses it (at line 11) by asserting that shooting black people is so routine as to not deserve to be treated as newsworthy in Dallas. In other words, he claims it’s the normal way shootings are covered in Dallas – at least when it comes to black people, they are not worthy of note. Roger thus takes advantage of Louise’s mention of a racial membership category to explain why nothing was subsequently said about the shooting of the child. This dis- counting of Louise’s suggestion that news of another shooting had been sup- pressed results in laughter from several participants, including Louise herself. This response by the other participants shows that they understood the inference Roger was making, and the sardonic humor embedded therein, thus demon- strating their appreciation of the ‘race + place’ explanation that Roger has produced.
Roger then continues his account, making explicit the reason ‘bullets are intended for little colored kids’; they ‘Keep them from growing up into big colored men’ (line 23). He thus invokes the image of the threatening nature of ‘big colored men’, which was particularly salient in the South during the Civil Rights era, to account for why it would be common practice in Dallas to kill black men while they are still ‘little’ and non-threatening, before they can grow up to become ‘big’ and threatening.23 Thus, although Louise’s reference to the race of the child did not appear to be designed as topically salient or as an account for action when it was introduced, Roger uses the reference opportunistically against her by topicalizing the category and using it as an explicit account for the (in)action of the law-enforcement authorities.
Concluding remarks
In this report, we have investigated the practical asymmetries that obtain be- tween different categories of racial membership as they are employed in talk-in- interaction. It is important to underscore that these asymmetries are produced or reproduced as situated interactional accomplishments. In each of the instances we have examined, one racial category stands – for the practical purpose at hand at just that moment – in an asymmetrical relationship with other categories by virtue of being treated as taken-for-granted, in contrast to the marked visibility of other categories. We have focused in particular on the surfacing of the ordinarily invisible category ‘white’, identifying three interactional environments in which ‘white’ is employed overtly, and we have described the mechanisms through which each can occur.
First, we have shown that ‘white’ may surface as an account for action ‘just in time’ to serve as an account. Such surfacing of ‘white’ reveals one problem for an ordinarily invisible racial category: it can’t be employed as an explicit
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 633
account without being made visible. Thus, occasions on which ‘white’ is needed as an account present recurrent occasions on which its ordinary invisibility can be disturbed.
Second, we have shown that ‘white’ may be exposed systematically in order to repair referential ambiguities arising in sequential environments in which race has been made relevant. This demonstrates the importance of sequential position in determining when and how some racial references are formulated, and particularly when otherwise taken-for-granted categories are employed overtly. Our data show how both speakers and recipients orient to the ambiguity that can result from asymmetrical practices for formulating references to taken- for-granted and non-taken-for-granted categories once the race MCD has been made relevant. Ordinarily, allusions to otherwise tacit racial categories beside (or in direct contrast to) overtly mentioned categories do not result in any apparent trouble for speakers or their recipients (as we show in Excerpt 2). Yet, it is the very invisibility of whiteness – i.e. its absence from person reference and descrip- tion even when race is locally relevant – that itself can be a source of referential ambiguity for locally subsequent (racial) references. Here, the routine, largely mundane organization of references to racial categories in talk-in-interaction systematically provides for the possibility of interactional difficulties, in the form ofreferentialambiguity,oflocallysubsequent,raciallyrelevantreferences.24 This practice reveals something of a ‘cultural paradox’ built into the asymmetrical character of tacit (racial) reference itself: the normal invisibility of ‘white’ as a race category turns out to be a source of trouble, and thus the very source of its overt usage.
Third, ‘white’ may surface through the operation of the recipient design consideration of descriptive adequacy. The operation of descriptive adequacy in what otherwise seem like gratuitous references to race suggests that the use of references to racial categories is not simply a consequence of a participant’s personal belief in the relevance or importance of racial categories, but of a par- ticipant’s orientation – as a speaker – to such categories as important resources for other participants – as recipients – in grasping the portrayal of a person, even when race is not being used to account for action. Thus, a recipient design consideration may underwrite apparently gratuitous racial references on many occasions, and it is the ordinary invisibility of whiteness (or another locally invisible racial category) as taken-for-granted that can occasion the overt racialization of an otherwise non-racial formulation. Moreover, we have shown that once employed in this way – thereby making the race MCD relevant – a now-relevant racial membership category can be seized upon opportunistically as an account for action. In this way, a formulation that is, in the first instance, produced to satisfy an endogenous-to-conversation consideration (recipient design) furnishes a systematic basis for subsequently employing race as an account for action.
Taken together, our findings demonstrate some of the ways racial categories are employed in conversation, showing one way that society’s social organ- ization can be found in its enabling sequential organization of conduct at the point of its production. The cases we have examined are exceptional in the sense
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
634 Discourse & Society 20(5)
that they represent occasions on which the otherwise taken-for-granted status of ‘white’ as seen-but-unnoticed is disturbed. Ordinarily, taken-for-granted categories are not mentioned, and thereby remain implicit and invisible, while other categories (from the same collection of categories) are mentioned, thereby routinely and recurrently reproducing the asymmetrical relationships between categories. This reveals the degree to which not disturbing the invisibility of whiteness (or any other locally taken-for-granted category) constitutes a largely unnoticed interactional achievement in its own right.
While we have focused primarily on cases in which ‘white’ is treated as an invisible category, our findings suggest that other racial categories can be treated as taken-for-granted, and that the interactional mechanisms that pro- duce and reproduce this invisibility, and provide for the surfacing of ordinarily invisible categories, may be practically the same no matter which category is treated as taken-for-granted. In saying this, we are not proposing that other ‘cir- cumstantially taken-for-granted’ racial categories are to be viewed in precisely the same way that ‘white’ is viewed, given the dominant place whiteness holds in society (Du Bois’ [2003 (1903)] concept of ‘double consciousness’ again comes to mind here). Nevertheless, empirically, ‘white’ does surface to satisfy descriptive adequacy in some contexts, thus demonstrating that the invisibility of whiteness is not inevitable regardless of the particulars of the context.
The unavoidable historical and cultural specificity of our data (or any other field recordings) has raised the question for a number of readers as to whether our findings may be specific to particular historical periods. In response, we would note that although some aspects of the lexical content of the data in question (e.g. the term ‘colored’) and the social knowledge bound to particular racial categories may be associated with particular historical contexts, the practices we have identified here appear empirically to be stable over time, operating in similar ways throughout our data. Thus, the use of race as an account for action during the 1960s may rely on different inferences about a racial category than those that would be mobilized by a mention of the same category in the 1990s or 2000s. However, the possibility of using race as an account, and the interactional mechanism through which this is achieved, is fundamentally the same across all the time periods represented in our data. Thus, while the content of category-bound knowledge and inferences may change over time, the mechanism through which these inferences are mobilized appears to have remained stable. Similarly, the content of speakers’ utterances in the sequential environments we have identified in which taken-for-granted cat- egories get overtly mentioned may be historically contingent, and (as we have shown) which category is treated as invisible may be context-specific, but the mechanisms through which an otherwise invisible category surfaces appear to be stable across contexts. Moreover, the apparent robustness and stability of these interactional mechanisms is further evidenced by data the first author is currently collecting in South Africa as part of ongoing research on the use of racial categories in talk-in-interaction. Although this data is more recent and was collected in a different national context from the data we present in this article (and includes a substantial proportion of second-language English
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 635
speakers), we have found thus far that the findings we present in this article are consistent with what we are observing in the South African data. However, by describing the operation of asymmetrically employed racial categories, we hope to have provided accounts of enduring features of race in everyday life that can be employed as analytic tools by scholars who wish to examine aspects of racial discourse that are shaped by their historical context and that have changed significantly over time.
Finally, because our findings demonstrate some ways in which the mundane invis- ibility of whiteness can be exposed as a part of routine interactional processes, they can also suggest how such invisibility might be subverted. Speakers who choose not to treat ‘white’ as an invisible category in the course of their day-to- day encounters with others can be understood as engaging in everyday acts of resistance against a society’s hegemonic whiteness – subverting that hegemony at the point of its mundane production, and using the basic machinery of sociality to do so.25 This reveals the importance of this basic machinery of talk- in-interaction for developing an understanding of both the reproduction of, and resistance to, the racial dynamics of everyday life.
  1. An earlier version of this report was presented at the 2008 Annual Conference of the International Communication Association in Montreal. We thank Celia Kitzinger for her helpful comments on an earlier draft of this report.
  2. Only a small proportion of the field recordings we examined for this project comes from interactions in which race was the announced topic of discussion, and the discussions in these cases were not prompted by researchers.
  3. Here, we present a case that does not employ the race MCD. We use this case to illustrate as simply as possible how membership categories can be employed as tacit accounts. Sacks’ own work demonstrates how this operates for the race MCD and we will not duplicate that here. All of the subsequent excerpts are taken from recorded data and concern racial reference.
  4. Another noteworthy feature of this case is that Passenger A can rely on the heterosexuality manifest in the husband–wife team-type collection of categories as an interactional resource in making this request. Thus, a female speaker would have difficulty in using a reference to ‘my wife’ as a tacit account for a request without being heard as flaunting her sexual orientation. In this way, heteronormativity is realized here through the asymmetrical availability of categories of sexual orien- tation – an availability that is built into such team-type membership categorization devices as ‘husband–wife’ and ‘family’. See Kitzinger (2005a, 2005b; Land and Kitzinger, 2005) for an examination of sequential environments in which this form of asymmetry surfaces in interaction.
  5. Whitehead (in press) identifies additional practices for the tacit formulation of race in talk-in-interaction. In addition, Sacks (1995: 42) has described how actions bound to particular membership categories (as well as other identifying features of some categories, e.g. recognizably ‘Jewish names’) can also be employed to tacitly refer
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
636 Discourse & Society 20(5)
to a person as a member of a category. Also relevant here is Sacks’ (1995: 47–8)
discussion of MCDs, such as race, that can be formulated as ‘two-set classes’.
  1. The asymmetrical granularity of the two place formulations (‘hotel’ is a specific establishment that serves alcohol, whereas ‘black section’ merely formulates a part
    of town) also contributes to the disqualification of the overtly racial formulation.
  2. A parallel body of literature by feminist scholars critiques ‘invisible masculinity’, whereby ‘Masculinity is the unmarked form: the assumption [is] that the world is
    male unless proven otherwise’ (Spender, 1980: 20).
  3. Kitzinger (2005a: 257) notes the parallels between whiteness and heterosexuality
    in this regard, and suggests that the conversation analytic approach she used to examine the invisibility of heterosexuality could also be used to examine whiteness and ‘any other hegemonic identity’.
  4. Sacks (1995: 479–82) has also shown how the connection between membership categories and ‘category-bound actions’ can account for some actions being treated as illegitimate or as ‘imitations’ when produced by someone seen as belonging to a different category.
  5. See Whitehead (2007) for a description of how the speaker in this case uses ‘proxies for race’ to establish a (black) racialized location.
  6. In this way, as Whitehead (in press) shows, Sammy is able to propose that the attack was racially motivated, without attributing the actions of the attackers to their racial category. That is, he shows that they attacked his friend because he was white, rather than proposing that they did so because they were black.
  7. The action Ken is about to formulate in the next part of his utterance would have made clear that the trouble source (‘these guys‘) refers tacitly to referents other than ‘colored people’. The repair that Ken produces in its place is thus not required to disambiguate the trouble source, but is used to produce an overt racial reference that can serve as an account for the forthcoming action. (We take up referential ambiguity in the next section.)
  8. Interestingly, in one case in our data, a not-taken-for-granted racial reference is apparently held back until nearly the end of a story about being pulled over for a traffic stop by a police officer, thereby apparently allowing recipients to suppose the officer is white. As the teller nears the end of the story, the race of the officer is sprung as a kind of ‘last straw’. For a discussion of the ‘suppression’ and ‘release’ in this case, see Jefferson (1996: 24).
    [GTS III, 34–5]
1 Ken: 2

5 6 7 8 9
11 Ken: 12
He sez uh ‘Can I see your license? Where’s your operator’s license?’ you know, ‘Yeah,’ I took my wallet out, he sez uh ‘Take
it out, please.’ I took it out, he looks on the back, he sez ‘mh
hm,’ Looks at it real carefully an’ sees I’m not eighteen. He sez ‘Well you know, you went two foot over that stop sign. Now uh, I’m very sorry but I’m gonna have to writechu out a ti-citation
on this.’ Well my jaw- ‘Well wha-wha-wha-wha-’ Y’hhhh he sez ‘Is against the law to go over the white line.’ An’ he gives me a
big long lecture
an’ he’s a colored guy.
An’ it burned me up you know, because if it was an adult, they sure wouldn’t stop an adult, you know, somebody- somebody thirty, thirty five years old,
14. Schegloff (1996: 450ff) distinguishes between locally initial and locally sub- sequent positions for references to the same person. Here, we use the term ‘locally
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
1 Ann: 2

5 6 7 8 9
  1. 10  Bob:
  2. 11  Cathy:
  3. 12  Ann:
14 Cathy:
Whitehead and Lerner: When are persons ‘white’? 637
subsequent racial reference position’ to refer to the position of a person reference produced after a locally initial racial reference has been employed (thus making race – the race membership categorization device – relevant for subsequent refer- ences to persons). This can be understood as a consequence of the ‘consistency rule’ Sacks (1995: 246) has described for the operation of MCDs. The consistency rule states that ‘if a category from some [membership categorization] device’s collection has been used to categorize a first Member of the population, then that category or other categories from the same collection may be used to categorize further Members of the population’ (Sacks, 1995: 246; emphasis in original).
  1. The speech production error ‘sart’ can be understood as a blend of ‘start’ and ‘support’ and as the first half of a ‘Spoonerism’ in which the first sounds of two words are swapped. This in itself does not preclude ‘white’ as the next word, but the subsequent cut off ‘s-’ does indicate that ‘support’ was going to be the next word.
  2. An orientation to descriptive adequacy can also be seen in the noticeable absence of racial identification. One reviewer offers the following field note: ‘I recently par- ticipated in a discussion between faculty members about a named student whom some of us knew by sight and others did not: when those who had apparently had her in their classes but couldn’t fit the name to a person asked what she looked like, people described her hair and her height; nobody made the obvious observation that (in our predominately white university) she is black.’ Here, on an occasion when race might furnish a pathway to recognition, it is apparently suppressed presumably to guard against possible challenges, or at least inferences about the speaker’s motives for employing race (cf. Whitehead, in press). Nevertheless, the relevance of race to satisfy descriptive adequacy can be found here in its noticeable absence.
  3. This orientation to what recipients may envisage during the telling of a story can be seen in the following excerpt. Briefly, a speaker (who is explaining what happened after she had hit her head while in the shower) adds a towel to her description of being ‘in the shower’ – as a kind of incremental afterthought. She next shifts the temporal location to match the insertion and then explicitly orients to what her recipients might have been envisaging before the repair – and within the story, before her father comes into the bathroom to help her. By adding the towel to her description, Ann makes herself ‘decent’ for both her father’s appearance and for her recipients.
    [MP3 Towel]
. . . an’ I knew I hit my head because it’s porcelain. (.)
And it was just me in the shower.

and a towel.
I was done with the shower.
by the way.=>Just in case any($)body car [es($)<.

[ugh huh huh hn

an’ my dad ended up having to=ah (0.5)
bre(h)ak the lo(h)ck¿

Note that this story is designed as a second story (Sacks, 1995: 764) whose primary recipient (Cathy) told the previous story, but the speaker turns to address Bob at
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
638 Discourse & Society 20(5)
‘I was done with the shower’ and ‘in case anybody cares’. These gaze shifts occasion Bob’s ‘Okay’. Here is one place where perhaps a ‘culture in action’ can be found in an orientation to the point of view of one’s recipients and perhaps where a membership category of a recipient (as male) may surface as relevant to action. Note that the storyteller had just complained about her brother barging in on her while she was in the shower as the reason she regularly locked the door. (See Lerner [2004: 155–6] for another instance of what might be termed ‘incremental modesty’.)
  1. Of course, for the speaker to report that the referent was black suggests that this was included in the news story she is recounting. However, the fact that the news story very likely identified him as black does not require her to mention it. Moreover, that it was probably a part of the story might be further evidence for the claim we are making here.
  2. We are not suggesting that the operation of descriptive adequacy of asymmetrically available racial categories is limited to environments where visible features of the referent are prominent. However, it was striking in our data that such otherwise seemingly unmotivated use of racial identification occurred in the environment of physical description.
  3. This reveals the importance of ‘categorizing the categorizer’ (see Sacks, 1995: 45) and other participants for understanding the operation of racial asymmetry (cf. Whitehead, in press).
  4. Giving his nephew’s height as an approximation (bolstered by the visual metaphor ‘growing like a weed’) suggests that he is basing his claims on visual inspection, and not on a report of a measurement.
  5. The category used here may be part of both the ‘ethnicity’ and ‘race’ collections, but Antonio’s emphasis on how he ‘looks’ makes relevant the phenotypic characteristics associated with race, rather than the cultural features associated with ethnicity. It may be the case, however, that the MCD they are employing here is a ‘blended collection’ of race and ethnicity categories reflecting a blurring, in practice, of the distinction between race and ethnicity.
  6. This case also demonstrates the conjoined use and intersecting import of race and sex categories to provide accounts for action, as well as the surfacing of a speaker’s orientation to the invisible masculinity found in the inference-rich sex categories. Even though Louise did not specify the gender of the child who was shot, Roger displays his assumption that the child was a male, and uses this category (along with the racial category Louise did supply) in his account for the authorities’ lack of action following this incident. Clearly, such an account would not have been possible if Louise had identified the child as a female, or if Roger had assumed that the child was female rather than male, since ‘big colored women’ would not carry the threatening connotation (particularly during that historical period) that ‘big colored men’ does.
  7. The production of overlapping talk that is built into the routine operation of turn- taking practices for conversation comes to mind as another case of ‘unavoidable’ systemic interactional trouble (Sacks et al., 1974: 706–8).
  8. This is an interactional tactic that is reminiscent of the early English Quakers’ refusal to use, for example, deferential titles and honorific pronouns. These speaking practices ‘challenged . . . the very fabric of social relations and social interaction’ (Bauman, 1983: 43).
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 639
Bauman, R. (1983) Let Your Words Be Few: Symbolism of Speaking and Silence among Seventeenth-Century Quakers. Cambridge: Cambridge University Press.
Bonilla-Silva, E. (2002) ‘The Linguistics of Color Blind Racism: How to Talk Nasty about Blacks without Sounding “Racist”’, Critical Psychology 28: 41–64.
Bucholtz, M. (2001) ‘The Whiteness of Nerds: Superstandard English and Racial Markedness’, Journal of Linguistic Anthropology 11: 84–100.
Bucholtz, M. and Trechter, S. (eds) (2001) Discourses of Whiteness: Special Issue of Journal of Linguistic Anthropology 11.
Buttny, R. (1997) ‘Reported Speech in Talking Race on Campus’, Human Communication Research 23: 475–504.
Collins, P.H. (2004) Black Sexual Politics: African Americans, Gender, and the New Racism. New York: Routledge.
Condor, S. (2006) ‘Public Prejudice as a Collaborative Accomplishment: Towards a Dialogic Social Psychology of Racism’, Journal of Community and Applied Social Psychology 16: 1–18.
Condor, S., Figgou, L., Abell, J., Gibson, S. and Stevenson, C. (2006) ‘“They’re not racist”: Prejudice, Denial, Mitigation and Suppression in Dialogue’, British Journal of Social Psychology 45: 441–62.
Cutler, C. (2007) ‘The Co-Construction of Whiteness in an MC Battle’, Pragmatics 17: 9–22. Du Bois, W.E.B. (2003 [1903]) The Souls of Black Folk. New York: Signet Classic. Frankenburg, R. (1993) White Women, Race Matters: The Social Construction of Whiteness.
Minneapolis: University of Minnesota Press.
Hartigan, J.J. (1999)
Racial Situations: Class Predicaments of Whiteness in Detroit. Princeton,
NJ: Princeton University Press.
Kitzinger, C. (2000) ‘Doing Feminist Conversation Analysis’,
Feminism and Psychology
10: 163–93.
Kitzinger, C. (2005a) ‘“Speaking as a Heterosexual”: (How) Does Sexuality Matter for

Talk-in-Interaction?’, Research on Language and Social Interaction 38: 221–65. Kitzinger, C. (2005b) ‘Heteronormativity in Action: Reproducing the Heterosexual Nuclear
Family in After Hours Medical Calls’, Social Problems 52: 477–98.
Kitzinger, C. (2007) ‘Is “Woman” Always Relevantly Gendered?’,
Gender and Language
1: 39–49.
Land, V. and Kitzinger, C. (2005) ‘Speaking as a Lesbian: Correcting the Heterosexist

Presumption’, Research on Language and Social Interaction 38: 371–416.
Lerner, G.H. (2004) ‘On the Place of Linguistic Resources in the Organization of Talk- in-Interaction: Grammar as Action in Prompting a Speaker to Elaborate’,
Research on
Language and Social Interaction 37: 151–84.
Lipsitz, G. (1995) ‘The Possessive Investment in Whiteness: Racialized Social Democracy

and the “White” Problem in American Studies’, American Quarterly 47: 369–87. Lipsitz, G. (1998) The Possessive Investment in Whiteness: How White People Profit from
Identity Politics. Philadelphia: Temple University Press.
McIntosh, P. (1988)
White Privilege and Male Privilege: A Personal Account of Coming to See
Correspondences Through Work and Women’s Studies. Wellesley, MA: Wellesley College
Center for Research on Women.
McKinney, J.R. (2003) ‘Diverse Perspectives on Doing Antiracism: The Younger

Generation’, in A.W. Doane and E. Bonilla-Silva (eds) White Out: The Continuing Significance of Racism, pp. 233–52. New York: Routledge.
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
640 Discourse & Society 20(5)
Sacks, H. (1972a) ‘On the Analyzability of Stories by Children’, in J.J. Gumperz and D. Hymes (eds) Directions in Sociolinguistics: The Ethnography of Communication, pp. 325–45. New York: Holt, Rinehart and Winston.
Sacks, H. (1972b) ‘An Initial Investigation of the Usability of Conversational Materials for Doing Sociology’, in D.N. Sudnow (ed.) Studies in Social Interaction, pp. 31–74. New York: The Free Press.
Sacks, H. (1984) ‘On Doing “Being Ordinary”’, in J.M. Atkinson and J.C. Heritage (eds) Structures of Social Action: Studies in Conversation Analysis, pp. 413–29. Cambridge: Cambridge University Press.
Sacks, H. (1986) ‘Some Considerations of a Story Told in Everyday Conversation’, Poetics 15: 127–38.
Sacks, H. (1995) Lectures on Conversation. Oxford: Blackwell.
Sacks, H. and Schegloff, E.A. (1979) ‘Two Preferences in the Organization of Reference to

Persons in Conversation and their Interaction’, in G. Psathas (ed.) Everyday Language:
Studies in Ethnomethodology, pp. 15–21. New York: Irvington.
Sacks, H., Schegloff, E.A. and Jefferson, G. (1974) ‘A Simplest Systematics for the

Organization of Turn Taking in Conversation’, Language 50: 696–735.
Schegloff, E.A. (1972) ‘Notes on a Conversational Practice: Formulating Place’, in D.N. Sudnow (ed.)
Studies in Social Interaction, pp. 75–119. New York: MacMillan and
The Free Press.
Schegloff, E.A. (1996) ‘Some Practices for Referring to Persons in Talk-in-interaction:

A Partial Sketch of a Systematics’, in B. Fox (ed.) Studies in Anaphora, pp. 437–85.
Amsterdam: John Benjamins.
Schegloff, E.A. (1997) ‘Whose Text, Whose Context?’,
Discourse & Society 8: 165–87. Simpson, J.L. (2008) ‘The Color-blind Double Bind: Whiteness and the (Im)Possibility of
Dialogue’, Communication Theory 18: 139–59.
Spender, D. (1980)
Man Made Language. London: Routledge.
Stokoe, E. and Edwards, D. (2007) ‘“Black this, Black that”’: Racial Insults and Reported

Speech in Neighbour Complaints and Police Interrogations’, Discourse & Society
18: 355–90.
van den Berg, H., Wetherell, M. and Houtkoop-Steenstra, H. (eds) (2004)
Analyzing Race
Talk: Multidisciplinary Perspectives on the Research Interview. Cambridge: Cambridge
University Press.
Van Dijk, T. (1987)
Communicating Racism: Ethnic Prejudice in Thought and Talk. Newbury
Park: Sage.
Van Dijk, T. (1992) ‘Discourse and the Denial of Racism’,
Discourse & Society 3:
Van Dijk, T. (1993)
Elite Discourse and Racism. Newbury Park: Sage.
Van Dijk, T., Ting-Toomey, S., Smitherman, G. and Troutman, D. (1997) ‘Discourse,

Ethnicity, Culture and Racism’, in T. van Dijk (ed.) Discourse as Social Interaction: Discourse Studies: A Multidisciplinary Introduction, Vol. 2, pp. 144–77. Thousand Oaks: Sage.
Verkuyten, M., de Jong, W. and Masson, K. (1994) ‘Racial Discourse, Attitude and Rhetorical Manoeuvres: Race Talk in the Netherlands’, Journal of Language and Social Psychology 13: 278–98.
West, C. and Fenstermaker, S. (2002) ‘Accountability in Action: The Accomplishment of Gender, Race and Class in a Meeting of the University of California Board of Regents’, Discourse & Society 13: 537–63.
Wetherell, M. and Potter, J. (1992) Mapping the Language of Racism: Discourse and the Legitimation of Exploitation. Hemel Hempstead: Harvester Wheatsheaf.
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009
Whitehead and Lerner: When are persons ‘white’? 641
Whitehead, K.A. (2007) ‘The Use, Management and Reproduction of Racial Common Sense in Interaction’, unpublished MA Thesis, Department of Sociology, University of California, Santa Barbara.
Whitehead, K.A. (in press) ‘Categorizing the Categorizer: The Management of Racial Common Sense in Interaction’, Social Psychology Quarterly.
Wilkinson, S. and Kitzinger, C. (2003) ‘Constructing Identities: A Feminist Convers- ation Analytic Approach to Positioning in Action’, in R. Harré and F. Moghaddam (eds) The Self and Others: Positioning Individuals and Groups in Personal, Political and Cultural Contexts, pp. 157–80. Westport, CT: Praeger.
KEVIN A. WHITEHEAD is a doctoral candidate in Sociology, and in the Language, Interaction and Social Organization (LISO) Interdisciplinary Emphasis, at the University of California, Santa Barbara. His research focuses on the development of an ethno- methodological, conversation analytic approach to studying race, and particularly on the ways in which racial categories are used, resisted, and reproduced in talk-in-interaction. A D D R E S S : Department of Sociology, University of California, Santa Barbara, CA 93106– 9430, USA. [email: kwhitehead@umail.ucsb.edu]
G E N E H . L E R N E R is Professor of Sociology and Linguistics at the University of California Santa Barbara and a member of the Language, Interaction and Social Organization (LISO) research group. His research centers on language use, body behavior and very young children insofar as these (together and separately) exhibit the formal structures and local organization of practical sequential action in interaction. A D D R E S S : Department of Sociology, University of California, Santa Barbara, CA 93106–9430, USA.
[email: Lerner@soc.ucsb.edu]
Downloaded from http://das.sagepub.com at UNIV CALIFORNIA SANTA BARBARA on September 4, 2009