Abstract
The police are more likely to arrest, write citations, and reciprocate with coercive responses that involve warnings and threats when citizens disrespect and resist the police. Warnings in natural discourse are considered to be benevolent and differ from threats only in the intention and attitude of those who issue such speech acts. This article examines how ordinary features of talk-in-interaction are adapted to meet the institutional exigencies of patrol work, and how the interactional order of closing sequences in traffic encounters are collaboratively produced by police and drivers alike. We analyze 50 traffic encounters involving mundane infractions such as speeding and running a red light from the United States using principles of conversation analysis. Our findings indicate that closing sequences can be viewed as mutually accomplished social action in three distinct ways that are similar to and different from ordinary conversations. The implications for police studies and procedural justice research are discussed.
Introduction
The police initiate traffic stops and give form to the occasion as an instance of bureaucratic interaction by conversationally opening the encounter in a way that syntactically marks their office. This institutional character is especially visible in traffic encounters that the police initiate for minor infractions such as failure to use a seatbelt, speeding, and running a red light. Findings from the United States indicate that the majority of police officers who pull over drivers explain the reason for the stop, which leads to increases in perceptions of legitimacy; perceptions of legitimacy decrease when the police do not provide a reason for the stop (Davis et al., 2018). These ordinary traffic encounters that unfold must be initiated by summoning drivers to pull over through the activation of sirens; the encounters must be conversationally opened through a bureaucratic request (e.g. “May I see your driver’s license”); the reason for the stop announced (e.g. “I’m pulling you over because you were doing 60 miles per hour in a 25 zone”); misunderstandings repaired; and finally, the encounter must be brought to an orderly close by having drivers sign the tickets (e.g. “I’m gonna ask you to sign here”). These steps constitute the structural segments of traffic encounters (Shon et al., 2020). How drivers respond to the police requests and questions during their encounters reflexively shape the outcome of the traffic encounter in social, legal, and moral ways (Snow, 2019; Van Maanen, 1978; Wells and Savigar, 2019).
A conversation-based study of police-citizen encounters is necessary because the recent public interest generated by the erosion of public confidence as citizens have questioned the legitimacy of the police given the highly publicized cases of police killings of unarmed civilians and the ensuing attempts at reform (Goldsmith, 2005; Murphy and Tyler, 2017). As traffic encounters represent the most pervasive site of legal interaction between the state and the public, how these types of incidents can be understood from a process-oriented view of social interaction using analysis of language rather than outcome-oriented studies that measure citizen satisfaction warrants further empirical scrutiny. Methodologically, a conversation-based approach to analysis is appealing because the raw data are plainly available for all to see; variables such as race, class, and gender are verifiable in the turns and in their implicativeness in the way that they shape the line-by-line sequences of talk. The close examination of sequences of talk illustrates a fundamental analytical concept in conversation analysis: sequential implicativeness. This analytic concept simply refers to the fact that what the second speaker does in her second turn has conditional relevance to what preceded in the prior turn (Schegloff, 1968: 1083).
The goal of conversation analysis is to demonstrate how intentions of speakers are made orderly for those who are engaged in the social interaction (Maynard and Clayman, 1991). Conversation analysis attempts to explicate how social structures are embodied in the context-free structures of talk itself, in their sequential organization, independent of the occasion for the interaction (Schegloff, 1991). The objective in conversation analysis is the direct study of social action, not the experience of social action as perceived and reported by the subjects themselves. Such an assumption leads to the view that social order is recurrent and that the order is produced in situ by the parties themselves. Questions of frequency and pervasiveness are tertiary to the discovery, description, and analysis of the structures of discourse. The direct study of social action, as engaged by the subject themselves, not their perception of their experience, fundamentally differentiates psychology-based studies of perception of experience from studies of conversation as direct social action. This method of analyzing social interactions thus begins with the “description and explication of the competences that ordinary speakers use and rely on in participating in intelligible, socially organized interaction” (Heritage and Atkinson, 1984: 1).
While natural social interactions such as dinner with friends are brought to a close in mutually ratified ways, by making future arrangements that result in terminal exchange of good-byes (Schegloff and Sacks, 1973), with no one person exercising undue advantage over another, the institutional nature of traffic stops affords the police with the authority to arbitrarily end encounters without moral culpability (Frank, 1982). This point is notable, for proper closings require the possibility of mentioning an unintroduced topic in the conversation—opening up a potential closing—in order to be considered socially appropriate. By closing, we refer to the way natural (e.g. dinner party) and institutional interactions (e.g. job interview) are accomplished and terminated using rituals that signal the end of a putative interaction. The police beckon drivers to interact with the activation of their sirens and horns, but they must also communicate their readiness to bring the encounter to an end. There are obvious differences between natural social interactions and traffic encounters as they relate to closing sequences. For example, drivers do not issue dinner invitations to police who have pulled them over. The police must bring the encounter to a close despite the application of punitive sanctions such as tickets and arrests.
The closing sequences in traffic encounters are an important topic of inquiry as they intersect with two established literatures in police studies. First, there is consensus that negative outcomes are shaped by citizens’ demeanor (Worden and Shepard, 1996). Citizens who disrespect the police are more likely to face punitive sanctions than those who comply with police directives (Mastrofski et al., 2002). Second, traffic encounters represent one of the most pervasive ways that “the law” interjects into the lives of people (Davis et al., 2018), and how the police are perceived to treat citizens has the potential to shape the legitimacy of the police as well as citizens’ trust in them (Madon et al., 2017; Murphy and Tyler, 2017). Although researchers in procedural justice have framed police legitimacy as a correlate of the treatment citizens receive from the police (Sunshine and Tyler, 2003), it could also be that outcomes of police–citizen encounters are shaped by citizens who reject police authority and their definition of the situation (McCluskey et al., 2018). However, even in instances where citizens disrespect the police, officers must persuade citizens to bring the encounter to a satisfactory close.
This article examines the way police and drivers mutually bring traffic encounters to a close. In particular, this article examines how ordinary features of talk-in-interaction are adapted to meet the institutional exigencies of patrol work, and how the interactional order of closing sequences in traffic encounters is collaboratively produced by police and drivers alike. We demonstrate the ways that the attitudes and perceptions of the participants can be empirically validated in the observable details of talk-in-action. By doing so, we critically examine the assumptions of procedural justice research which frames police legitimacy in relation to citizens’ perception and satisfaction. The implications for procedural justice and police behavior are discussed.
The intersection of procedural justice, demeanor, and closings
Citizens want to be treated with respect in their encounters with the police, and this desire shapes the interaction in significant ways. Research on procedural justice indicates that citizens are more likely to obey the law when they have been treated with respect (Hamm et al., 2017). When citizens are treated with respect or interactional justice, they are more likely to internalize the moral compunctions of law and perceive laws as legitimate thereby increasing compliance (Bradford et al., 2015; Jackson et al., 2012). It is in the interest of the police to treat citizens with deference, for their legitimacy increases when they are perceived to be benign in their actions, and citizens feel like they have a voice in the encounter (Nawaz and Tankebe, 2018). Citizens want to be able to articulate their opinions, and for their perspectives to count toward the decisions that the police make. Citizens do not want their opinions to be ignored nor do they want to be interrupted when they talk, or be disrespected by the police (Van Damme, 2017).
If citizens want to be treated with respect and dignity by the police, the police demand it at all costs from the public as a feature of their work (Westley, 1970). That the failure to show proper deference to the police results in punitive sanctions such as arrests, tickets, and physical punishment has been a finding that has been replicated for over five decades, beginning with William Westley (1953). Although legal variables technically determine when the police invoke their authority (Mastrofski et al., 1995; Terrill and Mastrofski, 2002; Tillyer and Engel, 2013), a substantial body of literature demonstrates that citizens shape police behavior through their own demeanor (Engel et al., 2000; Novak et al., 2002). While citizens’ demeanor can be measured in numerous ways, non-compliance, disrespect, and verbal and physical aggression are common ways that negative demeanor is embodied in police–citizen encounters (Engel, 2003; Klinger, 1996).
When citizens disrespect and resist the police, they are more likely to arrest, write citations, and reciprocate with coercive responses that involve warnings and threats (Engel et al., 2000; Novak and Engel, 2005; Novak et al., 2002). In fact, the police are most likely to disrespect citizens when they disrespect them first (Mastrofski et al., 2002). Although warnings in natural discourse are considered to be benevolent and differ from threats only in the intention and attitude of those who issue such speech acts (Fraser, 1998), prior sociolinguistic work on police communication indicates that the illocutionary sensibility of coercion is semiotically accomplished in police work. It has been argued that the presence of uniformed officers who carry guns transforms the nature and effect of their speech acts into perlocutionary ones (Shon, 2005). When citizens do not comply with police directives, they are essentially resisting the police definition of the situation, molding their moral identity as a disreputable one in a way that invokes a punitive response from the police (Schafer and Mastrofski, 2005; Van Maanen, 1978).
Research indicates that citizens’ demeanor during interactions with police is correlated with their characteristics such as race or socioeconomic status (Engel et al., 2012; Reisig et al., 2004). For example, Engel’s (2003) work demonstrates that racial minorities are more likely to demonstrate disrespect, non-compliance, and verbal resistance toward the police. This finding has parallels in police procedural justice literature, for citizens also impute malice and racial motivation to police who disproportionately target them for aggressive police action, which then leads to a decrease in the trust that the public has toward the police (Gau and Brunson, 2015; Tankebe, 2013; Tyler and Jackson, 2014). Research has also found that disrespect toward the police is more prevalent among poor African-American drivers, suggesting a social ecological effect on demeanor (Engel et al., 2012; Novak and Engel, 2005). Or, it could be the case that the higher levels of disrespect toward the police in communities of color may be a legacy of the lack of trust and confidence in the police from the outset, as a function of a strained history of aggressive police actions, which then becomes embodied in disrespect (Baldwin, 1990).
The findings from procedural justice literature indicate that citizens want to articulate their perspectives during their encounters, the voice of citizens an important index of satisfaction (Lowrey-Kinberg and Buker, 2017). Prior findings indicate that black people are suspicious of the police and less likely to impute benign intentions to them due to the high frequency of proactive contacts that occur, as well as the discourteous ways in which they are treated (Renauer and Covelli, 2011; Tyler et al., 2015). Similarly, the demeanor literature shows that citizens who disrespect the police are more likely to be ticketed and arrested than those who treat the police with deference (Mastrofski et al., 2002; Savigar-Shaw et al., 2021). While much of the procedural justice literature focuses on citizens’ perceptions based on what the police are perceived to have done and does not delve into citizens’ actual behavior that shapes the interaction (Waddington et al., 2015), recent works suggest that citizens’ disrespect toward the police results in a decline in the exercise of procedural justice by the police themselves (McCluskey et al., 2018; Nix et al., 2017).
While closing sequences have been the object of inquiry in the sociology of language (Schegloff and Sacks, 1973), especially in the context of medical encounters (Allwood et al., 2017) and emergency call-centers (Riou et al., 2017; Zimmerman, 1992), they intersect with the literature on procedural justice and demeanor because all interactions have to be brought to a close. Furthermore, as evident from prior work, institutional representatives, such as the police, must negotiate their encounters on two footings: clients who orient toward social frames while the police orient to institutional ones (Shon et al., 2013). This difference in footing has implications for how police and citizens interpret and behave during their encounters. For example, citizens sometimes treat officers as if they are talking to a friend at a social gathering; this assumption leads to the use of informal tone and language, which the police interpret as a sign of disrespect. Such imputations lead to the police chastisement, which then generate apologies from drivers. This difference in footing leads to multiple turns by both parties that are not relevant to the official business at hand (e.g. interturn expansion) and prolong the encounter, increasing the possible emergence of conflict talk. How mundane institutional interactions such as traffic stops are made meaningful and brought to an orderly close by both parties warrants an examination of how natural interactions might be modified to meet the exigencies of patrol work. This inquiry is all the more pressing given the recent spate of highly publicized cases of minority drivers who have died at the hands of the police during routine traffic encounters. It behooves us to illuminate how such disorder may emerge from the orderliness of mundane interactions.
In addition to surveys, there are other methods that can be used to explore how police and drivers interact with one another (e.g. Harkin, 2015; Snow, 2019; Wells and Savigar, 2019). According to the tenets of conversation analysis, closings are social achievements that have to be mutually ratified by both speakers. Conversations are mutually brought to a close when none of the parties continues her turn or when one of the parties signals her willingness to end the conversation through deployment of pre-closings such as “well” and “ok” (Schegloff and Sacks, 1973). A proper way to close a conversation entails reinvocations of earlier-made arrangements, making plans for future, and an explicit possibility of opening up the closing again with the introduction of a new unmentioned topic (Frank, 1982).
This natural model of a social interaction is difficult to extrapolate to institutional interactions like traffic stops for a number of reasons. Police and citizens do not operate on an equal footing. Drivers typically do not initiate the encounter with the police, and certainly do not end it of their own volition. Such encounters end only when the police say it is over and release the drivers. Traffic stops are initiated and completed by the police in a way that iterates the inequality between the two. The traffic encounters are also asymmetrical by design and structured in a way that conforms to the institutional order of talk rather than social ones. The police determine the turn-taking system, turn order, and turn size as a function of their institutionality. In natural social interactions the preceding components of discourse are democratically managed by the parties themselves. Possessing such sequential power gives the police their form and authority in addition to the coercive power that is ascribed to them. An examination of the closing sequences has the potential to illuminate cases where typical traffic encounters may deviate from this canonic form by providing a baseline reference for understanding the mundane ways that such encounters are interactionally organized and mutually brought to an end by both parties.
Shortcomings in the previous literature
The notion that citizen participation enhances police legitimacy is a bedrock of procedural justice research (Eterno et al., 2017). However, there are several difficulties with the assumptions about discourse that a procedural justice model presupposes. The idea that citizens who vocalize their opinions during police–citizen encounters serves as an iconic representation of democracy in action presupposes a model of communication that is rooted in abstraction rather than empiricism. The ease with which police and citizens exchange discourse is assumed prima facie and ignores the dialogic character of talk and legitimacy: “they [changes in police procedure] involve learning to provide people with opportunities for explanation before decisions are made; explaining how decisions are being made; allowing people mechanism for complaint; and in particular, treating people with courtesy and respect” (Tyler, 2011: 260). Simply put, procedural justice policing assumes that citizens voicing their opinion during encounters is a good thing, and attributes the problem in police–community relations to the consequences of discourteous police behavior. This way of framing the problem ignores the dialogic and intersubjective character of communication (Bottoms and Tankebe, 2012).
Simply declaring that providing people with opportunities to explain themselves before a decision is made ignores the institutional character of interactions and privileges a natural model of discourse. Such an idealistic position is emblematic of the type of discourse where speakers convey their intentions and messages unimpeded by structural constraints that are embedded in talk and the inherent inequality that exists in institutional interactions (Drew and Heritage, 1992). Such idyllic conceptions of discourse are primarily philosophical where speakers are unencumbered by differences in race, class, gender, and turn-taking formats. One might say that such a notion is nostalgic of “the original position” where speakers are unaware of the structural positions of their interlocuters and where conversations unfold in coercion-free contexts (Rawls, 1971). However, by no means is a dialogue between a patrol officer and a motorist indicative of an ideal speech situation in the original position where speakers can freely exchange ideas and views unhindered. It is this philosophical and idealistic model of dialogue that procedural justice research presupposes (Habermas, 1996). However, even in such a hypothetical ideal speech situation and in the original position, speakers would have to negotiate turns, repair misunderstandings, and allocate turn order. In such an ideal speaking situation, would men continue to interrupt women as they routinely do in social life? (Tannen, 1990); would persons of color receive equal opportunities to speak at length without being interrupted? (see Norander, 2017). Simply put, even in coercion-free speaking contexts like the Rawlsian original position or Habermas’ ideal speech situation, dialogue would not exist as a philosophical problem to be solved in abstraction but a practical problem to be resolved by the members themselves, thus relevant for a sociological inquiry.
Voice is usually interpreted as unfettered discourse that speakers are able to articulate, without concern for turn order, turn size, and turn allocation. Those requirements are negotiated endogenously in natural social interactions (Sacks et al., 1974). That impromptu character, as well as the moral ascriptions that are imputed in procedural violations that occur, is the stuff of sociology of everyday life (Sacks, 1984). However, those components are not managed locally in institutional interactions such as a job interview, courtroom, or when speaking to a patrol officer. In fact, applying the norms and conventions of social discourse to institutional settings becomes problematic as participants have trouble discerning the frame of the encounter (Goffman, 1974), which then obfuscates communication.
The limitations of assuming an idealistic model of conversation can be shown empirically in other institutional settings. For instance, linguistic forms matter, for they shape the type of resources that are activated in the context of criminal justice responses. When dispatchers ask callers to describe the emergency, it has been found that the use of the present perfect (e.g. “tell me exactly what’s happened”) increases the likelihood that callers respond with an informative and short report as opposed to the use of the past tense (“what happened?”). Framing the question in syntactically appropriate ways is important because the appearance of a fact-based report decreases as more time elapses from the beginning of the call (Riou et al., 2017). The problem is that open-ended narratives (e.g. “Tell me what happened?”) produce cumbersome responses that delay the next turn response by dispatchers. When citizens talk out of turn or say things that are not “procedurally consequential” (Schegloff, 1992), the attempt at making one’s voice heard actually impedes the official business at hand.
The problematic nature of open-ended interrogatives is evidenced in cases where potential openings to closing sequences pose interactional problems by leading to inapposite responses that prolong the encounter. Hence, applying a normative model of interaction to healthcare-giving environments hinders effective communication. Rather than using open-ended pre-closings and nonspecific and indeterminate terms (e.g. “see you around”), it has been found that concrete arrangement making (e.g. “I’ll see you tomorrow”) is a preferable way to close the encounter in the context of acute healthcare interactions (Allwood et al., 2017).
For the preceding reasons, it is necessary to recontextualize studies of police–citizen encounters, from asking citizens about their perceptions and satisfaction through surveys, to a more sociological one by examining the empirical sequences of talk (e.g. Harkin, 2015). That logically leads to the next question: How do the police close their encounters with drivers who have been pulled over for minor infractions? Do they give citizens free reign to voice their concerns? Do they follow the protocol as described in the sociology of conversation literature by modifying the interaction to its institutional contours? Are warnings and threats coercive speech acts as characterized by previous researchers?
Data and methods
This study uses audiotapes of 50 traffic stops as data. The data represent observations and audio-recordings of five unique police agencies in two different states in the United States (2001–2002). The audiotapes of the traffic stops were made available to one of the authors for secondary analysis. Out of the 50 traffic encounters, the police cited the motorists in 58% (29) of the cases, and released the drivers with a warning in 40% (20); one driver was arrested. This finding closely parallels national trends in the United States. In 2015, 19.2 million drivers were stopped, of which 49% were ticketed as a result of their contact with the police while 36% were released with a warning (see Davis et al., 2018).
All of the recorded traffic stops were transcribed. We have attempted to recapture the conversational details as much as possible, and will include those details when analytically relevant for the sake of clarity and accessibility (see Appendix 1). The traffic stops are identified as “
We adopt a two-fold analytic strategy in this article. First, the encounters were classified into thematically parsimonious categories, paying particular attention to the way encounters are brought to a close. This first step resembles the process of analytic induction described by qualitative researchers (Strauss and Corbin, 2008). Once this thematic category is developed, each category is analyzed for its conversational structures. These structures refer to the generalizable aspects of talk, features that are contextless and illustrate the sociological aspects of talk as a form of social action (Heritage and Atkinson, 1984). While nonverbal communication is important for understanding talk and its background context, it was not possible due to the limitations inherent in the data.
Closings as mutually accomplished social action
Traffic stops are structurally organized activities in that the police typically semiotically summon drivers to interact, initialize the encounter, announce the problematic relevancy, and bring the encounter to an orderly termination of the interaction. The notable point is that giving warnings to drivers, writing tickets, and arresting them involve speech acts that occur in specific sequential environments that parallels the social model of language as well as adaptations that are made due to its institutional context. In the sections below, we describe the three thematically occurring sequential contexts in which closing sequences are contingently shaped by drivers and patrol officers alike as well as their conversational structures.
Requesting signature as making future arrangements
Excerpt 1: CC2P TS #2 34 → PO: [okay] (4.7) sir I’m gonna 35 ask you to sign where it says signature of driver this is 36 not saying that you’re guilty only saying that you’re 37 gonna go to or call the xxx city courthouse on 38 or before march twelveth I’mm sorry april twelveth at which 39 time you can either plead innocent and they can give you a 40 court data if you wanna fight this ticket or you can plead 41 guilty and pay the fine 42 (.9) 43 → D: uh about how much? 44 (.6) 45 PO: I don’t know but I’ll give to you if you call 46 this number during working hours they should 47 be able to tell you how much 48 (.) 49 → D: okay 50 → PO: okay? Please be careful sir
Tickets are not one-way transactions; police officers do not simply hand them to drivers. They require collaborative action by both parties; moreover, one of the parties has to signal his or her readiness to bring the encounter to an end in some way. Across multiple police agencies, the police obtained a driver’s consent through a signature or by eliciting a promise to appear by checking a box on the ticket. This bureaucratic requirement that necessitates the driver’s consent shapes the encounter in notable ways.
The promise to appear at a future date can be considered as the collaborative roots of a closing sequence, and illustrates the ways police communication is similar to social discourse and is situationally modified to meet the institutional relevancies of police work. That is, similar to the way ordinary social interactions are brought to a close with pre-closing markers (e.g. “well,” okay), patrol officers signal the upcoming termination of the encounter with acknowledgment tokens such as “aw right” and “okay”. In line 34, the officer’s “okay” functions as a pre-closing marker to indicate the end to come, for in the very next breath, the officer requests the driver’s signature and then proceeds to explain the processes involved. The person who articulates the markers as pre-closing signals is almost always the police, not drivers. The capacity to utter and issue pre-closing markers thus embodies the linguistic legitimacy and authority of the police as well as the asymmetrical status of the participants, just as the way the bureaucratic request (e.g. may I see your driver’s license?) serves as an iconic index of police power.
As noted, the decision to close the encounter is locally managed by the participants themselves in ordinary social life; that the police initiate the closing sequences indicates the asymmetrical character of the encounter and the difference in power between the two participants. In addition to coercive power, the police also possess sequential power to control the flow of talk—to initialize it and to terminate it. This type of inequality—to open and close interactions—is generally not observed in ordinary social encounters. The mutual exchange of “okays” brings the encounter to an official end in lines 49 and 50. In these two ways, traffic encounters abide by the sociological rules of language use. While the police must abide by the rules of language to communicate their intent and meaning, they also possess a type of power that has been overlooked in the literature.
Although the linguistic markers are used in ways that parallel social discourse to signal closings, the differences also become evident. Notice that going to court at a later date or challenging a ticket represent one way in which future plans are solidified. Hence, when police write citations to drivers, court dates represent a way of closing down an encounter by making plans for the future. However, this future plan is solely shaped by the bureaucratically occasioned character of the event and the ensuing punitive action (see Feeley, 1979). While the possibility of opening up a closing is always present as a feature of social talk, this feature is absent in the traffic stop data. The encounters almost always came to a close when terminal markers (e.g. good bye, have a nice day) were exchanged. Drivers routinely asked about the price of the tickets that interrupted the officer’s turn-in-progress (line 43); these side sequences that expand the turn, however, only advance the action that is projected in the prior turn, and do not introduce a new topic. Hence, we can see the dispreferred ways in which the possibility of opening up closings are foreclosed due to the institutionality of talk. Thus, that the police initiate closings—not citizens—and that drivers are unable to open up a closing are some of the ways that the institutional character of talk is embodied in its sequences. The voices of citizens are structurally constrained as a feature of the structures of police talk.
Police warnings as contingent closings
Statements such as “I’m gonna let it slide this time but slow down” represent typical warnings that police give to drivers (Shon, 2005). Although warnings have been conceptualized as coercive acts in prior police literature (e.g. Novak and Engel, 2005), there is a social dimension to them that is closely related to the driver’s demeanor. As a non-punitive outcome that circumvents the monetary cost of a ticket, warnings that are given to drivers are desirable outcomes, and represent another prominent way that traffic encounters can be brought to a close. These warnings that patrol officers give to drivers occur toward the end of the encounter and constitute one way that traffic stops end. Police sometimes give warnings to drivers because they have personal connections to them or some institutional affiliation.
1
For all other citizens who do not have clout, the warnings police give should be viewed as contingent accomplishments that lead to a particular type of a closing sequence. This desired outcome, however, emerges in particular sequential environments, and involve impression management by both parties. Consider the following data: Excerpt 2: CC4P TS #8 19 PO: ten four 20 (2.8) 21 driver step to the rear of the vehicle 22 (12.1) 23 you got your regis you ran a stop sign right there 24 (.3) 25 → D: ↑I never saw it I’m sorry↓ I’m[ sorry 26 PO: [you got your 27 registration and proof of in[surance? 28 → D: [↑yes
There are several notable features of talk in excerpt 2. First, it shows the initial moments of the traffic stop as the officer attempts to open the encounter with a request for documents, only to self-correct and announce the problematic relevancy (line 23). Another way that police negotiate this segment in traffic stops is to formulate it as a yes/no interrogative (“Do you know why I stopped you?”). Second, notice the driver’s response to the announcement of the problematic relevancy in line 25: she proffers an excuse, thus exceeding the conditional requirements of the preceding turn by providing an account where none is warranted. Then she proffers an apology—twice. The driver compresses three potential turns into one snug utterance, in almost one breath, prosodically marking her willingness to admit her mistake, confess her crime, and apologize for her infraction. Third, when the officer is able to complete his bureaucratic request, the driver’s turn overlaps with his turn-in-progress to fulfill the semantic and grammatical requirements projected in the prior turn (line 28).
How should such conversational exchanges be interpreted and understood in the context of traffic encounters? What are the practical implications of proffering an unsolicited account that is compounded with an apology and alacrity with which the interrogatives are answered? Across four turns and two exchanges in the opening segments of a traffic stop, the driver manages to convey to the officer her moral character as a penitent and corrigible person. Moreover, she shows herself to be a deferential person, linguistically signaling her cooperative attitude by answering the officer’s question even before he completes asking it. Thus, the decision to apply or not apply the law is shaped early on, through the citizen’s demeanor in the opening turn.
These types of questions that police ask are mundane and simple, but they perform the task of discerning the status of drivers as a “governable” or a “rebel” (Muir, 1977). Police interpret demeanor as a test to evaluate citizens’ future adherence to law and the perceived efficacy of informal requests and warnings in securing compliance (McCluskey et al., 1999). Citizens who fail the attitude test are perceived as higher risk of future offending, a potential threat, and more deserving of punishment (Worden et al., 1996). Conformity to police interrogatives is one way in which respect toward the police is performed on a turn by turn basis, and this cooperative demeanor shapes the appearance of verbal warnings that police proffer to drivers rather than tickets. This is one way that demeanor of drivers and police are accomplishments, for they have to be collaboratively sustained across multiple turns early in the interaction. Warnings as a particular type of closing outcome are contingently accomplished by drivers and police alike.
Citizens’ request for warnings as attempted pre-closings
If police give warnings to drivers as a function of their cooperative demeanor, citizens also make proactive requests for warnings when they are informed that they are about to be ticketed. This act functions as an attempt at preemptive pre-closing by sending a signal to terminate the encounter in a non-punitive way through a politely formulated request. Consider the police response to such requests for warning: Excerpt 3: CC2P TS #9 22 D: I was just out (to be honest with you) 23 I was just headin home 24 (1.0) 25 (won’t) you please just give me a [warning? 26 → PO: [>um unfortunately I’m 27 → in the traffic section when I make a stop I have to write 28 → a ticket< >and we’ve got some complaints from the neighbors 29 → on this street so<= 30 D: =I gotcha
That the patrol officer has heard the request as a request rather than other speech acts can be verified in the following way. First, the officer’s turn (line 26) begins with a hesitation marker (um) and a disjunctive adverbial (unfortunately) that signals the dispreferred character of the emerging turn-in-progress and the rejection to follow in the clauses after. He then explicitly formulates an account (lines 27–29). In fact, the officer provides three separate excuses as to why he cannot let the driver go with a warning. The way the rejection to a request is structured is consistent with the social ways that invitations and requests are declined: delayed by hesitation markers, extended within turns, and avoidance of outright declinations (Pomerantz, 1984).
That the police also formulate their own accounts in the context of a closing sequence is significant for several reasons. Drivers have been caricatured for the creative excuses they come up with as a way of generating sympathy to avoid a ticket (Eagan, 1990; Kelley, 1989). Just as significantly, even the police who are applying punitive sanctions mitigate the effect of their coercive action through strategies that parallel the softening techniques of bad news delivery found in social discourse. Notice that the officer in excerpt 3 formulates his account using a deontic modal verb to indicate his obligation to write the ticket (“have to”); moreover, he formulates his punitive action using the institutional “we” and deflects blame to the community, who has complained about speeding through their neighborhood. Through softening the delivery of bad news, presenting the offense as an injury to the community, and convincing the driver they have avoided some possible penalty, the police attempt to “sell” the disposition to the citizen and avoid possible complaints or contests in court (Schafer and Mastrofski, 2005: 11). As such, accounts recalibrate the application of tickets in ways that are socially and culturally acceptable, in ways that neutralize personal responsibility, essentially, contesting the foundational basis of moral judgment (Scott and Lyman, 1968). An explicit rejection without attendant account holds the police morally accountable as a particular type of person. It is this unflattering social identity that the police attempt to avoid through this turn-within-a-turn expansion. The behavior of drivers and patrol officers are similar in this way in that they both attempt to avoid pejorative moral characterizations through the performance of accounts (Coates and Wade, 2004).
The police excuse (lines 26–29) thus serves to “reject” the attempted pre-closing signal to release the driver with a warning, and continues the path of closing sequences that the police delineate. The ability to frame discourse in this way, to control the footing of the encounter on institutional or social terms, illustrates the power of the police to control the sequential flow of talk. This discursive power the police possess is implicative as it shapes the legal and moral identity of drivers in ways that they define. Moreover, the ability of the police to switch from social footing to an institutional one illustrates their ability to traverse two linguistic systems in fluid ways, as second language speakers are wont to do (Schumann, 1995).
Discussion and conclusion
This article has examined the closing sequences of traffic stops. We have attempted to show that traffic stops are brought to an orderly close by both parties, deploying and modifying normative sociological properties of conversation to meet the exigencies of patrol work. Patrol officers signal the upcoming termination of the encounter with tokens such as “aw right” and “okay.” Drivers have to acknowledge and accept these pre-closing signals in order for the encounter to close; that is what makes closing sequences mutual accomplishments. Drivers may be overcome by feelings of injustice upon the receipt of a ticket but they reify and reproduce the sociality of everyday life and its normative encumbrances through an acknowledgment of the pre-closing signal. It is in this almost reflexive ratification of the preceding turn that the sociality of everyday life and the interactional order of police–citizen encounters are reproduced; moreover, the examination of the direct sequences of talk during a traffic stop illustrates the difference between an empirical study of social action and a perception-based study of police behavior. Our findings reveal that despite the pangs of punitive sanctions that drivers may experience at the receipt of a ticket, they conform to the sociological rules of conversation and social interaction. Rule compliance during traffic encounters, at least from a sociological perspective, is reflexive, as supported by the preference organization of adjacency pairs found in the closing sequences, and not solely related to feelings of satisfaction or perceptions of police legitimacy. The adherence to conversational rules by both parties suggests that conformity to sociological norms is just as meaningful as obedience to the law.
One of the criticisms of the procedural justice literature that we have made is that it frames citizens’ participation as a good thing, often without considering 2 content, turn design, or turn order, framed primarily in relation to citizens’ perceptions. Moreover, procedural justice literature mostly frames the unfavorable experiences of citizens as a function of discourteous police behavior and overlooks their own behavior that may contribute to police actions. The analysis we have employed attempts to move beyond the post hoc character of procedural justice police research (see Harkin, 2015). That the warnings police give to drivers can be viewed as contingent accomplishments, as a prelude to a particular type of a closing sequence, shows another way that our sociological approach to the study of police–citizen encounters differs from the extant work of procedural justice research. Our work shows that measures of citizen satisfaction may overlook the nuanced ways that the interaction is shaped by the actions of both parties. In particular, we have shown that drivers exercise tremendous agency, inverting the interactional order of police–citizen encounters, at least for a turn, through the request for a warning. Through such moves, citizens attempt to wrest control of the encounter by pre-emptively moving toward closing sequences on their own. This move, as described here and elsewhere, is a highly disrespectful one, one that challenges the authority of the police. Hence, the police should not be the only ones held accountable for disrespect toward citizens. Citizens also shape their own outcomes through their behavior. Drivers already exercise sequential voice in their encounters to force traffic police to enact the role of excuse-givers, constructing their accounts in unsolicited ways, necessitated by communal need, crafting their moral identity in positive ways through impression management. As shown here, drivers can use their voice to effectuate a positive impression management as well, which then leads to lenient dispositions. That is what makes police–citizen encounters collaborative acts.
Our findings militate against the tenets of procedural justice literature in another way. The model of speech presupposed by procedural justice researchers, we have argued, is philosophical rather than empirical. As we have shown here, this presupposition is not tenable, for our work suggests that the encounters that occur between the police and the public illustrate the reflexivity and intersubjectivity of traffic encounters as sites of complex social—sociolinguistic, empirical— interactions. Our work shows that there are numerous speech acts that occur in the brief interactions between patrol officers and drivers, in ways that reify and challenge the findings from previous studies. Rather than assuming the truth condition of the perceptions reported in surveys, we have argued that police–citizen encounters should be understood and interpreted in their structural and sequential environments in empirical ways, using actual verbal data rather than the impressions of trained coders. This way of viewing outcomes of encounters as social action that is mutually shaped and collaboratively accomplished may provide an alternative way of understanding the encounters between the police and the public.
For example, prior scholars have shown that the police are more likely to arrest, write tickets, and reciprocate with coercive responses that involve warnings and threats with citizens who are uncooperative (Mastrofski et al., 2002). Although warnings and threats have been operationalized as hostile acts (Fraser, 1998), the empirical details suggest otherwise. Police warnings, although theoretically coercive, occur in sequential environments that shape the trajectory of talk and the moral identity of drivers in prosocial ways. Police give warnings to drivers after they have demonstrated a version of a self that is respectful and deferential. The responses of drivers were unequivocal in the turn after the warning, for they expressed their gratitude for the leniency shown. Thus, warnings the police deliver are benign in intention and consequence, and these can be empirically verified in situ.
We found that drivers also make requests for warnings to police after they have been informed that they are about to receive a ticket. In those environments, the requests for warnings function as preemptive moves toward contingent closings, which then leads to the performance of two selves by the police: an institutional actor who usurps the extant social conventions to deflect blame and responsibility away from the self (Scott and Lyman, 1968), and one who flouts conversational norms and signals the enforcer-like attitudes through syntax (Muir, 1977). We might even hypothesize that police officers who flout conversational rules by withholding mitigation devices might be harbingers of complaints to come, for the absence of such acknowledgment tokens and delay markers are morally accountable. Although not shown here, when patrol officers reject citizens’ requests for warnings outright, without softening the rejection, drivers emitted responses that indicated the intensity and escalation of their adverse experiences. How the police formulate their questions and respond to questions from citizens thus affects the experiences of drivers in ways that can be empirically verified in vivo rather than in post hoc ways. Moreover, just as citizens are held accountable for their verbal behaviors, the police can also be held accountable for their verbal behaviors, for our work has shown that police and drivers alike explain, make requests and warnings, deliver warnings, and perform excuses in the performance of their legal and moral identities. They do so in ways that go beyond static responses to a survey.
Although it is tempting to conceptualize the attitudes of citizens and police by asking about their perceptions, such approaches overlook the empirical and interlocked character of speech acts that occurs in the context of patrol work. Citizens respond to police directives that orient to its social contours, and that ought not to be surprising, for they are immersed in it and orient to rules normative to that setting. The police, however, are able to switch from an institutional code to a social one, engaging in a code switching of sorts. This flexibility gives police room to maneuver in order to set the agenda of talk and to control its sequential flow. The police are able to articulate their turns so as to design the preference organization toward a particular orientation while citizens primarily orient to the social frames of utterances. Although the procedural justice literature has framed the legitimacy of the police as being contingent upon how the police treat members of the public, our empirical work shows that police–citizen encounters are collaborative outcomes that require the give and take of both parties. The police are able to open and close the encounters, but drivers also possess the capacity to mold the shape of the encounters through their responses at every structurally provided turn. Closings are accomplishments that require drivers to conform to the interactional order of police work in order to effectuate a favorable impression management that leads to lenient outcomes or affirm the pre-closing signals emitted by the police. Citizens also invert the moral order by requesting a warning, forcing the police to formulate unsolicited accounts and excuses. Citizens participate in ways that go beyond mere satisfaction and trust, and shape the encounter in interactionally significant ways.
The fact that police are able to explain the necessity of signing tickets to drivers and request their presence at a later court date as an important step in the closing sequence is one way in which both parties mutually perform their roles to bring the encounter to an orderly end. Moreover, we have argued that citizens are not able to vocalize their opinions when they want to, for the structures of talk in police communication and the inherent asymmetry in power unfold along the contours the police dictate. Citizens who deviate from this interactional order of police–citizen encounters are held accountable in several ways. In this way, the ability of the police to define the situation is expanded to include their ability to set the sequential environment of talk. If we may conjecture here, community leaders who tell citizens to vocalize their opinion irrespective of turn taking, turn order, and turn allocation in their interactions with the police is not sound advice, for it is likely to prolong the encounter that entangle both participants in arguments. One way that participants foreclose an argument is by leaving; citizens who attempt to shut down an argument by driving away has the potential to lead to fatal consequences.
Despite our findings, there are several limitations to our work. While we have provided an analysis of police–citizen encounters that came to a close without deaths of officers or drivers, recent highly publicized incidents indicate that certain traffic encounters may lead to tragic outcomes. We have not selected or examined these types of tragic encounters as units of analysis. A future project could collect and analyze mundane traffic encounters that end with police shootings to examine the verbal interactions that precede the police use of deadly force. Although we have selected and analyzed closing sequences, it just may be that openings and closings are closely linked, the former shaping the latter. It is this close connection between the two that may illuminate incidents where closings fail in a spectacular fashion. Finally, we have examined police–citizen encounters from a conversational analytic perspective, focusing our attention on the structures of talk and turn organization. For future works, it may be worthwhile to examine how race of drivers and gender of the police may affect the organization of talk in procedurally consequential ways.
Footnotes
Appendix 1
Acknowledgements
The authors are grateful to both reviewers for their insightful comments and suggestions. Any flaws that remain in the article, however, are their own.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
