How Talk Works: Studying Negotiation Interaction

Negotiation depends on communication. Whatever else goes on during a negotiation, parties attempt to manage their differences and reach agreements through exchanges of messages that make up sequences of moves and countermoves. Complementing language use, negotiation interaction is unavoidably situated within physical and social environments that can function as resources for negotiators: location (institutional, architectural), embodiment (posture, gesture, laughter, eye gaze), modes of communication (documents, symbol systems, telephones, e‐mails), and social relationships. Furthermore, even the “mental” elements of negotiation (goals, planning and strategizing, emotional reactions, evaluating outcomes, etc.) are communicatively constituted, made public, and mutually understood in and through interaction. More than simply representing and conveying information, communication is the means by which social actors create meanings, outcomes, identities, and relationships.

Some recently published research on doctor–patient interaction demonstrates convincingly how communication shapes outcomes. When patients come into a medical interview with multiple concerns, doctors who ask “Is there something else you want to address in the visit today?” are significantly more likely to elicit additional concerns than if they ask “Is there anything else you want to address in the visit today?” (Heritage et al. 2007), a difference of just one word. Open‐ended queries such as “What can I do for you today?” generate significantly longer and more detailed problem presentations from patients, compared with more closed initial turns such as, “So you're sick today, huh?” (Heritage and Robinson 2006). Patients who receive these open‐ended initial queries report significantly higher satisfaction with the encounter and better physician listening (Robinson and Heritage 2006). An earlier study found that if doctors make “online commentary” while they examine patients (e.g., “Well, I see a bit of redness in this ear”), they are much less likely to encounter subsequent patient resistance if they subsequently produce a “no problem” diagnosis (Heritage and Stivers 1999). These findings demonstrate convincingly how powerfully particular word choices and communicative actions can shape trajectories of unfolding talk.

Like doctor–patient interviews, police interrogations, and many other interactions, negotiations and mediations succeed or fail based in large part on how participants manage their talk. Negotiation training materials are rich with suggestions for managing such talk, but prescriptions tend to skew either toward global strategic considerations or toward isolated individual behavior. Often overlooked are the actual interchanges through which people set agendas, propose, offer, agree and disagree, and problem solve.

Much negotiation prescription relies on real‐time observation, role‐play, and written accounts of successful or problematic practices, but observation is subject to temporal and perceptual limitations. Recall is subject to the vagaries of memory. Role‐plays, while valuable, leave open the question of how “real life” might differ. Much that is crucial happens in negotiation quickly, in real time. Sometimes, these are critical moments, turning points. Other times, they simply represent the ebb and flow of dialogue, the moment‐to‐moment activities that shape meaning, realities, and outcomes.

So we believe in the value of looking closely at actual talk, studying how the participants in a negotiation session come together and do what they do. Doing so helps us understand how negotiation works, how to listen better, how to take action more effectively, and how to teach and train about the art and practice of negotiation. If skeptics have any doubt that such close study can be turned into practice, let us add one crucial feature to our summary of the doctor–patient dialogue study (Heritage et al. 2007) reported previously. Some physicians were trained to modify their preclosing questions to patients, and the striking results showed the positiveeffects of such training. People can learn to shape their talk‐in‐interaction to achieve greater success.

What do we know about the linguistic and embodied actions, sequences, and events through which people negotiate face to face or over the telephone? How might negotiation and dispute resolution theory, research, and practice benefit from more focused, substantive conceptualization of the interactions through which participants engage in agenda setting or preparation, value creation and distribution, and follow‐up? These and related issues are the subject of this special section of Negotiation Journal.

The impetus for this special issue began with a December 2008 symposium on negotiation and communication that was sponsored by the Program on Negotiation at Harvard Law School, Emerson College's School of Communication, and MIT's Department of Urban Studies and Planning. We began with a desire to bring communication theory and research into the mainstream of negotiation studies, but we recognized that the term communication covers a wide array of phenomena, theoretical perspectives, and research methods. Working across these dilutes our focus and risks diverting attention to paradigm disputes that, however, interesting, do not help us study negotiation. We decided therefore to focus on interaction, thereby narrowing the range of issues under consideration and bringing in expertise from multiple disciplines, including communication, sociology, and management. The term interaction tends to convey a set of approaches bearing a family resemblance that reflects the “linguistic turn” (Rorty 1967) that philosophy and social sciences took in the mid‐twentieth century, away from viewing language as a transparent medium of thought and toward action‐based theories of meaning. They appear under such names as language and social interaction, discourse analysis, conversation analysis, microethnography, and pragmatics (see Pomerantz and Fehr 1997; Grant et al. 2004; Fitch and Sanders 2005; Putnam 2005; LeBaron 2008). We present a brief summary of two prominent research traditions below.

Conversation analysis (CA) arose from ethnomethodology, which was developed by Harold Garfinkel (1967) and had roots in the social structural theories of Talcott Parsons and the phenomenological project of Alfred Schutz and others interested in challenging representational views of reality and instead attempting to describe peoples' lived experience of phenomena. Garfinkel treated social actors not as “judgmental dopes” but as active sense makers and creators of social reality. Ethnomethodology is the study of people's methods for leading social lives. CA was also deeply influenced by the microsociological studies of Erving Goffman, with his use of theatrical metaphors and vocabulary to describe how people enact behaviors, roles, scenes, and relations. CA began with the pioneering studies of Harvey Sacks and Emanuel Schegloff. Sacks's lectures (Sacks 1992), circulated like samizdat for years, and, finally published in 1992, lay out a systematic range of phenomena for detailed study. Since then, CA studies across a wide range of disciplines have yielded a wealth of findings about how people create social life, one turn at a time (see Atkinson and Heritage, 1984; Psathas, 1995; Schegloff, 2007).

Conversation analysts work with recordings of naturalistic interactions, creating detailed transcripts (relying on a system developed by Gail Jefferson; see Appendix A to the Special Section) highlighting words, selected nonverbal features, and timing. Analysis involves characterizing sequences, turns, actions, and overall organization. The analysis attempts to account for “Why that now?”— that is, how does this action, at this moment, create social structure and display participant orientations?

The analytic process eschews psychological inference in favor of describing what is actually occurring, working inductively through collections of like instances or extended explication of case studies. Researchers are cautious about generalizing, although they expect that a practice of talk shown in one moment will be likely to appear elsewhere. They use statistics and quantification only minimally, preferring to analyze the formal procedures social actors use in particular situations. In this sense, CA is an empirical enterprise, producing replicable findings that can be applied and extended.

Some examples of the phenomena of conversation that researchers using CA study include “adjacency pairs,” such as summons–answer, greetings, question–answer, offer–reply, compliment–response; repair; storytelling; arguments; laughter; preference structures; and turn taking. CA methods have been applied to everyday conversations, medical interviews, courtroom trials, classrooms, focus groups, broadcast interviews, talk shows, speeches, and employment interviews.

The term discourse analysis (DA) covers a wide range of theoretical and methodological perspectives (in fact, some would subsume CA as a form of DA) broadly concerned with how meaning is created and conveyed in discrete linguistic units. Discourse sometimes is treated as an all‐encompassing term for the study of language‐in‐use and is sometimes applied more specifically to the cultural or interpretive frames contained in messages (this “capital D” discourse is reflected in the work of Michel Foucault, Norman Fairclough, and others).

One prominent strain of DA is rooted in Speech Act Theory, built on the ordinary language philosophy of Ludwig Wittgenstein and the lectures of J. L. Austin and developed in particular by John Searle (see review in Levinson 1983). DA takes the individual utterance as the unit of analysis and seeks to specify the conditions under which utterances perform certain actions such as “assertives, directives, commissives, declaratives, expressives, and accreditives” (Fairhurst and Cooren 2004: 136). It owes intellectual debts more to linguistics and philosophy than to sociology. DA represents a prominent strain in organizational studies (Grant et al. 2004). For some scholars, such as Karen Tracy (2003), DA allows greater room for the interpretation and consideration of strategy and thus may be particularly useful in negotiation analysis.

CA and DA share some central commitments. Both focus on how people accomplish meaning and use language in social contexts and tend to treat meaning as action‐based, systematic, and coconstructed. Both take seriously the participants' point of view, explicating people's sense‐making and sense‐producing practices. Both involve an empirical science of interaction, with insistence on carefully describing what is actually going on rather than speculating about off‐stage phenomena such as cognition or forwarding ideological arguments. They privilege description and induction over experimentation and hypothesis testing. Both tend to emphasize microlevel phenomena. Finally, both insist on studying naturalistic interaction in situ, as much as possible.

Although the research articles in this field by and large reflect these commitments, they also display some differences on such issues as:

• the extent to which meaning is located in the individual speech act or in interactive sequences;

• the extent to which phenomena assumed to be within social actors' consciousness and volition are given emphasis;

• their willingness to make claims about strategy and motive, or to move to criticism;

• whether to employ existing lay uses of terms (DA) or to respecify them into a more technical vocabulary of interaction (CA); and

• the microlevel of detail at which analysis is conducted.

The articles that follow showcase conversation analytic and discourse analytic approaches to the study of negotiation. The first two work with the same data, an audiotaped recording of a telephone negotiation between two real estate agents. In his article, Doug Maynard analyzes the structure of negotiation talk, focusing on the core bargaining sequences that lie at the heart of negotiations. He uses the terms defer, demur, and deter to capture some of the systematic ways that negotiators respond to proposals and the further unfolding courses of action. Although these patterns are standard (perhaps even typical), they are by no means predetermined: participants routinely deviate from canonical forms to mark ongoing meaning, action, identity, and relationship. In the second article, Linda Putnam reviews discourse analytic research on how language use in negotiation informs strategy, relational development, identity management, emotional expression, issue development, and framing. In her original analysis, she demonstrates how negotiators jointly constitute risk, certainty, and loss–gain in how they frame the discussion and develop the issues.

We follow these articles with a close look at videotaped excerpts from an actual small claims court mediation session. The analysis points to a central concern for mediators: how to deal with the tension between demonstrating a commitment to take each party seriously while scrupulously avoiding the appearance of taking sides. Phillip Glenn's article examines several moments when this tension arises, how each moment gets managed, and how these moments mark turning points in the mediation. In addition, working with videotaped interaction invites us to attend to what is going on visually and nonverbally, providing a corrective to the temptation to make analysis unduly logocentric. When people negotiate face to face, they put their bodies as well as their uttered words into play, and phenomena of gaze, posture, and movement, as well as use of artifacts and documents, contribute closely to unfolding courses of action. The organization of talk intersects with the organization of embodied action.

Following this analysis, we close with a brief discussion by Larry Susskind from the perspective of a mediation theorist and practitioner. He raises issues that merit consideration for readers wanting to learn more about the powerful role interaction plays in negotiation processes and outcomes.