This is part of the “virtual book club” discussing readings for the symposium at the University of Missouri on October 7: Moving Negotiation Theory from the Tower of Babel Toward a World of Mutual Understanding.
My article, A Framework for Advancing Negotiation Theory: Implications from a Study of How Lawyers Reach Agreement in Pretrial Litigation, prompted a lot of reactions from David Matz, which led to the following conversation. We’ll start with an abstract of my article.
The prevailing negotiation theory tries to fit lots of square pegs into just two round holes – adversarial or cooperative bargaining. In the real world, negotiation comes in many different shapes, not just circles and squares. Analyzing law school textbook definitions of the traditional models, this article demonstrates that the two “round holes” in current negotiation theory are poorly defined.
This article also presents empirical accounts of actual pretrial negotiations to demonstrate that the theoretical models do not fit some real-life negotiations. It argues that it is time to replace the traditional models with a flexible framework that can accommodate virtually all legal negotiations and it uses cases from this study to illustrate the proposed framework. Instead of focusing only on bundles of characteristics for each theoretical model that are assumed to be highly correlated with each other, the framework unbundles the variables, which permits more accurate description of negotiations.
The variables in the framework are: (1) the degree of concern, if any, negotiators have for the other side, (2) the communication process used in trying to reach agreement, (3) the extent that negotiators create value in the negotiation, (4) the negotiators’ tone, (5) the use of power, and (6) the source of norms that negotiators use. These variables are likely to be causal factors affecting particular negotiation goals such as efficiency and satisfaction of parties’ interests. The conclusion discusses implications and recommendations for academics, practitioners, and instructors.
David: I just read your Cardozo piece on negotiation theory. I cheered. We have much to discuss. Let me rattle through some reactions, leaving coherence for another day.
First, and most important, you start where we should (and by and large don’t), by looking at real negotiations. Question: what are we looking for when we do that?
- Are we looking to see if they use “our” tools?
- Are we looking to see what errors they made, or how they could do it better?
- Are we looking to see what do to be successful?
- Are we looking to see why they fail if they do?
- You seemed to be looking to see if your data fit with our models, and where it didn’t, offering a new (or augmented) model which will describe the rest. Why is that the way to go?
John: Like an astronomer seeking to find evidence of a tenth planet in our solar system, I actually set out in this research project to document cases of what I called “ordinary legal negotiation” in an earlier article.
Unlike the familiar positional and interest-based models, where negotiators are oriented to maximizing their outcomes or finding solutions meeting both parties’ interests, in ordinary legal negotiation, the lawyers begin their analysis based on likely results in negotiation or trial. Of course, both sides typically shade their positions in self-serving ways, but the focus is on normal outcomes rather than exchange of extreme offers or discussion of interests.
In the legal context, it makes sense to refer to this as ordinary legal negotiation. In other contexts, negotiators use other norms such as going rates, standard terms, etc. So generically, this might be called “norm-based negotiation” or something like that.
I imagined that all the cases I studied would fit into one of these three models but found that most did not. The failure of this two- (or three-) model system to fit my data is what prompted me to try to develop another framework.
I interviewed lawyers about the cases they recently settled, starting from their first interaction with their clients in the case. I did find actual cases of ordinary legal negotiation but, more importantly, I documented cases showing how the system of theoretical negotiation models doesn’t work very well.
David: Second, some more general issues:
We should prescribe if we think we can improve on what ordinary negotiators really do. Certainly that is what Getting to Yes (GTY) aspires to, though it doesn’t use real negotiations as its base. Do your categories do that?
We should describe if we can then see in the facts patterns that suggest what can be improved, or patterns that suggest boundaries/limits built into the negotiating process. This is where I start.
John: In developing the dimensions of my framework, I started with the elements that were collectively ascribed to the two general models in my survey of law school negotiation texts. Then I disaggregated these elements into separate dimensions that could vary from each other. In other words, instead of assuming that negotiations could be characterized as coherent models, my data suggested that the variables would not all vary together.
For example, many of the descriptions of positional negotiation suggested that it would involve hostile interactions but I found that in many negotiations involving a traditional counteroffer process, the interactions were friendly and professional. In developing the labels for the different dimensions, I tried to use words that I thought people generally understand the same way.
This is a small dataset and I wrote that this was just a step toward developing a good framework. Indeed, my purpose in organizing this symposium was to use this and other work to advance this effort.
David: Third, you use lawyers account of their negotiations. I am not thrilled. These accounts suffer from biases due to faulty memory, self-serving interests, personal categories, etc.
I’ve mentioned before that I am eager to learn more about real negotiations as described by more detached authors, e.g., Thirteen Days in September by Lawrence Wright, describing the 1978 Camp David negotiation involving Carter, Begin, and Sadat.
Do you – or any readers of this blog – know of other serious, lengthy, descriptions of real negotiations?
John: Every research method has limits – none is perfect. I am in the middle of reading Wright’s book, which is fascinating. Each chapter describes one of the thirteen days of the negotiation – and also includes helpful background about the individuals, events, and context.
Of course, Wright relied on various sources, including interviews with some of the participants in the process. But these sources are subject to the same biases you mentioned. Indeed, even if Wright was a “fly on the wall” observing all the interactions in the negotiation (and had a swarm of “flies” observing all the private interactions of the principals and their advisors) he would have his own biases. Undoubtedly, the accounts would vary to some extent and the process of reconciling differences inevitably introduces errors (or at least some doubt). This is a single, very distinctive, case which produces some interesting insights, but they are hard to generalize to other situations.
Social scientists know that they cannot research an entire universe in a single study, especially recognizing that phenomena may vary in different contexts. They try to develop “robust” results by conducting multiple studies in different contexts, noting similarities and differences across contexts. They also try to “triangulate,” using a variety of data sources and research methods to develop more comprehensive and realistic understandings.
Lawyers were an especially good source for my research for several reasons. Lawyers regularly negotiate and it is useful to learn from experienced negotiators. I was particularly interested in negotiations in pretrial litigation because when analyzing negotiation of litigated cases, people usually focus only on the final stage of the process and I knew that the antecedents were likely to be important. So, as a law professor, I wanted to produce research to correct what I believe are misconceptions in the theory and teaching of legal negotiation. I also used this same data to write a companion article for legal practitioners and students, Good Pretrial Lawyering: Planning to Get to Yes Sooner, Cheaper, and Better.
I think it is also valuable to study actual negotiations by other repeat-player negotiators such as insurance claims adjusters, realtors, merchants, public officials, and even supreme court justices among others. There may also be some value in studying negotiations involving one-shotters, though this can be hard to arrange for a research project.
My data has its limits as you suggest and as I noted in the article. It is a small dataset relying on retrospective accounts by a single individual in each case. It is not as rich an account of cases as in Wright’s book.
But it provides data on a larger number of more typical cases. Importantly, it provides accounts of the entire “life” of legal negotiations beginning with the first interaction between lawyers and clients (though obviously the conflicts began before then).
Ideally in my research, I would have selected a set of disputes from their inception and followed them throughout their “lives.” I would have observed the interactions all the individuals involved, analyzed all the documents and other evidence, and repeatedly interviewed all the individuals. Obviously, this wouldn’t be practical. Moreover, my interpretations would still be subject to my biases and my interactions with the subjects presumably would affect the negotiation process. So researchers must make compromises, recognizing that no study can be definitive.
I assume that there are lengthy case studies of other diplomatic and public policy negotiations. They may or may not include some explicit theorizing based on these negotiations.
There are a number of studies of actual negotiations that do not provide the extended accounts you seek but nonetheless provide valuable insights.
High on my list is a delightful classic study by Stewart Macaulay, Non-Contractual Relations in Business: A Preliminary Study, 28 American Sociological Review 55 (1963). Based on interviews with businessmen and lawyers, he found that businessmen – presumably they all were men at that time – generally preferred to negotiate by avoiding legally-enforceable agreements and resolve disputes by staying away from lawyers as much as possible.
H. Laurence Ross published a classic study of how insurance claims adjusters do their work in his 1970 book, Settled Out of Court: The Social Process of Insurance Claims Adjustments. I imagine that there may be other such studies in the sociology of law literature.
Tamara Relis did an impressive study of mediations of actual medical malpractice cases in which she observed mediations and conducted semi-structured interviews of parties, lawyers, and mediators in the same case soon after the mediation sessions. Her analysis focused on themes in the mediation, particular differences in perspective between the professionals and the parties and gender dynamics. She published the findings in several articles and a book, Perceptions in Litigation and Mediation: Lawyers, Defendants, Plaintiffs, and Gendered Parties.
Andrea Schneider’s article, Shattering Negotiation Myths: Empirical Evidence on the Effectiveness of Negotiation Style, is based on surveys of lawyers using general characterizations of their counterparts in actual negotiations.
Readers: Can you suggest any other studies of actual negotiations (of any length and in any context)? Don’t be shy about mentioning any studies of your own.
David: Fourth, you have cleverly brought in an additional set of categories to enrich what we can see and to reflect what we think they are doing, and you have tried to link your new categories to the ones we conventionally use.
John: Thanks for your kind words, David. I tried to develop categories that seemed significant, related to the existing models, and were readily understandable.
David: Fifth, you use the behaviors lawyers describe to infer motivation. How safe is that? One can with little difficulty infer two or more motives from each behavior. We always want to know motive; in my judgment, we rarely if ever can, in others or even in ourselves. So we fool ourselves into believing we know something when we can’t.
John: It depends on what one means by “knowing.” No one can know others’ motivations with certainty – or perhaps even truly know one’s own motives. Yet I think it should be indisputable that motivations (and other cognitions) are important elements of negotiation. So I think it would be a serious oversight to ignore them. Researchers and readers should not take people’s accounts at face value but rather should consider alternative explanations and use good judgment in interpreting data.
David: Sixth, Wright’s account seems not to use any of the categories we profess to use. Is that because Wright didn’t have those categories in his head? Or because the parties didn’t have those categories in their heads? Is a long, international conflict with huge complex history a different animal from a litigotiation (my spellchecker just dropped dead), so using negotiation-field terms just doesn’t fit?
John: Andrea famously titled an article “Labels Suck” before she thought better of it and renamed it “Teaching a New Negotiation Skills Paradigm.” She describes why our theoretical concepts are useful but also problematic. She ultimately concluded that we do need such concepts – just better ones. I agree with her. Without theoretical concepts and causal theories, every experience is unique and we can’t apply lessons from past situations to future ones.
David: Seventh, you use terms that grow from Fisher and Ury with a little updating here and there. But of course much work has been done that doesn’t fit with GTY, and various authors have suggested that GTY is just off the mark and should be scrapped. But GTY holds the field because there is no coherent alternative in teaching materials, intellectual substance, or Harvard-sponsorship. So we have, as you point out, a chaos of efforts to understand and to make coherent the field.
Eighth, what’s the possibility that we should have three different kinds of negotiation thinking and that they need not overlap much? These might be (1) introductory teaching and training, (2) advanced teaching and training, and (3) advanced understanding of what a negotiation is.
For example, it may be that we teach introductory negotiation using GTY et al. not because it accurately describes the field, but because it feels accessible for students, and gives them confidence that they can really do this, two goals of significance.
Advanced negotiation teaching and training, for experienced negotiators, should reflect a reality they can understand and use. This may contradict GTY et al., but will allow them to be better negotiators (if indeed we know how to make them so).
And what I call advanced understanding might have to do with how scholars understand the negotiating process. This may provide very different vocabularies and categories than would be useful for professional negotiators trying to learn how to negotiation better. Or not. We probably have this situation now in that much of the research we have (e.g., in the Honeyman and Schneider et al. books), though it has nearly no impact on teaching and training.
John: I think that your suggestion for using GTY as the basis for introductory teaching is an absolutely terrible, horrible, no good, very bad idea. Not only that, I don’t like it.
I used to use GTY theory in my courses and I understand the appeal of an apparently simple story of two models, which often are presented as the right and wrong way to negotiate. It is comfortable and easy for us to continue on cruise control using concepts that students often find very appealing.
The problem is that this is like one of those annoying jingles you can’t get out of your head. And the misleading nature of the GTYtwo-model structure usually lasts indefinitely. Not everyone will take advanced training and even if people do, it’s hard to dislodge those darn ideas from their brains.
Based on the research in this article, I revised the chapter in my book, Lawyering with Planned Early Negotiation, to briefly summarize the GTY scheme, point out the problems, and then focus in depth on my framework. Take a look. And readers should feel free to use it in your courses.
I assume that there may be better conceptualizations than this and I hope this symposium will point us in such a direction. But I think it is a heck of a lot better than GTY, despite how useful it has been in the past. (Please pardon my strong language.)
David: Ninth, our field has focused on a tools approach: what can/should a negotiator do at a given moment?
I am increasingly of the view that this is inadequate or worse. “Tools” is a metaphor; we draw it from a setting in which one person works on a material. Tools exist based on a set of cause-and-effect assumptions. However, cause and effect is a very misleading way to think about negotiation because it is, at its heart, interactive.
We have had much writing about the interactivity, e.g., Raiffa, but it has not penetrated the basic canon; it sort of sits to the side. We respect it, but don’t quite know what to do about it. The Oil Pricing simulation and its kin is a good example of interactivity; we use it without quite knowing how to integrate it.
John: I agree that it is hard to make strong cause-and-effect claims about negotiation. There are so many variables that one can’t confidently predict the outcome of any particular interaction.
But without any tools – including theoretical concepts and causal theories – people are left to flounder without any ideas about what they might try. So, rather than eschewing tools, I think it is better to suggest that people consider what tools might be appropriate in particular situations, observe the results, and be prepared to try Plan B if Plan A doesn’t work the way they want.
Linda Putnam distinguishes prescriptions from heuristics which “provide guidelines based on deciphering a complex and dynamic process rather than offering specific directives for action.” I think that such heuristics can be valuable tools for negotiators.
David: Tenth, interactivity is an inevitable condition of negotiating. There are others. Together these inevitable conditions describe the experience of being a negotiator. One might read most of our tools as ways of coping with, perhaps overcoming, these conditions. Example of such conditions:
- Uncertainty about self and other. What will he accept? What will I accept? How important is one outcome or another? Am I screwing this up? How long will this take? Am I being a sucker? Is he conning me How do we respond to uncertainty? In thinking about decisions under uncertainty, we have the math of calculating decision trees. But what about the Kahneman et al. work on the impact of uncertainty on risk taking, assertiveness, perception and memory, etc.?
- We all acknowledge the impact of history (and its operating agent, memory) on our present. But where is it in thinking about negotiation? There is the history of relationship with the other; the history of experience with negotiating; the history of relationship of negotiator to subject matter; and the future relationship of parties and subject. We do have some good work on this, e.g., David Axelrod’s book, The Evolution of Cooperation.
- Our desire to read minds contrasts with how bad we are at it (see above).
- The impact of power on our sense of self and on the other. Power is at the heart of every negotiation from start to finish. It can’t not be. This is a huge topic, studied elsewhere but not to my knowledge brought into our domain. I am not discussing tools of power or sources of power. I am thinking about how power – its use and its imminence – impacts how we see, recall, decide, and act.
- Complexity is a hot topic but I cannot say that the books in our field offer much insight into understanding the negotiating process or the way we experience the process.
This tenth paragraph is for me the most interesting because it suggests how we might go forward. For example, one can look at each of the conditions I outline in paragraph ten and see them as hurdles to reaching an agreement. But in my experience, most negotiating parties do reach agreement. Not all, but most. How do they do that?
One part of the answer is this: Paragraph ten suggests hurdles built into every negotiation. But there are also forces that push toward agreement: most people enter negotiation considering the benefit of reaching an agreement; the prospect of future relationships encourage reaching a deal now; expectations of others not at the table often make reaching a deal more attractive than “failing” to do so.
How do the hurdles and the pro-agreement forces interact? This is something to learn from your lawyers and from Wright and other cases.
My inclination is to study real negotiations that have been subject to serious study and writing, to see what the negotiators did, successfully or not, to see what boundaries they ran into, and finally to see how their humanity (reactions to history, complexity, power, etc.) works as they negotiate. Why?
- I think we are more likely to help negotiators improve their negotiating if their and our thinking is based on a verifiable notion of the many realities at work in a negotiation.
- If there are lessons to teach about good negotiating, they should grow in the context of the forces inherent in the process.
- The humility inherent in negotiating (one can’t succeed alone) needs to be part of everything we teach.
- We should focus more on the humans doing the negotiating and perhaps less on the tools they wield. Tools have a role, but we may have foregrounded them so much that we forget that agreements are sought and reached, or not, by whole people.
John: These are great questions for readers and participants in our symposium to consider.
Thanks so much for engaging with me about the ideas in my article. I hope that this conversation will help us all think more clearly about the issues.