For too long there’s been enormous division and vigorous debate in our field about the wisdom, value, and ethics of a mediator providing an evaluation of issues involved in a dispute. Most of this debate has concerned a mediator stating how he or she thinks a judge, jury, or arbitrator might rule if a case is adjudicated. Many academics and trainers argue that this kind of evaluation is improper. Some courts prohibit it outright or at least discourage it. Ethical standards waffle on the topic. But many lawyers and parties consistently say they want their mediator to evaluate.
We are concerned that this argument has distracted us from carefully examining the effects of evaluation in mediation practice. We believe that, at least in cases where all parties are represented by lawyers (pro se litigants, we recognize, pose special concerns), everyone involved can benefit from a mediator’s evaluation – but only when done well.
A working definition
To “evaluate,” dictionaries tell us, means to assess, analyze, consider, value, weigh, judge, size up, or form anApplying this definition, we believe that most mediators evaluate constantly, from the moment of their first contact with a dispute. They assess parties, lawyers, issues, options, and potential avenues for agreement, forming opinions not only about legal issues and likely case outcomes but also about personalities, bargaining tactics, and the conflict’s impact on peoples’ lives and businesses.
Evaluation, in this broad sense, is essential to a mediator’s role. Parties and lawyers don’t hire mediators to be “potted plants”; they expect them to use their wisdom, judgment, and experience to facilitate settlement. And mediators act on their opinions, whether expressed or not. Every question or comment by an experienced neutral reflects her assessment of the best next step in the process. Thus, to us, whether mediators should evaluate is not the question. The question instead is how to evaluate effectively. That leads to two related questions: what should mediators do with their evaluative views, and how does what they do affect the participants and the process?
We think mediators handle evaluations in three general ways: they express them directly; suggest them implicitly or “leakily”; or keep them unspoken and hidden. Perhaps because so many mediators have been told that evaluation is wrong and abstinence is a common prescription, we have not been able to find or hear much about courses, trainings, or writing that focus on the different ways mediators can offer opinions and the effects of alternative approaches. In our view, the power and prevalence of evaluation, combined with the lack of training and discussion of the topic, has left many mediators ill-prepared to evaluate in ways that protect the integrity of the process and enhance the likelihood of success. Our discussion is intended to raise the topic of mediator evaluation anew and encourage our colleagues to join in the discussion.
Because it has been the subject of so much controversy, our analysis starts with and focuses substantially on direct evaluation: a clearly and explicitly communicated analysis or prediction that focuses on legal issues. We understand that direct evaluation can be risky, but we also believe it can be a useful tool in bringing parties to agreement.
First, however, a few clarifications. In our opinion, good evaluation does not include a mediator voicing his personal opinion about what is “right” or “fair” or a “just” outcome in a dispute. This is for two reasons: relevance and impartiality. A mediator’s view of fairness or justice is irrelevant because he will not be the one deciding the case if it is adjudicated. More important, once a mediator suggests to a disputant that a claim or defense is less than “just,” the listener may see him as biased. Even if a mediator can put his personal viewpoint aside, disputants may not believe he has done so. We understand that some mediators convey their personal views about fairness or justice, but we do not defend the practice.
We think direct evaluation, done well, should be an assessment and prediction about someone else’s viewpoint: how a judge, jury, or arbitrator is likely to decide a specific issue or the entire case if it does not settle. Good evaluation, then, is akin to a professional weather forecast, except that the mediator is predicting the atmosphere in a future courtroom or arbitration. Such predictions are relevant because parties almost always see adjudication as their most likely alternative to settlement.
We fail to see how predicting the likely outcome of a trial inevitably compromises a mediator’s impartiality. A weather forecaster’s prediction of rain on a day chosen for an outing does not make the listener think the meteorologist wants her to get soaked, and listeners understand that. In the same way, a mediator’s prediction that a jury may not respond well to a certain fact or argument, if communicated thoughtfully and skillfully, does not make the listener conclude that the mediator is partial to the other side.
We also don’t believe that an evaluation diminishes parties’ self-determination. Our weather forecaster’s prediction of the likelihood of rain doesn’t impinge on the family’s ability to decide whether to rent a tent for their wedding reception; it simply provides useful information for the planner (or litigant) to consider. If anything, a mediator’s offering a neutral view of the value of the parties’ likely alternative to agreement may actually enhance their ability to exercise self-determination.
What are the benefits?
Simply put, a mediator’s explicit evaluation can help parties overcome impasses caused by divergent views of the likely outcome in adjudication of a single legal issue or the entire case.
Evaluation is not magic, however. By the time they start mediation, most disputants have been living with their dispute for months or even years. The idea that they will reverse a strong opinion based on one mediator’s contrary viewpoint strikes us as bordering on hubris (although we understand that disputants may give special deference to retired judges). But even if a mediator’s evaluation doesn’t persuade someone to change his assessment entirely, it can infuse uncertainty or reduce confidence that a future adjudicator will see it his way. Sometimes a party recognizes the flaws in his case but hopes to hide them. A mediator’s assessment warns him that those flaws can unfortunately still be seen.
A mediator evaluation can also influence the views of individual parties. Lawyers often admit that their efforts to convey a realistic evaluation to a client have fallen on deaf (or at least resistant) ears. When the mediator’s evaluation matches the lawyer’s, a client may finally see the handwriting on the wall. If a lawyer had avoided raising doubts to a client for fear of being perceived as disloyal or insufficiently zealous, she might use a mediator’s evaluation as “cover” to introduce uncomfortable topics. A lawyer might also choose to talk only about best-case outcomes, letting the mediator articulate less pleasant possibilities. Later, after the mediator has left, the lawyer may suggest the client consider the mediator’s evaluation, even while maintaining that it’s overly pessimistic.
Evaluation can have other useful effects. A nonbinding opinion from a respected person who has listened carefully to arguments can give a litigant the feeling of a “day in court.” Given the small proportion of cases that go to formal adjudication, this is as close as most litigants will ever get to traditional justice – and much safer. A neutral opinion can also help a party deflect criticism from a spouse or advisor or get approval from a corporate supervisor, whether or not those individuals are present at mediation.
What about risks?
No doubt, evaluation is risky. We see it as the “surgery” of mediation, a tool to use carefully and sparingly, to the minimum extent needed, and only when disagreement about legal issues is the barrier blocking agreement. If the barrier is something other than a legal issue — for instance, strong emotions or poor bargaining techniques – offering an evaluation is the equivalent of prescribing antibiotics to cure a viral infection. It’s an ineffective treatment that carries the risk of side effects.
The largest, most obvious risk is that a disputant receiving an unfavorable evaluation will then perceive the mediator as aligned with the other side, no longer neutral, and may withhold confidential information and mistrust the mediator’s actions from that point on. The “medication” may have been successful, in that the neutral’s evaluation was thoughtful and thorough, but the mediation process is now on life support. Not surprisingly then, when we teach and train about evaluation, we focus on whether to use it at all and how to offer evaluative comments in ways that minimize the risk of losing the perception of
A closer look at evaluation reveals other, less-often-acknowledged risks. One is that the mediator’s prediction of the likelihood of success at trial may not reflect the way the judge or jury will look at it. Unless a specific case – or at least one with similar fact patterns and issues – has been tried many times, no one can know whether a single prediction is right or wrong. Research suggests that non-partisan evaluations are less subject to cognitive distortion and thus morebut this doesn’t mean that all mediator forecasts are correct or that every mediator would make the same prediction. Indeed, when one of us asked lawyer-mediators to evaluate likely outcomes in a simulated case, their responses varied widely. No mediator should think his evaluation is the only reasonable one.
Evaluation can also be dangerous if the disputant takes it as a signal that he cannot achieve his goals in a negotiation. This could trigger loss aversion, one of the strongest influences on human decision-making. Disputants do not always react to mediators’ evaluations with respectful appreciation; often they express denial, disappointment, even anger, and this may trigger strong feelings in the evaluator. Having put forward a thoughtful forecast, the mediator may react defensively to criticism of her work, putting her in opposition to the parties or lawyers – not a place where a mediator should ever be.
A final risk is that mediators will treat evaluation as an end, rather than what it should be: a useful but limited tool to overcome specific obstacles. Even the best evaluation, in other words, should not conclude or unduly narrow the process. It can help get a settlement process moving when it is stalled but should not be used to steer the negotiation toward any particular outcome.
Implicit or “Leaky” Evaluation
When parties have divergent views of the merits and less intrusive measures are insufficient, advocates of facilitative mediation recommend that mediators undertake reality- testing. But this has limitations. As Professor Deborah Kolb, a pioneer in our field, once observed, deciding to reality-test implies that the mediator has developed an opinion about what reality is, that the disputant’s view is different from that reality, and that the disputant would benefit from testing.
Reality-testing is intended to encourage disputants to consider their own views more carefully, not to communicate the mediator’s perspective, but we find it often does not work that way. Best practice may be to begin with questions posed in a scrupulously neutral manner (perhaps “Tell me how you see the future lost-profits claim.”) However, tough litigators (which describes most lawyers we see in mediation) typically respond with highly optimistic assessments (“I have very little doubt we’ll prevail on lost profits.”) If the lawyer is exaggerating knowingly, he may eventually moderate his arguments. Often, however, litigators remain obdurate, to maintain a positional strategy, to please clients, or because they passionately believe in the strength of their case.
At this point training programs often suggest that mediators move to more pointed reality-testing. But it’s very difficult to pose questions that are both non-evaluative and strong enough to shake the views of a stubborn litigant. Indeed, when we asked our teaching colleagues around the country if they could provide models of effective reality-testing in such a situation, we got no examples.
A facilitative reality-tester dealing with a stubbornly unrealistic disputant might say something like “I’m (really) having trouble understanding your take on the second paragraph in the Smith case … Can we go over it again, to be sure I understand well enough to convey your interpretation to the other room?” But a mediator who does this is, in essence, applying her internal evaluation covertly. That’s where the leaking may begin. One can imagine counsel saying to a client after the mediator leaves the caucus room: “Hmmm … She was being really careful, but did you notice that she spent a lot of time asking about how we’d prove lost profits? She said she ‘just wanted to be sure’ we’d thought about it, but she was awfully persistent. I wonder if she was thinking we’re wrong?” Eventually the divergence between the assumptions underlying a mediator’s queries and those supporting the disputants’ answers is likely to become apparent.
Communicating an evaluation through questions can also diminish its effectiveness. Research on how people are induced to change their mind has found that stating opinions explicitly is more effective than hinting throughAnd if repeated reality-testing generates suspicion about a mediator’s impartiality, by the time the mediator moves to direct evaluation, the audience may have shut down and shut her out.
Often, we think, a mediator communicates his evaluation unintentionally. He may do so through his questions, tone, less-than-neutral phrasing or body language, or by the issues on which he focuses. Unfortunately, however, when evaluation is leaked unintentionally, the mediator is unlikely to focus on how to deliver the opinion in a way that reduces its risks.
Mediators sometimes also “leak” their views intentionally, especially in non-American cultures. When we demonstrated explicit evaluation at a conference sponsored by the Dispute Resolution Section in New Delhi, for example, an Indian panelist said that he presented evaluations differently. Asked to demonstrate, he made a series of statements to a mock CEO focusing on issues in proving her company’s lost profits, a risk the mediator thought she was not taking seriously enough: “I’m sure you’ve discussed the issue of lost profits with your lawyer,” he said. “I would expect that you’ve given it careful consideration …” and so on. After a few sentences, no one in the room had any doubt about the mediator’s opinion. A facilitative American mediator would probably have used questions (“Have you discussed lost profits with your lawyer?”), but we doubt that any sophisticated party would fail to get the drift, particularly if the mediator returned to the issue.
Our field has not thought much collectively about the fact that mediators continuously and unavoidably evaluate internally and silently as they do their work. In this context, standards and codes that tell mediators to maintain impartiality and honor party self-determination seem laudable but insufficient.
How a mediator works is inevitably influenced by the way she evaluates – considers, assesses, sizes up – the personalities, legal and factual issues, bargaining patterns, and other aspects of a dispute. This is natural, but it is perhaps better done with eyes wide open. If a mediator internally discredits an argument or feels skepticism about one side’s motivations, it is likely to impact how she conducts the mediation. A mediator who believes that a claim is invalid, for instance, might steer the conversation toward other topics. And if a party asks a mediator for advice about what offer to make next, the mediator’s sense of whether the party’s bargaining strategy is “reasonable” may influence his answer. Opinions can also affect how a mediator communicates arguments and information in caucus-based bargaining. All evaluation creates temptations to favor one side or another. If internal evaluation is inevitable, then it is worth thinking deeply and teaching thoughtfully about how to deal with the feelings and perceptions it can generate.
If evaluation were accepted as a legitimate technique and understood to occur in explicit, indirect, and hidden forms, then teachers, trainers, and practitioners could turn to examining concerns that abstinence policies ignore. These include how to evaluate effectively while maintaining the overall facilitative nature of the mediation process; how to teach good evaluative techniques to students, litigators, and judges; and how to assess the benefits, dangers, and limitations of evaluative techniques and use them as wisely and sparingly as possible. Greater understanding and wider discussion, we hope, will help mediators use this controversial tool more effectively.