Ediscovery and the Science of Cooperation

TwitterLinkedInFacebookGoogle GmailYahoo MailAOL MailEmailPocketEvernoteInstapaperShare

PrirodneNaukeIn his most recent book, Harvard Law Professor Yochai Benkler surveys the sciences to summarize our current knowledge about what maximizes cooperation. Yochai Benkler, The Penguin and the Leviathan  (1st ed. 2011). The Penguin of the title refers to Tux, the emblem of the Linux operating system, representing the spontaneous appearance of cooperation without any influence from a central authority. The Leviathan represents Hobbesian top-down control.

In summary, science shows that cooperation is part of our genetic, epigenetic, and cultural makeup. Benkler provides examples of how scientific experiments show that people are cooperative by nature, wholly apart from any expectation of reciprocity, rewards, punishments, or other selfish interests. He shows how cooperation has conferred tremendous adaptive benefits on our species, and he believes that cooperation co-evolved with our language and intelligence. He considers cooperation to be a driver of evolution, coequal with competition and survival of the fittest.

To maximize cooperation, we need to know how to activate these innate tendencies in addition to regulating behavior from the top down. Benkler provides examples of the interplay between these two approaches, although he expressly disclaims any attempt to provide a unified field theory of cooperation.

At the pragmatic level, Benkler distills current psychological, social, and economic science into a set of factors that tend to encourage cooperation. No one factor is necessary with the exceptions of the essential underlying factors of communication and “framing.” The factors are:

  • communication;
  • framing, fit, and authenticity;
  • looking beyond ourselves: empathy and solidarity;
  • constructing moral systems: fairness, morality, and social norms;
  • reward and punishment;
  • reputation, transparency, and reciprocity; and
  • building for diversity.

What follows is my sense of how each factor might relate to maximizing cooperation in ediscovery.

Preliminarily, and as set forth below, litigators as a class are adept at the skills required to cooperate. And litigators can and do regularly cooperate with opposing counsel. They usually grant each other extensions of time and other professional courtesies as a matter of course, in order to build a degree of trust and a sense of shared humanity and professionalism and, of course, to avoid judicial wrath. When litigators negotiate and draft a complex settlement agreement or any other jointly-authored document, they are cooperating at a very high level. They may cooperate in order to cope with unreasonable clients or judges. The practice of their close relatives, transactional lawyers, is focused on finding cooperative solutions to help their clients create new business relations. Many transactional lawyers also litigate.

Ediscovery simply presents some unique challenges for several reasons. First, it raises many relatively new legal and factual issues, and so norms and expectations aren’t yet settled. Second, professional competence (including judicial competence) varies widely. Third, ediscovery is exceptionally expensive. And fourth, discovery requires the direct involvement of the litigants themselves more than any other part of litigation.


Satellite Communication ScetchIn ediscovery as in any other context, cooperation depends on good communication. Fortunately, litigators can communicate well.

The ability to communicate effectively is the core competence of all litigators. To litigate well, you must be able to hear, understand, process, and formulate and clearly articulate responses to your client, judges, opposing counsel, potential witnesses, consultants, and your peers. You must be able to ask questions clearly and in different ways, and to know when you’ve gotten a clear answer. You must be able to identify ambiguities in questions and answers and resolve them. You must be able to persuade, and to be persuaded.

That ability is not limited to communicating about the law, the facts, or their intersection. Narrative is often essential to good communication, and motives, emotions, and equities are often necessary to communicate narrative.

Because communication is a litigator’s stock in trade, litigators have the ability that Benkler identifies as the “central” prerequisite to cooperation. And, to the extent that their clients cannot communicate with each other effectively, litigators can act as a static-free channel.

Benkler’s prescription for ediscovery would be to adopt rules and safeguards that make it safer for the parties to communicate with each other about ediscovery and that protect them from the possible pitfalls of inadvertently communicating too much. I’ve proposed concrete examples of such safeguards here and here.

Framing, Fit, and Authenticity

Frame“Framing” is about how the immediate social context and expectations affect cooperative behavior. Benkler gives an example of an experiment involving the classic Prisoner’s Dilemma game, with a neat twist; the subjects in one group were told they were playing “the Wall Street Game,” while the subjects in the other group were told that they were playing “the Community Game.” The result was that the Wall Street group cooperated 33 percent of the time, while the Community group cooperated 70 percent of the time.

The implications for ediscovery are that the odds of cooperation will rise if the process of communicating about ediscovery is presented as an opportunity to minimize everyone’s costs instead of a way of pummeling the opposition. This means that not only the client’s expectations, but also her counsel’s expectations, should be framed that way.

Looking Beyond Ourselves: Empathy and Solidarity

Badge of the Civil Order of Social Solidarity (Spain)Litigation is the antithesis of empathy and solidarity. It’s the last resort – what happens when something can’t be resolved by other means. Still, it’s not completely crazy to talk about enhancing empathy and solidarity between opposing litigants (or their counsel). After all, mediation isn’t crazy, and a big part of mediation is facilitating communication without contentiousness so that each side can understand the other side’s point of view in order to find win-win outcomes.

Litigators are popularly seen as being constitutionally bellicose and argumentative. What most people don’t appreciate is that litigators need to first understand arguments made by the other side before we can effectively refute them.

Also, paradoxically, some of the most toxic litigation arises when good relationships go bad. With appropriate facilitation, the parties can get beyond whatever derailed their relationship and get back on track.

The takeaway for ediscovery is that cooperation will be facilitated if the parties can communicate through a static-free channel.

Constructing Moral Systems: Fairness, Morality, and Social Norms

Simple balance scales-01As the father of two 13-year-old boys, I could happily live out the rest of my life without ever hearing about fairness again.

Briefly then, Benkler breaks down fairness into fairness of intent, fairness of outcome, and fairness of process. He cites studies that show that people will be more or less cooperative depending on their perception of how fair the other party’s intent is, how fair the outcome will be, and how fair the process is.

Their perceptions are, in turn, affected by the larger cultural frames of morality and social norms. People want to be, or to appear to be, moral and normal, and so they will cooperate if their moral and normative context expects it.

In ediscovery, litigators and their clients will tend to cooperate with their opponents if they see cooperation as fair, moral, and normal.

Litigators know about fairness. Fairess is very closely related to reasonableness, and to litigate well, you must know how to be reasonable. After all, much of what litigators do and say is actually or potentially subject to judicial review for reasonableness. So litigators are trained by the requirements of their profession to be able to objectively assess what’s reasonable and, therefore, what’s fair.

The Sedona Conference plays a great role in creating authoritative norms specifically in connection with ediscovery. The Federal Rules of Civil Procedure go far in laying out what is fair and normal, and the proposed Amendments elevate the principles of cooperation and proportionality by giving them a home in Rule 1. State laws are increasingly patterned on the FRCP. Bar associations also propound relevant advice.

More can be done to develop norms about cooperation in ediscovery, not only by advisory groups and legal authorities but also by peer networks, as summarized in an earlier post, and below.

Reward and Punishment

Carrot and stick motivationReward and punishment have their place in facilitating cooperation. However, Benkler points out that certain rewards and punishments that are intended to enhance cooperation can have unexpected paradoxical effects, even making it normal or socially acceptable to be non-cooperative. For example, implementing a monetary fine for non-cooperative behavior seems to tell people that it’s OK to be non-cooperative as long as they pay the fine. This can result in less cooperation than simply relying on social norms. The possibility of paying a monetary penalty for non-cooperation becomes a simple cost of doing business, “crowding out” considerations of fairness and norms.

We want our litigators to accomplish ediscovery with as little judicial intervention (i.e., potential punishments) as possible. I think that Benkler’s discussion of this factor might be useful as advice to the judiciary: if the parties may be able to work cooperatively, don’t inadvertently or prematurely inject the spectre of sanctions.

Reputation, Transparency, and Reciprocity

Benkler finds that reputation networks such as EBay’s and Wikipedia’s foster cooperation because their processes are transparent and so their fairness is clear. Although Benkler doesn’t draw the parallel, I think that crowdsourced information, like Amazon’s product rating system, offers another interesting model for fostering cooperation.

I proposed a possible way to implement a reputation network for ediscovery in an earlier post.

Robert Axelrod’s The Evolution of Cooperation (2d Ed. 2006) describes how and why, in continuing relationships, reciprocity allows cooperation to trump competition. Using computer simulations of the Prisoner’s Dilemma, current science, and real-world examples, he shows how cooperation arises spontaneously in continuing relationships, even between opponents on the battlefields of war.

The lesson for ediscovery is that litigators who are likely to litigate against their opposing counsel again in the future will be more likely cooperate with them. And even within the course of a single extended lawsuit, each side’s optimum strategy is to cooperate at every turn until the very last.

So it might be in everyone’s interest for litigants to be able to select their counsel based at least in part on their reputation for successfully cooperating in ediscovery, and on their good relationships with other litigators in their communities.

Building for Diversity

This factor recognizes that there is no one-size-fits-all solution for designing a system to optimize cooperation. Benkler stresses individual and situational differences, and counsels that the system needs to allow for “asymmetric contributions.”

If there’s a lesson here for ediscovery, it would be that each case involves the interaction of unique personalities with their own motives and values, and that litigants need to understand that discovery obligations don’t always fall equally on both sides of the “v.”

Final Thought

In some circumstances, litigators are unable to cooperate. This can happen, for example, when the client is on an irrational vendetta. And it is not unheard of for certain litigators to bilk their clients by encouraging vendettas. There are also situations where it makes sense to have one litigator handle the competitive aspects of a case while another simultaneously handles the conciliatory aspects (such as settlement talks).

In such cases, it may make sense to engage ediscovery counsel in a limited scope representation. This can be initiated by a litigant, counsel, or the court.


TwitterLinkedInFacebookGoogle GmailYahoo MailAOL MailEmailPocketEvernoteInstapaperShare

2 thoughts on “Ediscovery and the Science of Cooperation

  1. The Honorable Ralph Artigliere was unable to post because of a WordPress problem (which I’m trying to resolve). He writes:

    When I teach about cooperation, I also emphasize the advantages of reducing uncertainty, not only in avoiding having to deal with the judge in sorting out disputes, but also in not having to deal with the opponent’s behavior in response to your unilateral path for discovery. A negotiated or cooperative route gives both sides a better ability to know what will occur and therefore a better opportunity to effectively and efficiently plan. This helps with budgeting, answering to clients and the court, and achieves stress reduction, which is no small matter in a trial lawyer’s life. Further, as the give and take occurs, credibility and trust are developed, which may help later when addressing negotiated settlement or issues in preparing to try the case.

    The best thing is that cooperation in discovery can and should occur without giving up an ounce of zealous advocacy.

    • This seems to be true not only in practice but also in theory. As you probably already know, in The Evolution of Cooperation, Robert Axelrod describes the tournament he organized, in which he invited game theorists to submit algorithms to find what strategy succeeded best in an iterated Prisoner’s Dilemma competition. It turned out that the most successful strategy was “Tit for Tat,” i.e., always cooperating on the first move and thereafter doing whatever the other player did on the previous move.

      Starting out by cooperating is crucial: an uncooperative start tends to lead to a vicious cycle, while a cooperative start tends to lead to a virtuous cycle.

Leave a Reply

Your email address will not be published. Required fields are marked *