The power of repeat, high-value interaction

IN Partnership with

Infrastructure dispute clauses can feel as impenetrable as Finnegans Wake. Drawing on game theory, this piece offers a “skeleton key”: design ADR to foster repeat, high-value cooperative interactions, not one-off adversarial battles

I REMEMBER struggling with Joyce’s Finnegan’s Wake. Every time I thought I might be getting an inkling as to what was going on in that book, I’d end up turning a page and despairing of ever understanding it. Then I stumbled across Joseph Campbell and Henry Morton Robinson’s A Skeleton Key to Finnegans Wake and the light flickered on (albeit dimly). Joyce had a point and had written a masterpiece. The genius of his book was deliberately inaccessible without deep engagement on the reader’s part and the help of an informed guide, but with those steps taken, the book, however difficult, was well worth the effort. Many will see a parallel with infrastructure ADR. Every time you think you’ve grasped the full range of ADR options available in the average project agreement, they seem to change in character, form, and substance. The closer you look at arbitration, for example, the more it starts sounding like, looking like, and costing like plain old litigation. The closer you look at interests-based mediation, the more it starts looking like a kind of low-key evaluative process. The closer you look at adjudication, the more it starts looking like a kind of expensive coin toss. The closer you look at what is meant to be the relational and collaborative “amiable negotiation,” the more it may start looking like just another distributive rights-claiming hurdle that one must cross. But this cannot be true. These procedures are meant to work together to provide a fair process leading to certainty at the lowest transaction cost reasonably possible. To achieve that goal, however, all these processes, like chapters in a book, require both deep engagement and an informed guide.

Duncan W. Glaholt, CArb, is the founder and principal of Glaholt ADR Inc., a provider of independent mediation and arbitration services to the construction industry in Canada and internationally. He’s counsel to leading construction law firm Glaholt Bowles LLP. He practised construction law since his call to the Ontario bar in 1979 and was active in the introduction of statutory adjudication into Ontario and Canada. Since 2019, he’s devoted his time to the mediation, arbitration, and resolution of construction disputes. He’s chaired arbitration, dispute review, dispute adjudication, and combined dispute boards and served as a court-appointed claims officer. He’s authored, co-authored, and edited several reference books on construction law and ADR including the recent Construction Dispute Mediation: Theory and Practice. He’s an adjunct professor at the University of Toronto and recipient of the CBA Ramon John Hnatyshyn Award for law reform and legal scholarship, the OBA Award of Excellence in Construction Law, and the Toronto Construction Association Lifetime Achievement Award.

Find out more

Finding a skeleton key in game theory How do we find deep engagement and an informed guide? Is there some kind of skeleton key that will unlock and render useful all of these ADR methodologies? The answer is “yes.” To find this answer, however, one must look beyond the names and conventional rules in each case and try to find the common thread. When these many well-intentioned processes work, what is it that is working? The first step in answering these important questions is to find the right guide. Here, the right guide is Robert Axelrod, and his small and easily understood book, now just over 40 years old: The Evolution of Cooperation, Basic Books, NY, 1985, reprinted in 2025. Axelrod and his researchers set out to answer a simple question, which I paraphrase this way: what is the most successful competitive strategy when one is blind to an opponent’s choices but directly affected by them? His way of scientifically answering this question was bold and insightful.

Designing ADR to evolve cooperation, not conflict Axelrod started by accepting the adversarial nature of all rights-claiming games, in which category I would firmly place all current forms of infrastructure dispute resolution, including adjudication, arbitration, amiable negotiation, mediation, and litigation. He devised and ran a “Prisoner’s Dilemma” computer tournament, where many iterations of a conventional Prisoner’s Dilemma game were played using strategies that were submitted in the form of computer programs by academics all over the world. Entrants were encouraged to include in their programming a feature that changed competitive strategy in response to the relative success of a previous strategy, over many iterations of the same game. In other words, just as in the life of any infrastructure project over many years, one could remember their own strategy and apply that knowledge to each subsequent iteration of the game. A conventional two-party Prisoner’s Dilemma game goes something like this: two people are charged with a serious crime. They are interviewed by the prosecutor separately. Each can either cooperate (by refusing to answer questions) or defect (by “rights claiming,” implicating the guilt of the other party as sole perpetrator of the crime). Each prisoner is blind to the other’s choice. If both stay silent, each gets a moderately negative outcome for non-cooperation (one year in jail). If one defects and the other does not, then the defector gets a high payoff (zero years in jail) and the non-defector gets a high penalty (10 years in jail). If both defect, by each blaming the other for the crime, the prosecutor gives each a moderately high penalty (five years in jail). After a statistically significant competition, U of T professor Anatole Rapaport was found to have submitted the winning strategy. His strategy was so robust that it won against all other strategies, no matter what, and yet was so simple that it could be contained in four lines of basic coding. His strategy was “tit-for-tat starting with cooperation.” In other words, one’s opening move in any competitive environment when one is blind to a competitor’s choices must always be cooperation, followed by reciprocation of the opponent’s last move. The important point here is that just like infrastructure disputes, one side’s choices are made iteratively over an extended period against the same opponent. Thus, strategic learning is possible, and many iterations of the same game give each player an opportunity to optimize the success of their negotiation strategy. Axelrod’s experimental methodology led him to the conclusion that among competitive commercial actors, cooperation evolves naturally from repeat, high-value interaction. That suggested the title of Axelrod’s book: The Evolution of Cooperation. Things get massively more complex when systems are involved instead of two-party disputes. This would occur, for example, where individual members of a DBJV recombine in some other form with the same public infrastructure entity on successive infrastructure projects. For two-party infrastructure disputes, however, the lesson of Axelrod’s game remains intact almost 40 years later. Thus, we have a true skeleton key to construction industry ADR: in distributive, rights-claiming situations, where two disputants are blind to each other’s choices yet dependent upon them, engagement in repeat, high-value interaction leads naturally to cooperative outcomes and discourages defection to adversarialism.

Curating repeat, high-value interactions in practice If I am right in identifying this as a skeleton key to infrastructure dispute resolution, the stepped dispute resolution processes are set out in any project agreement, including DRB, CDB, adjudication, arbitration, mediation, or amiable negotiation, or in whatever order or combination must be actively managed to generate repeat, high-value interaction between decision-makers. The class of “decision-makers” includes not only to the parties themselves, but also their lawyers and their experts, in each case both internal and external. In practical terms, curating repeat high-value interaction may include:

This skeleton key also allows us to reformulate the role of third-party neutrals hired to participate in contractual ADR processes. Rather than view their role narrowly (to adjudication, arbitration, recommendation, determination, or mediation, for example), it becomes possible to view their role more liberally as one of curating iterative repeat, high-value interaction and helping the parties to identify and contextualize reciprocation when it occurs, to re-frame reciprocation in a way that encourages repeat interaction by decision-maker, and generally to foster the use of “tit-for-tat beginning with cooperation” strategies. Adjudicators, arbitrators, and dispute boards can use this skeleton key before and after hearings and mediations to encourage the evolution of cooperation as an opening strategy and reciprocation as a follow-up. Mediators can use this skeleton key in pre-mediation organizational meetings, “education sessions” conducted in caucus, and early expert conferencing, for example. Even the simplest procedural decision can be an opportunity for repeat, high-value interaction. Now back to finishing Finnegan’s Wake.

Published May 20, 2026

The evolution of cooperation requires that individuals have a sufficiently large chance to meet again so that they have a stake in their future interaction. If this is true, cooperation can evolve in three stages

Encouraging the use of working groups, breaking big problems down into meaningful components, and using working groups to move iteratively from underlying technical issues to commercial issues, and finally to legal or interpretive issues Treating each stage of each step in the prescribed contractual dispute resolution process (scheduling, document production, etc.) as an opportunity for repeat high-value interaction among decision-makers Reducing transaction costs to encourage this approach by working virtually only, especially at the beginning

SPECIAL EDITIONS

Energy

Mining

Finance

Infrastructure

Litigation

Technology

Health Sciences

Insolvency and Restructuring

Mergers and Acquisitions

Rankings

Lawyers

Firms

Lexpert 500 Cross Border

Who's Who

Methodoloiogy

Toronto

Montreal

Ottawa

Vancouver

Calgary and Edmonton

Regina & Saskatoon

Winnipeg

Halifax

Atlantic Canada

LARGEST FIRMS

Legal Insights

Features

Cross Border

In House Lawyer

Big Deals

Premium

Lexpert TV

News

Lexpert Rising Stars

Canadian Law Awards

Awards

DIGITAL EDITIONS

Under what conditions will cooperation emerge in a world of egoists without central authority?

Axelrod, The Evolution of Cooperation, Rev. Ed. Basic Books 2006, p. 3

The beginning of the story is that cooperation can get started even in a world of unconditional defection . . . The middle of the story is that a strategy based on reciprocity can thrive in a world where many different kinds of strategies are being tried. The end of the story is that cooperation, once established on the basis of reciprocity, can protect itself from invasion by less cooperative strategies . .

Axelrod, The Evolution of Cooperation, Rev. Ed. Basic Books 2006, pp, 20, 21