When You Actually Have to Talk to Someone You Despise
Robert Mnookin is a Harvard Law professor who wrote a fairly important paper on a specific negotiation problem that most people encounter at some point. The idea is simple but uncomfortable: there are situations where refusing to negotiate with someone is the morally right thing to do, and there are situations where that refusal actually causes more harm than any deal you might cut with them. He calls these moments "devil's bargains." Mnookin's framework doesn't give you a neat algorithm. It gives you a set of questions to ask yourself before you walk into a room with someone you consider fundamentally objectionable. The first question is always the same: what happens if I don't negotiate? Not in some abstract moral sense, but in concrete terms. Who gets hurt? What resources get wasted? What precedents get set? The second question is equally uncomfortable: what am I willing to accept from this person, and where is my line? Most people skip straight to the third question, which is about strategy and tactics inside the room. But Mnookin's point is that you have to answer these first two questions clearly before you even think about your opening offer.
I dealt with a situation three years ago where I had to negotiate with a company that had systematically violated my own team's intellectual property. The legal angle was clear, but every lawyer I talked to said the same thing: litigation would take four to six years and cost upwards of two million dollars, and the outcome was genuinely uncertain. Walking away entirely wasn't realistic because they held patents we needed for our product roadmap. So I sat down with Mnookin's framework and wrote out what each path actually looked like. Fighting meant filing suit, which would have frozen our product development for an extended period and burned through cash reserves. Bargaining meant accepting some licensing terms we'd rather not have accepted, but it kept the product moving and preserved our funding. The devil's bargain here was obvious once you mapped it out, and the legitimacy cost of fighting was real — it would have validated a system where bigger companies can grind smaller ones down through procedural exhaustion. There is a counter-intuitive piece most people miss about this framework. The devil's bargain isn't primarily about your own moral discomfort. It's about the social cost of legitimizing certain kinds of actors. Mnookin argues that sometimes the reason you shouldn't negotiate isn't because you personally find the other party offensive, but because negotiating with them would send a signal that normalizes their behavior in ways that harm other people downstream. That's a harder calculation to make because it requires you to think beyond your immediate situation.
Another thing beginners get wrong is treating the devil's bargain as binary. It's not. You can negotiate while simultaneously building institutional safeguards that limit what you're giving away. Licensing with tight scope restrictions. Setting precedent through public settlement terms. Creating exit ramps in your agreements. The point isn't to find a perfect clean solution — there usually isn't one. The point is to make the costs explicit and then decide whether the alternative is actually worse. One significant limitation of this framework is that it works best when you have time to think. In fast-moving situations where the other party is applying pressure and demanding an immediate response, the deliberative quality of the analysis breaks down. I've seen people try to apply devil's bargain reasoning in settings where they had hours, not weeks, to make a decision. It doesn't land well under that kind of urgency. In those cases, having a pre-commitment rule — a decision you've already made about what kind of actors you will and won't deal with — is more useful than trying to work through the framework in real time. The other limitation is that the legitimacy question is inherently subjective. Two people looking at the same situation can reach opposite conclusions about whether negotiating would do more social harm than good. There's no neutral arbiter for that calculation, which means you're ultimately responsible for your own judgment call and whatever comes after it.
Get the Full Details
If you want to read the actual work, Mnookin's central essay is titled "Bargaining with the Devil: When to Negotiate When Not to Negotiate" and it was published in 1999 in the Yale Law Journal. You can find it through most academic databases. The core argument is about 35 pages and it's written in a way that's more accessible than most law review pieces. There isn't a single book-length treatment of this framework, which is a minor gap given how often it comes up in practice. The practical takeaway is straightforward. Before you decide whether to sit down across a table from someone you find morally objectionable, write down the consequences of not doing so. Then write down the consequences of doing so. Be honest about both lists. The gap between them is where your decision lives.
