Nobody Needs a Fourth Study

Duelling experts don’t produce information. They destroy it — and then everyone stops believing in studies at all.

5 min read

People seated around a meeting table with papers and laptops.


Here’s a sequence I’ve watched play out in more counties than I can count.

The developer commissions a noise study. It says the facility will comply. The opposition doesn’t believe it, reasonably, because the developer paid for it.

So the opposition raises money and commissions its own study. It reaches a different conclusion, using a different method, at a different location, answering a slightly different question.

The developer’s consultant writes a rebuttal identifying methodological flaws in the second study. The opposition’s consultant writes a rebuttal of the rebuttal.

Six months later, the county has four documents, two opposed camps who each believe the other’s expert is a hired gun, and — this is the expensive part — a population that has learned that studies are just what each side buys to say what it already wanted to say.

That last outcome is the real cost, and it’s permanent. You’ve spent six months and a lot of money teaching your community not to believe technical evidence. The next project doesn’t get to start from zero. It starts from negative.

Why more information makes it worse

There’s a comfortable theory that disagreement comes from an information deficit. Fill the gap, and reasonable people converge.

Dan Kahan’s work on cultural cognition found something else. On identity-loaded, high-stakes risks, polarization is often greatest among the most scientifically literate and numerate participants.

That’s not a typo, and it isn’t a story about stupidity. It’s a story about what reasoning capacity gets used for. In a dispute where a person’s standing in their own community is bound up in the outcome, sophisticated reasoning gets deployed to protect that standing. The most numerate resident in the room will produce the most sophisticated critique of your study — not despite being numerate, but because of it.

Which means dropping a 500-page technical document into a room that feels its way of life is under threat doesn’t resolve anything. It supplies ammunition, in both directions, and it reads as condescension besides.

The alternative isn’t fewer facts

It’s a different process for producing them.

Joint fact-finding is a specific procedure with a specific mechanism, and the mechanism is not “everyone gets along.” It’s that the parties commit to the method before they know what the answer is.

Four steps:

1. Scope the questions jointly. Stakeholders negotiate the questions the study will answer — not the answers. This is the step people skip, and it’s the one that matters most. In most technical disputes, the real fight is about which question got asked. A noise study measured at the wrong location, at the wrong time of day, on the wrong weighting scale, will produce a technically correct answer to a question nobody cared about.

2. Select the experts jointly. Each party nominates; the group agrees on a slate, or on criteria. Funding and prior positions disclosed as a matter of course — for everyone, including the community’s nominees.

3. Fix the method first. Protocols agreed in writing before any results exist. This is the load-bearing step. Pre-committing to a method that could produce an unwelcome answer is precisely what makes the answer binding when it arrives. You cannot object to a finding produced by a procedure you helped design and approved in advance, and everyone in the room knows it at the moment they sign off.

4. Interpret and publish jointly. No party spins first. Where the group can’t agree on interpretation, document the disagreement. A joint statement of agreed facts plus identified disagreements is a completely legitimate output — often more useful than forced consensus, because it tells the decision-maker exactly where the real dispute is.

What it costs, and what it buys

It costs a few weeks up front. Scoping takes meetings. Agreeing on an expert takes longer than hiring one.

It buys the elimination of the six-month argument that otherwise follows. One study instead of three or four. And a number in the permit that nobody has to relitigate at the microphone, because the people who would have objected to it helped design the thing that produced it.

For a developer, that’s the commercial case, and it’s straightforward: a jointly scoped study costs less than three commissioned ones and is the only version that ends the argument rather than relocating it.

For a community, the case is different but equally practical. You are almost certainly outgunned technically. Joint fact-finding is the mechanism that gets you a credible answer without needing to fund a parallel technical operation you can’t afford.

When it’s the wrong tool

I’d rather say this plainly than oversell the method, because overselling it is how it gets discredited.

Joint fact-finding does not work when:

  • The dispute is really about values. If the actual disagreement is whether this county should become an industrial county, no study resolves it. Running a technical process over a values dispute wastes months and produces cynicism. A values dispute wearing a technical costume needs to be named as one.
  • The decision is already made. If the entitlement is vested and the process is theater, joint fact-finding launders the theater. Say so instead.
  • A party will reject any unfavorable result. If someone won’t pre-commit, you’ve learned something valuable in week one rather than month six.
  • The timeline is too short. A real study takes time. A rushed joint process is worse than an honest acknowledgment that the schedule doesn’t allow one.

The underlying idea

There’s a principle from the trust literature that sits underneath all of this. Communication aimed at building trust in institutions mostly fails. What works is structure — arrangements that make agreement possible without requiring trust in the first place.

Joint fact-finding is that idea applied to evidence. It doesn’t ask a community to believe the developer’s consultant. It doesn’t ask a developer to accept the opposition’s. It builds a procedure that neither side controls and both side agreed to, and then it lets the procedure produce the number.

Verification buys the time in which trust can accumulate. That’s the whole trick. You don’t start with trust and use it to reach agreement. You start with a verifiable process, reach agreement, and let trust follow if it’s going to.

Most siting fights get that backwards, and pay for it with a fourth study nobody believes.

More insights