Devil's Advocate in Decision Making: The Vatican's Rule

A devil's advocate in decision making is a named assignment, not a personality type and not a mood in the room. One person is given the job, on that specific decision, of building the strongest case against it — with full access to the evidence and nothing to gain from the outcome. An analysis deserves trust only after someone with standing has tried to defeat it and failed.

For roughly four centuries the Catholic Church ran exactly that system.¹ As the canonization process was reorganized in the late sixteenth century² and refined over the two hundred years that followed, one official — the Promoter of the Faith, popularly the Devil's Advocate³ — was required to argue against every candidate for sainthood: natural explanations for reported miracles, selfish motives for deeds called heroic, gaps in the record, witnesses who disagreed. The cause could not advance until his written objections had been answered. Prospero Lambertini, who held the post before becoming Pope Benedict XIV,⁴ wrote the treatise on the process that governed it for generations.³

In 1983 a rewrite of the rules replaced the adversarial trial with a scholarly inquiry.⁵ The title survived; the duty to attack the case did not. Canonizations then ran at a pace no earlier era approached — John Paul II declared more saints than his predecessors had across the preceding centuries combined.⁶ One office does not explain all of it: the same reform cut the ordinary waiting period after a candidate's death from fifty years to five,⁷ reduced the number of verified miracles required,⁸ and many of the new saints were martyrs declared in groups. Several hurdles came down in a single motion. That pattern — the critic retired and the bar lowered at once, under the banner of streamlining — is the thing worth watching anywhere capital is committed.

Owners weighing an investment memo, a manager's pitch, a family loan or the purchase of another operating business meet the same mechanics. The file is almost always assembled by someone who looks good if it is accepted. The questions below test whether an analysis survived an advocate, or simply never met one.

Who, by name, was asked to defeat this case?

A good answer names a person and a date. "Our analyst was given two weeks to kill it; here is her memo, circulated before the committee met." The strength of the Vatican's arrangement was that the objector was an office rather than an attitude — someone whose standing depended on finding what was wrong.

A weak answer praises the room. "We encourage pushback." "The tires get kicked." These describe a place where objection is permitted, which is not a place where objection is required of a named person who will be judged on how well they did it.

Volunteered dissent is rationed by social cost. People hold back when the proposer is senior, enthusiastic or well liked. An assigned advocate is spared that arithmetic. The office exists so that the objection does not have to be brave.

Did the objector work from the same file as the sponsor?

A good answer sounds like equal footing: the person arguing against had the data room, the management meetings, the customer calls — the raw material, not a summary prepared by people who already believed. The Church's promoter worked from the full body of testimony, not the sponsor's brief.

A weak answer sounds like a courtesy review. The objector was sent the finished memo and invited to comment. Reviewing someone else's conclusion is a far lighter task than building a rival one from the same facts, and it produces marginal notes rather than a second opinion.

This is where well-meant processes quietly fail. The dissenter is real and the time is real, but the evidence was filtered by the case's advocate before its opponent ever saw it. Anyone who has sat through diligence on their own company knows how much the picture depends on who assembled it.

Who does the objector answer to?

A good answer can be drawn as a line on a chart. The person assigned to find the flaw does not report to the sponsor of the memo, is not angling for the next engagement from that sponsor, and is not a relative whose standing in the family depends on keeping the peace.

A weak answer wraps the conflict in courtesy. The reviewer "partners closely" with the banker who brought the idea. The outside firm was introduced by the seller's counsel. The same adviser serves both the family office and the cousin who wants the purchase approved. Speed is usually the reason offered for the overlap.

A critic who needs the sponsor's goodwill sands the finding down until it no longer threatens the timetable. Independence is not a tone of voice; it is the structural distance that lets an objection survive the people who want a yes. It is a fair thing to raise directly when putting questions to any wealth manager: who here is rewarded for the ideas that did not happen?

Which facts came from someone who loses if this goes ahead?

A good answer sorts the file by allegiance. Facts from the people proposing the action are unavoidable. The question is whether anything material was also checked with someone who gains nothing if it proceeds — a customer free to leave, a supplier with choices, a departed executive. Scuttlebutt, the habit of asking outside the room that holds the thesis, does not expire at the sale.

A weak answer treats the assembled file as the world. Every figure traces back to management or to an adviser management hired. Even a quality-of-earnings report — the outside check on whether reported profit will hold up — still answers to whoever commissioned it. The awkward customer was dismissed as unrepresentative, and nobody called another.

A tidy file can be a closed loop. Trust begins where the record contains something the case's supporters would rather have left out.

What would have to be true for this to be wrong, and where is that written down?

A good answer is a short list of concrete conditions, dated before the meeting: the new facility opens more than a year late; three customers turn out to be one buying decision; the founder's successor cannot hold the sales team. Each names the observable fact that would settle it, and what the group does if that fact arrives after everyone has grown fond of the plan.

A weak answer is a risk section written in a quieter voice than the thesis — "execution," "markets," "key personnel" — or a restatement in the negative: it would be wrong if the business were not as good as believed. Labels are not objections, and that one cannot be tested.

Writing it down matters as much as naming it. Spoken dissent evaporates; once an outcome is known, everyone remembers raising the concern that mattered and nobody remembers the rest. A dated page turns the later conversation from "did anyone see this coming?" into "we wrote that this was the risk — why did we accept it?" The strongest objections, once named, become the early warnings to watch long before trouble reaches the numbers.

What happened to the last person who slowed a decision down?

A good answer describes a consequence that does not punish being right. The objector keeps the fee, the seat or the role. The objection is recorded, answered in writing, and either met or left standing as a reason to wait. Rotating the job helps, so that no one becomes the house cynic.

A weak answer is a slogan followed by silence. "Candor is valued here." The real test is what became of the last person who delayed a closing. Inside a family the penalty is rarely a firing — it is the next gathering, the sibling who stops returning calls, the counsel of long standing who is suddenly not quite the right fit.

Where being right costs belonging, the room will produce agreement and call it judgment. And the longer a decision has been in process, the more everyone present has a stake in a yes, if only the time already spent.

Is a hurdle being lowered in the same motion that retires the critic?

A good answer separates the moves and names each one. Faster and easier are two different changes, and so is the removal of the person paid to object. If a committee is shortening its diligence period, relaxing a size limit and dropping the contrarian review in one meeting, each change deserves its own defence — and someone should say plainly what now does the critic's old work.

A weak answer calls the bundle streamlining. Speed and collegiality are presented as a single virtue, and the standard is said to be unchanged because a checklist still exists. A checklist nobody is assigned to use against the file is a filing ritual.

This is the Vatican's real lesson. The title outlived the duty, the waiting period shrank, and the number of approvals rose — and no one can now cleanly separate which change did what. Processes that change several things at once lose the ability to learn from their own results.

---

These questions belong to the week before a yes, not the post-mortem after it. They apply to any analysis that will move serious capital: an investment memo, a manager's pitch, a family loan, a plan to buy a business. The point is not to manufacture conflict but to make approval mean something — an idea that has survived a real adversary is evidence, and one that has not is momentum with a cover page.