History & Social Studies: Decision under period information

What you'll be able to do: Decide as the people in your period actually had to (with only what they knew and what they valued) which is the direct cure for writing history that makes everyone before 1990 look slow.

"They should have known"

That phrase, or a close cousin of it, appears in a huge number of mid-band history essays. Chamberlain should have known Hitler couldn't be trusted. The Tsar should have known reform was coming. The generals should have known the machine gun changed everything.

Every one of those claims uses information the writer has and the actor didn't. Psychologists call the underlying error hindsight bias: once you know an outcome, you reliably overestimate how predictable it was, and misremember your own prior uncertainty as confidence you never had. Fischhoff's original studies found that people who'd just been told an outcome rated it as more inevitable than people asked to predict it beforehand, while believing they'd have predicted it anyway. Historians have a name for the same failure applied to values rather than facts: presentism, judging the past by standards nobody in it held.

Both errors have the same fix. Take the actor's information and the actor's values away from yourself, and decide.

The two things you have to give up, not one

Most students who've heard of hindsight bias think it's about facts: don't use information from after the decision. That's half the method.

The other half is values, and it's the harder one to give up because you don't experience your own values as historically located, they feel like "just how you see things." A 1917 decision to keep fighting reads as obstinate cruelty if you bring 2026's cost-tolerance to it. It reads differently (not necessarily right, but legible) once you've also given up the assumption that a government's duty was to minimise casualties above all else, because that wasn't yet the settled value it is now.

This is why the method is "period information," not just "no spoilers." You are restoring a whole decision environment: what was knowable, and what was thought to matter.

How to run it

1. Pick the decision point precisely. Not "the outbreak of war" but the cabinet meeting, the memo, the specific morning.
2. Set the date lock explicitly, in the instruction below. This is the part students skip and the part that does the work.
3. State the decision as the actor had to state it: the options actually available, not the ones visible in hindsight.
4. Decide, in role, before asking what happened.
5. Only then compare your decision to the actual one, and to the outcome.
6. Ask what values, not just facts, you had to set aside to make the decision look reasonable rather than merely explicable.

It is [DATE]. I am [ROLE]. Brief me only on what would have been known to someone in that position on that date, no later events, no outcomes, no information that reached the public or the decision-makers afterwards. Also tell me what this decision-maker would have taken for granted as right or normal, even if we'd reject it now. Then give me the decision I actually face. Don't tell me what happened until I've decided.

The instruction to withhold the outcome is not decoration. If the tool answers with information from after the date, even a phrase like "which would prove disastrous": the exercise has already failed, and it fails quietly, because you won't notice the contamination while you're inside it.

Worked shape

A student essay argues that the July Crisis of 1914 shows Europe "sleepwalking" into a war nobody wanted, and marks decision-makers down for "failing to see" where mobilisation timetables were leading.

Run in role as the German Chancellor on 29 July, restricted to that date: the information available is fragmentary, contradictory and days old by the time it arrives; the value in play is not "avoid war" as a near-absolute good but national honour and alliance credibility, which most European statesmen of the period ranked as comparably weighty. Decided from inside that position, mobilisation looks less like blindness and more like a rational response to an information and incentive environment nobody involved could see whole.

That doesn't make the decision correct. It replaces "they should have known" with a specific, checkable claim: given what was knowable and what was valued, was this decision reasonable, and where exactly did it stop being so? That's an argument. "They should have known" is a verdict standing in for one.

Across the subjects

History. As above, the direct fix for judging decisions by their outcomes rather than their inputs.

Political science. Replay a policy decision with only the contemporary forecasts, not the outcome. Most "obviously flawed" policies had a respectable case in front of the people who made them.

Sociology. Reconstruct a reform debate using the period's own understanding of the problem, what a "solution" to poverty or crime meant before the concepts in use today existed.

Anthropology. Restrict yourself to what a fieldworker could observe and believe at the time a study was written, before later theory reframed it.

Geography. Planning and engineering decisions made without knowledge that came later, germ theory, plate tectonics, climate data. What looks like negligence is often the edge of contemporary science.

International relations. The natural home of this method. Crisis decision-making is almost always defended or condemned using information that reached the actors only afterwards.

Pitfalls

1. A leaked outcome. If the tool's briefing contains a hint of what happens next, the exercise is compromised, restart with a stricter instruction.
2. Restoring facts and forgetting values. The harder half, and the one that actually explains behaviour that otherwise looks irrational.
3. Using this to excuse rather than explain. Understanding why a decision looked reasonable is not endorsing it, and a good essay says both.
4. Stopping at "it's complicated." The method should produce a specific account of where reasoning was sound and where, even by period standards, it wasn't.
5. The tell: your account of a decision cites something nobody in the room could have known. That's not analysis with hindsight removed, it's analysis with hindsight still doing the work.

Try this today

Take a decision in your topic that your essay currently calls a mistake.

Set the date lock, restrict yourself to what was knowable and what was valued, and decide it yourself before checking what actually happened. Then write one sentence on exactly where the reasoning stops holding up, by period standards, not yours.