IDEOLOGICAL TURING TEST
Argue an opponent's view well enough that strangers can't tell you're not a believer.
What this is
The original Turing test: a machine passes if humans can't tell whether they're talking to a human or a machine. The ideological version: you pass if defenders of a position you disagree with can't tell whether you actually hold the position.
This is the most demanding version of steelmanning. It requires not just understanding the argument but inhabiting its motivations — feeling the world from inside the position. Most arguments would improve if both sides could pass this test before responding.
Steps
- 1.Pick a view you disagree with. Choose one you've ENGAGED with — not a strawman you've avoided.
- 2.Write a 300-word essay arguing FOR the position. In first person. As if you held it.
- 3.Specifically include the motivations: not just what the position says, but why someone reasonable might find it compelling.
- 4.If possible, show the essay to someone who actually holds the view. Ask: 'Does this sound like one of us, or someone pretending to be?'
- 5.If they say pretending, ask what you missed. Revise.
- 6.If you can't access a real holder, ask yourself: would I be embarrassed for the holder to read this?
What did writing the essay teach you about why intelligent people hold the view? What's now harder to dismiss?
Reflections you write below are saved to your trajectory — Claude reads the prose and adds a small dimensional shift to your map, the same way it does for daily dilemmas and diary entries.
Sign in to save your reflection — it'll feed into your trajectory the same way dilemma and diary entries do.
More on this practice
The economist Bryan Caplan proposed the test on his blog in 2011, borrowing the frame from Alan Turing. Turing's original imitation game asked whether a machine could converse well enough that a judge couldn't tell it from a human. Caplan's version asks whether you can argue for a view you reject well enough that its actual believers can't tell you're a critic in disguise. Passing means you've understood the position from the inside — not just its claims, but the felt reasons someone holds it.
It's the most demanding member of the steelmanning family. Charitable interpretation asks you to get the view right; steelmanning asks you to build its strongest form; the ideological Turing test asks you to inhabit it convincingly enough to be mistaken for a native. That last step exposes a particular kind of ignorance — the comfortable conviction that the other side is simply stupid or wicked, which survives only as long as you never have to reproduce their reasoning in a form they'd applaud.
The underlying idea is old. Mill, in On Liberty, warned that 'he who knows only his own side of the case knows little of that' — you don't really understand your own position until you can state the opposing one in its full force. The test makes Mill's warning operational and falsifiable: show your essay to a believer and find out, empirically, whether you understand them or only think you do.
Common pitfalls
- Writing a version dripping with tells — the faint sneer, the giveaway caricature — that no actual believer would produce. If they can spot you, you've failed the test by definition.
- Capturing the claims but not the motivations. The hardest and most important part is why a reasonable person finds the view compelling, not merely what it asserts.
- Picking a view you've never seriously engaged. You can't pass a test on a position you've only met through its opponents.
A worked example
You favor open borders and decide to write the restrictionist essay in the first person. You can list the slogans, but to pass you have to reach the reasons a thoughtful person holds the view: that a wage floor and a welfare state may depend on bounded membership; that communities have a legitimate interest in the pace, not just the fact, of change; that 'the world's poor' is a real moral claim and so is 'my unemployed neighbor.' You show it to a restrictionist friend. They say it sounds like one of them — except you missed the argument about assimilation capacity, which they consider central. You revise. Whatever you now think of the view, you can no longer pretend its holders are simply heartless, because you just made their case in a form they signed off on.
Thinkers in this lineage
- Bryan Caplan — Coined the 'ideological Turing test' in 2011, adapting Turing's imitation game to belief.
- Alan Turing — His 1950 imitation game is the original — indistinguishability as the test of a capacity.
- John Stuart Mill — On Liberty Ch. 2: knowing only your own side of an argument is barely knowing it at all.
Where to read further
- On LibertyJohn Stuart Mill · 1859
Chapter 2 is the classical argument for why you must be able to state the other side.
- The Scout MindsetJulia Galef · 2021
A modern treatment of the habits — including this test — that keep reasoning honest.
Pairs well with
- Steelmanning the opposite →
Write the strongest possible version of the view you most reject.
- Charitable interpretation →
Before arguing against a position, prove you understand it well enough that its holder would say "yes, that's it."
- Switch sides →
Argue both sides of a debate, alternating, until you no longer know which side you started on.
Kindred practices
- Devil's advocate — The lighter, looser cousin — argue the other side once, without the bar of fooling a believer.
- Red team — The institutional form — a group tasked with making the adversary's case as convincingly as possible.
Three doors lead onward.
- 01 · QUIZThe InheritorFind your archetype — exercises hit differently when tuned to who you are.CONTINUE ▶
- 02 · NEXT EXERCISEThe 60-second caseCompress your argument until 60 seconds is enough.CONTINUE ▶
- 03 · DAILYThe CrucibleA philosophical action to actually do today. Tomorrow you report back.CONTINUE ▶