What exactly is gaslighting?

It's a form of manipulation that leads someone to doubt their own perception of reality, their memory or their judgment. The term comes from Patrick Hamilton's play Gas Light, staged in 1938, in which a husband convinces his wife she's going mad. Sociologist Paige Sweet showed in 2019 that gaslighting is only fully effective when it rests on a pre-existing power imbalance.

Is an unproven cheating accusation gaslighting?

Not automatically. It becomes gaslighting when it's public, repeated, unsupported by evidence submitted to the competent bodies, and when it leads the accused to doubt the value of their own results. In July 2026, the FIDE Ethics Commission sanctioned Vladimir Kramnik for, among other things, psychological abuse and unfounded public accusations, holding that publicly naming identifiable players without institutional verification exposed them to unjustified reputational and psychological harm.

How do you tell a demanding coach from one who gaslights?

A demanding coach disputes your moves. A manipulative coach disputes your perception. The first says "this move is bad and here's why," a verifiable claim you can contradict. The second says "you never understood this position" or "you're misremembering what I told you," claims that are about your mind rather than the board and that can't be checked.

Why is chess such particular ground for this?

Because the facts are fully archived. Every game exists as notation, retrievable and replayable. So gaslighting can't be about the moves themselves, which are provable. It shifts to what isn't recorded: your intentions, your understanding, your real level, what you did or didn't see.

What should I do if I'm being gaslit at a chess club?

Keep the written traces, they're your main protection: game notation, messages, analyses. Test claims against the archived facts rather than against your memory. Have your games analysed by an independent third party. And if the matter involves a formal accusation, use the official channels, which exist precisely to take the case off the terrain of rumour.

Gaslighting in Chess: Accusations, Toxic Coaching and Denial

The game is over. You lost, and you're going through it together. Except the game he's describing isn't the one you played.

Your position on move 20, the one you thought was solid, was "completely lost long before that, you just didn't realize." Your plan, the one you'd prepared, you "never really understood it." The combination you found at the end, "anyone would have seen that." And then, on the way out: "you've always had this problem, I've told you ten times."

You don't remember him telling you ten times. You don't remember being lost on move 20. But he sounds so certain that you start wondering whether your memory is reliable.

There it is. It has a name.

Where the Word Comes From

The term comes from a play. Gas Light, written by British novelist Patrick Hamilton, premiered in London on 5 December 1938. The plot is simple and chilling: Jack Manningham wants his wife Bella committed so he can seize her inheritance. He never hits her. He moves objects and denies moving them. He lowers the gas lamps in the house, and when she notices the room darkening, he tells her she's imagining things.

The 1944 Hollywood adaptation with Ingrid Bergman made the story famous and gave the technique its name. The verb to gaslight entered psychological vocabulary much later, and everyday usage later still.

What's remarkable about Hamilton's device is that it rests on no lie about external facts at all. Manningham doesn't tell a false story. He attacks the measuring instrument directly: Bella's ability to trust what she perceives. Once that instrument is discredited, everything becomes negotiable.

What the Research Added

For a long time, gaslighting was treated as an individual pathology: a manipulator, a victim, a toxic relationship.

In 2019, sociologist Paige Sweet published a paper in the American Sociological Review that shifted the analysis. Her thesis: gaslighting isn't primarily a psychological phenomenon, it's a social one. It only works when it can lean on pre-existing inequalities and on available stereotypes.

Put differently, telling someone "you're imagining things" doesn't carry the same power depending on who says it. If the claim comes from a person whose authority is socially recognized, it doesn't merely contradict: it mobilizes a structure. The doubt it installs is reinforced by everything the surrounding community already takes for granted.

This point matters enormously for what follows, because it explains why gaslighting has nothing to do with simple disagreement. Two players of the same strength who disagree about a position are having a debate. A titled coach explaining to his student that the student misremembers what was said has something more than an argument.

The Chess Paradox

Here's what makes chess fascinating on this terrain.

Chess is probably the human domain where reality is best archived. Every tournament game is notated move by move. Every online game is recorded automatically. The position on move 20 isn't an impression: it's an object, reconstructable in three seconds, on which an engine will deliver a verdict to the hundredth of a pawn.

In a marriage, there's no notation. In a company, the meeting isn't replayable. In chess, it is.

So you'd think chess would be immune. It isn't remotely. It's simply displaced.

Since the moves are provable, the manipulation isn't about the moves. It's about everything else, meaning the vast part of chess experience that isn't recorded:

what you had understood about the position, which appears nowhere in the notation;
what you had calculated and discarded, invisible by construction;
your real level, of which your last game is only a sample;
what was said in training three months ago;
your intentions.

Notation proves what you played. It proves nothing about what you are. And that's exactly where the doubt settles in.

Three Forms, From Most Visible to Most Insidious

1. The Public Accusation Without Evidence

The most spectacular form, and recently the best documented institutionally.

On 3 July 2026, the FIDE Ethics and Disciplinary Commission issued its decision in case 12-2025, following complaints filed by the federation's Management Board and its Fair Play Commission. Grandmaster Vladimir Kramnik, world champion from 2000 to 2007, was found responsible for several breaches: harm to the dignity and respectful treatment of persons, harassment and cyber-harassment, psychological abuse, failure in his role as an example, failure to cooperate with the Fair Play Commission's investigation, and false or unjustified public accusations.

The commission's wording deserves precise quotation, because it describes a mechanism and not merely a fault: publicly associating identifiable players with cheating suspicions without sufficient institutional verification exposes them to unjustified reputational and psychological harm.

The sanction: a two-year worldwide ban from competition and official functions, the final twelve months suspended over a three-year probationary period, together with twelve months of unpaid service to the chess community.

What makes this mechanism so destructive is its asymmetry. The accused can't prove innocence: you can't demonstrate the absence of cheating, you can only note the absence of evidence of cheating. Meanwhile every good game he plays becomes suspect, including in his own eyes. He ends up playing against his own past.

This is a different situation from the Niemann affair, where the question of proof arose differently. Here, the adjudicating body characterized the behaviour itself.

2. The Coaching That Moves the Target

Far more common, far less visible, and with no body to adjudicate.

A normal coach, even a very harsh one, disputes your moves. He says: "this move is bad, here's the refutation." That's verifiable. You can open the engine, or bring a variation, or disagree. The discussion is about an object external to both of you.

The slide happens when the challenge changes target and lands on your mind:

"you never understood this structure";
"I've explained this to you already, you don't listen";
"you think you calculated, but you didn't calculate";
"everyone at the club can see this problem in you."

None of these sentences can be verified. None is about the position. They're about your memory, your attention, your understanding, your reputation. They are exactly the instruments with which you assess reality.

And on Sweet's analysis, their effectiveness doesn't come from their content: it comes from the position of the person saying them. The same reproach, from a player at your level, would be argued about. From someone who rates you, selects you or grades you, it settles in.

3. The Post-Mortem That Rewrites

The most ordinary form, often with no intent to harm.

Post-game analysis has a strange property: it takes place in a state of knowledge you didn't have during the game. Once you know where it turned, everything else looks like it led there. Move 20 "was already losing," when it was only losing in the light of move 34.

The winner, in perfectly good faith, tells the story of a game he controlled. He doesn't mention the four moments where he had no idea what to play, because he doesn't remember them either. His version is coherent, fluent, and false.

If you accept that version, the lesson you take from the game corresponds to nothing: that you were dominated from start to finish, when you may have been better on move 25. Repeated across a hundred games, this mechanism builds a self-image entirely disconnected from your results.

This isn't manipulation, it's a reconstruction bias. But the effect on you is identical, which is why it belongs here. The difference is in the intent, not the damage.

The Test: Gaslighting or Disagreement?

Three questions settle it fairly quickly.

Is the claim about a fact or about your perception? "This move loses a piece" is a fact, verifiable, arguable. "You never see anything" isn't a fact, it's a verdict on your perceptual apparatus.

Is it falsifiable? An honest criticism indicates what would contradict it. If no evidence could change your interlocutor's mind, it's no longer an analysis, it's a position.

What happens when you produce evidence? This is the decisive test. You open the game, you show the evaluation on move 20, it's level. An interlocutor in good faith updates his judgment. A gaslighting mechanism moves the target: "the evaluation means nothing," "that's not the point," "you can't take criticism." The evidence isn't discussed, it's bypassed, and you become the subject.

What Protects You, and Few Domains Have It

Here's the good news, and it's specific to this game.

You have the notation.

In almost every life situation where this mechanism operates, the victim has no record. She confronts the other person's memory with her own, and the other person's memory is more confident. In chess, you have a complete, timestamped record, external to both of you, that takes nobody's side.

Three concrete uses:

Notate your games, including online. Not just the moves, which the platform keeps: write one line at the end about what you thought of your position at the key moments and why you played what you played. It's the only record that will ever exist of your actual state of knowledge. Without it, the reconstructed version always wins.

Test claims against the archive, not against your memory. If you're told you were lost on move 20, don't search your recollection. Open the position. The answer exists.

Get analysis from an independent third party. Not an engine alone, which tells you that the move is bad without telling you what was humanly visible. A stronger player, outside the relationship. The method is detailed in our guide on how to analyze your games.

And for formal accusations: institutional channels exist for exactly this. Their function isn't bureaucratic, it's protective. They take the matter off the terrain where rumour beats evidence, and where, as FIDE noted, the harm occurs regardless of whether the facts are true.

What to Take Away

Gaslighting doesn't consist of lying about facts. It consists of discrediting the instrument with which you establish facts. That's why it works even in a domain where everything is archived: it simply shifts to what the archive doesn't contain, meaning your level, your understanding, your intentions.

Paige Sweet's research adds the decisive element: this mechanism only has force when carried by an asymmetry. It isn't a problem of character, it's a problem of position. A coach, a titled player, a club figure has a lever that a peer doesn't.

Your protection is concrete and you already have it: the written record. Chess is one of the rare places where you can answer "here's the position, let's look." You just have to look rather than remember.

After reading: go back to a game where you were told you'd been "lost for ages." Open it. Look at the evaluation at the move in question. You'll find out whether you accepted a version, or a fact.

This article is part of a series on psychology applied to chess. See also the Dunning-Kruger effect in chess. An upcoming article will cover emotional manipulation, which operates during the game rather than after it.

Sources

Sweet, P. L. (2019). The sociology of gaslighting. American Sociological Review, 84(5), 851-875.
Hamilton, P. (1938). Gas Light: A Victorian Thriller in Three Acts. Constable.
FIDE Ethics & Disciplinary Commission (2026). Decision in case 12-2025, 3 July 2026.
Fédération Internationale des Échecs. Anti-Cheating Regulations, FIDE Handbook.

Gaslighting in Chess: When Someone Rewrites Your Own Game