A motte-and-bailey is a pair of claims advanced as if they were one: a strong, contested claim and a weak, obvious one. The arguer asserts the strong version in practice but defends only the weak version when challenged, then resumes the strong version once the challenge passes. Philosopher Nicholas Shackel named the structure in 2005. The non-obvious part is that he explicitly did not call it a fallacy, and that distinction is what most popular uses miss.
The castle metaphor
Shackel introduced the term in The Vacuity of Postmodernist Methodology (Metaphilosophy, 2005), one of five rhetorical maneuvers he catalogued alongside Troll’s Truisms, Equivocating Fulcra, the Postmodernist Fox Trot, and Rankly Relativising Fields. His target was Foucault, Latour, and postmodern epistemology.
A medieval motte-and-bailey castle has two parts. The bailey is the desirable land, a courtyard where you actually want to live, but it is hard to defend. The motte is a raised earthwork with a tower, miserable to occupy but easy to hold. When raiders come you retreat to the motte; when they leave you return to the bailey. The rhetorical version maps cleanly. The bailey is the big exciting claim you want to assert. The motte is the modest claim you fall back to when pressed. The dishonest move is treating a successful defense of the motte as if it had vindicated the bailey. The term sat mostly unused until Scott Alexander popularized it in 2014.
Doctrine, not fallacy
Shackel insisted that a motte-and-bailey is a doctrine, not a fallacy. A fallacy is a defective argument. A doctrine is a body of propositions. The problem is not a single bad inference but a dialectical pattern: holding two positions of very different strength and shuttling between them as the rhetorical situation demands.
This matters because it changes the fair-use test. The structure is not automatically illegitimate. Asserting both a strong and a weak claim is fine. The abuse appears only when someone defends the motte, declines to abandon the bailey, and then proceeds as though the bailey were what survived. Honest use looks different. Someone who claims a policy will “eliminate” a problem, concedes under challenge that it only reduces it, and then argues for the policy on the reduction alone has retracted, not retreated. They never return to “eliminate.” The structure becomes a motte-and-bailey only on the return trip. The maneuver overlaps with equivocation but is not identical: equivocation is sliding between two meanings of one word, while a motte-and-bailey is sliding between two distinct propositions of unequal defensibility. The shared ingredient is vagueness, which is why broadened, fuzzy concepts are prime material. See [private link] for how harm concepts inflate into exactly the kind of elastic terms a motte-and-bailey needs.
Worked examples
Shackel’s original case was postmodernism. Bailey: there is no objective truth, reality is a social construct. Motte: our theories are shaped by social and historical context. The motte is nearly uncontroversial; the bailey is the radical thesis that does the work, and it hides behind the motte under fire.
Alexander’s most-cited example is a strain of feminism. Bailey: a set of specific, contested laws and social norms one must endorse to count as a feminist. Motte: feminism just means women are people. Challenge the bailey and the response is “so you think women aren’t people?”
Religion supplies another. Bailey: an interventionist personal God who answers prayers. Motte: God is the order of the universe, or God is love. The two are defended interchangeably.
The overuse problem
The label is easy to weaponize, and Shackel’s own framing warns against it. Because identifying a motte-and-bailey requires showing that the same person or doctrine holds both claims and refuses to drop the bailey, you cannot fairly pin it on a movement just because different people say different things. Different people holding different views is not a motte-and-bailey.
Alexander notes the structure has a mirror image: the weak man, where you attack an unrepresentatively feeble version of an opponent’s position. A debate can deadlock when one side weak-mans and the other retreats to its motte. Calling “motte-and-bailey” is itself often a cheap move, a way to dismiss a position by asserting without evidence that its defender is secretly committed to a stronger claim. The honest use names the specific strong claim, the specific weak claim, and the moment of retreat. This is why [private link] treats the demand for a concrete referent as the practical defense: pin the claim to specifics and there is no room to shuttle.
Try it
Map a live motte-and-bailey (30-45 min, your feed plus a notes file). Pick three contested claims you have seen argued this week. For each, write the bailey (what the claim asserts in practice, the version that motivates action) and the motte (the version defended when someone pushes back). Then apply the distinguishing test: when challenged, does the speaker abandon the bailey or just shelter in the motte and return to the bailey later? You are looking for the return trip. A claim where the strong version never comes back was not a motte-and-bailey, it was an honest retraction. A claim that snaps back to the strong version once the heat is off is the real thing. The exercise also catches your own: the cases hardest to map are often the ones where you do it too.
See also
- [private link] — how concepts broaden into the elastic terms this maneuver exploits
- [private link] — demanding specifics as the defense
Sources
- The Vacuity of Postmodernist Methodology — Nicholas Shackel, the originating paper (Metaphilosophy, 2005)
- All In All, Another Brick In The Motte — Scott Alexander, the essay that popularized the term
- Motte and Bailey Doctrines — Shackel’s own follow-up clarifying doctrine vs fallacy