Skip to content

Informal Fallacy

Prime #
916
Origin domain
Philosophy
Subdomain
informal logic and argumentation → Philosophy
Also from
Rhetoric, Law & Governance, Political Science
Aliases
Informal Logical Fallacy, Material Fallacy

Core Idea

An informal fallacy is a named, recurring pattern of argument whose persuasive force exceeds its logical force because of a defect in content or context rather than in logical form. The conclusion does not actually follow from the premises, yet a relevance-substitution, an illicit reframing, or an unmet hidden premise makes it feel as though it does — the argument wears the costume of a good argument while failing as one. The defining contrast is with the formal fallacy: a formal fallacy is invalid by the shape of the inference alone — affirming the consequent, denying the antecedent — and can be detected by inspecting the argument's structure with the content blanked out. An informal fallacy can be valid or cogent in form yet still fail as reasoning, because the defect is in the material the form operates on: a premise that does not bear on the conclusion (irrelevance), a premise no better warranted than the conclusion (a question-begging or unacceptable premise), a term whose extension or meaning shifts under argumentative pressure, an ambiguity exploited, a context that makes a locally-reasonable move illegitimate. The error is substantive, not structural — which is exactly why blanking out the content, the move that catches formal fallacies, fails to catch these.

The structural commitments are four. First, there is an argument — premises advanced in support of a conclusion, by a reasoner, in a context. Second, the argument has a recurring, recognizable shape that earns a name (no-true-Scotsman, nirvana fallacy, straw man, ad hominem, equivocation, false dilemma, slippery slope, begging the question) — the fallacy is a type, not a one-off mistake. Third, the defect is material: it lies in the content of the premises, the relevance of premise to conclusion, the acceptability of what is assumed, or the context of the move, not in the formal validity of the inference. Fourth, the fallacy is normatively charged: to call a move an informal fallacy is to fault it as bad reasoning — the term carries a built-in negative evaluation, and the point of naming it is to license rejection of the argument (or demand its repair).

The single most consequential fact the prime names is the form-passes-content-fails structure: an informal fallacy is the class of argumentative failures that survive a validity check. This is what makes them slippery and what makes a catalogue of named types valuable — because the inference is often locally well-formed, the failure is not exposed by logic alone but only by interrogating the material: is this premise relevant? is it acceptable independently of the conclusion? does this term keep one meaning? is the comparison class legitimate? does the context license this move? What informal_fallacy provides as a prime is the recognition that there is a whole category of reasoning defect distinct from invalidity — located in content, relevance, acceptability, and context rather than in form — and that these defects recur in named, diagnosable types, each with a characteristic signature and a characteristic test. The prime is the genus; named fallacies like its children no_true_scotsman and nirvana_fallacy are its species, and what they share is precisely being content-located, recurring, named, and faulted defects of argument.

How would you explain it like I'm…

Wolf In Sheep's Clothing

Sometimes an argument SOUNDS right but is actually sneaky and wrong. It's like a wolf wearing sheep's clothing — it looks like a good reason, but if you check closely, the reason doesn't really fit. The trick isn't in how the words are lined up; it's that what they say doesn't truly back up the point.

Sneaky Bad Argument

An informal fallacy is a common, named kind of bad argument that FEELS more convincing than it really is — not because the steps are lined up wrong, but because of a problem in the actual content. For example, attacking the person instead of their point, or pretending someone said something silly so it's easy to knock down. These mistakes happen so often that people gave them names, like 'straw man' or 'false dilemma.' The sneaky part is that the argument can be shaped perfectly fine and STILL be bad, because the trouble is in WHAT is said, not in the shape. So to catch it, you have to ask whether the reasons are actually relevant and fair, not just whether the steps connect.

Form Passes, Content Fails

An informal fallacy is a named, recurring pattern of argument whose persuasive force exceeds its logical force because of a defect in CONTENT or CONTEXT rather than in logical form. The conclusion doesn't really follow, yet an irrelevant move, a sneaky reframing, or an unmet hidden assumption makes it FEEL like it does — the argument wears the costume of a good argument while failing as one. This contrasts with a FORMAL fallacy, which is invalid by the shape of the inference alone (like affirming the consequent) and can be caught by blanking out the content and checking the structure. An informal fallacy can be perfectly valid in form yet still fail, because the defect lives in the material the form operates on: an irrelevant premise, a question-begging premise no better supported than the conclusion, a term that shifts meaning, an exploited ambiguity, a context that makes a move illegitimate. The key idea is the 'form-passes, content-fails' structure — these are reasoning failures that SURVIVE a validity check — which is why a catalogue of named types (straw man, ad hominem, equivocation, false dilemma, slippery slope) is useful: you catch them by interrogating relevance, acceptability, meaning, and context, not by logic alone.

 

An informal fallacy is a named, recurring pattern of argument whose persuasive force exceeds its logical force because of a defect in content or context rather than in logical form. The conclusion does not actually follow from the premises, yet a relevance-substitution, an illicit reframing, or an unmet hidden premise makes it feel as though it does — the argument wears the costume of a good argument while failing as one. The defining contrast is with the formal fallacy: a formal fallacy is invalid by the shape of the inference alone (affirming the consequent, denying the antecedent) and can be detected by inspecting the argument's structure with the content blanked out. An informal fallacy can be valid or cogent in form yet still fail as reasoning, because the defect is in the material the form operates on — a premise that does not bear on the conclusion (irrelevance), a premise no better warranted than the conclusion (question-begging), a term whose meaning shifts under pressure (equivocation), an exploited ambiguity, a context that makes a locally-reasonable move illegitimate. The error is substantive, not structural, which is exactly why blanking out the content fails to catch it. Four commitments build the prime: there is an argument (premises advanced for a conclusion, by a reasoner, in a context); the argument has a recurring, recognizable shape that earns a name (no-true-Scotsman, nirvana fallacy, straw man, ad hominem, equivocation, false dilemma, slippery slope, begging the question) so the fallacy is a TYPE not a one-off; the defect is material (in content, relevance, acceptability, or context, not in formal validity); and the fallacy is normatively charged (to name one is to fault the argument as bad reasoning and license its rejection or repair). The single most consequential fact is the form-passes-content-fails structure: an informal fallacy is the class of argumentative failures that survive a validity check, which is what makes a catalogue of named types valuable — the failure is exposed not by logic but by interrogating the material. The prime is the genus; named fallacies like no_true_scotsman and nirvana_fallacy are its species, sharing the property of being content-located, recurring, named, and faulted defects of argument.

Structural Signature

the argument (premises advanced for a conclusion by a reasoner) — the recurring, nameable shape (the fallacy as a type) — the material location of the defect (content, relevance, acceptability, context — not form) — the persuasion–warrant gap (felt persuasive force exceeds actual logical force) — the form-passes-content-fails invariant (it survives a validity check) — the normative charge (naming it faults the argument) — the type-specific diagnostic test (each species has its own check)

An informal fallacy is present when each of the following holds:

  • An argument in context. Premises are advanced in support of a conclusion, by a reasoning agent, within a discourse — the move occurs in deliberation, not in an indifferent medium.
  • A recurring, nameable shape. The defect instantiates a recognizable type that recurs across arguments and earns a name (straw man, ad hominem, equivocation, false dilemma, begging the question, no-true-Scotsman, nirvana fallacy), rather than being a unique one-off error.
  • A material (non-formal) defect. The fault lies in the content — an irrelevant premise, an unacceptable or question-begging premise, a shifted term, an exploited ambiguity, an illegitimate comparison class, a context-violating move — not in the logical form of the inference.
  • The persuasion–warrant gap. The argument's felt persuasive force exceeds the logical force its content actually licenses: a relevance-substitution, an illicit reframing, or an unmet hidden premise makes the conclusion feel as though it follows when it does not. This gap is what makes the move effective and worth naming — a defect that fooled no one would not recur.
  • The form-passes-content-fails invariant (the load-bearing fact). The argument can be valid or cogent in form and still fail, because the defect survives a structure-only validity check; blanking out the content (which catches formal fallacies) does not catch it. This is the structural reason the persuasion–warrant gap can open undetected: the costume of validity is intact.
  • The normative charge. To classify the move as an informal fallacy is to fault it — the term carries a built-in negative evaluation, and the classification licenses rejecting the argument or demanding its repair.
  • The type-specific diagnostic test. Each named species comes with its own content-interrogating check — is the premise relevant? acceptable independently of the conclusion? does the term keep one meaning? is the comparison class legitimate? — distinct from the single validity check that suffices for formal fallacies.

The components compose into a single object — a recurring, named defect of argument located in content rather than form, faulted as bad reasoning, and detected by interrogating the material rather than checking validity — and it is the material-defect-that-survives-validity pairing that generates everything downstream: that informal fallacies are a category distinct from invalidity, that they require type-specific content checks rather than a single formal test, and that the genus organizes a family of named species each sharing this structure.

What It Is Not

  • Not a formal fallacy. A formal fallacy is invalid by logical form alone (affirming the consequent, undistributed middle), detectable by inspecting structure with content blanked out. An informal fallacy can be valid in form yet fail in content — the defect is material, and the structure-only check that catches formal fallacies misses it. The two partition the space of argumentative error by where the defect lives: form versus content.
  • Not a bias. A bias (the embedding-nearest neighbor) is a systematic tendency of a cognitive or estimation process to deviate — a property of the reasoner's machinery, present whether or not any argument is advanced. An informal fallacy is a defect of an argument as offered — a property of the public reasoning move, named and faulted in the argument itself. A bias may produce a fallacious argument, but the fallacy is the flawed argument, not the underlying disposition; one is diagnosed by examining the inference offered, the other by examining the estimation process.
  • Not a falsehood. A false premise makes an argument unsound but need not make it fallacious — a valid argument from a false premise commits no fallacy, it is simply unsound. An informal fallacy is a defect of the reasoning move (irrelevance, question-begging, equivocation, illegitimate comparison), which can occur with perfectly true premises. Falsity is about the premises' truth; fallacy is about the argument's legitimacy.
  • Not a rhetorical device as such. A rhetorical device (metaphor, repetition, appeal to emotion as ornament) is a means of persuasion that need not be fallacious; many are legitimate. An informal fallacy is specifically a defective argumentative move — persuasion that substitutes for or corrupts the inference rather than dressing a sound one. The overlap (an emotional appeal used in place of a relevant reason is the fallacy of irrelevant appeal) does not make every device a fallacy.
  • Not the falsifiability standard. falsifiability is a demarcation criterion for empirical content; some informal fallacies (definitional rescue) are ways of evading falsification, but the fallacy is the evasive move, not the standard it evades — the disease, not the test.
  • Common misclassification. Calling any argument one disagrees with, or any persuasive move, an informal fallacy. The signature requires a recurring, named, content-located defect that survives a validity check — not mere disagreement, not a false premise, not legitimate rhetoric. Catch it by asking whether the inference is well-formed yet still fails because of its material (irrelevant premise, shifted term, illegitimate comparison, question-begging assumption); if the form is the problem it is a formal fallacy, and if nothing is wrong but the conclusion is unwelcome it is no fallacy at all.

Broad Use

Informal fallacy, read as a named content-located defect of argument whose persuasive force outruns its logical force, recurs across every arena where reasoners advance premises for conclusions. In informal logic and argumentation theory, its home discipline, it is the organizing category of the entire field: the standard catalogues of fallacies of relevance (ad hominem, appeal to force, red herring), of ambiguity (equivocation, amphiboly), of presumption (begging the question, false dilemma, complex question), and of weak induction (hasty generalization, slippery slope, false cause) are all enumerations of informal-fallacy species. In rhetoric and public debate, the named fallacies are the working vocabulary for diagnosing bad public arguments in real time — the straw man that distorts an opponent's position to refute the distortion, the ad hominem that attacks the arguer instead of the argument, the false dilemma that hides the third option, the appeal to nature that treats "natural" as a covert premise for "good," the nirvana fallacy that faults a real policy against an unattainable ideal. In law and adjudication, rules of relevance and evidence are partly codifications of informal-fallacy avoidance: character evidence rules guard against ad hominem reasoning, the prohibition on assuming facts not in evidence guards against question-begging, and the recurring balancing of probative against prejudicial value is precisely the demand that an item's persuasive force not exceed its logical force — that the jury be moved by what the evidence actually warrants rather than by an irrelevant emotional or character-based substitution — while appellate briefs routinely allege an opponent's argument rests on an informal fallacy. In science and its philosophy, the ad-hoc rescue (the no-true-Scotsman move applied to a theory's auxiliary definitions) and the appeal to ignorance recur as named failures of scientific argument, and the modern methodological literature names HARKing (hypothesizing after the results are known) and the narrative around p-hacking as post-hoc rationalizations — a result fished from the data is dressed as a prediction it was reverse-engineered to fit, a content defect (the hypothesis is not independently warranted of the result) that no check of the statistical form exposes, and methodological norms like pre-registration exist partly to forbid it. In politics and policy, the slippery slope, the false dilemma, the appeal to consequences, the appeal to authority, and the nirvana fallacy are the recurring defective moves of advocacy, named precisely so they can be flagged. In advertising and marketing, the false dilemma ("either our product or failure"), the appeal to nature, the bandwagon appeal, and the borrowed-authority endorsement are stock persuasive devices whose force depends on an inference the audience does not actually have warrant to draw. In AI and large-language-model systems, the failure surfaces in a sharpened form: a model can generate a plausible-but-unsound argument — fluent, well-formed, rhetorically confident, and content-defective — that exhibits exactly the form-passes-content-fails structure (a fabricated-but-relevant-sounding premise, a smuggled false dilemma, a confident equivocation), so that the named-fallacy catalogue becomes a tool for auditing generated reasoning, and the genus names what an "argument that sounds right but isn't" actually is. And in everyday reasoning and education, the catalogue of named fallacies is taught as a portable toolkit for evaluating arguments in any domain. Across all of these the recurring fact is identical: an argument whose inference may be well-formed but whose material — a premise's relevance, a term's stability, a comparison's legitimacy, an assumption's acceptability — is defective in a recognizable, nameable way, so its felt persuasiveness exceeds its real warrant, and the recurring consequence is that the argument can be rejected (or required to be repaired) on grounds a validity check would never surface. The breadth is genuine but bounded entirely to discourse: every instance presupposes a reasoner advancing a claim, and there is no fallacy in a system that is not arguing.

Clarity

Naming informal fallacy as a category separates two questions that are routinely run together when an argument is criticized: is the inference valid? — the formal question, answerable by inspecting structure — and is the argument legitimate given its content and context? — the informal question, answerable only by interrogating the material. The first licenses the powerful, content-blind validity check; the second voids the assumption that passing that check is sufficient. The clarifying force of the prime is to make explicit that an argument can survive the validity check and still be bad, and that the badness lives in a named, diagnosable place — relevance, acceptability, term-stability, comparison-class, context — rather than in the form. This converts a vague "something is wrong with this argument even though it seems to follow" into a determinate search over the catalogue of content-located defects: which named fallacy, with which characteristic test, is at work?

The prime also clarifies the internal organization of argument criticism by supplying the genus under which the named species fall. Without the category, each named fallacy looks like an isolated rhetorical curiosity; with it, the no-true-Scotsman move, the nirvana fallacy, the straw man, the equivocation, and the false dilemma are recognizable as instances of one kind of thing — recurring, named, content-located, faulted defects of argument — each diagnosed by interrogating the material rather than the form. The clarifying move is to show that informal fallacy is not a grab-bag of unrelated tricks but a structured category defined by the form-passes-content-fails invariant, so that learning to spot one species (ask what membership criterion predated the counterexample, for no-true-Scotsman; ask what the real alternative is, for the nirvana fallacy) is learning to run the general discipline — find the material defect a validity check would miss — on any argument in any domain.

Manages Complexity

Informal fallacy compresses an unbounded space of "this argument is bad even though it seems to follow" into a catalogue of named types, each with a characteristic signature and a characteristic test. The complexity reduction is large because the prime replaces case-by-case adjudication of every suspect argument with pattern-recognition over a finite list: rather than reconstructing from scratch why each defective argument fails, the reasoner matches it to a named species and applies that species' test. A confusing family of argumentative failures collapses to a structured taxonomy — relevance fallacies, ambiguity fallacies, presumption fallacies, weak-induction fallacies — and within each, a small set of named patterns with known diagnostics. The act of naming is itself the compression: "no-true-Scotsman" packages an entire diagnostic (was the exclusion criterion specifiable before the counterexample, independently of the predicate?) into a single label that can be invoked in seconds.

The compression is also what makes argument criticism teachable and portable, because the genus supplies a single meta-procedure beneath the many species. The reasoner does not memorize each fallacy as an unrelated rule; they internalize the general move — run the validity check, and if it passes, interrogate the material (relevance, acceptability, term-stability, comparison-class, context) for a named defect — and then deploy the specific test of whichever species the material defect matches. This converts the open-ended problem of "how do I tell a good argument from a bad one when the logic seems fine?" into a bounded two-step: check the form, then check the content against a catalogue. And because the catalogue is shared across domains, a defect learned in one arena (the ad-hoc rescue in science) is recognized in another (the no-true-Scotsman in politics), so the complexity of evaluating arguments across many fields reduces to applying one portable taxonomy rather than learning each field's bad arguments separately.

Abstract Reasoning

The informal-fallacy pattern licenses several reasoning moves, though all of them operate within deliberation. Run the form check first, then the content check: faced with a suspect argument, the reasoner first asks whether it is invalid by form (a formal fallacy, caught by blanking the content) and, if it is well-formed, switches to interrogating the material — because the informal fallacies are exactly the defects that survive the validity check, and conflating the two stages misdirects the search. Locate the defect in the right material dimension: the reasoner asks which kind of content-failure is present — irrelevance (does the premise bear on the conclusion?), unacceptability (is the premise warranted independently of the conclusion?), ambiguity (does a term keep one meaning?), illegitimate presumption (is the comparison class or partition legitimate?), or context-violation (does the setting license this move?) — because each dimension corresponds to a different family of named fallacies with a different test. Match to a named species and apply its test: rather than re-deriving why the argument fails, the reasoner identifies the recurring type and runs its characteristic diagnostic, inheriting the accumulated discrimination the named test encodes. Distinguish the fallacy from mere unsoundness or unwelcomeness: the reasoner checks that the defect is genuinely in the reasoning move (not a false premise that leaves the argument valid-but-unsound, and not simply a conclusion they dislike), because the normative charge of "fallacy" is reserved for legitimacy-failures of the inference, not for falsity or disagreement. And wield the accusation responsibly: because naming a fallacy is itself a rhetorical move that can be deployed in bad faith, the reasoner requires that the charge be demonstrated — show the irrelevance, the shifted term, the illegitimate comparison, the question-begging assumption — rather than asserted, since an unprincipled "that's a fallacy" is itself a defective argumentative move.

Knowledge Transfer

Because informal fallacy is the genus of content-located, named, faulted argument-defects, a diagnostic learned for one species transfers to any other discourse by re-identifying the argument, the recurring shape, and the material dimension of the defect — and the prime's reach is the reach of that one category across deliberative arenas. The general meta-procedure transfers intact: run the validity check, and if the inference is well-formed, interrogate the material for a named defect and apply its test. A reasoner who has learned to spot the ad-hominem irrelevance in a political debate deploys the identical move — "does this premise bear on the conclusion, or does it attack the arguer?" — in a scientific dispute, a legal argument, or a boardroom disagreement, because the relevance dimension and its test do not name a domain. The specific named tests transfer the same way: the no-true-Scotsman test (was the exclusion criterion specifiable before the counterexample, independently of the predicate?) carries unchanged from theology to software-engineering debate to political identity argument; the nirvana-fallacy test (what is the best feasible alternative, and does the faulting property discriminate among feasible options?) carries from medicine to economics to engineering to personal decision-making; the equivocation test (does the disputed term keep one meaning across the premises?) carries everywhere a term is used. In every transfer the practitioner runs the same diagnosis — identify the argument, confirm the inference is well-formed (else it is a formal fallacy), locate the material defect, match it to a named species, and run that species' content test — and the transfer is secure because none of these steps names a substrate within discourse: a logician, a litigator, a scientist defending a theory, a debate coach, and an editor checking an op-ed are all running the same content-defect-detection procedure, distinguished only by the subject matter the arguments are about. What does not transfer is any application beyond deliberation: there is no informal fallacy in a physical, biological, or formal system that is not arguing, because the category is constitutively about a reasoner advancing premises for a conclusion — the genus is portable across every arena of human reasoning and across none outside it.

Examples

Formal/abstract

The contrast between a formal and an informal fallacy, set side by side, is the prime in its purest worked form, because it isolates exactly the form-passes-content-fails invariant. Take first a formal fallacy — affirming the consequent: from "if \(P\) then \(Q\)" and "\(Q\)," infer "\(P\)." Blank out the content — replace \(P\) and \(Q\) with schematic letters — and the defect is visible in the structure alone: the inference pattern is invalid for any substitution, so a validity check on the form catches it with no appeal to what \(P\) and \(Q\) mean. Now take an informal fallacy — the no-true-Scotsman move, a child of this prime. Begin with a contentful universal "all \(X\) are \(Y\)," meet a counterexample (an \(X\) that is not \(Y\)), and redefine \(X\) as "true \(X\)" to exclude it, so the claim becomes "all (\(X\) and \(Y\)) are \(Y\)." Run the validity check on this inference and it passes: "all things that are both \(X\) and \(Y\) are \(Y\)" is a tautology, valid under every interpretation. The defect is not in the form — the form is impeccable — but in the material: the subject term's extension was redefined post hoc, reverse-engineered from the embarrassing case, so the universal was preserved at the cost of becoming empirically empty, its membership now inspecifiable independently of the predicate. This is the invariant exactly: the formal fallacy fails the structure check; the informal fallacy survives it and fails only when the content is interrogated (was the exclusion criterion specifiable before the counterexample, independently of \(Y\)?). The same split distinguishes a denying-the-antecedent error (formal, caught by blanking content) from a nirvana fallacy (informal: the argmax over the feasible set is well-formed, but an out-of-set ideal was admitted as the benchmark — a content defect a validity check never sees). The structural payoff is that informal fallacies are precisely the argumentative failures located outside the form, which is why a single validity check cannot exhaust argument criticism and a catalogue of content-interrogating tests is required.

Mapped back: The formal/informal contrast instantiates every component — an argument advanced for a conclusion, a recurring nameable shape (affirming-the-consequent versus no-true-Scotsman), a defect that is structural in the formal case and material in the informal one, the form-passes-content-fails invariant (the no-true-Scotsman inference is a valid tautology), the normative charge (both are faulted), and the type-specific test (validity-check versus the in-advance-criterion test) — and shows the prime's core claim that informal fallacies are the category of reasoning defect that survives a validity check.

Applied/industry

Three high-stakes arguments from three different professions run the genus across three different species, and reading each through the prime locates the defect exactly where a validity check would miss it. In a courtroom, the prosecution argues that the defendant is guilty of fraud, and offers as a premise that he was twice convicted of unrelated assaults years earlier; the inference "he is a person of bad character, therefore he committed this fraud" is locally well-formed yet rests on a premise that does not bear on whether this act occurred — the ad hominem / character species, a relevance defect, and it is exactly the case the rules of evidence target when they exclude testimony whose prejudicial force (it inflames the jury) exceeds its probative force (it warrants the conclusion). The prime's discipline exposes it: the form passes, so interrogate the material — does this premise raise the probability of the charged act, or does it merely substitute hostility for relevance? In a research manuscript, an author reports that a sub-group effect was "predicted" and is therefore strong confirmation, when the hypothesis was in fact selected after inspecting the data that displays it — HARKing, a post-hoc rationalization that is the presumption/acceptability species: the inference "the data confirm a prediction" is well-formed, but the premise "this was a prediction" is unwarranted independently of the result it was reverse-engineered to fit, and no audit of the statistical form (the p-value is real) surfaces the defect; only interrogating the provenance of the hypothesis does, which is why pre-registration — fixing the hypothesis in advance — is the structural corrective. In an AI deployment, a language model is asked whether a policy is sound and returns a fluent, confident argument that turns on the claim "every credible economist agrees," a fabricated-but-relevant-sounding appeal to authority whose persuasive force (it sounds authoritative and well-formed) far exceeds its logical force (the premise is both false and, even if true, non-probative) — a plausible-but-unsound generated argument that exhibits the genus's signature exactly, and that the named-fallacy catalogue is the right instrument to audit. All three are informal fallacies — every inference is locally well-formed, and each fails only when the content is interrogated — and the genus supplies the one unifying discipline: run the validity check (all three pass), then interrogate the material (an irrelevant character premise, an unwarranted predicted-this claim, a non-probative authority appeal), match each to its named species, and apply that species' test. None re-uses the machinery of the children, yet each is the same kind of defect the children are: a well-formed inference whose felt persuasiveness outruns its warrant because of a fault in content or context.

Mapped back: The courtroom, the manuscript, and the AI output each run the prime end-to-end — arguments advanced for conclusions, three distinct recurring nameable shapes (ad hominem, HARKing-as-post-hoc-rationalization, appeal to authority), material defects (irrelevance, unwarranted-prediction presumption, non-probative ground) that survive a validity check, the normative charge (each rejected or flagged as bad reasoning), and three type-specific tests (probative-vs-prejudicial, provenance-of-hypothesis, is-the-authority-relevant-and-real) — and demonstrate the transfer: a litigator, a methodologist, and an AI auditor are running the same content-defect-detection procedure, distinguished only by what the arguments are about.

Structural Tensions

T1 — Form versus Content (the Locating-the-Defect Tension). The prime's foundational tension is that argumentative error can live in two distinct places — the logical form (formal fallacy) or the material (informal fallacy) — and the diagnostics differ: the form-defect is caught by blanking the content and running a validity check, while the content-defect survives that check and is caught only by interrogating relevance, acceptability, term-stability, comparison-class, or context. The failure mode is wrong-check application: running a validity check on a content-located defect (and pronouncing the argument sound because it passes), or hunting for a material defect when the inference is simply invalid in form. Diagnostic: ask whether the defect remains when the content is replaced by schematic letters; if it disappears the fault is formal, if it persists only when the actual content is read the fault is informal, and only then does the catalogue of content tests apply.

T2 — Named Fallacy versus Legitimate Move (the False-Accusation Tension). Each named fallacy has a legitimate look-alike — the no-true-Scotsman resembles genuine essentialist refinement, the slippery slope resembles a valid causal chain, the appeal to authority resembles legitimate reliance on expertise, the ad hominem resembles a relevant credibility challenge. The tension is between flagging the fallacy and recognizing the sound move it mimics. The failure mode is over-application: branding every definitional refinement a no-true-Scotsman, every causal-chain argument a slippery slope, every expert citation an appeal to authority — weaponizing the fallacy label to dismiss legitimate reasoning. Diagnostic: require the specific condition that distinguishes the species from its legitimate twin (for no-true-Scotsman, that the criterion was post-hoc and inspecifiable independently of the predicate; for slippery slope, that the intermediate links are unsupported); absent that demonstrated condition, the accusation is itself an unprincipled rhetorical move.

T3 — Fallacy versus Unsoundness versus Disagreement (the Scope Tension). The normative charge "fallacy" is reserved for a legitimacy-failure of the reasoning move, but it is constantly stretched to cover a false premise (which leaves a valid argument merely unsound) or a disliked conclusion (which is no defect at all). The tension is between the precise category and its loose colloquial use. The failure mode is charge inflation: calling any unsound or unwelcome argument "a fallacy," which both misdiagnoses (a false premise needs a truth challenge, not a fallacy charge) and devalues the term. Diagnostic: ask whether the inference move is defective independently of the premises' truth; if the argument is valid and the only issue is a false premise, it is unsound not fallacious, and if nothing is wrong but the conclusion is unwelcome, no fallacy is present at all.

T4 — Genus versus Species (the Granularity Tension). Informal fallacy is the genus; the named fallacies are its species, and there is a standing tension between reasoning at the general level (interrogate the material for some content defect) and the specific level (apply this named test). The failure mode is granularity mismatch: stopping at "this is some informal fallacy" without identifying the species (so no specific test is applied and the diagnosis is unactionable), or forcing a defect into a familiar named species it does not quite fit (mislabeling a relevance failure as equivocation). Diagnostic: identify the material dimension of the defect first (relevance, acceptability, ambiguity, presumption, induction), then the species within it, and confirm the species' defining condition actually holds rather than reaching for the most familiar label.

T5 — Detecting the Fallacy versus Weaponizing the Catalogue (the Dual-Use Tension). The catalogue of named fallacies is a powerful detection tool, but invoking a fallacy name is itself an argumentative move that can be deployed in bad faith — "that's a straw man," "that's whataboutism," "that's a slippery slope" can shut down legitimate argument as easily as expose defective one. The tension is between the catalogue as a diagnostic and the catalogue as a rhetorical weapon. The failure mode is accusation-as-argument: asserting a fallacy label in place of demonstrating the defect, so the naming substitutes for the analysis. Diagnostic: require the accuser to exhibit the content defect the name denotes (show the distorted position for straw man, the shifted term for equivocation, the unsupported links for slippery slope); an undemonstrated fallacy charge is an informal fallacy in its own right — an irrelevant or question-begging move dressed as criticism.

T6 — Argumentation-Bound Category versus Substrate-Neutral Echo (the Scope-of-Transfer Tension). Stated abstractly, several informal fallacies have a bare structural echo (the no-true-Scotsman as a classifier tightening its boundary post-hoc; the nirvana fallacy as an argmax over the wrong comparison set) that surfaces in adaptive systems and optimization without any reasoner. But the category is constitutively about a deliberating agent advancing a claim, and its normative "bad reasoning" charge requires deliberation the bare structure lacks. The failure mode is over-transfer: importing the fallacy framing into a substrate where no one is arguing (faulting an overfit model or an evolutionary process for "committing a fallacy"), or missing the structural echo because it lacks rhetorical dress. Diagnostic: confirm a claim-defending agent is present before applying the fallacy framing; the bare structure may recur, but informal fallacy proper requires an argument, so the genus travels across every arena of discourse and none outside it.

Structural–Framed Character

Informal fallacy sits at the framed pole of the structural–framed spectrum, with a frontmatter aggregate of 1.0 — every diagnostic reads maximally framed. It is a named informal-logic category that carries a normative "bad reasoning" charge and is bound to argumentative practice; the prose should not pretend a substrate-neutral structure is doing the load-bearing work, because the prime is the genus of content-located argument-defects, and an argument-defect presupposes an argument.

Every diagnostic points to the framed pole. Vocabulary travels (1.0): the prime is stated entirely in the language of argument — premises, conclusion, relevance, acceptability, validity, the comparison class — and even its formal restatements (a no-true-Scotsman collapsing into a tautology, a nirvana fallacy as an argmax over the wrong set) presuppose a claimant advancing a generalization, with no meaning in a substrate that is not reasoning. Evaluative weight (1.0): "fallacy" is the name of an error — to classify a move as an informal fallacy is to fault it as bad reasoning — so the disapproving charge is maximal and constitutive, and the whole point of the category is to license rejection or demand repair. Institutional origin (1.0): its home is the discipline of informal logic and argumentation theory, and it cannot be stated without that discourse and its taxonomy of relevance, ambiguity, presumption, and weak-induction fallacies. Human-practice-bound (1.0): the category requires a deliberating agent who advances premises for a conclusion in a context — a system that is not arguing cannot commit an informal fallacy, and the normative charge requires exactly the deliberation the bare structural echoes (an overfit classifier, a mis-targeted optimizer) lack. Import-versus-recognize (1.0): invoking the prime imports the entire interpretive frame of argument-criticism — the apparatus of relevance, acceptability, comparison-class, and the named-fallacy catalogue — rather than recognizing a pattern wired into an indifferent medium.

The honest reading is that several individual informal fallacies have a content-neutral skeleton with faint echoes outside deliberation (its children's substrate-independence grades reach a 2–3 on the strength of those echoes), but the genus — a named, faulted, content-located defect of argument — is fully framed, because the very thing that makes a move an informal fallacy is that it is a defective argument, faulted as such. The substrate-independence grade is correspondingly low (a 2): every instance is human deliberation, with no physical, biological, or formal substrate. The 1.0 aggregate is fully justified, and the prose should keep the prime squarely in the argumentation-bound region where its content lives — more framed than its child nirvana_fallacy (0.6, whose argmax-over-the-wrong-set core is substrate-neutral) and matching its child no_true_scotsman (1.0, the paradigm framed fallacy), because as a category the prime carries the full framing of the discipline that defines it.

Substrate Independence

Informal fallacy is a low-substrate-independence prime — composite 2 / 5 on the substrate-independence scale. The category — a named, recurring defect of argument located in content, relevance, acceptability, or context rather than in logical form — is fundamentally an argumentation category, and outside the practice of a reasoner advancing a claim it does not apply, which holds the composite near the floor. On domain breadth (3) the category does recur across many arenas, but every one is deliberative: informal logic and argumentation theory (its home, where the standard fallacy taxonomy lives), rhetoric and public debate (the working vocabulary for diagnosing bad public arguments), law and adjudication (relevance and evidence rules as codified fallacy-avoidance), science and its philosophy (ad-hoc rescue, appeal to ignorance as named failures of scientific argument), politics and policy (slippery slope, false dilemma, nirvana fallacy as recurring advocacy defects), and everyday reasoning and education (the named-fallacy toolkit taught for evaluating any argument). The breadth is genuine but it is breadth across substrates of discourse, with no physical, biological, or formal instance. On structural abstraction (3) the genus — a content-located defect that survives a validity check, recurring in named types — is statable somewhat abstractly, but it presupposes a reasoning agent who advances premises for a conclusion, so it cannot be lifted out of deliberation the way a conservation law or a feedback loop can. On transfer evidence (2) the carry is real but lives entirely inside reasoning: the general meta-procedure (run the validity check, then interrogate the material) and the named tests (the in-advance-criterion test, the best-feasible-alternative test, the term-stability test) port across theology, science, law, and politics, but the transfer is among deliberative arenas rather than across genuinely unlike media — it is the same argumentative discipline re-applied, not a structure recurring in indifferent substrates. Consistent with its fully framed character (a named informal-logic category carrying a normative "bad reasoning" charge and bound to argumentative practice), the honest grade is low: this prime lives almost entirely inside human reasoning.

  • Composite substrate independence — 2 / 5
  • Domain breadth — 3 / 5
  • Structural abstraction — 3 / 5
  • Transfer evidence — 2 / 5

Relationships to Other Primes

One-hop neighborhood: parents above, mutual partners to the right, children below.Informal Fallacysubsumption: Nirvana FallacyNirvana Fallacysubsumption: No True ScotsmanNo True Scotsman

Foundational — no parent edges in the catalog.

Children (2) — more specific cases that build on this

  • Nirvana Fallacy is a kind of Informal Fallacy

    child of emergent informal_fallacy

  • No True Scotsman is a kind of Informal Fallacy

    child of emergent informal_fallacy

Neighborhood in Abstraction Space

Informal Fallacy sits in a sparse region of abstraction space (80th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely rather than landing on a neighbor.

Family — Inference & Evidence (26 primes)

Nearest neighbors

Computed from structural-signature embeddings · 2026-06-14

Not to Be Confused With

The most important confusion is with the formal fallacy, because the two together partition the space of argumentative error and the distinction is the prime's defining contrast. A formal fallacy is invalid by logical form alone — its defect survives blanking out the content and is caught by a structure-only validity check (affirming the consequent is invalid for every substitution). An informal fallacy is valid or cogent in form yet defective in content — its fault lives in relevance, acceptability, term-stability, comparison-class, or context, and survives the validity check, so the content-blind test that catches formal fallacies misses it entirely. The distinction is load-bearing because the diagnostics differ completely: the formal fallacy is found by inspecting structure, the informal fallacy by interrogating material. A reasoner who applies only the validity check pronounces every well-formed argument sound and misses the entire category of content-located defects; a reasoner who hunts for a material defect in a form-invalid argument mislocates a structural fault. The cleanest tell: replace the content with schematic letters — if the defect vanishes it was formal, if it persists only when the actual content is read it was informal.

A second genuine confusion is with bias, the embedding-nearest neighbor, because both name reasoning gone wrong and both come with a catalogue of named types. But they are different kinds of object. A bias is a systematic tendency of a cognitive or estimation process — a property of the reasoner's machinery (confirmation bias, anchoring, availability) that operates whether or not any argument is advanced, diagnosed by examining the process. An informal fallacy is a defect of an argument as offered — a property of the public reasoning move (the irrelevant premise, the shifted term, the illegitimate comparison), named and faulted in the argument itself, diagnosed by examining the inference. The relationship is causal but not identity: a bias may produce a fallacious argument (confirmation bias may generate a cherry-picking fallacy), but the fallacy is the flawed argument, not the underlying disposition, and the same fallacy can be committed with no bias at all (a deliberate, cynical straw man). Confusing them misroutes the remedy: a bias is addressed by debiasing the process (structured procedures, disconfirmation prompts), a fallacy by criticizing the argument (exposing the content defect and demanding repair). A practitioner who treats a fallacy as a bias looks to fix the arguer's mind when the fix is to fault the argument, and vice versa.

A third confusion is with mere unsoundness — an argument with a false premise. It is tempting to call any bad argument a fallacy, but a valid argument from a false premise commits no fallacy: it is simply unsound, its inference impeccable and only its premise untrue. An informal fallacy is a defect of the reasoning move itself — irrelevance, question-begging, equivocation, illegitimate comparison — which can occur even when every premise is true. The distinction is consequential because the corrective differs: unsoundness is repaired by challenging the truth of the premise (or replacing it), while a fallacy is repaired by fixing the move (supplying a relevant premise, stabilizing the term, correcting the comparison class). Conflating them sends a reasoner to dispute a premise's truth when the actual fault is the inference's legitimacy, or to allege a fallacy when the inference is fine and only a premise is false.

The remaining confusions run the other way — not with neighbors the prime must be distinguished from but with its own species, which must not be mistaken for the genus itself. Each named fallacy is an informal fallacy plus a specific extra commitment that fixes which material dimension fails and which diagnostic test applies; the genus is what is left when those specifics are removed, and any one species is recoverable by re-adding its commitment. no_true_scotsman is the genus restricted to the presumption/acceptability dimension with one added mechanism: a universal claim meets a counterexample and the subject term's extension is tightened post hoc to exclude it, so the universal survives by becoming a content-empty tautology. Strip those specifics — the universal, the counterexample, the extension-rigging — and what remains is the bare genus (a content-located defect surviving a validity check, here the valid tautology "all things that are both X and Y are Y"); re-add them and the species reappears, with its own test (was the exclusion criterion specifiable in advance, independently of the predicate?). It is not the genus, because most informal fallacies involve no universal claim and no extension-tightening at all. nirvana_fallacy is the genus restricted to comparison-class selection with a different added mechanism: a real option drawn from a feasible set is faulted against an idealized counterfactual admitted from outside that set, so a property absent from every feasible alternative is used to reject one of them. Strip those specifics — the feasible set, the out-of-set ideal, the discrimination invariant — and again only the bare genus remains; re-add them and the species reappears, with its own test (what is the best feasible alternative, and does the faulting property discriminate among the options actually available?). It too is not the genus, because most informal fallacies involve no comparison class and no idealized benchmark. The same relation holds for the wider family: a straw man adds the commitment that the opponent's position has been distorted before refutation (an irrelevance/relevance defect); an ad hominem adds that the arguer rather than the argument is attacked (irrelevance); an equivocation adds that one term shifts meaning across premises (ambiguity); a false dilemma adds that a partition is presented as exhaustive when a third option is suppressed (presumption); a slippery slope adds that intermediate causal links are asserted without support (weak induction); an appeal to authority/nature/ignorance adds that a non-probative ground (status, naturalness, absence of disproof) is substituted for a relevant reason (relevance); begging the question adds that a premise presupposes the conclusion (acceptability). In every case the species is the genus plus its distinguishing commitment, recoverable as a special case, and the genus is strictly more general than any of them — which is exactly why no single named fallacy, the children included, can stand in for the parent.

For a practitioner these distinctions decide both diagnosis and remedy. Mistaking an informal fallacy for a formal one applies a structure check to a content-located defect and pronounces a bad argument sound. Mistaking it for a bias looks to debias a mind when the fix is to fault an argument. Mistaking it for mere unsoundness disputes a premise's truth when the fault is the inference's legitimacy. And mistaking the genus for one of its species — reaching for "no-true-Scotsman" or "nirvana fallacy" when the defect is some other content failure — forces the wrong test onto the argument and misses the actual fault. The prime earns its place as the genus of content-located, named, faulted argument-defects whose persuasive force exceeds their logical force and which survive a validity check — distinct from the formal invalidity it is partitioned against, the cognitive tendency it is confused with, and the false premise it is mistaken for, and strictly more general than the named species (its children among them) that instantiate it — and its unifying discipline is to run the validity check first, then, if the form passes, interrogate the material against a catalogue of named species, each with its own content-interrogating test.

Solution Archetypes

No catalogued solution archetypes reference this prime yet.