Which fallacies are covered?

The common informal ones, each with its cue phrases: appeal to popularity, false dilemma, hasty generalisation, slippery slope, ad hominem, straw man, appeal to authority, circular reasoning and appeal to emotion. The full list is shown on the page.

Is a language model used?

No. There is no model and no network call. The detector is a fixed set of patterns, so the same passage always produces the same findings, and you can read exactly why each one was flagged.

Is the passage uploaded?

No. There is no account and no upload, and the text lives only in the tab where you pasted it, so unpublished work stays private.

Why does the tool explain each pattern?

Because the point is to learn to spot the pattern, not to accept a flag. Each finding carries a one-line description of what the fallacy does, so the cue and the reasoning behind it are visible together.

Logical Fallacy Detector

Q: Does it prove an argument is fallacious?

No. It matches a documented list of cue phrases and names the fallacy each usually signals. A flagged sentence may be perfectly sound, and an unsound argument may contain none of the cues, so treat a finding as a place to look rather than a verdict.

Your argument

Paste the passage to check — one argument per line

Findings

Possible issues

sentences with a cue phrase

Appeal to popularity
“Everyone knows”
Everyone knows the plan is wrong, so it must be.
Treats how many people believe something as evidence that it is true.
False dilemma
“Either we cut the budget or”
Either we cut the budget or the project fails completely.
Offers two options as if no others exist.
Slippery slope
“will lead to”
It will lead to chaos next year.
Asserts that one step inevitably leads to an extreme outcome.

Report

3 possible issue(s)

- Appeal to popularity ("Everyone knows"): Everyone knows the plan is wrong, so it must be.
  Treats how many people believe something as evidence that it is true.
- False dilemma ("Either we cut the budget or"): Either we cut the budget or the project fails completely.
  Offers two options as if no others exist.
- Slippery slope ("will lead to"): It will lead to chaos next year.
  Asserts that one step inevitably leads to an extreme outcome.

The cues this tool knows

Appeal to popularity — Treats how many people believe something as evidence that it is true.
False dilemma — Offers two options as if no others exist.
Hasty generalisation — Draws a sweeping conclusion from too little evidence.
Slippery slope — Asserts that one step inevitably leads to an extreme outcome.
Ad hominem — Attacks the person instead of the argument.
Straw man — Restates an opponent's position in a weaker form and attacks that.
Appeal to authority — Cites an authority without saying what they found or why it applies.
Circular reasoning — Uses the conclusion as its own support.
Appeal to emotion — Replaces an argument with a feeling.

Scope and limits

This is a lexical detector, not a logic engine. It matches a fixed list of cue phrases and reports what each phrase usually signals, so a flagged sentence may be perfectly sound and an unsound one may carry no cue at all.

The list is not exhaustive — every fallacy has forms without a stock phrase — and it makes no judgement about your conclusion. Treat a finding as a place to look, then decide for yourself whether the reasoning holds.

The logical fallacy detector scans a passage for a documented list of cue phrases and names the common fallacy each phrase usually signals — a prompt to look, not a verdict on the argument.

What is Logical Fallacy Detector?

The logical fallacy detector does one narrow thing and is honest about it. It holds a documented list of cue phrases for the common informal fallacies and reports where those phrases appear, naming the fallacy each one usually signals. An appeal to popularity is flagged by "everyone knows" or "most people agree"; a false dilemma by "either … or" or "only two options"; a slippery slope by "will lead to" or "next thing you know". The result is a list of sentences to look at again, each with the phrase that triggered it and a one-line explanation of the pattern.

That design is deliberate, because a logic engine is not what a writer needs at the desk. A text field cannot evaluate an argument's validity — that depends on meaning, on context and on premises the passage may not state — but it can reliably find the stock phrases that fallacies leave behind. Flagging those phrases puts the passages a reader might stumble over in one list, which is a useful first pass over a draft and, in a debate or an op-ed, a useful check before publishing.

The tool lists every rule it knows, with its cue phrases and a description, so the list is inspectable rather than mysterious. There is no hidden model and no network call: the same passage always produces the same findings, which makes the output something to reason about rather than trust blindly. A flagged sentence may be perfectly sound — "always" is sometimes the right word — and an unsound argument may carry no cue at all, and both of those limits are on the page.

Fallacies are worth knowing for more than editing. Naming the pattern — the straw man, the ad hominem, the appeal to emotion — is how you answer one, and seeing the list of cues is a compact way to learn to spot them. The detector is therefore a teaching aid as much as a proofreading pass: it shows what the phrases look like before they are pointed out to you in someone else's reply.

Nothing is uploaded and nothing is stored beyond the tab, so a passage stays private. The detection is lexical, the list is finite, and the judgement stays with the writer — which is the only honest way for a tool like this to work, and the page says so plainly.

A detector of cue phrases is not a substitute for reading closely, and it is worth saying what it is for. In an op-ed or a policy memo the phrases travel ahead of the argument: a reader spots "everyone knows" before they weigh the evidence, and the writer who removes it has to make the case instead. In a classroom the list doubles as a syllabus of the informal fallacies, since each rule names its pattern and shows the words that give it away. Either way the value is in the first pass, when a draft is long enough that the obvious weak spots are buried and a targeted list of them saves a slow, uneven read later.

How to use Logical Fallacy Detector

Paste the passage, one argument per line.
Read the findings: each names a fallacy, the cue phrase and the sentence it sat in.
Check the explanation of the pattern before deciding whether the passage is really affected.
Copy the report, or read the full list of cues the tool knows.

When to use Logical Fallacy Detector vs related tools

Use the logical fallacy detector as a first pass over an argument you are about to publish or submit. When you want to see the reasoning laid out rather than checked for cues, the Argument Mapping Tool shows the claim, its reasons and the warrants between them, and when the sentence itself needs sharpening the Thesis Statement Builder assembles and checks it.

If the passage is academic and the concern is register rather than logic, the Academic Writing Tone Checker measures that instead.

Privacy & Security

This tool runs entirely in your browser — no data ever leaves your device. There is no server round-trip, no upload, no logging, and no account required. Your input is processed locally using client-side JavaScript and is never stored, transmitted, or accessible to anyone else. When you close the tab, everything disappears.