Why Arguments Go Wrong

What is happening under the surface

Arguments go wrong because the brain treats "feels right" as "is right." The feeling of conviction is generated by cognitive shortcuts, not by the strength of the evidence. Fallacies are those shortcuts: they win the social game of an argument while skipping the logical work that would actually settle it.

When a statement is easy to process — familiar, repeated, and consistent with what you already think — it feels more true. This is processing fluency, and it is the engine of the bandwagon fallacy. A claim echoed by many people becomes easier to process, so it feels truer. Disliking a speaker makes their claim feel suspicious, and that feeling, not any flaw in the argument, powers the ad hominem attack. A distorted summary of an opponent's position is easier to attack than the real position, so the straw man offers the comfort of a clear victory without ever confronting the actual view.

A worked example

Suppose 100 people in a village say the spring water at the edge of town cures headaches. That is 100 testimonials, so intuitively the water works. But the number of believers is a fact about the villagers, not about the molecules in the water. If a laboratory measures the water and finds it chemically identical to ordinary tap water, then its real cure rate is whatever tap water has. The 100 people's belief does not move the measured effect from, say, 10% to 90%. Popularity is a property of the crowd, not of the claim.

The common misunderstanding

It is easy to assume these patterns are always fallacious. They are not. An attack on the speaker is legitimate when the immediate question is about the speaker's reliability. "You were paid by the company whose product you are recommending" is not an ad hominem; it is evidence of a conflict of interest. Similarly, "everybody knows" is not a fallacy if the claim really is about what people believe. Saying "this is the most downloaded app on the store" and citing the download count is a factual claim about popularity, not a bandwagon appeal. And a straw man is only a straw man when it distorts. Accurately summarizing an opponent's position, even unflatteringly, is not a straw man.

When naming fallacies does not help

Naming a fallacy is only useful when the shared goal is to decide what is true. In a negotiation about a raise, "I deserve more because I have worked here longer" is an appeal to fairness, not a logical proof, and calling it a fallacy would be a conversational error. In a personal argument about how someone made you feel, feelings are data, not invalid reasoning to be dismissed. Logic is a tool for deciding what to believe. When the real question is what to do, whom to trust, or how to repair a relationship, a fallacy diagnosis should yield to the actual human problem.