What makes reasoning go wrong — and how to find, name, and fix logical errors before your readers do
Every argument rests on a foundation of reasoning. When that foundation is solid — when claims follow from evidence, when conclusions are supported by relevant facts, when the logical steps are honest and complete — a reader has good reason to trust what the writer says. When the foundation is weak or hidden, the argument may still sound convincing, but it misleads. It takes the reader somewhere the evidence does not actually go.
This chapter is about learning to see beneath the surface of arguments, including your own. It introduces the principles of logical reasoning that make academic argument credible, and it identifies the most common reasoning errors — called logical fallacies — that undermine otherwise interesting claims. A fallacy is not simply a mistake of fact. It is a mistake in the structure of reasoning itself: a flaw in the way a writer moves from evidence to conclusion, or a maneuver that makes an argument seem stronger than it actually is.
By the end of this chapter, you will be able to recognize weak reasoning in what you read and, just as importantly, catch and correct it in your own writing. That ability is one of the most transferable skills academic writing teaches. It applies to every discipline, every professional field, and every conversation where honest, clear thinking matters.
Think of an argument you found persuasive recently — in an article, a conversation, or a speech. Now ask: Did the conclusion actually follow from the evidence? Were there hidden assumptions? Were any comparisons genuinely fair? By the end of this chapter, you will have a precise vocabulary for answering those questions — and for asking them about your own writing before your readers do.
To understand what makes reasoning go wrong, you first need a clear picture of what good reasoning looks like. Logical reasoning is the practice of drawing conclusions that are genuinely supported by the evidence and premises offered. In academic writing, this means that every step in your argument — from your opening claim to your final conclusion — should follow honestly and clearly from what came before it.
Academic argument is not simply a matter of stating a position and then filling pages with material that feels related to it. Strong academic argument is structured: a writer makes a claim, offers reasons for that claim, supports those reasons with evidence, and then explains how the evidence connects to the claim. Readers who follow this chain of reasoning should be able to see, step by step, why the conclusion follows from what was offered. If they cannot — if there are hidden jumps, missing steps, or maneuvers that substitute emotion or distraction for evidence — the argument is logically weak, regardless of how confidently it is written.
It helps to think of reasoning as a chain. Each link in the chain must hold. A claim like "Online education has expanded access to higher learning for working adults" needs to be supported by reasons: reasons like "working adults often cannot attend classes at fixed times" or "many working adults live far from college campuses." Those reasons, in turn, need to be supported by evidence — data, research, documented examples, or expert testimony. And the evidence needs to be explained in a way that shows it actually supports the claim, not merely that it exists in the same general neighborhood.
What logical reasoning requires is not genius or specialized training. It requires honesty, care, and discipline. It requires that writers ask themselves, at every step: Does this actually follow from what I just said? Am I leaving something out? Am I substituting something easier for the harder work of real evidence? Academic readers are trained to ask exactly those questions. Learning to ask them yourself — before your readers do — is the practical goal of this chapter.
Claim → Reasons → Evidence → Explanation of how evidence supports the claim
Every link must hold. A gap at any point — a jump from evidence to conclusion that is not explained — is where weak reasoning lives.
Every argument depends on things that are not always stated out loud. Two of the most important are assumptions and inferences. Understanding the difference between them, and learning when each creates problems, is foundational to everything else in this chapter.
An assumption is a belief that a writer takes for granted without stating or defending it. Assumptions are not always problems. All communication requires shared background assumptions to work at all. When you write for an academic audience, you can reasonably assume that your reader understands what a university is, what peer-reviewed research means, or what a thesis statement does. You do not need to define every term or defend every background belief.
The problem arises when a writer takes for granted something that the argument actually depends on — something that is contested or uncertain — without acknowledging or defending it. These are called hidden assumptions or unstated premises, and they can silently undermine an argument that looks strong on the surface.
"Students who take notes by hand learn more effectively than students who use laptops."
This claim might be supported by research. But it rests on at least one important assumption: that the type of note-taking device causes the difference in learning outcomes, rather than some other factor — such as differences between the kinds of students who tend to prefer one method over another, or the subject being studied, or the classroom environment. If that assumption is not examined and defended, the argument is incomplete.
Identifying hidden assumptions is a skill. Ask yourself: What would have to be true for this claim to hold? What is this argument taking for granted? When you find assumptions that are large, hidden, or contested, you need to either defend them or acknowledge that the argument rests on them.
An inference is a conclusion drawn from evidence. When you observe that a pattern exists in several cases and conclude that it probably holds more broadly, you are making an inference. Inference is essential to reasoning — without it, we could never move from particular facts to general conclusions. But inferences can be drawn too quickly, too broadly, or from insufficient evidence, which is where reasoning breaks down.
The key question to ask about any inference is: Does the evidence actually support this conclusion, or am I reaching further than the evidence allows?
"A student surveys twenty classmates and finds that fifteen of them feel anxious before major exams. From this, the student concludes: 'Most college students suffer from debilitating test anxiety.'"
The inference jumps from twenty students at one institution to all college students everywhere. It also slides from "feel anxious" — a normal and common experience — to "suffer from debilitating test anxiety," a much more serious clinical claim. Both jumps are unsupported by the evidence offered.
Most logical fallacies are, at their core, either hidden assumptions that should have been defended, or inferences that reach further than the evidence allows. Keeping this in mind as you study the fallacies below will help you understand why each one is a problem — not just as a named category to memorize, but as a recognizable pattern of reasoning that goes wrong in a specific way.
The hasty generalization is one of the most common reasoning errors in student writing, in journalism, in everyday conversation, and in political speech. It occurs when a writer draws a broad conclusion from too few examples, too narrow a sample, or too limited an experience. The name says it clearly: the generalization was made too quickly, before enough evidence was gathered to support it.
A hasty generalization follows a basic pattern: here are one, two, or a handful of cases; therefore, this is true of all (or most) cases. The problem is that a handful of cases — even striking, memorable, or emotionally powerful cases — do not constitute sufficient evidence for a sweeping general claim. Personal experience, while real and meaningful, does not automatically represent the experience of a larger population.
"I know three people who quit smoking without any professional help. Clearly, willpower alone is enough to overcome nicotine addiction."
Three cases are not sufficient to conclude that willpower alone generally overcomes a complex physical dependency. The three people the writer knows may be exceptional cases. There may be hundreds of others — invisible in this argument — who tried and could not succeed without support.
"The first two novels I read by this author were slow and difficult. She must be an inaccessible writer."
Two novels — and a subjective reading experience — do not establish a writer's entire body of work. The reader may have started with two unusual examples, or the judgment of "slow and difficult" may not be widely shared.
Academic writing deals with claims about the world that need to be reliable and defensible. If your argument rests on a generalization, your reader will want to know: How many cases? How representative were they? How was the sample selected? If you cannot answer those questions, you have likely made a hasty generalization.
The correction is almost always the same: qualify the claim honestly. Instead of writing "students who work part-time perform worse academically," write "some research suggests that students who work more than twenty hours per week may face academic challenges" — and then cite the research. The more carefully qualified version is not weaker. It is more honest and therefore more credible.
Students sometimes confuse a vivid example with proof. An example illustrates a point; it does not prove it. Always ask whether your examples are representative of a larger pattern before treating them as evidence of a general claim. One compelling story is not sufficient data for a sweeping conclusion.
These two fallacies are structurally different but equally common in academic and public argument. Both involve oversimplifying a complex situation in ways that make an argument seem more decisive than it actually is.
The false cause fallacy — sometimes identified by its Latin name, post hoc ergo propter hoc, meaning "after this, therefore because of this" — occurs when a writer concludes that because one event followed another, the first event caused the second. Timing and causation are not the same thing. The fact that B happened after A does not mean A caused B. Many events follow one another in sequence without any causal connection at all.
"After the new principal arrived, test scores at the school declined. Clearly, her leadership is responsible for the decline."
Test scores may have declined for many reasons: changes in student demographics, shifts in curriculum, testing format changes, broader social factors. Blaming the principal based on timing alone is a false cause argument.
"I started drinking green tea every morning, and my headaches went away. Green tea must cure headaches."
The headaches may have resolved for any number of reasons — stress reduction, better sleep, seasonal change, a different medication — that have nothing to do with green tea. Correlation does not establish causation.
In academic writing, false cause errors often appear when writers deal with social trends, health outcomes, educational results, or economic data. These are areas where many variables interact simultaneously, and isolating a single cause requires careful research design — controlled studies, comparison groups, statistical controls — not just observation of sequence. When you make a causal claim in your writing, ask yourself: Have I shown why A would produce B, or am I simply noting that B followed A?
The false dilemma — also called the either/or fallacy or black-and-white thinking — occurs when a writer presents a situation as if only two options exist, when in fact other options are possible. This fallacy works by artificially narrowing the choices available, which forces the reader to accept one of two presented options rather than considering the full range of possibilities.
"Either we increase funding for the military, or we leave our country defenseless."
This presents only two options — full funding increase or complete defenselessness — when many intermediate positions exist.
"You either support this policy or you don't care about children."
This forces the reader to choose between agreement and apparent moral failure, without acknowledging that thoughtful people may oppose a particular policy while deeply caring about children's welfare.
When you catch yourself using the word "either" in an argument, pause and ask: Are these really the only two options? Is there a middle ground or an alternative I'm not acknowledging? In argument essays, false dilemmas often appear when writers try to eliminate counterarguments too quickly. Academic argument needs to acknowledge that complex issues usually have more than two sides.
Both of these fallacies involve how a writer treats opposing viewpoints and the people who hold them. Both represent a failure of intellectual honesty — a willingness to fight easier battles rather than engage with the actual difficulty of the issue.
The straw man fallacy occurs when a writer misrepresents an opposing argument — usually by exaggerating, simplifying, or distorting it — and then attacks the distorted version rather than the actual argument. The name comes from the image of a fighter who, instead of sparring with a real opponent, builds a figure made of straw, attacks it easily, and then claims to have won the fight. Defeating an argument you invented is not the same as defeating the argument your opponent actually made.
Suppose an opponent argues: "We should require students to take at least one course in statistics to better understand data in the modern world."
A straw man response might be: "My opponent wants to force every student to become a mathematician. This is completely unreasonable."
The original argument was modest and specific. The straw man version inflated it into something much more extreme and easier to attack. No honest reader should be satisfied by a response that defeats a position the opponent never actually held.
In academic writing, straw man errors often appear in argument and counterargument sections. A student might summarize an opposing view so briefly and uncharitably that it looks obviously wrong, and then declare it defeated. This satisfies no careful reader. Academic argument requires representing opposing views fairly and at their strongest — a principle sometimes called the principle of charity.
The correction is simple in theory and challenging in practice: before you argue against a position, make sure you have understood it accurately. Quote it when possible. Summarize it carefully. Ask: Would someone who actually holds this view recognize my summary as fair?
The ad hominem fallacy — from the Latin for "to the person" — occurs when a writer attacks the person making an argument rather than engaging with the argument itself. It is a fallacy because the personal characteristics, history, motivations, or credentials of the speaker do not determine whether an argument is logically sound. A flawed person can make a valid argument. A credentialed expert can make a poor argument. The argument must always be evaluated on its own terms.
"We shouldn't trust Professor Lin's research on climate policy because she has been known to have strong political opinions."
Whatever Professor Lin's political opinions, this says nothing about whether her research methodology is sound or her data is accurate.
"Of course he supports higher taxes — he's never run a business in his life."
Whether or not someone has run a business may be relevant context in some discussions, but it does not determine whether an economic argument is logically valid.
Students sometimes confuse addressing a source's credibility — which is legitimate — with ad hominem. It is appropriate to note when a source lacks relevant expertise or has a significant conflict of interest. But this context should inform how the reader weighs the evidence, not serve as a substitute for engaging with the argument itself. Noting a conflict of interest is context; using it to dismiss an argument without engaging with its content is ad hominem.
The slippery slope fallacy occurs when a writer argues that one event or decision will inevitably lead to a series of increasingly extreme or negative consequences, without providing sufficient evidence for why each step in that chain would necessarily follow. Not every slippery slope argument is fallacious. Sometimes there is genuine, documented evidence that one change tends to lead to others — this can be a legitimate causal chain argument when the evidence supports it. The fallacy occurs when the causal chain is assumed rather than demonstrated: when the writer simply asserts that step A will lead to steps B, C, and D without explaining the mechanism or evidence that would produce each transition.
"If we allow students to retake one exam, soon they'll expect to retake every assignment, and eventually academic standards will collapse entirely."
This argument assumes that permitting one retake will automatically produce a cascade of ever-larger concessions. But each step in that chain requires its own evidence and explanation. The argument leaps from a specific, limited policy to total institutional failure without justifying the connections.
In academic writing, slippery slope arguments often appear in policy discussions, ethical debates, and technology arguments — areas where consequences are genuinely uncertain and writers may be tempted to project worst-case scenarios as inevitable outcomes. The correction is to either provide genuine evidence for the causal chain or qualify the claim honestly: "This policy could create pressures that, over time, might challenge academic standards in the following specific ways..."
Circular reasoning — sometimes called "begging the question" — occurs when the conclusion of an argument is simply assumed in one of the premises. The argument goes in a circle: A is true because of B, and B is true because of A. No independent evidence or reasoning actually supports the claim. The argument seems to offer support, but when you trace it carefully, it is simply restating what it set out to prove.
"This book is the most important ever written because it contains the most significant ideas ever expressed."
"Most important" and "most significant ideas" are essentially the same claim restated. The sentence offers no independent evidence for either.
"You should trust this website because it publishes only truthful information, and I know it publishes truthful information because it is a trustworthy website."
The claim (the website is trustworthy) is being used to prove itself. There is no independent evidence — just the claim circling back to support itself.
In academic argument, the evidence used to support a claim must be independent of the claim. If removing the claim would eliminate the "evidence," you have circular reasoning. Ask: What supports this conclusion that exists separately from the conclusion itself?
Circular reasoning is surprisingly easy to fall into because, from inside the argument, the logic can feel coherent. If you genuinely believe a claim, the reasoning that supports it can seem obvious to you — but that does not mean you have offered independent support. The correction is always to ask: What evidence, separate from the claim itself, supports this conclusion?
A particular challenge in academic writing is that data, statistics, and research findings can themselves be used in misleading ways — not because the numbers are fabricated, but because of how they are selected, framed, compared, or presented. Understanding how evidence can mislead helps writers both evaluate sources they read and use evidence honestly in their own writing.
Cherry-picking — also called selective evidence — occurs when a writer chooses only the evidence that supports a preferred conclusion while ignoring or downplaying evidence that complicates or contradicts it. The individual pieces of evidence may each be accurate, but the overall picture they create is distorted.
A writer arguing that a particular educational intervention works well might cite three studies showing positive results while not mentioning four studies that found no significant effect. Each cited study may be legitimate. The selection of only those three is the problem. The reader is given a systematically incomplete view of the evidence, which leads to an inflated sense of confidence in the conclusion.
In academic writing, honest use of evidence requires representing the state of the evidence fairly, including evidence that complicates your argument. If the research is mixed, say so, and then explain why you find the supporting evidence more persuasive. This acknowledgment does not weaken your argument; it demonstrates that you have engaged seriously with the full picture.
Statistical comparisons can mislead when the things being compared are not genuinely comparable — a problem sometimes called the apples-to-oranges comparison.
"Hospital A has a significantly higher patient mortality rate than Hospital B. Hospital A provides worse care."
This comparison may be entirely misleading if Hospital A specializes in treating critically ill patients while Hospital B primarily handles routine procedures. Comparing raw mortality rates between institutions with fundamentally different patient populations produces a number that tells you nothing meaningful about quality of care.
Academic writers need to ask, when using comparative data: Are these things genuinely comparable? What variables might explain the difference that I am not accounting for? The most careful writers specify the conditions under which a comparison holds and acknowledge when the comparison has important limits.
Both raw numbers and percentages can mislead when presented without context. A dramatic percentage increase sounds alarming but may involve very small underlying numbers. A large raw number may be insignificant when placed against a large population.
"Crime in this neighborhood increased by 100% last year."
This sounds alarming. But if the baseline was two incidents and the new total is four incidents, the percentage increase — while accurate — creates a very different impression than the raw numbers would.
"Country A reported ten thousand cases of this disease last year. Country B reported five thousand cases."
If Country A has a population of five hundred million and Country B has a population of ten million, Country B actually has a much higher rate of disease per capita. The raw numbers, without population context, are misleading.
This distinction is important enough to state carefully. Many studies find correlations — statistical relationships between two variables. Finding a correlation is meaningful and worth reporting. Concluding that the correlation proves causation is, in most cases, an overreach.
Many variables can produce correlations that have no causal relationship at all. Two things may rise and fall together because they are both driven by a third factor — this is called a confounding variable. Without carefully designed studies that control for alternative explanations, a correlation tells you that two things move together, not that one produces the other.
"Studies show that children who own more books perform better in school. Therefore, providing children with more books will improve their academic performance."
The correlation between book ownership and academic performance might reflect a causal relationship — or it might reflect other variables, such as family income, parental education level, or cultural attitudes toward reading, that produce both higher book ownership and better academic outcomes. The correlation alone cannot tell us. Researchers use careful study designs to try to move from correlation toward stronger causal inference. As a writer, be careful not to leap from "these two things are associated" to "one causes the other" without evidence of the causal mechanism.
| Type of Evidence Issue | What Goes Wrong | How to Correct It |
|---|---|---|
| Cherry-picking | Only favorable evidence is presented; contradictory evidence is ignored | Represent the full state of the evidence; acknowledge mixed findings |
| Apples-to-oranges comparison | Comparing things that differ in relevant ways; results are meaningless | Ensure comparable populations; specify conditions; acknowledge limits |
| Percentages without context | A dramatic percentage obscures a small baseline number | Provide both the percentage and the raw numbers with their base |
| Raw numbers without proportion | Large absolute numbers obscure small per-capita rates | Provide rates, proportions, or per-capita figures alongside raw totals |
| Correlation as causation | Two things that move together are assumed to cause each other | Name confounding variables; cite causal evidence if available; qualify |
Identifying fallacies in published arguments is a valuable skill. But the most practically important application of this chapter is learning to find and fix weak reasoning in your own drafts. This is harder than it sounds, because your own reasoning always feels coherent from the inside. When you write, you know what you mean, so the gaps in the argument are invisible to you. This is why revision matters — and why distance from your draft before revising helps you see it more clearly.
When revising an argument for logical strength, work through the following questions systematically. Do not rush. Read each major section of your draft with one of these questions in mind before moving to the next.
Generalizations: Is each broad claim based on sufficient evidence, or am I generalizing from too few examples?
Causal claims: When I argue that A caused B, have I shown the mechanism, or am I relying on sequence alone?
Either/or framing: Am I fairly representing the range of possible positions, or have I created a false dilemma?
Opposing arguments: Have I summarized opposing arguments fairly and at their strongest, or have I built a straw man?
Personal attacks: Have I addressed arguments, or have I commented on the people who hold them?
Consequence arguments: If I argue that something will lead to bad outcomes, have I shown why those outcomes would follow?
Circular passages: Is every major claim supported by evidence that is independent of the claim itself?
Evidence: Have I represented the evidence fairly? Are my comparisons genuinely comparable? Am I treating correlation as causation?
One of the most important insights this chapter can offer is that acknowledging the limits of your argument is not a sign of weakness. In academic writing, careful qualification — using language like "the evidence suggests," "in many cases," "under these specific conditions," or "one possible explanation" — signals that you understand your argument's scope and are not overclaiming.
Fallacies often arise from overclaiming: reaching further than the evidence allows, simplifying beyond what the situation supports, asserting causal relationships that have only been shown to be correlational. When you revise your reasoning, ask yourself not just Is this wrong? but Am I claiming more than I can actually defend?
The strongest academic arguments are often notable not for how sweeping they are, but for how precisely they define what they are and are not claiming — and how honestly they acknowledge what remains uncertain. Precision and qualification are signs of a writer who has thought carefully, not a writer who lacks confidence.
"Social media is destroying young people's ability to focus, and if we don't ban it from schools immediately, the next generation will be cognitively crippled."
"Several studies suggest a correlation between heavy social media use and reduced attention span in adolescents, though the causal mechanisms are still being investigated. Some researchers recommend limiting social media use during school hours as one possible intervention, though its effects would depend heavily on implementation."
The revised version does not simply back away from a claim. It replaces a collection of fallacies — hasty generalization, false cause, slippery slope — with an honest representation of what the evidence actually shows. For academic readers, the qualified version is more persuasive, not less, because it demonstrates intellectual honesty and familiarity with the actual state of the evidence.
The logical standards in this chapter apply equally across disciplines. A history student arguing that one event caused another, a biology student interpreting experimental results, an economics student reading policy research, and a literature student analyzing an author's argument are all doing the same fundamental work: checking whether conclusions follow honestly from evidence. The fallacy names may not appear in those disciplines' textbooks, but the underlying reasoning standards are universal.
Read each of the following short arguments. Identify the primary logical fallacy in each and explain why it qualifies as that fallacy.
For each paragraph below, identify the reasoning error or errors present, and then rewrite the paragraph to make the reasoning more logically sound. You may need to add qualifications, acknowledge complexity, or specify what additional evidence would be needed.
Identify the fallacies present. Rewrite this paragraph so the reasoning is honest and the claims are supported appropriately.
Identify the fallacies present. Rewrite this paragraph so the reasoning engages with the actual argument rather than attacking the person.
What reasoning error is present? What would a more careful version of this argument need to include?
Identify the circular reasoning in this paragraph. Rewrite it with a logically sound structure.
Choose one of the following general claims and write a paragraph identifying at least three significant assumptions the claim rests on. For each assumption, explain whether it would likely be accepted without defense by a general academic audience, or whether it needs to be acknowledged and defended.
Choose a topic relevant to education, technology, health, or your community. Write a one-page argumentative response that deliberately avoids all the fallacies covered in this chapter. When you have finished, complete the following self-evaluation, then exchange your draft with a classmate and complete the same evaluation for their argument.
Use this checklist when revising any argumentative writing for logical strength.
Chapter 29 gave you the tools to evaluate the logical structure of arguments — to distinguish valid reasoning from reasoning that only appears valid. Chapter 30 turns to a related and equally important question: how arguments sound to an academic audience. Academic tone is not just about being formal. It involves confidence without overstatement, caution and qualification, respectful disagreement, precision in word choice, and the ability to address controversial topics without inflammatory language. The logical standards of this chapter and the tonal standards of the next work together: a logically sound argument that is poorly toned will be dismissed; a well-toned argument built on fallacious reasoning will not survive close reading. Together, these two chapters complete Part V's treatment of academic argument.