Logic And Critical Thinking

This deck walks you through the building blocks of logic and critical thinking, starting with the basics of what makes an argument and moving through key ideas like validity, soundness, and the difference between deductive and inductive reasoning. You'll also get familiar with helpful tools such as syllogisms, truth tables, and the main logical operators that show up everywhere from philosophy to everyday reasoning.

It's a great fit if you're just starting out with formal logic, taking an introductory philosophy or critical thinking course, or simply want to sharpen how you think through questions and arguments. Even if you've never written a proof or opened a logic textbook before, the cards are designed to introduce each idea in small, manageable pieces so you can build confidence as you go.

Because many of the later concepts rest on the earlier ones, it's worth working through the deck in order the first time, then coming back later for spaced review. A helpful habit while studying is to try making up your own quick example for each new term, since logic really clicks when you see how it applies to statements you come up with yourself.

Formal logic is the study of inference conducted with purely formal content, where the validity of an argument depends solely on its logical form rather than on the meaning of its terms. At its core, logic is concerned with arguments, which are sets of statements (called premises) offered as reasons to support another statement (called the conclusion). When evaluating arguments, two central concepts apply: an argument is valid when, if all its premises are true, the conclusion must necessarily be true, and it is sound when it is both valid and has all true premises. Soundness is the only condition that guarantees a true conclusion.

A key distinction in logic is between deductive and inductive reasoning. Deductive reasoning moves from general premises to a specific conclusion that follows with certainty, so that given true premises the conclusion cannot be false. In contrast, inductive reasoning moves from specific observations to a general conclusion that is merely probable. A classic deductive example runs: all mammals are warm-blooded, dogs are mammals, therefore dogs are warm-blooded. An inductive counterpart would be: every swan I have observed is white, therefore all swans are probably white. Because of this difference, deductive arguments are judged valid or invalid in an all-or-nothing sense, while inductive arguments are judged strong or weak depending on how probable their conclusion is given the premises.

A syllogism is a classical form of deductive reasoning consisting of two premises and a conclusion, typically organized around a major premise, a minor premise, and a final claim. In a categorical syllogism, three terms interact. The major term is the predicate of the conclusion, the minor term is the subject of the conclusion, and the middle term appears in both premises but not in the conclusion, serving as the bridge that connects the other two. For an example, "all mammals are warm-blooded; dogs are mammals; therefore dogs are warm-blooded" uses "warm-blooded" as the major term, "dogs" as the minor term, and "mammals" as the middle term.

Categorical reasoning also operates with statements that assert relationships between two categories, expressed in four standard forms: "All S are P," "No S are P," "Some S are P," and "Some S are not P." The logical relationships among these forms are mapped by the square of opposition, a diagram that organizes how categorical propositions with the same subject and predicate relate through contradictory, contrary, subcontrary, and subalternation relations. Together, syllogisms and the categorical framework provide the historical foundation of deductive logic.

Propositional logic operates on whole propositions—that is, statements that are either true or false—and combines them using logical connectives. A truth table is a mathematical tool that determines the truth value of a compound proposition for every possible combination of truth values of its component statements, making explicit exactly when each connective yields a true or false result.

Conjunction (\(P \land Q\)) is true only when both P and Q are true. Disjunction (\(P \lor Q\)) is true when at least one of P or Q is true, and false only when both are false. Negation (\(\neg P\)) reverses the truth value of a proposition: if P is true, \(\neg P\) is false, and vice versa. The conditional or material implication (\(P \to Q\)) is false only when P is true and Q is false, and true in every other case. The biconditional (\(P \leftrightarrow Q\)) is true exactly when both propositions share the same truth value, whether both true or both false.

Compound statements can also be classified by their overall behavior. A tautology is true under every assignment of truth values to its variables, as with \(P \lor \neg P\). A contradiction is false under every assignment, as with \(P \land \neg P\). A contingency is neither a tautology nor a contradiction; its truth value depends on the values of its variables. These three categories cover every possible compound proposition in propositional logic.

Propositional logic supplies a set of rules of inference—reliable patterns for deriving conclusions from premises. Modus ponens reasons from \(P \to Q\) and P to Q. Modus tollens reasons from \(P \to Q\) and \(\neg Q\) to \(\neg P\). Hypothetical syllogism chains conditionals: \(P \to Q\) and \(Q \to R\) together yield \(P \to R\). Disjunctive syllogism resolves alternatives: \(P \lor Q\) and \(\neg P\) together yield Q. These patterns preserve validity, while faulty counterparts such as affirming the consequent (from \(P \to Q\) and Q inferring P) or denying the antecedent (from \(P \to Q\) and \(\neg P\) inferring \(\neg Q\)) do not.

Several classical laws govern how propositions behave. The law of excluded middle holds that for any proposition P, either P or \(\neg P\) is true; there is no middle ground, expressed as \(P \lor \neg P\). The law of non-contradiction states that a proposition cannot be both true and false at the same time and in the same sense, expressed as \(\neg(P \land \neg P)\). The law of identity holds that every object is identical to itself, formalized as \(P \equiv P\).

De Morgan's laws describe how negation distributes over conjunction and disjunction: \(\neg(P \land Q) \equiv (\neg P \lor \neg Q)\) and \(\neg(P \lor Q) \equiv (\neg P \land \neg Q)\). Working with conditionals requires care, since the contrapositive of \(P \to Q\), namely \(\neg Q \to \neg P\), is logically equivalent to the original, but the converse (\(Q \to P\)) and the inverse (\(\neg P \to \neg Q\)) are not. More generally, two statements are logically equivalent when they share the same truth value in every possible scenario, as with \(P \to Q\) and \(\neg P \lor Q\).

Fallacies are errors in reasoning that undermine arguments, and they come in two broad kinds. Formal fallacies arise from invalid logical form; informal fallacies arise from problems of content, context, or delivery. Two classic formal fallacies involve conditional arguments: affirming the consequent (if P then Q, Q, therefore P) and denying the antecedent (if P then Q, not P, therefore not Q). In both cases the conclusion does not follow, because Q could be true for reasons other than P, and Q might still be true even when P is false. Other formally invalid moves—those whose conclusions simply do not follow from their premises—are called non sequiturs.

Informal fallacies are far more numerous. Ad hominem attacks the person rather than the argument; straw man misrepresents an opponent's position to refute a distorted version; slippery slope asserts that one event will inevitably trigger a chain of extreme consequences without justifying each link. Several appeals lean on factors other than evidence: appeal to authority (argumentum ad verecundiam) leans on a supposed expert, especially one outside the relevant field; appeal to ignorance (argumentum ad ignorantiam) treats absence of disproof as proof; appeal to emotion manipulates feelings; appeal to nature conflates naturalness with goodness; and the bandwagon fallacy (argumentum ad populum) treats widespread belief as evidence of truth. Genetic fallacy judges claims by their origin; false dilemma presents only two options when more exist; red herring diverts attention with an irrelevant topic; tu quoque deflects criticism by accusing the accuser; circular reasoning assumes the conclusion in its premises (begging the question); and equivocation exploits ambiguous terms.

Hasty generalization leaps from a small or unrepresentative sample to a broad claim; post hoc ergo propter hoc assumes sequence implies causation; the no true Scotsman fallacy excludes counterexamples by redefining the group; loaded question smuggles in an unproven presupposition; composition transfers properties from parts to whole, while division transfers them from whole to parts. The middle ground fallacy (argument to moderation) assumes the truth is a compromise between positions; special pleading applies standards inconsistently to exempt oneself or one's case; the Texas sharpshooter cherry-picks data clusters after the fact; and whataboutism shifts attention to a different issue entirely. Recognizing these patterns is essential for any critical thinker, since fallacious reasoning can be persuasive even when its conclusions are unjustified.

The scientific method provides a systematic framework for testing claims about the world: observation leads to questions, which generate hypotheses, which are tested by experiment, with the resulting data analyzed to draw conclusions. A hypothesis is a testable, falsifiable prediction about the relationship between variables, formulated before experimentation. Karl Popper emphasized falsifiability—the requirement that a statement admit some possible observation or experiment that could prove it false—as the defining feature that separates science from non-science. Across this process, the burden of proof falls on the claimant: the person making an assertion must provide evidence, rather than placing the obligation on others to disprove it.

Experiments gain their power through control. A controlled experiment manipulates one variable (the independent variable) while holding others constant, comparing results against a control group to isolate causal effects. Statistical inference relies on the null hypothesis, the default assumption of no effect or no relationship between variables, with experiments aiming either to reject it or to fail to reject it. Rejecting a true null yields a Type I error, or false positive, while failing to reject a false null yields a Type II error, or false negative. In interpreting results, it is essential to distinguish correlation—two variables tending to occur together—from causation, where one variable directly produces the other. Occam's razor guides scientists toward the simplest explanation that accounts for the evidence, reducing unnecessary assumptions ("Do not multiply entities beyond necessity").

Scientific claims are further strengthened by community practices such as peer review, where independent experts evaluate methodology, accuracy, and significance before publication, and reproducibility, the requirement that experiments be repeatable by others with the same results. Together, these features—falsifiability, controlled experimentation, null hypothesis testing, peer review, reproducibility, and Occam's razor—form a self-correcting system that distinguishes empirical inquiry from mere speculation.

Human reasoning is systematically shaped by cognitive biases, mental shortcuts, and predictable errors that can distort even careful thinking. Confirmation bias leads us to search for, interpret, and remember information that confirms preexisting beliefs while discounting contrary evidence. Anchoring bias makes us rely too heavily on the first piece of information we encounter. The availability heuristic causes us to judge the likelihood of events by how easily examples come to mind, often overestimating vivid or recent occurrences. The Dunning–Kruger effect describes how people with low ability tend to overestimate their competence, while highly skilled individuals may underestimate theirs.

Other biases shape how we update beliefs and make decisions. Cognitive dissonance produces discomfort when we hold contradictory beliefs, often prompting rationalization rather than revision. Hindsight bias makes past events seem more predictable than they were—the so-called "I knew it all along" effect. The framing effect shows that people react differently depending on how information is presented, even when the underlying facts are identical, such as "90% survival rate" versus "10% mortality rate." Groupthink occurs when the desire for harmony in a group overrides realistic appraisal of alternatives. The sunk cost fallacy persuades us to continue failing endeavors because of past investment rather than current value, and motivated reasoning pushes us toward conclusions we want to be true.

Probability judgments are particularly vulnerable to predictable errors. The gambler's fallacy wrongly assumes that past random events affect future ones—for instance, believing that a coin is "due" for heads after a streak of tails—while in reality each independent trial is unaffected by previous outcomes. The base rate fallacy ignores how common an event actually is in the general population, focusing instead on less informative specific evidence. Bayesian reasoning counteracts some of these errors by systematically updating the probability of a hypothesis in light of new evidence, using Bayes' theorem \(P(H \mid E) = \frac{P(E \mid H) \times P(H)}{P(E)}\). Finally, the self-fulfilling prophecy illustrates how expectations can themselves alter outcomes, since believing a prediction often changes the behavior that brings it about.

Predicate logic, also called first-order logic, extends propositional logic with quantifiers and predicates, allowing statements about properties of objects and relations among them. The universal quantifier \(\forall x\, P(x)\) states that every element in a domain satisfies the predicate P, while the existential quantifier \(\exists x\, P(x)\) claims that at least one element does. This added expressiveness enables the analysis of arguments that propositional logic alone cannot capture, such as "all mammals are warm-blooded" or "some philosophers are empiricists." Together with categorical propositions and the square of opposition, predicate logic forms the modern backbone of formal reasoning.

Several advanced tools deepen our ability to evaluate arguments. Abductive reasoning, often called inference to the best explanation, starts from observations and selects the simplest, most likely explanation; it differs from inductive reasoning, which generalizes from specific instances to broader rules. Reductio ad absurdum proves a proposition false by assuming it true and showing the assumption leads to a contradiction or absurd consequence. Analogies argue that because two things are similar in some respects, they are likely similar in a further respect, with strength depending on the relevance of the similarities invoked. A counterexample is a single specific case that disproves a universal claim.

Understanding necessary and sufficient conditions clarifies how requirements relate to outcomes: a necessary condition must be present for an event to occur, while a sufficient condition guarantees it. For instance, being a mammal is necessary for being a dog, and being a dog is sufficient for being a mammal. The principle of charity requires interpreting an argument in its strongest, most reasonable form before critique, rather than attacking a weaker version. An enthymeme is an argument with an implicit premise that the audience is expected to supply, such as "she's a doctor, so she's smart," whose hidden premise is that doctors are smart. Finally, the paradox of the heap (the sorites paradox) shows how vague predicates resist sharp boundaries, illustrating that questions like "when does a collection of grains stop being a heap?" cannot always be answered with classical categories.