1.3 Deductive reasoning

Deductive reasoning is how mathematicians move from things they already know (or assume) to conclusions that must follow. Unlike guessing or pattern-spotting, deduction gives you certainty: if your starting assumptions are true and your logic is correct, your conclusion is guaranteed.

This section covers the structure of deductive arguments, the most common argument forms, formal logical systems, how deduction is used in proofs, and where it can go wrong.

Foundations of deductive reasoning

Deductive reasoning starts with premises (statements you accept as true) and applies logical rules to arrive at a conclusion. The key feature is that the conclusion isn't just likely true; it's necessarily true if the premises hold.

This matters because mathematics doesn't tolerate "probably." A theorem is either proven or it isn't, and deduction is the tool that gets you from assumptions to proof.

Logic and validity

A deductive argument is valid when its conclusion follows logically from its premises, regardless of whether those premises are actually true. Validity is about structure, not content.

For example, this argument is valid even though it's absurd:

• All cats are made of glass.
• Whiskers is a cat.
• Therefore, Whiskers is made of glass.

The structure is airtight. The premises are false, but if they were true, the conclusion would have to be true. That distinction between truth (are the premises actually correct?) and validity (does the conclusion follow from the premises?) is one of the most important ideas in logic.

Premises and conclusions

• Premises are your starting points: definitions, axioms, or previously proven results.
• Conclusions are what you derive from those premises using logical rules.
• The strength of any deductive argument depends entirely on whether the premises are true and whether the reasoning connecting them to the conclusion is valid.

Always check the assumptions hiding inside your premises. A perfectly valid argument can still lead you astray if a premise is wrong or smuggles in an unstated assumption.

Syllogisms vs enthymemes

A syllogism is a deductive argument with exactly two premises and one conclusion:

1. Major premise (general rule): All even numbers are divisible by 2.
2. Minor premise (specific case): 14 is an even number.
3. Conclusion: Therefore, 14 is divisible by 2.

An enthymeme is a syllogism with one premise left unstated. For instance: "14 is even, so it's divisible by 2." The missing premise ("All even numbers are divisible by 2") is implied. Enthymemes show up constantly in everyday reasoning and even in math when a step feels "obvious." The risk is that the unstated premise might be wrong, and you won't catch it if you never make it explicit.

Types of deductive arguments

These are the core argument forms you'll use repeatedly. Each one has a specific structure, and recognizing them helps you both build and evaluate arguments.

Modus ponens

"Mode of affirming." This is the most straightforward deductive form:

1. If P, then Q.
2. P is true.
3. Therefore, Q is true.

Example: If a number is a multiple of 10, then it's a multiple of 5. The number 30 is a multiple of 10. Therefore, 30 is a multiple of 5.

Modus ponens is the workhorse of direct proofs. You establish a conditional ("if...then") and then show the "if" part holds.

Modus tollens

"Mode of denying the consequent." This one works backward from a false result:

1. If P, then Q.
2. Q is false.
3. Therefore, P is false.

Example: If $n^2$ is odd, then $n$ is odd. Suppose $n$ is not odd (i.e., $n$ is even). Then $n^2$ is not odd. This is the logic behind contrapositive proofs: instead of proving "if P then Q" directly, you prove "if not Q then not P," which is logically equivalent.

Hypothetical syllogism

This chains two conditional statements together:

1. If P, then Q.
2. If Q, then R.
3. Therefore, if P, then R.

Example: If a function is differentiable, then it's continuous. If a function is continuous, then it has the intermediate value property. Therefore, if a function is differentiable, it has the intermediate value property.

This form lets you build longer chains of reasoning, which is exactly what happens in multi-step proofs.

Disjunctive syllogism

This uses process of elimination:

1. Either P or Q.
2. Not P.
3. Therefore, Q.

Example: A number is either rational or irrational. $\sqrt{2}$ is not rational. Therefore, $\sqrt{2}$ is irrational.

This form works when you can establish that the options are exhaustive (they cover every possibility). If there's a third option you haven't considered, the argument breaks down.

Formal deductive systems

Propositional logic

Propositional logic deals with statements that are either true or false, connected by logical operators:

• AND ($\wedge$): Both statements must be true
• OR ($\vee$): At least one statement must be true
• NOT ($\neg$): Flips true to false and vice versa
• IF-THEN ($\rightarrow$): The conditional connective

Propositions are represented by variables like P, Q, and R. You can evaluate compound statements using truth tables, which systematically list every possible combination of truth values for the component propositions and show the resulting truth value of the whole expression.

Predicate logic

Predicate logic extends propositional logic by adding quantifiers and variables, which lets you express statements about collections of objects rather than just individual propositions.

• Universal quantifier ($\forall$): "For all" or "for every." Example: $\forall x, x + 0 = x$ means "for every number x, adding zero gives x."
• Existential quantifier ($\exists$): "There exists." Example: $\exists x, x^2 = 2$ means "there exists some x whose square is 2."

Predicate logic is essential for formalizing most mathematical statements, since math is almost always about properties that hold across entire sets of objects.

Axioms and rules of inference

• Axioms are statements accepted as true without proof. They're the starting points of any formal system. For example, Euclid's postulates in geometry.
• Rules of inference are the allowed logical moves: modus ponens, modus tollens, substitution, and others.

Together, axioms and rules of inference define what you can prove within a system. A formal proof is a sequence of statements where each one is either an axiom or follows from previous statements by a rule of inference.

Applications in mathematics

Proofs and theorems

Deductive reasoning is what makes a proof a proof. A theorem is a mathematical statement that has been established as true through a valid deductive argument from accepted axioms and previously proven results.

The main proof strategies include:

• Direct proof: Assume the premises, apply logical steps, arrive at the conclusion.
• Indirect proof (contrapositive): Prove "if not Q, then not P" instead of "if P, then Q."
• Proof by contradiction: Assume the opposite of what you want to prove, then show this leads to a contradiction.

Mathematical induction

Despite the name, mathematical induction is actually a deductive technique. It proves that a statement holds for all natural numbers using two steps:

1. Base case: Show the statement is true for the first value (usually $n = 1$ or $n = 0$).
2. Inductive step: Show that if the statement is true for some arbitrary $n = k$, then it must also be true for $n = k + 1$.

Once both steps are established, deduction guarantees the statement holds for every natural number. Think of it like dominoes: you prove the first one falls (base case) and that any falling domino knocks over the next one (inductive step).

Reductio ad absurdum

Latin for "reduction to absurdity." To prove a statement P:

1. Assume P is false (assume $\neg P$).
2. Use valid deductive steps to derive a contradiction.
3. Conclude that P must be true, since assuming otherwise leads to impossibility.

The classic example is the proof that $\sqrt{2}$ is irrational. You assume it is rational, write it as a fraction in lowest terms, and show through algebra that both the numerator and denominator must be even, contradicting the "lowest terms" assumption.

Common deductive fallacies

These are mistakes that look like valid deduction but aren't. Recognizing them is just as important as knowing the valid forms.

Affirming the consequent

This is the most common logical error with conditionals:

• If P, then Q. Q is true. Therefore, P is true. (Invalid.)

Example: If it's raining, the ground is wet. The ground is wet. Therefore, it's raining. But the ground could be wet from a sprinkler. The conditional "if P then Q" doesn't mean P is the only cause of Q.

Denying the antecedent

• If P, then Q. P is false. Therefore, Q is false. (Invalid.)

Example: If you study, you'll pass. You didn't study. Therefore, you won't pass. But you might pass anyway for other reasons. The conditional only tells you what happens when P is true; it says nothing about what happens when P is false.

False dichotomy

This fallacy presents only two options when more exist. "Either this proof uses induction or it's wrong" ignores direct proofs, contradiction, and other methods. In formal logic, disjunctive syllogism is valid only when the "either/or" genuinely covers all possibilities. A false dichotomy sneaks in by pretending it does when it doesn't.

Deduction vs induction

Certainty and probability

• Deduction provides certainty. If the premises are true and the argument is valid, the conclusion must be true.
• Induction provides probability. Observing a pattern many times makes a conclusion likely, but never guaranteed.

This is why mathematical proofs use deduction. "It worked for the first million cases" isn't a proof; there could always be a counterexample at case one million and one.

General to specific vs specific to general

• Deduction typically moves from general principles to specific cases. You know all primes greater than 2 are odd; 17 is prime and greater than 2; therefore 17 is odd.
• Induction moves from specific observations to general claims. You notice that 3, 5, 7, 11, 13 are all odd and conjecture that all primes greater than 2 are odd.

The inductive observation might suggest the rule, but only deduction can prove it.

Roles in scientific method

In science, induction and deduction work together in a cycle:

• Induction generates hypotheses from observed data.
• Deduction derives testable predictions from those hypotheses.
• Experiments then test the predictions, and the cycle continues.

Mathematics relies more heavily on deduction, but inductive reasoning (noticing patterns, forming conjectures) often guides mathematicians toward what to try proving.

Evaluating deductive arguments

Soundness vs validity

These two terms are easy to confuse, so keep them straight:

• Valid: The conclusion follows logically from the premises (the structure is correct).
• Sound: The argument is valid and all premises are actually true.

A sound argument guarantees a true conclusion. A valid-but-unsound argument has correct logic but at least one false premise, so the conclusion might be false.

Identifying hidden premises

Many arguments depend on assumptions that aren't stated explicitly. To evaluate an argument properly:

1. Write out each premise explicitly.
2. Ask: does the conclusion follow from just these premises, or is something missing?
3. If something is missing, state it. Then ask whether that hidden premise is actually true.

This skill is critical for spotting weak arguments that seem convincing on the surface.

Diagramming arguments

Drawing out the structure of an argument can make complex reasoning much easier to follow. A simple approach:

• Write each premise and the conclusion in separate boxes.
• Draw arrows showing which premises support which conclusions.
• Look for gaps (missing premises) or unsupported jumps.

This is especially useful for multi-step proofs where it's hard to keep track of how everything connects.

Historical perspectives

Aristotelian logic

Aristotle (4th century BCE) developed the first systematic study of deductive reasoning, centered on the syllogism. His framework dominated Western logic for nearly 2,000 years and established the idea that valid reasoning follows fixed, analyzable patterns.

Development of symbolic logic

In the 19th century, mathematicians began replacing natural language with formal symbols. George Boole developed Boolean algebra (the basis for modern digital logic), and Gottlob Frege created the first comprehensive system of predicate logic. This shift made it possible to analyze arguments with far greater precision than Aristotle's syllogistic framework allowed.

Modern formal systems

Today, logic extends well beyond classical propositional and predicate logic. Areas like modal logic (reasoning about possibility and necessity) and applications in computer science and artificial intelligence build on the deductive foundations covered here. The field continues to grow as new problems demand new logical tools.

Limitations of deductive reasoning

Gödel's incompleteness theorems

In 1931, Kurt Gödel proved two results that shook the foundations of mathematics:

1. First incompleteness theorem: Any consistent formal system powerful enough to express basic arithmetic contains true statements that cannot be proven within the system.
2. Second incompleteness theorem: Such a system cannot prove its own consistency.

These theorems don't make deduction useless, but they do show that no single formal system can capture all mathematical truth. There will always be limits to what deduction alone can reach.

Paradoxes in logic

Paradoxes reveal places where logical systems run into trouble:

• Russell's Paradox: Consider the set of all sets that don't contain themselves. Does it contain itself? Either answer leads to a contradiction. This paradox forced mathematicians to develop more careful foundations for set theory.
• The Liar's Paradox: "This statement is false." If it's true, then it's false; if it's false, then it's true.

These aren't just curiosities. They've driven the development of more rigorous logical frameworks.

Criticisms and alternatives

Deductive reasoning requires precise premises and clear-cut true/false values, which doesn't always match the messiness of real-world problems. Alternative approaches like fuzzy logic (which handles degrees of truth rather than strict true/false) and non-monotonic reasoning (where conclusions can be revised as new information arrives) address some of these limitations. In practice, strong reasoning often combines deductive, inductive, and other strategies.