12.9 Logical Reasoning: Proof

How mathematical statements are justified. Includes axioms and postulates accepted without proof, definitions, theorems and what proving one requires, conjectures and why examples never suffice, disproving a general statement with a single counterexample, and indirect proof by assuming the opposite and reaching a contradiction.

Subject: Algebra 1 · 65 slides · symbolic lesson

Open the interactive version of this deck

What this lesson covers

The lesson, slide by slide

1. Lesson 12.9 Logical Reasoning: Proof

Title

Algebra 1 · Chapter 12 — Radicals and More Connections to Geometry

Logical Reasoning: Proof

2. By the end of this lesson you can

Objectives

Five outcomes, each one you can test yourself on with a pencil and no answer key.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 740-745 — the lesson these objectives are drawn from

3. What you already have

Warm-up

This whole book has used rules such as the distributive property. This lesson asks where those rules came from and how anything is ever established.

Discussion prompt

Why is it true that a times the quantity b plus c equals a b plus a c? Can you prove it, or is it assumed?

Hint: Try to derive it from something more basic.

Answer:

It cannot be derived from anything simpler within algebra — it is an axiom, one of a small number of properties assumed at the outset. Everything else in the book was built on such assumptions.

That is not a weakness. A system of reasoning has to start somewhere, and being explicit about where is what distinguishes mathematics from a collection of rules that happen to work.

4. Assumed, defined, or proved

Concept

A postulate or axiom is a property accepted without proof. A definition names a new idea in terms of existing ones. A theorem is a statement that must be proved before it may be used.

theorem — A statement that can be proven to be true. Once proved, a theorem may be used as a reason in the proofs of other theorems.

Every step of a proof must cite an axiom, a definition, given information, or a theorem already proved.

Figure (svg): How the parts of a mathematical system rest on each other

Everything in a proof traces back to something assumed or already established. That is what stops a chain of reasoning from going round in circles.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 740-741

5. Axioms, definitions and theorems

Section

Section 1

6. Three kinds of statement

Concept

Axioms are assumed, definitions introduce vocabulary, and theorems are earned. Knowing which is which tells you what may be used freely and what must be established first.

The basic axioms of algebra are the rules of Chapter 2.

  1. Axioms and postulates are accepted without proof.
  2. Definitions do not need proving; they only name things.
  3. Theorems must be proved before use.

Figure (svg): How the parts of a mathematical system rest on each other

Everything in a proof traces back to something assumed or already established. That is what stops a chain of reasoning from going round in circles.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 740-741 — the sections on axioms, definitions and theorems, and the Basic Axioms of Algebra summary

7. What rests on what

Picture it

Assumptions at the bottom.

Figure (svg): How the parts of a mathematical system rest on each other

Everything in a proof traces back to something assumed or already established. That is what stops a chain of reasoning from going round in circles.

The arrangement is not arbitrary. A proof may cite anything below it in the picture and nothing above, which is what keeps the reasoning from becoming circular.

8. Worked example: sort the familiar rules

Worked example

Placing Chapter 2's properties in the structure.

\[ \text{Are the commutative property, the definition of an integer, and } c(-b) = -cb \text{ axioms, definitions or theorems?} \]

Take the commutative property

Why: Assumed at the outset.

Take the definition of an integer

Why: Names a set using existing terms.

Take the third

Why: Provable from the axioms.

Note the consequence

Why: Only the last needed proving.

Figure (svg): How the parts of a mathematical system rest on each other

Everything in a proof traces back to something assumed or already established. That is what stops a chain of reasoning from going round in circles.

\[ \text{axiom}, \; \text{definition}, \; \text{theorem} \]

Verify: ask what would happen without each

Why: Removing the axiom would leave nothing to reason from; removing the definition would leave a word undefined; removing the theorem would only mean it had to be reproved when needed. That difference in consequence is what the three categories capture.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 740-741

9. Assumed, defined, or proved?

Sorting

Three categories.

Sort into buckets

Sort each statement by its status.

Axiom
a + b = b + a; a(b + c) = ab + ac
Definition
an integer is a whole number or its opposite; a midpoint is the point equidistant from both ends, on the segment
Theorem
c(a - b) = ca - cb; the Pythagorean theorem
ax
It is one of the basic properties assumed at the outset without proof.
def
It introduces a term by describing it with words already available.
th
It is a claim that can be, and must be, derived from the axioms.

Two of each, and the distinction is not about difficulty. The commutative property is simpler than most theorems and is nonetheless assumed rather than proved.

10. Worked example: why axioms are not proved

Worked example

A question worth asking once.

\[ \text{Why is the distributive property not proved?} \]

Ask what it would be proved from

Why: Something more basic.

Ask about that statement

Why: It would need proving too.

Note the difficulty

Why: The chain cannot go on forever.

Resolve it

Why: Some statements are assumed.

Figure (svg): How the parts of a mathematical system rest on each other

Everything in a proof traces back to something assumed or already established. That is what stops a chain of reasoning from going round in circles.

\[ \text{axioms end the regress} \]

Verify: consider the alternative

Why: If every statement required proof from an earlier one, no proof could ever be completed, since there would always be one more step behind. Choosing a small set of starting assumptions is what makes the whole structure possible.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 740-740

11. Trap: treating a theorem as obvious enough to assume

Trap

The trap

The statement is clearly true, so it can be used without proof.

Skip the proof for something that seems evident

Why: Obviousness felt like sufficient justification.

Many false statements seem evident and many true ones do not. Whether something needs proving is settled by its logical status, not by how convincing it looks.

The fix

Prove it from the axioms, then use it freely afterwards.

Establish it once, then cite it

Why: That is what theorems are for.

Goldbach's conjecture has seemed evident for nearly three hundred years and remains unproved.

12. What may a proof cite?

Faded example

Four kinds of reason.

Fill in the blanks

Every step of a proof must be justified by an axiom, a definition, given information, or a previously proved theorem.

Why: Those four sources are exhaustive, which is what makes a proof checkable by someone else. Anything cited that is not one of them is an unjustified assumption.

13. Which needs no proof?

Elimination

By its logical status.

Eliminate the wrong options

Which kind of statement is accepted without proof?

  • A. An axiom
  • B. A theorem
  • C. A conjecture
  • D. Any statement that seems obvious

Survives elimination: A

Why: Axioms are assumed deliberately and definitions need no proof because they only assign names. Conjectures are the interesting case: unproved and, unlike axioms, not accepted either.

14. Why be explicit about assumptions?

Socratic

The rules work either way.

Discussion prompt

Say what is gained by listing the axioms of algebra explicitly. Then say what the Greek geometers changed about mathematics.

Hint: Think about what a reader can check.

Answer:

Listing them makes every later claim checkable: a reader can follow a proof and verify that each step rests on something agreed. Without an explicit starting point there is no way to distinguish a proof from a persuasive argument, and no way to settle a disagreement about whether something has been established.

What the Greek geometers changed, about two and a half thousand years ago, was to stop treating mathematics as a collection of practical rules that worked and start demanding that each result be derived from a small set of stated assumptions. That shift is why mathematical results stay true rather than merely holding up so far, and it is the reason this lesson closes the book.

15. Proving a theorem

Section

Section 2

16. Every step needs a reason

Concept

A proof is a chain of statements in which each one is justified by an axiom, a definition, given information, or a theorem already proved. The justifications are as much a part of it as the algebra.

\[ c(a - b) = ca - cb \]

Once proved, a theorem may be cited in later proofs.

Figure (svg): A short proof with every step justified

The right-hand column is what makes this a proof rather than a calculation. Each line is licensed by something already accepted.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 741-741 — Example 1, Prove a Theorem, and its Study Tip on justifying every step

17. Statements and reasons

Picture it

Two columns, both required.

Figure (svg): A short proof with every step justified

The right-hand column is what makes this a proof rather than a calculation. Each line is licensed by something already accepted.

Without the right-hand column this is only a calculation. The reasons are what turn a sequence of true statements into an argument that establishes something.

18. Worked example: prove a distributive theorem

Worked example

This is Example 1 from the textbook.

\[ \text{Using } a - b = a + (-b), \text{ prove that } c(a - b) = ca - cb. \]

Rewrite the subtraction

Why: By the subtraction property.

\[ c [a + (-b)] \]

Distribute

Why: By the distributive axiom.

\[ c a + c(-b) \]

Use the known theorem

Why: That c times negative b is negative c b.

\[ c a + (-c b) \]

Rewrite as a subtraction

Why: By the subtraction property again.

\[ c a - c b \]

Figure (svg): A short proof with every step justified

The right-hand column is what makes this a proof rather than a calculation. Each line is licensed by something already accepted.

\[ c(a - b) = ca - cb \]

Verify: check every reason is legitimate

Why: Two steps cite the subtraction property, one cites the distributive axiom, and one cites a theorem proved earlier. All four are among the permitted kinds of justification, so the chain holds.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 741-741

19. Step to justification

Matching

Every line needs one.

Match the pairs

  • l1. c(a - b) = c[a + (-b)]
  • l2. = ca + c(-b)
  • l3. = ca + (-cb)
  • l4. = ca - cb
  • r1. subtraction property
  • r2. distributive property
  • r3. a theorem already proved
  • r4. subtraction property

Why: The same property is cited twice, in opposite directions. That is entirely legitimate: a property may be used to rewrite in either direction, since it asserts an equality.

20. Worked example: prove a rearrangement

Worked example

Guided Practice 1.

\[ \text{Prove that } (a + b) + c = (b + c) + a. \]

Swap the first two

Why: Commutative property.

\[ (b + a) + c \]

Regroup

Why: Associative property.

\[ b + (a + c) \]

Swap inside the bracket

Why: Commutative property.

\[ b + (c + a) \]

Regroup again

Why: Associative property.

\[ (b + c) + a \]

Figure (svg): A short proof with every step justified

The right-hand column is what makes this a proof rather than a calculation. Each line is licensed by something already accepted.

\[ (a + b) + c = (b + c) + a \]

Verify: check with numbers first

Why: Taking a, b and c as one, two and three gives six on both sides — which supports the claim without proving it. The proof covers every possible triple at once, which no amount of numerical checking could.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 741-741

21. Find the error in this student's work

Error analysis

The student proved a theorem about subtraction.

Annotate

On: \( \begin{aligned} c(a - b) &= ca - cb \\ &\text{because that is how the distributive property works} \end{aligned} \)

  • The distributive property is stated for a sum, not a difference, so it cannot be cited directly for a subtraction.
  • The subtraction has to be rewritten as an addition first, using the subtraction property, before the axiom applies.
  • The full proof needs four steps, two of which convert between subtraction and addition.

The conclusion is correct and the justification is not, which makes this a failed proof rather than a wrong answer. Citing an axiom in a form it was not stated in is the commonest way a proof goes wrong.

22. Rewrite before distributing

Faded example

The axiom is stated for a sum.

Fill in the blanks

c(a - b) = c[a + (-b)] \;\to\; ca + c(-b)

Why: The subtraction property converts the difference into a sum, which is the form the distributive axiom is stated for. Citing an axiom requires the expression to match its statement.

23. Which justification is not allowed?

Elimination

In a proof.

Eliminate the wrong options

Which reason may not be cited?

  • A. It is obviously true
  • B. By the distributive axiom
  • C. By the definition of subtraction
  • D. By a theorem proved earlier

Survives elimination: A

Why: Obviousness is not a logical status, so it justifies nothing. The other three are exactly the permitted kinds of reason, along with information given in the problem.

24. Why write the reasons down?

Socratic

The algebra is the same either way.

Discussion prompt

Say what the justification column adds to a proof. Then say what happens to a proof whose reasons are omitted.

Hint: Who is the proof for?

Answer:

The reasons let someone else check the argument line by line without having to reconstruct your thinking. They also force the writer to confirm that each step really is licensed, which is where errors are caught — the student above would have noticed that the distributive property does not mention subtraction.

Without them the proof becomes a sequence of assertions that a reader must take on trust or verify independently, which defeats the purpose. A proof is a public object, meant to persuade anyone who follows it, and the reasons are what make that possible.

25. Conjectures and evidence

Section

Section 3

26. Examples are not proof

Concept

A conjecture is a statement thought to be true but not yet proved. No number of confirming examples establishes a claim about infinitely many cases.

Goldbach's conjecture is still unproved after nearly three centuries.

  1. Examples give evidence, which raises confidence.
  2. A general claim covers cases no list can reach.
  3. Only an argument covering all cases proves it.

Figure (svg): Even numbers written as sums of two primes

The pattern holds for every case anyone has ever checked, which is billions of them. That is strong evidence and it is not a proof, because the claim is about infinitely many numbers.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 741-741 — the section on conjectures and Example 2 on Goldbach's Conjecture

27. Twelve confirmations

Picture it

And still not settled.

Figure (svg): Even numbers written as sums of two primes

The pattern holds for every case anyone has ever checked, which is billions of them. That is strong evidence and it is not a proof, because the claim is about infinitely many numbers.

Every even number ever tested has been a sum of two primes, into the quintillions. That is overwhelming evidence and no proof at all, which is a distinction worth taking seriously.

28. Worked example: Goldbach's conjecture

Worked example

This is Example 2 from the textbook.

\[ \text{Every even integer except } 2 \text{ is a sum of two primes. Does the list } 4 \text{ to } 26 \text{ prove it?} \]

Check the examples

Why: Every one works.

\[ 4 = 2 + 2, \; 26 = 3 + 23 \]

Ask what is claimed

Why: Every even integer.

Compare

Why: Twelve against infinitely many.

Conclude

Why: Evidence, not proof.

Figure (svg): Even numbers written as sums of two primes

The pattern holds for every case anyone has ever checked, which is billions of them. That is strong evidence and it is not a proof, because the claim is about infinitely many numbers.

\[ \text{unproved, and still open} \]

Verify: consider how much has been checked

Why: Modern computers have verified the conjecture for every even number into the quintillions without finding an exception, and it remains unproved. If that much evidence is not proof, twelve cases plainly are not either.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 741-741

29. Evidence against proof

Faded example

Two different things.

Fill in the blanks

Checking many cases provides evidence, but establishing a claim about infinitely many cases requires a proof.

Why: Evidence makes a statement worth believing and worth investigating. Proof makes it certain, and only an argument that reaches every case can do that.

30. Worked example: a pattern that fails late

Worked example

Why the caution is not merely pedantic.

\[ \text{Consider the claim that } n^2 + n + 41 \text{ is prime for every whole number } n. \]

Try small values

Why: Nought, one, two.

\[ 41, 43, 47 \text{, all prime} \]

Keep going

Why: Up to thirty-nine.

Try forty

Why: Forty squared plus forty plus forty-one.

\[ 1681 = 41 ^{2} \]

Conclude

Why: The pattern breaks.

Figure (svg): Evidence and proof distinguished

Evidence makes a statement worth believing and proof makes it certain. Mathematics is unusual in insisting on the second, and this lesson is about why.

\[ n = 40: \; 1681 = 41 \times 41 \]

Verify: notice how convincing forty cases would feel

Why: Anyone testing the first ten or twenty values would be confident the claim was true, and they would be wrong. That is precisely why mathematics demands an argument rather than a tally, however long the tally is.

31. Trap: treating a strong pattern as established

Trap

The trap

It works for every case I have tried, so it is true.

Generalise from a run of confirming cases

Why: The pattern was consistent and there was no reason to doubt it.

The expression above stays prime for forty consecutive values and then fails. A pattern's track record says nothing certain about the case you have not tried.

The fix

It works for every case I have tried, so it is worth trying to prove.

Treat evidence as a reason to investigate, not as a conclusion

Why: That is what a conjecture is.

Most theorems began as conjectures supported by exactly this kind of evidence.

32. Does this settle the question?

Sorting

Proving or disproving a general claim.

Sort into buckets

Sort each by whether it settles a general statement.

Settles it
finding one case that fails; an argument covering every case
Does not
checking twelve cases that work; checking a billion cases that work; the statement seeming obvious; a computer search finding no exception
yes
It either exhibits a failure, which disproves the claim, or argues for all cases at once, which proves it.
no
It confirms particular cases without reaching the ones untested, so the claim remains open.

Only two of the six settle anything, and one of those settles it in the negative. Confirming instances, however many, never close the question.

33. Why can examples never prove a general claim?

Hypothesis

Even a billion of them.

Predict first

What is the essential obstacle?

  • The claim covers cases no finite list can include
  • Examples might be chosen dishonestly
  • Computers make errors
  • Nothing; enough examples would suffice

Correct: The claim covers cases no finite list can include.

Goldbach's conjecture has been checked past four quintillion and remains open.

Why: A statement about every even integer concerns infinitely many numbers, and any list you can write is finite — so infinitely many cases remain untested no matter how long you continue. Honesty and computational accuracy are beside the point: the gap is one of logic rather than of diligence. That is why an argument covering all cases at once is the only thing that closes it.

34. Is a conjecture worth anything?

Socratic

It is not established.

Discussion prompt

Say what use an unproved conjecture has. Then say what makes one worth attention rather than dismissal.

Hint: Think about what mathematicians do with them.

Answer:

A conjecture directs effort: it says where a theorem is likely to be found and gives something specific to attempt. Whole areas of mathematics have been developed by people trying to prove one, and the tools built along the way often outlast the question.

What makes one worth attention is a combination of strong evidence, connections to other established results, and a statement simple enough to be worth settling. Goldbach's has all three, which is why it has occupied people for nearly three hundred years despite having no practical consequence anyone can name.

35. Counterexamples

Section

Section 4

36. One failure is enough

Concept

To show that a general statement is false, a single counterexample suffices. Proving and disproving are strikingly unequal tasks.

Disproving needs an instance; proving needs an argument.

  1. A general claim asserts something about every case.
  2. One case where it fails makes the claim false.
  3. No further work is needed once one is found.

Figure (svg): Two columns comparing what it takes to prove and to disprove a general claim

The two tasks are wildly unequal. Disproving a general statement needs one instance; proving it needs an argument that reaches every instance at once.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 742-742 — the section on counterexamples and Example 3, Find a Counterexample

37. An asymmetry worth noticing

Picture it

One instance against all cases.

Figure (svg): Two columns comparing what it takes to prove and to disprove a general claim

The two tasks are wildly unequal. Disproving a general statement needs one instance; proving it needs an argument that reaches every instance at once.

The imbalance is the whole point: a claim about everything is fragile in one direction and hard to secure in the other. That is what makes general statements worth stating carefully.

38. Worked example: disprove a general claim

Worked example

The method of Example 3 from the textbook.

\[ \text{Disprove: for all real } x, \; \sqrt{x^2} = x. \]

Read what is claimed

Why: For every real x.

Look for a likely failure

Why: Try a negative value.

\[ x = -3 \]

Test it

Why: The root of nine is three.

\[ \sqrt{9} = 3 \]

Compare

Why: Three is not negative three.

Figure (svg): Two columns comparing what it takes to prove and to disprove a general claim

The two tasks are wildly unequal. Disproving a general statement needs one instance; proving it needs an argument that reaches every instance at once.

\[ x = -3: \; \sqrt{(-3)^2} = 3 \ne -3 \]

Verify: check that one instance really is enough

Why: The claim said every real x, so exhibiting one that fails makes it false — no further cases need testing. Note also that the statement is true for every non-negative x, which is why it looked plausible.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 742-742

39. How many counterexamples are needed?

Elimination

To disprove a general statement.

Eliminate the wrong options

How many cases must fail?

  • A. One
  • B. Several, to be convincing
  • C. Half of all cases
  • D. All of them

Survives elimination: A

Why: A statement about every case is contradicted by one case where it does not hold. That asymmetry with proof is one of the most useful facts in this lesson.

40. Worked example: repair the statement

Worked example

What a counterexample teaches.

\[ \text{How should the claim } \sqrt{x^2} = x \text{ be corrected?} \]

Note where it fails

Why: Only for negative x.

Restrict the claim

Why: To non-negative values.

\[ \text{for } x \ge 0 \]

Or state it generally

Why: Using absolute value.

\[ \sqrt{x^2} = |x| \]

Check the repaired version

Why: At negative three.

\[ |-3| = 3 \;\checkmark \]

Figure (svg): Two columns comparing what it takes to prove and to disprove a general claim

The two tasks are wildly unequal. Disproving a general statement needs one instance; proving it needs an argument that reaches every instance at once.

\[ \sqrt{x^2} = |x| \text{ for all real } x \]

Verify: test the repaired claim on both signs

Why: At three the root of nine is three and the absolute value of three is three; at negative three the root is still three and the absolute value is also three. The corrected statement holds everywhere the original was tested.

41. Trap: producing more counterexamples than necessary

Trap

The trap

Here are five values of x where the claim fails, which shows it is false.

Give several instances for extra certainty

Why: More evidence seemed stronger.

One suffices, and completely. A general claim is false the moment a single case fails, so four of the five add nothing — which is worth knowing, since finding even one can take effort.

The fix

At x equal to negative three the claim fails, so it is false.

Exhibit one counterexample and stop

Why: The claim is already refuted.

The asymmetry runs the other way for proving, where no number of instances is ever enough.

42. Test a likely value

Faded example

Negative values often break claims about roots.

Fill in the blanks

\sqrt3 = \sqrt-3 = ___, \text___ ___

Why: The radical gives the positive root, so it cannot return a negative value of x. Testing a negative number is usually the first thing to try against a claim involving square roots.

43. Where would you look for a counterexample?

Sorting

Some values break claims more often.

Sort into buckets

Sort each candidate by whether it is a promising first test.

Promising
a negative number; zero; a fraction between 0 and 1; one
Less so
a typical positive whole number; a large positive number
yes
It sits at a boundary or behaves unusually, so claims often fail there first.
no
It is an ordinary case, and a claim that fails at all will usually already have failed at an edge case.

Negatives, nought, one and fractions below one are where general claims break most often, because each behaves differently from a typical positive whole number. Trying those first is the efficient search.

44. Why is disproving so much easier?

Socratic

The two tasks look symmetric.

Discussion prompt

Explain the imbalance between proving and disproving a general statement. Then say what it implies about how carefully general claims should be worded.

Hint: What does each task have to cover?

Answer:

Proving requires an argument reaching every case at once, while disproving requires exhibiting a single case — so one task is about all instances and the other about one. That is not a symmetry at all, which is why a claim that took centuries to prove can be refuted in a line if it happens to be false.

It implies that general claims should be stated with their conditions attached, since an unstated restriction is exactly where a counterexample will be found. The claim about square roots was true for every non-negative value and false only for negatives, so wording it carefully would have made it a theorem instead of a mistake.

45. Indirect proof

Section

Section 5

46. Assume the opposite and break something

Concept

In an indirect proof, the statement is assumed false and reasoning proceeds until something impossible follows. That impossibility shows the assumption was wrong, so the statement is true.

It is how the root of two is shown to be irrational.

  1. Suppose the statement is false.
  2. Reason carefully from that supposition.
  3. Reach a contradiction, so the supposition fails.

Figure (svg): The shape of an indirect proof

The contradiction is what does the work: if assuming the opposite forces something impossible, the opposite cannot hold. Nothing else about the original statement need be shown.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 742-745 — the key word indirect proof and Example 4, on proving a client's innocence

47. Three steps to a contradiction

Picture it

Assume, reason, break.

Figure (svg): The shape of an indirect proof

The contradiction is what does the work: if assuming the opposite forces something impossible, the opposite cannot hold. Nothing else about the original statement need be shown.

Nothing about the statement itself is demonstrated directly. What is demonstrated is that its denial cannot stand, which for a statement that is either true or false amounts to the same thing.

48. Worked example: the square root of two is irrational

Worked example

The proof promised back in Lesson 9.1.

\[ \text{Prove that } \sqrt{2} \text{ cannot be written as a fraction.} \]

Assume the opposite

Why: A fraction in lowest terms.

\[ \sqrt{2} = \tfrac{a}{b} \]

Square and rearrange

Why: Both sides.

\[ 2 b ^{2} = a ^{2} \]

Deduce a is even

Why: Its square is even.

\[ a = 2 k \]

Deduce b is even too

Why: Substituting gives 2b^2 = 4k^2.

Figure (svg): The indirect proof that the square root of two is irrational

The assumption that the fraction is in lowest terms is what the contradiction destroys. Since every fraction can be put in lowest terms, no fraction can equal the root of two at all.

\[ \sqrt{2} \text{ is irrational} \]

Verify: identify exactly what was contradicted

Why: The fraction was assumed to be in lowest terms and the reasoning showed both parts are even, so it was not. Since every fraction can be written in lowest terms, no fraction at all can equal the root of two.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 742-745

49. The shape of the argument

Faded example

Three moves.

Fill in the blanks

Assume the statement is false, reason from that until something impossible follows, and conclude that the statement is true.

Why: The structure works because a statement is either true or false: ruling out one leaves the other. The whole argument rests on the contradiction being genuine.

50. Worked example: an alibi as an indirect proof

Worked example

The method of Example 4, in its everyday form.

\[ \text{How does an alibi establish innocence?} \]

Assume the opposite

Why: Suppose the client is guilty.

Follow the consequence

Why: They were at the scene.

Bring in the evidence

Why: They were elsewhere then.

Conclude

Why: Impossible, so not guilty.

Figure (svg): An indirect argument used to establish innocence

An alibi is an indirect proof: guilt is assumed, an impossibility follows, and guilt is therefore rejected. The logical shape is identical to the one used for the root of two.

\[ \text{indirect proof} \]

Verify: compare the two arguments

Why: The root of two proof assumes a fraction and breaks lowest terms; the alibi assumes guilt and breaks being in one place. Both reject their assumption because it forced something that cannot be, which is exactly the same logical move.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 745-745

51. Trap: reaching a surprise rather than a contradiction

Trap

The trap

Assuming the opposite led to a very strange conclusion, so the statement must be true.

Treat an unexpected result as a contradiction

Why: It seemed unlikely enough to count.

Strange is not impossible. An indirect proof needs something that genuinely cannot be — a quantity both even and odd, a fraction both in lowest terms and not — rather than merely a result nobody expected.

The fix

Assuming the opposite forced a and b to be both coprime and both even, which cannot happen.

Identify precisely what has been contradicted

Why: Name the two incompatible facts.

Many true statements are surprising, so surprise carries no logical weight at all.

52. What must an indirect proof reach?

Elimination

After assuming the opposite.

Eliminate the wrong options

What ends the argument?

  • A. A genuine contradiction, something that cannot be true
  • B. A surprising result
  • C. A difficult calculation
  • D. A result that disagrees with intuition

Survives elimination: A

Why: Only an impossibility rules out the assumption. Being unexpected, hard or counter-intuitive says something about the reader rather than about the mathematics.

53. Why does the root two proof work?

Hypothesis

It never computes anything about root two directly.

Predict first

What does the contradiction actually establish?

  • That no fraction in lowest terms can equal root 2, and hence no fraction can
  • That root 2 is close to a fraction but not equal
  • That the calculation was done incorrectly
  • That a and b were badly chosen

Correct: That no fraction in lowest terms can equal root 2, and hence no fraction can.

The same argument works for the root of three, and for any prime.

Why: The final step is the important one: every fraction can be reduced to lowest terms, so ruling out those in lowest terms rules out all of them. Without that observation the proof would only have shown something about a particular kind of fraction. Choosing a and b badly is not possible either, since the argument never used anything about them beyond the assumption.

54. Why prove something indirectly?

Socratic

A direct argument would seem preferable.

Discussion prompt

Say when an indirect proof is the natural approach. Then say what makes irrationality a case for it.

Hint: What does the statement deny?

Answer:

It suits statements that assert something cannot happen, because a direct proof would have to examine every possible way it might and rule each out. Assuming it does happen gives you something concrete to work with, which a negative statement otherwise fails to provide.

Irrationality is exactly such a statement: it says no fraction whatever equals the root of two, and there are infinitely many fractions to consider. Assuming one does gives an equation to manipulate, and that single equation is enough to break the assumption for all of them at once.

55. Proving against disproving

Comparison

Fill the blanks from memory before you scroll back.

Comparison matrix

TaskWhat is neededHow much work
Prove a general statementan argument covering every caseno number of examples suffices
Disprove a general statementone counterexamplea single case is enough
Prove something cannot happenoften an indirect proofassume it does, and reach a contradiction

The middle row is the cheapest task in mathematics and the top row is among the most expensive. Knowing which you face determines how to spend your effort.

56. The procedure, in order

Pattern

To establish or refute a mathematical claim, these five moves cover it.

  1. Decide what kind of statement it is: axiom, definition, theorem or conjecture.
  2. If you suspect it is false, look for a counterexample at the edge cases first.
  3. If you believe it is true, look for a direct chain of justified steps.
  4. If it asserts that something cannot happen, try assuming it does and seek a contradiction.
  5. Write every step with its justification, so a reader can check the argument.

Step two before step three is deliberate: a few minutes hunting for a counterexample can save hours attempting to prove something false.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 740-745

57. Check yourself 1 of 3

Check

Statuses differ.

Check your understanding

Which kind of statement is accepted without proof?

  • A. An axiom (correct)
  • B. A theorem
  • C. A conjecture
  • D. A counterexample

Answer: A

Why: Axioms, also called postulates, are the properties assumed at the outset so that reasoning has somewhere to start.

Why B tempts people
A theorem is precisely a statement that must be proved.
Why C tempts people
A conjecture is unproved and, unlike an axiom, not accepted either.
Why D tempts people
A counterexample is an instance that refutes a claim, not a kind of assumed statement.

58. Check yourself 2 of 3

Check

One is enough.

Check your understanding

How many counterexamples disprove a general statement?

  • A. One (correct)
  • B. At least three
  • C. More than half the cases
  • D. It cannot be disproved by examples

Answer: A

Why: A statement claiming something about every case is false as soon as one case fails.

Why B tempts people
Further examples add nothing once one failure has been exhibited.
Why C tempts people
The claim asserts all cases, so a single exception refutes it.
Why D tempts people
Disproving is exactly the task that examples can accomplish.

59. Check yourself 3 of 3

Check

Assume the opposite.

Check your understanding

What must an indirect proof reach to succeed?

  • A. A contradiction (correct)
  • B. A surprising result
  • C. A numerical answer
  • D. A counterexample

Answer: A

Why: Something genuinely impossible must follow from the assumption, which shows the assumption cannot hold.

Why B tempts people
Surprise has no logical force; many true statements are surprising.
Why C tempts people
An indirect proof establishes a statement rather than computing a value.
Why D tempts people
A counterexample refutes a claim directly and is a different method.

60. Where this shows up outside the textbook

Real world

This is the courtroom question from the lesson opener. A lawyer defending a client often argues indirectly rather than demonstrating innocence directly.

Discussion prompt

Describe how an alibi establishes innocence, name the logical form of the argument, and say what mathematical proof in this lesson has the same shape.

Hint: What is assumed at the start?

Answer:

The lawyer supposes the client is guilty, which would place them at the scene at a particular time, and then produces evidence that they were somewhere else at that time. Being in two places at once is impossible, so the supposition fails and the client is not guilty.

That is an indirect proof: assume the opposite of what you want to establish, reason to something impossible, and reject the assumption.

The proof that the square root of two is irrational has exactly this shape. It supposes the root is a fraction in lowest terms, shows that both its parts must then be even, and observes that a fraction cannot be in lowest terms while both parts share a factor of two. The contradiction is different but the logical move is identical, which is why the courtroom is a fair illustration of a technique that looks purely mathematical.

61. How sure are you?

Commit first

Answer, then rate your confidence honestly. Confident and wrong is the combination worth finding.

Predict first

A claim about all even integers holds for the first billion cases tested. What follows?

  • It is proved, since a billion cases is overwhelming
  • It remains a conjecture
  • It is disproved
  • It becomes an axiom

Correct: It remains a conjecture.

\[ n = 40: \; 40^2 + 40 + 41 = 1681 = 41^2 \]

Why: The claim concerns infinitely many integers and a billion is a finite list, so infinitely many cases remain untested — no amount of checking closes that gap, which is a matter of logic rather than of diligence. Goldbach's conjecture has been verified past four quintillion and is still open after nearly three hundred years, which shows how little even extraordinary evidence settles. The expression n squared plus n plus forty-one is a sharper warning: it produces primes for forty consecutive values and then fails at forty, so a run of confirmations genuinely can end. An axiom, meanwhile, is something chosen as a starting assumption rather than something a claim graduates into by accumulating evidence.

62. Explain it to someone a year behind you

Explain it

They checked a pattern on ten examples and concluded it is always true.

Discussion prompt

In no more than four sentences, explain what their examples have and have not shown. Then give them the example that makes the point.

Hint: How many cases does the claim cover?

Answer:

A usable answer: ten examples give good evidence that the pattern is worth investigating, but the claim covers infinitely many cases and ten of them leaves infinitely many untested. Only an argument that reaches every case at once can prove it.

The example worth showing them is n squared plus n plus forty-one, which gives a prime for every whole number from nought to thirty-nine and then fails at forty. Forty confirmations in a row and the pattern still breaks.

63. Exit ticket

Exit ticket

Name the weakest spot before you close the deck. That is the one worth ten minutes tonight.

Predict first

Which of these would you least want to be handed cold on a quiz tomorrow?

  • Telling an axiom from a theorem
  • Justifying every step of a proof
  • Explaining why examples are not proof
  • Setting up an indirect proof

Correct: Whichever you picked is the right answer — and each one has a specific fix.

Why: Axioms and theorems are separated by asking whether the statement is assumed or derived. Justification is fixed by writing a reason beside every line and checking it is one of the four permitted kinds. The examples point is fixed by remembering that a general claim covers cases no list can reach. Indirect proof is fixed by writing the assumed opposite explicitly at the top, so you know what you are trying to contradict. Pick yours and do five of that kind tonight rather than twenty mixed ones.

64. Draw the lesson on one page

Connect it up

Do this on paper. It is worth more than rereading the slides.

Draw it

At the top of a page draw the three levels — axioms, definitions, theorems — and write two examples of each from this book, noting beside each level whether proof is required. Underneath, write out a short proof in two columns, statements on the left and reasons on the right, and check that every reason is an axiom, a definition, given information or an earlier theorem. In the middle, write a general claim, list several confirming examples, and then write one sentence saying exactly what those examples have established and what they have not. Beneath that, take a false general claim, find one counterexample, and then repair the claim by adding the condition its counterexample violated. In the lower half, set out an indirect proof in three labelled parts — the assumed opposite, the reasoning, and the contradiction — using either the root of two or an argument of your own, and name precisely which two statements are incompatible. Finally, in the margin, write the asymmetry between proving and disproving in one line.

Your indirect proof should name the contradiction explicitly rather than leaving it implied. If you cannot state which two things cannot both be true, the argument has reached a surprise rather than a contradiction.

65. What you can do now

Recap

Five things, and together they are what makes the rest of this book more than a list of rules.

If the question saysYour first move is
Prove this theoremFind a chain of justified steps from the axioms
Is this always true?Look for a counterexample before attempting a proof
Show this statement is falseExhibit one case where it fails
Show this cannot happenAssume it does and seek a contradiction
Does this list prove it?Ask how many cases the claim covers

That completes the course. Every technique in this book — solving, factoring, graphing, the quadratic formula, the distance formula — rests on the small set of axioms set out in this lesson, and each was established by exactly the kind of reasoning practised here. What began as rules to follow is, properly seen, a single connected argument.

McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof §12.9, pp. 740-745 — everything on these slides traces back here

Sources

  1. McDougal Littell Algebra 1: Concepts and Skills, Ch. 12 Radicals and More Connections to Geometry — Lesson 12.9 Logical Reasoning: Proof — Larson, Boswell, Kanold & Stiff, McDougal Littell / Houghton Mifflin, 2004, pp. 740-745

Want this taught 1-on-1? Alexander tutors Algebra 1 — $55/session, free consultation.

Book on Wyzant · Text (657) 465-8108