4/9 4. Validity and Soundness

Arguments and Their Structure

An argument is a group of statements made up of premises and a conclusion. Premises provide reasons, while the conclusion is the statement those reasons are intended to support.

For example:

Premise 1: If a number is divisible by 4, then it is even.

Premise 2: 12 is divisible by 4.

Conclusion: Therefore, 12 is even.

The structure of an argument concerns the way its statements are connected, rather than the topic being discussed. The following pattern is an example of modus ponens:

Any argument with this form is deductively valid when the conditional and premises are represented correctly. The same structure can be used in mathematics, science, everyday reasoning, or any other subject.

Takeaway: To analyze an argument, first separate its premises from its conclusion and then examine the pattern connecting them.

Deductive Validity

Deductive validity concerns whether the premises guarantee the conclusion. An argument is valid exactly when there is no possible case in which all its premises are true and its conclusion is false.

Validity depends on logical form, not on the actual truth of the premises. For example:

All birds are mammals.

Penguins are birds.

Therefore, penguins are mammals.

The form is valid because, if both premises were true, the conclusion would have to be true. However, the first premise is false, so the argument is not sound.

A true conclusion alone does not establish validity. Consider this pattern:

If the Moon is made of cheese, then $2 + 2 = 4$.

The Moon is made of cheese.

Therefore, $2 + 2 = 4$.

The conclusion is true, but validity is determined by the relationship between the premises and conclusion, not merely by the truth of the conclusion.

Takeaway: Ask whether the premises make a false conclusion impossible. That question tests validity.

Invalidity and Counterexamples

An argument is invalid when its structure permits all the premises to be true while the conclusion is false. Invalidity does not prove that the conclusion is false; it shows only that the premises do not guarantee it.

A counterexample makes this failure concrete. Consider:

If it rains, the sidewalk becomes wet.

The sidewalk is wet.

Therefore, it rained.

The premises could be true while the conclusion is false because a sprinkler, hose, or street-cleaning truck could have made the sidewalk wet. One possible case of this kind is enough to establish invalidity.

The general form is:

This form is invalid because $Q$ may have more than one possible cause.

Takeaway: To disprove validity, look for just one possible situation in which every premise is true and the conclusion is false.

Soundness

Soundness combines two requirements:

1. The argument is valid.

2. Every premise is actually true.

A sound argument is therefore guaranteed to have a true conclusion. For example:

All mammals are warm-blooded.

Whales are mammals.

Therefore, whales are warm-blooded.

This argument is valid, and both premises are true, so it is sound.

An argument can be valid but unsound when it has a false premise:

All reptiles are mammals.

Snakes are reptiles.

Therefore, snakes are mammals.

The form is valid because the conclusion would follow if the premises were true. However, the first premise is false, so the argument is unsound.

An invalid argument cannot be sound, even when all its premises and its conclusion happen to be true. Soundness requires both a truth-preserving structure and true premises.

Takeaway: Validity asks whether the reasoning works; soundness asks whether the reasoning works and starts from true premises.

Testing Validity with Truth Tables

A truth table tests the validity of an argument by listing every possible combination of truth values for its component propositions.

Use this procedure:

1. Translate the premises and conclusion into symbolic form.

2. List every possible assignment of truth values.

3. Find the rows in which all premises are true.

4. Check the conclusion in those rows.

5. If the conclusion is false in even one such row, the argument is invalid. If it is true in every such row, the argument is valid.

For the form

the only row in which both premises are true is the row where $P$ and $Q$ are true. The conclusion is true in that row, so the form is valid.

The central test is:

Is there a possible case in which all the premises are true and the conclusion is false?

If the answer is yes, the argument is invalid. If the answer is no, the argument is valid.

Takeaway: Truth-table testing focuses only on rows where every premise is true.

Deductive and Inductive Reasoning

Deductive and inductive reasoning differ in the kind of support the premises are intended to provide.

Deductive reasoning aims at certainty. If a deductive argument is valid and its premises are true, its conclusion cannot be false. For example:

All registered voters in District A received a ballot.

Jordan is a registered voter in District A.

Therefore, Jordan received a ballot.

Inductive reasoning aims at probability or support rather than certainty. For example:

The last 20 buses on this route arrived late.

Therefore, the next bus will probably arrive late.

The premises support the prediction, but the next bus could still arrive on time. Deductive arguments are evaluated as valid or invalid, while inductive arguments are commonly evaluated as strong or weak.

A failure to guarantee a conclusion is not automatically a defect in inductive reasoning. Predictions, generalizations, and many practical decisions depend on evidence that makes conclusions probable without making them certain.

Takeaway: Deduction seeks a guaranteed conclusion; induction seeks a conclusion supported by evidence.

Common Invalid Forms

Two common invalid forms result from treating evidence that is compatible with a conclusion as evidence that guarantees it.

Affirming the consequent has this form:

It is invalid because $Q$ may result from causes other than $P$. A wet sidewalk does not prove that rain occurred.

Denying the antecedent has this form:

It is invalid because $Q$ may occur for another reason. For example:

If the alarm is set, the light flashes.

The alarm is not set.

Therefore, the light does not flash.

The light might flash because of a power test or another system.

To evaluate an argument, keep two questions separate:

1. Does the conclusion follow from the premises? This concerns validity.

2. Are the premises true? This matters for soundness.

Takeaway: A conditional statement identifies a sufficient connection, not necessarily the only cause of its consequent.