Argument Evaluation

An argument is a set of statements in which some (the premises) are offered as reasons for believing another (the conclusion). Critical thinkers evaluate arguments by reconstructing them, testing their logical form, checking the truth of the premises, and engaging with the strongest version of the opposing case. This skill provides a systematic procedure for doing all four.

Agent affinity: paul (overall framing, elements of reasoning), elder (structural reconstruction)

Concept IDs: crit-argument-structure, crit-deductive-reasoning, crit-inductive-reasoning, crit-charitable-interpretation

The Evaluation Toolbox at a Glance

Operation 1 — Premise Extraction

Pattern: Isolate statements the author offers as reasons. These are the building blocks of the argument.

Indicator words: "because," "since," "given that," "in light of," "as shown by," "for the reason that."

Worked example. "Because the economy is slowing and interest rates are still high, a recession is likely within six months."

Premises:

• P1: The economy is slowing.
• P2: Interest rates are still high.

Notice that indicator words are cues, not guarantees. Some premises appear without indicators; some indicator words are used rhetorically without introducing real premises.

Operation 2 — Conclusion Extraction

Pattern: Identify the statement the premises are meant to support.

Indicator words: "therefore," "thus," "so," "hence," "it follows that," "we can conclude," "which shows."

Worked example (continued). The conclusion of the example above is:

• C: A recession is likely within six months.

Common mistake. Treating a rhetorical question as a conclusion. "Why should we trust them?" is not a conclusion; it is a question that implies one.

Operation 3 — Argument Mapping

Pattern: Show which premises support which conclusions, including intermediate sub-conclusions.

Worked example. "Social media harms adolescent mental health because it promotes comparison with curated images. Therefore, schools should restrict phone use during the day."

P1: Social media promotes comparison with curated images.
                |
                v
SC1: Social media harms adolescent mental health.  (sub-conclusion)
                |
                v
C:  Schools should restrict phone use during the day.

The map reveals that the argument has two logical steps, only one of which was explicitly signaled. Mapping often exposes gaps.

Operation 4 — Reconstruction in Standard Form

Pattern: Restate the argument as numbered premises followed by a single conclusion.

Worked example. "You shouldn't drink the water here. The EPA says it has lead contamination."

Standard form:

P1. The EPA says this water has lead contamination.
P2. [Hidden] Water with lead contamination is unsafe to drink.
P3. [Hidden] You should not drink unsafe water.
C.  You should not drink the water here.

Reconstruction often surfaces hidden premises. An argument is not a shapeless cloud of sentences — it is a chain, and every link needs to be visible before you can test it.

Operation 5 — Validity Check

Pattern: Ask whether, if the premises were all true, the conclusion would have to be true.

Logical basis: Validity is a property of the argument's form, not of its content. An argument can be valid with false premises, and invalid with true premises.

Worked example (valid).

P1. All ravens are birds.
P2. All birds have feathers.
C.  All ravens have feathers.

This is valid because the form (All A are B; All B are C; therefore All A are C) guarantees the conclusion whenever the premises are true.

Worked example (invalid).

P1. All ravens are birds.
P2. Some birds are flightless.
C.  Some ravens are flightless.

The premises are both true, but the conclusion does not follow. There is no logical guarantee connecting the particular birds that are flightless to ravens specifically.

The counterexample test. To show an argument form is invalid, produce an instance with the same form where the premises are true and the conclusion is false.

Operation 6 — Soundness Check

Pattern: An argument is sound if and only if it is valid and all premises are actually true.

Validity is a relationship among statements. Soundness is validity plus real-world accuracy. A valid argument with a false premise proves nothing about the world.

Worked example.

P1. All mammals lay eggs.          [FALSE]
P2. Whales are mammals.            [TRUE]
C.  Whales lay eggs.               [Valid form, unsound argument]

The form is valid. The first premise is false. The conclusion is false. Soundness failed at P1.

Discipline. When evaluating a valid argument, interrogate each premise independently. Which evidence supports it? Which evidence could refute it? What would change your mind about this specific premise?

Operation 7 — Strength Check (Inductive Arguments)

Pattern: Inductive arguments do not guarantee their conclusions; they make them more or less probable. Strength measures the probability given the premises.

Worked example (strong induction).

P1. A random sample of 10,000 adults from the population shows
    a 62% approval rating for policy X.
P2. The sample was drawn using standard random-digit dialing with
    stratification by age, region, and household income.
C.  Approximately 62% of adults in the population approve of policy X.

Strong because the sample is large, randomized, and stratified. Conclusion is probabilistic, not certain.

Worked example (weak induction).

P1. I know three people who got sick after eating at that restaurant.
C.  That restaurant causes people to get sick.

Weak because the sample is tiny, self-selected, and confounded with many other causes.

Strength indicators: sample size, representativeness, sampling method, absence of confounds, replicability. Weakness indicators: small n, selection bias, anecdote, single source.

Operation 8 — Steel-Manning (Charitable Interpretation)

Pattern: When reconstructing an argument you disagree with, build the strongest version possible before criticizing it. Add missing support the author could reasonably have given, interpret ambiguous phrases in the most defensible way, and acknowledge which parts of the opposing view are correct.

Worked example. Suppose the opposing claim is: "We shouldn't raise the minimum wage because it will cause job losses."

Straw-man version (weak). "They think rich employers matter more than poor workers."

Steel-man version (strong).

P1. Some studies find that minimum wage increases above the local equilibrium
    cause reductions in low-wage employment, especially in small businesses.
P2. Low-wage workers include many first-job seekers and those with low bargaining power.
P3. Job losses f