Two Separate Checks: Premises vs Connection
When you evaluate an argument, do two different jobs on purpose: (1) check whether the premises are acceptable, and (2) check whether the reasoning actually connects those premises to the conclusion. Many disagreements happen because people mix these up.
Check A: Are the premises acceptable?
A premise is acceptable when it is reasonable to believe in the context: it is well-supported, accurately stated, and not wildly controversial without evidence. “Acceptable” does not mean “certain.” It means “a fair starting point.”
- Evidence check: Is there a source, observation, measurement, or common experience backing it?
- Clarity check: Is it specific enough to be tested or discussed (not vague like “usually” with no details)?
- Fit check: Is it relevant to the conclusion, or just emotionally persuasive?
- Counterexample check: Can you easily think of cases where it fails? If yes, the premise may need narrowing.
Check B: Does the reasoning connect?
Even if every premise is acceptable, the argument can still fail if the conclusion doesn’t follow from them. Ask: “If I granted these premises, would I be forced (or at least strongly pushed) to accept the conclusion?”
- Bridge question: What is the step that moves from premises to conclusion? Is it stated or missing?
- Scope question: Do the premises support this exact conclusion, or only a weaker one?
- Alternative explanation question: Could the premises be true while a different conclusion is also plausible?
A quick two-step routine
- Underline the premises (the “because” parts) and circle the conclusion (the “therefore” part).
- Rate premise acceptability first, then rate reasoning connection second.
- Write one sentence for each: “Premises are (strong/medium/weak) because…” and “Connection is (strong/medium/weak) because…”
Validity vs Soundness (Intuitive, No Symbolic Logic)
For arguments that aim to be certain (not merely likely), philosophers often use two labels:
- Valid: If the premises were true, the conclusion would have to be true. Validity is about the structure of the reasoning, not whether the premises are actually true.
- Sound: Valid and the premises are actually true (or at least acceptable as true). Soundness is “good structure + true starting points.”
Valid but not sound (true connection, bad starting point)
Example:
- Listen to the audio with the screen off.
- Earn a certificate upon completion.
- Over 5000 courses for you to explore!
Download the app
Premise 1: All birds are mammals. (false premise) Premise 2: A robin is a bird. (true premise) Conclusion: Therefore, a robin is a mammal.This argument is valid because if Premise 1 and Premise 2 were true, the conclusion would follow. But it is not sound because Premise 1 is false.
Invalid (even if the premises are true)
Example:
Premise 1: If it is raining, the street is wet. Premise 2: The street is wet. Conclusion: Therefore, it is raining.The premises can both be true, but the conclusion doesn’t have to be true (a street can be wet for other reasons). So the reasoning is invalid.
How to test validity without formal logic
- “Could the premises be true while the conclusion is false?” If yes, the argument is invalid.
- Try a simple counter-scenario: Keep the premises the same, imagine another cause or case, and see if the conclusion still must be true.
Strength for Probabilistic Arguments
Many everyday arguments are not trying to guarantee the conclusion; they aim to make it more likely. For these, use strength rather than validity.
- Strong (probabilistic) argument: If the premises are true, the conclusion is very likely.
- Weak (probabilistic) argument: If the premises are true, the conclusion is only slightly more likely—or not much more likely at all.
How evidence can increase strength
Example (starting weak):
Premise: I saw one review saying the restaurant is good. Conclusion: The restaurant is good.This is weak because one review is a thin sample and could be biased.
Make it stronger by adding independent, relevant evidence:
Premise 1: The restaurant has 1,200 reviews with an average of 4.6/5. Premise 2: Several friends with similar tastes recommend it. Premise 3: A recent health inspection score is high. Conclusion: The restaurant is probably good.Notice the conclusion becomes probabilistic (“probably”), and the premises are more diverse and harder to dismiss all at once.
How evidence can decrease strength
Adding information can also weaken a probabilistic argument by introducing alternative explanations or exceptions.
Premise 1: This supplement has many positive testimonials. Conclusion: The supplement is effective.Weak already, but it becomes weaker if we add:
Premise 2: Testimonials are not controlled studies and can be influenced by placebo effects. Premise 3: People with bad results may be less likely to post. Revised conclusion: Testimonials alone do not strongly support that the supplement is effective.Beginner checklist for strength
- Quantity: Is there enough evidence, or is it a tiny sample?
- Quality: Is the evidence reliable (measurement, records, expert consensus) or shaky (rumor, cherry-picked anecdotes)?
- Independence: Are multiple sources truly independent, or repeating the same claim?
- Relevance: Does the evidence actually bear on the conclusion?
- Alternatives: Are there other plausible explanations the premises don’t rule out?
Practice: Rate Arguments on a Two-Axis Grid
You will rate each argument on two axes: Premise Credibility (how acceptable the premises are) and Reasoning Quality (how well the premises support the conclusion). Use this grid to keep the two evaluations separate.
| Reasoning Quality: Low | Reasoning Quality: High | |
|---|---|---|
| Premise Credibility: High | Good facts, weak connection | Strong overall argument |
| Premise Credibility: Low | Weak overall argument | Good structure, doubtful starting points |
Argument 1
Premise: Most people who drink coffee every day are addicted. Conclusion: Therefore, if you drink coffee every day, you are addicted.Your task: Rate it on the grid and write one paragraph explaining both ratings.
Argument 2
Premise 1: If a device is left in a hot car, its battery life can degrade. Premise 2: My phone was left in a hot car for six hours. Conclusion: Therefore, my phone’s battery life has degraded.Your task: Rate it on the grid and write one paragraph explaining both ratings.
Argument 3
Premise 1: This policy reduced traffic accidents in three similar cities. Premise 2: Our city is similar in population and road design. Conclusion: Therefore, adopting this policy will probably reduce traffic accidents in our city.Your task: Rate it on the grid and write one paragraph explaining both ratings.
Argument 4
Premise 1: Everyone I know who uses that budgeting app saves money. Conclusion: Therefore, that budgeting app causes people to save money.Your task: Rate it on the grid and write one paragraph explaining both ratings.
Argument 5
Premise 1: If a claim is on the internet, it must be true. Premise 2: This claim is on the internet. Conclusion: Therefore, this claim is true.Your task: Rate it on the grid and write one paragraph explaining both ratings.
Optional scoring method (if you like numbers)
Give each axis a 1–5 score. Then write your paragraph using this template:
Premise credibility: __/5 because ... Reasoning quality: __/5 because ... Overall: The argument is (valid/invalid) or (strong/weak) given its goal, because ...Quick Repair Activity: Make a Weak Argument Stronger
Choose one of the practice arguments you rated as weak. Strengthen it by doing one of these repairs:
- Repair type A (premise repair): Replace one shaky premise with a more credible, specific, or better-supported premise.
- Repair type B (conclusion repair): Narrow the conclusion so it matches what the premises actually support (often by adding “probably,” limiting the scope, or changing “causes” to “is associated with”).
Step-by-step repair method
- Name the weakness: Is it premise credibility, reasoning connection, or both?
- Pick one repair type: A (premise) or B (conclusion).
- Rewrite the argument: Keep it short—2–3 premises.
- Re-rate it on the grid: Did your change improve one axis without harming the other?
Example repair (from Argument 4)
Original:
Premise: Everyone I know who uses that budgeting app saves money. Conclusion: Therefore, that budgeting app causes people to save money.Repair by narrowing the conclusion (Repair type B):
Premise: In my social circle, people who use that budgeting app tend to save more money than before. Conclusion: Therefore, in my social circle, using the app is associated with increased saving (though it may not be the only cause).Repair by improving a premise (Repair type A):
Premise 1: A study comparing similar users found that app users increased savings by an average of 8% over three months. Premise 2: The study controlled for income changes during that period. Conclusion: Therefore, the app probably helps some users save money.
