How to Argue Well — Logic and Critical Thinking

Department of Philosophy | Level: Beginner | Duration: 25 minutes

Objectives

After this lesson, you will be able to:

Identify the structure of an argument (premises and conclusion)
Distinguish between validity and soundness
Recognize common logical fallacies
Spot bad arguments in everyday conversation
Construct a well-structured argument of your own

1. What Is an Argument?

In philosophy, an “argument” is not a shouting match. It is a structured set of claims:

Premises — statements offered as evidence or reasons
Conclusion — the statement that the premises are supposed to support

Example:

Premise 1: All guitars have strings.
Premise 2: This instrument is a guitar.
Conclusion: Therefore, this instrument has strings.

That is it. An argument is just premises leading to a conclusion. Everything else — rhetoric, emotion, volume, confidence — is decoration.

How to Find the Argument

In everyday conversation, arguments are rarely laid out neatly. Look for indicator words:

Premise indicators	Conclusion indicators
Because, since, given that, as	Therefore, so, thus, consequently
The reason is, it follows from	This means, which shows that
Considering that, due to	We can conclude that, hence

Example in the wild: “We should switch to a new framework because the current one has no community support and three unpatched vulnerabilities.”

Premise 1: The current framework has no community support.
Premise 2: It has three unpatched vulnerabilities.
Conclusion: We should switch to a new framework.

Now you can evaluate each piece separately. Are the premises true? Does the conclusion follow?

Practice Exercise

Find an opinion piece, tweet, or Slack message that makes a claim. Identify the premises and conclusion. Write them out in the format above.

2. Validity vs Soundness

These two words are the most important distinction in logic:

Validity

An argument is valid if the conclusion must be true whenever the premises are true. It is about the structure, not the content.

Valid (but absurd):

Premise 1: All cats are made of cheese.
Premise 2: Whiskers is a cat.
Conclusion: Whiskers is made of cheese.

The structure is perfect. If cats were made of cheese, Whiskers would indeed be cheesy. The argument is valid even though Premise 1 is obviously false.

Soundness

An argument is sound if it is valid AND all premises are actually true.

Sound:

Premise 1: All humans are mortal.
Premise 2: Socrates is a human.
Conclusion: Socrates is mortal.

Valid structure + true premises = sound argument. This is the gold standard.

Why This Matters

When someone presents an argument, you have two separate questions:

Is the structure valid? Does the conclusion follow from the premises?
Are the premises true? Is the evidence actually correct?

A bad argument can fail at either level. Many real-world disagreements are actually about premises (the facts), not logic (the reasoning).

3. Common Fallacies — The Rogues Gallery

A fallacy is a pattern of reasoning that looks convincing but is logically broken. Here are the ones you will encounter most often:

Ad Hominem (Attack the Person)

What it looks like: “You can’t trust her analysis of the codebase — she only joined the team six months ago.”

Why it is wrong: The quality of an argument does not depend on who makes it. A newcomer can be right; a veteran can be wrong. Evaluate the argument, not the person.

Watch out for: “Of course you would say that,” “What would they know about it?”

Straw Man (Distort the Argument)

What it looks like: Person A says “We should add input validation to the API.” Person B responds: “So you want to block all user input? That would make the product unusable.”

Why it is wrong: Person B is attacking a distorted version of Person A’s argument. Nobody said “block all input.” This makes it easy to “win” against a position nobody actually holds.

Watch out for: “So what you’re really saying is…”

False Dichotomy (Only Two Options)

What it looks like: “Either we ship this feature by Friday or we lose the client forever.”

Why it is wrong: There are almost always more than two options. You could ship a partial version, negotiate a new deadline, or address the client’s underlying concern a different way.

Watch out for: “Either… or…” when the options are not genuinely exhaustive.

Appeal to Authority (Trust Because of Status)

What it looks like: “The CEO thinks we should use this database, so it must be the right choice.”

Why it is wrong: Authority is not evidence. The CEO might know nothing about databases. What matters is the reasoning and evidence, not who said it.

Nuance: Expert opinion is evidence when the expert is speaking within their field of expertise. A database engineer recommending a database carries more weight than a CEO doing the same. The fallacy is treating authority as a substitute for evidence rather than a source of it.

Appeal to Popularity (Everyone Thinks So)

What it looks like: “This JavaScript framework has 80,000 GitHub stars, so it must be good.”

Why it is wrong: Popularity is not a reliable indicator of quality. Many popular things are mediocre; many excellent things are obscure.

Slippery Slope (If A Then Z)

What it looks like: “If we allow remote work on Fridays, soon nobody will come to the office at all, and the company culture will collapse.”

Why it is wrong: Each step in the chain needs its own justification. A leads to B only if you can show why it leads to B. Merely asserting a chain of consequences is not an argument.

Practice Exercise

Over the next 24 hours, listen for fallacies in conversations, meetings, news, or social media. Try to identify at least one example of each type above. Write down what was said and which fallacy it represents.

4. How to Spot Bad Arguments

Here is a quick checklist you can run mentally on any argument:

Find the conclusion. What is actually being claimed?
Find the premises. What reasons are given?
Check the structure. Does the conclusion follow from the premises? (Validity)
Check the premises. Are they actually true? What evidence supports them? (Soundness)
Check for fallacies. Is the argument relying on a logical trick instead of real reasoning?
Check for missing information. What is not being said? What assumptions are hidden?

Red flags:

Emotional language doing the work that evidence should do
Vague claims that cannot be tested (see PM-001 on the BS Test)
Arguments that attack people instead of ideas
“Everyone knows” or “It’s obvious” used as premises
False urgency (“We must decide NOW”) shutting down analysis

5. How to Construct Good Arguments

Building a good argument is the reverse of spotting a bad one:

Step 1: State Your Conclusion Clearly

Say what you are arguing for. No hedging, no burying it at the end.

“I think we should rewrite the authentication module.”

Step 2: Provide Premises That Actually Support It

Each premise should be independently verifiable and directly relevant.

“The current module has had 12 security incidents in the past year.”
“The code was written for a framework we no longer use.”
“A rewrite would take an estimated 3 weeks; patching would take longer over the next 6 months.”

Step 3: Address Counterarguments

The strongest arguments acknowledge the best case against them.

“The obvious counterargument is that rewrites are risky and often take longer than estimated. I’ve mitigated this by scoping it to authentication only and building in a 50% time buffer.”

Step 4: Be Honest About Uncertainty

If you are not sure about a premise, say so. This is not weakness — it is intellectual integrity.

“I’m confident about the security incident count (it’s in our logs). The rewrite estimate is less certain — it assumes no surprises, which is optimistic.”

Practice Exercise

Pick a decision you are currently facing (at work, in a project, in life). Write a structured argument for your preferred option using the four steps above. Then write the best counterargument you can. Does your original argument survive?

Key Terms

Term	Definition
Argument	A set of premises offered in support of a conclusion
Premise	A statement offered as evidence or reason
Conclusion	The claim that the premises are supposed to support
Validity	An argument’s conclusion follows necessarily from its premises
Soundness	An argument is valid and all its premises are true
Fallacy	A pattern of reasoning that appears convincing but is logically flawed
Ad hominem	Attacking the person rather than the argument
Straw man	Distorting an argument to make it easier to attack
False dichotomy	Presenting only two options when more exist

Self-Check Assessment

1. Is this argument valid? “All birds can fly. Penguins are birds. Therefore, penguins can fly.”

Yes, it is valid — the conclusion follows from the premises. But it is not sound, because the first premise is false (not all birds can fly). This illustrates why validity alone is not enough.

2. Someone says: “You can’t criticize this policy — you’ve never worked in government.” What fallacy is this?

Ad hominem. The quality of the criticism does not depend on the critic’s background. The argument should be evaluated on its own merits.

3. What is the difference between a valid argument and a sound argument?

A valid argument has correct logical structure — if the premises were true, the conclusion must be true. A sound argument is valid AND has premises that are actually true. Soundness is the higher standard.

4. “Either we adopt AI governance or we’ll have no governance at all.” What’s wrong with this?

False dichotomy. There are many forms of governance between “AI governance” and “no governance.” The argument artificially narrows the options.

Pass criteria: Can identify premises and conclusions in a real-world argument, correctly classify at least four fallacies, and construct a structured argument with counterargument acknowledgment.

Research Basis

Argument structure and validity/soundness from Irving Copi & Carl Cohen, Introduction to Logic (standard edition)
Fallacy taxonomy based on Aristotle’s Sophistical Refutations and updated by Douglas Walton, Informal Logic (1989)
Critical thinking pedagogy draws on Richard Paul & Linda Elder, Critical Thinking (2001)
Relevance to AI governance: tetravalent logic (T/F/U/C) extends classical true/false to handle uncertainty and contradiction — see logic/ directory
Belief state: T(0.88) F(0.02) U(0.07) C(0.03)