Formal Logic — The Architecture of a Valid Argument

Overview

Logic is not about being clever in arguments. It is the engineering discipline of thought — the study of which conclusions are actually held up by their reasons and which only appear to be. Once you can see the structure underneath an argument, you cannot unsee it. You will catch the load-bearing wall and the decorative trim, and you will know the difference between a claim that is true and a claim that is merely well-decorated.

This lesson teaches you to take any argument apart, lay its pieces on the table, and judge them. By the end you will be able to do something most adults cannot: agree that an argument is perfectly valid and still reject its conclusion because one of its premises is false — and explain exactly which one and why.

Background for Parents

This is the conceptual spine of the entire Agency pillar at this stage. The single hardest idea here — and the one that unlocks everything else — is the separation of validity from truth. A valid argument is one whose structure guarantees that if the premises are true, the conclusion must be true. Validity says nothing about whether the premises actually are true. A sound argument is both valid and built from true premises. Most bad reasoning your teenager will encounter is not invalid; it is valid-but-unsound — a clean logical structure resting on a smuggled-in false premise.

Your role is advise. Do not lecture through this. The learner can read and work the material independently. Where you add value is as a sparring partner: bring them a real argument from your own week — something from work, the news, a salesperson — and let them dissect it. Resist the urge to supply answers. Ask, "Which premise is doing the work? What happens if it is false?" The most common misconception to watch for is the belief that calling something "a fallacy" wins an argument. It does not. A fallacy means the reasoning fails; the conclusion might still be true for other reasons. Teach them to attack the reasoning, not to declare victory.

Lesson Flow

Session 1: Validity Is Not Truth (75 minutes)

Opening (15 minutes)

Read this argument aloud:

All cats are reptiles. All reptiles can fly. Therefore, all cats can fly.

Every statement in it is false. And yet the argument is perfectly valid. If the two premises were true, the conclusion would be forced — there would be no way to escape it. The structure is flawless; the inputs are garbage.

Now read this one:

Some of the greatest scientists in history were religious. Newton was one of the greatest scientists in history. Therefore, Newton was religious.

This one feels more respectable, and the conclusion happens to be true. But the argument is invalid — the conclusion does not follow. "Some" scientists being religious does not force any particular scientist to be. Newton's faith is established by historical fact, not by this argument.

Sit with the discomfort. A false-everything argument can be valid. A true-conclusion argument can be invalid. Validity lives in the structure. Truth lives in the content. They are two separate dials, and you must learn to read each one independently.

Core Instruction (40 minutes)

1. Premises and conclusion. Every argument is one conclusion plus one or more premises offered as support. Your first job, always, is to find the conclusion — the thing being argued for — and then identify the premises offered to hold it up. Signal words help: "therefore," "so," "thus," "which means" usually precede a conclusion; "because," "since," "given that" usually precede a premise. But signal words are unreliable. The real test is: which claim is this person trying to get me to accept, and which claims are they using to do it?

2. The four combinations. Lay these out in your notebook:

   	Premises true	Premises false
   Valid structure	Sound — accept the conclusion	Valid but unsound — reject the conclusion, attack the premise
   Invalid structure	Unsound — conclusion not earned	Unsound — conclusion not earned

   Only the top-left box earns your agreement. Every other box is a place where an argument fails, and your job is to identify which box a given argument lands in.

3. Validity defined precisely. An argument is valid when it is impossible for all the premises to be true and the conclusion false at the same time. That is the whole definition. To test for validity, you do not ask "is this true?" You ask: "Can I imagine a world where every premise is true but the conclusion is still false?" If you can, the argument is invalid. If you genuinely cannot, it is valid.

4. Deductive versus inductive. There are two great families of argument, and confusing them is the source of endless bad reasoning. A deductive argument claims its conclusion is guaranteed by its premises — if the premises hold, the conclusion must hold, with no gap. Everything above is deductive. An inductive argument claims only that its premises make the conclusion probable — "the sun has risen every day of recorded history, so it will rise tomorrow." Induction is how science and everyday life actually work, because we almost never have perfect premises. The key insight: an inductive argument is never "valid" or "invalid" in the strict sense — it is strong or weak, depending on how well the evidence supports the conclusion and how representative it is. When someone demands you "prove" something about the real world with certainty, they are misapplying a deductive standard to an inductive question. Most real knowledge is built from strong induction, not airtight deduction, and a mature thinker knows which kind of claim is on the table.

Practice (20 minutes)

Take the three or four arguments you harvested from the wild. For each one, in your notebook:

• Circle the conclusion.
• Underline each premise.
• Build the table above for it: is the structure valid? Are the premises true?
• Decide which box it lands in.

Do not yet try to win against these arguments. Just classify them. Accuracy of diagnosis comes before the cure.

Session 2: The Workhorse Forms (75 minutes)

Opening (10 minutes)

Most everyday deductive reasoning runs on a handful of forms. Two of them do most of the work, and two more imitate them well enough to fool nearly everyone. Learn all four together — the valid ones next to their counterfeits — because you will almost never meet one without the other nearby.

Core Instruction (45 minutes)

Write P for any statement and Q for any statement. The form is what matters, not the content.

1. Modus ponens (valid):

   • If P, then Q.
   • P.
   • Therefore, Q.
   • "If it rained, the street is wet. It rained. So the street is wet." Airtight.

2. Modus tollens (valid):

   • If P, then Q.
   • Not Q.
   • Therefore, not P.
   • "If it rained, the street is wet. The street is not wet. So it did not rain." Also airtight, and the one people use least. Train yourself to reach for it — denying the consequence is one of the most powerful moves in clear thinking.

3. Affirming the consequent (INVALID — the counterfeit of modus ponens):

   • If P, then Q.
   • Q.
   • Therefore, P.
   • "If it rained, the street is wet. The street is wet. So it rained." Wrong. A street sweeper, a burst pipe, or a sprinkler could have wet the street. This fallacy is everywhere — it is how superstition, conspiracy thinking, and bad medical reasoning all operate. "If the treatment worked, you would feel better. You feel better. So the treatment worked." No — you might have recovered on your own.

4. Denying the antecedent (INVALID — the counterfeit of modus tollens):

   • If P, then Q.
   • Not P.
   • Therefore, not Q.
   • "If it rained, the street is wet. It did not rain. So the street is not wet." Wrong, for the same reason — something else could have wet it.

The two valid forms work on the certain end (P guarantees Q; the absence of Q guarantees the absence of P). The two fallacies work on the uncertain end (Q does not guarantee P, because Q can have other causes). When you see an "if-then" argument, your reflex should be: which of the four am I looking at?

Practice (20 minutes)

Write four short arguments of your own — one in each form — about something real in your life. Then swap the valid and invalid versions and read them aloud to a parent or sibling without telling them which is which. Notice how natural the fallacies sound. That naturalness is exactly why they fool people.

Session 3: Informal Fallacies — How Reasoning Fails in the Real World (90 minutes)

Opening (10 minutes)

Formal fallacies break the structure. Informal fallacies are sneakier: the structure can look fine, but the content cheats. These are the ones you will meet a hundred times a day — in advertising, in political speech, in your friends' arguments, and, most importantly, in your own head when you want something to be true.

Core Instruction (50 minutes)

Build a reference page in your notebook with one example of each, drawn from your own life:

• Ad hominem — attacking the person instead of their argument. "You can't trust her budget analysis, she's never even had a real job." The job history is irrelevant to whether the numbers add up.
• Straw man — replacing your opponent's actual argument with a weaker, distorted version and defeating that. "You want a later curfew, so you basically want no rules at all." Defeating a claim no one made.
• Appeal to authority (the bad kind) — citing someone's status as proof, when they are outside their expertise or the matter is genuinely contested. A famous actor endorsing a supplement is not evidence about the supplement.
• False dilemma — presenting two options as if they are the only two. "Either we cut the program entirely or we go bankrupt." Usually there is a third, fourth, and fifth door.
• Slippery slope — claiming one step inevitably leads to an extreme outcome, without showing the links. "If we let students retake one test, soon no one will study for anything." The chain is asserted, not demonstrated.
• Appeal to emotion — substituting a feeling for a reason. Fear, pity, and pride are powerful; none of them make a claim true.
• Bandwagon — "everyone thinks this, so it's right." The number of believers is not evidence of correctness.
• Begging the question — assuming the conclusion inside one of the premises. "This policy is the right thing to do because it's what a just society would choose." That just restates the claim.

Two more deserve their own attention because they are the favorite tools of people trying to manipulate you:

• Equivocation — using the same word in two different senses inside one argument, so the conclusion only "follows" if you do not notice the switch. "Evolution is just a theory, and a theory is just a guess." The word "theory" means a casual guess in everyday speech and a rigorously tested explanatory framework in science. The argument trades on the slide between the two. Whenever a key word seems to be carrying the whole argument, stop and ask: does this word mean the same thing every time it appears?
• The motte and bailey — defending an extreme claim (the "bailey," the exposed and valuable ground) by retreating, when challenged, to a modest claim that is easy to defend (the "motte," the fortified tower), then advancing back to the extreme claim once the challenger backs off. Someone makes a sweeping assertion, you push back, and suddenly they say "I only meant that we should care about this" — which no one would dispute — and then minutes later they are back to the sweeping version. Watch for the position that shrinks under pressure and re-inflates the moment pressure lifts.

Practice (30 minutes)

Take one piece of persuasive writing — a real op-ed or a long advertisement — and annotate it line by line. Mark every fallacy you find. Then do the harder thing: find the parts that are genuinely well-reasoned. Most real arguments are a mix. The skill is not "everything is a fallacy"; it is precision about which parts hold and which do not.

A warning that separates serious thinkers from sophomores: spotting a fallacy does not refute a conclusion. This is itself a fallacy — sometimes called the fallacy fallacy. If your friend gives a bad argument that the bridge is unsafe, the bridge does not become safe because their reasoning was sloppy. The bad argument fails to establish the conclusion; the conclusion might still be true on other grounds. So the correct response to a fallacy is never "therefore you're wrong" — it is "that argument doesn't work; do you have a better reason?" Use the fallacy tools to evaluate reasoning, never to award yourself a win.

Session 4: Steelmanning and Synthesis (60 minutes)

Opening (10 minutes)

Anyone can knock down a weak version of an argument. The mark of a serious thinker is the opposite move: the steel man. Before you criticize a position, restate it in its strongest form — stronger than its own advocates usually manage — and only then respond. If you can defeat the best version, your criticism means something. If you can only defeat the worst version, you have learned nothing and persuaded no one.

Core Instruction (30 minutes)

Pick a position you genuinely disagree with — a real one, on a real issue you care about. In your notebook:

1. Write the strongest case for it that you can construct. Use real premises a thoughtful person would actually hold. No distortion, no mockery. This should be hard and slightly uncomfortable.
2. Now examine your own steel man with the tools from this lesson. Is it valid? Which premises are doing the work? Are they true?
3. Write your response — not to the cartoon version, but to the strong version you just built.

This is the entire discipline distilled: understand before you judge, attack the strongest form, separate structure from content.

Practice and Closing (20 minutes)

Read your steel man aloud to your parent. Ask them to push back. Defend a position you do not even hold, using only valid reasoning, for ten minutes. Then drop it and say what you actually think. Notice that you understand your own view better now that you have inhabited its opposite.

Assessment

• Learner can state the difference between validity and truth and give an example of a valid argument with a false conclusion and an invalid argument with a true conclusion
• Learner can take an unfamiliar argument, identify the conclusion and premises, and judge whether the conclusion follows
• Learner can recognize modus ponens and modus tollens and distinguish them from affirming the consequent and denying the antecedent
• Learner can name at least six informal fallacies and point one out in live conversation without prompting
• Learner can construct a steel man of a view they oppose that the opposing side would recognize as fair

Adaptations

• Simpler: Spend two full sessions on validity-versus-truth alone before introducing the named forms. Mastery of the core distinction matters far more than memorizing fallacy names.
• More challenging: Introduce truth tables and basic symbolic notation (∧, ∨, →, ¬). Have the learner prove the validity of modus tollens formally and show why affirming the consequent fails using a truth table.
• Different setting: Run the whole thing as a two-person debate club with a friend or sibling. Argue assigned sides, then switch. Logic is best learned against a live opponent.

Going Deeper

• Read Being Logical by D.Q. McInerny — short, clear, and exactly at this level.
• For the brave: An Illustrated Book of Bad Arguments by Ali Almossawi (free online), which puts a memorable picture to every fallacy here.
• Keep a "fallacy of the week" log for a month. Catch one in the wild, write it down with the source, and name it. The log will outlast the lesson.