ProjekinAja. Clear Thinking Stage 2 · Applied Judgment Lesson 2.6
● Stage 2 · Lesson 2.6

Argument Mapping & Charitable Criticism

Map the structure, build the strongest version, apply the method. This is the complete critical-thinking toolkit — applied from start to finish on the argument we've been building toward.

⏱ ~18 min read 🗺 Visual argument map 🎮 Steel-man builder

🎯 By the end you'll be able to

Draw a visual argument map from any complex argument; build the strongest charitable version of a position you initially disagree with; apply the 14-step method to complete critical analysis; and understand the full diagnosis of the running example — including which specific errors it commits and what a more defensible version would look like.

VISUAL TOOLArgument mapping — rendering structure visually

An argument map is a visual representation of an argument's logical structure: which claims support which, which premises depend on which assumptions, and where the challenges enter. Mapping forces you to make structure explicit before evaluation — the visual format makes it much harder to glide past hidden dependencies.

The running example as an argument map

Here is the argument we've been analyzing across Stage 2, rendered as a map. Read from bottom to top: supporting elements flow upward toward the conclusion.

CONCLUSION "Investing is wrong" PREMISE 1 (moral) "Wrong to profit from another's loss" PREMISE 2 (factual) "Investment profit always needs another's loss" HIDDEN ASSUMPTION "All investment activities are the same kind" CHALLENGE 1 Productive companies create new value — not zero-sum CHALLENGE 2 Buyer & seller may both prefer the transaction CHALLENGE 3 A seller does not necessarily lose by choosing to sell CHALLENGE 4 Some instruments are zero-sum but that doesn't cover all CHALLENGE 5 Moral principle not applied consistently to claw machines supports challenges
Argument map of the running example. Solid arrows = support; dashed red arrows = challenge. All five challenges attack Premise 2; Challenge 5 also attacks Premise 1 (the moral principle).

How to read the map

The map makes several things visible that are easy to miss in prose:

💡 Why map before evaluating?

Argument maps prevent the most common critical-thinking failure: responding to the conclusion without engaging the premises, or attacking the easiest premise while ignoring the others. A complete map ensures you address the argument's actual structure, not your impression of it.

SKILLSteel-manning — before you criticize

The map above identifies weaknesses. But before we exploit them, we owe the argument a steel-man: the strongest possible version of the concern, built charitably.

Perhaps the person making the argument is not actually claiming mathematical zero-sumness for all investment. Perhaps what they are genuinely concerned about is one or more of these:

The strongest charitable version of the argument might be: "Participating in financial markets without understanding what you own and what value it creates makes it easy to profit from activities you would condemn if you understood them clearly — including extractive financial practices, speculation divorced from production, and companies with harmful externalities. Uninformed investment is morally lazy at best."

That is a much stronger and more defensible position than "every investment profit requires an equivalent loss." It is also a position that: (a) does not condemn all investment; (b) is consistent with permitting investment in genuinely productive companies; and (c) has real empirical support. After steelmanning, our evaluation needs to engage this version.

THE COMPLETE METHODThe reusable 14-step method

This is the toolkit in its final, assembled form. Apply it to any complex argument you want to evaluate seriously — arguments about policy, science, ethics, business, or personal decisions. Each step is designed to prevent a specific category of critical-thinking failure.

🔑 The 14-Step Critical Analysis Method
  1. Clarify the exact conclusion. What is actually being claimed? State it in one precise sentence. Distinguish descriptive claims from normative ones.
  2. List the explicit premises. What reasons are offered in support? Number each one separately.
  3. Expose the hidden assumptions. What would the arguer need to believe for the premises to support the conclusion? Make these explicit — they are often where the argument is weakest.
  4. Define ambiguous key terms. Every argument has at least one term that is doing more work than its definition allows. "Investment," "loss," "wrong," "always" — pin them down before proceeding.
  5. Identify the argument type. Deductive (necessarily follows)? Inductive (probably follows)? Abductive (best explanation)? This determines the appropriate standard for evaluation.
  6. Test validity or inferential strength. For deductive: does the conclusion follow necessarily from the premises? For inductive: how strong is the probabilistic support? For abductive: is this really the best explanation?
  7. Evaluate the truth and evidence for each premise. Treat each premise as a separate claim requiring support. What would make it more or less credible?
  8. Check scope words. "All," "always," "must," "never," "every" — these are extremely strong claims that require exceptional evidence. Is the scope word justified, or would "some," "often," or "typically" be more accurate?
  9. Search for counterexamples. A single genuine counterexample refutes a universal claim. Look actively for cases where the premises are true but the conclusion fails, or cases where the key premise is false.
  10. Consider alternative explanations. Is this the only way to explain the evidence? What other conclusions could the same premises support?
  11. Separate factual from moral claims. These require different kinds of evidence and different evaluation standards. An argument that conflates them cannot be properly evaluated until they are separated.
  12. Test whether the moral principle is applied consistently. If a moral principle condemns X, it must also condemn Y when Y is morally relevantly similar to X. Identify cases that should receive the same verdict and check whether they do.
  13. Construct the strongest charitable version. Before finalizing your evaluation, build the best possible version of the argument. If you can refute the steelmanned version, your critique is strong. If you can't, you haven't finished thinking.
  14. Only then assign a fallacy or bias label. Fallacy and bias labels are summaries of a diagnosis, not substitutes for one. Use them after completing the analysis, not as a shortcut to skip it. Mislabeling is worse than not labeling.

Steps 1–4 are reconstruction. Steps 5–10 are logical and empirical evaluation. Steps 11–12 are domain-specific evaluation (moral arguments). Steps 13–14 are synthesis. The sequence matters: do not jump to step 14 without completing the prior steps.

INTERACTIVESteel-man builder

Work through a sloppy argument step by step. At each stage, form your own answer before revealing the model response. This is the same method we built toward across the entire course — applied here to a fresh argument.

🏗 Interactive · Steel-Man Builder

The argument you're working with:

"Social media is destroying society. Studies show it increases anxiety and depression in teenagers, and people are more polarized than ever. We should heavily regulate or ban social media platforms."
STAGE 1 Reconstruct in standard form

Before reading the model answer: identify the conclusion, the stated premises, and any implicit premises. What scope words appear? What hidden assumptions does the argument make?

STAGE 2 Construct the strongest version

Assume the arguer had access to the best evidence. What is the most defensible, precise version of this argument?

STAGE 3 Apply steps 5–14 of the method

Evaluate the steelmanned argument: argument type, validity, premise credibility, scope, counterexamples, consistency. What does it establish? What does it not establish?

Stages revealed: 0 / 3

FINAL DIAGNOSISComplete analysis of the running example

We can now apply all 14 steps to the argument we've been building toward since the beginning of Stage 2.

StepFinding
1. Conclusion"Investing is wrong" — a normative claim about all investment activity.
2. Explicit premisesP1: wrong to profit from another's loss. P2: investment profit always requires another's loss.
3. Hidden assumptionsAll investment activities are the same kind; "loss" means financial loss in a market transaction; the moral principle (P1) is correctly specified and consistently held.
4. Ambiguous terms"Investment" (covers equity, derivatives, speculation, fraud — very different structures); "loss" (requires specifying: compared to what baseline? voluntarily accepted?); "profit" (gross or net? from what activity?)
5. Argument typeDeductive in form. P1 is normative; P2 is empirical; C follows if both are true. The argument's weakness is almost entirely in P2 and the hidden assumption in step 3.
6. ValidityIf P1 and P2 were both true as stated, C would follow deductively. The argument is formally valid in structure — the problem is the premises.
7. Premise truthP1: plausible as a rough moral intuition but underspecified (which principle?). P2: false as a universal claim — productive company ownership is positive-sum; voluntary secondary-market transactions are not necessarily zero-sum; the category "investment" contains structurally different activities.
8. Scope words"Whenever" in P2 is a universal quantifier. A single genuine counterexample — a productive company creating real value, shared by willing owners — refutes it. Several such counterexamples are easily identified.
9. CounterexamplesA bakery owner who profits from bread sales; a passive index fund investor who shares in decades of economic growth; a venture capitalist who funds a successful medical device company. All are "investment," none are zero-sum.
10. Alternative explanationsThe evidence that inspires P2 (fraud, speculation, market manipulation) is real but does not require the universal generalization. The more parsimonious conclusion is that some investment is zero-sum or exploitative, not all.
11. Factual vs moralP1 is moral; P2 is factual. Separating them shows: even if P1 were uncontroversial, the argument fails because P2 is empirically too broad. Conversely, even if P2 were accurate for some activities, P1 still requires specification (which of Principles A, B, or C?).
12. Moral consistencyP1 as read (Principle A) would also condemn claw machines, salary negotiation, and any sale at a profit — activities the arguer likely permits. The principle is applied selectively to investment but not to structurally similar activities. This does not refute the conclusion, but it demonstrates the principle is not the arguer's actual belief.
13. Steel-manThe strongest version: "Speculative financial activity that extracts value without producing it concentrates wealth unfairly, and ordinary participation without understanding enables this. Investors should understand what they own and what value it creates." This is a defensible, narrower position that survives the counterexamples to P2.
14. Fallacy / bias labelP2 reflects zero-sum bias (the cognitive tendency to perceive all competitive situations as fixed-sum). The argument also commits a hasty generalization (inferring a universal claim from some cases) and a composition / category error (applying properties of some financial activities to all). The claw-machine point alone would be tu quoque — but it is legitimately used to test moral consistency.
🔑 The bottom line

The argument is not formally invalid — if both premises were true, the conclusion would follow. The problem is that Premise 2 is empirically false as a universal claim, and Premise 1 is underspecified (it is most naturally read as a principle that fails universalization). The psychology behind the argument is zero-sum bias — a documented cognitive tendency to perceive economic interactions as fixed-sum that is well-evidenced in behavioral economics research. In argumentative terms, the argument commits hasty generalization and a category error. The steelmanned version — a concern about speculation, exploitation, and uninformed participation — is more defensible, but it is not what the original argument says.

WEAKNESSCommon mistakes in argument mapping and criticism

✗ Skipping the map and jumping to the label.

"That's just zero-sum bias" or "that's a hasty generalization" as a first response — without working through the argument's structure — is a shortcut that prevents real understanding. The label is a summary of a diagnosis, not a substitute for one. If you can't explain exactly what the argument says, where it goes wrong, and what a stronger version would look like, you haven't completed the analysis.

✗ Steelmanning as performance rather than practice.

A pro forma "I'll steelman this: perhaps they mean..." followed by a slightly-less-weak version is not steelmanning. The test is: could the original arguer read your steelmanned version and say "yes, that captures my concern better than I stated it"? If not, you haven't built the strongest version — you've built a version that is easy to defeat.

✗ Treating a valid formal structure as a sound argument.

Validity (the conclusion follows from the premises) is a necessary but not sufficient condition for soundness. An argument can be valid and completely useless because a premise is false. Always evaluate the premises separately from the logical structure.

✗ Stopping at "the argument is flawed."

Finding flaws is only the diagnostic half of critical thinking. The productive half is: what would a better argument for the same concern look like? Charitable criticism identifies what is worth preserving from the original concern and builds on it, rather than simply dismissing.

PRACTICE🔨 Weekly exercise

This week's task — and an ongoing practice

Find a strong argument you disagree with — in a news article, a discussion, a book, or a conversation — and apply all 14 steps. Write your analysis down. Focus especially on steps 13 and 14: build the strongest charitable version before you assign any labels.

The discipline of mapping before evaluating, steelmanning before criticizing, and labeling only after diagnosing is the complete method. It takes longer than the usual "I disagree because..." response. It produces better thinking.

Carry these three habits forward from this course:

  1. Reconstruct before responding. Make the argument's structure explicit.
  2. Steelman before criticizing. Engage the strongest version, not the easiest.
  3. Diagnose before labeling. Fallacy names are conclusions, not evidence.

GLOSSARYKey terms introduced

Argument map
A visual representation of an argument's logical structure — which claims support which, which assumptions are hidden, and where challenges enter.
Steel man
The strongest possible version of an argument, built charitably — the opposite of a straw man. Always construct this before evaluating.
Soundness
An argument is sound when it is both valid (conclusion follows from premises) and has true premises. Validity alone is not sufficient for a good argument.
Zero-sum bias
The cognitive tendency to perceive economic and competitive interactions as fixed-sum (one side must lose what the other gains), even when they are positive-sum.
Hasty generalization
Drawing a universal conclusion from an insufficient number of cases — "some X are Y, therefore all X are Y."
Composition / category error
Treating qualitatively different things as if they are the same kind, or assuming a property of parts applies to the whole, or vice versa.
Tu quoque
"You too" — the fallacy of dismissing an argument because the person making it behaves inconsistently. Distinguished from a legitimate consistency challenge, which demands explanation of the morally relevant difference.
Abductive reasoning
Inference to the best explanation — choosing the hypothesis that best accounts for the available evidence. Not certain, but can be highly warranted.

ACTIVE RECALLCheck your understanding

Answer in your own words first, then reveal.

1What does an argument map show that a prose reconstruction does not? Give one specific benefit for evaluating a complex argument.›

An argument map makes the hierarchical structure of support and challenge visible: which premises support which conclusions, which assumptions underlie which premises, and where each challenge applies. One benefit: it prevents the common error of responding to the conclusion without engaging the premises — the map forces you to identify which premise a challenge attacks, rather than vaguely objecting to "the argument." It also makes hidden assumptions visible as nodes, forcing them to be evaluated explicitly rather than slipped past.

2In the 14-step method, why does step 14 (assigning fallacy labels) come last rather than first?›

Fallacy labels are summaries of a diagnosis, not the diagnosis itself. Jumping to a label ("that's just zero-sum bias") without completing the prior steps means you've identified a pattern but not verified that it actually applies to this argument's specific structure and specific premises. Premature labeling is also a form of dismissal — it stops thinking rather than structuring it. Only after completing reconstruction (steps 1–4), evaluation (5–12), and steelmanning (13) can you accurately describe what errors the argument commits and why the labels apply.

3Why is the running argument formally valid but unsound? Name the specific flaw in each of the two premises.›

Formally valid: if P1 and P2 were both true, C would follow necessarily — the logical structure is not broken. Unsound because: P2 is false as a universal claim — investing in productive companies is positive-sum, voluntary transactions can benefit both parties, and different financial activities have fundamentally different structures. P1 is underspecified — the moral principle it expresses (Principle A) fails the universalization test and is not applied consistently to structurally similar activities. Neither premise survives scrutiny, so the conclusion is not established.

4What are the three cognitive/argumentative errors the running argument commits? (Use the step 14 labels and explain briefly.)›

Zero-sum bias: the empirical belief that investment is fixed-sum reflects a documented cognitive tendency to perceive economic competition as zero-sum even when production creates new value. Hasty generalization: inferring that because some financial activities (fraud, pure speculation) involve wealth transfer at another's expense, all investment does — drawing a universal conclusion from insufficient cases. Composition / category error: treating qualitatively different financial activities (equity ownership, derivatives, fraud, gambling) as a single category and applying a verdict valid for some to all of them.

You've completed Clear Thinking. The method is yours to keep. →