Gsd-skill-creator critical-reading
Analytical evaluation of texts for argument quality, author purpose, bias, credibility, and rhetorical strategy. Covers the distinction between comprehension and critique, argument analysis (claims, evidence, reasoning, warrants), rhetorical appeals (ethos, pathos, logos), bias identification (selection bias, confirmation bias, loaded language, omission), source credibility evaluation (CRAAP test, lateral reading), propaganda techniques, media literacy, and multi-text synthesis. Use when evaluating arguments, identifying bias, assessing source reliability, analyzing rhetoric, or synthesizing across multiple texts on the same topic.
git clone https://github.com/Tibsfox/gsd-skill-creator
T=$(mktemp -d) && git clone --depth=1 https://github.com/Tibsfox/gsd-skill-creator "$T" && mkdir -p ~/.claude/skills && cp -r "$T/examples/skills/reading/critical-reading" ~/.claude/skills/tibsfox-gsd-skill-creator-critical-reading && rm -rf "$T"
examples/skills/reading/critical-reading/SKILL.mdCritical Reading
Critical reading is the practice of evaluating what a text claims, how it supports those claims, and whose interests it serves. It goes beyond comprehension -- understanding what a text says -- to judgment: is what it says true, well-supported, fair, and complete? In an information environment saturated with persuasion, misinformation, and motivated reasoning, critical reading is not an academic luxury but a survival skill.
Agent affinity: achebe (postcolonial reading, "An Image of Africa," whose story gets told), austen (irony detection, social observation), morrison (race and narrative, Playing in the Dark)
Concept IDs: read-author-purpose-perspective, read-argument-analysis, read-evaluating-bias, read-literary-analysis
Comprehension vs. Critique
Comprehension asks: "What does this text say?" Critique asks: "Should I believe it, and on what terms?" These are sequential, not competing, operations. A reader who critiques before comprehending misrepresents the text. A reader who comprehends without critiquing accepts claims uncritically.
Reading order:
- First, understand the text on its own terms (comprehension).
- Then, evaluate its claims, reasoning, and perspective (critique).
- Finally, situate it among other texts on the same topic (synthesis).
Argument Analysis
The Structure of Arguments
Every argument has three components:
| Component | Definition | Question to ask |
|---|---|---|
| Claim | The assertion the author wants you to accept | "What is the author arguing?" |
| Evidence | The facts, data, examples, or testimony offered in support | "What support does the author provide?" |
| Reasoning | The logical connection between evidence and claim | "How does the evidence actually support the claim?" |
Toulmin (1958) adds three more components for full analysis:
| Component | Definition | Question |
|---|---|---|
| Warrant | The underlying assumption connecting evidence to claim | "What unstated belief makes this argument work?" |
| Qualifier | The degree of certainty ("most," "probably," "in many cases") | "How strongly does the author state the claim?" |
| Rebuttal | Acknowledgment of counter-arguments | "Does the author address objections?" |
Worked example. "Schools should start later because adolescents' circadian rhythms shift during puberty, making early mornings biologically misaligned with teenage sleep needs (Carskadon, 2002). Districts that delayed start times saw improved attendance and reduced car accidents."
- Claim: Schools should start later.
- Evidence: Circadian rhythm research; district outcome data.
- Reasoning: Biological misalignment causes harm; delayed starts reduce that harm.
- Warrant (unstated): School policy should align with biological needs when possible.
- Qualifier: Implicit "should" -- a recommendation, not a certainty.
- Rebuttal: Not addressed -- what about transportation logistics, after-school jobs, childcare?
Common Logical Fallacies
| Fallacy | What it looks like | Why it fails |
|---|---|---|
| Ad hominem | Attacking the person, not the argument | A person's character does not determine the truth of their claim |
| Straw man | Misrepresenting the opposing view to make it easier to attack | Defeating a distorted version proves nothing about the real argument |
| False dilemma | Presenting only two options when more exist | Eliminates nuance and middle positions |
| Appeal to authority | "Expert X says so, therefore it's true" | Expertise increases credibility but does not guarantee truth |
| Hasty generalization | Drawing broad conclusions from limited examples | Insufficient sample |
| Slippery slope | Claiming one step inevitably leads to an extreme outcome | Causal chain is assumed, not demonstrated |
| Red herring | Introducing an irrelevant topic to distract | Deflects from the actual argument |
| Circular reasoning | The conclusion is assumed in the premise | "This book is good because it's well-written" defines good as well-written |
Rhetorical Appeals
Aristotle identified three modes of persuasion that remain the foundation of rhetorical analysis:
Ethos (Credibility)
The speaker's character, expertise, and trustworthiness. Readers should ask: "What qualifies this author to make this claim? Do they have relevant expertise? Do they have a financial or ideological stake?"
Pathos (Emotion)
Appeals to the audience's feelings -- fear, sympathy, outrage, hope. Emotional appeals are not inherently fallacious (suffering is real and worth responding to), but they become manipulative when they substitute for evidence or distract from weak reasoning.
Logos (Logic)
Appeals to reason through evidence, data, and logical argument. The strongest arguments combine all three appeals, but logos should bear the weight of the claim. An argument that relies primarily on pathos or ethos without logos is rhetorically effective but intellectually weak.
Bias Identification
All texts have perspective. The question is not whether a text is biased (all are) but whether the bias distorts the text's claims or omits essential information.
Types of Bias
| Bias type | How it works | What to look for |
|---|---|---|
| Selection bias | Choosing which facts to include and which to omit | What is NOT mentioned? What would a critic add? |
| Confirmation bias | Seeking evidence that supports a pre-existing view | Does the author consider disconfirming evidence? |
| Loaded language | Using words with strong connotations to influence perception | "Freedom fighters" vs. "insurgents" -- same people, different framing |
| Source bias | Relying on sources that share the author's viewpoint | Are sources diverse? Do they include opposing perspectives? |
| Omission | Leaving out context that would change interpretation | What background information is missing? |
| Framing | Presenting the same facts in a way that favors one interpretation | "Unemployment fell to 4%" vs. "96% of workers found employment" |
Achebe's Critical Lens
Chinua Achebe's "An Image of Africa" (1977) provides a model for reading bias at the cultural level. Achebe demonstrated that Joseph Conrad's Heart of Darkness, widely taught as a masterpiece, systematically dehumanizes Africa and Africans -- reducing an entire continent to a backdrop for European psychological drama. Achebe's analysis asks: "Whose humanity is assumed? Whose is denied? Whose story is being told, and whose is being erased?"
This lens extends beyond postcolonial literature. Every text constructs some perspectives as default and others as marginal. Critical reading notices these constructions.
Source Credibility Evaluation
The CRAAP Test
A widely used framework for evaluating sources (Meriam Library, CSU Chico):
| Criterion | Questions |
|---|---|
| Currency | When was this published? Has it been updated? Is the information still accurate? |
| Relevance | Does this relate to my question? Who is the intended audience? |
| Authority | Who created this? What are their credentials? Is the publisher reputable? |
| Accuracy | Is the information supported by evidence? Can it be verified? Are there errors? |
| Purpose | Why does this exist? To inform, persuade, sell, entertain? Is the purpose stated or hidden? |
Lateral Reading
Wineburg and McGrew (2019) found that professional fact-checkers evaluate sources differently from students and professors. Instead of reading a source deeply (vertical reading), fact-checkers immediately leave the source and check what other sources say about it (lateral reading). They ask: "What do independent, knowledgeable sources say about this author/organization/claim?"
Lateral reading is faster and more reliable than vertical reading for credibility assessment because it escapes the source's own framing.
Propaganda Techniques
Propaganda is persuasion that prioritizes the persuader's goals over the audience's understanding. Seven classic techniques (Institute for Propaganda Analysis, 1937):
| Technique | Method | Example |
|---|---|---|
| Name-calling | Attaching a negative label | "Elitist," "radical," "snowflake" |
| Glittering generality | Attaching a positive but vague label | "Freedom," "family values," "common sense" |
| Transfer | Associating with a respected symbol | Using flags, religious imagery, or scientific vocabulary |
| Testimonial | Celebrity or authority endorsement | "Nine out of ten doctors recommend..." |
| Plain folks | Presenting as ordinary, relatable | "I'm just a regular person who..." |
| Bandwagon | Claiming everyone agrees | "Millions have already joined..." |
| Card stacking | Presenting only favorable evidence | Showing poll results without methodology |
Multi-Text Synthesis
Critical reading reaches its highest form when a reader synthesizes across multiple texts on the same topic. Synthesis requires:
- Identifying each text's claim and evidence.
- Mapping agreements and disagreements. Where do the texts converge? Where do they diverge?
- Evaluating relative credibility. Which source has better evidence? More relevant expertise? Less conflict of interest?
- Constructing an integrated understanding that accounts for multiple perspectives rather than accepting any single source.
This is the reading equivalent of the math investigation team's synthesis protocol: converging findings are strengthened, diverging findings are preserved and investigated, and the reader constructs a richer understanding than any single text provides.
When to Use This Skill
- Evaluating arguments in persuasive or opinion texts
- Identifying bias in news, research, or historical sources
- Assessing source credibility for research purposes
- Analyzing rhetorical strategies in speeches, advertisements, or political discourse
- Synthesizing multiple sources on a contested topic
- Teaching media literacy
When NOT to Use This Skill
- For basic comprehension strategies -- use reading-comprehension
- For vocabulary and word learning -- use vocabulary-development
- For phonics and decoding -- use phonics-decoding
- For literary interpretation of fiction -- use literary-analysis (though critical-reading and literary-analysis overlap on author purpose and craft)
- For research process and search strategies -- use information-literacy
Cross-References
- achebe agent: Postcolonial critical reading. Achebe models how to read for cultural bias, representation, and whose story is being told.
- austen agent: Irony detection and social observation. Austen's irony requires the reader to hold the surface meaning and the critical subtext simultaneously -- a core critical reading skill.
- morrison agent: Race, narrative voice, and the construction of otherness. Morrison's Playing in the Dark is a masterclass in reading what a text reveals about its own cultural assumptions.
- reading-comprehension skill: Comprehension precedes critique. You must understand what a text says before evaluating whether it is true.
- information-literacy skill: Source evaluation and research skills extend critical reading into the research process.
References
- Achebe, C. (1977). An image of Africa: Racism in Conrad's Heart of Darkness. Massachusetts Review, 18(4), 782-794.
- Aristotle. Rhetoric. (Kennedy, G. A., Trans., 2007). Oxford University Press.
- Institute for Propaganda Analysis. (1937). How to detect propaganda. Propaganda Analysis, 1(2), 1-4.
- Toulmin, S. E. (1958). The Uses of Argument. Cambridge University Press.
- Wineburg, S., & McGrew, S. (2019). Lateral reading and the nature of expertise: Reading less and learning more when evaluating digital information. Teachers College Record, 121(11), 1-40.