Rhetoric and composition theory — from Aristotle to modern argument

Intuition Beginner

Every time you try to change someone's mind, you are doing rhetoric. A lawyer urging a jury to acquit, a friend convincing you to see a film, an advertisement selling a phone — each one builds an argument meant to move a particular audience. Rhetoric is the study of how that building works, and how to do it on purpose instead of by accident.

The Greek thinker Aristotle noticed that persuasive speakers lean on three different things at once. Ethos is the speaker's credibility — the reasons an audience should trust them. Pathos is feeling — making the audience care. Logos is reasoning — the evidence and the logic that connect a claim to proof. A strong argument uses all three; a weak one leans too hard on just one and collapses the moment that leg is knocked out.

Rhetoric also asks a handful of practical questions before you write a single word. Who is reading or listening? What do you want them to do or believe? And what situation are you all in together? Together these are the rhetorical situation: audience, purpose, context. Change any one of them and the whole argument has to change with it — a speech that wins over a town hall will fall flat in a courtroom.

Finally, classical rhetoric hands the writer a workshop of five stages, called the canons: invention (find what to say), arrangement (put it in order), style (choose the words), memory (hold it in mind), and delivery (voice, gesture, timing). You move through the canons whenever you brainstorm, outline, draft, rehearse, and present — even if you never call them by their Latin names.

Visual Beginner

The Aristotelian triad — three legs a persuasive argument stands on:

            ETHOS   (Who speaks? Can we trust them?)
              |
        +-----+-----+
        |           |
     PATHOS  <-->  LOGOS
    (feeling)     (reason)

The five canons (the stages of building any argument):
  1. INVENTION     - find what to say
  2. ARRANGEMENT   - put it in order
  3. STYLE         - choose the words
  4. MEMORY        - hold it in mind
  5. DELIVERY      - voice, gesture, timing

The rhetorical situation (match the argument to all three):
  AUDIENCE  <-->  PURPOSE  <-->  CONTEXT
   (to whom?)      (to do what?)  (in what setting?)

Worked example Beginner

Here is a one-sentence argument you might find in an opinion column:

"The city should fund all-day bus service, because ridership has risen 40 percent in two years and working riders depend on buses to reach their jobs."

We will break it apart using the Toulmin model, which splits an argument into pieces small enough to test one at a time.

• Claim (the point being urged): The city should fund all-day bus service.
• Data (the evidence offered): Ridership is up 40 percent in two years; working riders depend on buses to reach their jobs.
• Warrant (the bridge from data to claim): When demand for a public service rises and citizens depend on it, government should fund that service.
• Backing (why the warrant holds): Municipal governments exist to provide the transit residents rely on for work.
• Qualifier (how strong the claim is): Likely — unless the budget cannot bear it this year.
• Rebuttal (the honest exception): If the deficit is too large, funding may have to wait.

Once an argument is split this way, each piece can be attacked or defended on its own. A skeptic might grant the data but reject the warrant, arguing that rising demand does not oblige the city to pay. That is the power of the model: it shows where reasonable people disagree, instead of letting the argument stay a single muddy blob.

Check your understanding Beginner

Formal definition Intermediate+

Rhetoric, in Aristotle's definition, is "the faculty of observing, in any given case, the available means of persuasion" ^{[Aristotle 350 BCE]}. It is not a single technique but a discipline that studies how belief and action are produced under conditions of incomplete certainty.

A rhetorical situation (Bitzer 1968) is a triple

$$\mathcal{R}=(A,P,C)$$

consisting of an audience $A$ (the people whose belief or action the rhetor seeks to change), a purpose $P$ (the specific change sought), and a context $C$ (the occasion, prior discourse, and material constraints). An argument $\alpha$ is fitting for $\mathcal{R}$ when it is adapted to each component; a mismatch in any single one — wrong audience, wrong moment, unclear aim — defeats the other two.

The three pisteis. Aristotle partitions the available means of persuasion into three families: ethos (the credibility the speaker establishes), pathos (the disposition into which the audience's emotions are brought), and logos (the argument itself, its proofs and reasonings).

The five canons. Classical rhetoric organizes the production of discourse into five stages: invention (discovering arguments), arrangement (ordering them), style (diction and figures), memory (retention), and delivery (voice and gesture). These canons name a workflow rather than a fixed sequence; modern writers move among them recursively.

Stasis theory (Hermagoras, c. 150 BCE; transmitted by Cicero ^[Cicero]) is a procedure that locates the precise point of disagreement before arguments are built. Four stases are examined in order: conjecture (an sit — did it happen?), definition (quid sit — what is it?), quality (quale sit — was it just or unjust?), and translation (ad aliam quaestionem — is this the right forum at all?). Settling an earlier stasis closes the later ones, so a debate that skips stasis tends to argue past itself.

The Toulmin model (Toulmin 1958 ^{[Toulmin 1958]}) represents an argument as a six-part structure

$$(\text{Claim},\text{Data},\text{Warrant},\text{Backing},\text{Qualifier},\text{Rebuttal}),$$

in which the warrant $W$ licenses the inference from data $D$ to claim $C$. Formally, the model asserts that if $W$ holds, then $D$ licenses $C$ up to the strength of the qualifier $Q$ (e.g. probably, necessarily, presumably) against the rebuttal $R$. Toulmin's contribution was to insist that the warrant is itself a claim that can be examined, and that different argument fields admit different warrants.

Rogerian argument (Rogers 1951; codified for composition by Young, Becker, and Pike 1970) replaces the adversarial structure with four moves: (i) state the opposing view so faithfully that its holders would assent to the statement; (ii) list the conditions under which that view is valid; (iii) state one's own position; (iv) show how those who hold the opposing view would benefit from adopting it. The goal is not victory but a solution both sides can inhabit.

Genre (Bakhtin 1986; Miller 1984) is a socially recognized, recurrent rhetorical action with stabilizing formal features. A lab report, a legal brief, and a grant proposal are genres, each with its own audience expectations, evidence standards, and arrangement conventions. Genre theory holds that successful writing is partly a matter of recognizing which genre a situation calls for and then meeting (or deliberately stretching) its conventions.

Rhetorical analysis Intermediate+

This section applies the vocabulary above to real arguments and shows how the pieces interact — the work a rhetorician actually does.

Close reading: a one-paragraph argument

Consider a fundraising letter from a homeless shelter:

"Last winter our shelter turned away 300 families on the coldest nights. Your gift of $50 buys a family a week of warmth. Please give today — because no child should sleep in the cold."

Decomposed by appeal:

• Ethos: the shelter speaks with first-hand authority ("our shelter turned away..."); the figure 300 signals it has counted, not guessed.
• Pathos: "no child should sleep in the cold" names an image the audience cannot comfortably hold while doing nothing.
• Logos: a quantitative claim — $50 buys a week of warmth — plus an implied warrant that a small gift produces a measurable good. - Rhetorical situation: the audience is warm, comfortable, reading mail in autumn; the purpose is a donation; the context is the approaching winter. The argument is fitting because each element is tuned to that exact moment and reader.

Stasis in action

The question "Should universities abolish legacy admissions?" can stall in public debate because participants occupy different stases. One side argues conjecture — do legacy preferences measurably change who is admitted? Another argues definition — is "legacy" a merit category or a demographic one? A third argues quality — is it just? Productive disagreement requires the parties to agree which stasis they are in; otherwise each side scores points the other never contested, and the debate looks polarized when it is merely misaligned.

Toulmin: where disagreement lives

Returning to the bus-service example, the Toulmin model predicts where a debate will fracture. Data are testable; warrants are ideological. Two reasonable readers can accept identical ridership figures but disagree on the warrant that rising demand obliges public funding. The model thus converts a vague impression ("we just disagree") into a precise locus of dispute — and once the dispute is located, it can be argued on its own terms instead of re-fought at the surface.

Genre as constraint

A researcher who writes the same finding as a journal article, a press release, and a tweet is working in three genres with three different audiences and evidence standards. The article foregrounds method and citation; the press release foregrounds the human stakes; the tweet foregrounds a single quotable line. None is a "dumbed-down" version of another; each is a fitting argument for its own rhetorical situation. Rhetorical competence is partly the ability to recognize the genre a situation calls for and to honor its conventions without being trapped by them.

Bridge. This decomposition of an argument into claim, data, and warrant builds toward the modern composition-theory account of writing as a recursive, audience-shaped process, and it appears again in the formal study of logic, where the warrant is recast as a rule of inference. This is exactly the move that generalises from a single persuasive paragraph to whole genres of academic and civic writing; putting these together, the central insight is that rhetoric is not ornament added after the fact but the engineering of assent under conditions of incomplete proof. The bridge is that persuasion can be analyzed, taught, and revised precisely because it has parts, and the foundational reason those parts transfer across fields is that every act of arguing — whether in a courtroom, a lab report, or a family dinner — must still connect a claim to an audience through some warrant that the audience will accept.

Exercises Intermediate+

Advanced results Master

Classical rhetoric was long treated as a closed inheritance from Aristotle, Cicero, and Quintilian, but the twentieth century rebuilt it twice — once as "new rhetoric" and once as modern composition theory — and the two reconstructions together define the field today.

Chaim Perelman and Lucie Olbrechts-Tyteca's Traité de l'argumentation (1958) ^{[Perelman 1958]} restored rhetoric as a theory of reasoning that is neither formal deduction nor raw persuasion. Their central device is the universal audience: the imagined body of all competent reasoners against which an argument's strength is tested. An argument that convinces only a particular audience (a sect, a party, a market segment) is weaker than one that would compel the universal audience, and the rhetorician's task is to shift claims from the merely particular toward the universally compelling. This move re-grounds the classical appeals: logos becomes the appeal that survives scrutiny by the universal audience, while ethos and pathos address particular audiences with their local commitments.

Kenneth Burke's A Rhetoric of Motives (1969) ^{[Burke 1969]} recast persuasion in terms of identification. Where Aristotle asks how a speaker moves an audience, Burke asks how speaker and audience come to share an identity — to be "consubstantial." On this account rhetoric works not chiefly by argument but by creating a felt overlap in interests, symbols, and enemies. Burke's framework explains why political and advertising rhetoric so often bypass explicit claims: identification does the work that argument would have to do more laboriously.

Stephen Toulmin's model, treated above as a tool, is also a philosophical argument against the positivist idea that the only genuine reasoning is formal proof. In The Uses of Argument (1958) Toulmin showed that the warrant/backing structure is field-dependent — the warrants that count in law differ from those that count in science or ethics — yet the form of the argument (claim, data, warrant, and so on) is field-invariant. This dual claim licensed the study of reasoning in domains (ethics, policy, history) that the positivist tradition had dismissed as irrational.

Modern composition theory breaks from the classical picture by relocating rhetoric's object from the finished oration to the writing process. Janet Emig (1971), Donald Murray, and Linda Flower and John Hayes (1981) modeled writing as recursive cognitive activity — planning, drafting, revising, re-planning — rather than linear execution of the five canons. The pedagogical consequence was decisive: instruction should target the process that produces prose, not merely judge the product, because most "bad writing" is unfinished process rather than absent skill.

Genre theory (Carolyn Berkenkotter and Thomas Huckin 1993; Amy Devitt 2004) reframed genre as a dynamic social action rather than a fixed form. Genres evolve as communities use them; writers learn genres by participating in the communities that need them, not by memorizing templates. This explains why a student who writes fluent lab reports can still flounder in a history seminar: genre knowledge is community membership, and transfer across communities is itself a learned rhetorical skill.

The social-epistemic movement (James Berlin 1987; Patricia Bizzell) pushed further, arguing that rhetoric is never ideologically neutral: every writing classroom teaches some version of reality, self, and society, and "current-traditional" pedagogy (grammar, form, the five-paragraph essay) smuggles in a particular politics under the guise of neutrality. This critique does not invalidate the classical apparatus but it does demand that the rhetorician name the ideological work the apparatus is doing.

Synthesis. Putting these together, the long arc of rhetoric runs from Aristotle's triad of appeals, through the canons and stasis as a workflow for inventing arguments, to Toulmin's field-invariant anatomy of claim and warrant, Perelman's universal audience, Burke's identification, and the composition-theory turn from product to recursive process. This is exactly the central insight that unifies them: persuasion has a stable form (claim, data, warrant; audience, purpose, context) that generalises across two and a half millennia and across every field from law to physics, while the content it carries — which warrants count, which emotions move, which genre fits — is dual to the local community that does the arguing. The foundational reason the apparatus still teaches is that it is the bridge between a claim a writer already holds and the audience whose assent the writer needs, and the bridge is that every discipline is, at bottom, an argument community whose membership a writer earns by learning its warrants.

Connections Master

This unit depends on sentence-level craft — every argument is finally made of sentences — and so builds directly on the grammar sequence, especially the analysis of clause structure that lets a writer assemble compound and complex arguments without losing the reader 22.01.16. It also presupposes the prose-clarity principles of the writing chapter, because a muddy sentence defeats even a sound Toulmin schema 22.02.01.

Rhetoric supplies the analytic vocabulary that the literature chapter uses to explain how figurative language does persuasive work. Tone and mood are the literary descendants of pathos; figurative language carries the warrants of a poem or story in compressed form 22.03.07, and satire and parody are genres whose whole effect is an argument by ridicule, analyzable with stasis and the appeals 22.03.11.

The bridge to philosophy is direct: rhetoric studies ordinary-language argument under uncertainty, while the philosophy of language asks how reference, meaning, and truth make such argument possible at all. The warrant in a Toulmin schema is the informal cousin of a rule of inference, and the study of when everyday argument counts as valid belongs equally to both fields 20.12.01.

Historical & philosophical context Master

Western rhetoric begins in the Greek law courts of the fifth century BCE, where the absence of professional advocates meant citizens had to plead their own cases, and a market arose for teachers — the Sophists — who sold instruction in speaking well. Plato's dialogue Gorgias attacks these teachers for making the weaker argument seem the stronger, and this attack fixed "sophistry" as a term of abuse; but in Phaedrus Plato conceded that rhetoric, rightly studied and aimed at truth, could be a genuine art. Aristotle's Rhetoric (c. 350 BCE) ^{[Aristotle 350 BCE]} is the constructive answer: a systematic treatise that treats persuasion as a neutral technique, analyzable like any other craft, and that introduced the three appeals and the genres of deliberative, forensic, and epideictic speech.

The Roman orator Cicero consolidated the Greek inheritance in works like De Inventione ^[Cicero] and De Oratore, codifying the five canons and the stasis system that became the backbone of later Latin pedagogy. Quintilian's Institutio Oratoria (c. 95 CE) then grounded the whole curriculum in the moral character of the speaker — the "good man speaking well" — welding ethos to an ethical theory of education that survived the collapse of Rome through the medieval monastic and cathedral schools.

Rhetoric's prestige collapsed in the early modern period. The scientific revolution and then Enlightenment philosophy favored demonstration and plain style over ornate oratory; by the nineteenth century, rhetoric in the classical sense had retreated to the margins of the curriculum, surviving mainly as "composition" — the practical teaching of writing. Peter Ramus's sixteenth-century reassignment of invention and arrangement to logic left rhetoric with only style and delivery, a reduction whose effects persisted for four centuries.

The field's modern revival has two sources. In philosophy, Toulmin (1958) and Perelman and Olbrechts-Tyteca (1958) independently argued that reasoning outside formal logic was not thereby irrational, and reconstructed a theory of substantive argument. In the academy, the post-war expansion of American higher education created a mass need for writing instruction, and scholars such as Edward P. J. Corbett (Classical Rhetoric for the Modern Student, 1965) re-imported the classical apparatus into the composition classroom. Kenneth Burke's A Rhetoric of Motives (1969) ^{[Burke 1969]} widened the lens from formal oratory to the whole symbolic action of social life, a move that made rhetoric indispensable to criticism, advertising studies, and political theory.

A recurring philosophical question runs through this history: is rhetoric a legitimate art of reasoning, or a technique of manipulation that works upon the irrational part of the audience? Plato framed the sharpest version of the charge; the Aristotelian answer — that persuasion operates in the unavoidable space where demonstration cannot reach, because we must decide before certainty arrives — remains the field's working defense. Non-Western traditions developed parallel answers: the Indian Nyaya school of logic and debate codified rules for honest disputation; the Arabic tradition of balagha treated eloquence and clarity as inseparable; and the Confucian classics tied persuasive speech to the moral cultivation of the speaker. These traditions confirm that the problems rhetoric addresses — how to decide and persuade under uncertainty, how to match speech to audience and occasion — are not parochially Greek but arise wherever collective life requires reasoned disagreement.

Bibliography Master

@book{aristotle_rhetoric,
  author    = {Aristotle},
  title     = {Rhetoric},
  year      = {c. 350 BCE},
  translator= {W. Rhys Roberts},
  note      = {Books I--III; the three pisteis, the species of rhetoric, the canons, and the virtues of style}
}

@book{cicero_de_inventione,
  author    = {Cicero, Marcus Tullius},
  title     = {De Inventione},
  year      = {c. 84 BCE},
  note      = {Book I: the five canons and the stasis system}
}

@book{quintilian_institutio,
  author    = {Quintilian, Marcus Fabius},
  title     = {Institutio Oratoria},
  year      = {c. 95 CE},
  note      = {The orator as ``the good man speaking well''}
}

@book{toulmin1958uses,
  author    = {Toulmin, Stephen E.},
  title     = {The Uses of Argument},
  publisher = {Cambridge University Press},
  year      = {1958},
  note      = {The claim--data--warrant model and field-dependence}
}

@book{perelman1969new,
  author    = {Perelman, Chaim and Olbrechts-Tyteca, Lucie},
  title     = {The New Rhetoric: A Treatise on Argumentation},
  publisher = {University of Notre Dame Press},
  year      = {1969},
  note      = {Translation of the 1958 Trait\'e de l'argumentation; universal vs particular audience}
}

@book{burke1969rhetoric,
  author    = {Burke, Kenneth},
  title     = {A Rhetoric of Motives},
  publisher = {University of California Press},
  year      = {1969},
  note      = {Identification and consubstantiality}
}

@book{corbett1965classical,
  author    = {Corbett, Edward P. J.},
  title     = {Classical Rhetoric for the Modern Student},
  publisher = {Oxford University Press},
  year      = {1965}
}

@book{lunsford2016everythings,
  author    = {Lunsford, Andrea A. and Ruszkiewicz, John J.},
  title     = {Everything's an Argument},
  publisher = {Bedford/St. Martin's},
  year      = {2016}
}

@article{bitzer1968rhetorical,
  author    = {Bitzer, Lloyd F.},
  title     = {The Rhetorical Situation},
  journal   = {Philosophy \& Rhetoric},
  volume    = {1},
  number    = {1},
  pages     = {1--14},
  year      = {1968}
}

@article{flowerhayes1981cognitive,
  author    = {Flower, Linda and Hayes, John R.},
  title     = {A Cognitive Process Theory of Writing},
  journal   = {College Composition and Communication},
  volume    = {32},
  number    = {4},
  pages     = {365--387},
  year      = {1981}
}

@book{berlin1987rhetoric,
  author    = {Berlin, James A.},
  title     = {Rhetoric and Reality: Writing Instruction in American Colleges, 1900--1985},
  publisher = {Southern Illinois University Press},
  year      = {1987},
  note      = {Social-epistemic rhetoric and the politics of the composition classroom}
}

@article{miller1984genre,
  author    = {Miller, Carolyn R.},
  title     = {Genre as Social Action},
  journal   = {Quarterly Journal of Speech},
  volume    = {70},
  number    = {2},
  pages     = {151--167},
  year      = {1984}
}