Fallacy: Close Reading and The Beginning of Philosophy
by D. Vance Smith
The essay begins:
The formal study of fallacies began, as have so many other intellectual disciplines, with Aristotle. In a short treatise called Peri Sophistion Elenchon (On Sophistical Refutations, referred to in the Latin Middle Ages as De Sophisticis Elenchis), he classified thirteen types of fallacious argument—six that depend on slippages of language and seven that do not. As with much of Aristotle’s work, this emphatic division belied a dense cluster of problems that later expositors would tease out, and which this article will explore in a moment, beginning at the primal level of what, precisely, “fallacies” are. The change in terminology from “sophistical refutation” to “fallacies” belies an immense epistemic shift—nothing less, really, than the emergence of medieval philosophy out of classical thought. The Latin Middle Ages assimilated Aristotle’s treatise into the logical curriculum simply as “On Fallacies” (De Fallaciis), from fallare, “to deceive,” or “to trick.” In William of Sherwood’s influential twelfth-century Introduction to Logic, which expounds five of Aristotle’s logical texts, De Fallaciis is the sixth and final chapter. On the face of it, the word fallacia in medieval logic retains the implication of deliberate misleading, the problem that motivates Aristotle’s treatise, which is essentially a manual on how to spot the tricks an opponent might use to derail one’s argument. But deception is used one way in medieval philosophy, another in Aristotle’s.
Aristotle’s definition of sophistical refutations itself is deceptively complex, beginning with what seems to be the plainest declaration of what they are. The very first sentence of his treatise defines them as what they are not: they “appear to be refutations but are really fallacies.” From the start, sophistical refutations involve a dialectic of appearance: they are what they do not seem to be, and are not what they seem to be. Medieval discussions swerve around this problem by arguing that fallacies have both “semblance” or “appearance” (apparentia) and, seemingly paradoxically, nonexistence (non-existentia). What they mean by nonexistence needs to be qualified: a fallacy exists, of course, in the domain of language, and to that degree has the same ontological status as a valid syllogism. As William of Sherwood says in his foundational Introduction to Logic, fallacy and valid syllogism appear to be the same because they share a verbal identity: the resemblance comes “ex identitate sermonis.” But what leaves the fallacy in the realm of pure appearance only is that it doesn’t mean anything real: it diverges from valid syllogism “ex identitate rei.” But William (and subsequent commentaries) doesn’t say that what a fallacy refers to doesn’t exist; it is the fallacy itself that doesn’t have existence. To put this definition more precisely: a fallacy must meet the condition of appearing to be a syllogism and of “its nonexistence” (“non-existentiam eiusdem”). It exists only in apparentia, and its existence only there means that it does not actually exist.
For Aristotle, the phainomenon of fallacies is elusive, but so is the language we use to speak of them. They are “really” fallacies, paralogismoi, statements that lie outside of reason or of rational discourse (para-, outside, beyond + logizomai, to think, to calculate, to consider; the word is from logos, the Greek word that means anything from word to reason to order). That is, not only can we recognize sophistical refutations only by what they are not; we also cannot use the term itself to describe them, since they occupy a position outside of discourse. Sophistical refutations are actually not refutations at all (ouk elenchon legomen). Strictly speaking, a treatise on sophistical refutation is a logical impossibility, the first deception in a treatise full of them. Continue reading …
This essay by D. Vance Smith locates the beginning of philosophy in Aristotelian syllogistic logic, where fallacy is the precondition of rationality. Smith then turns to medieval commentaries, which treat fallacy as a nonreferential discourse and develop what is essentially a theorization of fictionality and its practices.
D. VANCE SMITH is Professor of English at Princeton University. His current work includes the completed Arts of Dying, a study of logic and death in medieval literature, and a book on medieval Africa, Blood Flowers.
In Woody Allen’s romantic comedy Annie Hall (1977), the world’s most famous technological determinist had a brief cameo that in some circles is as well-known as the movie itself. Woody Allen, waiting with Diane Keaton in a slow-moving movie ticket line, pulls Marshall McLuhan from the woodwork to rebuke the blowhard in front of them, who is pontificating to his female companion about McLuhan’s ideas. McLuhan, as it happened, was not an easy actor to work with: even when playing a parody of himself, a role he had been practicing full-time for years, he couldn’t remember his lines, and when he could remember them, he couldn’t deliver them. In the final take (after more than fifteen tries), McLuhan tells the mansplainer, “I heard what you were saying. You, you know nothing of my work. You mean my whole fallacy is wrong. How you ever got to teach a course in anything is totally amazing.” In the film, the ability to call down ex cathedra authorities at will to silence annoying know-it-alls is treated as the ultimate in wish fulfillment as Allen says to the camera, “Boy, if life were only like this!” Rather than a knockout punch, however, McLuhan tells the man off with something that sounds like a Zen koan, an obscure private joke, or a Groucho Marx non sequitur. There is more going on here than a simple triumph over someone else’s intellectual error. Isn’t a fallacy always self-evidently wrong?
JOHN DURHAM PETERS