The philosopher Michel de Montaigne had a very strange way of writing. After the publication of the first edition of his Essays (1580), he took a copy of his own published text, reread it, and added more bits of writing. These marginalia reflected his state of mind at the time of rereading. Sometimes he added new ideas, or introduced digressions into his argument; sometimes he inserted scraps of poetry, single words, or clarifications of points made earlier. Then he published a new edition, with his own marginalia incorporated into the main text. In this way Montaigne became the commentator on his own text, both the reader and the read.
The impulse to reread was accompanied by the impulse to amplify the text, to add more language. Only rarely did he take anything out. As he says in the essay "Of Vanity," "I add, but I do not correct" (I cite Donald Frame's well known translation). Yet this is not strictly true. There are a number of occasions in which he did delete text. And to these we should pay careful attention. For at those moments he allows us to see himself trying to position himself, getting things adjusted just right.
Thanks to the beautiful digital facsimile of Montaigne's own edition of the essays, made available on the website, "The Montaigne Project," hosted by the University of Chicago, we can focus in on those few moments when Montaigne actually took something out of his text, or corrected something other than a typographical error.
One such change comes at a key moment in what is these days Montaigne's most widely read essay. This is "Of Cannibals," his meditation on the European encounter with the New World. Near the center of his discussion of the cannibal rituals of the Tupi Indians in Brazil, Montaigne recounts how the prisoners who are about to be eaten defy their captors, demonstrating bravoura and arrogance by insulting them even as they are being led off to be killed. Montaigne prints part of what he calls a prisoner's "song," which the victim sings to his captors, defying them and recounting how he has eaten their own parents in earlier rituals. Montaigne expresses his admiration at the song, which, he says, "does not smack of barbarity," at all. And he comes to a conclusion: "Truly here are real savages by our standards; for either they must be so, or we must be." In other words, the savages are so extraordinary in their culture--so expressive of courage and heroism--that one cannot establish a hierarchy that puts our culture above theirs. They are not less "civilized" or less "virtuous" than we are, but are remarkable for their culture and its rituals. Thus if they look strange to us, we must look strange to them. So Montaigne concludes, "there is a marvelous distance between their character and ours" (I've altered Frame's translation slightly; he renders merveilleuse as "amazing").
It is worth looking at the manuscript of this last sentence. For Montaigne originally wrote something else. He originally wrote, "There is a marvelous distance between their constancy and ours"; "Il y a une merveilleuse distance entre leur constance et la nostre." Then he scratched out the French word "constance" and replaced it with the word "forme," writ large and circled (see the image, above). "Forme" is what Frame translates as "character."
Not constancy, then, but "forme." What are we to make of this correction? "Forme" is probably a word he got from reading Aristotle, who uses it to refer to the coherence of features that instill identity in an object. It is a word that Montaigne uses almost 200 times in the Essays, and it usually denotes some type of coherent entity, some identifying characteristic, such as when Montaigne asserts in the essay "Of Democritus and Heraclitus"(II.50) that his own "maistresse forme," his defining characteristic, is ignorance. In the essay on repentence (III.2) he famously asserts that even common and inglorious lives can offer material for moral philosophy since "Each man bears the entire form of the human condition" ("chaque homme porte la forme entiere de l'humaine condition"). Closer to "Of Cannibals," in the essay on custom, (I.23), he uses it a number of times to speak of differences between political systems, evoking what he calls the "forme of the polis" ("la forme de la police"). "Forme" is often juxtaposed and sometimes contrasted in Montaigne with another word, "condition," which seems to suggest the accidental factors that shape who we are (birth, nationality, fortune), whereas "forme" is more essential and defining.
In Montaigne's marginal correction to the essay on the cannibals, the shift in vocabulary has interesting implications. His description, in the first edition, of the cannibals' activity as a manifestation of "constancy," couches the description of cannibal culture in the language of classical moral philosophy. Constancy is understood to be a universal quality. Everyone, no matter what their origin, cultural values, "condition," or "forme," can be constant. Thus to describe the cannibal victim as "constant" makes him an admirable version of some general human type and, inevitably, a mere copy of such classical heroes as Cato and Regulus. All it tells us is that the cannibals are braver than we are, which we already knew. However, when Montaigne replaces "constancy" with the more powerful structural term "forme" he asserts the absolute difference between cannibal culture and European culture. Henceforth the cannibals are not "marvelous" because they demonstrate the same moral qualities as Europeans. They are "marvelous" because they are essentially different. They possess a different "form." With this correction of his text Montaigne stumbles on the terms that will later shape the academic discipline we know as anthropology. Never fear revision. You never know what you might discover.