The case of the Foucault-footnote or, why it’s ok to invent a source if your name is Borges.
This little bibliographic adventure began in an entirely unexpected way and resolved itself with a sort of vague inconclusiveness that I think would have pleased, or frustrated, both of the authors in the sub-title. If it could bring a smile to either of those now deceased entities I am fairly sure they would not agree on the source of the humour.
*A note to readers: This has been cleaned up for content since it was posted and I regret that most entries that get over 400 words are likely to endure a second edit after they appear. Nothing significant has changed, but odds are, it reads better. At least, I’m happier with it and that’s all that matters.
We start in Jorge Arditi’s A Genealogy of Manners.1 Arditi is a sociologist and while his topic sounds nicely historical, you will find that about two pages in it’s clear this will not turn out the way you hoped. Sociology and anthropology share some questions with historians but our methodologies are a little different. Arditi is engaged with one of the more intractable of historical questions that relates to behaviour and social norms. Historians try to answer these questions in their own complicated way, usually involving a great mass of comparative texts or close reading of some influential work from a given time like Pico de la Mirandola, or Thomas Aquinas, or Erasmus. Sociologists like their period authors as much as the next academic but they also like contemporary theorists and they will toss around names like Max Weber and Norbert Elias. Some of them really like Michel Foucault because he can be applied to almost any sociological problem.2
So what does Borges have to do with this? Well Arditi quotes Foucault who quotes from Borges, all in aid of a long, and not entirely intuitive, discourse on the arbitrary nature of language. At some point I expect Arditi will introduce Saussur, but for the time being he seems to be waiting in the wings, ready to pull sign and signifier from his valise like a rabbit from a top-hat.3
Foucault was tickled by Borges’ “well-known passage” about a certain Chinese encyclopedia and its bizarre and incongruous system of codifying the Emperor’s menagerie. Being a keen fan of Borges (but a often hostile reader of Foucault) I was intrigued, but search all I could, I found no citation to the original story that Foucault invokes, at least not in Arditi. The Foucault quote (with Borges, nestled within), is reproduced in full, in Arditi’s notes but Foucault himself does not name the story when he mentions it. He calls it only “a passage in Borges” as if there was only the one Borges from which a passage may come.4
And so, to the next layer of the reference onion. Foucault’s The Order of Things, where he describes this story, does not name the Borges source specifically either. I was no more bibliographically informed that when I left Arditi.5 Thankfully, Google came to my rescue since a search on “Borges Chinese encyclopedia” will get you close enough to the goal to cover the ground unassisted.
In 1942 Borges wrote the short non-fiction essay “El idioma analitico de John Wilkins”, for the journal La Nación, published 8 Feb 1942. It appeared in an English translation of an earlier anthology, Other Inquisitions 1937-1952, in 1964, with the title “John Wilkins’ Analytical Language”.6 Wilkins sounds like the sort of obscure intellectual that Borges would often invent but he was real, even if his work strains credulity. Wilkins (1614-72) wrote An Essay towards a Real Character and a Philosophical Language (London: John Martin, 1668) and Borges regretted that he had to describe the text, and its system of philosophical language, from secondary sources. An admirable qualification on the author’s part but, as we will see, not quite as sincere as it sounds.7 You lucky people can see a cover page thanks to the internets, something Borges may never have seen, even if the National Library of Brazil did own a copy. By then, Borges was blind so it would have done him no good, other than add to the already vast irony that his life had become.
What Wilkins (and Lull, and a dozen others) was trying to do was to construct a language that unlike all the others used a vocabulary that contained intrinsic meaning.8 This was distinct from all other language which was, through corruption or whatever, only able to identify things with some arbitrary sign. What Wilkins and his kind wanted was some incorruptible language that would resolve all ambiguity because its meaning would be intrinsic to its shape and construction. For Borges this was a spectacular exercise in futility because “there is no classification of the universe that is not arbitrary and speculative. The reason is quite simple, we do not know what the universe is.”9 Foucault needed 422 pages to say essentially the same thing. And having read Borges, and Paulo Rossi, I think Foucault may have missed the point of Borges’ example.
And here is the twist. While Borges admits that he didn’t get to see Wilkins’ book, he also didn’t see this Chinese encyclopedia. He tells his readers that Dr. Franz Kuhn, a very real and prolific sinologist and translator, was the source of the story.10 Borges does suggest that the encyclopedia (or the encyclopedeist) was apocryphal but diligent searches by later readers (more interested in the accuracy of the story than Foucault) have failed to find any trace of it, or anything like it, even in the work of Kuhn.
Now, having just complained, in my own tedious way, that Arditi and Foucault were guilty of some vague academic crime, I am now faced with the bibliographic source for this which appears to have done exactly that, or something even more serious. This is where I think Borges would find his germ of humour. If this was a joke, that’s a long time to wait for the punchline. If Foucault ever suspected that the Borges made up his encyclopedia, he doesn’t let on. Somehow, I think he would have been more angry than amused. Actually, that’s just what I hope.
1 Jorge Arditi, A Genealogy of Manners: Transformations of Social Relations in France and England From the Fourteenth to the Eighteenth Century, (University of Chicago Press: Chicago, 1998). This is one of the lovely gifts from Z and the mystery book list.
2 Yeah, when I started on this entry I knew the notes were going to get ugly and I’m not even going to try and list the relevant works or B and D dates for these people. I think the best thing is to relax, and try and enjoy the academic ride and I will try not to drown you in citations. Naturally, I will note my quotations. Hell, a failure to diligently cite sources is the reason I’m writing this in the first place. It is going to get a little ‘meta’ down here.
3 I speak of Ferdinand de Saussur and the semiotics of structural linguistics. Something of which I know just enough about to know how little I actually understand. I could have used Roland Barthes for that metaphor but his name sounds like a joke anyway and I don’t want to push it. Semiotics for Beginners reminds me that I missed an easy joke about pipes, but to it goes.
4 Arditi, pp. 6-7, 230.
5 Michel Foucault, The Order of Things: An Archaeology of the Human Sciences, Routledge Classics ed. (Routledge: London, 2002), p. xvi.
6 And this is where these innocent little notes turn into hard graft. After the 1942 publication of the essay it was collected in the 1952 Otras inquisiciones 1937-1952, which was the source of the English version. Odds are it was that English edition that Foucault read but modern readers will have to look to a different edition, still in print, if they want to see it (other than online, which is fairly easy). I pulled most of this bibliographic detail from Jorge Luis Borges, The total library, trans. Eliot Weinberger, Esther Allen, and Suzanne Jill Levine, (Penguin: London, 2001). The essay appears at pp.229-32.
7 Ibid., 229.
8 All of this is explained much more clearly than I will ever want to in Paolo Rossi, Logic and the Art of Memory: The Quest for a Universal Language, tran. Stephen Clucas, (Continuum International: London, 2006). As an added bonus for the brave, Rossi is rather scathing in his treatment of Foucault and his take on semiotics. It’s a really satisfying to read someone explain, with great clarity, that Foucault was completely wrong about something.
9 Borges, 231.
10 Ibid. Why Borges picked Kuhn (1884-1961) as his source is anyone’s guess. There is a discussion of this on a linguist list-serve, preserved here.