Foucault, of course, knew and said that the literary ‘ship of fools’ of Brant’s poem and Bosch’s picture were allegories and not transcriptions from contemporary social reality. He did not say that there existed medieval social practices resembling a literal actualization of Brant’s or Bosch’s image: none of the evidence he presents is taken as documenting the existence of boats entirely occupied or manned by the insane, navigating seas or rivers of Europe at will or at random. What Foucault is saying, by saying that there were actual ships of fools in the Middle Ages, is that some mad people travelled, for various reasons, in boats along European rivers in the course of the Middle Ages. Some were being got rid of by towns which had expelled them, or sent back to their own; others (very probably, to judge from Foucault’s data, the most numerous category) were pilgrims journeying to shrines reputed to cure madness.
Midelfort thinks that Foucault wildly over-estimates the former categories of insane voyagers, since the three individual cases of riverborne transportations which Foucault cites apparently turn out to be the only ones ever known to have occurred. Foucault certainly seems to have thought there were more, and that his handful of known cases were samples of a more frequent practice, though (to reiterate a point) he does not say that it was a systematic or universal one, and in the case of Nuremberg he suggests it was relatively short-lived. On Foucault’s own showing, it is manifestly dubious exactly how ‘often’ Rhineland cities ‘must’ have found insane passengers disembarking on their quays. This ‘must’ is probably one of the kind which actually (rather as with Megill’s ‘sans doute’) conveys a less than apodictic certainty; there are many notes of tentativeness in the course of these few pages.
If Midelfort is right and (as every subsequent commentator appears to take for granted) his archival searches conclusive and exhaustive, then Foucault made a wong inference, and, incidentally but interestingly, the secondary nineteenth-century sources that Foucault used here would seem to have been more impressively thorough and exhaustive than Foucault himself possibly supposed.
Where Foucault is far less evidently off-target is on the matter of his other category of data, the pilgrimages of the insane, the ‘highly symbolic’ boats of mad persons en route to the indubitably real and frequented pilgrimage shrines of Larchant, Gournay, Besançon and Gheel. One may indeed concede to Midelfort that they did not go alone, or unsupervised, or only by river, or only in the Middle Ages. Still, they did go, they were real, and they seem to have been quite numerous. Midelfort’s 1980 paper does at one point—as LaCapra says—advert to their existence, but he unfortunately fails there to advert to any possible connection between their existence and Foucault’s text.
It may reasonably be said that Foucault, writing at the extreme initial margin of his theme and period, elaborates his material with a certain degree of poetic licence, at times blurring together the possible symbolic meanings of material practices and cultural motifs. Maher and Maher show how this led to enthusiastically over-the-top retellings of his story by anti-psychiatric readers, some of which Maher and Maher proceed to read back into Foucault himself (1982:751 ff.). Foucault’s actual ships of fools are (a rhetorician would say) a slightly hyperbolic usage of the latter term. But this does not mean, as Megill contends, that Foucault’s text is in some sense being radically ambiguous about the mad in the Middle Ages. Foucault was not saying something ambiguous, he was saying that something was ambiguous, or ‘liminal’: namely, the cultural and symbolic status of madness in late medieval society.
