On mental and visual geometry.

Abstract

I THANK THOMAS JUNKER for his correction to the illustration of Blumenbach's skulls that accompanies the reprint of my essay, originally written for Discover Magazine (1994), in the revised version of The Mismeasure of Man (1996).1 Blumenbach's original depicts these skulls on a line, with the Caucasian example in the center. The version prepared by the Norton book designers converted this line into a wedge, with the Caucasian skull at the apex. The caption should have read modified ... Blumenbach rather than from ... Blumenbach, and the mode of modification should have been fully and explicitly specified. I did not prepare the figure, and I doubt that I ever saw it before publication, for I only obtained galleys of the text for proofreading. Nonetheless, the fault is entirely mine, as authors must assume full responsibility for any editorial changes or alterations. That is, authors should be proactive in overseeing every detail of a published book-a difficult task given standard procedures of publishing, where authors always work under time pressure and with partial versions missing various apparatuses of figures, footnotes, bibliographies, and so forth. But again, no excuses. I should have been more insistent. However, I am quite surprised by Junker's claim that my argument would have collapsed to a large extent if I had used Blumenbach's original figure-for a fundamental and conclusive reason that must be immediately apparent to any careful reader of my essay. My argument rests entirely upon Blumenbach's text. I never mention or cite his figure at all (except in a parenthetical remark, inserted by the editors to reference the added illustration). I don't think that I even knew about the figure when I wrote the article, for I worked a photocopy of Blumenbach's text alone. The version that accompanies my essay, drawn and inserted by the editors, does epitomize my argument in a useful way (especially for the nonprofessional readers targeted by Discover and, later, by my book), but it remains superfluous and additional to my intent and analysis. To summarize my argument: Blumenbach's original 1775 fourfold classification of human races follows his acknowledged master Linnaeus in recognizing four unranked divisions based on geography. His later 1795 version both adds a fifth (Malay) race and reorganizes the Linnaean geographic arrangement into a double hierarchy, moving outward in two directions a Caucasian ideal to least desirable Oriental and African endpoints via American Indian and Malay intermediaries-hence the importance of his Malay ad-