JEFFREY ROSEN'S ARTICLE is very well done, but, since it was written before the election results showed the power of the religious right, would he now revise his prognostication ("Supreme Mistake," November 8)? After all, the White House may now have a strong political incentive to nominate anti-Roe v. Wade judges--maybe not immediately, but, say, before the next congressional election, when a new fight over Roe would help solidify President Bush's extreme right, which no longer looks as far-out politically as before. Even to discredit it, the "moral values" issue needs deeper exploration by examining the internal contradictions in the concept.
Robert A. Senser
Reston, Virginia

JOHN B. JUDIS, RUY TEXIERA and Marisa Katz find profound significance in the supposed fact that Bush and John Kerry drew voters equally from those who had college degrees, while Bush supposedly trounced Kerry among voters who started but did not finish college ("30 Years' War," November 15). But the statistics published by The New York Times on November 7 show no such thing. According to the Times, there was almost no difference between the preferences of those who finished college and those who started but did not finish, and this held true in 2000 as well as in 2004. According to the Times, Kerry was favored by those who never finished high school and by those with graduate degrees, while all of the voters who were in between these educational extremes slightly favored Bush. Moreover, the Times data showed that the percentages voting for Bush increased in tandem with income, which is much more compatible with the Times' figures for voting as correlated with educational background than with the figures relied on by Judis, Teixeira, and Katz.
Richard Joffe
New York, New York

Richard Joffe misreads our article. We relied on the same data, from Edison Mitofsky Research, as the Times did. In the final exit poll tally, Bush won among all voters without college degrees by 53 to 47 percent, as we cited, and split college graduates. We didn't cite a figure for voters with some college but no degree, though Bush won these voters by 54 to 46 percent, up three percent from 2000. Bush's success among the noncollege-educated white voters was the prime reason for his victory in 2004.


Judge Richard A. Posner charges that Sherlock Holmes was not a scientific detective, lacked forensic competence, did not study the "criminal class" (do we still speak that way, judge?), and made little use of scientific instrumentation ("CSI: Baker Street," October 11). These assertions are all false. Counter-examples are manifold in the stories, though Posner tells us he has "only a faint recollection of stories read a half-century ago." When we first meet Holmes in "A Study in Scarlet," he is studying how bruises may still be induced on a body long after death. He has also just invented a new, and highly sensitive, test for the presence of blood, which will replace the guaiac method. We later learn that he has made a special study of cigar ash, of the telltale signs of a person's occupation on his hands, of handwriting, of footprints, et cetera. Not bad for 1887, but Posner actually thinks Sherlock ought to be studying DNA. And Holmes maintained extensive notebooks in which biographies of criminals as well as pillars of society were near at hand. Indeed, Watson tells us that Holmes was familiar with the details of a half- century of crime. Holmes employs the microscope as well as the magnifying glass, and his mastery of analytical chemistry is essential in several cases. Posner also asserts that the scientific method works by "confronting hypotheses with objective ... data that may falsify them." Holmes tells us that he never theorizes in advance of the facts because in that way one might twist facts to fit the theory rather than fitting the theory to the facts. As a practicing scientist, I find that I often work in the same way. It is the fluid interplay between observation and theory that makes for effective science. When Holmes reasons that the dog's nighttime silence in "The Hound of the Baskervilles" indicates that the criminal was known to the animal (i.e., it was an inside job), he illustrates the close kinship between observation and deduction. When he notes that only one of Sir Henry Baskerville's boots was stolen (an old one, not a new one), we understand that the hound of the Baskervilles is a flesh and blood creature, not a specter. And, finally, the judge fails to understand the enduring appeal of these stories. As Edgar Smith noted, "For it is not Sherlock Holmes who sits in Baker Street, comfortable, competent and self-assured; it is we ourselves who are there, full of a tremendous capacity for wisdom, complacent in the presence of our humble Watson, conscious of a warm well-being and a timeless, imperishable content. ... It is we who hear the pounding on the stairs and the knock upon the door. The swirling fog without and the acrid smoke within bite deep indeed, for we taste them even now." I hope that Posner brings to the bench a more open mind and a greater capacity for whimsy than he brings to literary criticism.
Robert A. Moss
Professor of Chemistry
Rutgers University
Investitured Member
The Baker Street Irregulars
Metuchen, New Jersey

De gustibus non est disputandum. But there is a mistake in Robert A. Moss's letter that is worth noting. Moss may not explicitly theorize whenever he examines data, but he is surely not a tabula rasa. What he notices in chemical data is shaped by his knowledge of chemistry. If he and I examined the same body of chemical data, we would see very different things because of very different theoretical presuppositions.