"This is not a random sample of the population," said a mathematician, standing at a buffet table and looking across a busy room at the Cosmos Club.
Indeed, it was not. In one corner, statistician Charles Roberts of the Smithsonian was discussing the mating habits of Bactrian camels: "It's a little bit like one Cadillac raping another," he was telling Seymour Selig of the Office of Naval Research and French science attache' Michel Clerget. A few feet away, statistician Nozer Singpurwalla of George Washington University was earnestly asking biomathematician Max Woodbury of Duke University, "Why should cancer-related events be binomial?"
Don Gross, the author of a standard textbook on something called "queuing theory," was reflecting that he doesn't mind waiting in line the way other people do, because his branch of science deals with the factors involved in the formation of waiting lines: "I tell myself, if it wasn't for lines like this, I wouldn't have a job."
All the scientists in this nonrandom group had spent the day at George Washington University delivering and listening to papers on the work of French mathematician Simeon-Denis Poisson, who was born 200 years ago. "Actually, his 200th birthday was last June," said Singpurwalla, one of the organizers of the event, "but it took a while to get it all together; we had so many sponsors, and we had to bring in people from other parts of the country and Dr. Bru was brought in from Paris by the French Embassy. I proposed that we delay it so that we could do a better job." The co-sponsors of the event (the GW Graduate School of Arts and Sciences and School of Engineering, the Office of Naval Research, the French Embassy and the Washington Statistical Society) did manage to get it all together within Poisson's bicentennial year.
Singpurwalla, a music lover as well as a mathematician, is unhappy that geniuses like Poisson, who work in a highly specialized field, get less attention than geniuses like Beethoven. "People remember great artists," he said, "but except for a small circle, they forget great scientists."
The people at the Cosmos Club have not forgotten Poisson, whose mathematical work is more important today than it was during his lifetime in fields that he could never have imagined. His research is used by insurance companies to estimate what kind and number of accidents will happen in a given time period, by the Navy to predict corrosion in pipes, by medical researchers and engineers, supermarkets and transportation systems.
Airlines and nuclear regulators rely on Poisson's work to predict the reliability of performance by airplanes and reactors. His key contribution, known as the "Poisson distribution," is useful for predicting the occurrence of rare events and evaluating the element of chance in human events.
Herbert Solomon of Stanford University studied his statistical reasearch on the relation of the size of a jury to the correctness of its verdicts and found that it is still valid today, when juries are shrinking along with a shrinking economy. "There is a significant difference between a jury of six and a jury of 12, although the Supreme Court has said there isn't," Solomon said. "When you reduce the size of a jury, the errors increase. It wouldn't hurt the Supreme Court or its clerks to read what Poisson published in 1837."
Across the room, Roberts had finished with camels and turned to Madame Pompadour, whose maiden name was Poisson. "I did research on the Poisson families in the United States," he was saying, "and I found four branches, all of which claim descent from her. All of them are proud to have her in the family, but some of them will still argue with you if you say that she was the mistress of Louis XV."