He was struck by the fact that so many of the diseases involved problems with processing sphingolipids, the fat molecules that transmit nerve signals.
He talked it over with Harpending, an expert in human population genetics. They came to believe this was an example of heterozygote advantage -- where having two copies of a mutated gene can mean disaster but one copy is helpful.
The most famous example of this is sickle cell anemia, which strikes people of African descent who have two defective copies of the hemoglobin B gene. As a result, they make red blood cells that are too curvy to carry oxygen to critical organs.
People who have only one bad copy make useful red blood cells that are deformed just enough to protect them from the malaria parasite, insulating them against the disease.
Instead of sickle cell anemia, Ashkenazi Jews had to contend with Tay-Sachs, Niemann-Pick and other diseases.
Instead of malaria resistance, Cochran and Harpending reasoned, Jews got an IQ boost.
The idea didn't come out of nowhere. Researchers have been drawn to the question of Jewish intelligence and genetic diseases at least since the 1920s, when some of the disorders were first being studied. Many physicians remarked on the unusual intelligence of their patients.
One of the first to conduct a systematic study was Dr. Roswell Eldridge, a neurogeneticist at the National Institutes of Health. He compared IQs of 14 children with torsion dystonia -- a neurological disorder afflicting Ashkenazi Jews that twists the body through uncontrollable muscle contractions -- against 10 of their healthy siblings and against unrelated Jewish students matched by age, sex and school.
The patients had an average IQ score of 121, compared with 111 for the control students, he found. Siblings had an average IQ of 119, compared with 112 for their matched controls. The results were published in 1970 in the medical journal Lancet.
Dr. Ari Zimran, director of Shaare Zedek Medical Center's Gaucher Clinic in Jerusalem, thought he would get similar results by studying the very bright patients he treated for Gaucher disease, another Ashkenazi genetic disorder in which excessive amounts of a fatty substance build up in certain organs, causing pain, fatigue and other symptoms.
His small study in the 1980s found no difference between IQs of patients and unaffected relatives. A larger study might have done so, Zimran said. But he decided not to pursue it.
"There is enough anti-Semitism," he said.
Cochran and Harpending are the first to make a broad case linking multiple Jewish genetic diseases to intelligence. Their theory draws on history, statistics, neurobiology and population genetics.
Jews first came to Europe in the 8th and 9th centuries, long before they were known for intellectual prowess, Cochran and Harpending say. They worked as traders before taking financial jobs made available by Christians who were forbidden by the Church from charging interest. By 1100, local registries listed most Ashkenazi Jews as lenders.
That set the stage for natural selection to do its work, Cochran and Harpending theorized. Jews didn't intermarry, keeping their gene pool closed. They were subjected to periodic persecution, which kept the population from outgrowing its professional niche.
According to the theory, the smartest individuals made the most money, and the wealthiest families had the most surviving children. The genes of the most intelligent Jews spread most, slowly raising the average IQ of the entire group.
Over 40 generations -- roughly 1,000 years -- an increase of just 0.3 points per generation would have added up to a cumulative advantage of 12 points, Cochran and Harpending theorized. Some of their other models projected a benefit of 16 to 20 IQ points.