Lewontin in Santa Fe circa 2003, delivering the SFI Stanislaw Ulam Memorial Lectures

Richard Lewontin, 92, a revolutionary geneticist, evolutionary biologist, and longtime member of the SFI Science Board, passed away in his home in Massachusetts on July 4. 

Lewontin delivered a critical jolt of complexity to the field of population genetics. Through fieldwork, new laboratory techniques, computation, and innovative statistical methods, his work pushed genetics past the simplistic assumption that all phenotypic variations are driven by isolated mutations on single genes. 

In 1960, with cytologist Michael White he published data on a curious pattern of statistical associations between different inversions in the chromosomes of Australian grasshoppers. Together with Ken-Ichi Kojima, also in 1960, Lewontin developed a model that described how fitness interactions, now commonly called epistasis, between two genes could produce the kind of correlation between the frequencies of alleles at the two genes that he had seen in the grasshopper inversions. They called this association “linkage disequilibrium,” which is still widely used in mapping genomic phenomena. Lewontin and Kojima studied the dynamics of this complex, two-gene system, identifying its fixed points and their stability as a function of the recombination between the genes. The model, which related the fitness phenotype to the degree of genetic variation in an individual, gave rise to decades of research on how multi-locus fitness structure interacted with recombination to produce patterns of chromosomal variation in populations.

“That was the introduction of complexity into population genetics, which had previously been mostly based on simple one-gene theory,” says SFI Science Board member Marc Feldman, who was a Ph.D. student when he learned about the Lewontin-Kojima model, which formed the basis for part of his thesis and subsequent research. Feldman is now a leading population biologist based at Stanford University. He credits Lewontin’s early work for inspiring his own research into natural selection and gene recombination. The questions that Lewontin asked about multi-gene systems were also fundamental to the subsequent work by John Holland and his students on genetic algorithms, much of which was done at SFI.

Lewontin published two landmark papers with John Hubby in 1966, which introduced the laboratory technique called “protein gel electrophoresis” into the study of population-level genetic variation The researchers used the technique to investigate the genetic variation in fruit flies and were surprised and puzzled by just how much variation they uncovered.

“That paper revolutionized evolutionary genetics,” Feldman notes, "Because you now had to ask, ‘what was the reason for that high level of genetic variation?’” The question re-ignited a debate, which had begun with differences between Lewontin’s thesis advisor, Theodosius Dobzhansky,  and H. J. Muller and his followers about the level of genetic variation that should exist in natural populations; this is frequently called the problem of “genetic load “As the size and complexity of genetic datasets grows, Lewontin’s question about the fundamental causes of observed patterns of genetic and genomic variation becomes ever more prescient.

Lewontin was deeply interested in the causes of phenotypic variation in all species, but in the early 1970s, there was a furor around claims by Arthur Jensen and William Shockley that the average IQ difference between white Americans and African Americans was due to the latter’s genetic inferiority. In 1972, he showed that there is  a great deal more genetic variation within human populations than between humans from different continents — a finding which Feldman says “directly contradicted the racist ideologies of the time.” It also highlighted the complexity of the causal relationship between genes and phenotypic traits. In 1975, Lewontin and Feldman co-authored “The heritability hangup," in Science, which argued that it was impossible to understand the differences in phenotypes between human groups from genetic or statistical analysis of within-population phenomena

Throughout his scientific career, Lewontin was an outspoken critic of popular accounts of racial differences, which invoked biology to lend credibility to assertions that complex phenotypic traits, like IQ, were simply pre-determined by DNA. He believed strongly that what science did and how discoveries were evaluated and used depended on the structure of the social system in which the science was carried out; that is science and society are in the same messy complex system.

"Dick was a tireless advocate for equality and never stopped fighting pseudoscientific race theories,” says SFI President David Krakauer. "He was put into a fit of despair when selection was applied to culture.” 

Lewontin’s later research stressed the reciprocal influence of organisms and their environments, adding what Feldman calls an "order of magnitude” more complexity to the Darwinian account of evolution. He called the way organisms influenced the environments in which they lived and evolved “construction”, and this idea gave rise to the theory of “niche construction” developed later by John Odling Smee, Kevin Laland, and Feldman

In 1996, Lewontin joined SFI’s Science Board, and in 2003 he delivered the annual SFI Stanislaw Ulam Memorial Lectures in Santa Fe —  a series that still continues as part of the Institute’s community lecture series. Feldman, who first invited Lewontin to SFI, says “Dick" was a good fit for SFI culture because he was "very open to new ideas," and also embraced interdisciplinary approaches. In his Harvard lab, he welcomed social scientists and philosophers as valued participants in group discussions. 

Lewontin felt compelled to communicate the uncertainty and complexity of science to the general public and to dispel popular misconceptions around simplistic genetic "causes" for complex human experiences. "He fought a lonely battle," says SFI External Professor Steve Lansing.

But he persisted, and contributed frequent columns to The New York Review of Books, delivered public lectures, and wrote numerous books, including Biology as IdeologyNot in Our Genes, The Triple Helix, and his iconic textbook, The Genetic Basis for Evolutionary Change.

"If one’s message is that things are complicated, uncertain, and messy,” Lewontin wrote in Biology as Ideology, “there are rather fewer ways to get the message across.”

Melanie Mitchell, SFI’s Davis Professor of Complexity, who ran the Institute’s first adaptive computation program, recalls Lewontin saying “I don’t believe in adaptation, but I sure as hell believe in computation.”

Lewontin and his wife Mary Jane Lewontin are survived by their four sons, seven grandchildren, and one great-grandchild, according to his obituary in The Scientist.


Read Lewontin’s obituary in New Scientist (July 5, 2021)

Read the tribute by his former student in the blog, "Why Evolution is True" (July 5, 2021) 

Read the obituary in The New York Times (July 7, 2021) 

Read the obituary in The Scientist (July 8, 2021)

Read the obituary in The Washington Post (July 8, 2021) 

Read the obituary in Nature (July 13, 2021)

Read the obituary in The Guardian (July 28, 2021)

Read the obituary in WIRED (September 3, 2021)

Watch Lewontin’s 2003 SFI Stanislaw Ulam Memorial Lectures: