Cambridge Economics supervisor criticised for racist and sexist views
Photo - UCL
Exclusive: Students have raised concerns about articles written by Martin Sewell, supervisor in Economics and research associate in Land Economy.
Though Sewell clearly and frequently references the work of other academics, some of his statements have been considered as contentious, offensive or explicitly racist.
For example, when writing on the significance of race in conjunction with crime, Sewell states, without reference to academic sources:
"The most likely reason for the high incidence of black crime is blacks' lower intelligence and greater impulsivity, which themselves are probably biological in origin".
On eugenics, again without clear sources to his statement, Sewell affirms: 'Hitler gave eugenics a bad name', although he does back up this argument by explaining that 'The modern objectives are actually highly desirable: eugenics can help eliminate genetic diseases, reduce personality disorders and increase intelligence via human biotechnology. Time to reconsider.'
The content can be freely read by anyone, inside or outside the university, on his website. Sewell has written on varying topics, including gender, race, intelligence, climate change and politics.
Following complaints, the Cambridge University Students' Union (CUSU) spoke to The Cambridge Student: "Obviously, an individual who expresses such deeply racist views, such deeply sexist views and who explicitly endorses national socialism cannot remain as a supervisor for Cambridge students: Cambridge is a diverse, multicultural community which stands against - and, indeed, refutes -everything he stands for.
CUSU added: "However, this raises further worrying issues regarding how the University could employ such an individual - the University must give its community concrete assurances that its recruitment procedures will become sufficiently robust to prevent such an unacceptable individual from being employed in future."
The University's Press Office was unable to comment when contacted.
TCS phoned Sewell, who claimed that "It all boils down to political correctness". Asked via email whether his website content might contradict his educative role at the University, Sewell responded: "Publishing novel material that is largely the result of synthesizing peer-reviewed scientific research is inherently educative. That communicating certain realities about the world in which we live may be construed as 'offensive and controversial' is a result of the politically correct climate of the West that we currently live in."
He concluded: "Political correctness is anti-scientific (and unjust), so opposing it by failing to conform provides a good example of keeping ideology out of science."
Martin Sewell graduated from Bristol University with a BSc with Honours in Mathematics, and received his Masters in Computing Science from Birkbeck College, University of London. After spending 8 years at UCL studying for a PHD regarding financial time series analysis and intelligent systems, Sewell came to Cambridge in March 2009 as a Senior Research Associate at the Department of Land Economy.
From January this year Sewell has provided undergraduate supervisions to 27 students for the Economics Tripos, at Homerton, Newnham, Queens' and St Edmund's Colleges.
Nicholas Tufnell & Laurence Tidy - Deputy News Editors
Story RSS Feed
Entries (RSS)
Comments (RSS)
N. Dar, Queens' College
I have Martin as a supervisor and would just like to make clear I’ve never felt him to be a racist person…not even remotely!
June 25th, 2012
W.L. Hamilton, Sanger Centre - students
Martin Sewell lacks formal qualifications in genetics, psychology, anthropology or evolutionary biology and it shows. His website contains the pseudoscientific ramblings of a committed racist. 1. Only a minority of human genetic variation is distributed between ‘groups’ of people; the majority is found within populations. 2. Ancestral groups that cluster genetically do not conform to subjective impressions of ‘race’ and skin colour. 3. Genetic differences that have been found between human groups relate to skin and hair colour, immunity and lactose digestion, not cognition. 4. Genes have a complex interaction with the environment, particularly for a multicomponent trait such as ‘intelligence’. Any one gene will have a small effect. 5. What we call ‘intelligence’ breaks down into many cognitive and neurological systems. You can’t lump it all into a single number measured with one test.
Martin Sewell uses flawed science to justify his racist and sexist views. That student and tax payer money goes toward his sallary is a disgrace.
July 21st, 2012
Martin V. Sewell, Clare Hall
1. Only a minority of human genetic variation is distributed between ‘groups’ of people; the majority is found within populations.
This is more of an egalitarian political statement than a scientific argument; it is true but irrelevant. For example, the statement is also true of dogs (Parker, et al. 2004) and humans plus chimpanzees (Long and Kittles 2003). Genetic variability is an indicator of evolutionary time; it ignores any dependence between the frequency of the alleles across different loci, so is not a useful test of phenotypic variability. Invoking such a statement in order to falsely trivialise the genetic differences between the races is so widely discredited that it is known as Lewontin’s fallacy (Edwards 2003). A scientist who wished to answer the question ‘to what extent are different races genetically distinct?’ would perform some form of cluster analysis using a sufficient number of loci. Such analyses have identified the race of individuals with close to 100 per cent accuracy (Tang, et al. 2005).
2. Ancestral groups that cluster genetically do not conform to subjective impressions of ‘race’ and skin colour.
This statement is complete nonsense, and is contrary to the evidence. Rosenberg, et al. (2002) ‘found that predefined labels were highly informative about membership in genetic clusters, even for intermediate populations’, Bamshad, et al. (2004) found that their genetic analysis ‘allocated individuals into clusters that were entirely concordant with self-assessed ancestry’ and Tang, et al. (2005) applied genetic cluster analysis of microsatellite markers and produced four major clusters ‘which showed near-perfect correspondence with the four self-reported race/ethnicity categories’ and even described major self-identified race/ethnicity category and genetic cluster as ‘effectively synonymous’.
3. Genetic differences that have been found between human groups relate to skin and hair colour, immunity and lactose digestion, not cognition.
This is another political and misleading statement. It is not necessary to identify the relevant genes in order to show that differences are genetic in origin. If you wish to answer the scientific question ‘to what degree is the IQ difference between races heritable?’, then you should consider all of the available evidence. Rushton and Jensen (2005) examined ten categories of evidence–the worldwide distribution of test scores, g factor of mental ability, heritability, brain size and cognitive ability, transracial adoption, racial admixture, regression, related life-history traits, human origins research and hypothesized environmental variables–and concluded that the mean IQ difference between blacks and whites in the United States is about 80 per cent heritable. The mean IQ difference between Africans in Africa and Europeans is likely only 50 per cent heritable because of the environmental differences between the two populations (especially in terms of nutrition and health) (Lynn 2006, p. 193).
4. Genes have a complex interaction with the environment, particularly for a multicomponent trait such as ‘intelligence’. Any one gene will have a small effect.
This is another true but misleading political statement, inferring that the genetic component of intelligence is unknown or low. Neither is the case. The following figures are from McGue, et al. (1993). The estimated proportion of IQ variance associated with genetic factors in children is 40-60 per cent and in adults is approximately 80 per cent. The estimated proportion of IQ variance associated with shared environmental factors is relatively constant at approximately 30 per cent for ages up to 20 years but then drops to 0 per cent in adulthood. The nonshared environmental variance is relatively constant and close to 20 per cent. The upshot of this is that education and other social factors have zero effect on adult intelligence.
5. What we call ‘intelligence’ breaks down into many cognitive and neurological systems. You can’t lump it all into a single number measured with one test.
This statement is not merely political, but false, and demonstrates a complete ignorance of the literature. The g factor has been known and studied since its discovery by Spearman in 1904. In the words of the most accomplished and respected researcher in the area of intelligence, Arthur Jensen, ‘virtually all present-day researchers in psychometrics now accept as a well-established fact that individual differences in all complex mental tests are positively correlated and that a hierarchical factor model, consisting of a number of group factors dominated by g at the apex (or the highest level of generality), is the best representation of the correlational structure of mental abilities’ (Jensen 1998, p. 106). Furthermore, a reanalysis of more than 460 data sets from the factor-analytic literature spanning more than seventy years confirmed only one highly general factor, g (Carroll 1993), and ‘the predictive value of mental tests derives almost entirely from this global factor’ (Gottfredson 1998).
References:
BAMSHAD, Michael, et al., 2004. Deconstructing the relationship between genetics and race. Nature Reviews Genetics, 5(8), 598-609.
CARROLL, John B., 1993. Human Cognitive Abilities: A Survey of Factor-Analytic Studies. Cambridge: Cambridge University Press.
EDWARDS, A. W. F., 2003. Human genetic diversity: Lewontin’s fallacy. BioEssays, 25(8), 798-801.
GOTTFREDSON, Linda S., 1998. The general intelligence factor. Scientific American Presents, 9(4), 24-29.
JENSEN, Arthur R., 1998. The g Factor: The Science of Mental Ability. Human Evolution, Behavior, and Intelligence. Westport, CT: Praeger Publishers.
LONG, Jeffrey C., and Rick A. KITTLES, 2003. Human genetic diversity and the nonexistence of biological races. Human Biology, 75(4), 449-471.
LYNN, Richard, 2006. Race Differences in Intelligence: An Evolutionary Analysis. Augusta, GA: Washington Summit Publishers.
McGUE, Matt, et al., 1993. Behavioral genetics of cognitive ability: A life-span perspective. In: Robert PLOMIN and Gerald E. McCLEARN, eds. Nature, Nurture & Psychology. Washington, DC: American Psychological Association, pp. 59-76.
PARKER, Heidi G., et al., 2004. Genetic structure of the purebred domestic dog. Science, 304(5674), 1160-1164.
ROSENBERG, Noah A., et al., 2002. Genetic structure of human populations. Science, 298(5602), 2381-2385.
RUSHTON, J. Philippe, and Arthur R. JENSEN, 2005. Thirty years of research on race differences in cognitive ability. Psychology, Public Policy, and Law, 11(2), 235-294.
TANG, Hua, et al., 2005. Genetic structure, self-identified race/ethnicity, and confounding in case-control association studies. The American Journal of Human Genetics, 76(2), 268-275.
August 19th, 2012
W.L. Hamilton, Sanger Centre - students
I will respond to several of the most erroneous points made in Martin Swewll’s reply (August 19nd 2012).
“Only a minority of human genetic variation is distributed between ‘groups’ of people; the majority is found within populations.”
This is emphatically not an “egalitarian political statement,” and to suggest that it is lazily evades the responsibility of providing scientific data that refutes it. Early studies by Lewontin (1) looking at diversity in 15 proteins found that 85% of allele frequency diversity was found within local populations, and only 7% was associated with racial groups. More modern and thorough approaches examining diversity across hundreds of non-coding genetic loci, such as Barbujani et al. (2) have reached very similar conclusions, with 84.4% of variation distributed within local populations and approximately 10% attributable to differences between continents. This means that if every human being on Earth were killed excluding a single group or “race”, for example the people of Peru, approximately 85% of human genetic diversity would be preserved.
While this proportion of within- vs between- group diversity is broadly similar in magnitude to other animals, as you suggest, the structuring of that between-group diversity is far more nebulous in humans. In non-human animals, genetic differences between sub-groups or “races” often involve fixation of alternative alleles and quantitative traits differ by several standard deviations, with clear-cut geographic boundaries known as ‘hybrid zones’ (3,4). In humans population structuring is far less substantial (4). Humans evolved recently in Africa and the entire human population is thought to have gone through a tight bottleneck of around 2,000 breeding individuals in our evolutionary past (5), which explains the low degree of genetic variation and limited population structuring seen in our species.
I have not committed Lewontin’s fallacy, which was to conclude (erroneously) that because genetic variation between groups of humans is small, the concept of race is rendered “meaningless.” Edwards’ solution to the fallacy (6), which you site yourself, states that if enough loci are considered then groups of humans do form genetic clusters reflecting recent shared ancestry (though very few, if any, alleles can act as race discriminators on their own). In my very next sentence I use the phrase “ancestral groups that cluster genetically”, which is lifted directly from Edwards’ refutation of the Lewontin fallacy. Please be more careful to read what your critics actually write in the future.
“Ancestral groups that cluster genetically do not conform to subjective impressions of ‘race’ and skin colour.”
This statement is not nonsense. In 1779 Johann Friedrich Blumenbach conceived a classification system that divided humans into five groups – Caucasioids (‘white’), Mongoloids (‘yellow’), Malayan (‘brown’), Negroid (‘black’) and American (‘red’). To my astonishment I have seen these terms used on your website, despite their being obviously anachronistic and oversimplified. The term ‘Mongoloid’, for example, would group Papua New Guineans with South-East Asians, despite these groups separating in genetic cluster analysis (7).
People labelled as ‘black’ by certain human cultures possess diverse ancestral relations and may be differently labelled in other social contexts. Australian Aborigines are called ‘blacks’ by the modern descendants of European settlers living in Australia, for example, despite their having left Africa over 40,000 years ago and forming a distinct genetic cluster to current African people who are also called ‘black’.
In Kenya, the word ‘Mzungu’ is used to describe ‘white person’ and also more generally ‘foreigner’. I have seen Kenyans describe a Nigerian person (who would be called ‘black’ by our culture, though had lighter skin than the Kenyans) and a white European as both being ‘mzungus’, and thought the two were sisters.
Many human populations are genetic mosaics reflecting rounds of migration and colonisation, further confounding any simplistic racial classification. People of Latin America, for example, possess a hodgepodge of DNA derived from Europeans, native Central Americans and Native North Americans. Caribbeans, despite all being considered ‘black’ by most Europeans, in fact possess a mixture of African and European DNA, and two black Caribbean people can produce white-skinned offspring, reflecting their mixed genetic heritage. Even groups that intuitively seem to form well-defined races, such as the Masaai people in Eastern Africa, in fact show substantial genetic introgression when their genetic makeup is examined in detail (8).
The ambiguity and context-dependent subjectivity of labels such as ‘black’ render statements like “black people are x” – where x can be anything from “less intelligent” to “better at long-distance running” effectively meaningless. To be given even a shred of scientific credibility, specifically defined groups of humans that form genetic clusters reflecting shared ancestry with limited outbreeding must be established. As the above cases demonstrate, the label ‘black’ manifestly fails in this regard. I can only hope that its repeated use on your website, and the frequency of “black people are x” and “white people are y” statements, reflect genuine and sincere ignorance on your behalf rather than anything more malicious.
The most genetically diverse peoples on Earth live in Africa, and this reflects the fact that every non-African human alive today is descended from a relatively small group of Homo sapiens that left Africa around 100,000 years ago. Most non-African diversity (excluding a small contribution to European and Asia-Pacific genomes from interbreeding with Neandertals and Denosovans respectively) thus falls within the much larger diversity found in Africa (7). Hence, to the extent that cognitive faculties contributing to our multi-component notions of “intelligence” turn out to have genetic influences, the greatest variation in these influences would be expected to be found in Africa. A homogeneous genetic trait found in all diverse African tribes and sub-populations but not found outside of that continent, for the reasons outlined above, seems unlikely to the point of absurdity.
“What we call ‘intelligence’ breaks down into many cognitive and neurological systems. You can’t lump it all into a single number measured with one test.”
I stand by this statement, and do not consider the so-called ‘g-factor’ to be a legitimate scientific “discovery” but rather an imagined concept.
William James, the famous Welsh linguist who organised languages into a branching family tree reflecting evolutionary descent (providing the inspiration for Darwin), Albert Einstein, who discovered special and general relativity, and Johann Sebastian Bach were all geniuses. But clearly, different cognitive systems were being deployed by these people. William James could have been hopeless at mathematics and Einstein a dud with languages, and nether able to play baa baa black sheep on the piano.
We know the human mind is highly partitioned because of two centuries of neuroscience, showing that lesions to some areas of the brain (e.g. Broca’s area) can selectively destroy language faculties but preserve other aspects of cognition. The brain is an extraordinarily complex system that in fact comprises many quasi-independent sub-systems, acting massively in parallel and providing a convincing but specious illusion of a serial, unified “self” (8). Words like “intelligence” are so vague that you might as well call it “brain goodness”, but this ignores the fact that some aspects of cognition can be well developed while others are less so.
How could the multi-layered richness of human cognitive achievement, from general relativity to Hamlet to the ceiling of the Sistine Chapel, possibly be conveyed by a single number? It can’t. And while I am quite sure that an individual’s genetic makeup influences cognitive development, interacting with innumerable environmental factors, it would be most extraordinary if all the diverse capacities of the human intellect that fall under the umbrella of “intelligence” turned out to be neatly partitioned between semi-subjective racial groupings, such that one “race” was, across the board, more “intelligent” than another. The mental contortions required to reach such a conclusion are more than my mind can handle, given all the data described above.
1. Lewontin, R.C., The apportionment of human diversity. Evol Biol 1972,6:381-398.
2. Barbujani G, Magagni A, Minch E, Cavalli-Sforza LL: An apportionment of human DNA diversity. Proc Natl Acad Sci USA 1997, 94:4516-4519.
3. Harrison RG: Hybrid zones and the evolutionary process. Oxford: Oxford University Press; 1993.
4. Barton NH. Population genetics: a new apportionment of human diversity. Curr Biol. 1997 Dec 1;7(12):R757-8.
5. Hawks J, Hunley K, Lee SH, Wolpoff M. Population bottlenecks and Pleistocene human evolution. Mol Biol Evol. 2000 Jan;17(1):2-22.
6. Edwards AW. Human genetic diversity: Lewontin’s fallacy. Bioessays. 2003 Aug;25(8):798-801.
7. Mark Jobling, Edward Hollox, Toomas Kivisild, Chris Tyler-Smith, Matthew Hurles. Human Evolutionary Genetics: Origins, Peoples and Disease. Garland Science 2003.
8. Tishkoff SA, et al. The genetic structure and history of Africans and African Americans. Science. 2009 May 22;324(5930):1035-44. Epub 2009 Apr 30.
9. Daniel C. Dennett. Consciousness Explained. Penguin; New Ed edition (24 Jun 1993).
November 2nd, 2012
W.L. Hamilton, Sanger Centre - students
Correction to the above post:
*Martin Sewell
*William Jones, not James.
November 2nd, 2012
W.L. Hamilton, Sanger Centre - students
I remain firmly of the opinion that the views expressed on Martin Sewell’s website are misunderstandings of current scientific data on human evolutionary genetics, and I disagree with his conclusions. However, I would like to make it clear now that I overstepped the mark in implying that he is a “committed racist” and “sexist”, and I voluntarily withdraw these accusations.
January 14th, 2013