The famed psychologist Leon J. Kamin passed away on December 22nd, 2017, at the age of 89.1 I did not know him personally, but in e-mail correspondences of the late 1990s and early 2000s he provided important feedback on analyses of genetic research that I was working on at the time, and I have been critically analyzing genetic research in the social and behavioral sciences ever since. Leon Kamin will be missed, and his writings were a major inspiration for me. His work will continue to inspire the next generation of critics of behavioral genetic and psychiatric genetic research, as well as the future debunkers of genetic (biological) determinism, and scientific racist theories of inherited racial (and social class) differences in intelligence and socially disapproved behavior.
In addition to his career as an academic psychologist, where a specific conditioning effect, the Kamin blocking effect, is named after him, Kamin wrote about issues such as the genetics of IQ, the genetics of criminal behavior, racial differences in IQ and reaction time, problems with twin studies of IQ and behavior, schizophrenia genetic research, racial head size and intelligence, claims of low African intelligence, problems with IQ genetic studies of adopted children (adoption studies), and “twin transfusion syndrome” and IQ.
Leon Kamin helped illuminate the false assumptions, methodological problems, biases, and even outright fraud in IQ genetic research. In the process he showed how faulty research had been used to harm and oppress people. He was a man of principle, who devoted much of the second half of his life to deciphering and re-analyzing the writings and original research publications produced by genetic researchers and their supporters, which included unpublished data and information when he was able to obtain it from the original researchers. For many years he was one of the main scholars that American journalists turned to when they sought an opposing perspective on the latest behavioral genetic “discovery” or theory. Kamin was frequently criticized by genetic researchers and their supporters, at times harshly and personally.2
In this article I will highlight and honor Kamin’s important work not only because of his scholarship, courage, and determination, but because his basic critique of IQ genetic and behavioral genetic research methods remains correct, and is especially relevant today in light of the ongoing failure to discover genes for “general intelligence” (“g”) at the molecular genetic level (see below).
As he described it in a 2005 letter to The New York Review of Books, Kamin was a victim of the McCarthy era witch-hunt of the 1950s on the basis of his membership in the U.S. Communist Party as a Harvard University undergraduate in the 1940s. In 1953, by then a teaching fellow at Harvard, he was subpoenaed by the U.S. Senate Subcommittee on Internal Security and was asked to provide the names of his former political associates. He declined to answer questions, citing his Fifth Amendment right to remain silent. He was subsequently subpoenaed by the McCarthy committee and answered questions about his own activities, but steadfastly refused to “name names.” Kamin was indicted for “contempt of the Senate,” but was acquitted on narrow technical grounds. Now unemployable in the United States, he moved to Canada and worked “happily” as an academic psychologist for fourteen years. The political climate in the U.S. changed greatly by the late 1960s, and in 1968 Princeton University invited him to become the Chair of its Psychology Department. In 1987 he returned to his home state of Massachusetts, becoming the Chair of the Northeastern University Psychology Department. After his retirement from Northeastern, he became an Honorary Professor at the University of Cape Town in post-Apartheid South Africa.
Here I will focus on Kamin’s pioneering critical analysis of “twins reared apart” (“TRA”) studies, also known as “separated twin studies,” including the more recent well-publicized and frequently cited Minnesota Study of Twins Reared Apart. This will be followed by brief descriptions of Kamin’s important contributions in most of the other areas mentioned above. In contrast to the common yet controversial studies of twins who grew up together, most people believe that TRA studies provide a definitive method of assessing the relative influences of genes and environments (nature and nurture) on human behavior and abilities. They are few in number and are very difficult to perform because twins are rarely separated in early life, and because it is difficult to identify, recruit, reunite, and study such pairs. In theory, because twins are genetically identical but grew up apart, TRA researchers were able to cleanly separate the potential influences of genes and environments on intelligence (IQ). As we will see, Kamin showed that the reality was something very different.
The handful of published TRA studies focused mainly on the behavioral resemblance or test score correlations of reared-apart MZ (monozygotic, identical) pairs, who are said to share 100% of their segregating genes.3 These are known as “MZA” pairs. Reared-apart DZ pairs (dizygotic or fraternal twins reared-apart) are known as “DZA” pairs, who like ordinary siblings are said to share an average 50% of their segregating genes. From the genetic perspective, unless twins grow up in very similar environments, a high MZA IQ score correlation indicates that genetic factors strongly influence intelligence, and that environmental factors play only a minor role. It should be noted that concepts such as “IQ,” “general intelligence,” and “heritability” are themselves controversial, and that the validity of the “heritability of IQ” concept is also controversial.4
The Science and Politics of I.Q.
While psychiatrist Don Jackson (1920-1968) set the standard in 1960 for casting serious doubt on genetic interpretations of comparisons between reared-together MZ and DZ twin pairs (the twin method), Leon Kamin set the standard in 1974 for casting serious doubt on genetic interpretations of studies of twins reared apart.5 In both cases, the basic assumptions and findings they called into question remain in dispute to this very day.
Kamin’s 1974 book The Science and Politics of I.Q. was intended mainly for an academic audience. It was a work of major scholarship, written by an academic psychologist who had not previously been involved in IQ genetic research, or had publically participated in the controversies surrounding it prior to 1972. He performed an amazingly in-depth analysis of IQ genetic studies ranging from reared-apart twins to reared-together twins, and from family members growing up together, to adopted children. He also examined claims that IQ scores are influenced by the prenatal environment. The search for “genes for intelligence” at the molecular genetic level did not begin until the 1990s, and in this era all claims in favor of genetic influences on intelligence were based on IQ test score correlations among twins, among adoptees and their biological and adoptive relatives, and on other comparisons among genetically related and non-related family members. Collectively, these are known as “kinship” or “quantitative genetic” studies.
Although Kamin was interested mainly in analyzing the original studies and documents, he devoted an additional chapter to showing how these original works were often misrepresented or misreported by the authors of secondary sources. The book’s title was itself provocative, since IQ testers and IQ genetics researchers claimed that their work was based only on science, and that it was separate from politics. In Kamin’s view, “the interpretation of I.Q. data seems never to be free both of policy implications and ideological overtones.”6 To “pretend” that politics and intelligence testing can be separated, he wrote, “is either naïve or dissembling.” He believed that psychometric IQ tests “served as an instrument of oppression against the poor—dressed in the trappings of science, rather than politics.”7 Kamin himself held leftist political views, which, as Princeton psychologist-historian William H. Tucker observed, he “never hesitated to announce publically.”8
Jensen and Herrnstein
The modern controversy over the role of genetic influences on IQ, and about the causes of racial differences in IQ scores, had been initiated by University of California, Berkeley psychologist Arthur Jensen (1923-2012) in his controversial 1969 article “How Much Can We Boost IQ and Scholastic Achievement?” Jensen argued that intelligence (IQ) test scores are “highly heritable,” and TRA studies were the centerpiece of this claim. Jensen also attempted to revive largely discredited hereditarian racial differences in IQ theories, where he argued that because (in his view) IQ was highly heritable within the Caucasian population, it was a “not unreasonable hypothesis that genetic factors are strongly implicated in the average Negro-white intelligence difference” of about 15 IQ points.9
Kamin became interested in IQ genetic studies three years later, following the appearance of a controversial 1971 article about IQ by Harvard psychologist Richard Herrnstein (1930-1994), published in the magazine Atlantic Monthly.10 This was followed by Herrnstein’s 1973 book I.Q. in the Meritocracy, where he expanded on his views.11 Herrnstein, who did not conduct human genetic research and whose expertise was studying the learned response of pigeons, argued that IQ scores had a strong genetic component and were a major determinant of people’s success in life, that economic classes differed in their innate (genetic) levels of intelligence, and that society would see increasing stratification and the creation of high and low castes based on inherited intelligence.12 He placed emphasis on the four TRA studies as providing the most convincing evidence in support of the “high heritability” of IQ, relying on an analysis of these studies that Jensen had published in 1970.13 Student activists at Princeton approached Kamin and urged him to read Herrnstein’s Atlantic Monthly article. In a 1973 interview, Kamin recalled, “I got interested in the problem of intelligence testing about a year ago, largely as a result of reading Professor Herrnstein’s article in the Atlantic.”14
“Genetic research on intelligence,” as a leading behavioral geneticist recently put it, “was in the eye of the storm of the nature–nurture debate in the social sciences” in the 20th century.15 In The Science and Politics of I.Q., Kamin entered the “storm” ignited by the Jensen and Herrnstein publications, and conducted a major in-depth critical analysis of IQ genetic research. His stated purpose was to examine “a single major question: are scores on intelligence tests (I.Q.’s) heritable?”16 Kamin pored over the original books and articles describing the four existing TRA studies. Jensen and Herrnstein had based much of their argument on these studies. Kamin’s analysis helped discredit the likely fraudulent TRA study by the British psychologist Cyril Burt (see below).
Documenting an Ugly History
In the first two chapters of The Science and Politics of I.Q., Kamin documented how early 20th century American psychologists, eugenicists, and politicians united to condemn entire groups of people to the status of genetic inferiority on the basis of the new alleged science of IQ testing. Kamin showed how these groups promoted eugenic policies, which included forced sterilization, and immigration laws aimed at preventing supposedly “feebleminded inferior stocks” from Eastern and Southern Europe from entering the United States. These included European Jewish refugees in the 1930s, who were denied entry to the U.S. and who later perished in Nazi death camps. Kamin, the son of a Rabbi, saw a direct path from the crimes of the past to the work of some of his contemporaries, who although they “speak with voices more cultured than” found in the writings of the crude Nordic (this word was used at the time by American eugenicists) supremacist psychologists of the 1920s, were again using intelligence tests to promote social and political policies based on claims that some classes, nationalities, and races were genetically inferior to others. “These teachers,” Kamin wrote, “once again assert that their effort is only to remove racial and economic conflicts from politics, and to place them on a firm scientific basis.”17 Like many of their predecessors of the 1920s and 1930s, both Jensen and Herrnstein believed that IQ was largely determined at birth (genes plus the prenatal environment18), and both implied that eugenic interventions would at some point be necessary to prevent a decline in national intelligence.19
The Cyril Burt TRA Study Scandal
Cyril Burt’s 1966 study was unique because he had reported that the twins had been placed into a wide range of families in different socioeconomic categories. Burt’s investigation was put forward by Jensen, Herrnstein and others as the only TRA study that showed that there was no correlation between the socioeconomic status (SES) of the homes into which MZA pairs had been placed, even as their IQ correlation remained high (.771). This suggested that placements had been made more or less randomly, and that MZA pairs’ high IQ correlation could not be caused by their environments. As Burt himself put it, his “figures should dispose of one of the commonest explanations advanced by thoroughgoing [IQ] environmentalists—namely, that the high correlations for the separated twins is due to the way the foster-parents were chosen.”20
Kamin’s first public discussion of the irregularities and contradictions he had found in Burt’s TRA publications took place at a March, 1972 lecture, delivered at Princeton. He gave another public talk on the issue in March, 1973 at Harvard, with Herrnstein in the audience. Unpublished transcripts of Kamin’s analysis were in circulation during this period.21 The first published presentation of his findings appeared in South Today, which was based on a March, 1973 talk he had given in Atlanta.22 In The Science and Politics of I.Q., published in October, 1974, Kamin presented a more detailed analysis of the Burt study in a chapter entitled “Separated Identical Twins.” Burt, who was knighted in 1946 for his work as a leading psychologist, had died in 1971 at the age of 88.
Kamin raised doubts about Burt’s claim in his TRA publications that he had identified and had given IQ tests to 53 MZA pairs—an exceedingly difficult task due to the rarity of such pairs. In his final 1966 TRA publication, Burt had reported an MZA IQ correlation of .771 for “group test” IQ.23 Kamin showed that Burt’s 1966 .771 MZA IQ correlation was identical to his reported 1955 correlation based on only 21 pairs, and to his 1958 correlation based on “over 30” pairs. The chance of obtaining identical correlations out to three decimal places on the basis of these differing sample sizes was extraordinarily low, suggesting that Burt had manipulated or even made up his data. Kamin also showed that Burt’s descriptions of the IQ tests he allegedly administered were vague and inconsistent. Kamin concluded, “The numbers left behind by Professor Burt are simply not worthy of our current scientific attention.”24 William Tucker subsequently showed that the probability that Burt could have found 53 twin pairs with the specific characteristics of his sample was also extraordinarily low.25
Although the major problems Kamin discovered were apparently unnoticed by mainstream psychology and by the genetically oriented researchers who cited Burt’s study, there were attempts by genetic researchers to “scoop” Kamin and to take credit away from him for discovering the flaws and irregularities in Burt’s publications. Jensen published a pre-emptive critique of Burt’s study while Kamin’s book was still in press (he mentioned Kamin only once, in a footnote), which was fast-track published in the journal Behavior Genetics in March, 1974.26 Jensen concluded that Burt’s “correlations are useless for hypothesis testing.”27 An obvious motivation of those attempting to beat Kamin to the punch was to create the impression that genetic researchers and IQ hereditarians were able to self-regulate their own research and practices, when clearly they weren’t.
Two years after the appearance of The Science and Politics of I.Q., Oliver Gillie, a reporter writing in the London Sunday Times, published an article entitled “Crucial Data Was Faked by Eminent Psychologist,” where he raised doubt that Burt’s research collaborator co-authors had ever existed.28 British psychologist Leslie Hearnshaw, until then an admirer of Burt, published a biography of Burt in 1979 in which he convincingly argued that Burt had invented much of his twin data.29 Hearnshaw’s conclusions were widely accepted by the early 1980s. Although subsequent investigations have led some genetic researchers and commentators to argue that the fraud charges leveled against Burt were unfounded, all sides of the issue agree that Burt’s “data” cannot be used.30
Had it not been for Leon Kamin’s meticulous detective work, it is likely that Burt’s TRA study would continue to be cited in contemporary textbooks as providing solid evidence in favor of the high heritability of IQ. This raises questions about possible additional cases of undetected fraud, and about the general existence of biased and unethical “p-hacking” practices in psychology and other areas of behavioral research, described in detail by cognitive neuroscientist Chris Chambers in his 2017 book The Seven Deadly Sins of Psychology: A Manifesto for Reforming the Culture of Scientific Practice.31 These problems could be reduced by the establishment of research “preregistration,” where investigators would be required to submit their proposed methods, definitions, and analyses before they collect their data.
The Three Remaining TRA Studies
The (then) three remaining TRA studies were published in 1937, 1962, and 1965 by Horatio Newman and colleagues in the United States (19 MZA pairs),32 James Shields in England (44 MZA pairs),33 and Niels Juel-Nielsen in Denmark (12 MZA pairs) respectively.34 Each of these studies was presented in a carefully detailed book, which included case histories and test scores for most pairs. As twin researcher Irving Gottesman (1930-2016) observed in 1980, however, “Few readers have actually read the original versions of these monographs, and have had to content themselves with condensations or reports of conclusions found in secondary sources.”35 While Herrnstein apparently was “content” to rely on Jensen’s account, Kamin closely examined these original works. Below I will summarize the main problems that Kamin found in these publications, and in TRA research in general. These problems include:
- The late or incomplete separation of most MZA pairs
- That many pairs grew up nearby to each other, and had substantial contact during important periods of their lives
- The fact that twins, who in most cases were volunteers responding to media appeals, had to have known of each other’s existence to be able to participate in the studies
- That the researchers’ assessment of the degree of twins’ separation frequently depended on trusting the (potentially unreliable) accounts of the twins themselves, and twins sometimes had financial and personal motivations to participate in the study
- That MZA pairs experienced similar or “correlated” environments due to the non-random placement of the adopted twins, which meant that they often grew up in similar cultural and socioeconomic conditions, and they sometimes were raised in different branches of the same family
- That there were problems with the intelligence tests used, including a lack of test standardization in some cases. This created the problem of age- and sex-confounds, which will inflate MZA intelligence test score correlations for non-genetic reasons
- That in some cases, twin pairs were recruited to a study because of their behavioral similarities, while dissimilar pairs were rejected
- The possible existence of “unconscious experimenter bias,” which may have led researchers who tested both members of an MZA pair to score them more similarly
- That subsequent reviewers performed questionable statistical procedures, which included deriving standardized IQ scores from non-standardized raw IQ scores, and the upward adjustments of some correlations
People generally believe, or are misled to believe, that TRA studies used only twin pairs who had been separated at or near birth, who were reunited as strangers at the time of the study, and who may have lived their entire (pre-study) lives not even knowing that they had a twin. In fact, this is rarely the case. Only 3 of the 75 pairs in the three remaining original studies were separated during the first year of life, were reared with no knowledge that they had a twin, and were studied at the time of their first meeting.36 Kamin showed that the Newman and Shields studies recruited volunteer twins through the use of media appeals, and that the twins had to have been aware of each other’s existence to be able to respond. They may have responded to the appeal, or they may have discovered each other, because of their similarities.
Kamin showed that in many cases, twins had substantial pre-study contact and sometimes had a close relationship to each other. Newman and colleagues clearly stated that they excluded some dissimilar MZA pairs due to the possibility that they might be DZA pairs, who were not included in their study. They acknowledged, “It seems possible that our [MZA] group is more heavily weighted with extremely similar pairs than with identical twins of less striking similarity” (italics added), which they viewed as an unintended bias resulting from the methods they used to recruit separated twin pairs.37 For Shields, twins separated as late as age 9, or for only 5 years during childhood, counted as MZAs.38 He counted twins “living next door to each other, brought up by different aunts” as a “separated” MZA pair (perhaps the twins were “separated” by a picket fence?).39 There are many more examples of questionably separated pairs of this type in the Shields 1962 study.40 In Juel-Nielsen’s 1965 study of 12 MZA pairs, “Ingegerd and Monika” were reared together with their mother between the ages of 7 and 14, and other pairs had a close relationship and experienced years of mutual contact.41
How Should Above-Zero MZA IQ Correlations Be Interpreted?
Some critics of behavioral genetic research argue that family, twin, and adoption studies of behavior are confounded by environmental factors, and that genetic interpretations of their results are therefore invalid. A confound is an unforeseen or uncontrolled-for factor that threatens the validity of conclusions researchers draw from their studies. As a pair of genetically oriented researchers recently put it, “When you can’t do experiments, you have to be very careful about something called confounding. Confounding is a pernicious problem that can make one thing look like it’s causing something else when, in actuality, it’s not” (italics in original). Kamin argued that genetic interpretations of above-zero MZA IQ correlations were confounded by environmental factors and methodological biases.
Kamin noted that MZAs and other adoptees were not randomly placed into available adoptive homes, but instead were selectively placed into such homes. Selective placement refers to adoption agencies’ and others’ practice of placing adoptees into homes matching the socioeconomic and perceived genetic status of the birth (biological) parents. MZA pairs were not assigned at birth to different randomly selected adoptive homes spanning the entire socioeconomic spectrum, as would occur in a true scientific experiment, and researchers were unable to observe them as they were growing up. Genetic interpretations of above-zero MZA IQ correlations, therefore, are based on the assumption that MZA pairs did not grow up in correlated environments. Kamin showed, however, that non-random placements and correlated environments were the norm in TRA studies.
Kamin identified age- and sex-effects as additional environmental confounds in TRA studies. Age-effects occurred when IQ tests were not properly standardized for age, meaning that MZA correlations could be spuriously inflated due to “a defective age standardization of the I.Q. test.” He made a similar point about potential sex-effects on IQ scores.42 MZA pairs, who are of course always the same age and sex, also share a common prenatal environment and a striking physical resemblance (e.g., height, attractiveness). Because in most societies men and women are socialized from birth to behave and think in differing gender-specific ways, the members of an MZA pair will behave more similarly for this reason alone.
Along with age-effects, sex-effects, and selective placement, as Kamin described them, MZA pairs share many additional environmental similarities and are subject to non-genetic cohort effects. The cohort effect concept refers to similarities in age-matched people’s behavior, preferences, beliefs, physical condition, and other characteristics that are caused not by heredity, but by experiencing stages of life at the same time in the same historical period and cultural milieu. “If you bring together strangers who were born on the same day in the same country and ask them to find similarities between them,” one critic observed, “you may find a lot of seemingly astounding coincidences.”43
In addition to sharing a common prenatal environment, I have shown previously that even hypothetical MZA pairs separated at birth, who never met each other and who spent their entire lives never knowing that they had a twin, grow up experiencing some or all of the following 14 behavior- and IQ-molding cultural/environmental influences in common: national, regional, language, ethnic/racial, religious (defined in part as “a cause, principle, or system of beliefs held to with ardor and faith”), striking physical resemblance, postnatal healthcare and nutrition, adoptee status (with accompanying abandonment issues), economic class, selective placement status, developmental stage, teaching methods and technological advances, birth cohort (age), and gender cohort (sex). As psychologist Susan Farber noted in her 1981 book Identical Twins Reared Apart: A Reanalysis, MZA pairs are “not so much similar to each other as they are similar to people of their eras and SES.”44
In summing up his TRA study findings in The Science and Politics of I.Q., Kamin wrote, “To the degree that the case for a genetic influence on I.Q. scores rests on the celebrated studies of separated twins, we can justifiably conclude that there is no reason to reject the hypothesis that I.Q. is simply not heritable.”45 He reached a similar conclusion in relation to IQ genetic research as a whole. Kamin was not concluding that heredity had no influence on IQ test scores, but rather that IQ genetics researchers, who bore the burden of proof, had failed to provide scientifically valid evidence that it did.46
In a 1976 book review, Harvard evolutionary geneticist and Kamin’s future collaborator Richard Lewontin wrote that Kamin had discovered, in IQ genetic research, “a pattern of shoddiness, carelessness, miserable experimental design, misreporting, and misrepresentation amounting to a major scandal.”47 In fact, as should be shouted from the rooftops of all of the world’s leading universities and media outlets, the evidence clearly shows that most MZA pairs found in the Newman, Shields, and Juel-Nielsen TRA studies were, to varying degrees, only partially reared apart.48 As sociologist Howard Taylor helped show, the claim that these studies investigated truly separated identical twins is a myth.49
And yet, in the eyes of many people the results of the more recent Minnesota TRA study have rendered Kamin’s analysis obsolete. Just a very brief overview of this study will show that this most certainly is not the case.
The “Minnesota Study of Twins Reared Apart” (MISTRA)
For many observers, the problems with the earlier studies, and the rejection of the Burt data, are old news. After all, it is widely believed that the more recent “Minnesota Study of Twins Reared Apart” (MISTRA) was methodologically sound, and that it corrected for many of the potential TRA study problem areas that Kamin had described in The Science and Politics of I.Q. The MISTRA is seen as having helped validate the findings of the earlier studies, and of behavioral genetic research in general. However, critical analysis of this study, based on the work of Kamin and others, reveals some largely overlooked major problem areas leading to a conclusion similar to the one that Kamin arrived at in relation to the Newman et al., Shields, and Juel-Nielsen studies.
The MISTRA, headed by University of Minnesota psychologist Thomas J. Bouchard, Jr., was conducted between 1979 and 2000, continuing to add newly recruited pairs as the study progressed. Its most famous and most cited publication was a 1990 article summarizing the findings, appearing in the prestigious journal Science.50 The researchers reported MZA correlations on IQ tests in the .64-.78 range, and concluded, “General intelligence or IQ is strongly affected by genetic factors.”51 At the time of the 1990 Science article, the sample consisted of 56 MZA and 30 DZA pairs. The final 2000 MISTRA sample consisted of 81 MZA and 56 DZA pairs. Bouchard and colleagues designated DZA pairs as the MISTRA control group early on in the investigation.52 The study was largely financed (about 60%) by the Pioneer Fund, an organization that was established to support eugenic and racial differences research.53
Major MISTRA Problem Areas
Kamin and his co-author Arthur Goldberger (1930-2009) addressed some MISTRA problem areas in their 2002 article, “Twin Studies in Behavioral Research: A Skeptical View.” When Kamin contacted Bouchard and requested access to the MISTRA raw data, Bouchard denied him access even under conditions where twin pairs would be identified only by code numbers, and where information about age would be omitted to guarantee non-identification.54 As Bouchard told a journalist, he “wouldn’t let Leon Kamin anywhere near” the MISTRA raw research material, although he was available to answer a “legitimate question.”55 This unethical “data hoarding” strategy violated basic scientific principles, especially since TRA studies are nearly impossible to reproduce (replicate) due to the increasing rarity of separated twins. Bouchard denied Kamin and others the opportunity to inspect the raw data, and to possibly arrive at a different set of conclusions than arrived at by the genetically oriented MISTRA researchers.56 The study’s findings should be rejected, and psychology textbooks should be rewritten, for this reason alone.
Based only on the published results, Kamin and Goldberger described “a number of serious problems in the design, reporting, and analyses by the psychologists engaged in the MISTRA.”57 Among these problems, they mentioned the selective placement of twins in the adoption process, that the twin sample consisted of a disproportionate number of highly similar pairs, that the MISTRA “contact time” formula used to assess twins’ pre-study contact was a “crude” measure that did not properly “capture the extent of a pair’s influence on each other,” that the MISTRA MZA pairs may have exaggerated or even lied about their degree of separation and behavioral similarity, and that the researchers failed to publish their full-sample control group DZA IQ correlations. To the best of my knowledge, the MISTRA full-sample DZA IQ correlations remain unpublished to this day.
The MISTRA was subject to problem areas 1-5 in the previous list of biasing and confounding factors that Kamin had identified in the earlier studies. The researchers’ conclusions in favor of genetics were further confounded by the above-listed 15 environmental influences potentially shared by even perfectly separated MZA pairs, a status that the MISTRA MZA pairs, who were volunteers, did not come remotely close to achieving.
Inspired by Kamin’s work, in my 2015 book The Trouble with Twin Studies: A Reassessment of Twin Research in the Social and Behavioral Sciences, I discussed in detail many additional potentially invalidating problem areas found in the MISTRA. These problem areas include, but are not limited to (1) the evidence suggests that, like the earlier volunteer-based studies, most pairs were only partially reared apart58; (2) in Bouchard’s own words, “several” key assumptions upon which he based his conclusions “are likely not to hold for cognitive abilities”59; (3) the researchers provided very little published case history information, and even failed to produce a table with basic demographic, degree-of-separation, and test score information for each studied pair; (4) the researchers’ belief that the MZA correlation “directly estimates heritability” was based on the completely false assumption that only genetic factors can account for above-zero MZA correlations60; (5) the researchers failed to publish their full-sample DZA control group IQ correlations, even though they published full-sample DZA correlations for most other non-IQ behavioral characteristics (such as “personality”); (6) the 1990 MISTRA Science article contained no DZA correlations, even though the researchers believed that determining whether the MZA correlation is significantly higher than the corresponding DZA correlation is “an important first step” in demonstrating “whether or not there is genetic influence on the trait”61; (7) based on the near-complete MISTRA DZA IQ results that have been published, there does not appear to be a statistically significant “important first step” difference between the mean (average) MZA and DZA correlations for either the Wechsler or the Raven tests, which were the main IQ tests used in the study62; (8) if the mean MZA behavioral trait correlation is not higher than the corresponding DZA correlation at a statistically significant level, this suggests that non-genetic factors alone are responsible for raising both correlations above zero; (9) Bouchard and colleagues believed that environmentally caused MZA behavioral resemblance should be counted as a genetic effect, thereby circularly assuming and concluding that twins’ behavioral resemblance is caused by genetic factors, which was the very issue under dispute63; (10) MISTRA-supplied single-case anecdotal stories of allegedly behaviorally very similar MZA pairs, such as the “Jim Twins,” the “Fireman Twins,” and the “Nazi and Jew Pair,” provide no scientifically valid evidence in favor of genetic influences on behavior64; and (11) the researchers’ conclusions were heavily influenced by their pre-existing IQ hereditarian biases, which naturally favored genetic interpretations of the data.65
To a far greater degree than what Kamin found in relation to the earlier studies, the MISTRA samples, methods, and findings have been systematically misinterpreted, misrepresented, and misreported in psychology textbooks and other influential secondary sources.66 Psychology textbook authors often endorse the MISTRA researchers’ conclusions with little critical analysis, and frequently reproduce the highly misleading anecdotal stories and photographs of pairs selectively released by the MISTRA researchers. Adding to this the American media’s scandalously uncritical celebration of the MISTRA since the early 1980s, and the many popular books about behavioral genetics and twin research that have appeared during the past thirty years, we are witnessing the latest (Internet-era) version of the “twins reared apart” deception that Kamin so brilliantly exposed in the 1970s.
Where Are the Genes?
Based on the common claim that the MISTRA and other behavioral genetic studies have confirmed the “high heritability” of general intelligence (g), molecular genetic researchers, armed with generous funding and modern gene-finding technology, should be able to identify the genes that underlie it. Bouchard admitted in 2014, however, that although “the high heritability of g has made it a popular target in the search for genes that influence behavior,” the “results to date have been dismal in comparison with expectation.”67 We have seen many gene discovery claims for general intelligence since the 1990s, yet follow-up studies were unable to confirm the original findings.
In a 2018 article, psychologist Robert Plomin, the world’s leading behavioral genetic researcher, claimed that genes for intelligence had been identified the previous year. At the same time, he acknowledged that attempts up to 2017 had failed: “Similar to results for many other complex traits, early results for intelligence were disappointing for more than 100 candidate gene studies and for seven GWAS [genome-wide association studies]. From the 1990s until 2017, no replicable associations were found.”68 History has shown that it is very unlikely that these recent claims will be confirmed, and Plomin has a 40-year track record of making claims of behavioral gene discoveries that turned out to be false-positive findings that could not be replicated.69 Until proven otherwise, we should assume that current and future gene discovery claims are false-positive findings as well.
As described by psychologist Ken Richardson in his 2017 book Genes, Brains, and Human Potential: The Science and Ideology of Intelligence, the IQ “gene bubble” is “bursting.” The ongoing failure to produce confirmed discoveries of genes that cause differences in so-called general intelligence provides additional evidence that Kamin and his colleagues and supporters were right all along, and that the behavioral geneticists, IQ hereditarians, and twin researchers were, and are, wrong.
Other Important Writings by Leon Kamin
The Twin Method
In several publications, Kamin correctly criticized genetic interpretations of “classical twin method” comparisons between reared-together MZ and same-sex DZ pairs, which remains the primary way that twins are used in human genetic research. In The Science and Politics of I.Q., following previous critics Kamin wrote that the twin method’s key assumption, which states that reared-together MZ and DZ pairs grow up experiencing similar or equal environments (the “equal environment assumption,” or “EEA”), “is on its face absurd” because “there is much evidence to demonstrate that in fact the environmental differences to which MZs and DZs are exposed are not equal.”70 In a subsequent publication Kamin argued that, because MZ pairs experience more similar environments than DZ pairs, from either a genetic or environmental perspective “one expects the correlation of MZs to be higher [for IQ] than that of DZs.” This indicated that MZ-DZ twin method comparisons “fail to reveal any unambiguous evidence for the heritability of IQ.”71
IQ Adoption Study Critique
Kamin also discussed in detail several major problems with genetic interpretations of IQ adoption studies, which were the subject of a chapter in The Science and Politics of I.Q. In addition, he found problems in a 1975 IQ adoption study published by the genetically oriented psychologist Harry Munsinger. This study supposedly showed that there was no relationship between an adopted-away child’s IQ and the SES of his or her adoptive parents. In a 1977 article, Kamin showed that there were irregularities and “serious errors” in Munsinger’s assessment and reporting of parental SES, and concluded that the study should “be dismissed.”72
The “Twin Transfusion Syndrome” Controversy
Munsinger published another article in 1977, where he analyzed the existing TRA data and case descriptions, and concluded that most twin studies underestimated IQ heritability. He claimed that the IQ correlation for MZAs of similar birthweight, who presumably did not experience twin transfusion syndrome, was 95%. Munsinger concluded that “postnatal social and cultural environmental influences have no effect on the population variation in IQ,” supporting the Jensen/Herrnstein position that IQ was largely determined at birth.73 In a detailed 1978 analysis, Kamin showed that Munsinger’s methods of classifying twins, and of assigning IQ scores, were “arbitrary and biased,” and that Munsinger had made “gross and biased errors in transcribing numbers from the original [TRA study] reports.”74 Kamin concluded that there is “no demonstrable basis for Munsinger’s claims about transfusion syndrome and the heritability of IQ,” and he raised questions about the validity of the heritability concept itself.
The Intelligence Controversy
In the 1981 general audience book The Intelligence Controversy (published in the United Kingdom as Intelligence: The Battle for the Mind), Kamin squared off with hereditarian IQ psychologist Hans Eysenck (1916-1997) in a debate on the genetics of IQ question. In his sections Kamin criticized aspects of IQ testing, and reviewed the case against genetic interpretations of TRA studies, adoption studies, and studies based on the twin method.75 Eysenck, on the other hand, defended IQ testing and concluded that a genetic interpretation of racial differences in IQ scores is “more likely to be correct.”76 Kamin took this opportunity to review the evidence that Burt’s findings were not reliable, and updated the account to include Gillie’s article and Hearnshaw’s biography of Burt. Kamin again reviewed the early history of IQ testing in the context of racism, the eugenics movement and forced sterilization, and psychologists’ and eugenicists’ use of IQ testing to label various races and national groups as genetically inferior.
Not in Our Genes
Kamin teamed up with Lewontin and British neurobiologist Steven Rose in their 1984 book, Not in Our Genes: Biology, Ideology, and Human Nature.77 Kamin and colleagues presented a comprehensive critical analysis of genetic determinist ideas, and of the evidence put forward in support of genetic theories of human behavioral differences. They also weighed in on the then-contentious sociobiology debate.
Not in Our Genes remains an important statement against genetic determinist ideology and the political implications and policies that flow from it. It also contained an important critical analysis of schizophrenia family, twin, and adoption studies, written mainly by Kamin.78 The authors provided an updated account of the Burt scandal, adding that “the implausibility of Burt’s claims should have been noted at once by any reasonably alert and conscientious scientific reader.”79
Turning to the other original TRA studies, Lewontin, Rose, and Kamin outlined many biases discussed by Kamin in The Science and Politics of I.Q., and pointed to the dubious “separation” of the twin pairs in these studies. “The technical use of the word ‘separated’ by the scientists of IQ,” they wrote, “obviously differs from the usage of the same word by ordinary people.”80 The authors questioned many aspects of IQ testing itself, seeing it as a flawed attempt to “rank order the world” on the basis of the “grand illusion of psychometry.”81
Crime and Genetics
Genetic and biological research on criminal behavior has a long and troubling history, very much tied into racist and eugenic theories and practices, where the main causes of criminal behavior were located in peoples’ genes, bodies, and brains. Supporters of these theories paid little attention to the real causes, which are rooted in economic and social inequality, racism, poverty, and other aspects of people’s social and political environments. Criminal biology research was largely discredited in the post-World War II era, only to make a comeback in the 1970s that continues to this day.82
In this context, in a 1986 article published in Scientific American, Kamin reviewed a widely discussed 1985 book by Herrnstein and James Q. Wilson (1931-2012) entitled Crime and Human Nature: The Definitive Study of the Causes of Crime. These authors argued that biology and genetics are major causes of criminal behavior.83 Kamin believed that, like the earlier IQ publications, Wilson and Herrnstein’s book was a “form of political argument, not science.”84 As always, Kamin closely examined the original research cited by Wilson and Herrnstein, and showed that these studies often did not support the authors’ claims. Wilson and Herrnstein confused correlation with causation, Kamin wrote, and the twin and adoption studies they cited “do not and cannot separate genetic and environmental variables, and the data therefore cannot be unambiguously interpreted.” Kamin concluded that Crime and Human Nature, which appeared during the conservative U.S. Reagan administration, reflected “the political climate of the times [and] makes it easy to understand why social scientists now rush to locate the causes of social tensions in genes and deep-rooted biological substrata.”
Racial Differences in Reaction Time
In a 1987 article, Kamin and Sharon Grant-Henry examined claims by Jensen and others that Caucasians outperform African-Americans in “reaction time” tasks, which record the elapsed time between a person being presented with a stimulus, and that person initiating a motor response to that stimulus. They found such claims to be “false,” and placed them in the context of asking readers to examine their own possible “unconscious racism,” in an American culture where racism is “endemic.”85
Kamin’s Critique of The Bell Curve
In 1995, Kamin published a scathing analysis of Herrnstein and Charles Murray’s The Bell Curve: Intelligence and Class Structure in American Life, a book that included yet another failed “scientific” attempt to claim that Caucasians are genetically more intelligent than African-Americans.86 Kamin refuted Herrnstein and Murray’s basic arguments, identifying their “two disastrous failings” as the “pathetic caliber” of the data they cited at “critical points” in the book, and (like Wilson and Herrnstein) their failure “to distinguish between correlation and causation.”87 After showing that Herrnstein and Murray’s arguments were faulty, and that some of the research they cited was produced by disreputable scientific racist investigators, Kamin concluded that the main thrust of the book was the authors’ use of “science” as a cover for a call for an end to affirmative action, and the promotion of other “reactionary” political policies. The reception given to The Bell Curve was political, Kamin wrote, and “had nothing to do either with its scientific merit or the novelty of its message.”88
Studies Claiming Low African Intelligence
Kamin published a 2006 critical review of publications by psychologists promoting racial differences in intelligence theories, including Jensen and Pioneer Fund leaders Richard Lynn and J. Philippe Rushton (1943-2012). These authors placed the average sub-Saharan African IQ at around 70. Kamin reviewed the problematic original studies that these claims were based upon, and pointed to these “ideological” researchers’ “palpably absurd conclusion” that half of the “African population [is] mentally retarded.” Their conclusion, he wrote, “should raise ‘serious questions’ about the test validity in the minds of critics.”89
Leon Kamin’s Legacy
Leon Kamin’s major accomplishment was to help distinguish science from pseudoscience, especially when pseudoscientific research and theories were used to harm and oppress people and groups. He was a truth-seeker and a man of principle, who was also motivated by a desire to help prevent a very ugly history from repeating itself.
Although published decades ago, The Science and Politics of I.Q., Not in Our Genes, and Kamin’s other writings remain works of major importance and scholarship. They serve as a model of how to perform a critical analysis of genetic research in the social and behavioral sciences, and also for showing how massively flawed and biased research can be used to promote regressive political agendas falsely claiming to be based on “science”—right up to current attempts to reduce immigration to the United States and Europe based, in part, on many political leaders’ implicit position that people of color are genetically inferior to Caucasians.
Contemporary supporters of behavioral genetic and hereditarian psychometric theories argue that the accumulated scientific evidence in favor of strong genetic influences on IQ and behavior in general is now overwhelming, implying that Kamin’s critique is outdated. In contrast, many critics of behavioral genetics, myself included, have pointed to faulty concepts such as IQ testing, “heritability estimates,” “model fitting,” and “personality,” in addition to environmentally confounded and otherwise faulty research methods such as family, twin, and adoption studies of behavior, as forming the false foundations of behavioral genetic and IQ hereditarian theories. They also point to the evidence showing that IQ scores have risen “massively” over time as education and living standards have improved,90 and to numerous other real-world and historical examples that disprove IQ hereditarian claims.91 An important aspect of Leon Kamin’s work was to inspire subsequent generations of critics, albeit still few in number, who have shown that his main conclusions remain in full force today—especially in light of the ongoing decades-old failure to discover “genes for behavior” at the molecular genetic level.
Leon Kamin should be remembered and honored as a champion of the oppressed in his role as an intellectual nemesis of the purveyors of the false ideologies of scientific racism and genetic determinism. While the American Psychological Association at times attempts to honor and reward the promoters of these ideologies,92 it is the task of others to remember and honor people who, like Kamin, helped show that these positions are based on beliefs, worldviews, economic interests, and prejudices that have a lot to do with politics, but have little if anything to do with science.
Abbreviations: EEA = equal environment assumption of reared-together twin studies; DZ = dizygotic (fraternal) twins; DZA = dizygotic (fraternal) twins reared apart; IQ = intelligence quotient; MZ = monozygotic (identical) twins; MZA = monozygotic (identical) twins reared apart; MISTRA = Minnesota Study of Twins Reared Apart; SES = socioeconomic status; TRA = twins reared apart (study).
Kamin, L. J., (1968), “‘Attention-Like’ Processes in Classical Conditioning,” in M. Jones (Ed.), Miami Symposium on the Prediction of Behavior, 1967: Aversive Stimulation (pp. 9-31), Coral Gables, Florida: University of Miami Press.
Kamin, L. J., (1974), The Science and Politics of I.Q., Potomac, MD: Erlbaum.
Kamin, L. J., (1977), Comment on Munsinger’s Adoption Study, Behavior Genetics, 7, 403-406.
Kamin, L. J., (1978), Transfusion Syndrome and the Heritability of IQ, Annals of Human Genetics, 42, 161-171.
Kamin, in Eysenck, H. J., vs. Kamin, L. J., (1981), The Intelligence Controversy, New York: John Wiley & Sons (published in the U.K. as Intelligence: The Battle for the Mind).
Kamin, L. J., (1981), “Commentary,” in S. Scarr (Ed.), Race, Social Class, and Individual Differences in I. Q., (pp. 467-482), Hillsdale, NJ: Lawrence Erlbaum Associates.
Lewontin, R. C., Rose, S., & Kamin, L. J., (1984), Not in Our Genes: Biology, Ideology, and Human Nature, New York: Pantheon.
Kamin, L. J., (1985), Criminality and Adoption [Letter to the Editor], Science, 227, 983.
Kamin, L. J., (1986), Is Crime in the Genes? The Answer May Depend on Who Chooses the Evidence, Scientific American, 254 (2), 22-27.
Kamin, L. J., & Grant-Henry, S., (1987), Reaction Time, Race, and Racism, Intelligence, 11, 299-304.
Kamin, L. J., (1995), “Lies, Damn Lies, and Statistics,” in R. Jacoby & N. Glauberman (Eds.), The Bell Curve Debate: History, Documents, Opinions (pp. 81-105), New York: Times Books.
Kamin, L. J., & Omari, S., (1998), Race, Head Size, and Intelligence, South African Journal of Psychology, 28, 119-128.
Kamin, L. J., & Goldberger, A. S., (2002), Twin Studies in Behavioral Research: A Skeptical View, Theoretical Population Biology, 61, 83-95.
Kamin, L. J., (2006), African IQ and Mental Retardation, South African Journal of Psychology, 36, (1), 1-9