The most important name in intelligence research in the second half of the 20th century is educational psychologist Arthur Jensen (1923-2012). From the late 1960s until his death, Jensen was the world’s foremost intelligence researcher who singlehandedly made methodological and scientific breakthroughs that greatly advanced intelligence research.

I never got to meet Arthur Jensen, but recently two videos were posted on YouTube that allowed me to get a glimpse into Jensen’s personality and how he responded to critics. The videos also reveal much about the man, the quality of critiques he faced, and science communication.

Panel on Jensen’s (1969) Article

The first video is a recording of a panel sponsored by Jensen’s university (the University of California) held to in response to an article he had written earlier that year (Jensen, 1969).

Jensen catapulted to infamy in 1969 when he published a 120-page article in Harvard Educational Review entitled “How much can we boost IQ and scholastic achievement?” Jensen’s answer was that IQ was not very malleable in most people. He went further, suggesting that socioeconomic and race differences in educational performance and intelligence could be partially genetic in origin, making these average intelligence differences among groups resistant to elimination.

The Jensen (1969) article did not fit in with the zeitgeist of the times. Discussing the article in my new book, In the Know: Debunking 35 Myths About Human Intelligence, I stated, “It would be extremely difficult to write something that would be more unwelcomed in the political and social milieu of 1969 than Jensen’s article” (Warne, 2020, p. 25).

The panel consisted of two sociologists (Aaron V. Cicourel and Arthur L. Stinchcombe), two geneticists (Joshua Lederberg and William J. Libby, Jr.), and an educational psychologist (Lee Cronbach), and was chaired by geneticist Curt Stern. (In 1972, Jensen wrote about how the panel came to be.)

Members of the 1969 panel on Jensen’s article. From left to right: Lee Cronbach (psychologist), William J. Libby, Jr. (geneticist), Arthur L. Stinchcombe (sociologist), panel moderator Curt Stern (geneticist), Joshua Lederberg (geneticist), Aaron V. Cicourel (sociologist), and Arthur Jensen (psychologist).

This is not a group of lightweights. All of these men are prominent enough today to be topic of a Wikipedia page, and Lederberg had already won a Nobel Prize. Cronbach would later by president of the American Psychological Association. These were serious scientists, and you can watch them in the video below.

The most notable aspect of the panel is how embarrassing it is to the sociologists. Generally, as the panelists’ disciplines become more scientifically rigorous, their arguments become more sophisticated. Between the two sociologists, Stinchcombe and Cicourel speak for a total of about 31 minutes to argue against Jensen’s (1969) paper, but they do little to rebut it. Stinchcombe repeatedly commits one of the most basic errors in the social sciences by drawing conclusions about cause from correlational data. Cicourel talks very little about Jensen’s paper and instead rants for several minutes about how Jensen doesn’t understand language acquisition and criticizes a different paper of Jensen’s (1968).

Both sociologists make assertions that Jensen is wrong, but provide little data or supporting citations to back up their beliefs. For example, Cicourel just states, without referring to any empirical evidence, that linguistic differences between Black and White Americans are so vast that they could entirely explain average IQ gaps between groups. Stinchcombe does slightly better by discussing Flynn effect data showing an average IQ score increase from World War I to World War II (Tuddenham, 1948) and data about the potential of preschool to raise IQ. But his criticisms are much weaker when he merely asserts that Jensen’s race and socioeconomic data are unreliable and that it is “outrageous” for Jensen to believe that within-family environmental influences could have a small impact on IQ.

Among the panelists, Cronbach speaks the least, and he fares much better than the sociologists. As an educational psychologist (like Jensen), he is clearly familiar with much of the data that Jensen draws on. He also mentions the higher IQ in World War II recruits, which he attributes largely to improved education. Cronbach suggests that race differences might also be caused by educational differences and differences in environment stimulation. (This is a plausible idea that has since been disproven; see Wicherts et al., 2004.) Cronbach also emphasizes that IQ is not fixed across the lifespan and can change (which is correct; see Moffitt et al., 1993). Finally, Cronbach argues that Jensen had not demonstrated that between-race differences in average IQ were genetic in origin.

Diagram that Lee Cronbach showed during the 1969 panel to support his claim that intraindividual IQ fluctuations can happen over the course of childhood and adolescence.

Lederberg goes further than Cronbach, arguing that Jensen (1969) is not only wrong that genetics might influence race differences in average IQ but that the issue was “undecidable.” Lederberg stated, “I do not see any simple methodology by which we can approach that question.” He was wrong. Within a few years, studies using three different simple methodologies would examine the issue: admixture studies (Loehlin et al., 1973; Rothhammer & Llop, 1976; Scarr et al., 1977), comparing IQs of Black and White children raised in equivalent environments (Tizard, 1974), and transracial adoption (Scarr & Weinberg, 1976). At least one of these methodologies should have been foreseeable to a Nobel-winning geneticist at the time.

Lederberg also took quite a bit of his time to downplay Jensen’s (1969) work and conclusions, saying that studies of groups are not relevant for educating or making decisions individuals. It was a poor argument because scientific research does not have to be relevant to particular individuals to provide improve humanity’s understanding of the world. He stated that the only relevant research on race differences in average IQ would be studies that identify specific genes that influence intelligence across groups. This was a far-off possibility in 1969, but in the 2010s the technology was developed to make this a reality.

Lederberg also, unfortunately, falls into the same traps as the sociologists on the panel, making assertions without evidence (“The notion that the environmental scale can be described in linear terms is the most outrageous suggestion I have heard here.”) and mistaking correlation for causation (e.g., when interpreting the relationship between African Americans’ degree of alienation from the school and their achievement).

The panelist that performs best is geneticist William J. Libby, Jr. His conclusion: “I think it’s a very good paper.” Libby also says that the paper is important and that it was good for Jensen to publish it “because both science and society are ready for it.”

I think that it should be read. I think that it should be thought about. I think that it should be discussed. And I think that in some rather specific ways, it should be acted on.

William J. Libby, Jr.

However, Libby points out errors in the work and emphasizes how Jensen sometimes oversimplifies the ways that genetics and environment would contribute to phenotypes like IQ. Libby also disagrees with Jensen’s belief that gene-environment interactions can be small, though he admits that this topic requires more research. Of the five panelists, Libby gives the most detailed response and is the most serious in considering Jensen’s work.

In his responses to these five scholars, Jensen is a gentleman. He counters claims by citing specific studies and shows how facts that the others present (like Cronbach’s correct view that education can raise IQ) are consistent with the 1969 paper (e.g., by explaining that education did not explain much individual variation within a cohort). My favorite moments are (1) when Jensen politely calls Stinchcombe’s remarks “conjectures” and “supposition” and (2) his response to Lederberg’s demand that environments between racial groups must be equalized before research on the genetic differences can happen. Jensen says this is utopian and would result in the research never happening.

Arthur Jensen at the 1969 panel to discuss his infamous article.

Jensen also takes umbrage with Lederberg’s argument that there is no simple methodology for investigating race differences in IQ, saying:

The fact that there is no simple way that to research this problem, I think, should not deter us. There is no simple way of putting a man on the moon, and yet we’re going to do that in a month or so.

Arthur R. Jensen

1980 The Phil Donahue Show Appearance

Watching Jensen’s 1980 appearance on The Phil Donahue Show is a different experience. Jensen was on the show to discuss his then-new book Bias in Mental Testing (Jensen, 1980), a hefty tome investigating the possibility of test bias in intelligence, achievement, and academic tests. The link to the video is below.

In the 1969 panel, Stern serves as a neutral moderator, but in 1980 Phil Donahue, as host of his eponymous show, is a hostile questioner who controls the conversation. Donahue knows what makes good TV, and he often steered the conversation to the most provocative idea related to the book, race differences in average IQ, even though that topic is not central to Jensen’s book. Donahue also occasionally grandstands to score rhetorical points with his audience.

The episode is more of a conversation than the 1969 panel was, but Jensen handles this format well. Donahue is most interested in simple ideas, such as “innate” (Donahue’s word) differences in intelligence, whether Jensen’s research will give support to racists, the potential for tests to be a tool to limit individuals’ potentials, the effects of labels, and questions about test content. They’re the types of topics that an interested layman would be expected to ask. Donahue was well prepared, but nothing he asked would have caught Jensen by surprise.

Arthur Jensen on the Phil Donahue Show in 1980.

Unsurprisingly, this discussion is much less technical, and Jensen does his best to respond in an accessible way. He doesn’t cite specific studies, and he doesn’t use words that non-experts would have difficulty understanding (like “phenotype” or “heritability”).

Despite his efforts, I don’t think that Jensen’s appearance changed many minds about the use of intelligence tests in society. Because Jensen doesn’t discuss data or the details of the evidence, he does not look like a vastly more informed expert when he disagrees with Donahue. Instead, they appear as equals; Donahue’s control of the show, his dominant posture over the sitting Jensen, and his charisma are hard to combat. Jensen has little more than the status of “college professor” and “book author” to make himself look more informed than Donahue.

Bias in Mental Testing, Jensen’s (1980) magnum opus on test bias and a book that pretty much put to rest the question of whether intelligence tests are biased against minority groups. (They are not, assuming examinees were born within the United States and speak English as a native language.)

Jensen’s performance is also a drawback. He is calm and subdued. This works well in the 1969 panel, but on TV against the animated Donahue, his replies often lack the conviction of Donahue’s assertions. Jensen doesn’t seem to be good at the witty retort, and he says little that would be memorable to viewers.

Audience Questions

Beyond Arthur Jensen’s presence, both videos have another interesting feature: questions from the audience directed at Jensen. Even though the 1969 panel audience consisted of college professors and the 1980 Donahue audience were all non-experts, there was an interesting commonality between them. Both audiences include people who make fools out of themselves by claiming that they know more about a topic than a person who has just published a lengthy article in a prestigious journal (in 1969) or an entire scholarly book (in 1980) about a topic. These audience members have a lot of hubris to think that they have some profound insight that Jensen never thought of, despite the years he spent researching these topics.

In the 1969 panel, the audience members with this tendency were those in the softer social sciences. For example, anthropologist Paul Kay sheds light only on his own ignorance when he claims that Jensen identified genetic influence by controlling for environmental variables and then labeling the residual variance as being genetic in origin. In reality, the reverse happens: behavioral geneticists control for shared genetic influence among people and then label the residual variance as being environmentally caused. (Jack King, a geneticist in the audience, points out Kay’s error later.) A sociologist, Dr. Blon(?), asked whether the testing situation may be a particularly hostile interracial contact that causes greater anxiety and lower performance in African American children because they know that tests are used to oppress people and limit their economic opportunities. (I found this suggestion bizarre because it would require children to think like a sociologist with a PhD.) Jensen easily counters this by pointing out that there are some tests where the score gaps are very small, and an oppressive society should not produce varying score gaps on tests. (Jensen would develop this further in the 1980s in his work on Spearman’s hypothesis.)

The most embarrassing audience comment in the 1969 recording, though, came from John Hurst, an education professor. Hurst attacked Jensen by claiming that there is no bias-free research and that Jensen’s work was the product of his political and social views:

I think that his paper reflects a systematic bias . . . which is a function of his own particular social and political orientation. . . . Hence, on these grounds–not on scientific grounds–I feel it is a very poor paper indeed.

John Hurst

Hurst claimed that he had documented this bias in a 130-page paper that he had submitted to the Harvard Educational Review. It was, apparently, never published. I think I can understand why.

The tendency for some audience members to display their ignorance was also apparent on The Phil Donahue Show in 1980. This was most embarrassing for the two people who identified themselves as teachers (pictured below). Both teachers display a classic flaw in critical thinking: generalizing from one’s own personal experience to draw a conclusion.

Two teachers in the audience of The Phil Donahue Show who argued to Arthur Jensen that intelligence tests were biased.

The teacher on the left was concerned about IQ scores have too much importance and that the scores are “a stamp” on children. When Jensen stated that he favored achievement testing more than intelligence testing in schools, she replied,

Even that sometimes is not very valid. Very often the child is afraid of a test. If they know achievement tests are coming, they will freeze up on you completely. Other times, it’s 50% guesswork. It really is.

Teacher on The Phil Donahue Show

The teacher on the right stated,

The tests are definitely culturally biased, not only . . . against Blacks but against anybody who has not had certain cultural experiences. . . . I would also like to point out the fact that I am also a product of Black schools. I did not score well on standardized tests until I took it upon myself to take special remedial courses. I am at a predominantly Jewish school, and I am excelling above the Jewish students, and I am a pre-med student, and I am at the top of my class. And I totally disagree with you.

Teacher on The Phil Donahue Show

Oh, Lord, grant me the confidence of a teacher who thinks that she knows more about test bias than a psychologist who wrote a 786-page book about the topic.

On the other hand, some audience members at The Phil Donahue Show seemed to be earnestly asking about the possible impacts of nutrition, the stability of IQ, and more. But the format of the Q&A portion of a talk show was not very conducive to giving these people a deep understanding of the science behind Jensen’s answers. I don’t blame Jensen for going on the talk show; if a nationally syndicated talk show wanted to talk about my book for an entire episode, I would agree to be on the show, too.

Likewise, the 1969 panel’s audience members often had great insights, such as political scientist A. James Gregor’s insight that Lederberg’s theory of “alienation” as a cause of lower average African American IQ was not plausible because Native Americans score higher than African Americans, despite both groups having a similar level of alienation. This anticipates Jensen’s (1998) criticisms of Ogbu’s theory of involuntary minority status as a cause of lower test scores.

Conclusion

Beyond seeing Jensen in action, I think that the videos provide important lessons in science communication.

First, know your audience. Jensen tailored his responses to the audiences, simplifying his message for non-experts, and clarifying complex topics clearly for scholars.

Second, be respectful. No matter how ridiculous or incorrect a comment from someone was, Jensen replied politely. (This is a lesson that I needed.)

Third, be clear and exact to avoid misunderstandings. Jensen never hesitated to clarify ideas so that he and his listeners were understanding one another. Whether he was explaining why he disliked the term “innate” or emphasizing his goal of treating people as individuals, he regularly tried to eliminate ambiguity.

Finally, there are tradeoffs to different forums of science communication. The Phil Donohue Show episode was probably watched by millions of Americans, but at a cost of exactness. The 1969 panel was very scholarly, but its reach was more limited, and most Americans would find it too technical and much less entertaining. Neither forum was ideal, but both were better for Jensen than not talking about his research would have been.

Together, the two videos run nearly three and a half hours. There are a lot of details that I haven’t summarized, and there are other lessons that people can drawn from them. The videos are a valuable resource, and I am glad that they are now available to everyone.

References

Jensen, A. R. (1968). Social class and verbal learning. In M. Deutsch, I. Katz, & A. R. Jensen (Eds.), Social class, race, and psychological development (pp. 115-174). Holt, Rinehart, and Winston.

Jensen, A. R. (1969). How much can we boost IQ and scholastic achievement? Harvard Educational Review, 39(1), 1-123. https://doi.org/10.17763/haer.39.1.l3u15956627424k7

Jensen, A. R. (1980). Bias in mental testing. The Free Press.

Jensen, A. R. (1998). The g factor: The science of mental ability. Praeger.

Loehlin, J. C., Vandenberg, S. G., & Osborne, R. T. (1973). Blood group genes and Negro-White ability differences. Behavior Genetics, 3(3), 263-270. https://doi.org/10.1007/BF01067603

Moffitt, T. E., Caspi, A., Harkness, A. R., & Silva, P. A. (1993). The natural history of change in intellectual performance: Who changes? How much? Is it meaningful? Journal of Child Psychology & Psychiatry & Allied Disciplines, 34(3), 455-506. https://doi.org/10.1111/j.1469-7610.1993.tb01031.x

Rothhammer, F., & Llop, E. (1976). Amerindian descent and intellectual performance in Chilean university students. Human Biology, 48(3), 455-464. http://www.jstor.org/stable/41462897

Scarr, S., Pakstis, A. J., Katz, S. H., & Barker, W. B. (1977). Absence of a relationship between degree of white ancestry and intellectual skills within a black population. Human Genetics, 39(1), 69-86. https://doi.org/10.1007/BF00273154

Scarr, S., & Weinberg, R. A. (1976). IQ test performance of Black children adopted by White families. American Psychologist, 31(10), 726-739. https://doi.org/10.1037/0003-066x.31.10.726

Tizard, B. (1974). IQ and race. Nature, 247(5439), 316. https://doi.org/10.1038/247316a0

Tuddenham, R. D. (1948). Soldier intelligence in World Wars I and II. American Psychologist, 3(2), 54-56. https://doi.org/10.1037/h0054962

Warne, R. T. (2020). In the know: Debunking 35 myths about human intelligence. Cambridge University Press. https://doi.org/10.1017/9781108593298

Wicherts, J. M., Dolan, C. V., Hessen, D. J., Oosterveld, P., van Baal, G. C. M., Boomsma, D. I., & Span, M. M. (2004). Are intelligence tests measurement invariant over time? Investigating the nature of the Flynn effect. Intelligence, 32(5), 509-537. https://doi.org/10.1016/j.intell.2004.07.002

css.php