The latest scientific scandal is a re-examination of the work of Hans Eysenck, a British psychologist who was very prominent in his lifetime. One recent study ranked him as the 13th most eminent psychologist of the 20th century (Haggbloom et al., 2002). A different study ranked him as the 46th most influential psychologist of all time (Diener, Oishi, & Park, 2014).
Eysenck was no stranger to controversy. He appears six times in Carl and Woodley of Menie’s (2019) database of controversies surrounding intelligence researchers. He was one of the first people to wade into the “Burt affair,” a (still not fully settled) dispute over whether Sir Cyril Burt committed scientific fraud. In 1971, he published the a book called Race, Intelligence and Education. (The very title is enough to make it controversial.) He also argued forcefully that smoking did not cause cancer (M. W. Eysenck, 2016).
Even though Eysenck died in 1997, he is the center of a new controversy. An article earlier this year cast doubt on his work on the integrity of his work regarding the influence of personality on physical health. Anthony J Pelosi (2019) chronicled Eysenck’s involvement with a physician and researcher named Ronald Grossarth-Maticek, who was conducting longitudinal studies on large samples of people in Germany and what is now Serbia.
Eysenck and Grossarth-Maticek’s research, published in the 1980s and 1990s, is literally unbelievable. A “cancer-prone” personality gave people a 121 times greater relative risk of dying of cancer. To show that this was a causal relationship, “cancer-prone” people randomly assigned to receive psychotherapy, and showed a 4-6 times lower relative risk than the “cancer-prone” placebo group. Results for heart disease are similar, and just as unbelievable.
Pelosi’s (2019) article chronicles a series of irregularities in Eysenck and Grossarth-Maticek’s research. It’s a dizzying series of contradictions, strange coincidences, and incredibly convenient data characteristics. To call this research “fishy” is a gross understatement. It is the scientific equivalent of a rotting carcass of a beached whale: the stench is overwhelming, and it urgently needs to be cleaned up.
Pelosi (2019) called for multiple investigations into Eysenck’s work on personality and disease, which he said is, “. . . one of the worst scandals in the history of science . . .” (p. 434). Kings College London obliged and identified 26 papers published in 11 journals that they deem “unsafe.” The committee doesn’t have the power to retract the articles (that is for each journal’s editor to do), but this type of labeling is often a precursor to later retraction.
But that may just be the beginning. Marks (2019) has identified 61 publications that he believes warrant retraction or correction. At Retraction Watch, Rod Buchanan, who wrote a biography of Eysenck, thinks the Kings College London investigation does not go far enough. Buchanan believes that 88 articles and chapters should be investigated and, probably retracted.
This is not some abstract academic debate. Grossarth-Maticek and Eysenck’s articles are still being cited, used in meta-analyses, and read approvingly. Fraudulent data showing that negative personality traits can cause cancer(!) can result in real harm if this influences treatment and policy.
I hope the journal editors do the right thing and retract these articles. This scientific grave wrong needs to be corrected.
Update: Grossarth-Maticek’s response is here, and (unsurprigingly) he denies any impropriety. The defense is vigorous but often strange. He ultimately fails to address many of the substantive concerns that Pelosi (2019) and Marks (2019) have about the data and the conclusions. It is mostly based on the silly claim is that Eysenck and Grossarth-Maticek’s research is groundbreaking because it is “multi-causal,” whereas Pelosi only considers “monocausal” models. The difference is that multi-causal research seems to refer to the possibility of statistical interactions, non-linear relationships, and other non-additive effects. It is absurd to think that a social scientist isn’t aware that causes may not be uniform in their effects. This is the sort of thing that people learn in their first year of graduate school (or even earlier than that). Grossarth-Maticek seems to claim that Pelosi doesn’t understand the research, which is clearly not true.
But the most bizarre portion of the response is when Grossarth-Maticek compares Pelosi’s and Marks’s calls for retraction to Nazi book burnings. Anyone who cannot see the difference between the two actions is being obtuse.
Lukas Jung and Hanna Steckelberg have an excellent article in Ruprecht, their student newspaper, about the controversy. The article includes a response from Grossarth-Maticek and is a superb example of science journalism. Also, according to Retraction Watch the editor of Personality and Individual Differences has decided not to retract the articles that Eysenck’s university said should be retracted. I’m sure it’s just a coincidence that the editor has coauthored with Eysenck and Eysenck’s son. [end of update]
Postscript: Eysenck’s influence
Eysenck’s impact on psychology is huge, but it may be waning. On Google Scholar, he hit his peak number of citations in 2013, and there has been a somewhat consistent decrease annually since then. Right now, he is on track for approximately 2,880 citations in 2019. If that projection is correct, it will be his lowest total since 2004.
In my fields of research, intelligence and educational psychology, Eysenck is already passé. His most widely cited work on intelligence, The Structure and Measurement of Intelligence (Eysenck & Fulker, 1979) garnered 19 citations last year (according to Google Scholar), which is impressive for the average scientist, but small potatoes for someone like Eysenck. In the field’s flagship journal, Intelligence, he was mentioned only 5 times last year and 3 times in 2017. In my upcoming book, In the Know: Debunking 35 Myths About Human Intelligence (available in late 2020 from Cambridge University Press), I don’t cite Eysenck at all in nearly 900 references.
Eysenck continues to exert much more influence in personality psychology, but within intelligence research, he barely makes a ripple any more. It seems strange that someone who was so influential and famous can be largely ignored in one of his areas of expertise. Maybe psychologists in other fields should follow suit.
Carl, N., & Woodley of Menie, M. A. (in press). A scientometric analysis of controversies in the field of intelligence research. Intelligence, 77. doi:10.1016/j.intell.2019.101397
Eysenck, M. W. (2016). Hans Eysenck: A research evaluation. Personality and Individual Differences, 103, 209-219. doi:10.1016/j.paid.2016.04.039
Haggbloom, S. J., Warnick, R., Warnick, J. E., Jones, V. K., Yarbrough, G. L., Russell, T. M., . . . Monte, E. (2002). The 100 most eminent psychologists of the 20th century. Review of General Psychology, 6, 139-152. doi:10.1037/1089-2618.104.22.168
Marks, D. F. (2019). The Hans Eysenck affair: Time to correct the scientific record. Journal of Health Psychology, 24, 409-420. doi:10.1177/1359105318820931
Diener, E., Oishi, S., & Park, J. (2014). An incomplete list of eminent psychologists of the modern era. Archives of Scientific Psychology, 2, 20-31. doi:10.1037/arc0000006
Pelosi, A. J. (2019). Personality and fatal diseases: Revisiting a scientific scandal. Journal of Health Psychology, 24, 421-439. doi:10.1177/1359105318822045