Obituary
In Appreciation: Lloyd G. Humphreys
Quintessential Scientist
I met Lloyd Humphreys during the first week of a three-year post-doc (quantitative methods training program, University of Illinois, 1987-1990). Soon we began meeting regularly, and I came to appreciate Lloyd’s extraordinary, brilliant mind. His wide-ranging command of applied research, psychometrics, and the study of individual differences left me in awe. When I shared this impression with my Minnesota graduate-school advisors, their response was simple: “Lloyd Humphreys is among the best differential psychologists and methodologists of the 20th century.” The late 1980s was a good time of life for Lloyd professionally. He was emeritus, had relatively few demands on his time, and I enjoyed hundreds of hours of one-on-one mentoring.
Over time, I learned about Lloyd’s early experimental work with Ernest Hilgard, his insights from taking a graduate seminar from Lewis Terman, his post-doctoral experience at Yale with Clark Hull, and his military work with J. P. Guilford. Lloyd discussed his many contributions: how statistical unidimensionality was not the same as psychological unidimensionality (a refinement of the multi-trait multi-method matrix, cognitive abilities being hierarchically organized), construct validity, systematic heterogeneity, the dimensions distinguishing achievement from ability tests, how the group contrast approach complements the prediction of individual difference in criterion performance, and why Inadequate Learning Syndrome is probably a bigger problem than the AIDS epidemic (Lubinski, 2003, in press; Linn, 1989). The more I listened to him and read his work, the more impressed I became with his intellectual depth, scientific integrity, and impatience with unctuous superficiality. I can’t imagine a better preparation for an academic position.
I can still vividly recall the apprehension I experienced when I handed him the first draft of what ultimately became our first joint publication. The next morning, I knocked on his door with trepidation, gingerly ventured in, and sat down to get his reactions. He promptly opened the manuscript (overflowing with penciled comments), and turned to a page that especially troubled him. Pointing to some underlined sentences, he inquired, “Why did you say this?” After re-reading my words, it was clear that this idea could have been communicated better. After readily admitting that, I took the opportunity to explain to Lloyd (with all the enthusiasm of a fresh PhD) that I thought it was important on first drafts to get the words down, although this haste sometimes results in saying stupid things. Lloyd calmly looked me in the eye and said, “I see. Well, I don’t like to say stupid things, under any circumstances.” Only those who knew Lloyd by acquaintance can begin to imagine the likely impact of these remarks.
On June 3, 2000, when the department of psychology at the University of Illinois dedicated their computational laboratory and quantitative reading room to Ledyard R Tucker and Lloyd G. Humphreys, respectively, Lloyd spoke at this event. He stressed that his work had always been aimed at application. The crowd gasped when he noted that “this has been true ever since my very first publication, 65 years ago!”
Lloyd placed a premium on science, rather than ideology, politics, or popularity (Page 1972). He was a generous and incredibly dedicated teacher, capable of seeing many perspectives. I will never forget his uncharacteristically beaming reaction to three reviews he received on his brilliant “Limited Vision in the Social Sciences” (Humphreys, 1991). One reviewer wrote that the manuscript ranked at least three standard deviations above the mean of the typical social science article and recommended prompt publication. A second reviewer suggesting that if Humphreys secured a coauthor and cut this piece down by about half, he might be able to develop something publishable. A last reviewer’s remarks were unlike anything I had seen before or since, and they tipped the decision in the positive direction. This evaluation was on a plain white sheet of paper, written and signed in a black felt tip pen. Because it only consisted of one sentence, it was easy to memorize: “I read this manuscript because I was ethically obligated to as a referee, but I hardly needed to as everything that Lloyd Humphreys does is great by my lights.” The signatory? None other than Paul E. Meehl.
Lloyd G. Humphreys was truly a professor’s professor, and the many students and colleagues he influenced will continue, in everlasting homage to this great man’s great work, to enjoy the privileges and pride that only come with being mentored by a world-class scientist.
References
Humphreys, L. G. (1991). Limited vision in the social sciences. American Journal of Psychology, 104, 333-353.
Linn, R. L. (1989). Intelligence: Measurement, theory, and public policy. University of Illinois Press.
Lubinski, D. (2003). Lloyd Girton Humphreys (1913-2003). Psychometrika, 68, 483-484.
Lubinski, D. (in press). Lloyd Girton Humphreys. Intelligence.
Page, E. (1972). Behavior and heredity. American Psychologist, 27, 600-661.
A Science-First Man, An Eternal Youth
Though it has been almost 30 years since I met Lloyd, I still remember him telling me (then a 22-year-old) that getting old was “hell,” or something like that. In the two years that I worked with him, I kept thinking that if Lloyd was this sharp in his 60s, what was he like in his youth? I have begun to realize that he never gave up the important aspects of his younger days, which were hard work, a strong belief in good data, psychometrics, and a resounding conviction in scientific psychology.
At his retirement symposium in 1985, Lloyd showed remarkable longevity as he continued his research long after he admitted to being “quite old.” In 1993 his research program was still strong, and he co-authored an article in the Journal of Applied Psychology. Lloyd’s commitment to his work must have been the key to his eternal youth.
By 2002 – nearly a decade later – I thought Lloyd must certainly be sipping tea on his ranch full-time. But that year he published two pieces in which he reminded the field about the nature of intellectual development. I actually believed I would retire before Lloyd!
The man was passionate about his data as well as his life. As a first year graduate student, I spent hours running data analyses while Lloyd sat with me, interpreting what he saw and waiting for me to share my insights. For my part, I stared blankly at some output from an early version of a structural equations modeling program. Our early research efforts were primarily oriented towards analyzing Project Talent data, which was already archived in the 1970s. He continued to analyze that data for another 25 years. Lloyd always felt the quality of the data, not the newness, was most valuable to research.
I remember a particular episode in which he asked me to help him compute some F-ratios from a table in a published article. I was confused as to why one would bother to re-analyze published results because, naively, I assumed that publication indicated “truth.” Lloyd concluded that our recalculated results were so strikingly different that the original authors must have faked their results. Lloyd tried, in the real spirit of truth, to publish his rejoinder. It wasn’t published, but it did leave a lasting impression in my mind about critically analyzing others’ results.
As for his life, he traveled extensively, but always seemed closer to his true self on his 60-acre ranch outside of Champaign, Illinois. He had a pickup truck, dogs, horses, a tractor, and a penchant for chopping wood, which he told me warmed you twice – once when you chopped it and again when you burned it. After an hour of chopping with him one afternoon, I was ready for the fireplace warmth.
Psychology will miss Lloyd Humphreys. He bridged the eras and factions of behaviorism, cognition, and the growth of professional psychology. He never lost touch with his youth, and he never lost sight of the science underlying psychology.
A Fearless Spirit
Lloyd was head of the psychology department at Illinois when I was hired. He was an outstanding department head for anyone in his first academic job. Lloyd provided resources and honest feedback, and kept the books balanced. I learned once that I had been denied an internal research board grant when I found a note from him under my door one morning. It said, almost in its entirety, “Sorry you did not get your research board grant. I have $300 in departmental funds you can use to hire a clerical worker if that will get you started on the research. You can do the pilot research and prepare a better proposal next time.” Nothing superfluous or Panglossian. Just an offer to provide resources and get the research started. You could buy a lot of clerical time for $300 then. I took the money and tried to gloss over the implied evaluation of my proposal. This was one of many similar incidents.
He provided unvarnished feedback on drafts of articles I, or any other junior faculty, post-doc, or graduate student asked him to read. He did not view research writings through the filter of wishful thinking that proud authors often use. He was outspoken in his devotion to the realities of what properly analyzed data – his and others – could and could not say. He did his best to protect those who asked for his comments on an article from stumbling about, blinded by rose-colored glasses. He seemed to see himself as an informal and undocumented collaborator when he was asked to comment on a draft of an article. He and Nobel Laureate Sir Francis Crick had similar beliefs when it came to feedback. Crick had noted in a BBC interview after receiving the Nobel Prize that: “Politeness is the poison of all good collaboration in science. The soul of collaboration is perfect candor. … A good scientist values criticism almost higher than friendship: in science criticism is the height and measure of friendship.”
Lloyd apparently viewed himself as a friend and collaborator, and did not want to poison our efforts with anything less than perfect candor.
In an era when too many get athlete’s mouth from jumping on and off political bandwagons, his personal integrity was impressive, though it did cause some problems. He thought individuals should be treated as individuals, not as members of a demographic group. He thought everybody should be judged on personal merit and afforded constitutionally guaranteed protections. This quaint view cost him his job at the Air Force Personnel Research Laboratory in San Antonio during the McCarthy era of the late 1950s. His final documented sin was to say at a social gathering that he thought the constitutional rights of communists should be protected. His view that individuals should be treated on the basis of their own merits did not change from then until his death. This got him branded as a right-wing racist 30 years later by those who failed to read his articles with the same care with which he wrote them.
In 1990, a professor in political science at Illinois received a research grant funded by the CIA to develop a program to model the global spread of AIDS. The CIA was interested in the modeling problem because AIDS was, and is, widely regarded as the most important destabilizing influence in the world. That program is now widely used in the United States and in other nations to project the spread of AIDS, and even to project population growth and shifts for policy planning. When word of the grant became public there was an uproar on campus – protests, pickets, confrontation with the professor – all of the trappings of a good campus brouhaha, except this one included death threats. The university administrators, with one exception, did nothing to support the professor. Lloyd, however, wrote one of his unique letters. In it he said, “I sympathize deeply with you in your interactions with the know-nothings who are attacking your research. I wish that I could be of help, but about all that I can offer is moral support.” There was a hand written asterisk attached to “know-nothings” with hand written a footnote: “You may quote me.”
In a second paragraph he added:
About 1969 or ’70, a new, powerful computer that had been designed by a member of our faculty was completed. Because the research had been supported by the Department of Defense, student protesters forced its removal from this campus to a location (military) on the Pacific Coast. It would have been a highly useful addition to our computer facilities and would have supported research projects far removed from military application. The next time you have an interview with a reporter, you might suggest that he or she look up that history.
The recipient of the letter carried it around with him for weeks and read it frequently. It was one of the few spots of light in the entire sorry episode.
Lloyd was an occasional contributor to the op-ed page of the local paper. His letters to the editor and the predictable responses they generated from readers livened an often dull page. The paper is less fun, and informative, to read now than when he was contributing. We are all poorer because of his death.
Humphreys a ‘Major Force’
I was one of Lloyd Humphreys’ last graduate students, having completed my PhD in the same year that Lloyd reached the age of mandatory retirement. However, I came to know Lloyd first through his publications rather than in person, mainly because he was acting dean of the college of liberal arts my first year of graduate school. This turned out to be quite a fortuitous circumstance, because in personal interactions, Lloyd always assumed that students were well read in the topics of individual differences in abilities and psychometric theory. For a graduate student to be unfamiliar with the history of the field was not an option. To remedy any shortcoming, the solution was to go to Lloyd’s office and take one of every article that he had published. Lloyd began publishing in 1935, and published widely and regularly in the decades that followed. The effort was very rewarding to the student, because the topics and writing style gave a very compelling sketch of Lloyd’s strong psychometric orientation and his fearlessness in confronting the sacred cows of psychology and public policy.
Lloyd was at his best when he thought that the science or psychometrics underlying someone’s research was dubious or weak. However, he was an equal-opportunity critic – he took on many luminaries, starting with William Sheldon’s somatypes in the 1950s, and moving on to Raymond Cattell, Arthur Jensen, and Robert J. Sternberg, among others in more recent years. He had no use for political correctness – one could always be sure that Lloyd wouldn’t pull any punches for the sake of diplomacy.
Lloyd also had an incredible memory, which would sometimes make it difficult for his coauthors. I remember writing a chapter with Lloyd in the late 1980s, and having several discussions about the extant research on abilities and group differences. As the director of the Air Force Personnel Laboratory during the 1950s, Lloyd was knowledgeable about many reports, based on large data sets, that had unfortunately never made it into peer-reviewed journals. Lloyd insisted that the data backed up his positions, and surely they did, but digging for citations just wasn’t something that he considered necessary to the enterprise.
Lloyd really never stopped working, even after retirement. An article he published in 2002 was classic Humphreys. He argued, as he had before, that problems arise with the peer-review system in psychology publications when the reviewers are not well trained in statistics, scientific method, and psychometric theory. Lloyd wasn’t ever much for small talk, but with his old manual typewriter, he was a major force for making psychology a quantitatively solid science.
APS regularly opens certain online articles for discussion on our website. Effective February 2021, you must be a logged-in APS member to post comments. By posting a comment, you agree to our Community Guidelines and the display of your profile information, including your name and affiliation. Any opinions, findings, conclusions, or recommendations present in article comments are those of the writers and do not necessarily reflect the views of APS or the article’s author. For more information, please see our Community Guidelines.
Please login with your APS account to comment.