Harry Harlow

Introduction

Harry Frederick Harlow (October 31, 1905 – December 6, 1981) was an American psychologist whose experiments on rhesus monkeys changed how developmental psychology understands love, attachment, and early social experience. Working almost entirely at the University of Wisconsin–Madison, he showed that infant monkeys separated from their mothers would cling to a soft, cloth-covered surrogate in preference to a wire surrogate that actually provided milk, and he concluded that what he called contact comfort — not feeding — is the foundation of the infant's bond. That finding dealt a severe blow to the drive-reduction or "cupboard love" accounts of attachment then dominant in behaviourist psychology. Harlow's later work, however, moved from surrogate mothers to prolonged social isolation and the deliberate induction of depressive states, and it is now among the most ethically condemned bodies of research in the history of psychology. He remains one of the most frequently cited psychologists of the twentieth century and one of the most contested.

Early Life and Education

Harlow was born Harry Frederick Israel in Fairfield, Iowa, the third of four brothers, to Mabel Rock and Alonzo Harlow Israel. Little is firmly documented about his childhood; in an unfinished autobiography he recalled a mother who was cold toward him and recurrent depressions throughout his life. His father tried a series of occupations, including farming, real estate, and a grocery business.

After a year at Reed College in Portland, Oregon, Harlow gained admission to Stanford University through a special aptitude test. He began as an English major and earned near-disastrous grades in his first semester before declaring psychology. He entered Stanford in 1924 and stayed on as a graduate student, working under the animal behaviourist Calvin Perry Stone and the vision researcher Walter Richard Miles, within a department shaped by Lewis Terman, the developer of the Stanford-Binet intelligence test. Terman became Harlow's principal mentor.

On completing his doctorate in 1930, Harlow changed his surname from Israel to Harlow. He did so at Terman's prompting, who feared that a name sounding Jewish would harm his career even though the family was not Jewish. The episode is often cited as a small but revealing marker of the antisemitism of American academic life in the period.

Academic and Professional Career

Harlow accepted a professorship at the University of Wisconsin–Madison immediately after finishing his doctorate and stayed there for the bulk of his career. He was refused adequate laboratory space by the psychology department, so he acquired a vacant building nearby and, with his graduate students, converted it into the Primate Laboratory — one of the first facilities of its kind in the world. Roughly forty students earned doctorates under him there. He established a breeding colony of rhesus macaques in 1932, in part because the nursery rearing his designs required gave him regular access to infants.

His earliest primate work was on learning rather than love. While working with the monkeys at the Henry Vilas Zoo in Madison, he developed the Wisconsin General Test Apparatus, which allowed standardized, repeated discrimination problems to be presented to many animals. Out of this came his discovery of learning sets, or "learning to learn": monkeys given hundreds of similar problems began solving new ones after a single trial. Harlow argued that this dissolved the apparent conflict between the Hull–Spence trial-and-error account and the Gestalt account of insight, since learning begins by trial and error and ends, as problems recur, in sudden solution.

Harlow held a series of influential posts outside the university: head of the Human Resources Research branch of the Department of the Army (1950–1952), head of the Division of Anthropology and Psychology of the National Research Council (1952–1955), and consultant to the Army Scientific Advisory Panel. He served as president of the American Psychological Association in 1958–1959, and later directed the university's Regional Primate Research Center. His honours included election to the National Academy of Sciences (1951), the Howard Crosby Warren Medal (1956), election to the American Academy of Arts and Sciences and the American Philosophical Society, the National Medal of Science (1967), and the American Psychological Foundation's Gold Medal (1973). After retiring from Wisconsin he held an honorary research professorship at the University of Arizona in Tucson, where he died on December 6, 1981.

Major Contributions to Psychology

Learning Sets and Primate Cognition

Before the attachment work, Harlow established that monkeys could acquire general strategies for problem-solving rather than only specific responses. This "learning to learn" finding, reported in a classic 1949 monograph, demonstrated transfer across problems and gave comparative psychology a method for measuring cognitive flexibility. It also implied that performance on any single test confounds the animal's current competence with its prior practice — a methodological caution that influenced later comparative work.

Contact Comfort and the Surrogate Mothers

Harlow's best-known experiments began in the late 1950s. Infant rhesus monkeys were separated from their mothers and housed with two inanimate surrogates: one built of bare wire and fitted with a feeding bottle, the other covered in soft terrycloth but providing no food. The infants spent the great majority of each day clinging to the cloth mother and went to the wire mother only to feed — roughly an hour a day. When frightened by a novel object, they ran to the cloth mother rather than the one that had nourished them, and with the cloth mother present they would venture out to explore and then return, behaviour that later attachment researchers would describe as the use of a secure base.

Harlow concluded that the mother-infant bond is not built on the reduction of hunger. He went on to vary the surrogates systematically, including a variable-temperature surrogate that let him separate warmth from softness, and surrogates that rocked. These later refinements mattered because the original comparison confounded feeding with several other properties at once; the follow-up work indicated that softness and warmth were specifically what the infants were seeking.

Maternal and Social Deprivation

Harlow then asked what happens when contact comfort is absent altogether. Monkeys reared in bare wire cages, and later in stainless-steel vertical isolation chambers, emerged grossly abnormal: they rocked, clutched themselves, bit their own limbs, stared blankly, and were unable to interact with other monkeys. Reintroduced to peers as adults they were either violently aggressive or terrified and withdrawn. Some isolates, later impregnated, proved unable to care for infants and in some cases mutilated or killed them. Harlow reported that isolation lasting beyond roughly the first months of life produced damage that could not be fully undone, and much of his later career was devoted to whether such damage could be reversed.

Models of Depression and Social Recovery

From the late 1960s Harlow deliberately set out to produce depressive states in monkeys and then to treat them, using isolation, separation from peers, and pharmacological and social interventions. This work yielded the finding that young, normally reared "therapist" monkeys could bring about substantial recovery in isolates — Harlow and Stephen Suomi's 1971 Proceedings of the National Academy of Sciences report on social recovery by isolation-reared monkeys remains one of his most cited later papers. Harlow's own severe depression, which led to hospitalization in 1968, was, by his own account, a major motive for this turn in his research.

Major Publications

  • "The Formation of Learning Sets" (1949)
  • "The Nature of Love" (American Psychologist, 1958)
  • "Affectional Responses in the Infant Monkey" (with Robert R. Zimmermann, Science, 1959)
  • "Social Deprivation in Monkeys" (with Margaret Kuenne Harlow, Scientific American, 1962)
  • "Total Social Isolation in Monkeys" (with R. O. Dodsworth and M. K. Harlow, Proceedings of the National Academy of Sciences, 1965)
  • "Social Recovery by Isolation-Reared Monkeys" (with Stephen J. Suomi, Proceedings of the National Academy of Sciences, 1971)
  • The Human Model: Primate Perspectives (1979)
  • From Learning to Love: The Selected Papers of H. F. Harlow (1986)

Impact and Legacy

Harlow's findings arrived at a moment when institutional care of children was unusually restrictive and when psychoanalytic and behaviourist accounts both treated the mother primarily as a source of feeding or of drive reduction. His demonstration that contact itself is a primary need fed directly into the emerging science of attachment: John Bowlby drew on it in the 1958 paper in which he set out the nature of the child's tie to the mother, and the broader climate of opinion it helped create contributed to reforms in hospitals and residential nurseries, including liberalized visiting and rooming-in. Later human research, particularly Mary Ainsworth's strange-situation work, established the specifics of human attachment on human evidence. Harlow's primate laboratory also trained a generation of researchers, and the partial reversibility of isolation damage — shown through peer "therapist" monkeys — became an early animal model for social interventions in developmental disorders.

The ethical verdict on much of this work is now overwhelmingly negative, and the criticisms are specific. The procedures included prolonged maternal separation, total social isolation lasting months and in some cases years, deliberately induced depression, and forced impregnation of isolates. Harlow and his colleagues named their apparatus in ways that made the cruelty explicit: the "pit of despair" for the inverted stainless-steel isolation chamber, the "iron maiden," and the "rape rack" for the restraint device used to impregnate females who had never experienced normal social contact. Animals died; many survivors never functioned socially again and could not be rehabilitated. Peter Singer's Animal Liberation (1975) cited Harlow's work as a clear case of laboratory cruelty whose scientific payoff could not justify the suffering, especially since less harmful human data pointed in the same direction. Gene Sackett, one of Harlow's own doctoral students, has said the experiments provided the impetus for the animal liberation movement in the United States, and another, William Mason, put it bluntly: Harlow "kept this going to the point where it was clear to many people that the work was really violating ordinary sensibilities, that anybody with respect for life or people would find this offensive."

A second line of criticism is scientific rather than moral. Critics note that much of what the surrogate studies showed had already been suggested by human observational and clinical data — Bowlby's clinical work, René Spitz's studies of institutionalized infants in the 1940s, and James Robertson's filmed records of hospitalized children — so the marginal knowledge produced by the most extreme procedures was small relative to the suffering they involved. Methodologically, the original wire-versus-cloth comparison differed on many dimensions at once, raising the possibility that infants were simply avoiding an aversive object. And cross-species generalization has limits: human infants attach to multiple caregivers, can recover substantially from early adversity with appropriate intervention, and develop in cultural settings with no laboratory analogue.

Harlow's defenders make three main points in reply. First, they argue that the early and less severe surrogate work — which produced the central finding — should be assessed separately from the later isolation studies, and that its scientific value is not seriously disputed. Second, they contend that the research produced practical knowledge that changed institutional practice for the better, and that it was conducted in an era before the Animal Welfare Act of 1966, before the 1985 amendments requiring attention to the psychological well-being of laboratory primates, and before institutional animal-care committees existed; the very framework now used to condemn Harlow was built partly in response to his work. Third, historians such as Lenny van Rosmalen and colleagues have argued from Harlow's newly disclosed hospital notes that the standard condemnations misread his intentions: he was driven by severe personal depression at a time when treatment options were poor, and he did, on his own evidence, obtain positive results that later informed models of depression and recovery.

Harlow did not publicly defend himself at length against the animal-welfare critique, and his joking vocabulary made his position worse rather than better. A further complication attaches to the reception of his ideas: the maternal-deprivation framing was later pressed into service in arguments against maternal employment and day care, a use that primate data of the kind he collected could not by itself settle. The current professional position is uncomfortable but stable. The findings are treated as real and historically important, the isolation and depression experiments are treated as unrepeatable, and Harlow's career is now taught as much as a case study in research ethics as as a contribution to attachment theory.

Learn More

References

  • Harlow, H. F. (1958). The nature of love. American Psychologist, 13(12), 673-685.

Last updated: 2026-10-20

Menu
Loading