I extend my thanks to S. Kuzma for library assistance; M. Mayer and J. Harris for secretarial assistance; F. Gaynor for computer advice; and L. Roberts and Dr. K. Shapiro for editorial advice. I also thank C. Quinn for typing early versions of the manuscript, and the late E. Seiling and A. St. Laurent of United Action for Animals for suggesting ways to monitor and summarize animal research. I am especially grateful to the American, National, and New England Anti-Vivisection Societies, who commissioned this report, for the opportunity to conduct this study.


This is the second in a series of critiques of the strategy of animal models in psychological research.* The purpose of the present monograph is to provide a comprehensive view of a particular area of study—the investigations of primate maternal deprivation begun in the 1950s by Harry Harlow and recently given major support through the appointment of Stephen Suomi, a former coworker of Harlow’s, as director of a new National Institute of Mental Health primate facility.

The presentation of an overview of a discrete research area signals an advance in the contemporary reevaluation of science’s use of nonhuman animals as subjects. Since Singer’s broad philosophic critique (1975), the dialogue of charges and countercharges has ranged in targets from the entire field to the single study. In terms of sophistication of analysis, most recently the dialogue has moved away from selective field-wide cataloguing of either benefits to humans or costs to animals to citation analysis, which in this context means the documentation of the frequencies with which animal-based research is cited in the literature. For example, on the side of research benefits, Miller (1985; Coile and Miller, 1984) has surveyed a limited set of American Psychological Association journals to buttress his claim that much research involving animals is noninjurious and that it results in concrete applications. There is clear and compelling evidence that discounts both of these claims (J. Kelly, 1985; M. Giannelli, 1985). Unfortunately, to date the profession’s control over avenues of publication has prevented the dissemination of these rebuttals in the principal professional journals. Taken together, Kelly and Giannelli present a convincing case through data from citation analysis that Miller’s and other claims are unwarranted both with respect to degree of injury to animals and clinical applications for humans.

For this reason alone we welcome the use of the resources of the animal rights community in bringing before the public a more balanced understanding of psychology’s use of animal subjects.

In addition, the publication of a comprehensive analysis of the literature of maternal deprivation is a very fortunate choice for reasons Stephens indicates directly in his introductory chapter. At this point in the current dialogue we must complement the philosophic discussion, the general survey and the singular instance of alleged abuse with evaluations at the level of areas of research. While large, the number of studies in the area of maternal deprivation employing primates lends itself to such an evaluation: although ongoing for over thirty years, the lines of its development are readily grasped; it has been the object of criticism on ethical grounds by the animal rights community but at the same time almost every litany of “benefits” includes reference to it; it is an exemplar case of the animal model strategy of psychological investigation, a strategy that itself has become a significant object of debate.

Stephens’ effort excels in its clarity and its dispassionate tone. Much of the text concretely portrays this set of over 250 studies in an organized and straightforward manner that makes the research program and its impact on the animal subjects available to expert and layperson alike. While often simply allowing the record of the literature, including critical discussion by the researchers, to speak for itself, Stephens also provides his own critique. It is incisive and yet stays close to the facts.

We learn that the effects of this research have been profoundly injurious to over 7,000 animals. Despite this “cost,” direct applications from the research are virtually nonexistent. Stephens is able to trace only one application, which he then finds is an effective intervention for reasons other than those suggested by the research. We learn further that since its inception maternal deprivation research has largely been the offspring of one “family” of researchers, predominantly students of Harlow, trained by him at the University of Wisconsin. One implication of this surprisingly inbred genealogy is that this particular research program is not the inevitable process of impersonal institutional forces that have a momentum of their own, but rather is the result of the momentous personal conviction of a small group of individuals. In this context, Stephens argues that the preponderance of the studies represent no more than “parametric tinkering” and unnecessary elaborations on Harlow’s original work.

But what of the theoretical advances and the conceptual clarifications provided by Harlow’s paradigmatic experimental design—disrupting the primate mother/infant relation? Surely these begin to justify this research. As a clinician, I am not surprised at Stephens’ finding of few direct applications from this research to the treatment of depressed, isolated, or abused individuals. However, as Midgely (1981) has pointed out with respect to various psychological research areas, readers approaching this literature, both psychologists and others, assume that at least there is much gain here theoretically, that both major theoretical disputes and heretofore unnoticed conceptual distinctions were settled or crystallized through it. “I had assumed that a deeper structure of thought must be present” (p. 332). But her examination of, for example, Harlow’s studies of “chambering” as a model for the production of human depression reveals them to be based on vague intuitions of similarities and to fail to provide any sharpening of our thinking about depression. In his comprehensive analysis Stephens confirms this. With the exception of the critique of the “primary drive” theory of motivation, a critique provided by numerous other research findings by the 1950s, there has been little theoretical payoff from the maternal deprivation studies. Psychoanalytic theory and clinical studies, the work of Spitz and Bowlby, not to speak of common sense, all point to the pernicious effects of early deprivation and separation of child from parent and the positive effects of enriched contacts, including affectionate contact. Beyond this, the many particular variables examined in this research pro-gram have yielded very little information that is conceptually clarifying. Many of the particular findings, beyond the primary theme of the gross effects of deprivation, are not generalizable from rhesus to pigtail monkeys let alone from the former to humans.

None of the early hopes for discovering a useful animal model for depression or for maternal isolation have been realized. That they have not is not an indictment of the ingenuity of these researchers. The reason for the failure runs deeper. The strategy of developing animal models for phenomena, which are essentially a function of cultural arrangement and of linguistic mediation, is a poor one. This is not to say that nonhuman animals do not suffer from depression, anxiety, apprehensiveness about separation and the like. It is only to assert that the constitutive and determinative features of human depression, its forms and occasions and, undoubtedly, therefore, the keys to its treatment are linguistically and culturally embedded. It is not a question of “further research,” of more carefully teasing out confounding variables, or of replacing the blow of the sledgehammer with a few gentle taps (to use maternal deprivation researcher Mason’s critical image of animal models). To understand depression in our particular contemporary culture, we cannot simply tap, gently or otherwise, our nonhuman primate “resources”—not in any case and certainly not when those animals are bereft of those natural settings which they require to be themselves and therefore to show us their own complex natures.

Yet we cannot deny that Harlow and family have had an impact on psychology and even on our contemporary culture. In retrospect the primary impact of their work is the provision of graphic metaphors for the phenomena under investigation. Stephens asserts that, “many of these experiments were little more than sensational illustrations of what had already been established in humans” (p. 58). Considering psychology as a natural science, this is not a function that is justifiable. However, taking psychology as a peculiar institution within our culture, these studies have given us a set of metaphors: radical isolation as an infant primate reduced to a whimpering ball of fur, utterly and irreparably destroyed; rejection as in an unforgiving spiny-porcupine monster. These images are part of our culture now, part of us, and perhaps, even part of what we imagine when we are feeling isolated or depressed. We did not so much gain in understanding from these studies – for the metaphors have no independent exploratory value – as were we changed by them. On ethical grounds it is clear that we cannot justify the time, money, and gross exploitation of animals for “benefits” such as these.

Prospectively, I believe that the chief contribution of these unthoughtful sledgehammer attempts to solve complex human problems will be in the areas of ethics — for they will have demonstrated to us the cruel excesses of our use of and disrespect for other sentient and intelligent beings. Michael Foucault’s description of the administration of justice in eighteenth century France provides a historical parallel (1979). At that time the trial of the accused involved undergoing an ordeal, a systematized suffering. It was believed that only through such a “trial,” in the original sense of that term, and through the confession exacted in the process could the truth be established. In this primitive form of justice the magistrate did not intend cruelty, yet, in part, the callousness of this judicial system led to the upheavals of the French Revolution. Aspects of present-day American psychology, notably exemplified by the animal model investigations of psychopathology under study in this monograph, also involve unintended or at least incidental suffering in the service of truth seeking. In the not too distant future, we will have the same repugnance at this contemporary “primitive” science with its experimental manipulations in the name of truth as we do now at Foucault’s accounts of justice in eighteenth century France.

It is a common saying that nothing is as American as motherhood and apple pie—as innocent, as fresh, as life-sustaining, as good. Yet through these studies of motherhood we are taken to the lower reaches of a medieval Dantesque inferno. As the present monograph objectively and effectively demonstrates, this is the case because what shows through in this research is not the subject matter of the investigation but its method—a primitive form of truth seeking through suffering. Ultimately, what we gain from this extended investigation is not an understanding of the “nature of love” (Harlow’s term) nor even cultural metaphors for love or the effects of its absence. In the final analysis, we are given an image of a science, a science that has no recourse but to reform itself.

Kenneth J. Shapiro, Ph.D., A.B.P.P.

Bates College

Lewiston, ME 02440

*The first is Brandon Kuker-Reines, Psychology Experiments on Animals: A Critique of Animal Models of Human Psychopathology (1982).



Coile, D. and Miller, N. (1984). How Radical Animal Activists try to Mislead Humane People. American Psychologist, 39, 700-701.

Foucault, M. (1979). Discipline and Punishment: The Birth of the Prison. New York: Vintage.

Giannelli, M. (1985). Three Blind Mice, See how They Run: A Critique of Behavioral Research with Animals. In M.W. Fox & L.D. Mickley (eds), Advances in Animal Welfare Science 1985/86. Washington: HSUS.

Kelly, J. (1985). Personal communication.

Kuker-Reines, B. (1982). Psychology Experiments on Animals: A Critique of Animal Models of Human Psychopathology. Boston: NEAVS.

Midgely, M. (1981). Why Knowledge Matters. In D. Sparlinger (ed.), Animals in Research. New York: Wiley.

Miller, N. (1985). The Value of Behavioral Research on Animals. American Psychologist, 40, 423-440.

Please cite as:

Stephens, Martin L. (1986) Maternal Deprivation Experiments in Psychology: A Critique of Animal Models.
Retrieved from