“In psychology there are experimental methods and conceptual confusion … the existence of the experimental method makes us think we have the means of solving the problems which trouble us; though problem and method pass one another by.”
– Ludwig Wittgenstein, Philosophical Investigations (1949)
Social psychology is that sub-discipline of the broader field of psychology that investigates how the thoughts, emotions, and actions of individuals are conditioned by situations. It is a staple of undergraduate education in psychology, has one of the largest memberships in the American Psychological Association, and not infrequently is the recipient of federal grants for research that could inform policy on pressing social problems. Few phenomenon outside cases of clinical disorders fall beyond its purview, and since the late 19th century when Norman Triplett experimentally investigated the ability of co-actors to enhance children’s performance on a simple laboratory task, it has accumulated a veritable mountain of facts concerning the human condition that fills textbooks and often garners public attention. Despite its successes, however, recent events raise questions about its health.
A treatise published last year in Behavioral and Brain Science built upon a controversial presentation made by Jonathan Haidt in 2011 before the Society for Personality and Social Psychology. In it, the question of a liberal bias among social psychologists was raised, and a firestorm of dispute and disbelief erupted as a result. This publication is the most recent addition to the conversation on the subject, and it attempts to drive Prof. Haidt’s point home further with evidence and admonitions that ideological diversity will improve the state of the field.
Also, in August 2015 Science published a project by the Open Science Collaboration—an organization that evolved in response to questions of fraud and poor publication practices in psychology. The researchers attempted something extraordinary in it: to replicate 100 published studies in the domains of cognitive and social psychology following as closely as possible the methods of the original research. The project was a success, but the replications less so. Although replication rates were low overall, they were devastatingly so in social psychology (near 25%).
The going answer to both of these problems seem to be: 1) “There’s nothing actually wrong” (e.g., the OSC’s findings have recently been argued in Science to point in a favorable direction once methodological corrections are made), and 2) “Dig in deeper and try harder.” That is, by increasing the transparency of researchers’ practices in the laboratory and tightening the reins on publication standards, it is expected that the self-correcting methods of science will be realized, elevating social psychology to a new level of respectability. And by laboring after political diversity, it is hoped that the blinders of ideological homogeneity will be cast off, thus attenuating bias in the field and opening up new areas of inquiry not previously known.
These answers to the problems in social psychology—including empirical showings that things may not so bad—square well with the natural science model the field has patterned itself after. But taking the problems to be real and nontrivial, as I do, other answers become attractive when a pervasive obsession common to both difficulties is seriously considered.
In 1974, Nobel Prize-winning physicist Richard Feynman addressed graduates of Caltech with a story about a tribe in the South Seas who so longed for the material goods they saw delivered by planes during wartime that they fashioned for themselves a runway lined with torches and a wooden hut for a tribesman to sit in with coconut headphones. They had the form right, Feynman said, but they lacked the one thing necessary to get planes to land: “utter honesty,” the “kind of leaning over backwards” honesty that is the defining attribute of the best practitioners of science.
This tribe Feynman called a Cargo Cult, and the sciences of, for instance, persons and societies that aspire towards the predictive power and control enjoyed by physics, Cargo Cult Sciences.
“A great deal of [these sciences’] difficulty,” Feynman said, “is the difficulty of the subject and the inapplicability of the scientific method to the subject [emphasis added].” This, of course, is a shocking claim to true believers, but given Feynman’s achievements, it seems gratuitous to dismiss him on the grounds that he simply does not understand what psychologists do or are about. Might the kind of “leaning over backwards” honesty demanded of scientists require social psychologists to make an acknowledgment that approaches this kind of severity?
The question of method has always been alive in psychology. Wundt, and Titchner after him, advocated a particular kind of introspective analysis as the key to psychological inquiry, and Freud celebrated the technique of free association in evaluating dream contents as a window into the apparatus of the unconscious. In social psychology it is well known that the research cited as the earliest work in the field was laboratory-based and experimental: viz. Triplett’s work referenced above, Floyd Allport’s studies of social facilitation, and Kurt Lewin’s work on leadership styles and cooperation among boys.
But it is my hunch that with the publication of John B. Watson’s Psychology as the Behaviorist Views It in 1913, a particularly heavy and expansive shadow was cast upon all psychology that persists today despite the fact that behaviorism failed to synthesize all of the discipline under its authority.
Behaviorists quickly ascended the ladder of rigor leading up to the natural sciences, and it used this achievement along with the vision-casting power of B.F. Skinner, to play upon the sentiment of physics envy already lurking in the hearts of psychologists. Its seat of judgment was hard won from those convinced of the irrelevance of animal behavior to human consciousness, but once there, it modeled for onlookers a spirit of severity in choice of method that aspired towards the strictest objectivity.
I know colleagues who would say that social psychology demonstrated itself immune to the behaviorist’s disapproving glare by its persistent interest in the inner-state of individuals (e.g., their motives and attitudes). But just because a difference in subject matter always stood clear, this does not mean that the behaviorists’ insistence on irreproachable methods did not carry over and breed insecurity. And when behaviorism faded from prominence, the work was already done: its posture of “model science” was translated into a pecking order of psychology’s subfields in which biological and cognitive psychologies stood at the top and social-personality psychology at the near bottom. In this position, the envy calcified, leaving us where we are today, with an inability to think beyond the pale of our methods, and no desire to do so.
Whether my hunch is right is almost beside the point. Even if different and multiple causes of method obsession are imaginable, as surely they are, I believe that it is a woefully inadequate education in the history of psychology that vitiates knowledge of the trends that have shaped our contemporary modes of conduct. I hold also that insufficient appreciation for the philosophical commitments that motivate experimental work with human participants furthers our intellectual myopia. For these reasons, I cannot help but draw a connection between this state of affairs and the wish voiced by Watson in 1913 to raise up future psychologists “ignorant” of philosophical problems.
Few students of social psychology pause to ask the question “Why?” They stand “on the cutting edge,” but when asked “of what,” or “why” that is where they stand, they are capable of giving only the shallowest answers. This is because they have been educated as specialists and technicians and thus have no language to speak otherwise. I think Watson got his wish.
All of this contributed to a risky move I made in the final meeting of my undergraduate social psychology class last fall: I announced to my students that in response to the extant problems in social psychology, I would be forsaking my old line of research, and embarking instead on an inquiry of recovery.
What if the measurements and experiments, which are presently so precious to social psychologists as the sine qua non of their work, were called into question and compared to the strategies assumed by individuals who wrote the earliest books bearing the name of the field? Of course Allport and William McDougall held affinities for the lab, but was this all or the best that they had to offer? It is true that Allport devoted a chapter to measuring personality, but it is one of the shortest he wrote, and neither he nor McDougall wrote chapters dealing solely with methods—what is inevitably and vitally treated in the first hundred pages of all modern social psychology textbooks.
More courageous still would be considering texts that reach outside the strict bounds of social psychology for inspiration: Donald Snygg and Arthur Comb’s Individual Behavior, for instance, or Wilhelm Wundt’s Völkerpsychologie. Although approaching the problem from a sociological framework, Edward Ross’ Social Psychology also surely holds something to offer. And dare I add to this list, Freud’s Group Processes and the Analysis of the Ego?
I sadly confess that I am unsure what these texts hold. Despite receiving the terminal degree in my field, I, along with most of my colleagues, was never asked to read these thinkers who all agree are pioneers—and recent ones at that; some wrote less than a century ago. Instead, I can only say what I intuit to be the case from reading excerpts from pieces written in the same era (e.g., Kurt Lewin’s Principles of Topological Psychology): that the answer must be, “No.” I sense that the approaches taken by our forebears were broader, more nuanced, and yet not lacking in the sophistication of the grandest methods our discipline now has to offer as proof of its importance.
It is for this reason that in the social psychology class I teach next fall I will jettison the modern textbook with all its accoutrements and return to the works of those authors above whom I typically treat as mere bullet points in the ever-so-short history segment of my course. It is, as I said above, an inquiry of recovery, and I am taking my students with me.
One final observation concerns not the matter of method in social psychology, but its epistemological grounding. To my knowledge, no student of the field seriously confronts this subject, and it is of great disservice to one’s education. Having recently taught an abbreviated seminar on the philosophy of social science, I can say that the ideas of Auguste Comte and John Stuart Mill, in whom modern social psychology has some of its roots, leads one to wonder whether there could ever be such a thing as a more conservative social psychology. These influential thinkers saw the culmination of science in the study of persons and societies as supplanting all former approaches to these matters. This sentiment, I trust it is plain to see, is deeply antithetical to the conservative idea that some old principles are indispensable to human flourishing. If conservatives avoid the ranks of social psychology, perhaps it is because they do not feel the need to treat experiments and quantities as the best or final arbiters of their social convictions. Asking them to give up this commitment and hold their beliefs in abeyance until the mechanisms of science establish their veracity is, in fact, asking them to make Comte’s positivism prior to their conservatism.
I believe that if social psychology is to find a remedy for its current ailments, and if it truly wishes to accommodate conservative thought in its quarters, assuming it is possible to do so, it must pursue a leaning-over-backwards-honesty not just in its application of scientific methods to its questions, but in considering whether these methods are the only or best to apply to its questions, and where the motivation to do so comes from. In doing this, the field will find itself making a conservative move in the truest sense of the phrase, even in the absence of conservative membership: looking to the past for answers to their now-besetting problems.
Books on the topic of this essay may be found in The Imaginative Conservative Bookstore.
As a fellow social psychologist* (who also teaches stats and methods), I look forward to your report of the outcome of your project. I too have never read those books, and am frankly at a bit of a loss as to how I would even go about my business of social psychology using other methods, especially since our reliance upon empirical (experimental and correlational) research is typically what we point to as defining our field as different from areas like social philosophy or cultural anthropology.
I have long railed against the general historical and philosophical ignorance that is rampant in psychology, despite the strong relevance of such areas to our work. I do hope, though, not to throw the baby out with the bathwater, as I consider it a good thing to put our clever ideas about the human condition to the test. I would hope that a “conservative” social psychology would be broad enough to encompass current data-driven approaches as well as those other approaches in which I have no training. In fact (and now I’m just shooting form the hip here), I wonder if our reliance on hard data is all that can save us from slipping into the same state of affairs that characterizes the humanities.
*technically social/personality
Charles, thank you for your thoughtful comment. I would simply ask in reply: Does putting our “ideas about the human condition to the test” strictly require us to turn to experimentation and measurement? Surely there are other ways of “testing;” surely verification of our convictions can proceed on other grounds. I would add too that I think we have more to learn from the greatest representatives of the humanities than we let on.
Good question. And that is exactly my problem. How else *do* we test them, if not by translating our questions into empirically testable hypotheses? When I teach methods, I have difficulty when we get to qualitative approaches, because they just look an awful lot me like “make it up as I go and subjectively eyeball the outcome”. I am open to the possibility that this is a blind spot for me, resulting from the kind of training I received in grad school.
The connection to the humanities that makes sense to me involves combining conceptual and empirical work. For example, my dissertation involved terror management theory, which is about taking Becker-inspired existentialism and using that to generate testable hypotheses.
So maybe my methodological horizons need to be broadened. I hope that you’ll post an article describing the outcome of this project.
I responded to your reply, but inadvertently posted it as a new comment below. I’m interested to hear your thoughts and am working on a longer essay now that deals with the problem of objectivity in social psychology. More to come, I hope.
My first experience in a Social Psychology class was in the Spring semester of 1967. Beyond the textbook (Newcom, Turner, Converse says memory), was a required reading of The Authoritarian Personality. I dutifully read the book, and went to my professor protesting the study, and especially that the “Fascism Scale” was so much horse manure. He concurred, but kept the book anyway. (It was a litmus test for who could spot [such manure] in such matters, I found out.)
Unfortunately, most of the students in that class took Adorno as gospel. Today, I see all over the Internet how “low-education voters” are trending Authoritarian, and I have flashbacks to that damned silly book.
Perhaps I should be grateful, as it was my introduction to how a researcher could skew questions to get the results he wanted. Researcher bias is becoming well known, but I dare say, a wise person could skew the MMPI results (or especially, any of its Human Resources Departments miniscule exams for ‘job fitness’.) — The “validation” is blatant if you know where to look for it.
It is interesting that you bring up Becker because it is exactly terror management theory that comes to mind when I think of empirical methods being used where they are not needed. I read Becker for the first time a couple of years ago. Here’s the conclusion I came to in a talk I gave last semester at Hillsdale College. Forgive me for quoting it at length, but I’m not sure I could state my position any better now.
“In graduate school, I became attracted to a body of research concerning the effect of having people think about their deaths on a range of outcomes including judgments of social deviants and attitudes towards those we do not consider as belonging to the same category as “us.” The ideas that motivated this research originated with a cultural anthropologist, Ernest Becker, who proposed on the basis of psychoanalysis, that the fear of death is common to all humans as self-aware beings. Cultural worldviews serve as buffers against the threat of death by promising to the individual who adheres to them literal or figurative immortality. It is for this reason that psychologists who experimentally induce reminders of personal mortality sometimes measure the outcomes I just mentioned – e.g., punishing social deviants is valued by society and, therefore, doing so serves as a mechanism for worldview validation that buffers the self against death anxiety.
Becker’s book on the subject did not rely on the methods of natural science. It was not experimental, not quantitative, not controlled or unbiased, whatever that means. It did not recruit participants to measure attitudes or observe unobtrusively. It did not require laboratory space, a government grant, or the approval of physics. It had much more in common with poetry, and yet it is some of the best psychology I have ever read. I believed his claims in the absence of scientifically obtained facts. Indeed, one is inclined to wonder what the empirical and experimental investigations of Becker’s ideas appearing in respected psychology journals adds to my valuation of and belief in the importance of death denial in shaping human action. You see, the researchers who undertook this work called themselves treating Becker’s claims as hypotheses – turning his statements of “I believe” into statements of “I predict” – and judging whether they held up to facts. I am suggesting that we did not need this work to believe or disbelieve Becker. The kind of idea he held out is one that is validated on entirely different grounds than empirical experimentation. It is validated through inhabiting his conceptions and judging through the process how well they synthesize the experience of reality. This is the only way such ideas can be “tested,” and does not the fact that his book received The Pulitzer Prize show that it passed? I take this as a powerful example of psychologists not appreciating where their methods are not needed.”
To summarize briefly: The empirical/experimental analysis of Becker’s theory requires that I hold his ideas at arm’s length until the hypotheses we derive from his concepts (which alters them in a fundamental way) are tested and confirmed, but such a procedure could never approach verification on the terms Becker’s ideas require. Why? Because they demand a commitment of all of ourselves – the whole of our beings – to his world of understanding. Only then and in that way will we ever see whether he was right or wrong. I would say too that this why experimentalists are opposed to Freud, Jung, and Rogers, to name a few.
I’ll end with this: I genuinely hope you do not take this as a criticism of your work that I have not also applied to myself. The very reason I am willing to forsake my old line of research is because I see similar problems in it. Our notions of objectivity are deeply flawed. Michael Polanyi’s “Personal Knowledge” reveals this most excellently.
I read THE AUTHORITARIAN PERSONALITY when it first came out. I did so at the recommendation of a friend. I was long since out of school by then. The circular reasoning was appalling. I wondered, do people really take this seriously? Evidently they do.
Very interesting read. I find it astounding while currently picking my way through Crime and Punishment that many of our modern views on behavior and psychology have not really changed much at all in almost 150 years. Dostoevsky’s descriptions could line up perfectly with any newsroom panel psychologist naming off a list of behaviors in the wake of a contemporary double murder. I don’t know if that is a result of no search for a “why” or what.
What are your thoughts on the synthesis of Psychology and Neurology? I am a huge fan of J Allen Hobson who is an expert in both and tries to balance the two (although he strays toward neurology). His approach allows for an injection of testable empiricism. (Honestly I’m of the persuasion that you can’t have comprehensible psychology without neurology andthe reverse) Notably his combined method explores dreams as chemical realities that can also point to stresses in our waking life. He also addresses the poetry of dreams as a physical reality. Our minds divine our dreams because we are peering briefly into the inner working of our own hardware as we wake. He discusses how our minds speak two chemical languages and the bleeding from one into the other creates many of these moments of insight.
Sorry to go off the rails a bit there.