I GOT MY IDEAL job, and it was for life. I was becoming well-known in experimental psychology, but its lack of application to real human affairs gnawed at me. I’d been hired by the most “basic” (some would say “constipated”) psychology department in America partly as a sop to student demands for “relevance,” and some of my colleagues already saw me as a wolf in sheep’s clothing. They were looking over my shoulder, and they were right to be leery. Experimental psychology and clinical psychology were just tools—mere methods—for approaching bigger questions. There were other tools to be tried, and I wanted to spread my wings to explore the bigger questions.
As I had been five years before, when I started at Cornell, I was fast and in a hurry. When I read an article, no matter how important, I sprinted to the finish and got onto the next item of work. I missed fine points. I cut off conversations once I got the gist, which usually happened sooner rather than later, and just as they had at Cornell, people saw me as brusque or even rude and self-obsessed. I did not step back and savor what I’d accomplished.
I was out of touch with the people who loved me and with the friends who liked me. I worked eighty hours a week, and when not overtly working, I was always thinking about psychology and philosophy. I paid too little attention to my wife and daughter, to my mother and father. My only friendships were with my students and colleagues.
I was pretty negative. Considerable and unexpected success bolstered my outward confidence and optimism, but underneath I was still anxious a lot. I brooded. When thwarted, I got depressed, but unlike my dogs, I never gave up. I readily saw what was wrong in psychology, but sometimes I also saw what was correct, once in a while down to its very core.
“NEAL IS REALLY pissed at you, Marty,” said Jay Weiss. It was a gorgeous late-April day in Rittenhouse Square in Philadelphia, the white apple blossoms exploding. The occasion was the annual meeting of the Eastern Psychological Association, and Jay had presented his doctoral research, conducted with Neal Miller at Yale using the triadic design in rats and looking at stomach ulcers. Jay found that the rats with inescapable shock got the most stomach ulcers.1 This debunked a famous study in monkeys by Joseph V. Brady purportedly showing that the “executives,” the monkeys who could turn off the shock, got the ulcers.2 But Brady made the mistake of not assigning his monkeys randomly to escapable or inescapable shock; rather he made the monkeys who started to respond earliest the executives. So the monkeys most upset by the shock became the executives and got the ulcers. Brady inferred that helplessness was good for you, and control was bad. Jay got it right and randomly assigned his rats in the triadic design, and his results were clear and opposite to Brady’s. It’s good to be the executive and bad to be the slave. Jay’s was a lovely piece of science.
“What’s eating Neal?” I asked, puzzled. The last thing I wanted to do was alienate the most powerful experimental psychologist in the world.
“Your articles about controllable and uncontrollable shock and predictable and unpredictable shock don’t cite him as having priority. Neal worked on this as Arlo Myers’s supervisor in 1956, even though they never published,” said Jay.
I had gotten pretty full of myself on the importance of learned helplessness, and this conversation was deflating. But it was also thought provoking. Naturally I thought that Steve and I “owned” learned helplessness (or at least Dick Solomon or Russ Leaf or Bruce Overmier did) and the triadic design as well. I had never heard of Arlo Myers, and when I later looked up his dissertation, it did not seem particularly on point.
But who “owns” what? Who should get the “credit”? Who has “priority”? That someone as illustrious and established as Neal Miller worried about such stuff told me this issue was serious and deep. Priority had engendered much bitterness and lasting enmity in science, and almost everyone tended to overclaim for themselves and undercite everyone else. I knew how annoyed and resentful I felt when someone failed to cite me.
Dick Solomon provided the answer. He was intellectually generous. He was a fountain of ideas—some good, more half-baked—and he spewed them forth every Thursday in his luncheon seminar. He encouraged us to do the same. I resolved to be like Dick and not like Neal. I had lots of ideas and was not afraid to give voice to them, even if they turned out to be wrong or even foolish. I erred on the side of not being boring rather than that of being right. Ownership in my view belonged to the person who ran with the idea, not to the person who thought it up. For the rest of my life, I spewed forth ideas, not vetting them as thoroughly as others did, and I encouraged my students and colleagues to run with them. When they later asked me to be first author of a resulting article or fifth author because I had thought the original idea up, I usually declined. I made the mistake of under-citing—but usually from lack of scholarship, not from lack of generosity in giving credit. I also made the mistake of over-claiming—but usually for the sake of pushing the limits of an idea, not for the sake of pushing myself or for publicity.
I didn’t have answers for the general problem of ownership of ideas, but with the hindsight of fifty years, I want to put ownership in perspective for the case of learned helplessness.
Learned helplessness illuminated a dimension consigned to silence by behaviorism and therefore unresearchable: the dimension of personal control. Control is both a cognition and a cognition about the future, both forbidden by behaviorism. To have a contingency between a voluntary response and an outcome is to control that outcome. The dog in the escape group controls the shock, and the dog expects to have control in the future.
Is the dog motivated to want control for its own sake? Do we want more than just lots of good stuff and little bad stuff? Do we also want control over the good stuff and the bad stuff? Is control a basic motivation?
“Control” meant something different in the 1960s than it means fifty years later. Then it was used not in the sense of personal control but almost entirely in sense of governmental or organizational control.
When our work was translated into German (and French, Portuguese, and Spanish), we were told something was “off” about our use of the term. In German, individual people don’t controllieren; only the government does. But something was blowing in the wind, and learned helplessness was one of the squalls. It was the idea that individual people, you and I, can and should control what is important to us.
Control certainly mattered in and of itself for the white-footed deer mouse in a 1967 experiment. This creature, when captured, was in a continuous wrestling match with the humans who thought they ran the lab. The humans set the lights to bright, and the mouse could press a bar to alter the level of lighting. The mouse stepped down the illumination to live in dim lighting. Maybe the mouse just liked dimness. So the next day the humans stepped the lights down to the chosen dimness, but the mouse again fought them and set its ambient lighting up to bright. The mouse seemed to care not about any particular level of light but about having control over its light.3
Not just mice. Old people as well.
When I was a graduate student, an ever-smiling undergraduate worked alongside us in Dick Solomon’s lab and was great fun to be around. Aside from bustling around the lab and enchanting all of us, she was also president of student government in the College for Women. Her name was Judith Seitz, soon to be Judy Rodin. Judy went on to a PhD at Columbia under one of the most illustrious social psychologists of the twentieth century, Stanley Schachter.
Just at the time I settled back in at Penn, Judy, by then an assistant professor at Yale, teamed up with Ellen Langer. Ellen was a pistol, and experimenting in the real world with real people came as second nature to her, as it did for Stan Schachter. She and Judy decided to look at the effect of control in a setting that often deprived human beings of it: a nursing home.4
Judith Rodin was an undergrad at Penn when I was a graduate student. She went on to become a pioneering researcher in personal control, then the provost at Yale, then president at Penn, and ultimately the president of the Rockefeller Foundation. Photo courtesy of the Rockefeller Foundation.
On the fourth floor of the nursing home, Ellen and Judy gave forty-seven patients a lecture. The lecture stressed responsibility: you decide how to arrange your room, and you decide how to spend your time. Each patient also picked out a houseplant that was theirs to keep and tend.
On the second floor, forty-seven other patients got a lecture stressing all the good things that were done for them. The staff arranged their rooms for them and scheduled how they spent their time. Each patient was given a houseplant and told that the nurse would take care of it.
Ellen and Judy decided to measure activity. The 1970s were lowtech rubber band and paper clip days, so they cleverly put white adhesive tape on the wheels of the wheelchairs and just measured how dirty the tape got. The patients on the “control” floor were more active and healthier, and eighteen months later only seven (15 percent) had died. Thirteen (30 percent) of the patients from the second floor had died.
Ellen and Judy argued that having control was a basic source of vitality and that being deprived of it might be lethal for an otherwise weakened person. For now, keep in mind that there were only two groups in the Langer and Rodin experiment, so it was unclear if having control enhanced vitality and/or not having control diminished it. (An ideal experiment would have included a third group that received no lecture at all.) It took another thirty years before Steve Maier figured out which was likely true.
It was an indication of the growing respectability of control as a legitimate object of study that Ellen went on to become the first tenured woman in the psychology department at Harvard and Judy became provost at Yale, then president of Penn, and finally head of the powerful Rockefeller Foundation.
Other influential theorists started to talk about control beginning in the late 1950s. I omit the great historical forerunners: Pico della Mirandola, Erasmus, Francis Bacon, Arminius, Friedrich Nietzsche, and Alfred Adler, to name just six. (I hope someday to do them justice.) The first modern was Robert White, who published on “competence” in 1959: people and animals, he asserted, play and explore for their own sake.5 This paper was widely read; indeed, it was assigned to me as an undergraduate and stuck in my memory when I went to work on learned helplessness. Julian Rotter, a social psychologist, developed the concept of “locus of control,” with some people being “externals,” believing that the environment or other people shape their lives, and others being “internals,” believing that they themselves shape their own lives.6 This attracted a large following in social psychology, but I did not read about it until after we published the first learned helplessness experiments. Stanford professor Albert Bandura, today the most-cited living psychologist, touted the concept of “self-efficacy.” An “efficacy expectation” is the belief that you, yourself, can bring about the outcomes you desire, whereas an “outcome expectation” is the belief that the outcome you desire will simply happen. He castigated us for not making this distinction. Self-efficacy equals control.7
Richard Lazarus at Berkeley proposed a cognitive “appraisal” theory. When an event occurs, the organism first assesses the consequences of the event (primary appraisal), then asks if it has the ability to cope with the event and its consequences. This is “secondary appraisal.”8 Coping was Lazarus-speak for control.
So learned helplessness, finding the downside of not having control, was really just one step in a vigorous march emphasizing the benefits of the presence of control. While we self-centeredly thought of our own work as huge, Steve and I were in fact just part of two intertwined developments making their way into mainstream psychology: the new acceptance of cognition and the new dimension of control. Our particular niche was the absence of control, helplessness. By the turn of the millennium, both control and cognition were firmly entrenched in mainstream psychology.
BY 1972, the learned helplessness was catching on. Other scientists had replicated it in rats and mice, and it would soon be found in gold-fish and cats and even cockroaches.9 This was a relief to me, since I hoped that we had found a general process, but at first I worried that we had tapped into some quirky human-dog submission pattern, and that learned helplessness was just a species-specific artifact. (Behold the external validity problem once again.) A similar worry was that the expectation Nothing I do matters was not general but somehow set off by trauma. Did learned helplessness occur across good and neutral events as well as bad ones? To find out, researchers gave food to rats independently of their actions, and these rats later had trouble learning to press a bar for food. They became passive. The same seemed to hold when lights and sounds were first presented independently of action.10
Yes, there was an “entitlement” effect, and “spoiled” rats learned to be lazy about good events and didn’t even try to control them.
But for all that progress, I had real doubts about the external validity of the learned helplessness endeavor. Was it just a laboratory curiosity that like so many others would gather dust on library shelves? Was it a model of real depression in the clinic? Did it capture the essence of what a human being experiences when her spouse dies, or bankruptcy occurs, or she is rejected by her boyfriend? The next logical step on the path to external validity was for some brave soul to do laboratory learned helplessness experiments in people.
The answer to my prayers came in the form of Donald Hiroto. As a graduate student at Oregon State University, Hiroto built a human shuttle box, a two-chambered device about four feet long. The human subject, wearing earphones, heard a very loud tone while he rested one hand in the shuttle box. If he moved his hand to the other side of the shuttle box, the tone stopped, and people easily learned to escape the tone. Against this background, on the previous day in another room, Hiroto recreated the triadic design: all wearing earphones, one group got the tone but could escape it by pushing a button. The second group was yoked and got an inescapable tone, and the third group got no tone. The next day in the shuttle box the results exactly paralleled those in rats and dogs. Two-thirds of the individuals who got the inescapable tone just left their hands in the shuttle box passively and did not even try to escape.
An exceedingly formal Japanese American, Don arrived at Penn as my very first postdoctoral student. He was obsessively thorough, exactly the right kind of researcher needed to build the systematic case for external validity. He insisted on calling me “Professor Seligman,” even though he was older and much better dressed than I.
“Professor Seligman,” Don said, “let’s do exactly what other people are doing in rats and find out if helplessness in people is general. If you are helpless about escaping noise, do you become helpless about solving cognitive problems and back again.” He then did exactly that, looking at the interchangeability of helplessness across two instrumental and two cognitive tasks. Three groups of people got anagrams: solvable (e.g., OTAIP), unsolvable (e.g., OTEIP), or nothing, and the next day they went to the shuttle box. The solvable group and the zero group escaped the loud noise well. Two-thirds of the people in the unsolvable group, however, just sat there passively and didn’t even try to escape the noise. Similarly, helplessness regarding the loud tone debilitated people in solving later cognitive problems.11
I was proud of this work and gushed about it to Tim Beck during one of our monthly lunches at Kelly and Cohen’s. (There was no Kelly, but it was still the most popular deli in West Philadelphia.) Tim was usually the gentlest and most nondirective of critics, so I choked on my grilled Reuben when he said, “Marty, if you continue down this path, you will waste your life.”
I was stirred but not yet shaken. Tim was telling me to take up new methods and to do a better job with external validity by turning my work to deal directly with real people and real problems. But it took a numinous dream a few nights later to shake my foundations: I somehow found myself in the New York Guggenheim Museum. I was slowly plodding up its famous curving ramp. There were rooms off to the right every few paces, and in the rooms people were playing with cards. They seemed to be tarot cards.
I asked, “Why is everyone playing with cards?”
Whereupon the roof of the museum opened, and God appeared—only his head and not his body. In case you are wondering, God is very old, he is male, he has a white, well-tended, curly beard, and he has a booming bass voice. He said, unforgettably, “Seligman, at least you are starting to ask the right questions.”