CHAPTER SEVEN
The Warmth of Minnesota
“I wanted to be the best husband and father I could be.”
Jason and I are in a huge, cavernous railroad station, something like Grand Central Terminal. We know that the train is about to leave, but we don’t have tickets and we don’t know which of the multilevel tracks our train will be departing from. We are racing around but cannot find the ticket window. Finally, we spot one in the far end of the station. “Wait here,” Jason says. “I’ll take care of it.” Wallet in hand, he dashes to the ticket window. When he returns, he is gasping for breath; he hands me a ticket and, pointing, says, “You’d better hurry; the train is about to leave on that track over there.”
“Aren’t you coming?”
“I’m afraid I can’t join you. You’ll have to go it alone from here on. Don’t worry, and don’t look back. You will be fine.”
Somehow, I am not surprised to hear him say that. And, in the dream, I am surprised at not being surprised. We hug, I take a deep breath, and I hop on the train just as it is beginning to move. I look back; Jason waves to me in a sad salute.
No need for Sigmund Freud to interpret this one. When I was seventeen years old, my brother, in defiance of our uncles, guided me to the right track, put me on a fast train, and, then, when my own family and career were secure, he died. He remained behind, forever at the station.
I missed Jason terribly. Next to Vera, he had been my closest friend and confidant. Once the initial shock of his death began to subside, I found myself turning inward, focusing on my own mortality. It was suddenly obvious to me that the Aronson family was cursed with a genetic defect that made us more cancer prone than the average family. Not only had that dreaded disease struck down my father and brother in their prime, but it had also killed two of my father’s siblings in their thirties. My first reaction, worried about financial security for Vera and the kids, was to take out a huge life insurance policy. If I was going to die young, I would not leave my wife and kids destitute. The Prudential Insurance Company was only too happy to sell a large policy to a twenty-nine-year-old professor in perfect health.
I began joking to my friends and colleagues about “shortgevity” running in my family. I believe I did that, ad nauseam, as a kind of memento mori; that lighthearted statement did for me what the skull on the desk did for a medieval scholar—serve as a reminder that life is too short and unpredictable to waste much time on unimportant or uninteresting things. So, I told myself, I had damn well better make use of whatever time I had left. I said to myself, “Okay, schmuck, stop wallowing in grief and self-pity. Let’s suppose, seriously, that you have only four or five years left to live; how do you want to spend your time?”
I wanted to spend more time with Vera and the kids. More than spending time, I wanted to be the best husband and father I could be; Hal, Neal, and Julie were now six, five, and three, and Joshua was an easy, happy baby. I also wanted to be the best teacher, the best researcher, and the best colleague I could possibly be. I always had those ambitions, but now my focus was sharper and more immediate. I figured out ways to spend more time with my family without taking anything away from the work I loved doing. I concentrated on being fully present, actively making plans with and for Vera and the kids.
I was amazed to learn how much time there is for everything if one doesn’t waste a good deal of it on trivia. I worked hard and well at the university, but almost all evenings and weekend days became sacrosanct family time. I refused to let course preparation, the writing up of research articles, colloquium invitations, or other business intrude on Saturdays and Sundays during the day. When I was home, I was really there. Although I fiercely guarded weekend days as a time to enjoy the children, weekend evenings were something else. Vera and I frequently hired babysitters so that after we tucked the kids in, we could step out, going to movies, theaters, or restaurants, or to one of the faculty parties that my sociable colleagues were forever throwing. Vera and I made a conscious decision to be more than the parents of four children. It was important to us to continue to think of ourselves as young lovers. On weekend nights when we did stay home, we conspired to keep our romance alive. After putting the kids to bed, we prepared simple but elegant late-night dinners with candlelight and wine.
Not all the changes I went through as a result of Jason’s death were positive and rational. I began drinking far more heavily. In the past I had enjoyed nightcaps with Vera as well as a martini or two with my graduate students whenever there was something to celebrate. But in Minnesota my drinking became more intense and more frequent. At parties I often got drunk. Vera and my friends told me that I was an affectionate drunk, not a belligerent one, but the fact is that they (usually Vera) had to tell me because, the morning after, my memory of my behavior was cloudy at best.
Vera also kept cautioning me that I was driving too fast and taking crazy risks. As I look back, I see what an irrational thing that was for me to do, given how hell-bent I was, at least consciously, on being present (and alive!) for my family for as long as possible. Many years later I learned that the three most common manifestations of grief among men are heavy drinking, reckless driving, . . . and singing sad, sentimental songs in the shower. Bingo: I was three for three. Fortunately, the intensity of my grief and my irrational behavior subsided within two years and did not result in any wrecked cars or strained friendships. Yet my dream of Jason at the train station has recurred all my life, serving as a reminder to keep things in focus and in perspective.
When Vera and I first arrived in Minneapolis, we rented a house some distance from the campus. The following September, when our lease was about to expire, we decided to buy a home of our own, though we didn’t have much money. After searching for a few months, we found only two houses that were large enough and in our price range that we both liked. They could not have been more different. One was a charming old Victorian, within walking distance of the campus. I liked it a lot, primarily because it meant that my colleagues and graduate students could casually drop in on us to discuss research or just schmooze. But that house was adjacent to an industrial area, without much space for our children to play. The other choice was a tract house in the suburbs, newer but without distinction, a thirty-minute drive from campus. It had two advantages: It had a large yard and was only a mile from Lake Owasso. Eventually, that was the house we chose, because we agreed it was better for the kids.
Not long after we moved in, I saw an ad for a used canoe and immediately bought it as a surprise for Vera and the children. When I arrived home on that freezing, bleak December day with the canoe lashed to the roof of my car, Vera burst into laughter.
“What’s so funny?” I asked.
“Ask Leon!” she said.
Of course! I had felt so much dissonance about buying the house in the suburbs that I needed to do something right away to justify that purchase. I somehow managed to ignore the obvious fact that it was the beginning of the long winter season in Minneapolis and would be another five months before the frozen lake would thaw out enough for the canoe to be usable. During the winter Minnesotans frequently take shortcuts by driving across lakes; they joke that the unofficial arrival of spring is the first car that falls through the ice. So, for five months, my canoe sat lazily in our garage. But it did more than take up space; it helped me come to terms with giving up my dream of living close to campus.
When I first moved into my office at the Laboratory for Research in Social Relations, I found, to my astonishment, that the lab was not the beehive of activity that it had been when I visited the previous spring. Indeed, things looked bleak. For one thing, the secretary, the very person whom Stan Schachter had assured me would know all the ropes and would relieve me of the onerous administrative work, was seven months pregnant and had one foot out the door. In addition, there were only a few graduate students. When Hal Kelley and Stan Schachter moved on, they had taken their favorite students with them, and those who remained were busily working on their dissertations or already committed to working with either Dana Bramel or Ben Willerman, my two colleagues at the lab. What to do? I needed a research assistant, and no one was available. Dana and Ben informed me that there was a wealth of talent in the Twin Cities and suggested that I advertise in the Minneapolis Tribune.
It seemed a strange way to hire a research assistant, but it turned out to be a great idea. I received a flood of responses and interviewed the eleven who seemed most qualified. Most of them could have done the job, but one interviewee struck me as being uncommonly bright. Her name was Ellen Berscheid. Ellen had earned a master’s degree from the University of Nevada and was working for Pillsbury Mills in marketing, “exploring,” as she put it, “ways to persuade housewives to buy cake mix.” Her behavior during the interview puzzled me. She told me she really wanted this job, yet she was not going out of her way to make a good impression. Far from it. From the moment she walked in the door, her demeanor was feisty and contentious; she seemed primed to argue with me about ideas great and small. I had the feeling that if I had said, “It’s a sunny day” (which it was), she would have replied sharply, “Nonsense. It is about to snow.”
But she was clearly the most qualified candidate, so I offered her the job. After a few months I was so impressed with her that I urged her to take some graduate courses and work toward a Ph.D., which she did, finishing in record time. Thirty-five years later, when Ellen was given the APA’s highest honor, the Distinguished Scientific Contribution Award, she wrote a brief professional memoir. On reading it, I was charmed to learn that she regarded our interview as having been a major turning point in her life. Because she had suffered from the pervasive sexist attitudes and treatment in her previous job, she wrote, she had walked into my office expecting more of the same. She was surprised and delighted that I had selected her on the basis of her qualifications alone, in spite of what she referred to as her “wrong gender and undeniably bad attitude.”
The new students were a smart and lively group. Dana, Ben, and I had recruited and selected them carefully, and by October the lab was humming with exciting research. The new graduate students included a brilliant Jesuit priest (Eugene Gerard) and a gifted mountain climber (Darwyn Linder). John Darley, having completed his course work at Harvard, now joined me at Minnesota as my majordomo and role model for the newer students. On a postdoctoral level, our greatest acquisition came out of the blue. One day I received a call from Leon Festinger, who told me that he had a first-rate new Ph.D., Elaine Hatfield Walster, who was having trouble getting an academic position, given the bias against hiring women at most universities at that time. Leon asked me if there was any way we could find a job for her at Minnesota.
“How good is she?” I asked.
“You know Stan Schachter?”
“Of course,” I said.
“You know Elliot Aronson?”
“A little bit,” I said.
“Well,” Leon said, “she is in that class.”
There was, however, no position open in our department for another social psychologist, so I thought that was the end of it. But a few weeks later, when I learned that the Student Activities Bureau was advertising for a research specialist to arrange freshmen orientation dances, I called the dean of students.
“This is a golden opportunity for you and for us,” I said. “Why hire a hack researcher to collect data on who goes to the dances and whether or not they had a good time? Why not hire a skillful experimental social psychologist who will make something interesting out of this opportunity and who will do the job right—maybe even discover something of scientific importance? If you hire the right person, we in the lab will work with him or her as well. You get a capable scientist with built-in consultants, and we get a good colleague.”
“Terrific idea,” the dean said. “Do you have anyone in mind?”
“It just so happens . . . ,” I said, and so Elaine came to Minnesota. She in fact did use this assignment as a vehicle for a groundbreaking study on dating preferences among college students, which was published in our leading journal. We gave her an office in the lab, where, just as I had hoped, she spent most of her time working and interacting with the grad students, who treated her as if she were a regular faculty member. She proved to be an excellent, intuitive researcher and became one of my all-time favorite colleagues. Because she had no teaching responsibilities, for Elaine the job was like having a high-paid postdoctoral fellowship. After two years she was offered and accepted an official position as an assistant professor in the Psychology Department. Eventually, Elaine and Ellen Berscheid teamed up to form what proved to be a highly productive research partnership that lasted for more than two decades.
The weather in Minnesota was frigid, but the social climate was warm and could not have been more different from the one I experienced at Harvard. There were no barriers between faculty members as a function of rank or tenure. Newly hatched assistant professors joked freely with their most distinguished colleagues, who included Starke Hathaway (the inventor of the Minnesota Multiphasic Personality Inventory, the single most important personality test of that era and since), David Lykken (a national expert on the physiology of lie detection), Norman Garmezy (a clinical psychologist ahead of his time in studying children’s resilience), and Gardner Lindzey (editor of the monumental two-volume Handbook of Social Psychology). There wasn’t any antagonism between the clinicians and the experimental psychologists, as was true in most departments at that time, with each side feeling that they were doing the important work—the clinicians in the study of real people, the experimentalists in the rigor of their science. I am convinced that the primary force underlying this atmosphere of respect was Paul Meehl. Meehl was extraordinarily smart, an excellent experimentalist, a masterful psychotherapist, a brilliant philosopher of science, and a first-rate psychometrician. When the acknowledged superstar of the department bridges all subdisciplines, how can any of the lesser stars quarrel with each other?
I believe that Meehl’s stature and ideas contributed to one other major difference between Harvard and Minnesota: While Harvard maintained its excellence by being extremely conservative when making a tenure decision, Minnesota maintained its excellence by making shrewd guesses about the future productivity of the young people on the faculty. I recall chatting with Meehl during my first year, telling him how pleased and surprised I was that they had taken a chance on me by offering me a tenured position when I was only two years out of graduate school. He said this was no gamble at all.
“Really?” I said. “Why?”
“Passion,” he said.
“Passion?”
“Past productivity is a good predictor of future productivity,” he said, “but the surefire predictor is passion. Anyone reading your work can easily see the joy you take in doing it.”
Meehl also saw no reason to follow any set rule in the timing of promotions. “When people are doing good research,” he said, “they should be promoted rapidly.” One year later he came bounding into my office with a big grin on his face and announced that the senior faculty had unanimously voted to promote me to full professor. This promotion came as a bolt from the blue. I had not asked for it, nor was it made in response to an offer from another university. They promoted me simply because they felt I deserved it. It took me many more years in academia to realize just how rare an attitude that was.
If the social and intellectual atmosphere in the department as a whole was warm and congenial, the atmosphere in our lab would have to be described as downright cozy. We all enjoyed working together, teaching together, collaborating, and just hanging out. Because I was finally working with a critical mass of like-minded experimental social psychologists, I was able to institute Tuesday-night research meetings for all of the social psychology faculty and graduate students. We would meet at one of our homes and, over beer and pretzels, serve as informal consultants on one another’s research ideas, experimental procedures, data analysis, and so on. We even organized a faculty-student softball team that was talented enough to challenge the fraternity jocks for the intramural championship. We never missed an opportunity for a party.
Vera and I treated the grad students like members of the family. The students got to know our kids, and hung out a lot at our house. At one of the Tuesday-night meetings, Josh, who was not quite three at the time, was playing around with some of the students. It was way past his bedtime, but he was having so much fun that he refused to heed Vera’s calls. Finally, Vera came into the living room and scooped him up. Trying to wriggle out of her grasp, he extended his arms toward Darwyn Linder and cried, “Darwee, help! Help me!” For months after that, the other students greeted Darwyn with Josh’s plea.
At the University of Minnesota I discovered that the architectural design of a building can have a profound effect on the productivity and creativity of the people in it. The layout of the lab was ideal. We occupied about half of the fourth floor of Ford Hall. The faculty offices and experimental rooms were clustered together, and all were adjacent to the most important place in the lab—a huge room at the end of the hall, where all of the social psychology graduate students had their desks. That room had three or four blackboards and two coffeepots, one of which was always kept brewing as the other was being emptied. We called that room the “bullpen” because it was the place to hang out and shoot the bull. Our conversations ranged from the trivial (gossip and sports) to the vital. We constantly threw out research ideas, and from them, experiments took shape.
One afternoon I came into the bullpen to find Gene Gerard and Darwyn Linder doubled up with laughter. Gene was reading passages aloud from a book by British civil servant C. Northcote Parkinson. It was Parkinson’s tongue-in-cheek thesis that, in a government bureaucracy, work will always expand to fill the time available, and he had hilarious examples to bolster his case.
After listening for a few minutes, I said, “You know, Gene, that’s not simply funny. It’s also interesting.”
“Uh-oh,” Darwyn said. “Be careful, Gene. I know that look. You are about to get roped into doing an experiment.”
“I’ll ignore that crack,” I said. “But it happens you’re right. Look, it’s true that, in a bureaucratic organization, people mark time a lot; they pretend to be working so that the work does expand to fill the time available. They are doing this because they are simply sitting around doing boring stuff and watching the clock. But what if a person is given a task to do—and because of an accidental occurrence, he has all day in which to do it. So he goes over it a few times, polishes it, hones it, and so on. Then, a short time later, he is given a similar task but is allowed to leave the office and go home as soon as he finishes it. I’m thinking that his first experience might lead him to define the task as one that requires a great deal of time—and therefore, he might spend more time on it than it deserves even though he is now wasting his own time and not the bureaucracy’s time.”
“You mean,” Gene said, “once the time expands, it stays expanded in that person’s mind?”
“You got it.”
And so Gene and I did the experiment; it was a simple, straightforward procedure, pretty much as we outlined it in the bullpen. While a subject was working on a tedious task, the experimenter (Gene) was abruptly called out of the room to take an urgent long-distance phone call. We simply varied the amount of time he was allegedly on the phone. In one condition, he was gone for far longer than the participant needed to finish the task. In the other condition, he was gone for a shorter time, but just adequate for the participant to finish comfortably. In the next phase, Gene gave the subject a similar task, told him that he was free to leave as soon as he finished, thanked him for his participation, and went back to his office. The subjects who had been in the “excess time” condition took far longer to complete the second task than those in the “adequate time” condition. We had gone beyond Parkinson’s law! We had shown not only that a task expands to fill the time available but also that the person doing it will then define the task as one that requires extra time.
The bullpen members loved this experiment, and we kept applying it to our own lives. For example, the first time I was invited to present a special address at an APA symposium, the invitation arrived in July, when I had plenty of leisure time; I spent five days preparing my presentation. A couple of years later I received a similar invitation, but this time I had other obligations and had only two days to prepare. At first I panicked and almost declined the invitation. After all, writing a speech takes five days! Then I realized, after some mental effort, that maybe it didn’t always take five days. I did it in two.
One of the fundamental questions in social psychology is: Why do people like or dislike each other? In the early 1960s virtually all of the answers that psychologists offered were inspired by a simplistic application of behaviorism: We like people who reward us in one way or another. We like those who are physically attractive because it is aesthetically rewarding to gaze upon beauty. We like people who are competent and smart because, if they are doing things for us, we can trust them to get the job done. We like people who share our beliefs and values and, most important, who seem to like us—who shower us with praise and express interest in us and our activities. Conversely, we certainly don’t like people who cause us pain, embarrassment, or discomfort.
These explanations were what any layperson would refer to as “just common sense.” As my grandmother would have said, “You had to get a whole entire Ph.D. to learn this? I could have told you that when you were still in diapers.” (In fact, my colleagues and I jokingly referred to this kind of research as “bubba psychology,” bubba being Yiddish for grandmother.) It’s not that the guiding principle is wrong; it’s that it is incomplete and grossly oversimplified. One of the reasons I was so excited about the initiation experiment was that no one’s grandmother could have predicted our results.
And so, when I decided to plunge further into the area of attraction, I wondered about other exceptions to behavioral predictions. For example, if the presence of a competent person is rewarding to us, then it follows that the more competent a person is, the more we should like him or her. But another person’s competence might present some cognitive complications that are unique to human beings. What if a person is too competent, coming across as unapproachable and superhuman? His very competence might make us uncomfortable in his presence. We might like that person more were he to show some evidence of fallibility. Hypothetically, if Herb is a brilliant mathematician as well as a great basketball player and a fastidious dresser, I might like him better if, every once in a while, he added a column of numbers incorrectly, blew an easy layup, or appeared in public with a gravy stain on his tie.
My musings were further reinforced (dare I say) by an incident that had taken place about a year earlier. In the spring of 1961 the American military believed that invading Cuba at the Bay of Pigs would lead to the overthrow of Fidel Castro. The invasion was a fiasco and a huge embarrassment to our nation and to President Kennedy in particular, as he had authorized it. Yet immediately after that fiasco, the Gallup Poll showed that Kennedy’s popularity had actually increased. How could this be? One reason might have been the desire to support our president in time of crisis, another the fact that Kennedy took full responsibility for the blunder. But a third possibility struck me: Kennedy was an extremely attractive man. He was young, handsome, smart, and witty. He was a war hero, had written a Pulitzer Prize-winning book, was the youngest president in our nation’s history, and had a beautiful, talented wife and two cute kids (one boy, one girl). In short, he seemed damned near perfect. Perhaps, I thought, committing that blunder increased his attractiveness by making him seem more human—more like the rest of us.
To test this idea I designed a simple experiment with Ben Willerman and Joanne Floyd. The participants were students at the university. We told them that we were auditioning candidates to represent Minnesota on The College Quiz Bowl, a popular show in which students from different universities competed against each other. We explained that we wanted them to help us select the team by giving us their frank impression of one of the candidates. Each student then listened to an audiotape in which Ben interviewed one of four stimulus persons, all played by a willing grad student: a nearly perfect candidate, a nearly perfect candidate who commits a clumsy blunder, a mediocre candidate, and a mediocre candidate who commits a clumsy blunder. On each tape Ben asked the candidate a set of extremely difficult questions like those on College Bowl. The grad student, using the same tone of voice, played each stimulus person, differing only in his answers.
Thus, the “nearly perfect” candidate answered 92 percent of the questions correctly, and when Ben asked him about his activities in high school, he modestly admitted he had been an honor student, the editor of the yearbook, and a member of the track team. The mediocre candidate answered only 30 percent of the questions correctly, and during the interview revealed that he had received average grades in high school, had been a proofreader on the yearbook staff, and had tried out for the track team but had failed to make it. On the other two recordings, the stimulus person committed an embarrassing pratfall, clumsily spilling a cup of coffee all over himself. The audiotape included sounds of commotion and clatter and the anguished voice of the candidate saying, “Oh, my goodness, I’ve spilled coffee all over my new suit.”
And in fact our participants rated the nearly perfect person who committed a blunder as the most likable and the mediocre person who committed the same blunder as least likable. (The blunderless perfect person came second in attractiveness, and the blunderless mediocre person finished third.) Clearly, there was nothing inherently attractive about the simple act of spilling a cup of coffee. The same action that added an endearing dimension to the perfect person, making him more likable, made the mediocre person appear that much more mediocre and, hence, less likable.
Our findings became widely known among social psychologists as “the pratfall effect.” And, from that moment on, around the lab, every time one of our graduate students screwed up in some way, he or she would claim, “I did it on purpose, so that you would like me more!” To which I would reply, “All well and good, but before you blunder, you had better be sure that you are nearly perfect to begin with.”
In those days the most famous popularizer of behaviorism was the supersalesman Dale Carnegie. His book How to Win Friends and Influence People, written in 1936 and translated into thirty-one languages, remains one of the most widely read self-help books of all time. If you want to get people to like you, to buy your product, or to change their behavior in any way, Carnegie advised, you should “dole out praise lavishly,” that is, reward people by acting as though you sincerely like and appreciate them and are interested in the things they are interested in. The more attractive you seem to find a person and the more attractive your own characteristics, the more that person will like you.
Carnegie’s salesmanship ran contrary to my own experience. If a person I hardly knew showered me with praise, I would suspect an ulterior motive: Is he trying to sell me something? But even if he or she doled out praise in a manner that was above suspicion, it might not be as valuable to me as someone who showed a bit more discernment in doling. Imagine that you’re a young man at a cocktail party, and you meet a woman, Peggy, for the first time and have a pleasant conversation with her. After a while you excuse yourself to refill your glass. As you are returning, Peggy has her back to you and is engrossed in conversation with another person. You hear her talking about you, so, naturally, you pause to listen. She has no ulterior motive in what she is saying; she doesn’t even know you are eavesdropping. Thus, if she tells her friend that she was impressed by you, that she likes you, that she finds you bright, witty, charming, gracious, honest, and exciting, you will probably like her. What a smart, perceptive person she is! If you hear her tell her friend that she dislikes you, that she finds you dull, boring, dishonest, stupid, and vulgar, you will probably like her less. What an insensitive clod she is, to draw conclusions about you after one five-minute chat!
So far, nothing unusual. Now, however, imagine that you attend seven cocktail parties on seven consecutive nights, and, miraculously, the same thing happens. You chat with Peggy for several minutes, you leave, and when you return, you overhear her talking about you. Her comments might be uniformly positive, uniformly negative, or varied—they might start negative and become positive, or start positive and go south. Which of these four possibilities would lead you to like Peggy most? Least?
According to reward-reinforcement theorists (and Dale Carnegie), you should like Peggy the most in the first situation, where she is doling out praise lavishly, and you should like her least in the second situation, in which she is saying only negative things about you. Because positive statements are rewarding, the more of them the better, and because negative statements are punishing, the more of them the worse. But I had a different idea. Because human beings are cognitive animals who are forever trying to figure things out, we pay close attention to changes in a person’s feelings toward us. Thus, if Peggy begins by disliking me, but the more she gets to know me the more her liking increases, this will have greater positive impact on my feelings about her than if she had been uniformly positive—because it constitutes a gain. In contrast, if she liked me at first but the more she got to know me the less she liked me, this would be painful because it constitutes a loss—and therefore I would like her least of all. I called this the gain-loss theory of attraction.
In physics, Albert Einstein constructed some brilliant thought experiments that were definitive. Alas, in social psychology, our thought experiments are never definitive. However, they can be good starting points; then we have to bring them into the laboratory. But how could we invent a credible scenario that captured the essence of the seven cocktail parties, given that the whole shebang had to unfold in less than an hour? We needed to invent a situation in which the subject interacts with a preprogrammed confederate, eavesdrops while the confederate evaluates him or her to a third party, has another conversation with the confederate, eavesdrops again, converses again, eavesdrops again, and so on through seven pairs of trials. What sensible cover story could we concoct? But Darwyn Linder and I came up with one.
When the subject (a female college student) arrived at the lab, Darwyn greeted her and led her to an observation room connected to the main experimental room by a one-way window and an audio-amplification system. Darwyn told her that two women were scheduled for that hour: One would be the subject, and the other would help perform the experiment, and because she had arrived first, she would be the helper. Darwyn asked her to wait while he left the room to see if the other woman had arrived. A few minutes later, through the one-way window, the subject was able to see Darwyn enter the experimental room with another female student (Darcy Oman, an undergraduate, who served as our confederate). Darwyn asked Darcy to be seated and said that he would return shortly to explain the experiment to her. He then reentered the observation room and began the instructions to the real subject, who believed herself to be the helpful confederate.
Darwyn told her she was going to assist him in performing a verbal conditioning experiment on the other student (Darcy); that is, he was going to reward Darcy every time she used a plural noun in conversation. He told the subject these rewards would increase the frequency with which Darcy would say plural nouns, but his interest lay in whether the use of plural nouns generalizes to a new situation. That is, will people continue to use more plural nouns when they are talking to a different person, someone who has not rewarded them for doing it? Darwyn explained that he would try to condition Darcy by subtly rewarding her with a nod, a smile, and an “mmmm hmmm” every time she said a plural noun. “Will she continue to use an abundance of plural nouns when she talks to you,” he asked the subject, “even though you will not be rewarding her?” He added that our volunteer’s tasks were to listen in and record the number of plural nouns Darcy used while talking to Darwyn and then to engage Darcy in a series of conversations in which the use of plural nouns would not be rewarded—so that Darwyn could listen and determine whether generalization had taken place. Darwyn then told the subject they would alternate in talking to Darcy until each had spent seven sessions with her.
Darwyn made it clear to the subject that Darcy must not know the purpose of the experiment, lest the results be contaminated. He explained that he would tell Darcy that the experiment was about attraction, and ask her to carry on a series of seven short conversations with the subject. Between each of these conversations, both she and the subject would be interviewed—Darcy by Darwyn and the subject by an assistant in another room—to find out what impressions they had formed. During the seven meetings Darwyn had with Darcy, the subject was in the observation room, listening to the conversation and dutifully counting the number of plural nouns used by Darcy. Because she had been led to believe that Darcy thought that the experiment had to do with people’s impressions of one another, it was understandable for Darwyn to ask Darcy to express her feelings about the subject. Thus, the subject heard herself being evaluated by a fellow student on seven successive occasions. This setup is more complicated in the telling than the doing. To the participants who were embedded in the experiment, it was completely believable: Only four of the eighty-four subjects were suspicious of our procedure.
There were four major experimental conditions: (1) positive—the successive evaluations of the subject made by the confederate were all highly positive; (2) negative—the successive evaluations were all highly negative; (3) gain—the first few evaluations were negative, but they gradually became more positive; and (4) loss—the first few evaluations were positive, but they gradually became negative. The results confirmed our predictions: The subjects in the gain condition liked Darcy significantly better than those in the positive condition. (So much for Dale Carnegie’s advice to dole out praise lavishly.) Similarly, the subjects in the loss condition disliked Darcy more than those in the negative condition.
Just as the layout of the lab and the bullpen facilitated creative conversations and generated friendships among grad students and faculty, the layout of the Minnesota faculty club did the same for instructors at all levels, from new assistant professors to the most eminent. It was another stark contrast with Harvard, whose faculty club was elegant: starched tablecloths, excellent food, and waiters who served us. Yet I ate there just a few times and only when invited by a senior professor. The faculty club at Minnesota was informal, with mediocre food served cafeteria style, on plastic trays. Most of us in the Psychology Department ate lunch there almost every day, and what our lunches lacked in ambience they gained in the fun and intellectual stimulation of our conversations. We brought our trays to a large round table at the far end of the dining hall, which was known as “the psychology corner.” But it was hardly exclusive; we were often joined by colleagues from other departments, such as philosopher of science Herbert Feigl and novelist Jack Ludwig.
It was at these lunches that I got to know Gardner Lindzey, who was to become one of the closest friends I have ever had. At first I was in awe of him; he was twelve years my senior and had written or edited two of the most important books in social psychology. Gardner was the star of the round table, an entertaining storyteller and a catalyst of conversation. And he seemed to know everything about everyone’s research. At one moment he would be asking Kenneth MacCorquodale detailed questions about some of the animal learning experiments he was doing, and the next he would be probing David Lykken to tell more about the lie-detection experiments that he had begun. But he would always find a way to make each person’s research relevant to the interests of the other people at the table. He knew everything I had ever published, asking detailed questions about how it was that I had decided on this or that specific procedure.
Gardner was a broadly educated scholar who was not only a fountainhead of information; he was also an accomplished gossip. He knew everything that was going on in the profession (say, who was leaving which university to teach somewhere else, who was promoted, who had been denied tenure) and also who was about to get a divorce and who was sleeping with whom. When I asked him one day why we hadn’t seen Jack Ludwig at the table for a while, he looked at me in astonishment. “Haven’t you read Herzog yet?” he said. “It’s been out for two months! You must read it immediately. It’s about people you know.” Herzog, of course, was written by Saul Bellow, who, a few years earlier, had been a visiting professor at Minnesota. None of us (except Gardner, apparently) knew that during that time, Bellow’s wife had had an affair with Jack, who had been one of Bellow’s close friends. Bellow was furious, and although he and his wife had gone to a psychotherapist for counseling, it didn’t help and they divorced. Bellow got his revenge the writer’s way, by writing a novel in which he barely disguised the characters. He is Herzog, and Jack is Valentine Gersbach, who has an affair with Herzog’s wife. (Jack had a club-foot; Valentine has a wooden leg.) Herzog goes to a psychiatrist, Dr. Edvig, who seems so matter-of-fact about his patient’s situation that Herzog accuses Dr. Edvig of being in love with his wife himself. The psychotherapist Bellow had gone to see, Gardner told me, was . . . Paul Meehl. Later I asked Paul about the way Bellow had characterized him. “Oh, these novelists,” he said.
After a few of these stories, I started to tease Gardner by calling him the Hedda Hopper of American psychology. Paul chastised me for that. “You are underestimating our friend,” he said. “Gardner’s gossip is not limited to the U.S. It is international in scope.”
One morning Gardner phoned and asked me if I wanted to join him for a “businessman’s lunch.”
“What’s that?” I asked.
“There are some great restaurants in downtown Minneapolis. The faculty club is pleasant, but you don’t want to ruin your palate by eating there all the time, do you?”
“Certainly not,” I replied.
“Be outside your building at 11:30. I’ll swing by and pick you up.”
In that first lunch Gardner and I reminisced about our days at Harvard. He had gotten his doctorate with Gordon Allport in the late 1940s and had taught there for a few years thereafter. He asked me who my favorite Harvard people were. I named a few.
“You were at 9 Bow Street, weren’t you?” he asked.
“Yes, I was.”
“That’s funny. I notice that you didn’t mention Jerry Bruner as one of your favorite people.”
“To tell you the truth, Jerry and I didn’t get along very well.”
“I hated the bastard too,” said Gardner gleefully.
“I didn’t say that.”
“I heard what you said. And, like I said, I hated him too.”
There are many reasons that Gardner and I became close friends, but that got us off to a flying start.
This was the beginning of a number of businessman’s lunches Gardner and I had, usually once or twice a week, occasionally with David Lykken or Paul Meehl. I quickly learned that Gardner’s major criterion for a good restaurant was not the food but the quality of the bar. Our lunches lasted for well over two hours and were always preceded by two or three martinis and followed by a cognac. Amazingly, each of us was able to work hard and well in the afternoons. Since our friendship began while I was in my heavy-drinking period, Gardner’s style fitted my needs at the time, but the lunches and the friendship continued even after I cut back to one martini. As for Gardner, well, I never knew anyone who enjoyed drinking more and who could hold his liquor so well and with such grace. He was capable of abstaining for weeks at a time without a craving, and in a close friendship of almost fifty years, I never saw him out of control.
For the next few years Gardner and I had many provocative conversations and strenuous arguments about research and theory in social psychology. Once he proclaimed that Solomon Asch’s experiment on conformity was the single most important experiment in the field. I said, “Nonsense! There was no theory behind it, no control group—it was just a demonstration that people conform more than we imagined. Now, the Festinger/Carlsmith one-dollar to twenty-dollar experiment was really important—it had theory; it was counterintuitive; it changed the field! Who would have guessed that people who told a lie for one dollar rather than twenty dollars would come to believe that lie?”
After one of these exchanges Gardner invited me to coedit the second edition of the Handbook of Social Psychology. Because I had always considered the Handbook to be the bible of our field, this offer was like, well, being asked to edit the next edition of the Bible. The Handbook contained up-to-date thinking, research, and theorizing by the world’s most eminent social psychologists. I agreed, figuring that I would learn a great deal from working with Gardner (as indeed I did).
A few weeks later, while I was still in a glow over the honor of being an editor, Gardner announced that he expected me to contribute a chapter as well. This time I demurred. “There are a few areas where I could do a decent job of writing a chapter,” I said. “But as an editor I would want to recruit the best scholar working on a topic, and the best scholar ain’t me.”
“How about writing a chapter on experimental methodology?” he asked.
“I’m no methodologist,” I said.
“But you did teach a seminar on that at Harvard.”
“Well, yes, but that was mostly a hands-on course, where I helped students find the best way to test a given hypothesis.”
“Precisely the kind of thing that would be great for the Handbook ,” said Gardner. “All you need to do is take a good hard look at your own experiments and try to make explicit the wealth of tacit knowledge that you have, so that others can understand why you did what you did.”
“But teaching people how to do experiments is in the doing, not in the reading,” I protested. “It is part of a master-apprentice relationship. You learn to be a great chef by working with one, not by reading a cookbook. This may even be truer in social psychology than in cooking. We don’t have standardized procedures because, most of the time, each new hypothesis requires the invention of a new procedure tailored to test it. Leon had to invent the ‘one-dollar, twenty-dollar’ procedure, Jud and I had to invent the initiation procedure, and so on. We are always flying by the seat of our pants. It’s an exciting way to work, but it makes it impossible for anyone to write a chapter called ‘How to Do Experiments in Social Psychology’ that will be of any use to the student who is just starting out.”
“Don’t be so damn rigid,” said Gardner. “Nobody has ever tried to do it before. This could be a unique contribution. Think of it not as something that would replace working with a master experimentalist but as a set of ideas that might serve as an adjunct to the apprenticeship. Why not give it a try? After you write the chapter, if you don’t like the way it turns out, as editor, you can simply reject it.”
Although I was not convinced, I decided to plunge in—but not by myself. I recruited my favorite former graduate student, Merrill Carlsmith, who was more excited about the challenge than I was. As Gardner had suggested, we started by looking at our own experiments, to articulate the thinking that went into the design and implementation of each one. We were trying to write a cookbook without recipes. It wasn’t easy.
We began by tackling the central criticism directed against experimental social psychology at that time: The mere fact that experiments were taking place in a laboratory setting made them “unrealistic.” Merrill and I asked what proved to be an important question—what does “realism” mean? And the answer led us to clarify the difference between what we called “mundane” realism from what we called “experimental” realism. For example, several critics had claimed that Stanley Milgram’s obedience experiment was unrealistic. After all, how often in our day-to-day lives are we asked to inflict a series of electric shocks on an innocent stranger? But that criticism missed the point: Milgram’s study was unrealistic only in the mundane sense. In terms of experimental realism, the procedure was real as hell to the people participating. They sweated real sweat. Merrill and I concluded that in some situations, mundane realism is necessary; if you ask people to read a communication that you planted in a dummy newspaper, the newspaper had better look real. Usually, however, mundane realism is of no importance as long as what is happening in the laboratory has real consequences for the subjects.
Merrill and I then tackled the nuts and bolts of constructing high-impact experiments. We spelled out the experimenter’s most important challenge: How to take a compelling conceptual variable, the thing you want to understand, and translate it into a concrete set of empirical steps. The trick is to achieve a high degree of scientific rigor without losing, in the translation, the essential, complex beauty of the idea. This is harder than it sounds; many a grand, important topic gets boiled down into something trivial or simplistic by the time it has been operationalized into a precise sequence of experimental events.
The next problem is to determine how to measure the effect of your intervention. In social psychology the most frequently used measure is a rating scale of some sort, usually in the context of an interview. For example, in the initiation experiment, we asked subjects to rate how much they liked the group they had joined; in the one-dollar or twenty-dollar experiment, Merrill asked subjects to rate how much they had enjoyed the boring task of packing spools. Such measures are the easiest to administer and are often the only feasible way you can get the information you want. The experimenter must not forget, however, that a rating scale is only a stand-in for behavior. Whenever possible, it is preferable to observe what people actually do rather than to ask them what they think or how they feel. In the forbidden-toy experiment, if we want to determine whether we have succeeded in reducing a toy’s appeal, it is better to watch and see whether the kids resist the temptation to play with the forbidden toy than to ask them to tell us how much they like the toy. That’s why Jonathan Freedman’s replication of our experiment was an improvement on the original. He used the behavioral measure; he observed what the children did.
Writing this chapter turned Merrill and me into better experimenters, because it made us confront our shortcomings as well as our skills in the laboratory. Moreover, it forced us to think creatively about solutions to the problems we had posed. Getting a behavioral measure is ideal, but what if doing so is impossible, either ethically or practically? Is there a better alternative than going back to the familiar rating scale? Merrill and I invented a compromise that we dubbed “the behavioroid” measure. Suppose you want to know if watching a poignant film depicting hunger in urban America will make people more altruistic. Showing the film is easy, but you can’t very easily follow the audience around for the next few weeks to see if they become more helpful to others. However, you can ask the subjects whether they are willing to sign an official-looking contract committing themselves to a specific altruistic action, such as volunteering to work in a soup kitchen. Of course, getting a person’s “behavioroid” commitment is not quite the same as seeing whether he or she actually turns up for duty in a soup kitchen. But it is far preferable to simply asking people to rate their feelings about the poor or about the value of soup kitchens.
Finally, we gave considerable attention to the postexperimental interview—the debriefing of the subjects. Because so many experiments in social psychology require deception, Merrill and I argued that it is of the utmost importance for experimenters to disclose everything that took place and to explain why we had done what we had done. Full disclosure is a vital part of our implicit contract with our subjects. Because no one enjoys being told that he has been duped, the disclosure must be done with tact and sensitivity, or the subject will feel that he or she has been a gullible fool. The experimenter can explain that because the cover story was so carefully crafted, just about everyone accepted it; far from being “gullible,” the subjects behaved perfectly normally and reasonably. And then we explain the goal of the research—why deception was necessary and what we hope to have learned about human nature as a result of our methods. We make sure that each and every one of our subjects leaves the experimental room with her self-esteem intact, in at least as good a shape as she was when she first came in, with the bonus of having learned something interesting. I often spent as much time chatting with my subjects after the experiment as I did during the experiment itself.
When we were colleagues at Harvard, my old mentor David McClelland once told me that he had run only one social psychology experiment in his whole life, when he was a grad student at Yale. He was observing a young man through the one-way mirror when the guy suddenly stood up, unzipped his fly, and began to examine his penis in the mirror, apparently looking for a rash or abrasion. I asked Dave how he dealt with it in the debriefing.
“I didn’t debrief him,” he said. “It would have been too embarrassing. For both of us.”
I clucked my tongue. “Dave,” I said, “in this business, you always debrief. There are no exceptions. If the subject were ever to learn, perhaps from a friend who was in the experiment, that the mirror was one-way, he would have been mortified. It is always up to the experimenter to find a gentle, perhaps in this case even a humorous, way of telling him what happened.”
And so we finished our chapter, and I, in my role as coeditor of the handbook, accepted it. Much to our own surprise, Merrill and I had done just what Gardner had hoped we would. By forcing ourselves to make our tacit knowledge explicit, we produced a document that graduate students in social psychology have been relying on for almost five decades. That methodology chapter of the Handbook underwent its most recent revision in 2010. My newest coauthor is Kevin Carlsmith, Merrill’s son, who was a toddler when Merrill and I were writing the original. I admit I enjoy the irony. A project that I had initially rejected, thinking it impossible, became one of my most enduring contributions to the field. As my grandmother would have said, “Go figure!”
In 1964 Gardner left the University of Minnesota to become chair of the Psychology Department at the University of Texas at Austin, with the mandate to make it into one of the strongest departments in the country. A year later he urged me to join him and become head of the social psychology program. I had been at Minnesota only three and a half years and was reluctant to leave. True, all of the graduate students I had supervised now had good job offers, so I had no ongoing mentoring obligations.
I talked it over with Vera, who said she could go either way. On the one hand, she had developed strong friendships in Minneapolis and was hesitant to pull up stakes yet again. On the other hand, as a young mother with four small children, she found the severity and length of the winters particularly difficult. The kids would look through the windows at a bright and sunny winter day and scream to go outside to play. Then, after ten minutes, frozen to the bone, they would scream to be let back in. (Imagine the time it takes to get one little kid into and out of a snowsuit, hat, boots, and mittens and multiply that times four, and you have some idea of what Vera was up against.) And so she left the decision to me.
A reasonable person would have stayed put. I loved everything about the Psychology Department, and I had no desire to move at all; even if I had, Texas would not have been near the top of my list. I’m embarrassed to confess that I held a negative stereotype about the state, and my impression of Texas as a raw and wild place had been exacerbated by the 1963 assassination of President Kennedy in Dallas. But, in the end, my friendship with Gardner trumped both my rational judgment and my prejudice. We moved to Austin.
在线 阅读网：hTTp://wwW.yuedu88.com/
“I wanted to be the best husband and father I could be.”
Jason and I are in a huge, cavernous railroad station, something like Grand Central Terminal. We know that the train is about to leave, but we don’t have tickets and we don’t know which of the multilevel tracks our train will be departing from. We are racing around but cannot find the ticket window. Finally, we spot one in the far end of the station. “Wait here,” Jason says. “I’ll take care of it.” Wallet in hand, he dashes to the ticket window. When he returns, he is gasping for breath; he hands me a ticket and, pointing, says, “You’d better hurry; the train is about to leave on that track over there.”
“Aren’t you coming?”
“I’m afraid I can’t join you. You’ll have to go it alone from here on. Don’t worry, and don’t look back. You will be fine.”
Somehow, I am not surprised to hear him say that. And, in the dream, I am surprised at not being surprised. We hug, I take a deep breath, and I hop on the train just as it is beginning to move. I look back; Jason waves to me in a sad salute.
No need for Sigmund Freud to interpret this one. When I was seventeen years old, my brother, in defiance of our uncles, guided me to the right track, put me on a fast train, and, then, when my own family and career were secure, he died. He remained behind, forever at the station.
I missed Jason terribly. Next to Vera, he had been my closest friend and confidant. Once the initial shock of his death began to subside, I found myself turning inward, focusing on my own mortality. It was suddenly obvious to me that the Aronson family was cursed with a genetic defect that made us more cancer prone than the average family. Not only had that dreaded disease struck down my father and brother in their prime, but it had also killed two of my father’s siblings in their thirties. My first reaction, worried about financial security for Vera and the kids, was to take out a huge life insurance policy. If I was going to die young, I would not leave my wife and kids destitute. The Prudential Insurance Company was only too happy to sell a large policy to a twenty-nine-year-old professor in perfect health.
I began joking to my friends and colleagues about “shortgevity” running in my family. I believe I did that, ad nauseam, as a kind of memento mori; that lighthearted statement did for me what the skull on the desk did for a medieval scholar—serve as a reminder that life is too short and unpredictable to waste much time on unimportant or uninteresting things. So, I told myself, I had damn well better make use of whatever time I had left. I said to myself, “Okay, schmuck, stop wallowing in grief and self-pity. Let’s suppose, seriously, that you have only four or five years left to live; how do you want to spend your time?”
I wanted to spend more time with Vera and the kids. More than spending time, I wanted to be the best husband and father I could be; Hal, Neal, and Julie were now six, five, and three, and Joshua was an easy, happy baby. I also wanted to be the best teacher, the best researcher, and the best colleague I could possibly be. I always had those ambitions, but now my focus was sharper and more immediate. I figured out ways to spend more time with my family without taking anything away from the work I loved doing. I concentrated on being fully present, actively making plans with and for Vera and the kids.
I was amazed to learn how much time there is for everything if one doesn’t waste a good deal of it on trivia. I worked hard and well at the university, but almost all evenings and weekend days became sacrosanct family time. I refused to let course preparation, the writing up of research articles, colloquium invitations, or other business intrude on Saturdays and Sundays during the day. When I was home, I was really there. Although I fiercely guarded weekend days as a time to enjoy the children, weekend evenings were something else. Vera and I frequently hired babysitters so that after we tucked the kids in, we could step out, going to movies, theaters, or restaurants, or to one of the faculty parties that my sociable colleagues were forever throwing. Vera and I made a conscious decision to be more than the parents of four children. It was important to us to continue to think of ourselves as young lovers. On weekend nights when we did stay home, we conspired to keep our romance alive. After putting the kids to bed, we prepared simple but elegant late-night dinners with candlelight and wine.
Not all the changes I went through as a result of Jason’s death were positive and rational. I began drinking far more heavily. In the past I had enjoyed nightcaps with Vera as well as a martini or two with my graduate students whenever there was something to celebrate. But in Minnesota my drinking became more intense and more frequent. At parties I often got drunk. Vera and my friends told me that I was an affectionate drunk, not a belligerent one, but the fact is that they (usually Vera) had to tell me because, the morning after, my memory of my behavior was cloudy at best.
Vera also kept cautioning me that I was driving too fast and taking crazy risks. As I look back, I see what an irrational thing that was for me to do, given how hell-bent I was, at least consciously, on being present (and alive!) for my family for as long as possible. Many years later I learned that the three most common manifestations of grief among men are heavy drinking, reckless driving, . . . and singing sad, sentimental songs in the shower. Bingo: I was three for three. Fortunately, the intensity of my grief and my irrational behavior subsided within two years and did not result in any wrecked cars or strained friendships. Yet my dream of Jason at the train station has recurred all my life, serving as a reminder to keep things in focus and in perspective.
When Vera and I first arrived in Minneapolis, we rented a house some distance from the campus. The following September, when our lease was about to expire, we decided to buy a home of our own, though we didn’t have much money. After searching for a few months, we found only two houses that were large enough and in our price range that we both liked. They could not have been more different. One was a charming old Victorian, within walking distance of the campus. I liked it a lot, primarily because it meant that my colleagues and graduate students could casually drop in on us to discuss research or just schmooze. But that house was adjacent to an industrial area, without much space for our children to play. The other choice was a tract house in the suburbs, newer but without distinction, a thirty-minute drive from campus. It had two advantages: It had a large yard and was only a mile from Lake Owasso. Eventually, that was the house we chose, because we agreed it was better for the kids.
Not long after we moved in, I saw an ad for a used canoe and immediately bought it as a surprise for Vera and the children. When I arrived home on that freezing, bleak December day with the canoe lashed to the roof of my car, Vera burst into laughter.
“What’s so funny?” I asked.
“Ask Leon!” she said.
Of course! I had felt so much dissonance about buying the house in the suburbs that I needed to do something right away to justify that purchase. I somehow managed to ignore the obvious fact that it was the beginning of the long winter season in Minneapolis and would be another five months before the frozen lake would thaw out enough for the canoe to be usable. During the winter Minnesotans frequently take shortcuts by driving across lakes; they joke that the unofficial arrival of spring is the first car that falls through the ice. So, for five months, my canoe sat lazily in our garage. But it did more than take up space; it helped me come to terms with giving up my dream of living close to campus.
When I first moved into my office at the Laboratory for Research in Social Relations, I found, to my astonishment, that the lab was not the beehive of activity that it had been when I visited the previous spring. Indeed, things looked bleak. For one thing, the secretary, the very person whom Stan Schachter had assured me would know all the ropes and would relieve me of the onerous administrative work, was seven months pregnant and had one foot out the door. In addition, there were only a few graduate students. When Hal Kelley and Stan Schachter moved on, they had taken their favorite students with them, and those who remained were busily working on their dissertations or already committed to working with either Dana Bramel or Ben Willerman, my two colleagues at the lab. What to do? I needed a research assistant, and no one was available. Dana and Ben informed me that there was a wealth of talent in the Twin Cities and suggested that I advertise in the Minneapolis Tribune.
It seemed a strange way to hire a research assistant, but it turned out to be a great idea. I received a flood of responses and interviewed the eleven who seemed most qualified. Most of them could have done the job, but one interviewee struck me as being uncommonly bright. Her name was Ellen Berscheid. Ellen had earned a master’s degree from the University of Nevada and was working for Pillsbury Mills in marketing, “exploring,” as she put it, “ways to persuade housewives to buy cake mix.” Her behavior during the interview puzzled me. She told me she really wanted this job, yet she was not going out of her way to make a good impression. Far from it. From the moment she walked in the door, her demeanor was feisty and contentious; she seemed primed to argue with me about ideas great and small. I had the feeling that if I had said, “It’s a sunny day” (which it was), she would have replied sharply, “Nonsense. It is about to snow.”
But she was clearly the most qualified candidate, so I offered her the job. After a few months I was so impressed with her that I urged her to take some graduate courses and work toward a Ph.D., which she did, finishing in record time. Thirty-five years later, when Ellen was given the APA’s highest honor, the Distinguished Scientific Contribution Award, she wrote a brief professional memoir. On reading it, I was charmed to learn that she regarded our interview as having been a major turning point in her life. Because she had suffered from the pervasive sexist attitudes and treatment in her previous job, she wrote, she had walked into my office expecting more of the same. She was surprised and delighted that I had selected her on the basis of her qualifications alone, in spite of what she referred to as her “wrong gender and undeniably bad attitude.”
The new students were a smart and lively group. Dana, Ben, and I had recruited and selected them carefully, and by October the lab was humming with exciting research. The new graduate students included a brilliant Jesuit priest (Eugene Gerard) and a gifted mountain climber (Darwyn Linder). John Darley, having completed his course work at Harvard, now joined me at Minnesota as my majordomo and role model for the newer students. On a postdoctoral level, our greatest acquisition came out of the blue. One day I received a call from Leon Festinger, who told me that he had a first-rate new Ph.D., Elaine Hatfield Walster, who was having trouble getting an academic position, given the bias against hiring women at most universities at that time. Leon asked me if there was any way we could find a job for her at Minnesota.
“How good is she?” I asked.
“You know Stan Schachter?”
“Of course,” I said.
“You know Elliot Aronson?”
“A little bit,” I said.
“Well,” Leon said, “she is in that class.”
There was, however, no position open in our department for another social psychologist, so I thought that was the end of it. But a few weeks later, when I learned that the Student Activities Bureau was advertising for a research specialist to arrange freshmen orientation dances, I called the dean of students.
“This is a golden opportunity for you and for us,” I said. “Why hire a hack researcher to collect data on who goes to the dances and whether or not they had a good time? Why not hire a skillful experimental social psychologist who will make something interesting out of this opportunity and who will do the job right—maybe even discover something of scientific importance? If you hire the right person, we in the lab will work with him or her as well. You get a capable scientist with built-in consultants, and we get a good colleague.”
“Terrific idea,” the dean said. “Do you have anyone in mind?”
“It just so happens . . . ,” I said, and so Elaine came to Minnesota. She in fact did use this assignment as a vehicle for a groundbreaking study on dating preferences among college students, which was published in our leading journal. We gave her an office in the lab, where, just as I had hoped, she spent most of her time working and interacting with the grad students, who treated her as if she were a regular faculty member. She proved to be an excellent, intuitive researcher and became one of my all-time favorite colleagues. Because she had no teaching responsibilities, for Elaine the job was like having a high-paid postdoctoral fellowship. After two years she was offered and accepted an official position as an assistant professor in the Psychology Department. Eventually, Elaine and Ellen Berscheid teamed up to form what proved to be a highly productive research partnership that lasted for more than two decades.
The weather in Minnesota was frigid, but the social climate was warm and could not have been more different from the one I experienced at Harvard. There were no barriers between faculty members as a function of rank or tenure. Newly hatched assistant professors joked freely with their most distinguished colleagues, who included Starke Hathaway (the inventor of the Minnesota Multiphasic Personality Inventory, the single most important personality test of that era and since), David Lykken (a national expert on the physiology of lie detection), Norman Garmezy (a clinical psychologist ahead of his time in studying children’s resilience), and Gardner Lindzey (editor of the monumental two-volume Handbook of Social Psychology). There wasn’t any antagonism between the clinicians and the experimental psychologists, as was true in most departments at that time, with each side feeling that they were doing the important work—the clinicians in the study of real people, the experimentalists in the rigor of their science. I am convinced that the primary force underlying this atmosphere of respect was Paul Meehl. Meehl was extraordinarily smart, an excellent experimentalist, a masterful psychotherapist, a brilliant philosopher of science, and a first-rate psychometrician. When the acknowledged superstar of the department bridges all subdisciplines, how can any of the lesser stars quarrel with each other?
I believe that Meehl’s stature and ideas contributed to one other major difference between Harvard and Minnesota: While Harvard maintained its excellence by being extremely conservative when making a tenure decision, Minnesota maintained its excellence by making shrewd guesses about the future productivity of the young people on the faculty. I recall chatting with Meehl during my first year, telling him how pleased and surprised I was that they had taken a chance on me by offering me a tenured position when I was only two years out of graduate school. He said this was no gamble at all.
“Really?” I said. “Why?”
“Passion,” he said.
“Passion?”
“Past productivity is a good predictor of future productivity,” he said, “but the surefire predictor is passion. Anyone reading your work can easily see the joy you take in doing it.”
Meehl also saw no reason to follow any set rule in the timing of promotions. “When people are doing good research,” he said, “they should be promoted rapidly.” One year later he came bounding into my office with a big grin on his face and announced that the senior faculty had unanimously voted to promote me to full professor. This promotion came as a bolt from the blue. I had not asked for it, nor was it made in response to an offer from another university. They promoted me simply because they felt I deserved it. It took me many more years in academia to realize just how rare an attitude that was.
If the social and intellectual atmosphere in the department as a whole was warm and congenial, the atmosphere in our lab would have to be described as downright cozy. We all enjoyed working together, teaching together, collaborating, and just hanging out. Because I was finally working with a critical mass of like-minded experimental social psychologists, I was able to institute Tuesday-night research meetings for all of the social psychology faculty and graduate students. We would meet at one of our homes and, over beer and pretzels, serve as informal consultants on one another’s research ideas, experimental procedures, data analysis, and so on. We even organized a faculty-student softball team that was talented enough to challenge the fraternity jocks for the intramural championship. We never missed an opportunity for a party.
Vera and I treated the grad students like members of the family. The students got to know our kids, and hung out a lot at our house. At one of the Tuesday-night meetings, Josh, who was not quite three at the time, was playing around with some of the students. It was way past his bedtime, but he was having so much fun that he refused to heed Vera’s calls. Finally, Vera came into the living room and scooped him up. Trying to wriggle out of her grasp, he extended his arms toward Darwyn Linder and cried, “Darwee, help! Help me!” For months after that, the other students greeted Darwyn with Josh’s plea.
At the University of Minnesota I discovered that the architectural design of a building can have a profound effect on the productivity and creativity of the people in it. The layout of the lab was ideal. We occupied about half of the fourth floor of Ford Hall. The faculty offices and experimental rooms were clustered together, and all were adjacent to the most important place in the lab—a huge room at the end of the hall, where all of the social psychology graduate students had their desks. That room had three or four blackboards and two coffeepots, one of which was always kept brewing as the other was being emptied. We called that room the “bullpen” because it was the place to hang out and shoot the bull. Our conversations ranged from the trivial (gossip and sports) to the vital. We constantly threw out research ideas, and from them, experiments took shape.
One afternoon I came into the bullpen to find Gene Gerard and Darwyn Linder doubled up with laughter. Gene was reading passages aloud from a book by British civil servant C. Northcote Parkinson. It was Parkinson’s tongue-in-cheek thesis that, in a government bureaucracy, work will always expand to fill the time available, and he had hilarious examples to bolster his case.
After listening for a few minutes, I said, “You know, Gene, that’s not simply funny. It’s also interesting.”
“Uh-oh,” Darwyn said. “Be careful, Gene. I know that look. You are about to get roped into doing an experiment.”
“I’ll ignore that crack,” I said. “But it happens you’re right. Look, it’s true that, in a bureaucratic organization, people mark time a lot; they pretend to be working so that the work does expand to fill the time available. They are doing this because they are simply sitting around doing boring stuff and watching the clock. But what if a person is given a task to do—and because of an accidental occurrence, he has all day in which to do it. So he goes over it a few times, polishes it, hones it, and so on. Then, a short time later, he is given a similar task but is allowed to leave the office and go home as soon as he finishes it. I’m thinking that his first experience might lead him to define the task as one that requires a great deal of time—and therefore, he might spend more time on it than it deserves even though he is now wasting his own time and not the bureaucracy’s time.”
“You mean,” Gene said, “once the time expands, it stays expanded in that person’s mind?”
“You got it.”
And so Gene and I did the experiment; it was a simple, straightforward procedure, pretty much as we outlined it in the bullpen. While a subject was working on a tedious task, the experimenter (Gene) was abruptly called out of the room to take an urgent long-distance phone call. We simply varied the amount of time he was allegedly on the phone. In one condition, he was gone for far longer than the participant needed to finish the task. In the other condition, he was gone for a shorter time, but just adequate for the participant to finish comfortably. In the next phase, Gene gave the subject a similar task, told him that he was free to leave as soon as he finished, thanked him for his participation, and went back to his office. The subjects who had been in the “excess time” condition took far longer to complete the second task than those in the “adequate time” condition. We had gone beyond Parkinson’s law! We had shown not only that a task expands to fill the time available but also that the person doing it will then define the task as one that requires extra time.
The bullpen members loved this experiment, and we kept applying it to our own lives. For example, the first time I was invited to present a special address at an APA symposium, the invitation arrived in July, when I had plenty of leisure time; I spent five days preparing my presentation. A couple of years later I received a similar invitation, but this time I had other obligations and had only two days to prepare. At first I panicked and almost declined the invitation. After all, writing a speech takes five days! Then I realized, after some mental effort, that maybe it didn’t always take five days. I did it in two.
One of the fundamental questions in social psychology is: Why do people like or dislike each other? In the early 1960s virtually all of the answers that psychologists offered were inspired by a simplistic application of behaviorism: We like people who reward us in one way or another. We like those who are physically attractive because it is aesthetically rewarding to gaze upon beauty. We like people who are competent and smart because, if they are doing things for us, we can trust them to get the job done. We like people who share our beliefs and values and, most important, who seem to like us—who shower us with praise and express interest in us and our activities. Conversely, we certainly don’t like people who cause us pain, embarrassment, or discomfort.
These explanations were what any layperson would refer to as “just common sense.” As my grandmother would have said, “You had to get a whole entire Ph.D. to learn this? I could have told you that when you were still in diapers.” (In fact, my colleagues and I jokingly referred to this kind of research as “bubba psychology,” bubba being Yiddish for grandmother.) It’s not that the guiding principle is wrong; it’s that it is incomplete and grossly oversimplified. One of the reasons I was so excited about the initiation experiment was that no one’s grandmother could have predicted our results.
And so, when I decided to plunge further into the area of attraction, I wondered about other exceptions to behavioral predictions. For example, if the presence of a competent person is rewarding to us, then it follows that the more competent a person is, the more we should like him or her. But another person’s competence might present some cognitive complications that are unique to human beings. What if a person is too competent, coming across as unapproachable and superhuman? His very competence might make us uncomfortable in his presence. We might like that person more were he to show some evidence of fallibility. Hypothetically, if Herb is a brilliant mathematician as well as a great basketball player and a fastidious dresser, I might like him better if, every once in a while, he added a column of numbers incorrectly, blew an easy layup, or appeared in public with a gravy stain on his tie.
My musings were further reinforced (dare I say) by an incident that had taken place about a year earlier. In the spring of 1961 the American military believed that invading Cuba at the Bay of Pigs would lead to the overthrow of Fidel Castro. The invasion was a fiasco and a huge embarrassment to our nation and to President Kennedy in particular, as he had authorized it. Yet immediately after that fiasco, the Gallup Poll showed that Kennedy’s popularity had actually increased. How could this be? One reason might have been the desire to support our president in time of crisis, another the fact that Kennedy took full responsibility for the blunder. But a third possibility struck me: Kennedy was an extremely attractive man. He was young, handsome, smart, and witty. He was a war hero, had written a Pulitzer Prize-winning book, was the youngest president in our nation’s history, and had a beautiful, talented wife and two cute kids (one boy, one girl). In short, he seemed damned near perfect. Perhaps, I thought, committing that blunder increased his attractiveness by making him seem more human—more like the rest of us.
To test this idea I designed a simple experiment with Ben Willerman and Joanne Floyd. The participants were students at the university. We told them that we were auditioning candidates to represent Minnesota on The College Quiz Bowl, a popular show in which students from different universities competed against each other. We explained that we wanted them to help us select the team by giving us their frank impression of one of the candidates. Each student then listened to an audiotape in which Ben interviewed one of four stimulus persons, all played by a willing grad student: a nearly perfect candidate, a nearly perfect candidate who commits a clumsy blunder, a mediocre candidate, and a mediocre candidate who commits a clumsy blunder. On each tape Ben asked the candidate a set of extremely difficult questions like those on College Bowl. The grad student, using the same tone of voice, played each stimulus person, differing only in his answers.
Thus, the “nearly perfect” candidate answered 92 percent of the questions correctly, and when Ben asked him about his activities in high school, he modestly admitted he had been an honor student, the editor of the yearbook, and a member of the track team. The mediocre candidate answered only 30 percent of the questions correctly, and during the interview revealed that he had received average grades in high school, had been a proofreader on the yearbook staff, and had tried out for the track team but had failed to make it. On the other two recordings, the stimulus person committed an embarrassing pratfall, clumsily spilling a cup of coffee all over himself. The audiotape included sounds of commotion and clatter and the anguished voice of the candidate saying, “Oh, my goodness, I’ve spilled coffee all over my new suit.”
And in fact our participants rated the nearly perfect person who committed a blunder as the most likable and the mediocre person who committed the same blunder as least likable. (The blunderless perfect person came second in attractiveness, and the blunderless mediocre person finished third.) Clearly, there was nothing inherently attractive about the simple act of spilling a cup of coffee. The same action that added an endearing dimension to the perfect person, making him more likable, made the mediocre person appear that much more mediocre and, hence, less likable.
Our findings became widely known among social psychologists as “the pratfall effect.” And, from that moment on, around the lab, every time one of our graduate students screwed up in some way, he or she would claim, “I did it on purpose, so that you would like me more!” To which I would reply, “All well and good, but before you blunder, you had better be sure that you are nearly perfect to begin with.”
In those days the most famous popularizer of behaviorism was the supersalesman Dale Carnegie. His book How to Win Friends and Influence People, written in 1936 and translated into thirty-one languages, remains one of the most widely read self-help books of all time. If you want to get people to like you, to buy your product, or to change their behavior in any way, Carnegie advised, you should “dole out praise lavishly,” that is, reward people by acting as though you sincerely like and appreciate them and are interested in the things they are interested in. The more attractive you seem to find a person and the more attractive your own characteristics, the more that person will like you.
Carnegie’s salesmanship ran contrary to my own experience. If a person I hardly knew showered me with praise, I would suspect an ulterior motive: Is he trying to sell me something? But even if he or she doled out praise in a manner that was above suspicion, it might not be as valuable to me as someone who showed a bit more discernment in doling. Imagine that you’re a young man at a cocktail party, and you meet a woman, Peggy, for the first time and have a pleasant conversation with her. After a while you excuse yourself to refill your glass. As you are returning, Peggy has her back to you and is engrossed in conversation with another person. You hear her talking about you, so, naturally, you pause to listen. She has no ulterior motive in what she is saying; she doesn’t even know you are eavesdropping. Thus, if she tells her friend that she was impressed by you, that she likes you, that she finds you bright, witty, charming, gracious, honest, and exciting, you will probably like her. What a smart, perceptive person she is! If you hear her tell her friend that she dislikes you, that she finds you dull, boring, dishonest, stupid, and vulgar, you will probably like her less. What an insensitive clod she is, to draw conclusions about you after one five-minute chat!
So far, nothing unusual. Now, however, imagine that you attend seven cocktail parties on seven consecutive nights, and, miraculously, the same thing happens. You chat with Peggy for several minutes, you leave, and when you return, you overhear her talking about you. Her comments might be uniformly positive, uniformly negative, or varied—they might start negative and become positive, or start positive and go south. Which of these four possibilities would lead you to like Peggy most? Least?
According to reward-reinforcement theorists (and Dale Carnegie), you should like Peggy the most in the first situation, where she is doling out praise lavishly, and you should like her least in the second situation, in which she is saying only negative things about you. Because positive statements are rewarding, the more of them the better, and because negative statements are punishing, the more of them the worse. But I had a different idea. Because human beings are cognitive animals who are forever trying to figure things out, we pay close attention to changes in a person’s feelings toward us. Thus, if Peggy begins by disliking me, but the more she gets to know me the more her liking increases, this will have greater positive impact on my feelings about her than if she had been uniformly positive—because it constitutes a gain. In contrast, if she liked me at first but the more she got to know me the less she liked me, this would be painful because it constitutes a loss—and therefore I would like her least of all. I called this the gain-loss theory of attraction.
In physics, Albert Einstein constructed some brilliant thought experiments that were definitive. Alas, in social psychology, our thought experiments are never definitive. However, they can be good starting points; then we have to bring them into the laboratory. But how could we invent a credible scenario that captured the essence of the seven cocktail parties, given that the whole shebang had to unfold in less than an hour? We needed to invent a situation in which the subject interacts with a preprogrammed confederate, eavesdrops while the confederate evaluates him or her to a third party, has another conversation with the confederate, eavesdrops again, converses again, eavesdrops again, and so on through seven pairs of trials. What sensible cover story could we concoct? But Darwyn Linder and I came up with one.
When the subject (a female college student) arrived at the lab, Darwyn greeted her and led her to an observation room connected to the main experimental room by a one-way window and an audio-amplification system. Darwyn told her that two women were scheduled for that hour: One would be the subject, and the other would help perform the experiment, and because she had arrived first, she would be the helper. Darwyn asked her to wait while he left the room to see if the other woman had arrived. A few minutes later, through the one-way window, the subject was able to see Darwyn enter the experimental room with another female student (Darcy Oman, an undergraduate, who served as our confederate). Darwyn asked Darcy to be seated and said that he would return shortly to explain the experiment to her. He then reentered the observation room and began the instructions to the real subject, who believed herself to be the helpful confederate.
Darwyn told her she was going to assist him in performing a verbal conditioning experiment on the other student (Darcy); that is, he was going to reward Darcy every time she used a plural noun in conversation. He told the subject these rewards would increase the frequency with which Darcy would say plural nouns, but his interest lay in whether the use of plural nouns generalizes to a new situation. That is, will people continue to use more plural nouns when they are talking to a different person, someone who has not rewarded them for doing it? Darwyn explained that he would try to condition Darcy by subtly rewarding her with a nod, a smile, and an “mmmm hmmm” every time she said a plural noun. “Will she continue to use an abundance of plural nouns when she talks to you,” he asked the subject, “even though you will not be rewarding her?” He added that our volunteer’s tasks were to listen in and record the number of plural nouns Darcy used while talking to Darwyn and then to engage Darcy in a series of conversations in which the use of plural nouns would not be rewarded—so that Darwyn could listen and determine whether generalization had taken place. Darwyn then told the subject they would alternate in talking to Darcy until each had spent seven sessions with her.
Darwyn made it clear to the subject that Darcy must not know the purpose of the experiment, lest the results be contaminated. He explained that he would tell Darcy that the experiment was about attraction, and ask her to carry on a series of seven short conversations with the subject. Between each of these conversations, both she and the subject would be interviewed—Darcy by Darwyn and the subject by an assistant in another room—to find out what impressions they had formed. During the seven meetings Darwyn had with Darcy, the subject was in the observation room, listening to the conversation and dutifully counting the number of plural nouns used by Darcy. Because she had been led to believe that Darcy thought that the experiment had to do with people’s impressions of one another, it was understandable for Darwyn to ask Darcy to express her feelings about the subject. Thus, the subject heard herself being evaluated by a fellow student on seven successive occasions. This setup is more complicated in the telling than the doing. To the participants who were embedded in the experiment, it was completely believable: Only four of the eighty-four subjects were suspicious of our procedure.
There were four major experimental conditions: (1) positive—the successive evaluations of the subject made by the confederate were all highly positive; (2) negative—the successive evaluations were all highly negative; (3) gain—the first few evaluations were negative, but they gradually became more positive; and (4) loss—the first few evaluations were positive, but they gradually became negative. The results confirmed our predictions: The subjects in the gain condition liked Darcy significantly better than those in the positive condition. (So much for Dale Carnegie’s advice to dole out praise lavishly.) Similarly, the subjects in the loss condition disliked Darcy more than those in the negative condition.
Just as the layout of the lab and the bullpen facilitated creative conversations and generated friendships among grad students and faculty, the layout of the Minnesota faculty club did the same for instructors at all levels, from new assistant professors to the most eminent. It was another stark contrast with Harvard, whose faculty club was elegant: starched tablecloths, excellent food, and waiters who served us. Yet I ate there just a few times and only when invited by a senior professor. The faculty club at Minnesota was informal, with mediocre food served cafeteria style, on plastic trays. Most of us in the Psychology Department ate lunch there almost every day, and what our lunches lacked in ambience they gained in the fun and intellectual stimulation of our conversations. We brought our trays to a large round table at the far end of the dining hall, which was known as “the psychology corner.” But it was hardly exclusive; we were often joined by colleagues from other departments, such as philosopher of science Herbert Feigl and novelist Jack Ludwig.
It was at these lunches that I got to know Gardner Lindzey, who was to become one of the closest friends I have ever had. At first I was in awe of him; he was twelve years my senior and had written or edited two of the most important books in social psychology. Gardner was the star of the round table, an entertaining storyteller and a catalyst of conversation. And he seemed to know everything about everyone’s research. At one moment he would be asking Kenneth MacCorquodale detailed questions about some of the animal learning experiments he was doing, and the next he would be probing David Lykken to tell more about the lie-detection experiments that he had begun. But he would always find a way to make each person’s research relevant to the interests of the other people at the table. He knew everything I had ever published, asking detailed questions about how it was that I had decided on this or that specific procedure.
Gardner was a broadly educated scholar who was not only a fountainhead of information; he was also an accomplished gossip. He knew everything that was going on in the profession (say, who was leaving which university to teach somewhere else, who was promoted, who had been denied tenure) and also who was about to get a divorce and who was sleeping with whom. When I asked him one day why we hadn’t seen Jack Ludwig at the table for a while, he looked at me in astonishment. “Haven’t you read Herzog yet?” he said. “It’s been out for two months! You must read it immediately. It’s about people you know.” Herzog, of course, was written by Saul Bellow, who, a few years earlier, had been a visiting professor at Minnesota. None of us (except Gardner, apparently) knew that during that time, Bellow’s wife had had an affair with Jack, who had been one of Bellow’s close friends. Bellow was furious, and although he and his wife had gone to a psychotherapist for counseling, it didn’t help and they divorced. Bellow got his revenge the writer’s way, by writing a novel in which he barely disguised the characters. He is Herzog, and Jack is Valentine Gersbach, who has an affair with Herzog’s wife. (Jack had a club-foot; Valentine has a wooden leg.) Herzog goes to a psychiatrist, Dr. Edvig, who seems so matter-of-fact about his patient’s situation that Herzog accuses Dr. Edvig of being in love with his wife himself. The psychotherapist Bellow had gone to see, Gardner told me, was . . . Paul Meehl. Later I asked Paul about the way Bellow had characterized him. “Oh, these novelists,” he said.
After a few of these stories, I started to tease Gardner by calling him the Hedda Hopper of American psychology. Paul chastised me for that. “You are underestimating our friend,” he said. “Gardner’s gossip is not limited to the U.S. It is international in scope.”
One morning Gardner phoned and asked me if I wanted to join him for a “businessman’s lunch.”
“What’s that?” I asked.
“There are some great restaurants in downtown Minneapolis. The faculty club is pleasant, but you don’t want to ruin your palate by eating there all the time, do you?”
“Certainly not,” I replied.
“Be outside your building at 11:30. I’ll swing by and pick you up.”
In that first lunch Gardner and I reminisced about our days at Harvard. He had gotten his doctorate with Gordon Allport in the late 1940s and had taught there for a few years thereafter. He asked me who my favorite Harvard people were. I named a few.
“You were at 9 Bow Street, weren’t you?” he asked.
“Yes, I was.”
“That’s funny. I notice that you didn’t mention Jerry Bruner as one of your favorite people.”
“To tell you the truth, Jerry and I didn’t get along very well.”
“I hated the bastard too,” said Gardner gleefully.
“I didn’t say that.”
“I heard what you said. And, like I said, I hated him too.”
There are many reasons that Gardner and I became close friends, but that got us off to a flying start.
This was the beginning of a number of businessman’s lunches Gardner and I had, usually once or twice a week, occasionally with David Lykken or Paul Meehl. I quickly learned that Gardner’s major criterion for a good restaurant was not the food but the quality of the bar. Our lunches lasted for well over two hours and were always preceded by two or three martinis and followed by a cognac. Amazingly, each of us was able to work hard and well in the afternoons. Since our friendship began while I was in my heavy-drinking period, Gardner’s style fitted my needs at the time, but the lunches and the friendship continued even after I cut back to one martini. As for Gardner, well, I never knew anyone who enjoyed drinking more and who could hold his liquor so well and with such grace. He was capable of abstaining for weeks at a time without a craving, and in a close friendship of almost fifty years, I never saw him out of control.
For the next few years Gardner and I had many provocative conversations and strenuous arguments about research and theory in social psychology. Once he proclaimed that Solomon Asch’s experiment on conformity was the single most important experiment in the field. I said, “Nonsense! There was no theory behind it, no control group—it was just a demonstration that people conform more than we imagined. Now, the Festinger/Carlsmith one-dollar to twenty-dollar experiment was really important—it had theory; it was counterintuitive; it changed the field! Who would have guessed that people who told a lie for one dollar rather than twenty dollars would come to believe that lie?”
After one of these exchanges Gardner invited me to coedit the second edition of the Handbook of Social Psychology. Because I had always considered the Handbook to be the bible of our field, this offer was like, well, being asked to edit the next edition of the Bible. The Handbook contained up-to-date thinking, research, and theorizing by the world’s most eminent social psychologists. I agreed, figuring that I would learn a great deal from working with Gardner (as indeed I did).
A few weeks later, while I was still in a glow over the honor of being an editor, Gardner announced that he expected me to contribute a chapter as well. This time I demurred. “There are a few areas where I could do a decent job of writing a chapter,” I said. “But as an editor I would want to recruit the best scholar working on a topic, and the best scholar ain’t me.”
“How about writing a chapter on experimental methodology?” he asked.
“I’m no methodologist,” I said.
“But you did teach a seminar on that at Harvard.”
“Well, yes, but that was mostly a hands-on course, where I helped students find the best way to test a given hypothesis.”
“Precisely the kind of thing that would be great for the Handbook ,” said Gardner. “All you need to do is take a good hard look at your own experiments and try to make explicit the wealth of tacit knowledge that you have, so that others can understand why you did what you did.”
“But teaching people how to do experiments is in the doing, not in the reading,” I protested. “It is part of a master-apprentice relationship. You learn to be a great chef by working with one, not by reading a cookbook. This may even be truer in social psychology than in cooking. We don’t have standardized procedures because, most of the time, each new hypothesis requires the invention of a new procedure tailored to test it. Leon had to invent the ‘one-dollar, twenty-dollar’ procedure, Jud and I had to invent the initiation procedure, and so on. We are always flying by the seat of our pants. It’s an exciting way to work, but it makes it impossible for anyone to write a chapter called ‘How to Do Experiments in Social Psychology’ that will be of any use to the student who is just starting out.”
“Don’t be so damn rigid,” said Gardner. “Nobody has ever tried to do it before. This could be a unique contribution. Think of it not as something that would replace working with a master experimentalist but as a set of ideas that might serve as an adjunct to the apprenticeship. Why not give it a try? After you write the chapter, if you don’t like the way it turns out, as editor, you can simply reject it.”
Although I was not convinced, I decided to plunge in—but not by myself. I recruited my favorite former graduate student, Merrill Carlsmith, who was more excited about the challenge than I was. As Gardner had suggested, we started by looking at our own experiments, to articulate the thinking that went into the design and implementation of each one. We were trying to write a cookbook without recipes. It wasn’t easy.
We began by tackling the central criticism directed against experimental social psychology at that time: The mere fact that experiments were taking place in a laboratory setting made them “unrealistic.” Merrill and I asked what proved to be an important question—what does “realism” mean? And the answer led us to clarify the difference between what we called “mundane” realism from what we called “experimental” realism. For example, several critics had claimed that Stanley Milgram’s obedience experiment was unrealistic. After all, how often in our day-to-day lives are we asked to inflict a series of electric shocks on an innocent stranger? But that criticism missed the point: Milgram’s study was unrealistic only in the mundane sense. In terms of experimental realism, the procedure was real as hell to the people participating. They sweated real sweat. Merrill and I concluded that in some situations, mundane realism is necessary; if you ask people to read a communication that you planted in a dummy newspaper, the newspaper had better look real. Usually, however, mundane realism is of no importance as long as what is happening in the laboratory has real consequences for the subjects.
Merrill and I then tackled the nuts and bolts of constructing high-impact experiments. We spelled out the experimenter’s most important challenge: How to take a compelling conceptual variable, the thing you want to understand, and translate it into a concrete set of empirical steps. The trick is to achieve a high degree of scientific rigor without losing, in the translation, the essential, complex beauty of the idea. This is harder than it sounds; many a grand, important topic gets boiled down into something trivial or simplistic by the time it has been operationalized into a precise sequence of experimental events.
The next problem is to determine how to measure the effect of your intervention. In social psychology the most frequently used measure is a rating scale of some sort, usually in the context of an interview. For example, in the initiation experiment, we asked subjects to rate how much they liked the group they had joined; in the one-dollar or twenty-dollar experiment, Merrill asked subjects to rate how much they had enjoyed the boring task of packing spools. Such measures are the easiest to administer and are often the only feasible way you can get the information you want. The experimenter must not forget, however, that a rating scale is only a stand-in for behavior. Whenever possible, it is preferable to observe what people actually do rather than to ask them what they think or how they feel. In the forbidden-toy experiment, if we want to determine whether we have succeeded in reducing a toy’s appeal, it is better to watch and see whether the kids resist the temptation to play with the forbidden toy than to ask them to tell us how much they like the toy. That’s why Jonathan Freedman’s replication of our experiment was an improvement on the original. He used the behavioral measure; he observed what the children did.
Writing this chapter turned Merrill and me into better experimenters, because it made us confront our shortcomings as well as our skills in the laboratory. Moreover, it forced us to think creatively about solutions to the problems we had posed. Getting a behavioral measure is ideal, but what if doing so is impossible, either ethically or practically? Is there a better alternative than going back to the familiar rating scale? Merrill and I invented a compromise that we dubbed “the behavioroid” measure. Suppose you want to know if watching a poignant film depicting hunger in urban America will make people more altruistic. Showing the film is easy, but you can’t very easily follow the audience around for the next few weeks to see if they become more helpful to others. However, you can ask the subjects whether they are willing to sign an official-looking contract committing themselves to a specific altruistic action, such as volunteering to work in a soup kitchen. Of course, getting a person’s “behavioroid” commitment is not quite the same as seeing whether he or she actually turns up for duty in a soup kitchen. But it is far preferable to simply asking people to rate their feelings about the poor or about the value of soup kitchens.
Finally, we gave considerable attention to the postexperimental interview—the debriefing of the subjects. Because so many experiments in social psychology require deception, Merrill and I argued that it is of the utmost importance for experimenters to disclose everything that took place and to explain why we had done what we had done. Full disclosure is a vital part of our implicit contract with our subjects. Because no one enjoys being told that he has been duped, the disclosure must be done with tact and sensitivity, or the subject will feel that he or she has been a gullible fool. The experimenter can explain that because the cover story was so carefully crafted, just about everyone accepted it; far from being “gullible,” the subjects behaved perfectly normally and reasonably. And then we explain the goal of the research—why deception was necessary and what we hope to have learned about human nature as a result of our methods. We make sure that each and every one of our subjects leaves the experimental room with her self-esteem intact, in at least as good a shape as she was when she first came in, with the bonus of having learned something interesting. I often spent as much time chatting with my subjects after the experiment as I did during the experiment itself.
When we were colleagues at Harvard, my old mentor David McClelland once told me that he had run only one social psychology experiment in his whole life, when he was a grad student at Yale. He was observing a young man through the one-way mirror when the guy suddenly stood up, unzipped his fly, and began to examine his penis in the mirror, apparently looking for a rash or abrasion. I asked Dave how he dealt with it in the debriefing.
“I didn’t debrief him,” he said. “It would have been too embarrassing. For both of us.”
I clucked my tongue. “Dave,” I said, “in this business, you always debrief. There are no exceptions. If the subject were ever to learn, perhaps from a friend who was in the experiment, that the mirror was one-way, he would have been mortified. It is always up to the experimenter to find a gentle, perhaps in this case even a humorous, way of telling him what happened.”
And so we finished our chapter, and I, in my role as coeditor of the handbook, accepted it. Much to our own surprise, Merrill and I had done just what Gardner had hoped we would. By forcing ourselves to make our tacit knowledge explicit, we produced a document that graduate students in social psychology have been relying on for almost five decades. That methodology chapter of the Handbook underwent its most recent revision in 2010. My newest coauthor is Kevin Carlsmith, Merrill’s son, who was a toddler when Merrill and I were writing the original. I admit I enjoy the irony. A project that I had initially rejected, thinking it impossible, became one of my most enduring contributions to the field. As my grandmother would have said, “Go figure!”
In 1964 Gardner left the University of Minnesota to become chair of the Psychology Department at the University of Texas at Austin, with the mandate to make it into one of the strongest departments in the country. A year later he urged me to join him and become head of the social psychology program. I had been at Minnesota only three and a half years and was reluctant to leave. True, all of the graduate students I had supervised now had good job offers, so I had no ongoing mentoring obligations.
I talked it over with Vera, who said she could go either way. On the one hand, she had developed strong friendships in Minneapolis and was hesitant to pull up stakes yet again. On the other hand, as a young mother with four small children, she found the severity and length of the winters particularly difficult. The kids would look through the windows at a bright and sunny winter day and scream to go outside to play. Then, after ten minutes, frozen to the bone, they would scream to be let back in. (Imagine the time it takes to get one little kid into and out of a snowsuit, hat, boots, and mittens and multiply that times four, and you have some idea of what Vera was up against.) And so she left the decision to me.
A reasonable person would have stayed put. I loved everything about the Psychology Department, and I had no desire to move at all; even if I had, Texas would not have been near the top of my list. I’m embarrassed to confess that I held a negative stereotype about the state, and my impression of Texas as a raw and wild place had been exacerbated by the 1963 assassination of President Kennedy in Dallas. But, in the end, my friendship with Gardner trumped both my rational judgment and my prejudice. We moved to Austin.
在线 阅读网：hTTp://wwW.yuedu88.com/