James Merrill Carlsmith

Introduction

James Merrill Carlsmith (April 12, 1936 - April 19, 1984) was an American social psychologist whose name is permanently attached to one of the most frequently reproduced experiments in the history of the field. As co-author with Leon Festinger of "Cognitive Consequences of Forced Compliance" (1959), he helped show that people paid a trifling sum to tell a lie came to believe the lie more than people paid generously to tell the same lie. That counterintuitive inversion - less justification, more attitude change - became known as the insufficient justification effect and gave dissonance theory its most quotable demonstration.

Beyond that single study, Carlsmith had a quietly wide career. He worked with Elliot Aronson on how the threat of punishment changes children's desires, with Mark Lepper on attributional accounts of social behaviour, and he wrote two books that shaped how the discipline taught itself: a mainstream social psychology textbook and a methods text that trained a generation of researchers in the craft of the laboratory experiment. He spent nearly all of his professional life at Stanford University, where he died at the age of 48.

Early Life and Education

Carlsmith was born in New Orleans, Louisiana, on April 12, 1936. Public biographical detail about his childhood is thin; he is usually remembered under his professional name, J. Merrill Carlsmith, and most obituary material focuses on the work rather than the upbringing.

He took his undergraduate degree at Stanford University in the mid to late 1950s, at a moment when Stanford was one of the most exciting places in American social psychology. Festinger had arrived there in 1955, and the department was thick with experiments on group pressure, opinion change and social comparison. It was in that environment, while still a young researcher working alongside Festinger, that Carlsmith became involved in the tedious "peg-turning" experiment that would make his name. His contribution was substantial enough to earn second authorship on what became one of the discipline's canonical papers - an unusual honour for someone who had not yet finished a doctorate.

He then moved east for graduate work, completing his Ph.D. at Harvard University in 1963 with a dissertation titled Strength of Expectancy: Its Determinants and Effects. The topic - how strongly a person expects an outcome and what determines that expectation - foreshadowed his later interest in attribution, the study of how people explain events and behaviour. Several reference works list Elliot Aronson as his doctoral adviser, reflecting the long professional association between the two men; accounts of exactly who formally supervised the dissertation vary, so the connection to Aronson is best described as a sustained mentorship and collaboration rather than a single documented role.

Academic and Professional Career

After Harvard, Carlsmith returned to Stanford University as a faculty member, and Stanford remained his institutional home for the rest of his life. He was part of a department that, in the 1960s and 1970s, was central to experimental social psychology on the American West Coast, alongside colleagues and co-authors such as Aronson, before Aronson moved to the University of California, Santa Cruz.

His research career can be divided loosely into three overlapping phases. In the first, running from the late 1950s into the mid-1960s, he was mainly a dissonance researcher, testing how the size of an incentive or the severity of a threat changes subsequent attitudes. In the second, from the mid-1960s onward, his interests shifted toward attribution theory and toward questions of punishment, reward and the socialisation of children. The third phase was pedagogical and methodological: he invested heavily in textbooks and handbooks rather than in a single research programme, an unusual choice that made his influence broader but his personal "brand" less visible than that of his collaborators.

He was known as a careful experimenter with a strong interest in design rather than in rhetoric. Students who trained under him included Phoebe C. Ellsworth, later a distinguished psychologist of emotion and law, which gives some measure of the methodological seriousness he demanded.

Carlsmith died on April 19, 1984, at Portola Valley, California, at the age of 48. Stanford issued a memorial resolution in his name, which remains one of the few authoritative accounts of his career. He was married to the social psychologist Lyn Carlsmith (1932-2011), herself a respected researcher on aggression and punishment; they had three children.

Major Contributions to Psychology

Forced compliance and insufficient justification

The experiment published as Festinger and Carlsmith (1959) had an almost absurdly simple structure. Male undergraduates were asked to perform two deliberately monotonous tasks - turning small square pegs a quarter turn at a time, and winding and unwinding spools - for roughly half an hour to an hour. Afterwards, the experimenter explained that he needed help: the next participant was waiting outside and needed to be told that the task was interesting and enjoyable. Would the subject do it? One group was offered $1 for this small lie, a second group $20 (sums with considerably more psychological weight in the 1950s than today), and a third group told no lie at all.

Later, participants were asked privately how much they had actually enjoyed the tasks. The $20 group rated them about as boring as everyone else did; the size of the payment fully explained having told the lie, so nothing needed adjusting. The $1 group, by contrast, rated the peg-turning as significantly more interesting. A dollar was not enough to justify having lied, so the cheapest available way to resolve the inconsistency was to change what they believed about the task itself.

This was "insufficient justification," and it inverted what reinforcement theories predicted. It also carried an obvious practical corollary that dissonance researchers have repeated ever since: if you want durable internal attitude change rather than temporary compliance, use the smallest incentive that will produce the behaviour, not the largest.

Severity of threat and forbidden behaviour

Working with Aronson, Carlsmith extended the logic from adults who lie to children who are tempted. In the 1963 "forbidden toy" study, nursery-school children were shown a set of attractive toys and told that a particular one was off-limits. Half were threatened with a severe punishment for touching it, half with only a mild one. When later asked how much they liked each toy, children who had refrained under a mild threat rated the forbidden toy as less desirable than did children restrained by a severe threat. A big punishment supplies an external reason for not playing with the toy, leaving desire intact; a small one leaves the child to conclude that the toy simply was not very appealing. The finding became a standard demonstration of how external pressure can crowd out self-generated attitudes, and it anticipated later work on intrinsic motivation.

Attribution, expectancy and research method

Carlsmith's dissertation work on the strength of expectancy fed into a broader concern with how people read their own behaviour. Reference works credit him with sustained collaboration with Mark Lepper on attribution theory, exploring how the causal explanations people generate for their own actions shape later attitudes and motivation - the intellectual context in which Stanford work on undermining intrinsic interest with extrinsic reward was developed.

His final contribution was to method. Social Psychology (1970), written with Jonathan L. Freedman and David O. Sears, went through multiple editions and became one of the standard texts of the era. Methods of Research in Social Psychology (1976), with Phoebe C. Ellsworth and Elliot Aronson, and revised in 1990 with Marti Hope Gonzales, codified hard-won practical knowledge about control conditions, confederates, deception and debriefing. Given how much of mid-century social psychology depended on elaborate deceptions, a serious book about how to do it honestly and interpretably mattered.

Major Publications

"Cognitive Consequences of Forced Compliance" (1959), Journal of Abnormal and Social Psychology, 58(2), 203-210, with Leon Festinger. The $1/$20 experiment and the most cited work of his career.
"The Effect of Severity of Threat on the Devaluation of Forbidden Behavior" (1963), Journal of Abnormal and Social Psychology, 66(6), 584-588, with Elliot Aronson. The forbidden-toy study with nursery-school children.
"Studies in Forced Compliance" (1966), with Barry E. Collins and Robert L. Helmreich. A methodological follow-up comparing novel formats for inducing compliance.
Social Psychology (1970), Prentice-Hall, with Jonathan L. Freedman and David O. Sears. A widely used introductory textbook, revised in 1974, 1978 and 1981.
Methods of Research in Social Psychology (1976), Addison-Wesley, with Phoebe C. Ellsworth and Elliot Aronson. Second edition published in 1990 with Marti Hope Gonzales.

Impact and Legacy

Few graduate students in social psychology pass through without meeting the $1/$20 study. It appears in every introductory textbook, it anchors the concept of insufficient justification, and it shaped practical advice in education, management and public health: minimise external pressure if you want lasting change. That durability is Carlsmith's legacy.

It is also a legacy under open challenge. The most serious theoretical objection arrived in 1967, when Daryl Bem proposed self-perception theory. Bem accepted the data entirely and offered a different mechanism: participants did not suffer any discomfort at all, they simply observed their own behaviour ("I said it was fun; I was barely paid") and inferred their attitude the way an outside observer would. Observers told only the procedure reproduced the same pattern, which was difficult for dissonance theory to explain away. Through the 1970s a series of "crucial experiments" - by Bem and McConnell, by Snyder and Ebbesen, and by Fazio, Zanna and Cooper in 1977 - produced an awkward compromise rather than a verdict: dissonance processes seem to operate when behaviour is clearly counter-attitudinal and outside a person's latitude of acceptance, self-perception processes when the behaviour is closer to what the person already believed.

Methodological objections ran alongside the theoretical ones. Critics including Chapanis and Chapanis, Rosenberg, and Elms and Janis questioned whether the dependent measures registered what they claimed to register. Martin Orne's 1962 account of demand characteristics applied with uncomfortable directness: a man who has just been paid a dollar to mislead a stranger is unlikely, when questioned minutes later, to say he did it for money. Rosenfeld, Giacalone and Tedeschi later pressed the same point as impression management, arguing participants were curating how they looked rather than reporting how they felt. In 1972, Joel Cooper and Edward E. Jones tested a related attributional alternative - Harold Kelley's suggestion that low-incentive subjects assume they are unusual in complying - and found no support for the "illusion of uniqueness," while the inverse incentive effect reproduced once again. Defenders noted that the original 1959 design had anticipated some of these worries: the key interview was conducted away from the laboratory by a different person, presented as unrelated to the experiment.

The most serious empirical challenge is recent. Vaidis, Sleegers and colleagues published a multilab replication of the induced-compliance paradigm in 2024, across 39 laboratories and close to 4,900 participants. Writing a counterattitudinal essay did move attitudes, which supports the broad claim that behaviour shifts belief. But high-choice participants changed no more than low-choice participants, meaning the freedom-of-choice manipulation that dissonance theory treats as essential failed to produce its signature effect. The authors concluded that the results call into question whether the induced-compliance paradigm provides robust evidence for cognitive dissonance.

Dissonance researchers have answered in several ways. Some argue that the massive multilab design, while statistically powerful, necessarily strips away the face-to-face contact, cover stories and careful suspicion-probing that the effect seems to require; the original studies were small, finely tuned laboratory encounters. Others note that the replication confirmed the basic counter-attitudinal attitude shift, and that dissonance theory's modern successors - notably the "new look" model emphasising freely chosen behaviour that produces an unwanted, foreseeable consequence - specify conditions the older work did not. Neither side claims the question is settled, and the current consensus is best stated negatively: the phenomenon is real at least sometimes, its boundary conditions are more demanding than mid-century textbooks suggested, and the classic $1/$20 result is now cited as a demonstration rather than as settled fact.

There is also an ethical dimension to the record. The 1959 study relied on substantial deception, and contemporary accounts of the procedure indicate that debriefing was not what modern review boards would require. The same is true of much research from the period, but because the study is taught so widely, it now functions as a teaching case in two registers at once: a model of elegant experimental logic, and an illustration of how far research-ethics standards have moved.

Learn More

References

Festinger, L., & Carlsmith, J. M. (1959). Cognitive consequences of forced compliance. Journal of Abnormal and Social Psychology, 58(2), 203-210.
Aronson, E., & Carlsmith, J. M. (1963). Effect of the severity of threat on the devaluation of forbidden behavior. Journal of Abnormal and Social Psychology, 66(6), 584-588.
Carlsmith, J. M., Collins, B. E., & Helmreich, R. L. (1966). Studies in forced compliance: I. The effect of pressure for compliance on attitude change produced by face-to-face role playing. Journal of Personality and Social Psychology, 4(1), 1-13.
Bem, D. J. (1967). Self-perception: An alternative interpretation of cognitive dissonance phenomena. Psychological Review, 74(3), 183-200.
Chapanis, N. P., & Chapanis, A. (1964). Cognitive dissonance: Five years later. Psychological Bulletin, 61(1), 1-22.
Bem, D. J., & McConnell, H. K. (1970). Testing the self-perception explanation of dissonance phenomena: On the salience of premanipulation attitudes. Journal of Personality and Social Psychology, 14(1), 23-31.
Cooper, J., Jones, E. E., & Tuller, S. M. (1972). Attribution, dissonance, and the illusion of uniqueness. Journal of Experimental Social Psychology, 8(1), 45-57.
Fazio, R. H., Zanna, M. P., & Cooper, J. (1977). Dissonance and self-perception: An integrative view of each theory's proper domain of application. Journal of Experimental Social Psychology, 13(5), 464-479.
Orne, M. T. (1962). On the social psychology of the psychological experiment: With particular reference to demand characteristics and their implications. American Psychologist, 17(11), 776-783.
Freedman, J. L., Sears, D. O., & Carlsmith, J. M. (1981). Social psychology (4th ed.). Prentice-Hall.