Social Intelligence: The New Science of Human Relationships

by Daniel Goleman

  Early theorists of social intelligence sought to find an analog of IQ that applied to talent in social life. Guided by the nascent field of psychometrics, they looked for ways to assess differences in social aptitudes that would be the equivalent of, say, the differences in spatial and verbal reasoning measured by IQ.

  Those early attempts fizzled, largely because they seemed to measure only people’s intellectual grasp of social situations. For instance, one early test of social intelligence assessed cognitive abilities like identifying what social situation a given sentence would be most appropriate for. In the late 1950s David Wechsler, who developed one of the most widely used measures of IQ, basically dismissed the importance of social intelligence, seeing it merely as “general intelligence applied to social situations.”7 That judgment suffused psychology, and social intelligence dropped off the major maps of human intelligence.

  One exception was the complex model of intelligence put forth by J. P. Guilford in the late 1960s; he enumerated 120 separate intellectual abilities, thirty of which had to do with social intelligence.8 But despite extensive efforts, the Guilford approach was unable to yield meaningful predictions of how well people actually operated in the social world. More recent models relevant to social intelligence—Robert Sternberg’s “practical intelligence” and Howard Gardner’s “interpersonal intelligence”—have gained more traction.9 But a cohesive theory of social intelligence that clearly distinguishes it from IQ and that has practical applications has eluded psychology.

  The old view saw social intelligence as the application of general intelligence to social situations—a largely cognitive aptitude. This approach casts social intelligence merely as a fund of knowledge about the social world. But this approach makes this capacity no different from “g,” general intelligence itself.

  But what, then, distinguishes social intelligence from g? There is no good answer as yet to this challenge. One reason is that psychology as a profession is a scientific subculture, one into which people are socialized as they go through graduate school and other professional training. As a result psychologists tend to view the world largely through the mental lens of the field itself. This tendency, however, may be skewing psychology’s ability to comprehend the true nature of social intelligence.

  When ordinary people were asked to list what makes a person intelligent, social competence emerged as a prominant natural category. But when psychologists who were considered experts on intelligence were asked to come up with a similar list, their emphasis was on cognitive abilities like verbal and problem-solving skills.10 Wechsler’s dismissive view of social intelligence seems to live on in the implicit assumptions of his field.

  Psychologists who sought to measure social intelligence have been stymied by startlingly high correlations between their results and the results of IQ tests, suggesting there may be no real difference between cognitive and social talent.11 This was a major reason social intelligence research was largely abandoned. But that problem seems to result from the skewed definition of social intelligence as simply cognitive ability applied to the social arena.

  That approach assesses interpersonal talent in terms of what people claim to know, asking whether people agree with assertions like “I can understand other people’s behavior” and “I know how my actions make other people feel.”

  Those questions come from a recently developed social intelligence scale.12 The psychologists who constructed the test asked fourteen other professors of psychology, a so-called “expert panel,” to define social intelligence. The resulting definition was “the ability to understand other people and how they will react to different social situations”—in other words, pure social cognition.13 Even so, the psychologists knew that definition would not suffice. So they made up some questions getting at how people actually get on socially, such as asking them if they agree with the assertion “It takes a long time before I get to know other people well.”

  But their test, like others, would do well to go further and assess the low-road abilities that matter so much for a rich life. Social neuroscience is detailing how multiple ways of knowing and doing spring into action as we engage with others. These ways include high-road abilities like social cognition, to be sure. But social intelligence also calls on low-road functions like synchrony and attunement, social intuition and empathic concern, and arguably, the impulse for compassion. Our ideas of what makes a person intelligent in social life would be more complete if they encompassed these abilities as well.

  Such abilities are nonverbal, and they occur in the span of microseconds, more quickly than the mind can formulate thoughts about them. Though low-road abilities may seem trivial to some, they shape the very platform for a smooth social life. Since low-road abilities are nonverbal, they elude what can be picked up in a paper-and-pencil test—and most current tests for social intelligence are such.14 In effect, they quiz the high road about the low, a questionable tactic.

  Colwyn Trevarthen, the developmental psychologist at the University of Edinburgh, argues compellingly that the widely accepted notions of social cognition create profound misunderstandings of human relations and the place of emotions in social life.15 While cognitive science has served well in linguistics and artificial intelligence, it has limits when applied to human relationships. It neglects noncognitive capacities like primal empathy and synchrony that connect us to other people. The affective revolution (let alone the social one) in cognitive neuroscience has yet to reach intelligence theory.

  A more robust measure of social intelligence would include not only high-road approaches (for which questionnaires are fine) but also low-road measures like the PONS or Ekman’s test for reading microexpressions.16 Or it could put test-takers in simulations of social situations (perhaps via virtual reality), or at least obtain other people’s views of a test-taker’s social abilities. Only then would we arrive at a more adequate profile of someone’s social intelligence.17

  In a little-remarked scientific embarrassment, IQ tests themselves have no underlying theoretical rationale supporting them. Rather, they were designed ad hoc, to predict success in the classroom. As John Kihlstrom and Nancy Cantor observe, the IQ test is almost entirely atheoretical; it was merely constructed to “model the sorts of things which children do in school.”18

  But schools themselves are a very recent artifact of civilization. The more powerful force in the brain’s architecture is arguably the need to navigate the social world, not the need to get A’s. Evolutionary theorists argue that social intelligence was the primordial talent of the human brain, reflected in our outsize cortex, and that what we now think of as “intelligence” piggybacked on neural systems used for getting along in a complex group. Those who would say that social intelligence amounts to little more than general intelligence applied to social situations might do better to reason the other way around: to consider that general intelligence is merely a derivative of social intelligence, albeit one our culture has come to value highly.

  ACKNOWLEDGMENTS

  Many people have contributed to my thinking in preparing this book, though the conclusions drawn are my own. I owe a particular debt of gratitude to those topic experts who reviewed sections of my book, especially: Cary Cherniss of Rutgers University; Jonathan Cohen, Princeton University; John Crabbe, Oregon Health and Science Center and Portland VA Hospital; John Cacioppo, University of Chicago; Richard Davidson, University of Wisconsin; Owen Flanagan, Duke University; Denise Gottfredson, University of Maryland; Joseph LeDoux, New York University; Matthew Lieberman, UCLA; Kevin Ochsner, Columbia University; Phillip Shaver, University of California at Davis; Ariana Vora, Harvard Medical School; and Jeffrey Walker, JPMorgan Partners. If readers find factual errors in the text, please notify me through my website (www.Danielgoleman.info), and I will endeavor to correct them in future printings.

  Among others who sparked my thinking, I thank:

  Elliot Aronson, Stanford University; Neal Ashkanasy, University of Queensland, Brisba
ne, Australia; Warren Bennis, USC; Richard Boyatzis, Case Western Reserve University; Sheldon Cohen, Carnegie Mellon University; Jonathan Cott, New York City; Frans de Waal, Emory University; Georges Dreyfus, Williams College; Mark Epstein, New York City; Howard Gardner, Harvard University; Paul Ekman, University of California at San Francisco; John Gottman, University of Washington; Sam Harris, UCLA; Fred Gage, Salk Institute; Layne Habib, Shokan, N.Y.; Judith Hall, Northeastern University; Kathy Hall, American International College; Judith Jordan, Wellesley College; John Kolodin, Hadley, Mass.; Jerome Kagan, Harvard University; Daniel Kahneman, Princeton University; Margaret Kemeny, University of California at San Francisco; John Kihlstrom, U.C. Berkeley; George Kohlrieser, International Institute for Management Development, Lausanne, Switzerland; Robert Levenson, University of California at Berkeley; Carey Lowell, New York City; Beth Lown, Harvard Medical School; Pema Latshang, New York City Department of Education; Annie Mckee, Teleos Leadership Institute; Carl Marci, Harvard Medical School; John Mayer, University of New Hampshire; Michael Meaney, McGill University; Mario Mikulincer, Bar-Ilian University, Ramat Gan, Israel; Mudita Nisker and Dan Clurman, Communication Options; Stephen Nowicki, Emory University; Stephanie Preston, University of Iowa Hospitals and Clinics; Hersh Shefrin, University of Santa Clara; Thomas Pettigrew, University of California at Santa Cruz; Stefan Rechstaffen, Omega Institute; Ronald Riggio, Claremont McKenna College; Robert Rosenthal, University of California at Riverside; Susan Rosenbloom, Drew University; John F. Sheridan, Ohio State University; Joan Strauss, Massachusetts General Hospital; Daniel Siegel, UCLA; David Spiegel, Stanford Medical School; Ervin Staub, University of Massachusetts; Daniel Stern, University of Geneva; Erica Vora, St. Cloud State University; David Sluyter, Fetzer Institute; Leonard Wolf, New York City; Alvin Weinberg, Institute for Energy Analysis (retired); Robin Youngson, Clinical Leaders Association of New Zealand.

  Rachel Brod, my principal researcher, provided easy access to far-flung scientific sources. A huge thank-you goes to Rowan Foster, who’s always ready for what’s needed and who keeps everything running smoothly. Toni Burbank continues to be a superb editor and pleasure to work with. And as always, I feel endless gratitude to Tara Bennett-Goleman, remarkably insightful partner in writing and in life, and a guide to social intelligence.

  NOTES

  Prologue: Unveiling a New Science

  1. The soldiers at the mosque were reported on All Things Considered, National Public Radio, April 4, 2003.

  2. On least force necessary, see, for example, law enforcement competence models in MOSAIC Competencies: Professional & Administrative Occupations (U.S. Office of Personnel Management, 1996); Elizabeth Brondolo et al., “Correlates of Risk for Conflict Among New York City Traffic Agents,” in Gary VandenBos and Elizabeth Bulatao, eds., Violence on the Job (Washington, D.C.: American Psychological Association Press, 1996).

  3. To see the way this expands our discourse, consider empathy versus rapport. Empathy is an individual ability, one that resides within the person. But rapport arises only between people, as a property that emerges from their interaction.

  4. My intent here, as in Emotional Intelligence, is to offer what I see as a new paradigm for psychology and its inevitable partner, neuroscience. While the concept of emotional intelligence has met with pockets of resistance in psychology, the notion has also been embraced by many others—most particularly by a generation of graduate students who have made it the focus of their own research. Any science advances through the pursuit of provocative and fruitful ideas rather than the lockstep pursuit of safe but sterile topics. My hope is that the new understanding of relationships and the social brain presented here will stimulate a similar tide of research and exploration. This refocusing on what happens in interactions, as opposed to within the person, as the basic unit of study has been called for, but largely neglected, within psychology. See, for example, Frank Bernieri et al., “Synchrony, Pseudosynchrony, and Dissynchrony: Measuring the Entrainment Prosody in Mother-Infant Interactions,” Journal of Personality and Social Psychology 2 (1988), pp. 243–53.

  5. On tantrums, see Cynthia Garza, “Young Students Seen as Increasingly Hostile,” Fort Worth Star-Telegram, August 15, 2004, p. 1A.

  6. The American Academy of Pediatrics recommends that children under age two not watch TV at all and that older children watch no more than two hours a day. The report on television and toddlers was presented by Laura Certain at the Pediatric Academic Societies annual meeting, Baltimore, April 30, 2003.

  7. Robert Putnam, Bowling Alone (New York: Simon & Schuster, 2000).

  8. Cited in “The Glue of Society,” Economist, July 16, 2005, pp. 13–17.

  9. On Hot & Crusty, see Warren St. John, “The World at Ear’s Length,” New York Times, February 15, 2004, sec. 9, p. 1.

  10. The data on checking e-mail are cited in Anne Fisher, “Does Your Employer Help You Stay Healthy?” Fortune, July 12, 2005, p. 60.

  11. Global average TV viewing was reported by Eurodata TV Worldwide, One Television Year in the World: 2004 Issue (Paris: Médiamétrie, 2004).

  12. On Internet use, see Norman H. Nie, “What Do Americans Do on the Internet?” Stanford Institute for the Quantitative Study of Society, online at www.stanford.edu/group/siqss; reported in John Markoff, “Internet Use Said to Cut into TV Viewing and Socializing,” New York Times, December 30, 2004.

  13. The earliest reference to the term “social neuroscience” I have found as yet is in a 1992 article by John Cacioppo and Gary Berntson. See “Social Psychological Contributions to the Decade of the Brain: Doctrine of Multilevel Analysis,” American Psychologist 47 (1992), pp. 1019–28. The year 2001 saw the publication of an article hailing the emergence of this new discipline under an alternative term, “social cognitive neuroscience,” by Matthew Lieberman (now at UCLA) and Kevin Ochsner (now at Columbia University). See Matthew Lieberman and Kevin Ochsner, “The Emergence of Social Cognitive Neuroscience,” American Psychologist 56 (2001), pp. 717–34. Daniel Siegel coined the phrase “interpersonal neurobiology” to link the interpersonal and neurobiological dimensions of the human mind so that we can understand the development of mental well-being in a fuller way; this marks another root of social neuroscience. See Daniel Siegel, The Developing Mind: How Relationships and the Brain Interact to Shape Who We Are (New York: Guilford Press, 1999).

  14. It has taken a decade for social neuroscience to reach critical mass as a field, but now there are dozens of scientific laboratories dedicated to this research. The first conference on Social Cognitive Neuroscience was held at UCLA, April 28–30, 2001, with thirty speakers and more than three hundred attendees from several countries. In 2004 Thomas Insel, director of the National Institute for Mental Health, declared that a decade of research had by then demonstrated that social neuroscience had come of age as a field. The search for the social brain, he predicted, would yield data valuable for the public good. See Thomas Insel and Russell Fernald, “How the Brain Processes Social Information: Searching for the Social Brain,” Annual Review of Neuroscience 27 (2004), pp. 697–722. In 2007 Oxford University Press will launch a journal called Social Neuroscience, the field’s first.

  15. The phrase “social brain” has come into common usage in neuroscience within the last few years. For instance, an international science conference on “The Social Brain” was held in Goteborg, Sweden, March 25–27, 2003. The same year saw publication of the first scholarly collection on the subject, Martin Brüne et al., The Social Brain: Evolution and Pathology (Sussex, U.K.: John Wiley, 2003). The first international conference on the social brain was held in Germany, at the University of Bochum, in November 2000.

  16. For the original definition of social intelligence, see Edward Thorndike, “Intelligence and Its Use,” Harper’s Magazine 140 (1920), pp. 227–35, at 228.

  17. A caveat: Those readers seeking the standard review of the psychological concept “social intelligence” will not find it here; for that, I recommend the excellent summary by John
Kihlstrom and Nancy Cantor. My intention here is to encourage a new generation of psychologists to expand beyond the limits of present concepts by integrating findings from social neuroscience, rather than adhere lockstep to the standard categories psychology has called “social intelligence.” See John Kihlstrom and Nancy Cantor, “Social Intelligence,” in Robert Sternberg, ed., Handbook of Intelligence, 2nd ed. (Cambridge, U.K.: Cambridge University Press, 2000), pp. 359–79.

  18. Thorndike, “Intelligence,” p. 228.

  PART I

  Chapter 1. The Emotional Economy

  1. When I refer to the amygdala or any other specific neural structure, I usually mean not just that region but its connective circuitry to other neural areas as well. The exception occurs when I discuss some aspect of the structure itself.

  2. Brooks Gump and James Kulik, “Stress, Affiliation, and Emotional Contagion,” Journal of Personality and Social Psychology 72, no. 2 (1997), pp. 305–19.

  3. This investigative function gets carried out through the amygdala’s links to the cortex, which guides our attention to explore uncertainties. When the amygdala starts firing in reaction to a possible threat, it directs cortical centers to fixate our attention on the possible danger, and we feel distress, uneasiness, or even a bit frightened as it does so. So if someone has a high level of amygdala activation, their world is an ambiguous and perpetually threatening place. A devastating trauma, like being mugged, can ratchet up the amygdala’s vigilance of the world, heightening levels of the neurotransmitters that keep us scanning for threats. Most of the symptoms of post-traumatic stress disorder, like overreaction to neutral events that are vaguely reminiscent of the original trauma, are signs of such an overreactive amygdala. See Dennis Charney et al., “Psychobiologic Mechanisms of Posttraumatic Stress Disorder,” Archives of General Psychiatry 50 (1993), pp. 294–305.