SULLOWAY: During the last two decades I have experienced a major shift in my career interests. I started out as a historian of science and was primarily interested in historical questions about people's intellectual lives. In trying to understand the sources of creative achievement in science, I gradually became interested in problems of human development and especially in how Darwinian theory can help us to understand the development of personality. I now consider myself a psychologist, in addition to being an historian.

JB: How did you make that leap?

SULLOWAY: This leap was determined by the kinds of questions I was asking. I was initially drawn to the problem of why scientists accept new ideas. If you survey the history of science, it is apparent that most individuals who have accepted radical innovations did not do so simply because they knew of some line of evidence that other people were unaware of. Darwin is a good case in point. He came back from the Beagle voyage and displayed his famous Gal‡pagos specimens in London. Within six months of his return, most of the top naturalists in Britain had seen Darwin's Gal‡pagos finches and reptiles, and hence the crucial evidence that converted Darwin to evolution (and that we now consider the textbook case of evolution in action). John Gould, who was one of the greatest ornithologists of the nineteenth century, knew far more about Darwin's Gal‡pagos birds than Darwin did. Gould corrected numerous mistakes that Darwin had made during theBeagle voyage, such as thinking that many of the finches from the Gal‡pagos Islands were the forms that they have come to mimic though biological evolution. For example, Darwin had mistaken the warbler finch for a warbler, and he had thought the cactus finch was a member of the Icteridae--a completely different family of birds. Gould corrected these errors and also showed Darwin that some of the other birds he had not recognized as finches were part of a single closely related group. Darwin was stunned by this and other crucial information that he received from Gould in March of 1837, and Darwin immediately became an evolutionist. The strange thing is that Gould did not. He remained a creationist even after The Origin of Species was published. Hence the man who knew more saw less, and the man who knew less saw more. It struck me that this puzzling episode in intellectual history had something to do with temperament, or character, or personality. It certainly didn't have anything to do with the scientific evidence per se. Darwin, Gould, and many other contemporary naturalists all knew about the same evidence. This leads to the inference that people who make creative leaps in science, and in other fields, do so in part because of their personalities-and more particularly because of their ability to think in new and unconventional ways. In short, I became interested in psychology.

JB: Was this a purely intuitive leap of mind?

SULLOWAY: There was certainly a lot of intuition involved in the leap. Fortunately, the intuitive leap was then followed up by hypothesis testing, which is a method that saves us all from becoming either astrologers or psychoanalysts.

I see the mind as an exquisitely engineered device÷not literally engineered, of course, but designed by the mimic of engineering that we see in nature, natural selection. That's what "engineered" animals' bodies to accomplish improbable feats, like flying and swimming and running, and it is surely what "engineered" the mind to accomplish its improbable feats.

JB: How did this idea creep into your consciousness?

SULLOWAY: It was partly intuition, and it was partly just hard evidence. In the early 1970s I began reading everything I could find in personality psychology, especially the literature on cognitive style, and I also began doing research in this area. Eventually I stumbled onto the topic of birth order, on which I subsequently spent two decades doing research. Birth order, however, was just the tip of the iceberg in this research project. The minute one begins to deal with the issue of family dynamics, one also encounters other important factors that are causing personality to develop the way it does.

JB: What was your background?

SULLOWAY: I was a first-year graduate student when I developed the interests that have marked my work on scientific creativity. I was just beginning to do my preliminary course work for a degree in the history of science. At that time I anticipated writing a doctoral dissertation on Darwin's life. I had done quite a bit of research on Darwin. For example, I had retraced theBeagle voyage around South America and I had made a series of films on Darwin's voyage. I also knew a great deal about Darwin's conversion to evolution, and the specific reasons why Darwin converted; and I had begun to write various papers on these topics-papers that eventually became published articles. In hindsight, I had stumbled onto a problem-Darwin's conversion-that completely changed by career. At one point I seriously considered getting a joint degree in psychology, and did most of the necessary course work in this field. Although I did not end up taking a joint degree, I had entered into what became a kind of hybrid career path. I continued to do considerable reading and research in psychology; I kept up my previous interests in evolutionary biology; and I also continued with my researches in the history of science-particularly on the topic of revolutions in science.

JB: Where were you at the time?

SULLOWAY: I was a graduate student at Harvard University. About two years into my graduate studies period I became a Junior Fellow in the Society of Fellows, and this was a wonderful experience. Being a Junior Fellow freed me to work in any area that I wanted. I was no longer under the direct supervision of anyone in my department. It was a terrific experience, and I thrived on the independence it provided.

JB: Let's talk about the thesis that led you to your bookBorn to Rebel.

SULLOWAY: Essentially what I stumbled on in 1970, and then empirically verified over a 20-year period, is that aspects of personality that are under environmental control are strongly influenced by family niches. Birth order is particularly important in this regard, because it is a systematic source of differences in family environments. But birth order is not a cause, in and of itself. Rather, it's a surrogate, or a proxy, for patterns of family dynamics that are actually molding personality. For example firstborns are bigger than their younger siblings. They also are older and tend to have more status. In competition with their siblings, there are certain strategies that eldest children can employ that younger children cannot. A younger child can decide to hit an elder sibling, but this is usually not a smart idea because the elder sibling can hit back harder. In general firstborns tend to be more aggressive; they use strategies and tactics that take advantage of their greater physical size.

There is an important dimension of personality called "agreeableness/antagonism"—one of the Big Five—that exhibits significant differences by birth order. This birth-order difference reflects difference in the niches that firstborns and younger children typically occupy. Firstborns tend to occupy the niche of a surrogate parent. Acting as a surrogate parent-that is, assisting with child-rearing duties-is a great way to curry favor with parents. For this reason, firstborns tend to identify more closely with their parents, and they also tend to identify with whatever their parents value. Parents value a child's doing well in school, so firstborns are conscientious, do their homework, generally do better at school, and tend to be over-represented as academics and in Who's Who. The niche of the responsible achiever is particularly likely to be open for an eldest child. Once this niche is taken, it is difficult for a younger sibling to compete effectively for the same niche, although they often try. The typical strategy of younger siblings is to see whether they can compete successfully in a niche already occupied by an elder sibling. If they cannot, then the best strategy is for the younger sibling to branch out—to become more open to experience—and to try to find some alternative niche where they will not be directly compared with their elder siblings. If an elder brother is a great spear-thrower and a younger cannot top that, they might as well take up the bow and arrow. And if there is another older sibling already specializing in the bow and arrow, then it pays to invent the crossbow. The general rule, then, is do something different that adds value to the family unit as a whole. Like Darwin's famous finches, younger siblings are busy diversifying: They are trying to radiate adaptively away from whatever specialized abilities are already represented by siblings who are older than themselves.

These "contrast effects" between siblings explain the relationship between birth order and certain kinds of creativity. Younger siblings are much more likely to accept radical innovations in science and in social thought. Within their own families, they are at the bottom of the pecking order, so they tend to identify more with the underdog and to champion egalitarian causes. Younger siblings were the earliest backers of the Protestant Reformation, and after it the Enlightenment. Most lost causes in history have been supported by younger siblings and opposed by firstborns. This historical difference goes directly back to the kind of psychological differences in strategic niches that siblings occupy within the family constellation.

JB: You have stated that younger siblings have more in common with their peers than their siblings.

SULLOWAY: On average, firstborns are more similar in personality to firstborns in other families than they are to their own younger siblings. Similarly, a youngest child in one family is often more similar to a youngest child in another family than to his or her own elder siblings. Still, all laterborns are more similar to one another, on average, than they are to firstborns.

JB: How did you test this hypothesis?

SULLOWAY: There are several ways of testing it. In my book Born to Rebel, I engaged in two major empirical assaults on this problem. The first method of attack involved historical evidence. I gathered data on more than 6,500 participants in major revolutions in science, politics, and social thought. In addition, I arranged for each individual's position in each controversy to be validated by half a dozen or more expert historians. Overall, I asked 110 historical experts to examine my lists of participants in revolutions, and to assess whether these lists were representative of participants as a whole. My experts were also asked to nominate missing individuals, and they rated every participant on a scale of acceptance and rejection. Obtaining these expert ratings involved a tremendous amount of work, in part because I did it in person. I flew a quarter of a million miles around the world as I gathered these expert ratings from scholars in England, France, Germany, Italy, and America. My second line of research involved a reassessment of the birth-order literature as a whole. There are more than 2,000 publications on this subject, and what was needed was a meta-analysis to determine whether there are more significant findings than would be expected by chance. In my meta-analysis I tested specific hypotheses about sibling strategies, using the Big Five personality dimensions as my guide. That is, I expected firstborns—relative to laterborns—to be more (1) conscientious, (2) aggressive, (3) conventional, (4) extraverted in the sense of being dominant (laterborns are more extraverted in the sense of being sociable), and (5) emotionally volatile, in the sense of being quicker to anger. All five of these hypotheses were confirmed by my meta-analysis, which involved a statistical survey of 196 birth-order studies controlled for social class and sibship size.

JB: What sort of grant support did you have?

SULLOWAY: My collaboration with my 110 expert raters was done when I was a MacArthur Fellow, and this fellowship was a opportune source of support for my project. Being a MacArthur Fellow was a boon to my ability to get on with the massive amounts of empirical research for this project and to overcome one of the most obvious objections to it, namely: If I have selected the historical samples, why should anyone trust my results? It was essential that the classification of my historical participants as supporters or opponents of radical change be done by people other than myself. As a MacArthur Fellow, I spent every penny of my stipend on research and living expenses.

JB: What procedures did you use after you gathered the results?

SULLOWAY: After I had assembled my samples for each of the 121 historical events in my study, I coded every individual for up to 256 different background variables. One of the most unusual features about Born to Rebel is that it surveys more than a hundred potential causes of radical thinking, and attempts to rank order these influences in terms of overall influence. Is social class a good predictor of radicalism? This variable is in my data base, so I can answer this question: Social class is not a good predictor. Is age a good predictor?: Yes, age is, just as Max Planck and others have thought, although age is not as good a predictor as either social attitudes or birth order. I also tested a special sub-set of variables-those related to sibling strategies and family dynamics-many of which also turned out to be significant predictors of radicalism. For example, age spacing between siblings is a significant predictor: Large age gaps between brothers and sisters cause the effects of birth order to dissipate. Conflict with parents is also a significant predictor of radicalism, and it is especially important for firstborns. Laterborns do not need to have the Wicked Witch of the West as a mother in order to become radicals: They have their older siblings to induce this behavioral predilection. But firstborns who grow up in happy families typically identify with parents and authority. Significant conflict with a parent tends to undermine this pattern of identification and causes firstborns to identify instead with the underdog. When I tested all of these different variables simultaneously, the single best predictor of radicalism proved to be birth order. But birth order is hardly the only significant predictor. The next two predictors in importance are social attitudes and age, followed by parent-offspring conflict.

JB: Your sampling of participants in radical revolutions seems to involve highly accomplished people who were successful enough to become historical figures. Would the same results apply if you had included the average person in your samples?

SULLOWAY: There are two ways we can answer this question. The first is to take my sample of 6,500 historical figures and rank them on a scale of eminence. I have done this, using 18 different eminence measures. There are some people, such as Darwin and Newton, who are particularly eminent. But when we go down the list, in order of eminence, we come to people who are so obscure that even Newton or Darwin scholars have not always heard of them. After we have stratified individuals by eminence, the question we may ask is whether there is any dilution of a general birth-order effect as we go up or down the scale? In other words, are larger effects are associated with eminence? As it turns out, the most obscure people in my sample show virtually the same effects for the influence of birth order as do the most eminent people. It is true that I have not included individuals in my study who are so obscure there is no biographical information about them. But by extrapolation, if there are biases in my study owing to the selection of eminent figures, we should be able to detect their extent when the samples have been stratified by eminence.

The second way to tackle this problem is to study ordinary people. Fortunately, this research has already been done. As I have previously mentioned, there are more than 2,000 published studies on birth order. Much of my own contribution in Born to Rebel was to try to make sense out of this extensive literature. This literature has been repeatedly criticized because many of the studies are not well designed or controlled for important background variables. The simplest way to solve these problems is to throw away all the studies that are not well designed. If we take the remaining 196 studies that arecontrolled for class and sibship size, we may ask how many significant findings are there in this set of 196 studies. As it turns out, there are 86 significant findings. The key question, then, is how often would this number of significant findings occur by chance? The procedure used to answer this question is called meta-analysis. The answer is that we would expect to get 86 significant results by chance once in a billion times. In fact, the birth-order literature is in surprisingly good shape compared to most other research areas in psychology.

JB: Let's talk about the intellectual antecedents.

SULLOWAY: There is a vast literature on birth order and personality, and, of course, on many of the other variables that I studied in Born to Rebel, including gender and parent-offspring conflict. Freud, for example, based his theory of personality development on parent-offspring conflict, and most aspects of family dynamics that I studied have also been extensively studied by other people. In my opinion, one of the most useful contributions of Born to Rebel was my effort to simultaneously assess many different influences that theorists from Freud to the present have thought were important.

JB: Two questions—What about the only child, and what about women? It seems like all the example I've heard you talking about are males.

SULLOWAY: I included a chapter in my book on women. In this connection I made a special effort to find historical samples where a substantial proportion of women participated in radical events—precisely so I could say something substantive about sex (and sexual differences). In general, women who end up in the history books as supporters of radical causes tend to be an unusual group. To begin with, they are much more liberal than the average man in the population. They are also more likely to have experienced substantial conflict with a parent, and they are far more likely to have been laterborn (and usually lastborn). In other words, the women who made it into the history books are typically the rebels of the family. These are individuals who boldly transgressed into a man's world because they were not willing to sit there and do what women were generally supposed to do prior to the 20th century. Their first "revolution" was getting into my sample. The historical revolution they later participated in, and that brought them to my attention, was a second revolution for them. Because I possess a reasonably large proportion of women in certain radical movements in my study—for example, in the Protestant Reformation and in 61 social reform movements that I studied in American history—I can say with confidence that birth-order effects in radical temperament hold for women as well as men.

Only children pose another interesting question. I view only children as the ideal controlled experiment. They are what it is like to have no birth-order effects at all: Only children have no siblings, hence they have no sibling rivalry. Two predictions follow from these circumstances. One is that only children ought to be intermediate on many personality traits. This follows because they are not being pushed by a younger sibling into being particularly conscientious or aggressive; and they are not being pushed by an elder sibling into being particularly daring or unconventional. Hence only children ought to be somewhere in the behavioral middle. And this is where they turn out to be. Secondly, only children are free to occupy any niche they wish to in childhood—for example, they do not have to worry about who is going to move in to occupy a niche that they vacate. For this reason, they are free to roam around. As a result, they ought to be more variable than average in their personality traits and interests, and they are. Only children are the most unpredictable group. Their behavior is difficult to predict precisely because their childhood options are greater than for people who grow up with siblings.

JB: What were some of the reactions to your book?

SULLOWAY: There have been a variety of reactions to the book, some that I anticipated and some that I did not. One of the most surprising reactions involved the accusation that I was a "determinist". This accusation took two forms: one involving determinism in a general sense and the other involving genetic determinism. I was puzzled by both forms of this accusation. If one reads my book carefully, it is obvious that sibling strategies are not strictly "determined". Rather, they are self-determined. Individuals have considerable choice as to which strategies they adopt in family life. For example, younger siblings are—on average—less aggressive than their elder siblings, but younger siblings always have theoption of being aggressive. Nothing stops them from punching an older sibling in the nose. But such aggressive acts are generally ill-considered, because older siblings can punch back harder. Younger siblings learn this lesson early on and behave accordingly. Most of the choices that siblings make in the course of human development are voluntary. Hence these choices are self-determined. It's really a mincing of words to call such actions "determined". We all know that it is unwise to cross the street when a giant Mack truck is likely to run us over. This fact, to which most of us wisely adapt, does not mean that all of our actions are predetermined. In short, some things in life are determined, and other things are not; but I hardly see this circumstance as something to get worked up about.

JB: You're talking about probabilities, you're not claiming that every firstborn has these characteristics.

SULLOWAY: Right, mine is a "probabilistic" account of behavior, in part because there are so many different variables that influence personality, including gender, parent-offspring conflict, birth order, and lots more that I document in my book. One can legitimately accuse me of being a multi-determinist. My book tells a very complex story and, in this story, there is lots of room for individual choices.

The second form of the determinist accusation directed against my book involved attempts to portray me as a genetic determinist. The few reviewers who tried to make this point did not understand the difference between a purely genetic argument and a developmental one. It is true that Born to Rebel is very much a Darwinian book, but this is hardly the same as being an argument for genetic determinism. One of the most subtle features of my argument in Born to Rebel is that one can propose a Darwinian argument that is highly environmentalist. Normally we don't hear about these kinds of arguments because this aspect of the story of human development is not well understood.

Here's the argument in a nutshell. Based on Darwinian theory, I argue that offspring are predisposed (genetically) to compete for parental investment. The role of the environment inevitably comes in because individuals—based on the contingencies of birth order, gender, and age spacing—tend to occupy different family niches. This part of the argument is not at all based on genetic determinism. There are no genes for being firstborn or genes for being laterborn. Siblings become very different in large part because different family environments—or niches, if you will—lead them to adopt differing strategies in their efforts to get out of childhood alive. Because firstborns are bigger than their younger siblings, it is easier for them to employ aggressive and tough-minded tactics, which then become part of their personality. This part of the theory is very much anenvironmental and interactionist argument. My reasoning in Born to Rebel is like Pinker's argument inThe Language Instinct. There's undoubtedly a hard-wired capacity for humans to engage in verbal communication, a capacity that other apes do not possess. But the country we grow up in determines which language we learn to speak. In the same way, we are hard-wired in a Darwinian sense to compete with our siblings for parental investment, but the particular aspects of each person's personality are the product of characteristics of the family environment in which one grows up, just as speaking German in one country, and French in another country, are appropriate linguistic differences produced by the same language instinct. In short, my argument is not just about nature; nor is it just about nurture-it is a combined nature/nurture argument, in which much of the psychological details are clearly on the environmental side.

Most readers of my book correctly understood this point. In an interview with Ted Koppel on "Nightline", Stephen Jay Gould emphasized this general logic when he said that birth order provides one of the best demonstrations of the power of the environment and is, on this account, a wonderful antidote to the kinds of genetic determinist arguments espoused in "The Bell Curve". I find it ironic to have been accused of being a genetic determinist by some people, and yet to have been publicly defended against this accusation by one of the leading critics of such views.

JB: Interesting that Gould and Pinker, who frequently disagree, appear to support your ideas. What do the adaptationists—John Maynard Smith, George Williams, Richard Dawkins—have to say about your book?

SULLOWAY: I don't know what Maynard Smith or George Williams think. I gave a lecture on my ideas at the Human Behavior and Evolution Society in 1995, where Richard Dawkins was the keynote speaker, and he seems to have been impressed with the argument. He referred to my paper several times in his keynote speech, at the end of the conference.

JB: What about Dan Dennett?

SULLOWAY: After the publication of Born to Rebel, Dan sent me a cordial letter saying that he had read my book and that, in general, he agreed with my argument. I am not surprised because, for a sophisticated Darwinian such as Dennett, there is not much that is really controversial about the book. It makes good sense that, if offspring are competing for parental investment, they will devise strategies to implement this competition in their favor.

JB: Are there any particular people mounting the attack?

SULLOWAY: The critics have not been connected by any single discipline. The most interesting responses to the book are now coming from psychologists who are busy trying to test and replicate some of my findings. This is becoming an interesting source of potential controversies for the following reasons. There are already more than 2,000 studies on birth order, and more than half of those studies show no significant findings. How can this be, if birth order has an important influences on personality? The answer is twofold. The first part of the answer is that self-report data are not all that reliable. If I had been able to ask Robespierre whether he was a mean and vindictive fellow, I don't think he would have replied in the affirmative. If I had been able to ask Darwin's staunch American opponent, Louis Agassiz, whether he considered himself reluctant to accept new ideas, he would rightfully have said, "No, I am very open to new ideas. I was a pioneer in the development of glaciation theory." Agassiz's openness to the theory of the Ice Ages is not inconsistent, however, with his vehement opposition to evolution. Evolution was a radical innovation, whereas glaciation theory was a somewhat conservative innovation closely allied to catastrophism. Agassiz later used glaciation theory as a conceptual weapon against evolution, claiming that each Ice Age had extinguished life on earth, requiring a new Creation by God to repopulate the planet. When one asks someone a question such as "Are you open to new ideas," most people interpret the question in ways that fit their own particular values and biases. We are all open tosome things. What we want to understand is how do birth order and other influences on personality channel our predispositions to be open to experience in specific ways. Personality tests are not particularly good at capturing these context-sensitive effects.

In Born to Rebel I was careful to identify the social and intellect context of each the innovations I was studying. For each scientific revolution that I studied, I operationalized the social context in terms of how ideologically radical the innovation was, how long the revolution took to be resolved, and various other measures of "radicalism." These markers of controversiality proved to be excellent predictors of the size of birth order effects. In addition, these contextual markers were also significant predictors of the effectiveness of other explanatory constructs, such as age, parent-offspring conflict, and social attitudes. In my book, I was continually dealing with person-by-situation interaction effects. Psychologists are now trying to replicate my findings without worrying about the context. Another problem with such studies is that self-report data tend to yield fairly small birth-order effects. We know from considerations of statistical power that one needs a sample of between 500 and 1,000 individuals to be reasonably sure that one is not missing a true effect owing to sampling error. The average study in psychology involves about 250 individuals. Psychologists have been designing studies to test my claims, based on samples of 200-400 subjects. These studies are generally incapable of answering the question that the investigators are asking, which is a waste of time and effort. Unfortunately, most psychologists—to this day—do not appreciate the issue of statistical power.

I recently designed a study myself to get around these dual problems of statistical power and self-report biases. The sample already includes about 3,500 subjects, and some of the questions I have asked are aimed at tapping objective indicators of behavior. For example, if I ask individuals to tell me how empathetic they are, using a 9-step scale, I know that I am not always going to get a realistic self-appraisal. In addition, most people don't know where they really lie on an objective measure of empathy. They might know that they are higher than the average person, but they do not know whether they are in the 60th percentile or the 70th percentile—we don't go around wearing "empathy badges" that identify us like men and women. And so there's a lot of imprecision in answers to questions of this sort. Small effects, including those for birth order and other aspects of family dynamics, are easily missed. So what I have done in my study is to include a second set of questions, which ask respondents to rate themselves relative to their friends, spouses, and siblings. Consider the approach entailed in a direct sibling comparison. We generally know (or think we know) whether we're higher or lower than a sibling on most personality traits, and so the method of direct sibling comparison serves to anchor each personality scale with a concrete comparison. We might be in error as to where we place ourselves on such scales-in absolute terms-but we are probably close to the truth in assessing the relative difference between ourselves and a sibling. When people compare themselves with a sibling, it turns out that the correlations between birth order and personality are at least twice as large compared with when subjects assesses themselves without reference to anyone else.

JB: You're talking about statistical results, but a lot of people are reading your book and thinking about it on the personal level.

SULLOWAY: Well these two ways of viewing the matter are not inconsistent. I employ statistical techniques and large samples just to be sure that I am right about the relationships I am studying. Once a researcher obtains the correct answer by this method, findings can be illustrated by anecdotes, which represent the level of personal truth that lay readers seek in a book such as mine. Anecdotes have a wonderful power to convey emotional truths. But I do not consider anecdotal evidence to be a proof of anything-on this important point I depart company with most historians, who actually think they've proven something when they tell a story. A story proves nothing; it just demonstrates that people have been clever enough to find evidence to fit their hypotheses. The approach I took in Born to Rebelinvolved testing my hypotheses using large statistical samples, and then illustrating the various relationships I had documented by telling one or more stories that brought these relationships to life. For example, laterborns are more likely to challenge the status quo, and they are more likely to cause their parents aggravation by doing all sorts of outrageous things. A person who exemplifies this tendency is Voltaire—he got his start as a poet when his family, to amuse themselves, had Voltaire and his elder brother Armand engage in poetry contests. The family soon discovered that Voltaire was a terror at satirical poetry—and he was probably aiming many of his scathing ditties at his elder brother, whom he didn't particularly like. The family put an end to these poetry contests. The father subsequently became concerned that his younger son would end up wasting his life in such an unfruitful profession as literature. "You will starve to death," he warned his son. But a poet had been born, and Voltaire became the richest literary figure in all of eighteenth-century Europe through the sales of his ribald poems, plays, and books. His brother Armand, by the way, became a religious fanatic. What is Voltaire most famous for? His scathing critiques of the Catholic church!

Here is another story about Voltaire that I cannot resist telling. Voltaire once witnessed his father having a vehement argument with his gardener. Voltaire's father was a stubborn man. He finally dismissed the gardener, saying to him, "I hope you find an employer who is as gracious and kind as I am." Voltaire thought this remark was ridiculous-that his father, one of the most irascible people he knew, would tell the employee he had just fired that he would be lucky to find another employer as even-tempered as himself. Soon after, Voltaire went to see a play. It turned out that there was a scene in the play just like the Voltaire had witnessed between his father and the gardener. After the play was over, Voltaire went to see the playwright and asked him if he would substitute, in the next performance of the play, a few words that were closer to his father's own remarks. Voltaire then went home and invited his father to attend the play. His father accepted, and as the father sat through the play, there finally came the scene with the gardener. Voltaire wrote of this episode that "My good father was rather mortified." This story reflects the use of the satirical knife blade, and the turning it in his victim, that Voltaire did to his enemies throughout his career. Some noblemen became so outraged by Voltaire's satirical broadsides that they had him beaten, or arranged for him to have a nice long stay in the Bastille. In any event, these are the kinds of biographical stories that bring a figure like Voltaire alive; and they also illustrates the kinds of unconventional and irreverent qualities that younger siblings have displayed throughout history.

JB: How has your own birth order affected your personality and your life?

SULLOWAY: I was the third of four boys, but I'm a functional youngest child because my brother Brook is nine years younger than I am (and from a second marriage). For nine years, I therefore grew up without a younger sibling, and I do not think that Brook had much of an influence on my personality. But my two older brothers did have an influence on me; we were each about two and a half years apart, and there was a lot of fighting among us. I think I have a pretty typical laterborn set of personality characteristics. As someone who has existed as an academic for more than two decades without ever holding a formal job, I have had an unconventional career.

JB: Are you familiar with Judith Harris's work on nurture?

SULLOWAY: Yes, she has focused on the influence that peer groups have on children. In response to the findings by behavioral geneticists that most environmental influences are not shared by family members, she and a few other psychologists have argued that the family has only limited influence on personality. An alternative viewpoint, to which I subscribe, is that families do not represent a shared environment. Hence they influence siblings in different ways, which is not the same thing as having no influence. I believe that Harris is correct to emphasize the importance of peer groups, but she is too single-minded when she denies the importance of systematic within-family differences. Actually, the two approaches (family niche theory and peer group influences) overlap in important ways. For example, some family members are probably influenced by their peer groups more than others, and we would especially expect this to be the case for younger siblings because they are more open to experience. It appears that middle children, in particular, are the most closely identified with peer groups rather than with the family. One can perform a very simple test of this claim, as Catherine Salmon did in a recent doctoral dissertation at McMaster University. One asks people to respond 10 times to the question "Who am I?" Middle children are significantly less likely than firstborns or lastborns to answer "I am a Brockman" or "I am a Sulloway"—that is, middle children do not identify themselves by using the family label. Why is this? From a Darwinian point of view, we know that middle children are at a disadvantage—they don't have the benefit of being first, which leads to greater parental investment because firstborns are closer to the age of reproduction. The lastborn has the benefit of being the last child the parents are going to have, so parents will tend to invest heavily in this child so that it will not die in childhood. The offspring who tend to get lost in the shuffle are middle children. How do they respond? They become peer oriented. If a person is not favored within the family, it is a wise strategy to build one's bridges to other sources of support.

JB: What conclusions will a father or mother take away from your book with regard to the raising of their children?

SULLOWAY: I do not directly address the issue of childrearing in my book, although any reader can draw numerous relevant conclusions on this subject. This is an issue, however, that I do discuss in public lectures. One obvious implication of my researches is that sibling rivalry is not pathological. Many people feel that if rivalry exists among offspring, the parents must have done something wrong. This is mistaken: sibling rivalry predates the dinosaurs. Sibling competition shapes creative behavior—it's part of the process by which children sharpen their endearing little claws and get ready for life. It is a considerable relief for parents to understand this point. Secondly, parents need to understand why siblings engage in rivalry—such competition is part of the effort to feel special within the family, to feel that one is not discriminated against. Ultimately, sibling competition is all about optimizing parental investment. What each sibling wants is special time with each parent, and when parents provide such moments, it makes children happy. In fact, this is a useful bit of practical information, if parents have not already discovered it. By being different, each sibling is trying to develop a special set of interests, a special niche, causing parents to pay attention to them and to them alone.

JB: Where are you headed in your future research?

SULLOWAY: I consider the findings in Born to Rebel to be just a preliminary outline of the many problems that we are now facing trying to understand personality development. Also, the book provides only a bare introduction to understanding how we can apply Darwinian theory to understanding all of the learned adaptations of childhood. Adaptations in childhood are not just random; they occur for a purpose, and this purpose is to get one's genes into the next generation. There is a whole class of potential future studies that can be done on these issues. These studies are going to require an even stronger interface between evolutionary biology and developmental psychology. I believe this area of research is going to be a very exciting one for the future.

My own future research is going to be more psychological than historical, so that I can answer some of the questions that I could not answer using historical data. In Born to Rebel I developed statistical models that combined the predictive power of birth order, parent-offspring conflict, temperament, and other variables in explaining what historical figures actually did during times of radical social and intellectual change. We can do a far better job in this regard by working with living individuals because we can ask specific questions about developmental history—for example, the nature of strategies employed in dealing with siblings, and to what extent these strategies (and associated personality characteristics) predict adult behavior. The jump to research on living subjects is a bit like moving from a 19th-century locomotive to a 20th-century jet in terms of the sophistication that one can hopefully achieve, and few of these kinds of studies have been done.

In order to achieve the kind of understanding of families that we need to have, we require studies in which all members of the family are studied simultaneously. When psychologists wanted to study an influence such as birth order in the past, they collected data on firstborns and laterborns selected from different families. We miss too much with this approach. I'll give you an example of why we want to study individuals growing up in the same family. Suppose you are a firstborn. Your usual strategy for dominating your younger siblings would be to act like a tough-minded Clint Eastwood (who, incidentally, is a firstborn like most of the other Hollywood macho types—John Wayne, Sylvester Stallone, Bruce Willis, and all of the actors who have played "James Bond"). But suppose a firstborn happens to be shy. Shy people do not generally chose to employ strong-arm tactics—they tend to be retiring and physically timid. And this shy behavioral disposition undermines their ability to occupy the typical firstborn niche. So a shy firstborn is likely to develop a different set of strategies for dealing with siblings. Such individuals might try to keep younger siblings in their place by being moody, or by giving younger siblings who have offended them the cold shoulder. There are many other strategies that people can employ in place of strong-armed tactics. The minute one opts for one set of strategies over another, the door is opened for a younger sibling to adopt some of the strategies that are not being employed. If one is comparing two individuals from different families, one misses these kinds of —coadaptations.— It should be kept in mind that personality development takes place on a kind of chess board. The moves that one family member makes are dictated by the moves that have already been made by other family members on the same board. Extraordinary as it may seem, very few studies have been done of personality development from this perspective. From an intuitive psychological point of view—but also from a Darwinian point of view—this is the best way to study human development.

JB: Will these studies be conducted in Western countries?

SULLOWAY: Since most psychologists live in the Western world, this is where the bulk of these studies will be done. But since psychologists always love to see cross-cultural replications, we will begin to see studies done in places such as Africa or South East Asia. Eventually such studies will be done around the world, and we should definitely expect some interesting twists on the story of human development as we go from one culture to another.

JB: Last words?

SULLOWAY: I have to say that I had no idea what I was getting into when I stumbled onto the project that culminated in Born to Rebel. Looking back 26 years later, it has been one of the most interesting things I possibly could have done. I have never gotten bored trying to understand what makes human beings tick. And to have recognized, two decades into the project, that Darwinian theory was a major player in understanding individual human differences was an exciting insight as well. The mysteries of human development have been a wonderful subject to devote my life to, and I hope to stay interested in these problems, and to continue to make progress in trying to resolve them.