Sperlonga – a town halfway between Rome and Naples – has a splendid beach, and on that beach, in the summer of 1965, I met a college student from England who happened to be reading Intelligence and Experience by J. McVicker Hunt. Her helpful synopsis of that book –“a critique of the claim that intelligence is impervious to experience” – piqued my curiosity. By implication, psychology was a subject where people argued with one another. But dissent had not been seen as virtue in the grammar school from which I had just graduated. This conversation stayed with me as I made my way to Naples where, along with other student volunteers from all over Europe, I worked at the Roman site of Herculaneum … and ate pizza for the first time in my life.

In September of that same year, back in England, I became a student at the University of Sussex which in the mid sixties was a so-called “new” university and – unlike the “old” universities – prided itself on being flexible. For example, it recognized that students might not want to pursue the major for which they had been accepted. So, although I had been admitted to study English literature, I could switch if I wanted to. And that was what I did on day one. The recollection of a casual remark by my somewhat ponderous English teacher bolstered my decision to switch. Jumbo – as we pupils affectionately called him – remarked of an essay that I had written on Shelley: “You seem to be arguing with him.” By implication, if I had a genuine literary sensibility, I would not be doing anything so crass as arguing. Well, if I was not suited to English literature, maybe I would be a better fit for the fledgling discipline of psychology.

There were two hurdles to negotiate. During that period at Sussex, you could study psychology at Sussex in one of three different ways – by emphasizing a biological/experimental approach, a social psychological approach, or a developmental approach. Which to choose? Happily, although for no rational reason, I chose a developmental approach. As for hurdle number two, I telephoned my parents and told them the unexpected news. Knowing little about university life and still less about psychology, they gave me their blessing.

Despite its trendy aspirations, Sussex retained the ancient tutorial system. This meant meeting one and sometimes two tutors each week to discuss the essay you had written for them in the previous three or four days. The regime was demanding but effective. Two tutors stick in my mind. One – an elderly philosophy tutor – returned my essays covered with remarks, all in red ink. Horrified at first, I gradually acknowledged that, not only was this what I had opted for, that is, active discussion, but also that, up to a point at any rate, the red ink meant that my arguments were being taken seriously. The other tutor was a kindly psychologist. Stretching out his long legs, and puffing on his pipe, Dr. Sants embodied the contentment that unhurried rumination can offer.

The training I received in psychology at Sussex was idiosyncratic rather than rigorous: a marvelous set of lectures on Freud, delivered with great aplomb – and an impeccable Viennese accent – by Professor Marie Jahoda; no lab experiments whatsoever; a shambolic course on statistics in which I eventually learned how to do t-tests (for both matched and independent samples!); a meandering exploration of the history of psychology, starting with Germany in the nineteenth century; and last but not least, exposure to Piaget.

Somewhat bizarrely, I came to think of Piaget as an intellectual safe haven. All students at Sussex who were not studying mathematics, or the natural sciences, had to do an introductory course in philosophy. Several of my friends were smitten; they abandoned their original major and switched to philosophy. Like them, I sensed the appeal but the thought of spending weeks, months, years, reading philosophy articles in the library hoping to gain traction on the intractable was not appealing. Piaget, on the other hand, also asked epistemic questions, but he sought to answer them with empirical research. That seemed more manageable and more fun.

In any case, I was sufficiently hooked on developmental psychology, and especially on cognitive development, that I applied for various graduate programs. The University of London explained that my training in psychology was inadequate: I would need to do an interim year before being admitted to the doctoral program. Harvard took one look at my lamentable GRE (Graduate Record Exam) score in mathematics and promptly averted its gaze in horror. (In my defense, I note that my decidedly lopsided grammar school education in England meant that I had done no mathematics after the age of fifteen.) Happily, Oxford overlooked my sketchy preparation in psychology and even sketchier preparation in mathematics and offered me a place as a doctoral student to work with Peter Bryant.

Doctoral study at the Department of Experimental Psychology in Oxford in 1968 included no course work, no lab work, no statistical training, no IRB (Institutional Review Board) submission, and no pre-registration. Life was simple: you devised an experiment, carried it out, and then you did another one. Impressed by Piaget’s findings, I devised an experiment to look at infants’ perseverative errors – their tendency to search back at a former hiding place rather than the most recent – when searching for a hidden object. Peter Bryant persuaded the Oxford health authority to let me recruit babies at local baby clinics and provided me with a pilot subject – Daniel, his eleven-month-old son. Daniel made me realize that getting babies to search for hidden objects required: (1) a fascinating object to hide (a bunch of keys proved to be the ideal solution) and (2) easy-to-displace covers with which to conceal the object (hanging leather flaps did the job). But with these practical problems resolved, more challenges lay in store.

The babies I recruited appeared not to have read The Child’s Construction of Reality. To my dismay, they did not make perseverative errors. I began to wonder if I was cut out for developmental psychology; after all, I was unable to replicate a supposedly classic finding. The chair of the department, Professor Larry Weiskrantz, came to my rescue. I told him about my decidedly non-perseverative babies, and he reminded me that monkeys with lesions in the frontal cortex also manage to avoid perseverative errors as long as you let them search for a hidden peanut straightaway; but they are error-prone if obliged to wait for a few seconds. And indeed, when I tested babies with a five-second delay, I got the same pattern.

In this period, research with babies was a fairly exotic enterprise in the UK. Rudolph Schaffer in Glasgow ran a baby lab, and served as advisor to Harry McGurk who later became a good friend. But other than that, I was on my own, unencumbered by findings from other laboratories. In some ways, that was a boon. I had the feeling of making my way through some unexplored rainforest. In other ways, it was a problem. With the exception of Piaget, there was little by way of a theoretical framework and in the end, despite its pretentious title (The Development of Representation in Young Children), my thesis was mostly an empirical documentation of the infant as an information-seeker, a creature who spends a lot of time exploring objects, especially if they are novel, or visually comparing displays, especially if they are different from each other rather than identical.

Writing up the thesis was a learning experience. Having become accustomed at Sussex to writing regularly, the act itself came easily. But as an undergraduate always having to deal with the next essay, I had given little thought to re-reading, much less re-writing. Peter Bryant put an end to that. Privately, I ended up dubbing Peter “the Ernest Hemingway of Developmental Psychology.” Certainly, my thesis drafts were trenchantly reviewed in that spirit. Unnecessary subordinate clauses, redundant adjectives, rambling asides … all were excised. It was a crash course in how to write with clarity and brevity.

One other lesson was delivered in a more incidental fashion. Some of my fellow graduate students were studying anthropology. They seemed more venturesome than the psychology students that I worked alongside: anthropology students went to faraway places, learned a new language (often unwritten), came down with mysterious illnesses, and returned to England with tales far more exciting than any I could muster from my afternoons in Oxford baby clinics. More importantly, they were dubious about the idea, deeply embedded in my notion of development, that entities, such as historical periods, cultures, or stages of cognitive development, can be tidily ordered into a sequence running from the more primitive to the more advanced. It was my first glimpse of an alternative to the meliorism that pervades much of developmental psychology and indeed much of social science.

In 1971, toward the end of my doctoral studies, Jerome Bruner visited Oxford, with the prospect in mind of joining the psychology department. I managed to buttonhole him during his visit and was excited to be offered a two-year postdoctoral fellowship, the first year spent at what would prove to be the final year of the Center for Cognitive Studies at Harvard, and the second back at Oxford in the wake of Bruner’s migration there. The postdoc was a wonderful opportunity to write up my doctoral work for publication, to work with Aidan Macfarlane on newborn babies and to watch Bruner’s theoretical panache doing battle with Oxford scientism.Footnote ¹

My first teaching position was at the University of Lancaster, another of the “new” universities. Indeed, it was so new that it lacked a psychology department altogether. Professor Phil Levy together with three newly appointed lecturers (me included) had the fun task of deciding what a psychology degree should include. I continued my research with babies but soon realized that being a full-time lecturer and running a baby lab was a tough combination. Testing children in schools was much easier and as enjoyable, not least because you could have a conversation with your participants. In retrospect, I was grateful for the experience of working with babies and enjoyed watching that field flourish, but I was also happy to expand my horizons by working with older children.

One Saturday morning over coffee in Lancaster’s covered market, I spotted a newspaper advertisement for a developmental psychologist at the Free University in Amsterdam. Looking up from the newspaper, I saw a Van Gogh poster on the café wall, obviously a sign that I should apply. Indeed, some months later, in the summer of 1976, I said farewell to my Lancaster colleagues, filled my ageing Vauxhall Viva with a jumble of books, clothes, and household items, and took the ferry across the North Sea. I soon found an apartment to rent (within earshot of the sea lions at the Amsterdam Zoo), bought a bicycle for my commute to the department (situated beside the beautiful Vondelpark), and started to learn Dutch.

Many of the students specializing in developmental psychology at the Free University had clinical interests. Indeed, the department was affiliated with a residential institute – the Paedologisch Instituut – that offered treatment to children with a variety of emotional difficulties. My lectures struck some of these students as completely impractical. They had a point: sorting out exactly why babies make perseverative search errors, to take one example, was unlikely to be of any immediate remedial help. Faced with student dissatisfaction, I decided to embark on a fresh line of research.

As a newly appointed “lector” I was expected to give an inaugural lecture. Combining various strands in the literature, I chose to talk about: Het Kind als Psycholoog (The Child as Psychologist) and came to the realization that research on children’s understanding of mental states could be extended to emotion. Together with two wonderful colleagues, Tjeert Olthof and Mark Meerum Terwogt, I designed a questionnaire to probe children’s developing ideas about emotion. A developmental pattern emerged: younger children focused mainly on the overt, situational aspects of emotion, when and where particular emotions were provoked; older children were also attentive to its mental aspects. This initial study led onto further studies, including some with a cross-cultural component. We recruited children in the UK and the Netherlands, of course, but also in China, Japan, and a remote area of Nepal.

Eventually, Francisco Pons, a Swiss colleague with much more psychometric acumen than me, was able to build on these findings to create the Test of Emotion Comprehension (TEC), a tool that can diagnose the level of emotion understanding shown by children from three to eleven years of age. By now, the TEC has been translated into over twenty languages and has uncovered impressive similarity in the way that children in radically different cultural settings conceptualize emotion. For example, few four-year-olds but most six-year-olds realize that expressed emotion and felt emotion can diverge; few four-year-olds but most six-year-olds realize that what you feel depends on the beliefs you bring to that situation and not on its objective features.

After three productive and happy years in Amsterdam, I returned to the UK, taught for a brief period at the London School of Economics, and then for a much longer period – two decades – at Oxford. I pursued two lines of research at Oxford; they were relatively distinct but with subterranean connections. First, I continued to think about children’s understanding of emotion, spurred on, in part, by the burgeoning research on children’s understanding of mind. Three books emerged on that broad theme: two co-edited volumes – Developing Theories of Mind and Children’s Understanding of Emotions – plus a book of my own – Children and Emotion – that drew together much of the research I had done on children’s understanding of emotion.

The other line of research focused on the development of the imagination. The subterranean connection to the work on emotion? In the final chapter of Children and Emotion, I discussed autism and its impact on children’s understanding of emotion. Leo Kanner’s classic account of children with autism also underlined their difficulties with pretend play. By implication, an early limitation to the imagination implied a cascade of later challenges. But I knew from slogging through Play, Dreams, and Imitation that Piaget, for one, did not hold children’s imagination in high esteem; he treated it as an early cul de sac rather than an Appian Way leading to intellectual enlightenment. So, were children’s early imaginings a help or a hindrance to subsequent cognitive progress? With that question in mind, I took a closer look at children’s pretend play, especially their collaborative pretend play. Greatly helped by Bob Kavanaugh, a visitor from Williams College and in collaboration with various students, I also looked at children’s counterfactual thinking, their reasoning about imaginary worlds, their diagnoses of rule violation, as well as their more metaphysical doubts and queries.

Oxford also provided a great opportunity to think about development in infancy again: my colleague, Kim Plunkett, and I served as advisors to several doctoral students. Kim had created a state-of-the-art baby lab where babies’ eye movements could be filmed, analyzed, and quantified. I occasionally regaled the students with tales from my (to them) distant past, when babies’ looking times were recorded on an event recorder, and my evenings were spent unfurling the resultant roll of recording paper to measure looking times with the help of a ruler.

Toward the end of my stay in Oxford, I co-edited a book on children’s imagination – Imagining the Impossible – and synthesized my own empirical work in a book, The Work of the Imagination. The final chapter of that book also turned into a stepping stone for further research. Reviewing the various chapters, I realized that I had not said much about the relation between language and the imagination. In retrospect, it seemed obvious that children’s imagination feeds on language, more specifically, on what they are told, especially about places, events, and phenomena that they have not seen, or cannot see for themselves. That final chapter was couched primarily in terms of a fairly wild speculation about the evolutionary connection between the emergence of the imagination and of language in homo sapiens. That evolutionary connection was scarcely open to empirical investigation but, as later research began to show, it could definitely be studied in a developmental analysis of how children learn about things they have not seen, but can imagine, based on what other people tell them.

Moving with my wife and our three young sons to Harvard in the summer of 2001 was a good deal more complicated than moving myself to Amsterdam. A Vauxhall Viva would not do the job. The shipping container that we needed for our household belongings (including books, filing cabinets, and toys) was about forty feet long. Even then, it was a challenge to get everything in. Seeing our bicycles loaded and the doors finally shut tight was an emotional moment. Based on my one-year postdoc at the Center for Cognitive Studies, I knew that the USA was separated from the UK by more than a common language. A month after our arrival, the USA faced the trauma of 9/11, whereupon the Stars and Stripes, defiantly hoisted even in Massachusetts, became a frequent reminder that we were living on American soil.

Still, it did not take too long for continuities and connections to be established. Melissa Koenig, who had joined me as a postdoc, and several welcome visitors from Europe – Rita Astuti, Fabrice Clément, and Francisco Pons – contributed to an emerging theme: children’s learning from the testimony of other people. A steady stream of experimental papers highlighted the way that preschoolers select among their informants – based on who is familiar rather than unfamiliar, knowledgeable rather than ignorant, and from their ingroup rather than an outgroup. A theoretical paper with Melissa underlined the scope of children’s reliance on testimony, especially with respect to hard-to-observe phenomena from the domains of both religion and science. Subsequent collaboration with Kathleen Corriveau led in due course to an ambitious study of how children learn about religion and science in three markedly different cultural settings: the US, Iran, and China.

One striking conclusion is that in all three cultures, confidence in the existence of ordinarily invisible scientific phenomena – germs or oxygen, for example – tends to be firmer than confidence in the existence of ordinarily invisible religious phenomena, God or Heaven, for example. I continue to puzzle about the explanation and significance of that finding. Meantime, the intertwined theme of testimony and imagination yielded two books, one focused on learning from testimony – Trusting What You’re Told: How Children Learn from Others – and one providing an update on the development of the imagination – Children’s Imagination.

Teaching at the Harvard Graduate School of Education proved to be both challenging and enjoyable. Many of the master’s students that I taught were taking their first course in psychology and as the years went on, an increasing number came from overseas. Over the course of more than twenty years, I gradually honed an introductory course intended to convey a sense of the big questions in the field – along with tentative answers – and given the diverse audience, I included a good deal of comparative and cross-cultural material. After endless tinkering with the course notes that I prepared, I eventually organized them into a book: Child Psychology in Twelve Questions, a relatively slim and somewhat idiosyncratic introduction to the field, but one that might serve as an enjoyable alternative to the blockbuster textbooks I had managed to avoid in my own student days. My fantasy is that on a bus from New York to Boston, or a train from Xi’an to Beijing, it will catch the eye of a young traveler and pique their curiosity.

In his reflections on Aspects of the Novel, the English novelist, E. M. Forster, made an influential distinction between flat and rounded characters: flat characters are useful for authors because they “never need reintroducing, never run away, have not to be watched for development, and provide their own atmosphere – little luminous disks of a pre-arranged size, pushed hither and thither across the void or between the stars; most satisfactory.” Flat children are also useful for developmental psychologists – they do things on time and confirm our cherished theories and expectations. Accordingly, we developmentalists are tempted – for good scientific reasons – to look for flat children and to write convincing experimental reports about how they behave.

But round children have their virtues, and we need to cherish them too. As Forster notes: “the test of a round character is whether it is capable of surprising in a convincing way.” If I think about ongoing research in developmental psychology, I see all sorts of round children who disconcert us – children whose utterances don’t fit together with their actions, children who stubbornly resist changing their ideas but are also amazingly receptive to adults’ assertions (true, false, and nonsensical), smart infants who grow into naive children, and so forth. There is a lot to discuss.

No memoir is complete without a piece of sage advice. Mine would be to not take contemporary findings too seriously – most of them will prove to be blinkered, partial, and quaint as the years pass. On the other hand, I would also counsel against caution – there’s a lot to be said for joyful commitment.