Diving deep into my archives I recovered an old file in which I describe how I got to study prosocial behavior. The file is dated 21 February 2000. I reproduce it below – and here’s a pdf.
The Autobiography of A Problem
In an otherwise useless book, Deutscher (1973) employed the distinction between the logic of a problem as portrayed in scientific publications and the autobiographical ‘logic’ of its discovery and treatment. The distinction is familiar to any social scientist. The way we present our results is seldom the same as the path we followed to reach them. Following current conventions in social science, I wrote the articles in my dissertation as a series of questions, that seemed worthy of giving attention given the state of the knowledge available in the literature. I tried to make progress where substantial gaps were obvious and my capacities allowed me to. But how did I get to my account of ‘the current state of knowledge’? It is the autobiography of my research problems that I want to report now.
Looking back on the development of my PhD-project, I can see a continuous line of development since the beginning of my education as a sociologist. I studied sociology at Nijmegen University, The Netherlands. Although the Institute was officially of Catholic signature, its ecclesiastical undertones were hardly visible to me. The kind of sociology I was taught was mainly empirical macro-sociology. Questions on stratification and mobility, racial attitudes, voting, homogeneity of marriages inspired most of the research going on at the time. The catholic origins of the Institute were only visible in repeated, cross-sectional surveys measuring values, attitudes and their behavioral correlates such as voting and religious denomination. In line with my interest in philosophical questions on scientific methods, religious beliefs and even metaphysics, I wrote a master thesis on the beliefs of clients of a ‘New Age’-center – which was quite a new phenomenon at the time (Bekkers, 1997).
A Different World
After my graduation I found a PhD-position at the Interuniversity Center for Social Science Theory and Methodology (ICS), Utrecht University. The project consisted of half a page of ideas on the sociology of consumption (ICS, 1997), and was supervised by Harry Ganzeboom, a former teacher from my first year of sociology in Nijmegen, and Nan Dirk de Graaf, who had been a lecturer in Nijmegen all along. In essence, the aim was to identify a new pattern of consumption, associated with a post-materialist value orientation (Inglehart, 1977). The project had some connections with my master thesis, because ‘New Age’-clients were high in post-materialism. But this was not the direction that I followed. As I entered the ICS, I discovered that the research school, of which my sociology-teachers were a part, was based on a combination of rational choice theory and Popperian methodology. I knew Popper’s philosophy of science and was very sympathetic to it, but with rational choice theory I was rather unfamiliar. At the ICS I learned the principles of rational choice theory. My first serious meeting with it was unsettling, because I thought that cynical assumptions on human nature were central to its explanatory power. Although I didn’t like the assumption of material self-interest, I tried to work with it. But I also came to see its shortcomings. According to a simple version of rational choice theory, rational actors in social dilemmas should not contribute to public goods. Because I saw the consumption of ‘environment-friendly’ products as a part of the post-materialist lifestyle, I tried to explain this phenomenon with reference to self-interest only. At first, this was an intellectual problem. Many rational choice-theorists I read apparently saw it as a challenge to explain away ‘seemingly’ altruistic acts as the ultimate result of self-interest. But many experiments show that people do cooperate, even in ‘one shot’-dilemmas, which are specifically structured to induce ‘rational’, selfish behavior. How could this be possible? Then the question caught me: why do people do things that are no good for themselves anyway?
After I had got acquainted with the ‘participation paradox’ in theory, I discovered more and more examples of unselfish behavior in real life. There were even much stronger examples in real life than in the world of experimental gaming: donation of money to charity, donation of bodily organs, donation of blood, and so on. These examples made my questions on the participation paradox even more urgent. From personal experience I got the suspicion that there had to be important individual differences in altruism. Maybe the steady percentage of people that keep on playing ‘cooperation’ in these games is a group of ‘hard core altruists’.
From ‘Cooperation’ to ‘Prosocial Behavior’
At this point, the project took a decisive turn. The central idea of the project shifted from the identification of a post-materialist lifestyle to the explanation of ‘altruistic’ behaviors such as buying environmentally friendly products. It was one of my supervisors, Harry Ganzeboom, who suggested that I take a look into social psychology. He remembered the term ‘prosocial behavior’ as precisely the category of behaviors that I was interested in. A blind search in the library formed the basis of all the work I did from that point onwards. As a kind of naïve mountaineer I started reading the literature. I didn’t know that so much work had already been done, and I was also unfamiliar with social psychology altogether. I did not only read substantive papers, but also dived into methodological issues such as the debate on consistency from the seventies.
I made some important discoveries that were surprising for a sociologist with an interest in questions on philosophy of science. I discovered that there are large differences in method and theory between social psychological studies, game theoretical approaches and sociological treatments of the subject. The substantive papers gave me an enormous list of independent variables to consider in an explanatory model, including personality characteristics. To many sociologists, personality is non-existent, or – at best – not relevant (read: not interesting). In the meantime, I sharpened my dependent variable, because I discovered that the term ‘prosocial behavior’ was too broad to deal with in unity. I discovered that I was interested in understanding how prosocial behavior in the absence of clearly material interests for the beneficiary him/herself came about. I selected dependent variables where situational factors cannot fully explain prosocial behavior, because they lack the altruistic motive: to make the other better off. When situational factors cannot do the job, we have to look on the person-side. I concluded that there must be something in the heads of people playing these prisoner’s dilemma games that makes them cooperate, even when the situation is specifically designed to make them defect. From the social psychological literature I learned that there were indeed important individual differences in game behavior. As an explanation, the obvious candidate in personality was some individual difference in altruism. The rational choice perspective I had become familiar with, made me think of altruism as having a preference for outcomes of other players. At first it was the Margolis (1982)-model that fit into this picture. Reading his book, however, I wondered how people decide (or: how it is determined) whether they value outcomes for others. The concept of social orientations that I met in the work of Liebrand and Van Lange appealed very much to me, because this model was more clear and simple.
However, sociologists are not very fond of introducing individual differences out of the blue. My supervisors kept asking me why these ‘social orientations’ were so important. I had no real answer at that time, but I came back to the question later on. In the meantime, I had become acquainted with the concept of empathy through the experimental work of Daniel Batson (1991). I didn’t like his general question (“Does real altruism exist?”) and I also didn’t like his situational interpretation of empathy. But the concept of empathy in itself stroke me as an important clarification of the simple idea of altruism inherent in social orientations. Empathy seemed to be an interpretation of why people care for others. Looking for an individual difference measure of empathy, which I found in the work of Mark Davis (1994), I also came across a series of articles by Nancy Eisenberg (Eisenberg, 1982; Eisenberg & Miller, 1987; Eisenberg et al., 1989; Eisenberg & Fabes, 1994). As a developmental psychologist, she had also been active in the field of socialization and parenting. This research opened up a new perspective on the old question of my supervisors why social orientations were so important. I reasoned that parenting styles are not randomly distributed among the population, which implies that conventional sociological variables such as education, income, and social status had to be of importance for individual differences in prosocial dispositions. It was also probably through her work that I discovered that a strong debate had been going on in social psychology in the late 1970s and early 1980s between supporters of dispositional and situational explanations of social behavior: the ‘consistency controversy’. Here I found another enormous amount of material, that needed delimitation in order not to get lost. I felt that the ‘interactionist’ perspective that emerged from the consistency debate could be an important candidate to integrate the different perspectives of social psychology, sociology, and rational choice theory.
Selling the Project: What’s New?
The next step was the formulation of a research problem that contained new elements. While I was getting to know the literature, I had mainly been interested in what had been discovered previously, and not so much in what had not been dealt with. Repeating what others have done before you creates clarity and order, but is not very useful to get your PhD-project funded. I had to come up with some strong points on which I could improve upon the existing body of research. I identified conceptual problems: prosocial behavior is often considered to be one single category of behavior, while this is not the case. I saw some awareness of this problem glimmering between the lines, but not a full realization of it. That’s why I took a fresh look at the literature that I had been dealing with for so long, with a new question in mind: how can I distinguish between different kinds of prosocial behavior to understand why so many variables are connected with it, and why they have different kinds effects in different studies? I divided the unified category of prosocial behavior in different dimensions, and used insights from the consistency debate to understand differential effects of person-and situation-variables. Et voila: research question 1 & 2.
My third research question emerged from an imaginary discussion I held between a sociologist and a social psychologist. The sociologist asks: ‘Why are individual differences important? They may add some explained variance to your regression model, but they do not alter the effects of other independent variables.’ The social psychologist started talking about interaction-effects, but was interrupted by the sociologist with the argument that inclusion of interaction-effects does not necessarily change the magnitude of main effects. Then a pair of new discussion partners entered the arena: the classical sociologist, and the developmental psychologist. The former asked the rational choice theorist: ‘Do you really think that individual differences in altruism are normally distributed among the population?’ The developmental psychologist added: ‘Of course not. Since we took over socialization research from sociology, we have discovered much of the origins of prosocial dispositions.’ The classical sociologist corrected this view: ‘But you neglected many of the truly sociological variables…’ This imaginary discussion was the beginning of my third research question, which I developed more and more later on.
Of course, there was the occasional disappointment during my exploration that others had already published my ‘revolutionary’ ideas years before I came to think of them and wrote them down. An example is the idea to divide prosocial behavior into theoretically meaningful categories of behavior, in order to predict differential personality effects on helping. I put the idea in my NWO-proposal (submitted in may 1998), but found it back in January, 2000, in Staub’s extensive volume (1978, vol. 1, p. 122-123) when ploughing through the pile of literature that was simply too much when writing the proposal.
Another example was the vignette-part of the study. When tracing the literature on person-situation-interactions in January, 1999, I recognized the idea to test interaction-effects with hypothetical scenarios that I put in the NWO-proposal in may 1998 in Carlo et al. (1991). And they were already late, since I found out in January 2000 that the very same idea is presented in the discussion of a taxonomy of helping situations in Pearce & Amato (1982, p. 75).
Batson, C. Daniel (1991). The Altruism Question: Toward a Social-Psychological Answer. Hillsdale, NJ: Lawrence Erlsbaum Associates.
Bekkers, René (1997). New Age in sociologisch perspectief. Nijmegen: Undergraduate thesis.
Carlo, Gustavo, Eisenberg, Nancy, Troyer, Debra, Switzer, Galen, & Speer, Anna L. (1991). The Altruistic Personality: In What Contexts Is It Apparent? Journal of Personality and Social Psychology, 61: 450-458.
Davis, Mark H. (1994). Empathy: A Social Psychological Approach. Madison, WI: Brown & Benchmark.
Deutscher, Irwin (1973). What we say/what we do: sentiments & acts. Glenview, Ill.: Scott, Foresman and Company.
Eisenberg, Nancy & Fabes, Richard (1994). Prosocial Development. In Damon, William & Eisenberg, Nancy (Eds.). Handbook of Child Psychology, (Fifth Edition), New York (etc.): John Wiley & Sons, Vol. 3, pp. 701-778.
Eisenberg, Nancy & Miller, Paul A. (1987). Empathy, sympathy, and altruism: empirical and conceptual links. Empathy and its development. Cambridge: Cambridge University Press, pp. 292-316.
Eisenberg, Nancy (1982). The Development of Prosocial Behavior. New York: Academic Press.
Eisenberg, Nancy, Miller, Paul A., Schaller, Mark, Fabes, Richard A., Fultz, Jim, Shell, Rita & Shea, Cindy L. (1989). The Role of Sympathy and Altruistic Personality Traits in Helping: A Reexamination. Journal of Personality, 57: 41-67.
ICS (1997). Postgraduate Program in Social Science Theory and Methodology 1997. Utrecht: ICS.
Inglehart, Ronald (1977). The Silent Revolution: Changing Values and Political Styles Among Western Publics. Princeton: Princeton University Press.
Margolis, Howard (1982). Selfishness, Altruism and Rationality. Cambridge: Cambridge University Press
Pearce, Philip L. & Amato, Paul R. (1980). A Taxonomy of Helping: A Multidimensional Scaling Analysis. Social Psychology Quarterly, 43: 363-371.
Staub, Ervin (1978). Positive Social Behavior and Morality. New York (etc.): Academic Press.