William Brustein 04/06/2000
The Logic of Evil

Overview

Q: Professor Brustein, could you begin by providing an overview of your project?

A: Right. Well, the project is a study of Nazi party membership, in the years before the Nazi party came to power. So it really deals with 1925-1933. And the genesis of the project goes really back to my childhood. I had lost relatives in the Holocaust and I was born after the war but I remember many discussions around the dinner table and how questions of "how could this have happened…Nazis came to power…the craziness of Hitler," so I became interested in the issue of Nazism from early on. But as I thought about doing this project, trying to understand why people joined the Nazi party and who they were, I waited until I was at least tenured because I knew that this project was so ambitious, would take many many years. In terms of the major research questions, I mean, they're pretty simple. I wanted to know who actually joined the party, what kinds of people, and more importantly, why. And the question of why has been very difficult because the period I'm dealing, this is pre-survey period, and so there have been very little and very non-systematic work done on motivations for joining the Nazi party when people had a choice. And so the questions of who and why are the principle research questions. The major publication from the research is a book titled The Logic of Evil: The Social Origins of the Nazi Party 1925-1933, and that was published by Yale University Press in 1996.

Theory

Q: What was the state of the literature on the subject of membership in the Nazi party at the time you began your project?

A: Let me talk a little bit about the literature, in terms of the theoretical literature, and then a bit about, if I may, about the problems with the empirical issues. In terms of the literature, there have been a number of studies that, during the Nazi period, that emphasized what I refer to as a mass society, a type of Durkheimian look at the rise of the Nazi party, and as well as the kinds of people who were attracted to it. And the emphasis there was on people suffering from Durkiem's anomie, or normlessness, that the people who were attracted to the party were people who were the dregs of society, and this certainly comes out in some of Hannah Arendt's work on totalitarianism, that the dregs of society were the ones who were swooped up by the demagogic messages of Hitler, and these were people who wanted to get even, with whether it was the Jews or with the middle and upper classes. And so much of the early literature emphasized the irrational appeal of Nazism as well as the types of people who joined, people being outside of major social networks. A second thrust in the literature, and one that develops a bit later and you find it in works of like Theodore Geiger and then in works of Seymour Martin Lipset, was the class analysis of Nazism. Lipset's work is perhaps the most important here, because for Lipset, he saw that different social classes, in times of particularly trauma or economic crisis, that different segments of class tend to associate with forms of political extremism. The working class he saw moving in times of crisis towards Communism, the lower middle class he saw moving towards Fascism. The understanding coming out of the works of people like Lipset, this class analysis, was that Nazism, as another variant of fascism, was a reaction of the lower middle class towards modernization. And then there were variants of that, Walter Burnham, a political scientist, had argued that it wasn't just old members of the lower middle class, but it was those members of the lower middle class who were not immunized against the Nazi virus. That is, members weren't sheltered by major organizations or networks, and in particular he said that people who didn't have the German Catholic Church's social organization or the labor unions in Germany, it was those people in the lower middle class particularly who had a greater affinity towards Nazism, that void gave them a new church. So that was another thrust in the literature. And then a third, which I found to be fairly atheoritical, was one that the Nazi party was a catch-all party of all different social classes, and this was driven a lot by empirical studies, particularly of aggregate voting returns from major Weimar elections, and that what scholars had found, these were people like Richard Hamilton, Jürgen Falter, was that people who voted for the Nazis generally came from the upper class, they came from the middle class, they came from the lower classes, but what drove them to the party was just dissatisfaction with the system. But I found that to be very lacking in terms of theory; it was just trying to explain certain results, empirical results.

Q: So what was your point of departure?

A: Well mine was that I have been for years involved in rational choice. And I've seen how rational choice had been used to explain the attraction in other contexts for certain political parties, and that what I saw in the literature was an absence of looking at the decision to join the party or support the party, an absence of looking at it as an individual decision, as weighing costs and benefits. And I felt that if rational choice could be used to explain why people join or don't join political parties, or select between party A and B in other contexts, why not see how far it could go with respect to this prototype of the most evil political party perhaps of or social movement of the 20th century. So I wanted to see what I could do with a rational choice model. But I also wanted, from an empirical basis, I wanted to study this from an individual level, because the theory is a theory that starts at the individual level and that heretofore most studies had been aggregate level studies and were familiar with problems of ecological fallacy. But to do that necessitated getting access to individual membership records, and that's the story of getting into the Berlin document center and basically how this project was designed.

Q: Can you describe your theory for us? How does it fit into the state of sociological theory? What were the primary propositions you were interested in exploring?

A: Sure. With respect to the theory…rational choice is not a very well excepted theory in sociology. In fact, many argue, and I talk about this often in the theory classes I teach, is that sociology, which is anchored more in Durkheimian and Weberian sociology, developed as a reaction to utilitarianism to the Scottish moralists and the English utilitarians, in those works of Jeremy Bentham and John Stewart Mill, there was certainly an emphasis on individuals as a unit of analysis, the fact that individuals are self-interested purposeful actors, and that norms played a minor role in influencing decisions in individuals took. So sociology, very normative, developed a reaction, and sociology still has problems with rational choice; it's a very different situation obviously, in political science and in economics. But in my use of rational choice, what I generally have argued, and I've done this in works previous to the logic of evil, was that when it comes to political decision-making, that there are two levels. In the theory that I developed, there are two levels or stages, explaining why people join the Nazi party. One level is the what I call interest compatibility, that is, that individuals are either attracted or not attracted to parties when there are choices, when there are other alternatives out there, based on their perception of what their self-interests are, and in my use of the model, I emphasize material self-interest. So if party A is offering a program that is at variance with the material interest or at least the perception of material interest of an individual, it's very likely that that individual won't be attracted, that individual will look for other, more viable alternatives. But that, in explaining the joining of an extremist movement such as the Nazi party, interest compatibility is necessary, but certainly not sufficient because the rational actor would obviously free ride, that is that the rational actor would perhaps maybe vote for the party that fit closest to his or her consolation of material interests, but wouldn't make the extra effort of joining the party because of the cost involved. Particularly when you are talking about an extremist party, such as the Nazi party. There was the cost of time commitment to party activities, wearing the uniform, and who knows, in those early years you could have been a target of people who obviously, and many were, who didn't like the party. There was the cost of monetary contribution, there were dues that had to be paid. And there was the cost of being ostracized by your friends, your family, your group, or even losing your job if your employer didn't like it. So the rational actor would say "yeah, with secret ballots I'll vote for the party, but I'm not going to make the next step, and join the party." So my model had to somehow address the issue of how you get people who have an interest to take the next step and overcome free-riding anxieties. And what I argued is that one has to look at selective incentives and disincentives, that if the outcome is a public good that everyone is entitled to if the party comes to power, then the rational individual will free ride, because why make the extra effort if I'm going to get it anyways. So, my model says well what selective incentives are there for people to go beyond just voting to the next step of joining? Then again, I look also at disincentives too, because even if there are selective incentives that only joiners get but no one else is entitled to; but if the disincentives are so high, that is that, again you could be fined, you could lose your job, that there's a strong likelihood that you would be punished for joining, that they could, in a sense, outweigh the selective incentives. So the model needs to take into account selective incentives and disincentives. And what I argue is that to the extent that the disincentives could be lowered so that the act of joining becomes very low cost, you should find more of those people who have that interest compatibility joining. And in the same way, as the selective incentives grow, increase, you should have more of those people who would, under normal circumstances free-ride, making the next step of joining. So it's a two-stage model.

Design

Q: How do you take that theoretical structure and translate it into a set of empirical propositions to study?

A: What I did there is I argued that, well, based on the material interest of different groups of individuals… in fact what I did was I broke down classes into certain strata, for example, looking at workers; I looked at 21 different branches of the working class, in terms of where people were employed, whether they were employed in the chemical industry or in woodworking or construction. And, what I did was I broke down into certain categories what their particular consolations of material interest would be on a number of important issues that were being discussed during the Weimar period. And then I did a very elementary form of content analysis on the different political programs of the nine major political parties in Germany. And look at it in terms then of, well, this group of workers or this group of, lets say, lower middle class should be more likely to support party A or party B. In particular I was interested in the Nazi party. And I said, we should expect then, if the model, at least at the first level, is accurate, that workers with these kinds of interests, should be over-represented in the Nazi party, versus let's say workers in this category, who's material interests do not fit very well with the Nazi party program, but fit better with another political party's program. So at the first stage I tested that way, I took each class, broke it down into constituent parts, made predictions about what we should expect to find in terms of membership: over-representation or under-representation.

Q: Now, you're not able to conduct a survey of individuals and so instead studied the behavior of groups but that raised methodilocigal questions of its own, such as the ecological fallicy and a set of problems associated with aggregate analysis. How did you deal with those issues of aggregate analysis?

A: Sure. Well, Steve, let me now, if I can, talk a little bit about the data, because gets up to one of the key issues. Before I embarked upon the empirical part of the project, I communicated with a number of historians and political scientists who had studied the Nazi party, and the consensus was that if you wanted to do the kind of study that I had envisioned, that somehow you would have to get access to the Nazi party's master file, which holds the 11 million official membership records of everyone who was ever in the party. What they're talking about is everyone who was ever in the party from 1925-1945. I have to say something about 1925, obviously the party started earlier, Hitler joins in 1920, it really gets going in 1920, but the party collapses once Hitler is involved in the Putsch in November of 1923, and it’s in 1925 in February that the party is reconstituted, and that the membership numbers are given out. So the records really date from the reconstitution of the party, when it begins to become a mass party in 1925. So the consensus was that if you wanted to do the kind of study that I envisioned you’d have to get access to those 11 million records. And that was no easy thing because the 11 million records were captured by the U.S. army, they were held in Munich at the end of the war. And they were brought to Berlin, set up at the Berlin Document Central, which was a former Gestapo telephone-monitoring center, so it had a lot of the security apparatus that would be needed. And there were problems getting access to these records because of the confidential nature, and this was a time particularly in the late 1980s when a number of records had been stolen, not a large number, but particularly of people who were in West German government as well as in business, they had paid a lot of money it seemed to get their cards removed. This was a time with the Waldheim controversy, so there was a lot of anxiety about letting scholars in. But I got, once I realized that to do the kind of study I needed to do, it’s a long story that I won’t go into it now; I got permission, I got in there, I recruited a research team and had my research team from the University of Minnesota work along with the leading scholar in Germany on Nazism, that's Jürgen Falter, and he had his research team. So the two research teams in 1989 worked together and collaborated. Now the primary data are from the cards, and there are 11 million cards, and there are many many problems that you envision trying to do this kind of research because the cards are listed alphabetically by surname. So its not as if you can go in there and say "oh, I’ll look at this box" and there are all these file drawers there, "Aa" starts with the first surname, and maybe I’ll take a few C’s and a few D’s. The problem is that last names, surnames, cluster regionally in Germany, so if you’re a Schmidt, basically Schmidts come from northern Germany. So if you’re going to do a national sample you got to be very careful, and there have been some preliminary studies done years ago which methodologically are very questionable because they chose a few letters. So Jürgen Falter, a very well known methodologist, worked with me and we designed a sampling strategy. We spent six months there and the sampling strategy was a disproportionate sampling strategy, where based on some preliminary weighting of the likelihood of a card showing up because more people joined after 1930, between 30 and 33, than joined between 25 and 30. We developed a weighting scheme, went in, and our sample consists of 42,000 cards, which provides us sufficient observations for a stratified sample for the years, as well as a complete national sample. So much of the information is at the individual level. And as a sociologist, the key thing that I was interested in was occupational, because I’m trying to make these arguments about within class. And it took us years to finally take all of the occupations and put them in a manageable form to work with. But what we did here, I brought over a German group that had been trained by Falter, to the University of Minnesota to work with my group, and it was funded by the National Science Foundation, using the 1925 and the 1933 German censuses. We needed people who were totally fluent, obviously brought up in the culture, in the languages, because we needed to take the occupations we found and place them in the appropriate class or status category. And this took years of painstaking, careful work to do.

Measurement

Q: So the occupation of individuals was noted on the Nazi membership cards?

A: Yes, Beruf, career, occupation, that was the key thing for me. There were other things on the cards: marital status, gender, where you joined the party, if you left the party. We were surprised to find so many people who quit the party in the early years; if you re-entered the party, and then there was a spot at the bottom of the card: comments about the individual, and very interesting things you’ll learn how the Ortsleider would write things about "oh, this person did this or that," and it was the ongoing record of that individual, because on some of them you’d have comments of that "this person died on the eastern front," etc. So we had that, but at the same time, those were individual level data. But then, my colleague Falter had spent 15 years putting together what’s called the Weimar County Data Collection, and that data collection was more of an aggregate data collection. And what he had done, it’s not only counties but it’s communities, for all the counties and most every community in Germany during the Weimar period, he and his team over 15 years have collected contextual data. So once we had our individual and we could position our individual in a community, in a village, we now had over 800 different attributes of that village: its former political history, how it had voted that community in elections, the religious makeup of that community, things like unemployment levels throughout years of that community. So that we had tremendous information on contextual effects, so that we for each of our observations, our members, we could put that person into a context.

Q: How could you make inferences about the nature of the occupational membership of Nazi party members? Did you have to compare Nazi membership to more general census data?

A: Yes, and indeed, that’s what we did. The major comparisons are the membership records, the membership sample compared to the general German population. We used the 1925 census rather than 33; our study goes from 25 to 33, because it’s a better census, it asks a lot more questions to our study than the 33 study did. But yes, the base population is the census population of adult Germans, because to join the party you had to be 18 years or older, and thus we took from the census the active population that was 18 years and older.

Q: Did you make any effort to compare the membership of the Nazi party with the memberships of other parties?

A: Wonderful question, Steve. That was something we wanted to do from the outset. But the problem is that there are many things you dream of and you just wished there were data available. Those data were not available on party membership. They’ve been destroyed; they were never put together in terms of any type of national sample. You could get hodge-podge or little regional studies done on the Communist party or the Social Democratic party, but it was because of the key interest in Nazism that these records were kept alive and brought to Berlin. So that was something that if we could have done, would have certainly made the study stronger, our findings, I think, more impressive.

Data Analysis

Q: Could you describe the nature of the statistical analysis you did in comparing Nazi members with the census data?

A: Most of it is pretty rudimentary. There were a lot of bivariate correlations and multiple-regression analysis. But most of it I think one would say is more descriptive; I made certain propositions about what we should expect for example with women. That was a key finding, that at first went against what I had predicted. But that for example, would you expect to find more married women or single women during the Nazi party, given the Nazi party’s position on women in the workplace? And so I made predictions. So what we show is basic, with tables, basic findings of the percentage of our sample who are single women, married women, compared to what we find in the general German population of women. So they’re simple, really simple statistics.

Q: Was your sample so large that you generally did not worry about statistical significance, or it was not an issue?

A: Yeah, it was not an issue because the sample was so large and it’s right from the population.

Interpretation

Q: What were the major findings of your study?

A: Well, I think one of our major findings is that heretofore, most people who had studied the Nazi party had argued that it was lower middle class, middle class predominantly. And that the working class in Germany, blue-collar working class, had very little affinity toward the Nazi party because it was seen as a proto-capitalist type party, it was seen as a bourgeois party, and that the working class had the social democratic party, and the communist party to join or to support. And at most, studies had given maybe one quarter of Nazi party membership, what would be working class. In particular, this came out of most of the aggregate level voting analysis. In every year, we found over 40 percent of our membership came from blue-collar working class. Very different, and part of it is that much of the literature has been written by people from more of a leftist political tradition, and the hope that workers would never have joined this party. But that from my rational-choice model, I had predicted that you should find certain kinds of workers that would have been attracted to the party; particularly very nationalist workers, workers who weren’t in labor unions, workers who were in the types of industries whose material interest came closer to what was being offered by the Nazi party and that’s a whole other story in terms of how we I think have misunderstood what the positions the Nazi party took particularly on certain key economic issues. So one important finding was the substantial and significant working-class support for the Nazi party. Another one was in terms of our ability to look at intra-class differences. Again, not much work had been done there, and I think there was a lot of support for the hypotheses that I had that you should find among lower middle-class groups such as the farming group, that the livestock and dairy farmers should have been over-represented in party membership, whereas those that worked in grain growing and wine production should have been under-represented. Again, it follows from the theory. And again, that was supported very strongly by the data findings. One of the most interesting findings was with women. Now, we know that the party was, as most political parties and particularly extremist parties, was very underrepresented with respect to women. And that we had generally seven percent of our membership was female. But I had predicted that women would be underrepresented in the party. Particularly because of the party’s famous stance on the three K’s: on the Church, on the Kitchen (women should be at home), and on Children. And the idea that many had argued well, the party had said women should return to the hearth and give up their jobs. And so you’d imagine if a woman has a nice job, is interested in working, this is not the party to support. But what the data show that women were particularly over represented, vis-à-vis the German population of women, in the working class, and in the white-collar, because most women worked in clerical and service, in the white-collar and service. And so I said, "my god, what’s wrong with my theory," but then I scratched a little and I said, "When you look at my theory, let’s see where it takes me and try to make sense of this finding." And again, and the data allowed us to do this, because my theory had argued that people would join a party or at least be attracted to a party that catered to their consolation of material interest. And when I looked at the issue of what was the Nazi party really saying about women in the labor market, and what the party was saying when you read the literature closely in the party’s position papers, it said that women who are in a relationship, who are married, living with someone, or out of home with a male who has a job, and this is in the depression when there was a pretty high unemployment rate, that those women who are what the party calls it "double earners," should give up their jobs. But it says nothing about single women, and whether single women, in fact it says if you have a job and you’re supporting yourself, you should be able to keep that job. And so then I said "aha, that’s the case, let’s see what we find here." And so, with the kinds of data we had, we broke our women down in terms of unmarried or married. And in fact, what we found was that between 94 and 90 percent of the females who joined the party were unmarried. Which makes sense from this whereas when we compared it to the males, we only got about 50 – 52 percent of unmarried males, because it was not an issue there, joining the party. And so again, the theory helped us find that the question is yes, women were over represented, but it’s the single women who felt that hey, this is fine what the Nazi party is saying, because what it’s really saying is that married women, who could be our competitors in the labor market, should be at home, which would enhance our competitive position for jobs. And so what we did was then we were able to then not look at the marriage and the married/unmarried women overall, but we looked at it in terms of the levels of unemployment. And we found indeed that in communities with higher levels of unemployment where that incentive would even have been stronger, that we found the highest rates of unmarried women joining.

Q: It seems that if your theory is right there may be a difference between the membership of the party and the voting coalition behind the party. Do you have any thoughts about the nature behind that relationship?

A: Yeah. Well, first of all, when you look at most of the empirical work that had been done—indeed you’re right, it has been done based on the electoral analysis—what is usually said is that the distinction between voting and joining is not made. That most of these people say the attraction to the Nazi party is based on whatever they found analyzing the electoral. What I do at a theoretical level is try to distinguish between the acts of voting and joining. Indeed, maybe there is a difference there, in terms of, maybe, because my theory predicts that different types of people should join from those people who voted. But I’m not yet at the point to say that what I found in terms of membership might not still be appropriate if there had been some way to study those who voted. Ok, at the same time, taking into a context, controlling for the selective incentives and disincentives. But what I’m saying that I think there are such problems with the ecological fallacy issues, that they weren’t able to capture perhaps what we were able to capture at the individual level. And if there was some way, better ways to overcome the ecological fallacy issues that these other studies were based on, they might have come up with some closer findings, or findings that mirrored what we found, stronger worker class support for the party. But I saw nothing in terms of my selective incentives and disincentives that would account for a differential of let’s say 15% between what they argued 25% working class support, but we found over 40%.

Q: Maybe for our particular audience, you could describe the ecological fallacy for us, and how that’s involved potentially in a study of this kind.

A: Well, with the ecological fallacy is that scholars are trying to often times say certain things about the kinds of people who join, about individuals. And they’re trying to say it from an aggregate level. They’re inferring certain attributes to individuals that they find at an aggregate level.

Q: Could you give us an example?

A: Well, at an aggregate level they found that 62%…. There’s a study, and I’ll give you the example from what I was working with. You have the 1930 election. In the 1930 German election of September, the Nazi party ended up with overall 18-19% of the popular vote. And what most of the studies have done, they had taken either the 35 Gau in the major districts of Germany, or in some cases some smaller sub-unit of that; and had found well, differentials here; that in this county we had 28% of the popular vote went for the Nazis. And in this county we find that 68% of the working population is lets say, working in crafts, artisans, and that only maybe 12% of the population is in farming. And they do this for a number of counties. And what they then find, aha, that the higher Nazi votes tended to come from counties that had similar attributes, that had higher proportions of their population lets say in the crafts. And therefore, it is more likely that people in the crafts would join the Nazi party. Well the problem there is that you don’t know whether that 28% of the vote actually came from those people in the crafts, or maybe, what if it came from the farmers plus people in business? But you’re making an assumption here that because the county has these characteristics that the individuals who actually voted for or joined must also have those characteristics. So that’s a more general idea of the ecological fallacy, when you infer certain things at the individual level from an aggregate level.

Q: Great. Could you reflect on the implications of your study for the future rational choice theory in sociology; what have you learned about rational choice theory as a general framework for analysis?

A: Well, I feel that, again, mine is just one study, and there are certainly many problems. I mean you pointed out well, it would be great if we could compare what we had with the Nazi party records, lets say the Communist party or the Social Democratic party. But when I got into this, there had been very very little, if any, attention given to that the decisions that individuals made about joining this movement, that those decisions could have all been rational, that is, individuals looked at it and said "my god, the costs in joining this movement are such, the benefits are such, and then because the benefits outweigh the costs, that would motivate my decision." There was really none of that in the literature. And so what I've tried to do was just to change the equilibrium here, or make it more of equilibrium in terms of lets take what is generally assumed as a major theory in the social sciences, and see whether it might have some relevance for extremist political behavior. So I do hope in one way that people would start to look at sociologists in particular who are very interested in studying social movements, and extremist political parties, that not to assume automatically, that because a movement, in terms of its outcome, could be so evil and so irrational as a Nazi party and to then assume that if the consequence is irrational or the consequence at an aggregate level is irrational, that at the origins, or at the individual level that those motivations have to also be irrational. Because what I argue in the book is that evil as an outcome can have very ordinary, mundane, rational origins, that those people who joined the party in 1928, 1930, could not have known what 1945 was to bring. They were looking at it as most people do: well, what do I see now on the landscape; what’s in my interests, what are not in my interests, and act on that. So it is that thinking that I hope sociologists would at least entertain, in terms of looking at social movements and political parties. It’s not that I’m arguing that all these other explanations that are more cultural, irrational, are irrelevant; but what it comes down to is that when you’re trying to understand the motivations of millions of people who joined a movement like the Nazi party, that made rational choice so attractive to me is that when you look at people’s preferences, you see that almost everyone’s preference scale, there is wealth maximization. What makes wealth maximization so important is the element of fungability, that it’s exchangeable. Money can be exchanged for other things. Some people may put wealth up at the top of their preference scale and always act on that, other people, like Mother Theresa, may put it very low. But even somebody like a Mother Theresa, if you were to give her a choice of "would you like $10,000.00 or $1.00, Mother Theresa would go for $10,000, not because she would enrich herself, but she could use that because of the fungability, to buy the things or get the things for the causes she cares about. So its that belief in fundability, and in the ability of wealth maximization as part of rational choice to explain motivations that to me makes this so promising, but not the only explanation.

Q: Well, there’s a number of avenues we can take here. One avenue though, is the question of whether wealth maximization is the only assumption that we can make about rational behavior. You could have made some other conclusion. Can you say something about…?

A: Sure. It’s the one I chose, but if you look at rational choice now, it’s a very large tent, and there are many different people under it. And there are people who go from a very rigorous, a parsimonious approach, who look at it in terms of strict wealth, to people who bring in things like ability, friendship, who basically have inflated the category to include so many different elements. And my fear there, particularly when I looked at this study, is I wanted to take a more parsimonious model, because I felt that it was more testable. And material interest to me was much more testable. But that’s not to argue that other elements such as people joining because their friends join, or likeability, or questions of ostracism aren’t rational as well. But I made a decision, and in the book I lay out that this is my form of rational choice. It’s not the only form. And believe me, there are, and you know, Steve, there are so many people in this field who feel that the form that I took is just a very small, and one that’s not the most appropriate rational choice.

Q: Could you characterize the reaction to your study?

A: Well yeah, there’s been a lot, and in a way I certainly enjoyed it. I’d start with the public reaction, or the outside of academia. I was surprised that a week before the book was published, after it came out, there was a review by James Sheehan, a historian at Stanford and the New York Times, review books. And that basically unleashed a number of reviews and public discussion. It was also; it came out at this time that Daniel Goldhagen’s, Hitler’s Willing Executioners, came out, which sold millions of more copies. But there was a lot of interest, public interest, in the issue of Nazis and Jews. And I hope you can have some time where I can talk about the issue of anti-Semitism, because that really is now where I’m going with this project, and a lot of it came out of the Goldhagen controversy. But the book has been, my book has been very controversial. Not as controversial as Goldhagen’s, but controversial nonetheless. Because a lot of people just don’t like the idea that first that I argued that anti-Semitism from what I could see played a marginal role, at least in the motivations of individuals to join this party. And much of the literature had emphasized the importance of anti-Semitism in the rise of the Nazi party and why people joined, and I didn’t find much support, at least with respect to what I did. There was controversy that I could argue that people were attracted to the Nazi party because its economic programs were so innovative. And in fact, we like to forget this but John Menard Keynes, a very well-renowned and respected economist, gave a speech in Hamburg, Germany, in 1932, before the Nazis came to power, and in it he praised the Nazi party’s economic programs, at the most foresighted programs as the program, in terms of looking at western political parties, most likely to address the worldwide economic depression. But we read back into history and people don’t like to remember those points. But a lot of people didn’t like the fact that I argued that people were attracted to the party because, and particularly kinds of people, because it made a lot of economic sense. They really believed that it offered them the best chance to improve their economic material livelihoods. And so that was a bit controversial. And then, I mean scholars certainly had their problems. It was controversial with respect to some of the methodological issues involved, but particularly the rational choice. The sociologists, most sociologists, were very much bothered by that. The historians came after me, many of them, and they didn’t like the idea that you could use any kind of grand theory to explain this idiosyncratic, in their minds, movement, which I tried to argue was not the case. So yes, indeed, a lot of controversy. But the anti-Semitism one was certainly a key one.

Q: Well let me press on that. You have data about the occupations, but you don’t have data on the degree of Anti-semanitism held by the individuals, or members of the party. How can you weigh these competing explanations then, having lots of evidence on one side and little on the other?

A: Good you asked me that; let me just say in terms of yes, indeed, you’re right. And what I relied on there, to a large extent, was secondary literature. And also some of the significant studies that had been done in terms of content analysis of Hitler’s speeches, and of the major German Nazi newspaper, the Volkische Beobachter. And what many had found was that between 1928 and 32, the key electoral period for the party, the discussion of Jews and anti-Semitism really recedes from these party programs and speeches, compared to what they had been 25 to 28 and from 33 on, once the party is in power. That was one thing. Another thing I found in my own research was that the Nazi party didn’t have any monopoly on anti-Semitism; that the context in Weimar Germany was very, was highly anti-Semitic; that I found, and my students that did research, found that when we read party programs as well as the party newspapers from the Communist party, the Social Democratic party, the Nationalist party, that there was very, there wasn’t a significant difference in what was being said in many of those papers from what the Nazis were. So I argued that Nazis certainly had no monopoly, and this by itself couldn’t explain why, that the Nazis stood out particularly from the other groups, with respect to their anti-semanitism. But also my findings, it made no sense, I tried to make sense of the findings in terms of why would dairy and livestock farmers be more anti-Semitic, because they were more, they were over represented in the party than let’s say, grain growers. Or why we found higher representations among workers in construction industry, in woodworking, in textiles, than let’s say in chemical, in metallurgy. And we looked at in terms of were there more Jews owning these kinds of companies; we found no proof of that, and so the data didn’t make sense from the anti-Semitic. But I didn’t feel like I’ve done a sufficient job there. So this gets me into the new research. The new research, which is being funded by the National Science Foundation, as well, is that what we’re trying to do is the first truly comparative and systematic study of anti-Semitism before the Holocaust, in different countries. And we’re trying to test whether in fact Germany was this exception, as many have argued. It’s a multi-country study. And, at least for the empirical part, we’re doing first we have coded the American Jewish yearbooks from 1899, the first year of publication, to 1939, and they are the best data source for anti-Semitic acts, and all types of acts, whether they are laws or a demonstration or anything violent, for the period 1899 to 1939, by year, by country. And so we have this one data source that we’re coding and we’re going to analyze to see whether Germany is any way an outlier here. The second, and this is the most important part, is that we are analyzing, examining, the major newspapers. Not anti-Semitic newspapers, but the major daily newspapers from each of the five countries in the study, and Germany is one of the countries, from 1899 to 1939, sampling the 15th day of the month, for each newspaper, looking for any article that talks about Jews or Jewish issues. And at the same time, we’re doing critical discourse moments, with respect to these newspapers. That is, that it’s like Krystalnacht, the pogrom that took place after the assassination of German Counselor in Paris in November 1938. To compare how newspapers both within countries and across countries dealt in detail with major events dealing with Jews. And so this will be the first major systematic empirical study of popular attitudes towards Jews. We’re taking newspapers again because this is a period when surveys weren’t done. And the assumption being that newspapers both reflected popular attitudes as well as shaped popular attitudes. So we’re trying to get at this issue because I feel I didn’t get at it sufficiently in the Logic of Evil, and I thing Goldhagen doesn’t do a very good job of it when he talks about German exceptionalism with anti-Semitism; you can’t talk about German exceptionalism if your only case is Germany. You have to talk about and study other cases, and this is what we’re trying to do.

Q: So from this particular study, you have made a strong case that there is some economic foundation for support of the membership in the Nazi party, even though you haven’t necessarily eliminated anti-Semitism as an important force for at least some people.

A: Yeah, and I think that perhaps under-rated, or I didn’t give enough attention to it. Again, it was very difficult to measure, so my evidence is very circumstantial; it’s very, um, the Logic of Evil not at all do a scientific study of the role of anti-Semitism.

Q: I wonder if we can bring things down to somewhat of a more mundane level. Can you tell me a little bit about the publication strategy that you took, the form of publication, the publisher you chose, and that sort of thing?

A: Well, I wanted to get a press that had a very good reputation in, particularly political science, and I went with the political science editor. But I had contacted a few presses, I had contacted at first Princeton, Chicago, and Yale. And I had told them that I was sending it out simultaneously, and a lot of presses as you know don’t like that strategy, but that I did, and I was not happy with, particularly, Princeton. I was talking to Princeton and I felt that the editor there that was interested, a very junior editor, wasn’t forceful enough with it. And particularly one review came back and said, and again, this has to do with the controversial nature of the book, said that "what Brustein is doing in this work is that he’s praising the virtues of Nazism and exonerating the Germans." And obviously as someone who has lost relatives because of this and as a social scientist, I felt that was so ridiculous, and so undeserved. And so I wasn’t very happy, but it was that John Covell, and he had been recommended to me, as a political science editor, senior editor at Yale. Yale has a history of publishing in the area of strong work in political science, comparative politics, and also in terms of Jewish studies, and because of it, my prior experience with my first book with University of California Press I wasn't happy. It was too big. And you know, books get lost. I had confidence that Yale, being smaller, was going to give it the kind of attention that I wanted. And everything has come out like I had hoped, by the way. And that I was so impressed by the editor there, and in fact the editor had to weather a storm, again because of the controversial nature of this book, and that particularly a number of historians at Yale who had written on the subject and found my work to be very much at variance to what they had written. But my editor stood up, and so, in fact the controversy over publishing this book led to a featured article on me and the book in the Chronicle of Higher Education, a couple of years before the book came out. But it was with Yale. So the major strategy was to get out a major monograph, and I was very happy with the copy-editing that Yale did, and with the marketing of the book that Yale has done. But at the same time, some of the more highly empirical pieces we published in the form of articles, I did with a member, this project involved the number of students, undergrads and grads, and so I published pieces with them as well as with my colleague in Germany, Falter. So the publications, the major thing was the book, but as well there were a number of articles and highly visible journals in the social sciences.

Q: Well, the articles of course are a way to reward collaborators that might not be easy to reward in the form of a book. But they also give away some of your story, maybe before you’re ready to actually pin it down in a book form. So were there qualms about publishing articles before the book?

A: No, because most of the articles have come out really after the publication of the book. But the articles that came out went in a different way. They went into the topics in a way that the book doesn’t, and particularly because they were collaborative pieces so you want to bring in the contributions of other people. You see, I wrote this book much more from the rational testing of one theory. So that seems to be strongly followed with respect to the book. The articles are less so because Falter, my colleague in Germany, for example, is not a rational-choice aficionado. And that wouldn’t fit him. And from the students, one of them, Brian Ault, was much more interested in the issue of political geography. So the articles we did together, that article, was much more in terms of mapping, which you don’t find much in the book. Another article that a couple of my students did, dealt with it, Ault and Berntson, was on gender. And though I talk a little bit about gender, they’re interested in identity theory, which I am not, and so they set their article up more in terms of testing identity theory against rational choice. And since they come out more strongly for identity theory, I wasn’t going to sign my name to the article.

Back to abstracts & scripts menu.