This file is also available in Adobe Acrobat PDF format
TWO STORIES, both reported in a Milan newspaper, the Corriere Della Sera:1 The first comes from Genoa and the latter from Rome:
[A]t 3 o'clock in the morning, Samlal Ali, a 24-year-old Moroccan, was selling cigarettes to passers-by. A car with four youths inside stopped in front of him. The man sitting next to the driver leaned out and said something to the Moroccan. He then grabbed him by the neck and pulled his head into the car. Then, at 80 kph, the Moroccan was dragged along the asphalt for a kilometer, hanging onto the car, his head trapped in the window, crashing into parked cars, rubbish bins, and lamp-posts. The men in the car hit him repeatedly. The nightmare was only ended when the car was stopped by a police patrol. The four men, Massimiliano Bordonaro and Constantino Carta, both 23, Davide Cavallaro, 18, and Giovanni Mariani, 30, have been arrested for attempted murder.
[T]wo Egyptians were attacked by a group of seven skinheads. One of them, Phami Sbavvio, 33, was treated at the Grassi hospital for cuts and grazes on his arms and legs. This assault happened at just after one o'clock in Viale delle Repubbliche Marinare as the two foreigners were on their way home. The gang was probably waiting in ambush like the one sprung on the young leftists the other week, when the former secretary of the WWF Fabio Converto was left bleeding on the ground in the front of the station. With his fair hair he had been mistaken for a Pole. And yesterday, once again in front of the Ostia station, a group of skinheads wearing bomber jackets casually told a TV crew, 'They did well to whack that Tunisian. They come over here, take our jobs and steal our women.'
Yesterday evening, the mayor Fancesco Rutelli appealed to all Romans not 'to tolerate this brutal and mindless violence.' Also because between now and the elections the Wild West of Ostia-Fiumicino is liable to explode. Standing for the Right is the ex-fascist Teodoro Buontempo . . . who has never concealed his sympathies for the skinhead Nazis.
Only two stories in a newspaper, but they drive home the depth of the problem. It is again a time of refugees in Western Europe. They are making their way from former colonies of the West in Africa and Asia and of Russia in Eastern Europe, attempting to escape poverty or find asylum, sometimes intending to return, sometimes not. Many have found refuge, contributing to the countries that have given them sanctuary. But a storm has been gathering around them. Its force is felt, most conspicuously, in acts of violence committed by individuals clumped at the periphery of contemporary European societies. Less visibly but no less consequentially, political institutions at their center show the stress. These strains of intolerance--their strength, their deeper-lying sources, their impact on the politics of the society as a whole--deserve exploration.
Two problems provided the point of departure for our study. The first concerns the distinctive status of race in marking others as outsiders. Even people who are diligently unreflective have shown themselves capable of impressive powers of imagination in finding, or formulating, lines to mark off those who belong to their group and those who do not. Over a range of historical conflicts, however, a limited number of cleavages have been particularly prominent--among them, class, religion, ethnicity, and nationality, and last, but hardly least, race.
Whatever was true of other eras, it seemed to us that differences of race now cut deeper. In taking this view, we did not doubt the divisiveness of other cleavages, particularly ethnicity and nationality. With the myriad clashes from Yugoslavia through Rwanda, it would call for a special measure of obtuseness to overlook the murderous variety of group differences. Yet, differences of race, we reasoned, can be stigmatizing in a way that other differences need not be. However long African immigrants live in their new country, however well they learn its language, with however much self-restraint and dignity they bear the blows that fall on them, they are marked off by their color. Just so far as they are visibly different, they cannot escape notice of their difference. And the visible sign of their difference, the color of their skin, is indiscriminately connected to a complex of associations--emotional, symbolic, historical, even psychosexual--so much so as to make blacks specially vulnerable to the prejudices of others. All who have come from outside a society aiming to make their way within it labor under a burden of prejudice. But blacks, we feared, bear a heavier burden.
The second problem has to do with the roots of prejudice. It once was common to conceive of prejudice as an intrinsically--and narrowly--psychological phenomenon. So conceived, prejudice was not connected to the actual world, to the frictions, abrasions, and conflicts for the limited goods on offer, whether material or symbolic. Prejudice was psychological in the specific sense of being irrational, and both psychological and irrational in the still narrower sense of being principally rooted in the interior lives of individuals, in the inner conflicts and emotional wounds that they suffered in the course of their early development rather than in the stream of their experiences as adults in the larger society and economy.
But fashions in explanation come and go. Psychological analyses in general, and personality-centered ones in particular, have fallen out of style. Once familiar chords of explanation--of authoritarian submission and dominance, of overcontrolling fathers and overwhelmed children, of dogmatism and intolerance of ambiguity--now have a recognizable quality of quaintness, where they are recognized at all. No form of explanation, it is true, disappears entirely from the market of ideas, and it is part and parcel of the ordinary commerce of social science for those versed in one form of analytic transaction to think little of those who practice another. It is therefore all the more striking that it is psychologists themselves who have undertaken root-and-branch critiques of personality-based analyses of prejudice2 and that even their canonical reviews of prejudice research now pay only perfunctory attention to personality-centered explanations, summarizing the work, it almost seems, out of historical courtesy.3 The explanatory parade has marched on, with considerations of group identity and group interest, the social, the cognitive, and the economic, taking the lead and personality factors lagging far behind. Yet the new perspectives, illuminating as they are, nonetheless seemed to us to miss the distinctively irrational, emotional, and expressive character of prejudice highlighted by the classical personality-centered perspective. So we took as one of our principal objectives the aim of demonstrating that prejudice is rooted less in the actual interplay of social and economic life than in the deep-lying folds of individuals' psychological makeup.
This book is a record of how, under the pressure of our own results, we have been required, if not to reject then fundamentally rethink both of the major ideas we held at the start.
FORMS OF PREJUDICE: THE "SWITCH" EXPERIMENT
There is, regrettably, no shortage of sites for a study of prejudice and politics. France, for example, is an obvious possibility. The National Front, under Jean LePen, openly campaigns on a plank of returning France to the French, working to incite hostility toward immigrants, not to mention taking pride in a succession of anti-Semitic thrusts. The Front, moreover, has continued to increase its margin of popular support, now approaching one out of every six or seven votes, in the process acquiring sufficient electoral strength to splinter the traditional right in the most recent regional elections. Germany is an equally obvious site. The German People's Union, a party on the far right running on an antiforeigner and anti-European platform, made a striking breakthrough in the German state of Saxony-Anhalt; and Eastern Germany generally, not to mention a swath of Western Germany, has been pock-marked by skinhead violence and public agitation against immigrants. For that matter, Austria, with its thick history of anti-Semitism and the transformation of the traditionally conservative Freedom Party, in the 1980s, to a party of the extreme right, surely has much to recommend it as a site for the study of prejudice and politics. Then, too, Belgium, with its deepening internal divisions and the ballooning of support for the "Flemish Block," appears nearer crisis than any of the others.
All of these countries illustrate two crucial conditions for the study of prejudice and politics. The first is a deep strain of intolerance, in Western Europe now characteristically focused on immigrants or foreigners. The second is the emergence of at least one political party publicly committed to mobilizing public resentment against immigrants or foreigners. The second condition matters as much as the first. To realize the full potential of the politics of prejudice and group conflict, private grievances need a public vehicle.
Italy satisfies the first condition, and not merely because of the rash of hate acts against immigrants. Charges that immigrants are sopping up public benefits have become routine; so, too, have claims that immigrants promote crime, spread disease, and increase unemployment. Italy satisfies the second condition, too. Intolerance has visibly leached from the margins of the political system to very near its center. In the mid-1990s at least three of the political parties--Alleanza Nazionale, the Lega Nord, and Forza Italia--bid for public support by campaigning against the new immigrants. In a way inconceivable in the United States for all the nativism of the American tradition, political argumentation in Italy can be xenophobic and chauvinistic; and it is difficult to believe that the wave of prejudice against immigrants has crested.
But to understand the forces responsible for this eruption of intolerance, it first is necessary to fix what constitutes prejudice. How should it be defined? Can prejudice be pinned down? And what distinguishes a person who is prejudiced from one who is not? On the face of it, the answers to these questions are straightforward. There may be problems at a practical level. Perhaps because of the pressure to say the socially acceptable thing, perhaps because of the lack of time, it may not be possible to pin down the level of prejudice toward a particular group in a standard public opinion interview. And out of a concern for theoretical fastidiousness, some definitional crossing of t's and dotting of i's no doubt is in order. But a consensus has been reached on the core meaning of prejudice.
This consensus summarizes what thoughtful people have come to believe constitutes the heart of prejudice and it includes such obvious features as sterotyping, thinking ill of others without justification, and rigidity. But how to think about prejudice is, we have become persuaded, a question that should be thought through again. Speaking in the abstract, it sounds reasonable to say that prejudice involves characterizations of others that are stereotypical. Yet who is to say which characterizations are stereotypical? Are there truly objective standards to determine which assertions about groups, apart from the indisputably pathological, are true and which are false? And if there are not, is the definition of prejudice only a matter of convention, of political correctness if you will?
Our concern is not about the meaning of words. Our concern is to clarify the meaning of prejudice so that we can pick it out in the actual world and catch hold of that in which it truly consists. And the view that we take of prejudice, which we detail in the next chapter, has turned out to have potentially profound implications about the nature of prejudice that we did not see at all at the start.
We started with the presumption that for all the varieties of group conflict, the cleavage over race--the cleavage between black and white, above all--has defined the most fundamental terms in which we approach issues of bigotry and discrimination at the end of the century. Saying this may give the impression of being ethnocentric about ethnocentrism itself, implying that the problem of intolerance cannot rightly be understood outside of an American context. All that we can say is that in our research project in Italy, all who took part, Italians as much as Americans, believed that blacks must bear a special burden by virtue of being black.
Indeed, our shared conviction that differences of race are specially stigmatizing was our common reason for selecting Italy as a site for the study of prejudice and politics. For there are two distinct streams of immigration, one from Africa--from Morocco, Tunisia, Algeria, Senegal, and Somalia--the other from Eastern Europe--from Poland, Albania, and the former Yugoslavia. The first stream of immigrants is thus very largely (though not completely) black; the second, white. Here, then, is a natural experiment. Immigrants, whether from Africa or Eastern Europe, bear a burden of intolerance by virtue of being immigrants. But if we show that immigrants to Italy who are black bear a heavier burden by virtue of being black, then we should demonstrate that differences of race cut deeper than differences of ethnicity or nationality.
This idea of the special burden of blacks was one of the animating hypotheses of our study, though to speak of it as a hypothesis does not capture our certainty of its truth at the start of the study. All of us were sure it was so, and we only went about the business of conducting an experiment to demonstrate it was so because knowing something to be the case is not a substitute for showing it to be the case. So we took advantage of computer-assisted interviewing to conduct a specially designed experiment, the "Switch" experiment, in order to drive home the distinctiveness of race.
There is a standard litany of praises for genuinely randomized experiments, and no doubt we would have recited it just as others have but for the "Switch" experiment. So, before seeing the results of this experiment, we would have said that experiments have the power to persuade because they have the power to surprise--the power, that is, to demonstrate convincingly that the world is not as it has habitually been taken to be. But in speaking of the power of experiments to surprise we would tacitly have had in mind that it is others who would be caught up to find the world is not as they too confidently assumed it to be. In fact, it turned out that the results of the "Switch" experiment took us by surprise. Although it was designed to demonstrate that differences of race cut deeper than those of ethnicity and nationality, it instead showed that the wave of prejudice and group conflict now washing over Western Europe is more menacing than has been recognized because the readiness to categorize others as belonging to a group other than one's own is more indiscriminate than we had imagined.
This indiscriminateness throws a new light, we believe, on the nature of prejudice, exposing the fundamental sense in which prejudice truly is blind. For it is not, in the end, about the particular ways in which a group either is different or is said to be different; it is instead, at its core, about the fact that it is judged to be different. It is, we think, important to work toward a stronger grip on the nature of contemporary forms of intolerance. But we shall argue that prejudice, in addition to deserving attention in its own right, also merits attention because its consequences, when they spill over into politics, are not merely individual but societal.
THE VULNERABILITY OF THE LEFT: THE "RIGHT SHOCK" MODEL
It is the conviction of every informed observer that the eruption of anger and resentment over immigrants in Western Europe strengthens the hand of the political right there. Our results agree. But they go farther by exposing the basis of the vulnerability of the left.
The pivot point of our account is a cluster of values, including the importance of guaranteeing order, upholding authority, maintaining discipline, which we label "authority" values. This cluster of values is part of the core platform of the political right; and what is more, commentators on the right as well as the left would agree that they can provide a political base for opposition to immigrants and immigration. The rhetoric of immigrants' intruding their foreign customs and manners into daily life, of taking jobs away from native citizens, and of profiting from public assistance all resonate naturally and effectively with the right's emphasis on order, authority, and tradition. So very nearly everyone believes and so we shall show.
The core of our contribution consists of two further lines of argument. The first concerns the character of the causal connection between authority values and prejudice. It is standardly argued that the more firmly and consistently that citizens are committed to the values of authority, the more susceptible they are to the intolerance of others, very much including immigrants. The possibility that struck us as pivotal, however, is that the causal relation may run in both directions. Commitment to the values of order and authority can stoke hostility to immigrants. But hostility to immigrants also can stoke the appeal of the values of the right.
We draw out the implications of this hypothesis of reciprocal causation as we proceed, but we want to draw attention to one of the intuitions underlying it. The study of prejudice has emphasized, above all, the sources of hostility toward outgroups in the psychology and social circumstances of individuals: for example, those with the advantage of an extended education are, by virtue of their years of formal schooling, less susceptible to the strains of intolerance than those with comparatively little of it; or, again, those who live at the margins of society are more vulnerable to the appeals of prejudice than those situated at its center. But the level of intolerance in a society also can rise and fall along with the stream of changes within it. As pioneering studies have shown, a surge in the inflow of immigrants,4 for example, can cause spikes in the aggregate levels of hostility toward immigrants; so, too, can a slump in the economy.5
The second line of argument is this. The more we have worked to specify the sources of prejudice, the more important it has seemed to us essential to acknowledge the embeddedness of the prejudice in the world of actual events, and although the design of our study does not permit us to get a direct grip on it, it has supplied the key intuition informing our formal account of the interplay of prejudice and politics, the "Right Shock" model. It is necessary, we suggest, to take account of a class of changes in the economy and society. Generalizing over the specific forms these changes can take, this set of societal changes can economically be represented as external shocks. Under the assumptions of the "Right Shock" model, hostility to immigrants is taken to be a function of the force of these shocks independent of the impact of the circumstances and makeup of individuals. The core idea is twofold: first, that the level of hostility to immigrants spikes in response to an external shock; second, that given the reciprocity of the causal connection between hostility to immigrants and authority values, insofar as prejudice increases in response to an external shock, the appeal of the values of the right will increase in response to the increase in prejudice.
Our third line of argument traces the strange filigree of political ideology in contemporary Italy and, we believe, Western Europe. For a study of the belief systems of ordinary citizens to focus on ideology may seem strange in itself. A half century into the systematic study of public opinion, it has been a wearying tale of minimal levels of attention and knowledge. If ordinary citizens often can barely make out the shapes of immediate and salient figures in politics, if they so frequently have not learned what goes together with what, let alone why, it cannot be surprising that their grip on the abstractions of political thought, and above all, of the complexities of political ideology has proven to be so conspicuously weak. Nonetheless we will suggest that a key to the politics of exclusion lies in the interplay between what citizens think that they think ideologically and what they actually think.
It is consensually agreed that the correspondence between the two--between ideological self-conceptions and actual ideological commitments--is imperfect. Many who think of themselves as being on the political left actually adhere to the beliefs of the right, and vice versa. But because it is perfectly obvious that ordinary citizens will make mistakes about ideological matters, the nature of the mistakes they will make has seemed so obvious as to be taken for granted. Just so far as mismatches between citizens' ideological self-conceptions and their actual ideological commitments follow from a failure to understand, in Philip Converse's classic phrase, "what goes together with what," the political consequences should wash out. It should be approximately as likely that those who think they are on the political left hold the values of the right as that those who see themselves as on the right hold the values of the left.
And there undoubtedly is a spray of random error. But there is also a pattern of systematic error. As we shall show, the cluster of authority values--the importance of guaranteeing order, securing respect for authority, maintaining discipline--has nearly as potent an appeal to those on the political left as to those on the right, their natural constituency. It is not, it should be underlined, that the European left's conception of itself is deformed at its core. Those on the left who are best positioned to understand the values of the left reject sharply these values of the right, as we shall see. It is another matter for the rank and file of the left. Handicapped by their limited education and understanding of the principles of the left, they are open to persuasion. The result: on one of the most basic dimensions of partisan conflict, the right has the support of its own constituency plus a large part of the left's.
The politics of immigration represents, we fear, a point of deep vulnerability for the political left in Italy--and, we believe, in Western Europe generally. What has so far provided an electoral defense against it is the power of traditional political self-conceptions to dominate the voting calculus. There is a knot of ironies to untangle here, and not the least of these, our results suggest, concerns the value of tolerance. It is the cosmopolitan and enlightenment value par excellence, yet its future may hinge on the tenacity of tradition and habits of allegiance.
AN INTEGRATED THEORY: THE "TWO FLAVORS" MODEL
What makes people susceptible to prejudice? Our touchstone has been two classic theories of prejudice. The first locates the sources of prejudice inside individuals. More particularly, the roots of intolerance, whether toward Jews, blacks, or an array of outgroups, are said ultimately to be sunk in people's emotional needs and inner psychological conflicts. There are a number of different ways in which these psychological processes may be conceived, and it certainly is conceded that a variety of societal and economic factors may aggravate or mitigate the problem of intolerance. But in this first approach the key premise is that prejudice is ultimately rooted in personality factors.
The second classic theory of prejudice and group conflict flies under the banner of "realistic conflict." It locates the sources of prejudice not in the interior needs of individuals, but in the objective conditions of social life. More particularly, it contends that the key mechanism generating prejudice is competition for scarce resources. The competition may be over economic well-being or social standing. Either way, those who belong to one group fear that they will be less well-off if those who belong to another group are better-off. The result: prejudice and group conflict.
Both of these classic theories, many will recognize, represent versions of more generic approaches to the explanation of human choice. The first views the choices that people make as an expression of their enduring emotions and attitudes toward others and themselves, their core beliefs about the way that the world is and how it should be understood, and their deep-rooted sense of what is meaningful, fulfilling, and valuable in their lives. The second approach treats the choices that people make as the product of a more objective, often material, and usually self-interested calculus. The first kind of explanation thus tends to have an expressive, psychological flavor; the second, an instrumental, rational choice flavor.
These two explanatory flavors seem naturally to clash. Though each can be variously formulated, psychologically oriented accounts (even when social in their orientation) have lent themselves to an interest in the inner needs and cognitive processes of individuals, while instrumentally oriented accounts have been contrastingly tough-minded, uninterested in the subjective, the irrational, or the emotional, and centered instead on the realistic calculus of advantage and disadvantage. Certainly judging by the record of current research, those with a taste for the first have an aversion to the second; those with a taste for the second, an aversion to the first.6
We were tempted to enter the fray. The arguments against personality-based accounts of prejudice, though cogent, are not hat-doffing, and we saw a way around the problems of measurement that have frustrated inquiry for so long. For that matter, at the start we were skeptics not about the validity but rather of the relevance of realistic conflict analyses applied to the particular problem of immigrants in Italy. No doubt a belief that Italians were worse-off just so far as immigrants were better-off lay behind some of the hostility directed at them. But it seemed to us a good bet that it would prove epiphenomenal, or at any rate marginal, to the main forces at work. In the interests of parsimony we therefore began by doing our level best to see if considerations of economic well-being (variously conceived) could either be eliminated or relegated to peripheral status. But whatever tack we took, the impact of considerations of economic well-being could not be ignored.
Our initial results thus required us to rethink our objectives. Rather than arguing on behalf of one explanatory approach and against another, our aim should be to show how each, if viewed from a larger perspective, could be fitted together with the other to form a larger, encompassing account of prejudice. But how could these two very different explanatory approaches, one expressive, psychological in flavor, the other instrumental, rational choice in flavor, possibly be blended together?
Mechanisms are the key to explanation. To understand is to grasp the means by which one thing leads to another, at any rate at the level that the social sciences are capable of yielding understanding.7 What we were in search of, accordingly, was a common mechanism, a way of explaining in the very same terms how very different kinds of factors could lead people to be susceptible to prejudice. Only after wrestling with the problem of conflicting explanations of prejudice for several years did we suddenly see that the work of Henri Tajfel supplied a key. In the argument and analysis that follow in later chapters, we spell out Tajfel's insight into the centrality of categorization for the understanding of prejudice and group conflict. Here we wish to underline our debt to his work. For it suggested a pivotal mechanism to account for the impact of a wide variety of ostensibly quite different kinds of causal factors, running from social class at one explanatory pole to personality at the other.
Once we saw the opening, we had to follow it up and develop a causal model--we have dubbed it the "Two Flavors" model since its objective is to integrate explanations that are expressive or psychological in flavor and those that are instrumental and more rational choice in flavor. The model offers a comprehensive account of an array of explanatory factors--comprehensive but not complete. Given the design of our study, it focuses on factors at the level of individuals and--ironically, given that it is rooted in the work of Tajfel and his students--passes over factors at the level of groups. But integrating different explanations, even if the integration inevitably is incomplete, seems to us a worthwhile objective. Much of the intellectual energy in the study of prejudice has gone into the clash of competing explanations. It is now appropriate for more effort to go to pulling them together, to showing that instead of conflicting necessarily one with another, they may complement one another. And by pulling them together, we do not mean ritualistically acknowledging that each captures a part of the phenomenon of prejudice; we mean, rather, genuinely integrating them by showing how they can be brought together under a common explanatory framework.
It is the object of science to show that initially dissimilar phenomena, more deeply understood, can be understood as aspects of an overarching theoretical framework. It should similarly be the goal of the social sciences, imperfect as they are as sciences. And just for this reason, it has seemed to us a good idea to test the reach of the "Two Flavors" model, to see if it could account for not just what it was designed to account for, but also what it would, at first sight, not even apply to. Italy notoriously is split by a cleavage between North and South. From its national formation, Northern Italians have shown a formidable measure of prejudice toward Southern Italians. We say prejudice, but not in the classical sense. There was never a presumption of biological inferiority, of an inherent and gross lack of intelligence or ability. Indeed, if anything, the suggestion has run the other way around--that Southerners, thanks to a special shrewdness, have managed to evade the responsibilities of life and yet enjoy the pleasures of the day. But, to sketch the familiar historical portrait of Southern Italians as painted by their Northern compatriots, they lack--indeed take pride in lacking--essential qualities of character: honesty, independence, the willingness to work.
Northern Italians' view of Southern Italians, though not necessarily unique in every aspect, is so sunk in the soil of Italy's national experience that it may seem fundamentally incomprehensible apart from it. And from many perspectives this surely is so. But both the results of the "Switch" experiment and the logic of the "Two Flavors" model suggested a different perspective. What is central, both suggested, is the highlighting of differences between one group and another; the particular points of difference highlighted, and the specific factors responsible for their being highlighted, are secondary. A truly demanding test of the value of this reasoning, it occurred to us, was to apply the "Two Flavors" model to the evaluations that Northern Italians make of Southern Italians. There surely are one hundred ways that Southern Italians are recognized to differ from immigrants to Italy from Eastern Europe and Africa. And the operational version of the model had been expressly constructed to account for the evaluations that Italians make of immigrants. If, using exactly the same explanatory measures, and making use of exactly the same estimation procedures, the "Two Flavors" model could account for the hostility of Northern Italians toward Southern Italians as well for the hostility of Italians from both the North and the South toward immigrants, we could take a genuine measure of confidence in our argument. And the detailed analysis suggests that we should.
Our analysis rests on a representative, national survey of Italians.8 There naturally are limits to what can be learned from a public opinion interview, and some of the subjects we wish to explore, prejudice most obviously among them, are famously elusive. Yet if you wish to know what others think, whether about immigrants or a possible breakdown in social order or the appeal of a new direction in politics, there is no substitute for asking them. Our survey, the first of its kind in Italy, offers a portrait of the causes and consequences of intolerance--a portrait that is partial, one that is certainly not free of blemishes (particularly at the level of measurement), but, we believe, one that is illuminating nonetheless.9
We are aware of, and want to underline, the limits of our analysis. We have worked to put our arguments to the test, through cross-validation, by multiple forms of statistical analysis and by conjoining randomized experimentation and representative sampling. But, it seems to us, the more one enters into the spirit of inquiry, the clearer it becomes that the goal is less to establish what is true than, progressively, to expose what is false. There never is certainty, but there sometimes is progress, and the latter is possible just because the former is unattainable. As Hilary Putnam, citing Charles Pierce, has remarked: "It's as if we were walking on unfirm ground, on swampy grounds, and that was good because, if the ground were firm, there would be no reason to go anywhere."10
Return to Book Description
File created: 8/7/2007
Questions and comments to: email@example.com
Princeton University Press