CC BY
40
RECONCILING THE GENERAL AND THE UNIQUE: AREA STUDIES, CASE STUDIES, AND HISTORY VERSUS THEORETICAL SOCIAL SCIENCE
Fred Eidlin
INTRODUCTION: THE UNDERLYING TENSION
BETWEEN THE GENERAL AND THE UNIQUE
A substantial part of social science research consists of area studies, case studies, and historical studies. And this is not surprising. Political and social events and processes always take place in specific contexts - for example, in specific countries, specific states, and specific communities. And, political and social phenomena can be observed only in concrete cases, contexts, and events. Area studies, case studies, and history constitute the political science equivalent of the laboratory in the experimental sciences.
Yet the relationship between the specific, the concrete, and the unique, on the one hand, and the generalizing spirit ofscience, on the other hand, is highly problematic. This relationship remains a continuing source of tension in the social sciences. The widespread concern with scientific status is usually associated with the view that any real science should search for general laws. Defenders of such a view contend that all genuine sciences seek to generalize. They attempt to explain specific events, cases, and phenomena by bringing them under general theories and laws. Many social scientists who take the ideal of a science of society view studies that focus on the specific, the concrete, and the unique as lacking “in rigor and scientific potentiality ... as descriptive and relativistic ... and normally idiosyncratic in terms of [their] findings1.
1 Ward, Robert E. 1975. Political Science and Area Studies. Pye, Lucien, (ed). 1975. Political Science and Area Studies: Rivals or Partners? Bloomington, IN, and London: Indiana University Press, p. 27.
Other social scientists have been deeply troubled about the ascendency of such a program in their disciplines - a program prescribing reduction of the study of politics and society to a search for general laws. Each case, each event, each phenomenon, each political regime, they argue, is infinitely rich in unique, irreducible detail. Hence the program of reducing all of political and social science to a search for general laws is bound to be futile, even to impede the search for useful knowledge. Consequently, as Lucien Pye has put it, social scientists have tended to divide “... between those who would be the boldest in striving for the outermost limits of generality and those who would be most precise and penetrating in understanding the individual case2.
To be sure, the polarization is not as sharp as it used to be. Few area specialists who make use of case study methodology, and few researchers who make use of historical data nowadays ignore social scientific methods and theories3. Many approaches and theories in the social sciences facilitate genuine comparative and generalizing research - for example, structural functionalism, institutionalism, systems approaches, social movement theory, modernization theory, elite theory, small group theory, and organization theory. At the other pole of the divide, those stressing the generalizing, law-seeking essence of science usually acknowledge, at least in principle, the importance of in-depth knowledge ofparticular countries, cases, events, and histories.
Most of what has been written in recent years
2 The Confrontation between Discipline and Area Studies. Lucien Pye (ed) Political Science and Area Studies:
Rivals or Partners? Bloomington, IN, and London: Indiana University Press, 6.
3 Ibid, p. 18.
PARATIVE POLITICS ■ 2 (19) / 2015 5
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
about the relationship between the particular and the general in political science acknowledges, at least in principle, that each side has something to contribute to the other.
But that is as far as it goes. Under the surface, the truce between the two poles has been uneasy, and the problems that separate them are far from being resolved. To be sure, there exists a substantial literature on comparative method. But it is not of much help, either in bridging the theoretical problems underlying the gap, or in providing useful guidance for the conduct of research. Moreover, the state of theory in comparative political and social science is not very satisfying. There may exist quantitative cross-cultural studies that are genuinely comparative. And, there may be interesting case studies, historical studies, and single- and multi-country studies that are rich in theoretical insight. But they rarely lend themselves to systematization and testing. Most importantly and interestingly, no coherent account has yet emerged as to how the generalizing spirit of social science might be fruitfully and systematically integrated with the uniqueness of case studies and historical and area studies.
Some social scientists have explicitly attempted to bridge the gap. Robert K. Merton and Samuel H. Beer, for example, long ago argued in favor of generalizing approaches that do not call for formulation of universal laws and theories4. Merton retains the ideal of universal laws and theories. Yet, in his view, until a mature social science emerges social scientists should focus on theories of the middle range. These are “theories that lie between the minor but necessary working hypotheses that evolve in abundance during day-to-day research and the all-inclusive systematic efforts to develop a unified theory that will explain all the observed uniformities of social behavior,
4 Johnson, Chalmers. 1975. Political Science and East Asian Area Studies. Pye, Lucien, (ed). 1975.
Political Science and Area Studies: Rivals or Partners? Bloomington, IN, and London: Indiana University Press: pp. 78-97.; Beer, Samuel H. 1963. Causal Explanation and Imaginative Re-Enactment. History and Theory. 3(1) 1963: pp. 6-29; Merton, Robert K. 1968. On Sociological Theories of the Middle Range. Robert K. Merton (ed.), Social Theory and Social Structure. London: The Free Press: pp. 39-72.
6 СРАВНИТЕЛЬНАЯ ПОЛИТИКА ■ 2
social organization and social change.” Beer, in contrast, is hostile to the ideal of what he calls “the universalist model of causal explanation.” Beer is, by no means opposed to scientific method which, he contends, “holds much promise for social and historical studies.” He favors the aims of “identifying uniformities in human and social behavior ... discovering “causal connections between types of events ... and framing “general statements describing such connections which can be used to explain events in the past - or even to predict events in the future.” Nevertheless, Beer considers “the universalist model of causal explanation” to have been “barren of success...” stating that it “cannot fail to daunt the social scientist who takes it seriously”5.
Most importantly, neither Merton nor Beer nor anyone else, as far as I am aware, has worked out an approach that systematically integrates the generalizing thrust of science with the concern for uniqueness that characterizes area, historical, and case-studies. It is precisely this task that the present paper addresses. Merton is far from having worked out such an approach. His argument fails to go beyond providing legitimation for a vaguely-defined range of kinds of middle-range theorizing that were and still are widespread in the social sciences. Beer goes farther than Merton by holding up two studies as concrete examples of what he calls imaginative reconstruction. And he discusses their methodology in some detail6. However, though clearer than Merton about the kind of approach he endorses, Beer too fails to elaborate a continuous link between studies that focus on the unique, and those that strive for generalization-even universal generalization, as the present paper seeks to do.
This paper puts forward an approach that provides a methodological bridge between the generalizing impulse of science and the richness and uniqueness of particular countries, cases, and events. I will refer to this bridge as “contextually limited generalization.” This method has much in common with Merton’s theories of the middle range, Beer’s method
5 Beer, Samuel H. 1963. Causal Explanation and Imaginative Re-Enactment. History and Theory. 3(1) 1963: p. 39.
6 Ibid.
(19) / 2015
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
of imaginative re-enactment, and Chalmers Johnson’s style analysis7. I assume, along with Beer, that the ideal of explanation in terms of universal laws, often called the “Covering Law Model,” is fundamentally flawed. This ideal, I contend, has been subjected to devastating criticism, is at odds with most of what goes on in the advanced natural sciences and, as Beer points out, has been barren of success8. Nevertheless, it remains firmly anchored in the training and professional norms of mainstream political science, It is largely acquired, without careful scrutiny, through scope and methods textbooks and courses which train students in the conduct of inquiry. The Covering Law Model is usually presented to students without any indication that it might be seriously problematic. Since this ideal is so influential among political scientists who take the ideal of science seriously, I will have to confront it before proceeding to elaborate a method that assumes it to be fundamentally mistaken.
SCIENTISM: SCIENCE AS NEWTONIAN MECHANICS
If criticism of the Covering Law Model is as devastating as I have claimed, if it is at odds with the practice of the advanced natural sciences, if it has been barren of success, why does it remain so firmly anchored in the training and professional norms of mainstream of the social sciences?
Interestingly, the often heated debate about science in social science rarely if ever counterpoises a positivist account of science to some alternative, non-positivist account of science. Defenders of mainstream positivist approaches to social science typically equate criticism of positivist-inspired social science to criticism of science itself. They typically respond to critics by enumerating the well-known and formidable achievements of the advanced natural sciences, as well as the commonsense advantages of a scientific approach. They rarely 7 8
7 Johnson, Chalmers. 1975. Political Science and
East Asian Area Studies. Pye, Lucien, (ed). 1975.
Political Science and Area Studies: Rivals or Partners? Bloomington, IN, and London: Indiana University Press: pp. 78-97.
8 Beer, Samuel H. 1963. Causal Explanation and Imaginative Re-Enactment. History and Theory. 3(1) 1963: pp. 6-29
defend the positivist account of science per se, as opposed to defending science, more generally. They simply assume that the positivist account of science represents, more or less, what the scientific enterprise is about. When confronted with arguments pointing to the poverty of positivist social science, they typically respond by arguing that the social sciences are still young and underdeveloped. In other words, they have faith that, somehow, some day, the research programs they are following will lead to a rich and fruitful social science. The positivist image of science is reinforced, not only by the positivist spirit of the classics of social sciences, as well as by the image of science that is taken for granted in modern society.
Social scientists and philosophers who have reflected on the social sciences may differ as to how much the social sciences should model themselves after the natural sciences. Yet what goes unexamined throughout most discussions about the nature of the social sciences is a shared image of the natural sciences - the image of Newtonian mechanics. Not only defenders of positivist-inspired social science, but also those who argue that the social sciences cannot be genuine sciences tend to assume that “science” means science in the Newtonian image. Almost everyone, whether sympathetic or opposed to integrating the generalizing spirit of science with a concern for uniqueness, assumes the same image. They assume that if social science is to be truly scientific, it will have to be causal and nomothetic in the image of Newtonian mechanics.
The philosopher of social science, Ian Jarvie, compares this mainstream image of science in social science with a cargo cult. “Scientific success, Jarvie writes, like all success,” tends to have its worshipers. ... Nearly all religions promise to deliver one or another sort of goods; they will save your soul or bring you to nirvana.“
A cargo cult is a religion which promises literally to deliver goods-like the cargo in a ship’s hold. If the cultists carry out their religious performances properly some desirable goods will, it is believed, be delivered to them. In this case the deity was scientific method, the ritual was unprejudiced observation and patient induction, and the cargo the deity would deliver if worshiped by means of the correct ritual
COMPARATIVE POLITICS ■ 2 (19) / 2015 7
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
would be a science of society comparable in vigor and success to natural science9.
Jarvie takes note of Francis Bacon’s argument that there should be no worship of the success of science. All that was needed was “a patient application of those methods which had led to success and would no doubt do so again.” Naturally, Jarvie continues, Bacon was misunderstood. “[M]any would-be social scientists took it that he was saying 'worship not the false god of science but the True God of Scientific Method which grants success in all fields of inquiry.’ Thus ... scientific method became the object of a cult among social scientists -a cargo cult10.
As far as I am aware, there are no arguments appealing to science that would pose problems for the method presented below. By science
I mean, of course, not the cargo cultist scientism of mainstream social science, but of the actual practice of the advanced natural sciences.
THE AIM OF SCIENCE
What then is the aim of science? An initial answer might be that it depends on which science, which scientist, and in which situation. Generally speaking, science seeks explanations for problems or puzzles. Different kinds of problems require different kinds of explanations. I will assume, following Popper that it is the aim of science to find satisfactory explanations, of whatever strikes us as being in need of explanation. By an explanation (or a causal explanation) is meant a set of statements by which one describes the state of affairs to be explained (the explicandum) while the others, the explanatory statements, for the ‘explanation’ in the narrower sense of the word (the explicans of the explicandum)11. And what kind of an explanation may be satisfactory in Popper’s view? A satisfactory explanation is one that is couched “in terms of testable and falsifiable universal laws and initial conditions”12.
9 Jarvie. 1964a. Explanation in Social Science. British Journal for the Philosophy of Science, vol. 15, no. 57 (1964)
10 Jarvie. 1964a. Explanation in Social Science. British Journal for the Philosophy of Science, vol. 15, no. 57 (1964)
II Popper, Karl R. Objective Knowledge: An Evolutionary Approach. Oxford: Oxford University Press. p. 191.
12 Popper, Karl R. Objective Knowledge: An Evolutionary Approach. Oxford: Oxford University Press. p. 191.
8 СРАВНИТЕЛЬНАЯ ПОЛИТИКА ■ 2
But is this not precisely the kind of explanation prescribed by the Covering Law Model? In fact, this is the way Popper’s account of Scientific Method is usually [mis]understood. Even Beer, who clearly respects Popper’s work and takes it seriously, explicitly identifies Popper as an advocate of the Universalist Model. Yet, as will be shown below, Popper rejects the view that it is usually laws that are the unknowns scientists seek and elaborates methodological proposals closely resembling those that Beer advocates.
The reasons for such widespread misinterpretation of Popper’s position are partly historical. The Covering Law Model was originally advanced by Carl G. Hempel who did in fact view the aim of science as a search for universal laws. Hempel’s Covering Law Model was frequently cited as authority by social scientists and philosophers of science in the Logical Positivist succession long before Popper’s Logic of Scientific Discovery [LScD] (1959) was first published in English translation in 1959. When the English translation of LScD finally appeared, it was pre-interpreted as a positivist work, rather than as the radical critique of Logical Positivism that it actually was. In fact, the Covering Law Model is often referred to as the Popper-Hempel model.
Unlike Hempel, Popper does not hold that it must be laws that are the unknowns in the problems to which science seeks solutions. It may also be that something in the initial conditions is unknown, mistaken, or otherwise flawed. And the problem at hand is often solved by filling out or otherwise repairing the initial conditions. In fact, when a law-like regularity is observed, this may be the beginning rather than the end of a search for explanation. Such an observation may give rise to a new problem -that is, explaining why the regularity obtains.
Even in the advanced natural sciences, many regularities are explained by appealing to underlying structures. Since these structures belong to initial conditions, it is the revised initial conditions rather than newly-discovered laws that actually do the explaining. For example, in genetics explanation in terms of DNA structure, or in physics explanation of why light is refracted in a certain way by a crystal in terms of the crystal’s molecular structure; or in evolutionary biology, explanation of the survival
(19) / 2015
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
of an organism with reference to its environment and the characteristics of the organism that allow it to survive in this environment. The law of natural selection plays a trivial role in such biological explanations. In all these examples, it is structures (initial conditions) that are the unknowns. And it is the object of scientific research to discover these structures. When such structures are found, it is they that account for the order in experience. To be sure, many universal laws figure in this kind of explanation. These include, for example, laws concerning the behavior of light, laws of chemistry and physics which relate DNA structure to the genetic makeup of organisms. But such laws are not the object of search. They are taken for granted.
Many areas in the natural sciences are concerned exclusively with generalizations that hold only within delimited contexts. There are, for example, many generalizations that are true of dogs, but not of other mammals. And there are many generalizations about mammals do not hold for other vertebrates. Scientists rarely think in terms of the highest degree of universality. More often than not, they look for contextually limited generalizations, and are entirely satisfied when they find satisfactory ones.
Even Newton’s laws, which are commonly held up as the very model of universal laws, are contextually limited. The prejudice that the ultimate aim of a mature sc0ience is to discover universal laws is so compelling because the context of Newtonian mechanics is so broad as to encompass all of our everyday experience. Yet as broad as this context may be, it is still not a universal context. Newton’s prime assumption was that space is everywhere flat and infinite as it is in the world of our immediate experience. As Jacob Bronowski points out, this assumption was criticized even in Newton’s time by Leibniz. And, he notes, it is not even probable in our own experience. “We are used to living locally in a flat space,” Bronowski writes, “but as soon as we look in the large at the earth, we know it not to be so overall.”
The earth is spherical; so that the point at the North Pole can be sighted by two observers on the equator who are far apart, yet each of whom says, ‘I am looking due North’. Such a state of affairs is inconceivable to an inhabitant of a flat earth, or one who believes that the earth is flat overall
as it seems to be near him. Newton was really behaving like a flat-earther on a cosmic scale: sailing out into space with his foot-rule in one hand and his pocket-watch in the other, mapping space as if it were everywhere as it is here. And that is not necessarily so.... [I]n laying out space as an absolute grid, Newton had given an unreal simplicity to our perception of things13.
It is not my intention to denigrate the aim of searching for broader generalization, even searching for universal laws when such is appropriate to the problem at hand. To unify and simplify what is known by discovering more widely applicable generalizations is among the many different aims of science that scientific research may pursue. However, I wish to point out that scientific problems almost always arise within limited contexts and are often solved or explained by generalizations that apply only within such limited contexts.
SOURCES OF ORDER IN INDIVIDUAL BEHAVIOR AND IN SOCIETY
What accounts for the patterns, regularities, and invariance in individual behavior and in society? Obviously, some of the determining factors belong to the individual, others to the social environment in which individuals act. Some regularities in the behavior of individuals result from the physical and social settings in which they find themselves. Some are attributable to regularities in such factors as the aims, resources, and dispositions of the individuals themselves.
The structure of the social environment is manmade, notes Popper, in the sense that its institutions and traditions are “the results of human actions and decisions. But this does not mean that they are all consciously designed, and explicable in terms of needs, hopes, or motives.” Social institutions are rarely the products of conscious design. Most have “just ‘grown’ as the undesigned results of human actions”14-Once social institutions come into existence, they take on a life of their own. They become, to some extent, autonomous and independent ofthe
13 Bronowski, The Ascent of Man. Boston: Little, Brown, 1973. pp. 240-241.
14 Popper, Karl R. The Open Society and Its Enemies Vol. II, The High Tide of Prophesy: Hegel and Marx. Princeton, NJ: Princeton University Press. p. 93.
COMPARATIVE POLITICS ■ 2 (19) / 2015 9
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
will (and even awareness) of those who created them, as anyone who has attempted to change a social institution (such as a bureaucracy or a language) knows very well.
The task of the explanatory or theoretical social sciences is, in Popper’s view, is to discover and explain “the less obvious dependencies in the social sphere.” It is to discover ‘the difficulties which stand in the way of social action-the study, as it were, of the unwieldiness or the brittleness of the social stuff, of its resistance to our attempts to mold it and to work with it”15. Hopes, fears, ambitions, and aspirations explain little because of the wide gap that always exists between human aspirations and achievements. This is so because “social life is not only a trial of strength between opposing groups: it is action within a more or less resilient or brittle framework of institutions and traditions, and it creates -apart from any conscious counteraction-many unforeseen reactions in this framework”16.
The theoretical social sciences (along with much of natural science) usually ask questions about kinds or types of events or phenomena, and they almost always make use of a method that consists of constructing types of situations or conditions, that is to say, the method of constructing models, making use of situational analysis. By situational analysis Popper means a kind of tentative explanation of some human action that appeals to the situation in which the agent finds himself. “The central idea underlying situational analysis is thus that we can construct models of typical social and political situations and that this is the only means we possess of understanding social events17.”
Situational analysis, situational logic, or the logic of the situation is a simple and intuitive notion. “We assume,” writes Jarvie, that people have certain aims, that they also have certain means (restricted by their physical nature and by the social set-up of institutions and traditions), and certain knowledge and beliefs about their means and about the social set-up. Armed with ail this, they act to achieve their aims within the social situation created by traditions, institutions, and the
15 Ibid, p. 94.
aims and actions of other people”18. “Situational logic is explanation of human behavior as attempts to achieve goals or aims with limited means”19. A person, for the purposes of social science, can be viewed as in pursuit ofcertain goals or aims, within a framework of natural, social, psychological and ethical circumstances. These circumstances constitute both the means of achieving his aims and constraints on that achievement. A person’s conscious or unconscious appraisal of how he can achieve his aims might be called sorting out the logic of the situation he is in20.
Situational analysis assumes a physical world in which we act. This world contains, for example, physical resources which are at our disposal and about which we know something (often not very much). Beyond this, however, situational logic must also assume a social world, populated by other people, about whose goals we know something (often not very much) and furthermore, social institutions. They exist in an objective sense - that is, independently of any individual’s subjective understanding of them21. They have properties that are “mapable” and at least partly outside the awareness and control of those who participate in them - even of those who supposedly control them. In situational analysis every complex social situation, institution, or event is seen as the result of a particular configuration of individuals, their dispositions, beliefs, and environment. We may be unable at any point in our investigation to give complete explanations in terms of individuals, but the ideal of eventually doing so remains a regulative principle in much of social science research. In a large domain of social science research, we continue to attempt to reduce our explanations to statements about the dispositions, beliefs, resources, and interrelationships of individuals. The individuals may remain anonymous, and only typical dispositions may be attributed to them22. * &
18 Jarvie. Revolution in Anthropology. London: Routledge
& Kegan Paul. p. 18.
19 Jarvie. Concepts and Society (London: Routledge & Kegan Paul. p.5.
20 Ibid, p. 4
21 Popper, Karl R. The Logic of the Social Sciences. Theodor W. Adorno et al., The Positivist Dispute in German Sociology, trans. by Glyn Adey and David Frisby. London: Heinemann. p. 103.
22 Watkins, J. W. N.. Ideal Types and Historical Explanation. The British Journal for the Philosophy of Science, vol. 3 . In John O’Neill (ed.), Modes of
17 Popper, Karl R. La rationalite et le Statut du principe de rationalite. Emil M. Claasen, ed., Les fondements philoso-phiques des systemes economiques. Paris: Payot. p. 143.
10 СРАВНИТЕЛЬНАЯ ПОЛИТИКА ■ 2 (19) / 2015
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
To this, it may be objected that there exist irreducible social wholes - that is, social and political entities and phenomena whose behavior cannot be explained entirely (if it can be explained at all) in terms of individuals. How, for example, can a “tradition” or “culture” or ‘spirit of the times” or organizational esprit de corps possibly be explained in terms of beliefs, dispositions, resources, and interrelations of individuals. Unlike psychologistic methodological individualism Popper’s situational individualism is compatible with several aspects of a holistic view, Social “wholes” exist, which are more than the sum of their parts. These “wholes” include social groups as well as social institutions in the widest sense of the word. They cover a wide variety “from customs to constitutions and from neighborhoods to states”23. Situational analysis assumes that social “wholes” affect the aims of individuals and that “the social set-up influences and constrains the individual’s behavior”24.
What Popper and Agassi, among others, deny is that social wholes have distinct aims and interests of their own. Only individuals, strictly speaking, can have aims. An institution (or other social whole) may have aims and interests only when individuals give it aims or act in accord with what they consider should be its interest. A society or institution cannot have aims and interests of its own25. Traditions exist, as do cultures, “group spirits,” and other social institutions. They are more than the sum of the individuals that constitute them, and may exhibit emergent properties. They can exist before the individuals who make them up at any given time belong to them and can survive while maintaining continuity or identity and spirit after all the individuals they contained at any given time have left them26. But it is individuals who carry these traditions, spirits, etc. If enough individuals in a society abandon or alter their behavior or attitudes, whether
Individualism and Collectivism. London: Heinemann: 1976, pp. 167-168..
23 Agassi, Joseph. Methodological Individualism. The British Journal of Sociology, vol. 2 ,1976. In O’Neill:
pp. 188.
24 Ibid, pp. 186.
25 Ibid, pp. 188.
26 Popper, Karl R. Towards a Rational Theory of Tradition. Conjectures and Refutations: The Growth of Scientific Knowledge. New York: Hamer & Row. Ch. 4. 1965
intentionally or unintentionally, that tradition will change or die.
What about those aspects of social situations that appear to be psychological and therefore subjective, such as wishes, motives, memories, and associations? In situational analysis such concrete psychological experiences are replaced by abstract and typical (objective) elements of the situation such as ends or knowledge. The person with certain wishes becomes a man whose situation may be characterized by the fact that he pursues certain aims, and a man with certain memories and associations becomes a man whose situation can be characterized by the fact that he is equipped objectively with certain theories or with certain information. We then hypothesize that the persons or agents in our analysis will act in a manner that is adequate or appropriate - that is conforming to the situation. In Popper’s words:
This enables us then to understand actions in an objective sense so that we can say: admittedly I have different aims and I hold different theories (from say Charlemagne); but had I been placed in his situation thus analyzed-where the situation includes goals and knowledge - then 1, and presumably you too would have acted in a similar way to him27.
The theoretical reconstructions ofsituational analysis will inevitably be rough, rudimentary, oversimplified, and over schematized. Consequently, they will usually be false. However, they can be good approximations to the truth, and the fact that they are objective hypotheses permits us to learn from their falsity. Like scientific theories, situational analyses are rational, empirically criticizable, and capable of improvement or competitive comparison with alternative models of the same situation28.
No creative action (like a decision) can ever be fully explained. Nevertheless, we can try to give an idealized reconstruction of the problem
27 Popper, Karl R. Logic of the Social Sciences. Theodor W. Adorno et al., The Positivist Dispute in German Sociology, trans. by Glyn Adey and David Frisby. London: Heinemann. 1976. p. 103.
28 Popper, Karl R. La rationalite et le Statut du principe de rationalite. Emil M. Claasen, ed., Les fondements phi-losophiques des systemes economiques. Paris: Payot, 1967. p. 144-145; Popper, Karl R. Logic of the Social Sciences. Theodor W. Adorno et al., The Positivist Dispute in German Sociology, trans. by Glyn Adey and David Frisby. London: Heinemann. 1976. p. 103.
COMPARATIVE POLITICS ■ 2 (19) / 2015 \\
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
situation in which the agent found himself/herself. In this way the action may be made “rationally comprehensible” or “understandable” - that is to say, adequate to his situation as he said it29.
PERSPECTIVE AND CONTEXT
What remains unclear at this point is the relationship of these models, types, or constructs of social science to the reality they are presumed to represent. Almost everyone would agree that such constructs are abstractions from reality, which reflect the interests of the social scientists who construct and make use of them.
The role of perspective or point of view is most apparent in history. It is now widely accepted that there is no such thing as a universal history that has simply to be uncovered or revealed by the historian, and that historical facts do not speak for themselves. Rather, the historian imposes a perspective that determines not only which facts are relevant but, also, what are to be considered as facts. What is often called “the history of mankind” turns out upon closer examination, to be nothing more than “the history of political power.” - one of countless histories of mankind that could conceivably be written30. Each generation has its own problems, its own interests, its own background assumptions, which is one reason why history is continuously being rewritten. The title of Russian historian, Yuri Polyakov’s book, Nashe nepredskazuyeme proshloye (Our Unpredictable Past) illustrates this nicely. Many conflicts among historical interpretations result from differing points of view. The “facts” (I mean the infinite number of uninterpreted facts as they exist independently of anyone’s selection and interpretation) do not change. But facts are never known independently of some statement of them. And every statement of fact presupposes selection and interpretation in the light of some preexisting point of view or perspective.
Perspective also plays a role in the generalizing natural sciences. In the generalizing natural sciences, however, perspective is provided by 29 30
29 Popper, Karl R. Objective Knowledge: An Evolutionary Approach. Oxford: Oxford University Press, 1971. p. 179.
30 Popper, Karl R. The Open Society and Its Enemies Vol. II, The High Tide of Prophesy: Hegel and Marx. Princeton, NJ: Princeton University Press. 1966, p. 270.
prevailing theories, paradigms, or research programs. But even in the generalizing sciences, perspective or point of view is determined by prevailing cognitive interests. For example, Newton’s laws work very well within the world as we experience it. It is only when the cognitive interest of the physicist extend to objects moving at very high speeds that Einstein’s new perspective becomes necessary.
What is the difference between perspective and context? While the two terms may sometimes be used synonymously, I use the word “context” here to refer to the objective settings of the phenomena to be explained - to the conditions that actually obtain, rather than to any particular reconstruction of them. I mean the human agent’s objective situation, as opposed to any particular reconstruction a social scientist might attempt of that objectively existing situation. Any reconstruction of the context will, of course necessarily, single out only those features of the context or setting relevant to the investigator’s theoretical framework and the problem under investigation. For example, a historian may be interested in explaining why Hitler ordered a certain military action that at first glance appears irrational or out of character for Hitler. Normally, the features that would go into an explanation of Hitler’s military decision making would have nothing to do with such elements of his personal situation as what he had had for dinner, what he had done for entertainment, etc., even though such elements clearly belong to his total life situation. Yet it may be that this particular military decision can be attributed to indigestion or to a nightmare Hitler had had the night before. It is the historian’s stock in trade to provide as rich a reconstruction of the actor’s situation or context or setting as is necessary to make his action rationally comprehensible -that is, adequate to the situation as he saw it.
When the historian becomes the
social scientist his/her interest shifts from explanation specific events to generalized
explanation. His/her interest may turn, for
example, to Hitler’s military decision making style. He abstracts from each particular
instance of Hitler’s military decision making those contextual features common to all these instances. In making such an abstraction, the social scientist inevitably loses much of the
12 СРАВНИТЕЛЬНАЯ ПОЛИТИКА ■ 2 (19) / 2015
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
richness of the historian’s reconstructions. As a result, his generalized reconstruction of Hitler’s military decision making may fail to explain some particular decision, since it will not include the bout of indigestion or the nightmare that was crucial for the explanation of that particular decision.
Such reconstruction of Hitler’s military decision making style would seek to model reality as it actually is. But it would select out for description only those aspects of reality deemed relevant to the explanation of a certain type of problem. The fact that it oversimplifies the situation and thus may not provide a “complete” explanation of any particular instance of Hitler’s military decision making poses no impediment to the ideal of generalizing social science. It merely reflects what is widely accepted at the level of common sense, namely, that society and human behavior exhibit enough orderliness to make social science possible and useful but that they contain enough haphazardness and variation that even a mature social science will have to integrate its organizing theories with this variation and haphazardness.
A generalized reconstruction of Hitler’s military decision making would, of course, be useful to the historian interested in explaining any particular instance of it. The historian would simply enrich his/her reconstruction of Hitler’s behavior in that specific instance to explain whatever problem is imposed by his own perspective. On the other hand, another social scientist might be interested more broadly in military decision making by contemporary European heads of state with dictatorial control of their respective regimes. Such situational reconstructions would lose much of the richness that could be provided in reconstruction of Hitler’s own peculiar military decision making situation, all the detail about Hitler’s personality, knowledge, and theories would be lost. Some similarities relating to typical personality features (if there are any) of contemporary European dictators might remain. The details of German political culture, German military organization, and other features peculiar to Germany under Hitler would be lost. Nevertheless, the similarities of the military decision making situations of European
dictators may share enough features to make such a generalized situational reconstruction theoretically interesting. Again, the model or reconstruction would refer to facts in the real world. In other words, it would refer to facts (concrete features of the decision making situations of concrete individuals) that bear upon their behavior. But because of the level of abstraction of the model, it will naturally not account for all the variance in any given instance.
SOCIAL SCIENCE THEORY
Most of what is called “theory,” as well as “middle range generalizations,” in the social sciences consists of generalized reconstructions of types of social situations or settings or events. These reconstructions may be cast at any level of abstraction, from the historian’s richly detailed account of a particular event to the organization theorist’s highly abstract model of a complex organization. There may be continuity across levels of abstraction. Highly abstract contextually limited generalizations may be used as empirical hypotheses to explain particular events or situations. Conversely, particular events and situations may be generalized. These contextually limited generalizations may be rough, rudimentary, oversimplified, and over-schematized, but, like theories in the natural sciences, they are objective, empirically criticizable, and capable of improvement or competitive comparison with alternative models of the same situation or type of situation.
Both history and the generalizing social sciences take for granted the laws of physics, chemistry, and biology, and many trivial law like generalizations of sociological and psychological character. A historian does not waste time while reconstructing a historical event, such as the assassination of Julius Caesar, spelling out the laws of physics that account for the blood spurting out of Caesar’s arteries, or the laws of biology that account for the death of a human organism. Neither does the generalizing social scientist bother to articulate all the many laws of nature and trivial sociological and psychological generalizations taken for granted in social science theorizing.
COMPARATIVE POLITICS ■ 2 (19) / 2015 13
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
In the explanation of types or kinds of events, initial conditions can be completely replaced by the construction of a model that incorporates typical initial conditions. In other words, a social phenomenon that is puzzling and in need of explanation is explained by showing it to be a special case of a generalized situation (or typical set of initial conditions).
SITUATIONAL ANALYSIS AND THE UNITY AND CONTINUITY OF THE GENERAL AND THE UNIQUE
In genetic and historical explanations, the focus of explanatory interest is almost always uponinitial conditions; the laws or generalizations are usually trivial and are taken for granted. If, for example, the question to be answered is something like, “What caused the cold war?” or “Why does Germany have an authoritarian political culture?” or “Why does a two-party system prevail in the United States?” or “Why did the Soviet Union invade Czechoslovakia in 1968?” the explanation will take the form of a narrative account. Such an account will consist of some combination of statements of fact plus statements of generalizations from which the previously puzzling (i.e. explained) phenomenon can be validly deduced. Such a genetic or historical account may make use of nontrivial generalizations discovered by social scientists, but need not necessarily do so. For example, generalizations taken from theoretical literatures related to the nature and causes of hostility among states may be brought to bear upon the problem of explaining the cold war. Generalizations and theories derived from the study of political culture may be brought to bear upon explanation of authoritarianism in German political culture. And generalizations derived from the study of party systems may be used to explain the two-party system in the United States. But, for the most part, historical and genetic explanations take for granted all kinds of law like generalizations, mare of them being trivial and unstated, and focus on the problem of producing an adequate reconstruction of initial conditions.
Social science theory, as has been shown, reflects an interest in typical settings or initial conditions. Organization theory, role theory, small-group theory, and game theory are only a
14 СРАВНИТЕЛЬНАЯ ПОЛИТИКА ■ 2
few examples of such typical initial conditions that are used as explanations in social science. When, for example, small-group theory is used to explain a specific instance of the behavior of a particular small group, the “explanation” amounts to an assertion that the initial conditions in this instance represent an occurrence of typical initial conditions in small-group theory. Such theories, like maps, attempt to model the social world faithfully. And yet, just as different kinds of maps reflect the differing perspectives and purposes of those who make and use them (e.g. road maps, topographical maps, population maps), so do different bodies of social science theory model the social world from different perspectives and for different purposes.
It is important to note that such models of typical initial conditions may be constructed at any level of generality. For example, the setting of one particular small group, such as one congressional committee or the U.S. Supreme Court, may be reconstructed and such a mode] used to explain specific instances of the group’s behavior. This is, as a matter of fact, just what some scholars who study congressional committees and the Supreme Court do. They seek to model (at least roughly) the institutional setting they study, so that specific instances of the institution’s behavior can be understood and explained by reference to this model. Such generalized settings resemble genetic or historical explanations in that it is mainly initial conditions that do the explaining. But, as in the case of genetic explanations, explanations in terms of generalized social situations include lawlike statements - trivial as well as nontrivial.
(19) / 2015
СРАВНИТЕЛЬНЫЙ АНАЛИЗ КОНЦЕПЦИЙ И ИНСТИТУТОВ
Общее и частное
Региональные исследования, кейс-стади и исторические исследования против теорий, разработанных в рамках социальных наук
Фред Эйдлин, доктор политических наук, приглашенный профессор Европейского колледжа Тартуского университета (Эстония)
Аннотация. В статье рассматривается проблематика соотношения общего и частного в социальных и политических науках. Выделяются различные точки зрения на данную проблематику: часть исследователей объясняет частные случаи при помощи общих теорий и законов, другие же исследователи акцентируют внимание на том, что каждый случай индивидуален и не поддается генерализации. Автор выдвигает теорию о том, что с помощью методологии можно объединитьанализ общего и уникального.
Ключевые слова: региональные исследования, кейс-стади, исторические исследования, социальные науки, политические науки, естественные науки.
Reconciling the General and the Unique
Area Studies, Case Studies, and History versus Theoretical
Social Science
Fred Eidlin, PhD in Political Science (University of Toronto), Visiting Professor, Euro College, University of Tartu, Estonia
Abstract. The article examines the problem of relations between the General and the Unique in the social and political sciences. The author highlights the different views on this perspective: some scientists explain specific cases by bringing them under general theories and laws while others researchers emphasize that each case, each phenomenon is unique and shouldn’t be generalization. The author theorizes that there is a methodological bridge between the generalizing spirit of social science and uniqueness of events and cases.
Key words: Area Studies, Case Studies, History, Social Science, Political Science, Natural Science.
COMPARATIVE POLITICS ■ 2 (19) / 2015 15
