Valiant attempts to define and describe giant cultures have become familiar in the 20th century. Spongier, Toynbee, Sorokin, and Kroeber have each offered impressive systems involving the conception of a number of exclusive, durable, mortal macrocultures that have come to be called "civilizations." That these attempts have aroused considerable interest derives, no doubt, from a feeling that our own civilization might be facing the possibility of coming to an end, of "dying" if you will, as others apparently have in the past.

    But most scholars, while praising the authors for their learning and audacity, have raised central questions about the validity of the whole approach. Do these civilizations, ranging over thousands of miles and years, really have meaningful internal relationships? Should anyone attempt to characterize them as if they were historical personalities? Is it possible to unravel the maze of history and compartmentalize it in this fashion? And if a Spengler or Toynbee does all this, has he discovered anything real or has he only found a way of simplifying history that happens to be convenient for him?

    The answers given to these questions, at the time they were raised, were overwhelmingly negative. The civilizations of Spengler and Toynbee behaved themselves so beautifully because they were fictitious creations. Since they were not real, they could not live or die and they certainly could not have personalities.

    But time has given a different answer. Many historians write today with a sharpened awareness of cultural integration and characterization. They seek relationships between politics, economics, and aesthetics, and they dismiss cause-and-effect political history as out of date, something that belongs to the 19th century What they have rejected in the system builders is their dogmatic periodization. The basic concepts have stood. Civilizations do have meaningful inner relationships, they can be characterized, they can be distinguished from one another.

    In the last year of his life, Rushton Coulborn wrote that a task "of pressing importance for the comparative study of civilized societies ... is the establishment of an outline body of doctrine for the whole field secure enough for all scholars working within it to accept." In the essay that follows I hope to make a beginning on the task this thoughtful scholar believed to be so important. I have tried to sift out the areas of agreement that exist among the students of comparative history who, together, have studied most of the civilizations of the world. What, from these areas of agreement, seem to be the recurrent, the usual, the normal characteristics of most civilizations? What is a normal civilization?

    Coulborn liked to apply Thomas Kuhn's term, paradigm, to a body of doctrine accepted by almost everyone working within a given field. What I am attempting here, of course, might be called a model of civilization. Others may modify this, or propose alternatives. Out of this kind of interaction, a paradigm for the study of civilization could arise.

    Let me stress again that most of the generalizations in this essay are drawn from the observations of comparative historians, and not from an exhaustive study of history itself. In fact, because I do not have an adequate general knowledge of many civilizations, I have tried to set aside my own observations of history or at least to relegate them to the notes. It will, however, be necessary to illustrate some points with examples, which I shall draw from my knowledge of Chinese, Russian, Indian, Egyptian, and Western history . . .

    Each civilization has a history of its own. We might, therefore, refer to those who have considered the history of one civilization in relation to the history of others as comparative historians.

    There have been quite a few comparative historians through the centuries: Orosius, Ibn Khaldun, Bodin, Le Roy, Vico, Danilevsky, Burckhardt, Brooks, and Henry Adams. But there is no doubt that the 20th-century crisis of the two world wars, coupled with significant developments in the methods of archaeology and anthropology, cleared the way for the emergence of the most knowledgeable and interesting comparative historians who have ever lived. Of these the most influential have been Spengler, Toynbee, Sorokin, and Kroeber.

    Oswald Spengler, an obscure German schoolmaster, began his Der Untergang des Abendlandes as an effort to comprehend the political events that preceded World War I. Eventually he enlarged his original scope, believing that politics could only be understood in relation to artistic and philosophical developments. His book is primarily a comparison of Western and Greco-Roman civilizations, with side glances at a half-dozen others. He is dogmatically insistent on the independence of civilizations from external influences. He writes with leaping flashes of intuition which may be exciting, inspiring, annoying, or baffling. His book has been translated by C.E Atkinson as The Decline of the West (1932).

    The British historian, Arnold Toynbee, began publishing Study of History (1934-61) a few years before the outbreak of World War II. Toynbee draws from a wider knowledge of history than Spengler; documents his material more thoroughly, and has at his command an astonishing amount of information on numerous and disparate subjects. He delineates more than twenty civilizations, though even for him Western and Classical civilizations provide the core of his support. He is less concerned than Spengler with characterizing civilizations, more concerned with the criteria by which they are to be determined. The Study is both pleasing and exhausting because of its winding side paths, many carried to footnotes and annexes, where they often lead to observations more significant than the point they are illustrating.

    As Toynbee worked steadily on his staggering one-man task, Russian-born Pitirim Sorokin mobilized a small army of Harvard graduate students to help him tackle similar questions from a sociological viewpoint. He concentrates even more than Spengler and Toynbee on the Classical-Western tradition. His Social and Cultural Dynamics (1937-41) overflows with statistics on all conceivable components of a culture, from types of art and systems of truth to methods of government and practice of war. Sorokin and his students have not only studied them, they have tried to weigh and measure them. Basing his writing on these data, Sorokin is often difficult to read. Fortunately his books are salted with pugnacious overtones, particularly in the footnotes. Though a harsh critic of his fellow comparative historians, he is also a staunch defender of what they are trying to do.

    A.L. Kroeber, born four years before Spenglei; did not publish his Configurations of Culture Growth (1944) until the latter part of World War n (though it was virtually completed nearly a decade earlier). Kroeber is more temperate and less conclusive than the others. He suggests areas for further investigation without trying to supply final answers to his own questions. He has consistently attempted to reconcile conflicting views of his fellow writers in the field without interpreting them, as Sorokin tends to do, in his own terms. He approaches civilizations from an anthropological viewpoint, seeing them as more complex, but not magically different, from simpler cultures. He is the most comfortable of the four in dealing with non-Western cultures. He writes in a mellow style, sometimes overelaborated by his anxiety to qualify and avoid overstatement.

    A number of other 20th-century scholars have made significant contributions to comparative history The British Egyptologist, Flinders Petrie, anticipated Spengler and Toynbee in his bold, scarcely-supported speculations about the nature and development of civilizations, the relations between them, and the nature of progress. Christopher Dawson, British Catholic historian, approaches the subject as a student of comparative religion, avoids constructions of central theories and, like Kroeber, stresses the importance of intermediate cultures. Quincy Wright, drawing heavily on Toynbee for theoretical support, concentrates on the structures of international societies. Shepard Clough focuses on the relationship between the development of civilizations and their economic system. Philip Bagby, an American anthropologist, outlined methods and suggested studies that ought to be undertaken in comparative history, but his early death prevented him from completing any significant study of his own.

    Two other comparative historians, writing more recently, have made important contributions. Rushton Coulborn concentrated on the origin and revival of civilizations. He stressed the tendency of civilizations to endure and recover, the importance of "style" in determining their delineation, and the necessity of remaining receptive to partial comparisons ("uniformities") without imposing a rigid, overall structure. Carroll Quigley has written a concentrated, sophisticated explanation of The Evolution of Civilizations (1961). His style is simple, direct, and often humorous; he is particularly concerned about explaining the correlation in the development of psychic, governmental, and economic aspects of a civilization. Coulborn and Quigley represent a new generation: their tools are more refined than those of their predecessors, and they are less dogmatic. Perhaps because they have been in a position to support themselves with more certainty they are less outrageous and have therefore attracted less attention. This is unfortunate, because they have much to say and there is still much to be said. . . .

    Civilizations are large and complex cultures, usually distinguished from simpler cultures by a greater control of environment, including the practice of agriculture on a large scale and the domestication of animals. They are technically advanced enough to use metals and to employ the wheel for transportation. These economic advantages give them enough of a surplus of food and necessities to free some of their members, at least in part, from subsistence work. This freedom usually leads to the building of cities, and the development of more complex art forms and some kind of writing to convey ideas and to maintain records. For whereas a simpler culture changes so slowly that it is usually studied in static terms, a civilization changes rapidly enough to be considered chronologically: it has a history (Quigley 1961 pp. 69-76).

    Usually civilizations incorporate a multiplicity of cultures and languages. But they have never expanded indefinitely, and it has been possible to distinguish them not only from their component cultures, but also from other civilizations. When Marco Polo traveled to China, he was aware that he was seeing a distinct civilization. The Chinese and Europeans each lived their own lives, and contact between them was rare. There was no society which included Europe and China. They were clearly separate entities. The geographical distinction is not always so deal; but civilizations that form and develop separately tend to remain distinct even after they come into physical contact.

    Toynbee sums up the incorporative but distinct characteristics of civilizations when he describes them as institutions that "comprehend without "being comprehended by others" (1934-61, Vol. 1, p. 455, n. 1).

    These civilizations must have a certain degree of integration. Their parts are defined by their relationship to each other and to the whole. If the civilization is composed of states, these states will have more relation to one another than they do to states outside the civilization. They might fight more, and engage more frequently in diplomatic relations. They will be more interdependent economically There will be pervading aesthetic and philosophical currents.

    The degree of integration that exists will vary from one civilization to another and it will vary within a civilization from time to time. Sometimes the parts will be so closely related that a change in one part will affect all the others, both geographically and in terms of ideas and attitudes. Sometimes the parts will be loosely related, so that a change in one part will have very little effect on the others. These degrees of integration are familiar in other systems. If you pull a man out of a football team, you still have a team that works, less efficiently perhaps, but it does work and is sdll clearly identifiable as the same team. If you knock a neutron out of an atom you get an isotope with considerably different properties. Similarly, the Polish Partition involved the removal of a state, but still the system worked, less efficiently perhaps, but it did work and was clearly identifiable as the same state system. But the Reformation involved a replacement of qualities so integrated in the system that the change led to a reconstitution of the civilization with considerably different properties.

    Civilizations are composed of a multitude of integrated "systems"-regional and provincial systems of government, agricultural and industrial districts-each of which is broken down still further. I am taking this use of the term "system" from Sorokin. Alternatively, using Kroeber's term (1948, p. 311), civilizations can be seen as being composed of "patterns"-systems of art, philosophy, religion that are again broken down into various schools and movements. The patterns are the arrangements that give the parts a relationship to one another and to the civilization as a whole, whereas systems have their own unity, regardless of whether they happen to form a part of a still larger system. The subsystems are like blocks that a child uses to build a castle. The patterns are like strands in a woven rug. Patterns are best studied in relation to the system they compose; we should study Impressionist art with reference to the society in which it existed, although we can study Russian government to see how it functions without concerning ourselves with Russian history . . .

    Marco Polo was struck by the fact that the Chinese ate differently, thought differently, and had different customs from the Europeans. The patterns of their activities and ideas gave distinctive qualities to their respective cultures. Since the patterns relate to one another in a culture, it is usually possible to distinguish the outstanding characteristics of various peoples.

    Concentration on the outstanding differences between Englishmen and Americans enables the social historian to characterize each country Concentration on their similarities enables him to characterize their common culture, and thereby to distinguish the civilization to which they belong from other civilizations. To an American the differences might be important: the Englishmen might seem old-fashioned, remote, charming. To the Chinese, the similarities might be overriding: the Westerners seem technologically-oriented, dynamic, superficial. Individual cultures and whole civilizations thus can be characterized in a few terms or a single word.

    Though we may characterize civilizations as a whole, we can best see the characteristics operating by comparing individuals, preferably individuals who are operating under identical conditions. For example, if you give a volleyball to a group of Western soldiers stationed in Korea, they will form two teams and play a competitive game. If you give the same ball in the same area to Korean soldiers, they will form a circle and kick the ball to one another. The Westerners keep score, play aggressively, achieve recognition by blending their abilities with those of teammates and discuss the game after it is played. The Koreans keep no score, yield readily to other participants, achieve recognition by adroit individual manipulation, and talk little about the game after it is over. The way in which the individuals behave, and the form their recreation takes, reflects their society But at the same time, their culture is what it is because they do these things; the form of their game fits the individuals and the individuals fit the form: each modifies the other. The Westerners say it is "like the Koreans" to play in the manner they do. What they mean is that the Koreans' approach to recreation seems consistent with the approach to other things they do, that they have characteristic ways of behavior.

    All the characteristics of a civilization, then, tend to relate to and modify one another. Nations tend to borrow from one another, developments in art and history in one area tend to be modified by those in other areas, and all these interacting and modifying elements tend to give an image to the civilization as a whole. This image, in turn, pervades the civilization and tends to influence and modify the disparate elements. Once these characteristics become established, they tend to persist through this reciprocal reinforcement, even though the civilization is undergoing momentous change. Thus Spongier and Toynbee take it for granted that the civilization of Homer is the civilization of Diocletian.

    Spengler sees this image as a "soul" that appears at the beginning of a civilization's existence and pervades and directs it throughout. This view has been modified by Kroeber who sees the "soul" only as a generalization about the relatedness of the patterns to the whole (Spengler 1932, pp. 179-80; Kroeber 1957, pp. 101-102). Though I concur with Kroeber's modifications, Spengler certainly has conveyed the early appearance of the image in saying that the soul exists at the "birth" of the civilization: We cannot be sure a civilization is there until we can discern an image.

    Spengler has also been criticized for overdrawing his characterizations and for overstressing their pervasiveness. For him, "pure and limitless space" describes the West, "the sensuously-present individual body" the Greco-Roman, a "wandering way" the Chinese, and so on (1932, pp. 183, 190). But I find these sharply drawn contrasts useful. We all know they are simplifications and that they may have to be modified for different situations, but if civilizations are going to be described at all, we must try to pick out those characteristics that make them unique. A watered-down, excessively elaborated description is difficult to work with. Somerset Maugham has pointed out that if a novelist is too detailed and exact in his characterization, his character appears senseless and inconsistent. We are no longer able to apprehend his image.

    Inevitably characterization is a matter of individual judgment and inevitably it will reflect the personality of that individual. But this is also true, of course, of the writing of any narrative history: When data are plentiful, the observer must select; when they are lacking, he must draw inferences. The danger in characterization is that once it has been made it tends to commit the observer in further observations he may make. If he says that Western civilization is Faustian - that its representatives tend to have an indomitable urge to explore, penetrate, or meddle-he may feel he ought to apologize for individuals who do not fit this characterization, or he may seek further evidence to prove that after all, despite appearances, they do fit. But this is a problem inherent in all hypothetical formulations. Characterizers, model builders, and image-makers must expect their creations to be modified or even destroyed by empirical data. . . .

    If civilizations have an internal consistency, if they have discernible, unique characteristics, then they can be distinguished from one another. Not that they do not interact and collide and occasionally destroy one another. But once they have a chance to develop, once they become sufficiently large and complex, they can withstand a considerable amount of buffeting and still retain their identity One civilization rarely receives material from another without changing the nature of that material to fit its own patterns. Anything that can be transmitted without change is concerned with basic, mechanistic functions - and if such things are not transmitted they may be reinvented anyway when the need arises.

    A good measure of agreement has already been reached on the methods for delineating civilizations, and on when and where these civilizations existed. Disagreements exist on whether more stress should be placed on the existence of specific patterns such as language, religion, technological development, and forms of art, or on the existence of historical processes and distinctive phases of development. Disagreements also persist on the margins of time and space, on whether smaller, less developed, or interrupted cultures should be called civilizations at all, and whether long, irregular periods of history should be studied as one or more civilizations. Out of this discussion separate civilizations are generally distinguished in the following areas:

- the Far East between 2000 BC and the present

- India between 2500 BC and the present

- Egypt between 4000 BC and 300 BC

- the Middle East between 4000 BC and the present

- the Mediterranean between 3000 BC and AD 1500

- Western Europe between AD 700 and the present

- Central America between AD 1 and AD 1600

- Western South America between AD 1 and AD 1600

    Further, there is a pronounced but less frequent tendency to distinguish an Islamic civilization around the southern Mediterranean between 500 and the present, an Orthodox civilization in eastern Europe at roughly the same period, and a civilization in Japan, since possibly 400 BC, that has been sufficiently distinct from China to merit separate classification.

    The delineation of civilizations is usually unsatisfactory, often esoteric and sometimes rather quibbly, but it is important because in any attempt to portray the history of a civilization it is necessary to understand the reasons for its limits and divisions; because a measure of agreement on civilizations by the comparative historians will make their own writing more useful for comparison with one another; and because generalizations about recurrence will have more meaning when the structures of the civilizations in which recurrences take place are better understood.

    The relationship between historical recurrences and civilizations is particularly important. If we are interested in Napoleon, we might explain his failure to unify Europe in terms of his own shortcomings, or in terms of a peculiar recuperative vitality manifested in Europe. But was it the man or the situation that prevented unification? Or was it a fortuitous combination of both? We begin to look for other examples in history Alexander also failed to create a long-standing empire. Was this because he and Napoleon faced similar situations? And how was it that these two most famous and gifted men failed to do what an Augustus, a Chandragupta, or an Ivan III did succeed in doing? If we study the careers of these men we can come up with some answers that probably will throw more light on the career of Napoleon. But we shall have a deeper understanding still if we also try to understand the context of the civilization in which the empire builders lived, if we can discern whether they were, to use one of Toynbee's more preposterous phrases, "philosophically contemporaneous."

    In discussing the delineation of civilizations, there is a tendency to be apologetic because there are so few examples. Toynbee is sorry because he can find only 30 while the lucky entomologist has all those millions of specimens to work on. He asks us to be patient because in a few hundred thousand years, if all goes well, which (he says) it probably will not, we shall have many more samples. More anthropologically-oriented historians, on the other hand, think we can partly make up for this difficulty by making studies of intermediate and primitive cultures, which will he easier to handle and will throw more light on methods of study in tackling the larger systems (Kroeber 1957, pp. 158-59; Dawson 1957, p. 425).

    This is all very well when you are making suggestions for other people to carry out, but when you are thinking of tackling these problems yourself it becomes frightening. Toynbee's maximum of 30 civilizations, or even the assessments of Spongier, Kroeber, Coulborn, and Quigley at between 8 and 15 major civilizations, are already enough to occupy anyone for a lifetime. Anyone now living was born none too soon. As for the minor and derivative and humble cultures, they will be useful in time for criticism and verification. But, understandably, comparative historians have been gravitating toward Spengler's intuitive generalizations rather than Toynbee's perhaps more sophisticated delineations. There are not too few civilizations, there are too many-for any one man. So let the pioneers draw their material from an insufficient sample, as Spengler and Toynbee have done, and let these hypotheses be modified or destroyed by the detailed work of scholars, by reconsideration, and by time. . . .

    The extent of consensus thus far attained might he gauged by comparing the delineations of six comparative historians who have been most explicit on the subject (see Table 1.1). In Table 1.1,1 have taken the liberty of giving a date (preceded by 'c.') where the writer has indicated perhaps no more than a millennium. A question mark following a date indicates that the date is inferred. A question mark without a date indicates that no inference was attempted.

    Coulborn's datings in primary civilizations are invariably earlier than those of the others, probably because he is trying to indicate where "civilization" has its origins rather than determine whether "civilizations" could be said to exist. Toynbee, Kroeber; Bagby, and Quigley would agree, I imagine, that high-level cultures existed in these periods, but they would either deny that they had attained the level of civilization, or, more likely, contend that the classification of these periods is still uncertain.

    Some comments on the individual areas may be helpful in decoding the chart.

    The Far East. Toynbee dates the Sink civilization from 1500 BC to AD 172 and the Far Eastern civilization from AD 500 to 1853. Quigley makes a similar division for his two civilizations: the Sinic from 2000 EC to AD 400, and the Chinese from 400 to AD 1930. Toynbee thinks the separation of modem Chinese civilization is open to question (1934-61, Vol. XH, pp. 173-83). He and probably some of the others would agree with Coulborn that a high level of culture existed in China before 1500 BC, but the question seems to be whether this culture should be classified as a civilization. Spongier rejects the separate classification of Japan (1932, Vol. II, p. 49); Bagby thinks the classification is still open to question; Coulborn apparently includes it with the Chinese.

    India. Toynbee adds a second civilization, the Hindu (AD 800 to the present), but he also grants this is debatable. The disagreement on whether there was an Indian civilization as far back as the 3rd and 4th millennium is similar to the disagreement on the dating of the Chinese.

    The Middle East. This seems to be the most difficult area to classify Kroeber apparently would have included an "Ancient Near Eastern Civilization" in an unfinished "roster" published after his death (1962, p. 21). This would probably have corresponded to the Mesopotamian-Sumerian-Babylonian civilization distinguished by others. Toynbee's Sumero-Akkadian civilization represents a revised view developed in Re-considerations. Toynbee and Quigley distinguish a separate Hittite civilization in the 2nd millennium BC, Kroeber and Coulborn consider the Jews to have had a separate civilization beginning in the 2nd millennium BC, while Quigley includes the Jews, Phoenicians, and Carthaginians in a Canaanite civilization (2200-100 BC). Kroeber and Coulborn distinguish an Iranian civilization beginning in the 1st millennium BC. Toynbee discerns a Syriac civilization that seems to have specialized in hatching religions (1100BC-AD969).

    Egypt. Considerable agreement here. No one seems to have accepted Petrie's view that successive civilizations existed in Egypt.

    Mediterranean. Recent research on Crete indicates that there were Mycenaean influences both on Crete and on Greece. At present, however, the culture of Crete seems to be considered as distinct from that of Greece. Spengler regards the civilization of Crete as an Egyptian offshoot, a view unsupported by later scholars. Bagby considers it a "secondary" culture, not a civilization. Agreement on Classical civilization is fairly close, though Kroeber includes the Byzantine civilization with the Roman.

    Eastern Europe. Spenglec, Kroebec, and Bagby incorporate Byzantium in their Magian, Mediterranean, and Near Eastern civilizations respectively Russia gets mixed reviews. It is treated as a separate civilization by Spengler and Toynbee, as part of Byzantine civilization by Coulborn and Quigley, as European by Kroebci; and as uncertain by Bagby Coulbom's last position was to regard the Byzantine as a separate civilization.

    Western Europe. The only disagreement here is on whether to include the Dark Ages or to start Western civilization at some point where recovery has begun.

    America. Originally Toynbee distinguished three civilizations in Central America. He reduced them to one in his Reconsidemtwns (1934-61, Vol. XII, pp. 173-83). According to Coulborn, Kroeber apparently concluded, after the publication of Configurations, that there was enough evidence to warrant delineation of civilizations in Central and South America. In his revision of Anthropology (1948) he still refers to Meso-American and Andean "high cultures" (pp. 777-801). Spengler never elaborates on Peruvian civilization. He couples it once with the Mexican (1932, Vol. II, p. 46), but he never classifies it....

    Although civilizations change continuously, they maintain their identity for centuries. The change comes about because the civilization is a going, functioning system. When the actions of men break restraining customs and set off processes of development in part of the system, this in turn sets off related development in other parts of the system.

    Despite change, civilizations maintain their identity once well-integrated patterns have been established. If changes are to be induced, they must have some relationship to these patterns, and leaders who fail to take these patterns into consideration are likely to be replaced.

    It seems natural in describing these processes to fall back on organic terms: to talk about the growth or unfolding of a civilization on the one hand and about its exhaustion and death on the other. This gives pain to many students of society who feel that after all, Spengler notwithstanding, cultures arc not organisms (Spengler 1932, p. 104). These terms arise, however not only because no better ones arc available but also because the analogies arc so tempting. Men, like cultures, change constantly yet maintain their identity Yet many do seem to maintain only a steady state, to go on living without becoming. If they do develop, it is because something inside them drives them to it or because qualities they have are allowed to develop by the culture in which they live. Once a man's patterns become established, it is difficult for him or anyone else to change them. Eventually a man runs down, disintegrates, dies, or is overwhelmed by external circumstances or killed by accident or design.

    Civilizations are living systems; men are living systems too.

    Often the characteristics of a civilization become manifest rather rapidly, so that Spengler could speak of the birth of its soul. The appearance of Gothic cathedrals in the 11th century dramatically commemorates the birth of Paustian Man. What really has happened, of course, is that we have identified a civilization through certain characteristics, we trace these characteristics back as far as we can, and the earliest point at which we think they emerge we call the beginning a civilization. When the civilization has begun to develop, it may continue to exist long after the cultures from which it originally sprang have lost their identity through mergei; division, disintegration, or destruction. Thus the Jewish civilization survived the Mesopotamian, the Byzantine survived the Classical.

    The confluence of patterns that had to do with the origins of the first civilizations may have been related to environmental challenges, but there is still considerable uncertainty about the nature of these challenges, and disagreement about whether civilizations were invented more than once. But there must have been some compelling reason for man to change his way of life, moving from a nomadic existence based primarily on hunting to a sedentary agricultural life in the groat river valleys of the world (Coulbom 1959, pp. 67-109, modified in "Structure and Process in the Rise and Fall of Civilized Societies" [1966], p. 410). The challenge involved in this transition must have been immensely difficult, requiring astonishing changes in political and economic concepts, and probably producing many failures before two civilizations - the Sumerian and the Egyptian - emerged nearly at the same time. Toynbee suggests a possible human challenge: the migration of peoples (which might have an environmental origin) often forces another people to move and face the challenge of a new situation, or else produces an intermixture of peoples that sometimes precedes the flowering of a new civilization (1946-57, Vol. 1, pp. 75-79; also Dawson 1957,p.8).

    In view of the importance of understanding the nature of patterns, you might suppose that there would be a great deal of concern about the origins or causes of civilizations. This has not been the case. There has been a tendency to avoid the study of origins that derives, I suspect, from the comparative historians' rejection of the narrative historians' cause-and-effect approach to history The historians of that period tended to look for the cause of World War I in the sequence of diplomatic events in the years preceding, without giving sufficient consideration to the social atmosphere of the entire civilization. The implication of looking for cause in preceding events was that ultimate cause was to be found in whatever came first-the origin of the situation-rather than in overall relationships. But more recently, as narrative historians have responded more to rational and less to sequential cause, comparative historians like Coulbom and Quigley seem to have been able to pursue the study of origins without apology

    It is clear that the understanding of a particular aspect of a culture is bound up with the understanding of the culture itself. And if we can see it in its formative phase, we can distinguish better the patterns that arc inclined to become more elaborated and more obscured in later phases. Moreover, if you are studying parallel developments in different cultures, anomalies may sometimes be explained not in the contemporary events, but in pattern variations that derive from formative periods. Thus the Russian Revolution in many aspects followed paths analogous to earlier European revolutions. We should expect, therefore, to learn a good deal about the Russian Revolution by comparing it, as Crane Brinton has done, with the French and the English. But in many respects the society emerging from the Russian Revolution differed considerably from the France of the 19th century These differences derive partly from the intervening development of the industrial revolution, but also partly from differences in the patterns of Russian and French history

    Some form of ritualized religion is repeatedly associated with the early stage of cultural development. Comparative historians agree almost unanimously on its dominant role in the formation of cultures. This is as true of those who, like Coulborn and Spongier, regard religion purely as a factor to be considered in the study of societies, as it is of those who, like Toynbee and Dawson, regard religion as "the foundations on which the great civilizations rest." Some kind of theocratic leadership seems to be necessary to inspire the support of the members of a culture that is still too fragile to withstand internal disunity Often this religious unity seems to be closely related to an economic system that places strong emphasis on the value of the land and on family relationships. Sometimes it is closely connected with government, as in the Iranian and Chinese civilizations; at others distinct, as in the Classical and Western.

    And if religion forms the unifying element in most developing or reviving civilizations, the nature of that religion will surely influence the process of secularization that accompanies a civilization's development. Secularization-the gradual freeing of patterns from the dominance of the original ritualized religion - seems to be a recurrent and necessary process. This process, which may last many centuries, involves the freeing of all political, economic, and aesthetic patterns from their close ties with religion. Without secularization, civilization cannot develop. Apparently no religious crystallization or synthesis can be maintained except at the cost of internal ossification, the smothering of all processes of development by an autocratic priesthood.

    The study of origins, then, and particularly the study of religion will frequently have tremendous relevance to the understanding of secular problems developing in later phases. And comparisons of secular problems will frequently be clarified by comparisons of origins. . . .

    Once a civilization has achieved a measure of coherence, with an established relationship between its components, potentialities for change become increasingly limited. This applies to systems at all levels and to the patterns that compose them. The limits of a particular civilization depend largely but not entirely on the character it assumed in its formative period. Once established, the various patterns of a civilization tend to develop in relation to one another until each has achieved its potential.

    For example, the development of geometry was the Greek style of mathematics, a pattern in their culture. It appears to be related to the Greek emphasis on proportion and their preference for visible and tangible bodies-hence the preference for integral numbers and the avoidance of negative numbers and fractions. It is easy to see that once the full possibilities of geometry are reached, creativity in the field must die out, or mathematical activity must shift to some other form, or some way must be found to add new elements to geometry that will give scope for further development. If such additions and changes were made, however, the resulting configurations might bear very little relation to what had been called geometry They would come to be classified under a new name and to be thought of as components of a different kind of mathematics.

    But if geometry develops in an integrated society, if it is one manifestation of a Greco-Roman preference for tangibly present forms, related to the nude statue, the "sensuous cult of the Olympian gods," and the politically individual city-states (Spengler 1932, p. 183), it is likely to change only if the whole culture is likewise in a process of transition. Otherwise, if new elements were added, geometry would lose its relationship to its culture. Such changes do take place, but rarely unless the culture as a whole is on the threshold of disintegration. In the case of the Greeks, the changes did not take place. "What they would do with their geometrical and whole-number manner of style, they achieved. Other mathematical possibilities . . . were simply left to be realized by other peoples and other times" (Kroeber 1948, p. 330).

    The geometry pattern, then, is inherently limited. Clearly all patterns tend to reach culminations unless new material is gradually and constantly added. And when the limits of possibilities are sensed, there is likely to be some casting around. After a vein or complex of veins has been developed, miners face a dilemma. They can explore subordinate, low-yield areas further; they can go back and try to find some ore overlooked when the major veins were freely yielding; they can look for new veins; they can close the mine and dig another; they can give up.

    Art patterns arc somewhat similar to the patterns of a mine. After an idea or major complex of ideas has been developed and explored, the developing artist is faced with a dilemma. Whereas fifty years previously artists were creating freely and prolifically, now they find that the additions they can make to the existing pattern arc of a secondary or claborative nature. Some will content themselves with doing this, some will repeat the patterns of immediate predecessors, some will return to earlier periods hoping that when development came, some possibilities were overlooked. Some will cast about, will feel uncomfortable, will experiment, will challenge the old patterns instead of trying to develop them further, Some will give up.

    Individual culminations of patterns, occurring in succession, form a culmination for the civilization as a whole. The culmination tends to come rather early in the life of the pattern, long before the creative phase has completely ended. Manifestations of creativeness appear long after the underlying factors that give rise to them have begun to change. In China, for instance, the Hundred Schools of philosophy (c. 500-300 BC) had given form to the civilization's outlook before the first of a series of great empires was formed. Sculpture had two important periods of development during the Han (200 BC-AD 100) and T'ang (AD 500-600) Empires, poetry reaching a long, high active period (AD 200-800) through the time of the Three Kingdoms and the T'ang, painting reaching a peak during the period of the Sung (AD 950-1100). Though there is development in drama and the novel in post-Sung periods, they seem less dynamic than the earlier periods. By the time of the highly refined Sung period, decline was setting in, but as early as the Three Kingdoms period most of the patterns had been set, and the philosophical patterns, which proved to be central to Chinese style, had been clearly delineated nearly a millennium before general decline had set in.

    Once a pattern or series of related patterns have passed a culmination, three possibilities remain open: they may disintegrate and disappear; they may become fixed in a steady state, or they may experience a period of transition in which disintegration takes place while new material is being added, before the onset of further development. Civilizations, like lesser systems, face these alternatives: they either disintegrate, ossify, or reconstitute themselves and develop further

    In all systems, in all patterns, there are forces working both for integration and for disintegration. In earlier periods of development, the former dominate until the culminating point of a given pattern or system. If the subsequent disintegration continues unchecked in a civilization for a long; period, the civilization may cease to exist. It dies. Between AD 100 and 700 the Roman Empire virtually disappeared from the western Mediterranean. The culture in this area changed so rapidly that hardly anyone would find in it a unity and continuity of existence. The lesser patterns that compose a civilization of course disintegrate and disappear frequently, even when its total processes may tend toward integration and unity

    It may happen that the process of integration is checked through the ossification of the most significant patterns. A strong central government, by maintaining the system as it is without permitting normal changes, may enable the civilization to hold the line for a very long time. But it will be sterile, its forms endlessly repeated, its creativity dried up, its activities without meaning beyond mere survival. This appears to have happened to the Egyptian civilization and more recently in the Chinese and Islamic civilizations. In a civilization capable of strong centralized control of political and economic functions, it can happen again. Ossification occurs far more commonly within the subsystems of a civilization. These often become so overelaborated that they are no longer able to function, but they may continue to exist in a ceremonial capacity while their functions are taken over by other systems that have relevance to a particular problem. The changes in the relationships between the British monarchy and parliament, or between the Japanese emperor and shogunate, will serve as examples.

    The process of disintegration may be resolved by internal effort, or by the impingement from outside of new material and new attitudes, probably resulting in the violent collapse of the old, unadapted framework. But then, because new answers are being sought, or because the internal disruption makes possible the absorption of alien ideas and artifacts, a new period of development may take place. What appeared to be a disastrous disintegration of old patterns turns out to be the creation of a condition from which a new and perhaps greater development can take place. The collapse and recovery of the T'ang empire and the emergence of the Western Renaissance are examples of reconstitutions of this kind. If men had adhered to cherished Confucian or Christian patterns, their respective civilizations might have disappeared. As it was, they were able to make a transition to unfamiliar patterns, the pattern of Chinese empire was given a viable foundation, the Renaissance took place, and the modern phase of Western history became one of fresh development.

    This concept of reconstitution raises hob with those of us who are trying to arrange history in an orderly manner. If disintegrating civilizations can recover and go on to further phases of development, it is obviously going to be very difficult to cut them into orderly segments, with this phase lasting 312 years and that phase 435 years. It is going to be difficult to make accurate predictions about the doom of those still in existence. Revelation can do that just about as accurately and far more impressively To make matters worse, the frequent intrusion of alien civilizations in these periods makes it difficult to decide whether recovery has taken place or whether the disintegrating civilization has simply been replaced by another. Is the Babylonian civilization a separate entity or a second phase of Sumerian civilization? To describe the fall of the Roman Empire must you, as Gibbon did, tell the story of Byzantium as well?

    What does emerge from this concept, though, is a striking similarity between formative periods and periods of disintegration. In both, things are in flux, and while men have great opportunities to share their destinies, they also face great risk of failure. And men pay for the glories, achievements, and security of periods of integration with the loss of their capacity to do much more than manage a going concern. They can achieve great things within the system, but they cannot change the system. So yop can say that there are periods of high integration and periods of low integration. In the latter there are both disintegrating and formative forces in action, and only in retrospect can you be certain which proved dominant.

    After all these modifications, how much is it possible to generalize about the formation and development of civilizations? Perhaps a symphonic analogy would be useful: The theme or themes of a symphony are roughly equivalent to the central idea or the symbols by which we characterize civilizations. The completion of the exposition is equivalent to the culmination of a civilization. The development is equivalent to disintegration, in which the original compact material is exploited, considered, elaborated, but little new is added. The recapitulation may be compared to those aspects of development that have to do with the reexaminarion of earlier times, the seeking for the restoration of basic foundations. The coda might represent the final collapse, the end, or it might prove to be a codetta leading to new themes, enabling new combinations and development to take place, the equivalent of reconstitution. If you were listening to a familiar symphony you would know what the coda signified, but if you were hearing it for the first time, you would have to wait and see. Should the development be carried too far, without recapitulation or coda, the thematic material becoming more worn out and repetitive, the symphony, like some civilizations, would become ossified, losing its interest. If the terminating coda is followed after a pause by a new movement beginning new themes in a new time and key, this is the equivalent of a heroic age ... in an affiliated civilization, the formation of a new set of patterns.

    Civilizations, like symphonies, retain characteristic patterns notwithstanding fluxes of formation, disintegration, and reconstitution. . . .

References

Bagby, Philip. 1958. Culture and History. London: Longmans, Green. Coulborn, Rushton. 1959. The Origin of Civilized Societies. Princeton, NJ: Princeton University Press.

----. 1966. "Structure and Process in the Rise and Fall of Civilized Societies."

Comparative Studies in Society and History 8:404-31. Dawson, Christopher. 1957. The Dynamics of Wirld History. London: Sheed and

Wild. Kroeber; A.L. 1944. Configurations of Culture Growth. Berkeley: University of

California Press.

----. 1948. Anthropology. 2nded. New York: Hal-court, Brace.

----. 1957. Style and Civilizations. Ithaca, NY: Comell University Press.

----. 1962. A Roster of Civilizations and Culture. New York: Viking Fund Publications.

Quigley, Carroll. 1961. The Evolution of Civilizations. New York: Macmillan.

Sorokin, Pitirim. 1937-41. Social and Cultural Dynamics. 4 vols. New York: American Book Co. (Revised and abridged by the author in one volume, Boston: Porter Sangent, 1957.)

Spengler, Oswald. 1932. Decline of the West. Trans. C.E Atkinson. London: Alien and Unwin.

Toynbee, Arnold J. 1934-61. A Study of History. 12 volumes. London: Oxford University Press. (First 10 volumes abridged by D.C. Somervell, Oxford University Press, 1946-57.)