Chapter One: Outside and Inside History

This paper was given as a lecture opening the academic year 1993-4 at the Central European University in Budapest, that is to say it was addressed to a body of students essentially drawn from the formerly communist countries in Europe and the former USSR. It was subsequently published as `The New Threat to History' in the New York Review of Books, 16 December 1992, pp. 62-5, and, in translation, in a number of other countries.

It is an honour to be asked to open this academic year of the Central European University. It is also a curious sensation to do so, since, though I am a second-generation English-born British citizen, I am also a central European. Indeed, as a Jew I am one of the characteristic members of the central European diaspora of peoples. My grandfather came to London from Warsaw. My mother was Viennese, and so is my wife, though she now speaks better Italian than German. My wife's mother still spoke Hungarian as a little girl and her parents, at one stage of their lives in the old monarchy, had a store in Hercegovina. My wife and I once went to Mostar to trace it, in the days when there was still peace in that unhappy part of the Balkans. I have had some connections with Hungarian historians myself in the old days. So I come to you as an outsider who is also, in an oblique way, an insider. What can I say to you?

I want to say three things to you.

The first concerns central and eastern Europe. If you come from there, and I assume that almost all of you do, you are citizens of countries whose status is doubly uncertain. I am not claiming that uncertainty is a monopoly of central and east Europeans. It is probably more universal today than ever. Nevertheless, your horizon is particularly cloudy. In my own lifetime every country in your part of Europe has been overrun by war, conquered, occupied, liberated and reoccupied. Every state in it has a different shape from the one it had when I was born. Only six of the twenty-three states which now fill the map between Trieste and the Urals were in existence at the time of my birth, or would have been if they had not been occupied by some army: Russia, Rumania, Bulgaria, Albania, Greece and Turkey, for neither post-1918 Austria nor post-1918 Hungary is really comparable to Habsburg Hungary and Cisleithania. Several came into existence after the First World War, even more since 1989. They include several countries which had never in history had the status of independent statehood in the modern sense, or which had it briefly--for a year or two, for a decade or two--and then lost it, though some have since regained it: the three little Baltic states, Belarus, Ukraine, Slovakia, Moldova, Slovenia, Croatia, Macedonia, not to go further eastwards. Some were born and died in my lifetime, like Yugoslavia and Czechoslovakia. It is perfectly common for the elderly inhabitant of some central European city to have had, successively, the identity documents of three states. A person of my age from Lemberg or Czernowitz has lived under four states, not counting wartime occupations; a man from Munkacs may well have lived under five, if we count the momentary autonomy of Podkarpatska Rus in 1938. In more civilized times, as in 1919, he or she might have been given the option which new citizenship to choose, but since the Second World War he or she has been more likely to be either forcibly expelled or forcibly integrated into the new state. Where does a central and eastern European belong? Who is he or she? The question has been a real one for great numbers of them, and it still is. In some countries it is a question of life and death, in almost all it affects and sometimes determines their legal status and life-chances.

However, there is another and more collective uncertainty. The bulk of central and eastern Europe belongs to that part of the world for which diplomats and United Nations experts since 1945 have tried to devise polite euphemisms: `under-developed' or `developing', that is to say, relatively or absolutely poor and backward. In some respects there is no sharp line between the two Europes, but rather a slope to the east and to the west of what we might call the main mountain-range or crest of European economic and cultural dynamism, which ran from north Italy across the Alps to northern France and the Low Countries, and was prolonged across the Channel into England. It can be traced in the medieval trade routes and the distribution map of gothic architecture, as well as in the figures for the regional GDP within the European Community. In fact, today this region is still the backbone of the European Community. However, insofar as there is a historical line separating `advanced' from `backward' Europe it ran, roughly, through the middle of the Habsburg Empire. I know that people are sensitive in these matters. Ljubljana thinks of itself as a great deal nearer the centre of civilization than, say, Skopje, and Budapest than Belgrade, and the present government in Prague does not even wish to be called `central-European' for fear of being contaminated by contact with the East. It insists that it belongs exclusively to the West. However, my point is that no country or region in central and eastern Europe thought of itself as being at that centre. All looked somewhere else for a model of how really to be advanced and modern, even, I suspect, the educated middle class of Vienna, Budapest and Prague. They looked to Paris and London, just as the intellectuals of Belgrade and Ruse looked to Vienna--even though by most accepted standards the present Czech Republic and parts of the present Austria formed part of the advanced industrial part of Europe, and culturally Vienna, Budapest and Prague had no reason at all to feel inferior to anyone else.

The history of backward countries in the nineteenth and twentieth centuries is the history of trying to catch up with the more advanced world by imitating it. The nineteenth-century Japanese took Europe as their model, the west Europeans after the Second World War imitated the American economy. The story of central and eastern Europe in the twentieth century is, broadly, that of trying to catch up by following several models one after the other and failing. After 1918, when most of the successor countries were new, the model was Western democracy and economic liberalism. President Wilson--is the main station in Prague named after him again?--was the region's patron saint, except for the Bolsheviks who went their own way. (Actually, they too had foreign models: Rathenau and Henry Ford.) This did not work. The model broke down politically and economically in the 1920s and 1930s. The Great Depression eventually broke multinational democracy even in Czechoslovakia. A number of these countries then briefly tried or flirted with the fascist model, which looked like the economic and political success story of the 1930s. (We are inclined to forget that Nazi Germany was remarkably successful in overcoming the Great Depression.) Integration in a Great German economic system did not work either. Germany was defeated.

After 1945 most of these countries chose, or found themselves being made to choose, the Bolshevik model, which was essentially a model for modernizing backward agrarian economies by planned industrial revolution. It was therefore never relevant to what is now the Czech Republic and to what was until 1989 the German Democratic Republic, but it was relevant to most of the region, including the USSR. I do not have to tell you about the economic deficiencies and flaws of the system, which eventually led to its breakdown, and still less about the intolerable, the increasingly intolerable political systems it imposed on central and eastern Europe. Still less do I have to remind you of the incredible sufferings it imposed on the peoples of the former USSR, particularly in the iron age of Joseph Stalin. And yet I must say, although many of you will not welcome my saying so, that up to a point it worked better than anything since the break-up of the monarchies in 1918. For the common citizens of the more backward countries in the region--say Slovakia and much of the Balkan peninsula--it was probably the best period in their history. It broke down because economically the system became increasingly rigid and unworkable, and especially because it proved virtually incapable of generating or making economic use of innovation, quite apart from stifling intellectual originality. Moreover, it became impossible to hide the fact from the local populations that other countries had made far more material progress than the socialist ones. If you prefer putting it another way, it broke down because ordinary citizens were indifferent or hostile, and because the regimes themselves had lost faith in what they were pretending to do. Still, however you look at it, it failed in the most spectacular manner in 1989-91.

And now? There is another model which everyone rushes to follow, parliamentary democracy in politics and the extremes of free-market capitalism in economics. In the present form it is not really a model, but chiefly a reaction against what has gone before. It may settle down to become something more workable--if it is allowed to settle down. However, even if it were to do so, in the light of history since 1918 there is not much likelihood that this region, possibly with marginal exceptions, will succeed in joining the club of the `really' advanced and up-to-date countries. The results of imitating President Reagan and Mrs Thatcher have proved disappointing even in countries which have not been laid waste in civil war, chaos and anarchy. I should add that the results of following the Reagan--Thatcher model in the countries of its origin have not been brilliantly successful either, if you will permit a British understatement.

So, on the whole, the people of central and eastern Europe will go on living in countries disappointed in their past, probably largely disappointed with their present, and uncertain about their future. This is a very dangerous situation. People will look for someone to blame for their failures and insecurities. The movements and ideologies most likely to benefit from this mood are not, at least in this generation, those which want a return to some version of the days before 1989. They are more likely to be movements inspired by xenophobic nationalism and intolerance. The easiest thing is always to blame the strangers.

This brings me to my second and main point, which is much more directly relevant to the work of a university, or at least to that part of the work which concerns me as a historian and university teacher. For history is the raw material for nationalist or ethnic or fundamentalist ideologies, as poppies are the raw material for heroin addiction. The past is an essential element, perhaps the essential element, in these ideologies. If there is no suitable past, it can always be invented. Indeed, in the nature of things there is usually no entirely suitable past, because the phenomenon these ideologies claim to justify is not ancient or eternal but historically novel. This applies both to religious fundamentalism in its current versions--the Ayatollah Khomeini's version of an Islamic state is no older than the early 1970s--and to contemporary nationalism. The past legitimizes. The past gives a more glorious background to a present that doesn't have much to celebrate. I recall seeing somewhere a study of the ancient civilization of the cities of the Indus valley with the title Five Thousand Years of Pakistan. Pakistan was not even thought of before 1932-3, when the name was invented by some student militants. It did not become a serious political demand until 1940. As a state it has existed only since 1947. There is no evidence of any more connection between the civilization of Mohenjo Daro and the current rulers of Islamabad than there is of a connection between the Trojan War and the government in Ankara, which is at present claiming the return, if only for the first public exhibition, of Schliemann's treasure of King Priam of Troy. But 5,000 years of Pakistan somehow sounds better than forty-six years of Pakistan.

In this situation historians find themselves in the unexpected role of political actors. I used to think that the profession of history, unlike that of, say, nuclear physics, could at least do no harm. Now I know it can. Our studies can turn into bomb factories like the workshops in which the IRA has learned to transform chemical fertilizer into an explosive. This state of affairs affects us in two ways. We have a responsibility to historical facts in general, and for criticizing the politico-ideological abuse of history in particular.

I need say little about the first of these responsibilities. I would not have to say anything, but for two developments. One is the current fashion for novelists to base their plots on recorded reality rather than inventing them, thus fudging the border between historical fact and fiction. The other is the rise of `postmodernist' intellectual fashions in Western universities, particularly in departments of literature and anthropology, which imply that all `facts' claiming objective existence are simply intellectual constructions--in short, that there is no clear difference between fact and fiction. But there is, and for historians, even for the most militantly anti-positivist ones among us, the ability to distinguish between the two is absolutely fundamental. We cannot invent our facts. Either Elvis Presley is dead or he isn't. The question can be answered unambiguously on the basis of evidence, insofar as reliable evidence is available, which is sometimes the case. Either the present Turkish government, which denies the attempted genocide of the Armenians in 1915, is right or it is not. Most of us would dismiss any denial of this massacre from serious historical discourse, although there is no equally unambiguous way to choose between different ways of interpreting the phenomenon or fitting it into the wider context of history. Recently Hindu zealots destroyed a mosque in Aodhya, ostensibly on the grounds that the mosque had been imposed by the Muslim Moghul conqueror Babur on the Hindus in a particularly sacred location which marked the birthplace of the god Rama. My colleagues and friends in the Indian universities published a study showing (a) that nobody until the nineteenth century had suggested that Aodhya was the birthplace of Rama and (b) that the mosque was almost certainly not built in the time of Babur. I wish I could say that this has had much effect on the rise of the Hindu party which provoked the incident, but at least they did their duty as historians, for the benefit of those who can read and are exposed to the propaganda of intolerance now and in the future. Let us do ours.

Few of the ideologies of intolerance are based on simple lies or fictions for which no evidence exists. After all, there was a battle of Kosovo in 1389, the Serb warriors and their allies were defeated by the Turks, and this did leave deep scars on the popular memory of the Serbs, although it does not follow that this justifies the oppression of the Albanians, who now form 90 per cent of the region's population, or the Serb claim that the land is essentially theirs. Denmark does not claim the large part of eastern England which was settled and ruled by Danes before the eleventh century, which continued to be known as the Danelaw and whose village names are still philologically Danish.

The most usual ideological abuse of history is based on anachronism rather than lies. Greek nationalism refuses Macedonia even the right to its name on the grounds that all Macedonia is essentially Greek and part of a Greek nation-state, presumably ever since the father of Alexander the Great, King of Macedonia, become the ruler of the Greek lands on the Balkan peninsula. Like everything about Macedonia, this is a far from a purely academic matter, but it takes a lot of courage for a Greek intellectual to say that, historically speaking, it is nonsense. There was no Greek nation-state or any other single political entity for the Greeks in the fourth century BC, the Macedonian Empire was nothing like a Greek or any other modern nation-state, and in any case it is highly probable that the ancient Greeks regarded the Macedonian rulers, as they did their later Roman rulers, as barbarians and not as Greeks, though they were doubtless too polite or cautious to say so. Moreover, Macedonia is historically such an inextricable mixture of ethnicities--not for nothing has it given its name to French mixed-fruit salads (macedoine)--that any attempt to identify it with a single nationality cannot be correct. In fairness, the extremes of emigrant Macedonian nationalism should also be dismissed for the same reason, as should all the publications in Croatia which somehow try to turn Zvonimir the Great into the ancestor of President Tudjman. But it is difficult to stand up against the inventors of a national schoolbook history, although there are historians in Zagreb University, whom I am proud to count as friends, who have the courage to do so.

These and many other attempts to replace history by myth and invention are not merely bad intellectual jokes. After all, they can determine what goes into schoolbooks, as the Japanese authorities knew, when they insisted on a sanitized version of the Japanese war in China for use in Japanese classrooms. Myth and invention are essential to the politics of identity by which groups of people today, defining themselves by ethnicity, religion or the past or present borders of states, try to find some certainty in an uncertain and shaking world by saying, `We are different from and better than the Others.' They are our concern in the universities because the people who formulate those myths and inventions are educated people: schoolteachers lay and clerical, professors (not many, I hope), journalists, television and radio producers. Today most of them will have gone to some university. Make no mistake about it. History is not ancestral memory or collective tradition. It is what people learned from priests, schoolmasters, the writers of history books and the compilers of magazine articles and television programmes. It is very important for historians to remember their responsibility, which is, above all, to stand aside from the passions of identity politics--even if we feel them also. After all, we are human beings too.

How serious an affair this may be is shown in a recent article by the Israeli writer Amos Elon about the way in which the genocide of the Jews by Hitler has been turned into a legitimizing myth for the existence of the state of Israel. More than this: in the years of right-wing government it was turned into a sort of national ritual assertion of Israeli state identity and superiority and a central item of the official system of national beliefs, alongside God. Elon, who traces the evolution of this transformation of the concept of the `Holocaust' argues, following the recent Minister of Education of the new Israeli Labour government, that history must now be separated from national myth, ritual and politics. As a non-Israeli, though a Jew, I express no views about this. However, as a historian I sadly note one observation by Elon. It is that the leading contributions to the scholarly historiography of the genocide, whether by Jews or non-Jews, were either not translated into Hebrew, like Hilberg's great work, or were translated only with considerable delay, and then sometimes with editorial disclaimers. The serious historiography of the genocide has not made it any less of an unspeakable tragedy. It was merely at variance with the legitimizing myth.

Yet this very story gives us ground for hope. For here we have mythological or nationalist history being criticized from within. I note that the history of the establishment of Israel ceased to be written in Israel essentially as national propaganda or Zionist polemic about forty years after the state came into being. I have noticed the same in Irish history. About half a century after most of Ireland won its independence, Irish historians no longer wrote the history of their island in terms of the mythology of the national liberation movement. Irish history, both in the Republic and in the North, is passing through a period of great brilliance because it has succeeded in so liberating itself. This is still a matter which has political implications and risks. The history that is written today breaks with the old tradition which stretches from the Fenians to the IRA, still fighting in the name of the old myths with guns and bombs. But the fact that a new generation has grown up which can stand back from the passions of the great traumatic and formative moments of their countries' history is a sign of hope for historians.

However, we cannot wait for the generations to pass. We must resist the formation of national, ethnic and other myths, as they are being formed. It will not make us popular. Thomas Masaryk, founder of the Czechoslovak Republic, was not popular when he entered politics as the man who proved, with regret but without hesitation, that the medieval manuscripts on which much of the Czech national myth was based were fakes. But it has to be done, and I hope those of you who are historians will do it.

That is all I wanted to say to you about the duty of historians. However, before I close, I want to remind you of one other thing. You, as students of this university, are privileged people. The odds are that, as alumni of a distinguished and prestigious institute you will, if you choose, have a good status in society, have better careers and earn more than other people, though not so much as successful businessmen. What I want to remind you of is something I was told when I began to teach in a university. `The people for whom you are there', said my own teacher, `are not the brilliant students like yourself. They are the average students with boring minds who get uninteresting degrees in the lower range of the second class, and whose examination scripts all read the same. The first-class people will look after themselves, though you will enjoy teaching them. The others are the ones who need you.'

That applies not only to the university but to the world. Governments, the economy, schools, everything in society, is not for the benefit of the privileged minorities. We can look after ourselves. It is for the benefit of the ordinary run of people, who are not particularly clever or interesting (unless, of course, we fall in love with one of them), not highly educated, not successful or destined for success--in fact, are nothing very special. It is for the people who, throughout history, have entered history outside their neighbourhoods as individuals only in the records of their births, marriages and deaths. Any society worth living in is one designed for them, not for the rich, the clever, the exceptional, although any society worth living in must provide room and scope for such minorities. But the world is not made for our personal benefit, nor are we in the world for our personal benefit. A world that claims that this is its purpose is not a good, and ought not to be a lasting, world.

Chapter Two: The Sense of the Past

The following chapters try to sketch the relations of past, present and future, all of which are the historian's concern. The present chapter is based on my introductory paper to the 1970 conference on `The Sense of the Past and History' of the journal Past and Present. It was published in number 55 of that journal (May 1972) under the title `The Social Function of the Past: Some Questions'.

All human being are conscious of the past (defined as the period before the events directly recorded in any individual's memory) by virtue of living with people older than themselves. All societies likely to concern the historian have a past, for even the most innovatory colonies are populated by people who come from some society with an already long history. To be a member of any human community is to situate oneself with regard to one's (its) past, if only by rejecting it. The past is therefore a permanent dimension of the human consciousness, an inevitable component of the institutions, values and other patterns of human society. The problem for historians is to analyse the nature of this `sense of the past' in society and to trace its changes and transformations.

I

For the greater part of history we deal with societies and communities for which the past is essentially the pattern for the present. Ideally each generation copies and reproduces its predecessor so far as is possible, and considers itself as falling short of it, so far as it fails in this endeavour. Of course a total domination of the past would exclude all legitimate changes and innovations, and it is improbable that there is any human society which recognizes no such innovation. It can take place in two ways. First, what is officially defined as the `past' clearly is and must be a particular selection from the infinity of what is remembered or capable of being remembered. How great the scope of this formalized social past is in any society naturally depends on circumstances. But it will always have interstices, that is matters which form no part of the system of conscious history into which men incorporate, in one way or another, what they consider important about their society. Innovation can occur in these interstices, since it does not automatically affect the system, and therefore does not automatically come up against the barrier: `This is not how things have always been done.' It would be interesting to enquire what kinds of activities tend to be thus left relatively flexible, apart from those which appear to be negligible at one time, but may turn out not to be so at a later date. One may suggest that, other things being equal, technology in the widest sense belongs to the flexible sector, social organization and the ideology or the value system to the inflexible. However, in the absence of comparative historical studies the question must be left open. Certainly there are numerous extremely tradition-bound and ritualized societies which have in the past accepted the relatively sudden introduction of new crops, new means of locomotion (such as horses among North American Indians) and new weapons, without any sense of disturbing the pattern set by their past. On the other hand there are probably others, insufficiently investigated, which have resisted even such innovation.

The `formalized social past' is clearly more rigid, since it sets the pattern for the present. It tends to be the court of appeal for present disputes and uncertainties: law equals custom, age wisdom in illiterate societies; the documents enshrining this past, and which thereby acquire a certain spiritual authority, do the same in literate or partly literate ones. A community of American Indians may base its claim to communal lands on possession from time immemorial, or on the memory of possession in the past (very likely systematically passed on from one generation to the next), or on charters or legal decisions from the colonial era, these being preserved with enormous care: both have value as records of a past which is considered the norm for the present.

This does not exclude a certain flexibility or even de facto innovation, insofar as the new wine can be poured into what are at least in form the old containers. Dealing in second-hand cars appears to be a quite acceptable extension of dealing in horses to gypsies, who still maintain nomadism at least in theory as the only proper mode of life. Students of the process of `modernization' in twentieth-century India have investigated the ways in which powerful and rigid traditional systems can be stretched or modified, either consciously or in practice, without being officially disrupted, that is in which innovation can be reformulated as non-innovation.

In such societies conscious and radical innovation is also possible, but it may be suggested that it can be legitimized in only a few ways. It may be disguised as a return to or rediscovery of, some part of the past which has been mistakenly forgotten or abandoned, or by the invention of an anti-historical principle of superior moral force enjoining the destruction of the present/past, for example a religious revelation or prophecy. It is not clear whether in such conditions even anti-historical principles can lack all appeal to the past, that is whether the `new' principles are normally--or always?--the reassertion of `old' prophecies, or of an `old' genre of prophecy. The historians' and anthropologists' difficulty is that all recorded or observed cases of such primitive legitimization of major social innovations occur, almost be definition, when traditional societies are thrown into a context of more or less drastic social change, that is when the rigid normative framework of the past is strained to breaking-point and may therefore be unable to function `properly'. Though change and innovation which comes by imposition and importation from outside, apparently unconnected with internal social forces, need not in itself affect the system of ideas about novelty held within a community--since the problem whether it is legitimate is solved by force majeure--at such times even the extreme traditionalist society must come to some sort of terms with the surrounding and encroaching innovation. It may of course decide to reject it in toto, and withdraw from it, but this solution is rarely viable for lengthy periods.

The belief that the present should reproduce the past normally implies a fairly slow rate of historic change, for otherwise it would neither be nor seem to be realistic, except at the cost of immense social effort and the sort of isolation just referred to (as with the Amish and similar sectarians in the modern USA). So long as change--demographic, technological or otherwise--is sufficiently gradual to be absorbed, as it were, by increments, it can be absorbed into the formalized social past in the form of a mythologized and perhaps ritualized history, by a tacit modification of the system of beliefs, be 'stretching' the framework, or in other ways. Even very drastic single steps of change may be so absorbed, though perhaps at great psycho-social costs, as with the forced conversion of Indians to Catholicism after the Spanish conquest. If this were not so it would be impossible for the very substantial amount of cumulative historical change which every recorded society has undergone to have taken place, without destroying the force of this sort of normative traditionalism. Yet is still dominated much of rural society in the nineteenth and even twentieth centuries, though `what we have always done' must plainly have been very different, even among Bulgarian peasants in 1850 from what it had been in 1150. The belief that `traditional society' is static and unchanging is a myth of vulgar social science. Nevertheless, up to a certain point of change, it can remain `traditional': the mould of the past continues to shape the present, or is supposed to.

Admittedly to fix one's eyes upon the traditional peasantry, however great its numerical importance, is somewhat to bias the argument. In most respects such peasantries are often merely one part of a more comprehensive socio-economic or even political system within which somewhere changes take place uninhibited by the peasant version of tradition, or within the framework of traditions allowing for greater flexibility, for example urban ones. So long as rapid change somewhere within the system does not change the internal institutions and relations in ways for which the past provides no guide, localized changes can take place rapidly. They may even be absorbed back into a stable system of beliefs. Peasants will shake their heads over city-dwellers, notoriously and proverbially `always seeking something new', the respectable city-dwellers over the nobility at court, dizzily pursuing an ever changing and immoral fashion. The dominance of the past does not imply an image of social immobility. It is compatible with cyclical views of historic change, and certainly with regression and catastrophe (that is failure to reproduce the past). What it is incompatible with is the idea of continuous progress.

II

When social change accelerates or transforms the society beyond a certain point, the past must cease to be the pattern of the present, and can at best become the model for it. `We ought to return to the ways of our forefathers' when we no longer tread them automatically, or can be expected to. This implies a fundamental transformation of the past itself. It now becomes, and must become, a mask for innovation, for it no longer expresses the repetition of what has gone before, but actions which are by definition different from those that have gone before. Even if the literal attempt to turn the clock back is made, it does not really restore the old days, but merely certain parts of the formal system of the conscious past, which are now functionally different. The most ambitious attempt to restore the peasant society of Morelos (Mexico) under Zapata to what it had been forty years earlier--to expunge the era of Porfirio Diaz and return to the status quo ante--demonstrates this. In the first place it could not restore the past literally, since this involved some reconstruction of what could not be accurately or objectively remembered (for example the precise boundaries of common lands in dispute between different communities), not to mention the construction of `what ought to have been' and was therefore believed, or at least imagined, to have actually existed. In the second place, the hated innovation was not a mere alien body which had somehow penetrated the social organism like some bullet lodged in the flesh and which could be surgically removed, leaving the organism substantially as it was. It represented one aspect of a social change which could not be isolated from others, and consequently could be eliminated only at the cost of changing far more than the operation envisaged. In the third place, the sheer social effort of turning the clock back almost inevitably mobilized forces which had more far-reaching effects: the armed peasants of Morelos became a revolutionary power outside their state, though their horizons were local or at best regional. Restoration under the circumstances turned into social revolution. Within the borders of the state (at least so long as the power of the peasants lasted) it probably turned the hands of the clock back further than they had actually stood in the 1870s, cutting links with a wider market economy which had existed even then. Seen in the national perspective of the Mexican revolution, its effect was to produce a historically unprecedented new Mexico.

Granted that the attempt to restore a lost past cannot literally succeed, except in trivial forms (such as the restoration of ruined buildings), attempts to do so will still be made and will normally be selective. (The case of some backward peasant region attempting to restore all of what still existed in living memory is analytically comparatively uninteresting.) What aspects of the past will be singled out for the effort of restoration? Historians are likely to note the frequency of certain calls for restoration--in favour of the old law, the old morality, the oldtime religion and so on, and might well be tempted to generalize from this. But before they do so they ought perhaps to systematize their own observations and seek guidance from social anthropologists and others whose theories might be relevant. Moreover, before taking too super-structural a view of the matter, they might recall that attempts to restore an actual dying or dead economic structure are by no means unknown. The hope of a return to an economy of petty peasant proprietorship, though it might be little more than a big-city pastoral in nineteenth-century Britain (it was not, at least initially, shared by the actual landless rural labourers), was nevertheless an important element in radical propaganda, and occasionally more actively pursued.

A distinction ought nevertheless to be made, even in the absence of a useful general model of such selective restoration, between symbolic and effective attempts of this kind. The call for a restoration of old morality or religion is intended to be effective. If successful, then ideally no girl will have, say, premarital sexual intercourse or everyone will attend church. On the other hand the desire to restore, literally, the bombed fabric of Warsaw after the Second World War, or conversely to pull down particular records of innovation such as the Stalin monument in Prague, is symbolic, even allowing for a certain aesthetic element in it. One might suspect that this is so because what people actually wish to restore is too vast and vague for specific acts of restoration, for example past `greatness' or past `freedom'. The relationship between effective and symbolic restoration may indeed be complex, and both elements may always be present. The literal restoration of the fabric of parliament on which Winston Churchill insisted could be justified on effective grounds, that is the preservation of an architectural scheme which favoured a particular pattern of parliamentary politics, debate and ambience essential to the functioning of the British political system. Nevertheless, like the earlier choice of the neo-gothic style for the buildings, it also suggests a strong symbolic element, perhaps even a form of magic which, by restoring a small but emotionally charged part of a lost past, somehow restores the whole.

Sooner or later, however, it is likely that a point will be reached when the past can no longer be literally reproduced or even restored. At this point the past becomes so remote from actual or even remembered reality that it may finally turn into little more than a language for defining certain not necessarily conservative aspirations of today in historical terms. The Free Anglo-Saxons before the Norman Yoke, or Merrie England before the Reformation, are familiar examples. So, to take a contemporary illustration, is the `Charlemagne' metaphor, which has been used, ever since Napoleon I, to propagate various forms of partial European unity, whether by conquest from the French or German side or by federation, and which patently is not intended to re-create anything even remotely like the Europe of the eighth and ninth centuries. Here (whether its proponents actually believe in it or not), the demand to restore or re-create a past so remote as to have little relevance to the present may equal total innovation, and the past thus invoked may become an artefact or, in less flattering terms, a fabrication. The name `Ghana' transfers the history of one part of Africa to another, geographically remote and historically quite different. The Zionist claim to return to the pre-diaspora past in the land of Israel was in practice the negation of the actual history of the Jewish people for more than 2,000 years.

Fabricated history is familiar enough, yet we ought to distinguish between those uses of it which are rhetorical or analytic and those which imply some genuine concrete `restoration'. The English radicals of the seventeenth to nineteenth centuries hardly intended to return to pre-conquest society; the `Norman Yoke' for them was primarily an explanatory device, the `Free Anglo-Saxons' at best an analogy or the search for a genealogy, such as will be considered below. On the other hand modern nationalist movements, which can almost be defined, in Renan's words, as movements which forget history or rather get it wrong, because their objectives are historically unprecedented, nevertheless insist on defining them to a greater or lesser extent in historical terms and actually attempt to realize parts of this fictitious history. This applies most obviously to the definition of the national territory, or rather to territorial claims, but various forms of deliberate archaism are familiar enough, from the Welsh neo-Druids to the adoption of Hebrew as a spoken secular language and the Ordensburgen of National Socialist Germany. All these, it must be repeated, are not in any sense `restorations' or even `revivals'. They are innovations using or purporting to use elements of a historic past, real or imaginary.

What kinds of innovation proceed in this manner, and under what conditions? Nationalist movements are the most obvious, since history is the most easily worked raw material for the process of manufacturing the historically novel `nations' in which they are engaged. What other movements operate in this way? Can we say that certain types of aspiration are more likely than others to adopt this mode of definition, for example those concerning the social cohesion of human groups, those embodying the `sense of the community'? The question must be left open.

III

The problem of systematically rejecting the past arises only when innovation is recognized both as inescapable and as socially desirable: when it represents `progress'. This raises two distinct questions, how innovation as such is recognized and legitimized, and how the situation arising from it is to be specified (that is how a model of society is to be formulated when the past can no longer provide it). The former is more easily answered.

We know very little about the process which has turned the words `new' and `revolutionary' (as used in the language of advertising) into synonyms for `better' and `more desirable', and research is badly needed here. However, it would seem that novelty or even constant innovation is more readily accepted as far as it concerns the human control over non-human nature, for example science and technology, since so much of it is obviously advantageous even to the most tradition-bound. Has there ever been a serious example of Luddism directed against bicycles or transistor radios? On the other hand, while certain socio-political innovations may appear attractive to some groups of human beings, at least prospectively, the social and human implications of innovation (including technical innovation) tend to meet with greater resistance, for equally obvious reasons. Rapid and constant change in material technology may be hailed by the very people who are profoundly upset by the experience of rapid change in human (for example sexual and family) relations, and who might actually find it hard to conceive of constant change in such relations. Where even palpably `useful' material innovation is rejected, it is generally, perhaps always, because of the fear of the social innovation, that is disruption, it entails.

Innovation which is so obviously useful and socially neutral that it is accepted almost automatically, at all events by people to whom technological change is familiar, raises virtually no problem of legitimation. One would guess (but has the subject actually been investigated?) that even so essentially traditionalist an activity as popular institutional religion has found little difficulty in accepting it. We know of violent resistance to any change in the ancient holy texts, but there appears to have been no equivalent resistance to, say, the cheapening of holy images and icons by means of modern technological processes, such as prints and oleographs. On the other hand certain innovations require legitimation, and in periods when the past ceases to provide any precedent for them, this raises very grave difficulties. A single dose of innovation, however great, is not so troublesome. It can be presented as the victory of some permanent positive principle over its opposite, or as a process of `correction' or `rectification', reason prevailing over unreason, knowledge over ignorance, `nature' over the `unnatural', good over evil. But the basic experience of the past two centuries has been constant and continued change, which cannot be so dealt with except sometimes, at the cost of considerable casuistry, as the constantly necessary application of permanent principles to circumstances ever changing in ways which remain rather mysterious, or by exaggerating the strength of the surviving forces of evil.

Paradoxically, the past remains the most useful analytical tool for coping with constant change, but in a novel form. It turns into the discovery of history as a process of directional change, of development or evolution. Change thus becomes its own legitimation, but it is thereby anchored to a transformed `sense of the past'. Bagehot's Physics and Politics (1872) is a good nineteenth-century example of this; current concepts of `modernization' illustrate more simple-minded versions of the same approach. In brief, what legitimates the present and explains it is not now the past as a set of reference points (for example Magna Carta), or even as duration (for example the age of parliamentary institutions) but the past as a process of becoming the present. Faced with the overriding reality of change, even conservative thought becomes historicist. Perhaps, because hindsight is the most persuasive form of the historian's wisdom, it suits them better than most.

But what of these who also require foresight, to specify a future which is unlike anything in the past? To do so without some sort of example is unusually difficult, and we find those most dedicated to innovation often tempted to look for one, however implausible, including in the past itself, or in what amounts to the same thing, `primitive society' considered as a form of man's past coexisting with his present. Nineteenth- and twentieth-century socialists doubtless used `primitive communism' merely as an analytical prop, but the fact that they used it at all indicates the advantage of being able to have a concrete precedent even for the unprecedented, or at least an example of ways of solving new problems, however inapplicable the actual solutions of the analogous problems in the past. There is, of course, no theoretical necessity for specifying the future, but in practice the demand to predict or to set up a model for it is too strong to be shrugged off.

Some sort of historicism, that is the more or less sophisticated and complex extrapolation of past tendencies into the future, has been the most convenient and popular method of prediction. At all events the shape of the future is discerned by searching the process of past development for clues, so that paradoxically, the more we expect innovation, the more history becomes essential to discover what it will be like. This procedure may range from the very naive--the view of the future as a bigger and better present, or a bigger and worse present so characteristic of technological extrapolations or pessimistic social anti-utopias--to the intellectually very complex and high-powered; but essentially history remains the basis of both. However, at this point a contradiction arises, whose nature is suggested by Karl Marx's simultaneous conviction of the inevitable supersession of capitalism by socialism, and extreme reluctance to make more than a few very general statements about what socialist and communist society would actually be like. This not merely common sense: the capacity to discern general tendencies does not imply the capacity to forecast their precise outcome in complex and in many respects unknown circumstances of the future. It also indicates a conflict between an essentially historicist mode of analysing how the future will come about, which assumes a continuing process of historical change, and what has so far been the universal requirement of programmatic models of society, namely a certain stability. Utopia is by nature a stable or self-reproducing state and its implicit a-historicism can be avoided only by those who refuse to describe it. Even less utopian models of the `good society' or the desirable political system, however designed to meet changing circumstances, tend also to be designed to do so by means of a relatively stable and predictable framework of institutions and values, which will not be disrupted by such changes. There is no theoretical difficulty in defining social systems in terms of continuous change, but in practice there seems little demand for this, perhaps because an excessive degree of instability and unpredictability in social relations is particularly disorienting. In Comtean terms `order' goes with `progress', but the analysis of the one tells us little about the desirable design of the other. History ceases to be of use at the very moment when we need it most.

We may therefore still be forced back upon the past, in a way analogous to the traditional use of it as a repository of precedents, though now making our selection in the light of analytical models or programmes which have nothing to do with it. This is particularly likely in the design of the `good society', since most of what we know about the successful functioning of societies is what has been empirically learned in the course of some thousands of years of living together in human groups in a variety of ways, supplemented perhaps by the recently fashionable study of the social behaviour of animals. The value of historical enquiry into `what actually happened' for the solution of this or that specific problem of present and future is undoubted, and has given a new lease of life to some rather old-fashioned historical activities, provided they are teamed with rather new-fangled problems. Thus what happened to the poor displaced by the massive railway building or the nineteenth century in the hearts of great cities can and ought to throw light on the possible consequences of massive urban motorway building in the late twentieth century, and the various experiences of `student power' in medieval universities are not without bearing on projects to change the constitutional structure of modern universities. Yet the nature of this often arbitrary process of dipping into the past for assistance in forecasting the future requires more analysis than it has so far received. By itself it does not replace the construction of adequate social models, with or without historical enquiry. It merely reflects and perhaps in some instances palliates their present inadequacy.

IV

These casual remarks are far from exhausting the social uses of the past. However, though no attempt to discuss all other aspects can be made here, two special problems may be mentioned briefly: those of the past as genealogy and as chronology.

The sense of the past as a collective continuity of experience remains surprisingly important, even to those most dedicated to innovation and the belief that novelty equals improvement: as witness the universal inclusion of `history' in the syllabus of every modern educational system, or the search for ancestors (Spartacus, More, Winstanley) by modern revolutionaries whose theory, if they are Marxists, assumes their irrelevance. What precisely did or do modern Marxists gain from the knowledge that there were slave rebellions in ancient Rome which, even supposing their aims to have been communist, were by their own analysis doomed to failure or to produce results which could have little bearing on the aspirations of modern communists? Clearly the sense of belonging to an age-old tradition of rebellion provides emotional satisfaction, but how and why? Is it analogous to the sense of continuity which infuses history syllabuses and makes it apparently desirable for schoolchildren to learn of the existence of Boadicea or Vercingetorix, King Alfred or Joan of Arc as part of that body of information which (for reasons which are assumed to be valid but are rarely investigated) they are `supposed to know about' as Englishmen or Frenchmen? The pull of the past as continuity and tradition, as `our ancestors', is strong. Even the pattern of tourism bears witness to it. Our instinctive sympathy with the sentiment should not, however, lead us to overlook the difficulty of discovering why this should be so.

This difficulty is naturally much smaller in the case of a more familiar form of genealogy, that which seeks to buttress an uncertain self-esteem. Bourgeois parvenus seek pedigrees, new nations or movements annex examples of past greatness and achievement to their history in proportion as they feel their actual past to have been lacking in these things--whether this feeling is justified or not. The most interesting question concerning such genealogical exercises is whether or when they become dispensable. The experience of modern capitalist society suggests that they may be both permanent and transitional. On the one hand late-twentieth-century nouveaux riches still aspire to the characteristics of the life of an aristocracy which, in spite of its political and economic irrelevance, continues to represent the highest social status (the country chateau, the Rhineland managing director hunting elk and boar in the implausible surroundings of socialist republics, and so on). On the other, the neo-medieval, neo-Renaissance and Louis XV buildings and decor of nineteenth-century bourgeois society gave way at a certain stage to a deliberately `modern' style, which not only refused to appeal to the past, but developed a doubtful aesthetic analogy between artistic and technical innovation. Unfortunately the only society in history which so far gives us adequate material for studying the comparative pull of ancestors and novelty is Western capitalist society in the nineteenth and twentieth centuries. It would be unwise to generalize on the strength of a sample of one.

Finally, the problem of chronology, which takes us to the opposite extreme of possible generalization, since it is hard to think of any known society which does not for certain purposes find it convenient to record the duration of time and the succession of events. There is, of course, as Moses Finley has pointed out, a fundamental difference between a chronological and a non-chronological past: between Homer's Odysseus and Samuel Butler's, who is naturally and quite unHomerically conceived of as a middle-aged man returning to an ageing wife after twenty years' absence. Chronology is, of course, essential to the modern, historical sense of the past, since history is directional change. Anachronism is an immediate alarm-bell for the historian, and its emotional shock-value in a thoroughly chronological society is such as to lend itself to easy exploitation in the arts: Macbeth in modern dress today benefits from this in a way in which a Jacobean Macbeth obviously did not.

At first sight it is less essential to the traditional sense of the past (pattern or model for the present, storehouse and repository of experience, wisdom and moral precept). In such a past events are not necessarily believed to exist simultaneously, like the Romans and Moors who fight one another in Spanish Easter processions, or even out of time: their chronological relation to each other is merely irrelevant. Whether Horatius of the Bridge contributed his example to later Romans before or after Mucius Scaevola is of interest only to pedants. Similarly (to take a modern example) the value of the Maccabees, the defenders of Masada and Bar Kokhba, for modern Israelis has nothing to do with their chronological distance from them and from one another. The moment when real time is introduced into such a past (for example, when Homer and the Bible are analysed by the methods of modern historical scholarship) it turns into something else. This is a socially disturbing process and a symptom of social transformation.

Yet for certain purposes historical chronology, for example in the form of genealogies and chronicles, is evidently important in many (perhaps in all?) literate, or even illiterate, societies, though the ability of literate ones to maintain permanent written records makes it possible for them to devise uses for them which would seem to be impracticable in those relying purely oral transmission. (However, though the limits of oral historical memory have been investigated from the point of view of the requirements of the modern scholar, historians have given less attention to the question how far they are inadequate to the social requirements of their own societies.)

In the broadest sense all societies have myths of creation and development, which imply temporal succession: first things were thus, then they changed thus. Conversely, a providential conception of the universe also implies some kind of succession of events, for teleology (even if its objects have already been achieved) is a kind of history. Moreover, it lends itself excellently to chronology, where such exists: as witness the various millennial speculations or the debates about the year 1000 AD, which pivot on the existence of a system of dating. In a more precise sense, the process of commenting on ancient texts of permanent validity or of discovering the specific applications of eternal truth implies an element of chronology (for example, the search for precedent). It is hardly worth mentioning that even more precise calculations of chronology may be required for a variety of economic, legal, bureaucratic, political and ritual purposes, at least in literate societies which can keep a record of them, including, of course, the invention of favourable and ancient precedents for political purposes.

In some instances the difference between such chronology and that of modern history is clear enough. The lawyers' and bureaucrats' search for precedent is entirely present-oriented. Its object is to discover the legal rights of today, the solution of modern administrative problems, whereas for the historian, however interested in their relation to the present, it is the difference of circumstances which is significant. On the other hand this does not seem to exhaust the character of traditional chronology. History, the unity of past, present and future, may be something that is universally apprehended, however deficient the human capacity to recall and record it, and some sort of chronology, however unrecognizable or imprecise by our criteria, may be a necessary measure of it. But even if this should be so, where are the demarcation lines drawn between the coexisting non-chronological and chronological past between the coexisting historical and non-historical chronologies? The answers are by no means clear. Perhaps they might throw light not only on the sense of the past of earlier societies, but on our own, in which the hegemony of one form (historical change) does not exclude the persistence, in different milieux and circumstances, of other forms of the sense of the past.

It is easier to formulate questions than answers, and this paper has taken the easier way rather than the more difficult. And yet, perhaps to ask questions, especially about the experiences we tend to take for granted, is not a valueless occupation. We swim in the past as fish do in water, and cannot escape from it. But our modes of living and moving in this medium require analysis and discussion. My object has been to stimulate both.