Writing Russia’s Future: Paradigms, Drivers, and Scenarios

The development of prediction and forecasting in the social sciences over the past century and more is closely linked with developments in Russia. The Soviet collapse undermined confidence in predictive capabilites, and scenario planning emerged as the dominant future-oriented methodology in area studies, including the study of Russia. Scenarists anticipate multiple futures rather than predicting one. The approach is too rarely critiqued. Building on an account of Russia-related forecasting in the twentieth century, analysis of two decades of scenarios reveals uniform accounts which downplay the insights of experts and of social science theory alike. It is commonplace amongst many observers today to note that Russia’s richness in natural resources has a key role to play in its future development, but that excessive interference from western powers will prove counterproductive. After all that Russia has been through in recent times, the population as a whole is suspicious of such interference. It is suspicious too of the term ‘democracy’, associating the word with utter disorder, and will therefore prefer instead stability as essential for the exploitation of the country’s underdeveloped resources. So far, so 21 st century. Except that these forecasts come from two analysts predicting Russia’s future almost a hundred years ago, in 1919 (Landfield, 1919, 33; Story, 1919, 85). That such century-old forecasts can be readily transposed to the present day tells us as much about forecasting as it does about Russia, raising questions of the specific versus the general, and of the empirical versus the ideological and cultural as key drivers in shaping the future. Such questions in turn highlight epistemological issues at the heart of social science, touching on its identity as either a positivist and reductive endeavour akin to the natural sciences, or as an enterprise most usefully approached from an interpretivist position, recognising human action as essentially contextual and intuitive. The relationship between predictive social science and scholarship on the future of Russia is a close one and can be divided into three phases. First, from the end of the 19 th to the middle of the 20 th century, predictions of the future tended to be methodologically unsophisticated and narrative-based. Second, the Cold War saw vast resources and scholarly effort put into developing new predictive approaches in the social sciences, with the aim of forecasting Soviet behaviour. Such approaches were themselves subject to detailed analyses and critiques by scholars such as Daniel Bell and Zbigniew Brzezinski in the 1950s and 1960s (Bell, 1958, 1965; Brzezinski, 1969). Third, from the end of the Cold War and the collapse of the Soviet Union onwards numerous academic and policy-oriented reviews of social science’s predictive possibilities emerged from a broad acknowledgement that the analytical and scholarly community had not foreseen the events of 1989-1991 (Cox, 1998; Lipset & Bence, 1994; Seliktar, 2004). From that time on, scenario analysis has become the dominant approach of futureoriented research in relation to Russia, both in the West and in Russia itself, as well as becoming widely used in relation to other countries, regions, and issues. However, no detailed critique of these scenario-based accounts of Russia’s future has yet appeared. This article seeks to fill that particular lacuna. It sets out the nature of future-oriented scholarship in relation to Russia in the context of broader debates about the predictive possibilities of social science, and provides a critique of scenario-based accounts of Russia’s future published in the past two decades. I argue that futureoriented scholarship across the decades has sought to connect drivers with outcomes, using a range of approaches. The focus on ‘getting it right’ which dominated such analysis during the Cold War, and led to a crisis of predictive confidence at its end, is too simplistic for forecasting the future of complex multi-vectored systems, such as a country or a region, where the connections between drivers and outcomes are almost infinite. Few scholars today claim that social science can provide accurate ‘point forecasts’ amidst such systemic complexity. The scenario approach insteads starts from the premise that getting it right is irrelevant, and that what matters is an awareness of and preparedness for multiple possible outcomes arising from the actions of key drivers. Such an approach has some strengths, but has also resulted in increasingly uniform future-oriented analyses, consisting of a standard range of possibilities so broad as to be of little use to the policy world. The best of the scenario accounts are replete with insights and draw on robust and original research. As with much future-oriented writing, a large element of their value comes from analysis of the present rather than predictions of the future. Scenarios commendably insist on developing clear paths from the present to each possible future; if someone is anticipating state collapse, then identifying the mechanism by which it might happen represents the minimum analytical expectation. However, for all its strengths, the scenario approach as a whole suffers from rigidity of method, an inability to deal effectively with complexity, and a disavowal of the predictive possibilities of its own analysis and of the social sciences more broadly. The standardised framework insisted on by scenario development undermines both the expertise of the scholar and the genuine possibilities which exist for predictive social science. The article concludes with a normative case for the return of single-future forecasts bringing discipline-based theory together with context-aware expertise. Futurology, social science and Russia before the Soviet collapse Starting from the end of the 19 th century, predictive writing in relation to Russian socio-political affairs can be seen to have developed in line with the nascent discipline of political science in particular, and of the social sciences more generally. According to Mark Bevir’s survey of political studies from that era to the present day, the dominant form of political analysis at the turn of the twentieth century was straightforward narrative, informed by a developmental, evolutionary historicism (Bevir, 2006, 584-585). The maxim ‘history is past politics, politics is present history’ adroitly sums up this approach. 1 The view that serious scholars were well advised to stay away from the present, let alone the future, remained common (Cooper & Layard, 2002, 2). When analysts in the field of Russian studies, often those with a foot in both the scholarly and the policy worlds, did nonetheless essay future-oriented writing, such work likewise took a narrative developmental historicist approach (Martin, 1906; Obsta principiis pseud., 1878). This was theoretically and methodologically unspecific work, which by today’s standards would be deemed lacking in scientific rigour, yet it could still prove immensely astute. Writing in 1906, the German government official, Rudolf Emil Martin, predicted ‘that the Russian Empire is slowly but surely approaching a Reign of Terror’, and that the years 1904 and 1905 represented ‘only a slight foretaste of the things which the Russian Empire will have to face in the future’. Although Martin claimed that his predictions were ‘scientifically correct’, they are essentially a narrative account of a straightforward continuity-based forecast, which discerned accurately the social, economic, and ethnic reasons behind the 1905 revolution and saw as likely the continuation of such unresolved incompatibilities (Martin, 1906, viii, 304-05). The Bolshevik revolution of 1917 brought to power in Russia a future-oriented regime par excellence, with Soviet Communism’s teleological eschatology setting the future workers’ utopia as the goal which motivated all present policy. Unsurprisingly then, future-oriented analysis took on an ideological hue. Scholars who were broadly sympathetic to the Soviet regime contributed their forecasts which bought in to the Marxist view of history (for example, Cole, 1941; Schapiro, 1955), dissidents – as dissidents do – begged to differ (Amalrik, 1970; Voinovich, 1987), and émigré writers prepared for the coming collapse of Communism (Il’in, 2008a, 2008b [1948-1954]). Alongside these ideologically normative approaches to the writing of the future, the Cold War also saw vast amounts of money and intellectual effort in the West poured into attempts to discern in a 1 Variously attributed to the Swiss scholar Johann Bluntschli, and the historians Sir John Seeley and Edward Freeman respectively (Bevir, 2006, 584; Burke, 2001, 3; Hartogensis, 23 May 1927) more scientific way the course of future events in the Soviet Union. As discussed in more detail later, attempts to develop scientifically rigorous predictions received a great boost during these years, with initiatives such as Project RAND, the Club of Rome, and the Commission on the Year 2000 (Bell, 1965). Despite such investment in future-oriented research, however, it is now widely accepted that the largely unforeseen Soviet collapse in 1989-1991 stands as the nadir of Western futurology with regard to Russia. As Mark Perry wrote, in his account of the US Central Intelligence Agency in those years, it was extraordinary that although it spent half its budget on Soviet analysis, the CIA did not appear to realise that the Soviet Union was on the verge of radical change (Perry, 1992, 308). Perry’s criticism oversimplifies. According to one 1972 survey of Western methodologies in the field of Soviet studies, ‘predicting the downfall of the Soviet regime has been a favorite academic pastime in the West for well over half a century. Probably no other regime has ever survived so many prophecies of inevitable catastrophe’ (Dziewanowski, 1972, 367). Of course astute analysts foresaw the inevitable generational change in the Soviet leadership in the 1980s, even noting the quiet rise of Mikhail Sergeevich Gorbachev, and the prospects which this offered for reform (Brown & Kaser, 1982). A number of scholars and analysts had likewise anticipated the long-term decline in and eventual collapse of the Soviet system (Brzezinski, 1969; Collins, 1986; Levin, 1977a, 1977b, 1977c; Todd, 1976). It remains the case though that, definitively by the early 1980s, such predictions lay outside the analytical mainstream; the radical nature of Gorbachev’s reforms and the rapid and peaceful collapse of the Soviet system took the expert and scholarly community, taken as a whole, largely by surprise. Such a summary certainly became broadly accepted (Seliktar, 2004), and led to a distinct turn in the study of Russia’s future, ushering in the era of the scenario approach, which continues today to dominate future studies of Russia both in the West and in Russia itself. Scenario-based analyses are the focus of the latter half of this paper. Before considering the rise of the scenario, however, let us look in more detail at the survey of future studies set out above, starting with the notion that the Soviet collapse was by and large not predicted. Predicting and not predicting the Soviet collapse Before turning to contemporary, scenario-based accounts of Russia’s future, my purpose in focusing on the apparent failure of the Russian studies community to predict the collapse of the Soviet Union is twofold, and not primarily concerned with assessing the veracity of that claim, as others have already expertly done (Cox, 1998; Lipset & Bence, 1994; Seliktar, 2004). First, I use the Soviet collapse to exemplify the methodological and epistemological difficulties facing the social scientist when it comes to prediction at the complex and large-scale level of a particular country or political system. Second, I argue that the notion of a collective predictive failure on the part of the Russian studies community in relation to the Soviet collapse became a totemic paradigm, which has shaped future-oriented research in relation to Russia and beyond ever since. The justification for the almost ubiquitous use of scenario-based futurology in this endeavour since the fall of communism, to which the second half of this paper is devoted, is explicitly based on the apparent failures of the expert community to predict the end of the Cold War and the break up of the Soviet Union. In post-Soviet accounts of the collective predictive failure of scholars and analysts with regard to the Soviet collapse, a select few names stand out as having correctly foreseen the end of the Soviet system. In this panoply of accurate forecasters can be found scholar (Collins, 1986), dissident (Amalrik, 1980), journalistic commentator (Levin, 1977a, 1977b, 1977c), and even thriller writer (James, 1982). Of these the most prescient is the British newspaper columnist Bernard Levin’s uncannily accurate 1977 account, in a series of three articles, of how the Soviet Union would come to an end. Levin explained that he had kept his conclusions to himself for a long time because they seemed ‘so heartening, and so strange’ (Levin, 1977a). His final predictive assertion bears setting out at length in order to establish its argument and precision. He forecast ‘that a new and utterly unprecedented Russian Revolution is coming, in which no shot will be fired’ (Levin, 1977b), and that the man who would strike the match to set this revolution burning would rise up through the party ranks – as indeed Mikhail Gorbachev was already doing at that time. ‘I do not know his name or what he looks like, but I know he is there. For do you seriously suppose – now we extend the same questions into another area – that Mr Dubcek ... and the other Czech liberationists who led the doomed revolt [of 1968] came up one night like mushrooms, or arrived in a rocket from Outer Space? They came up through the system ... . And if you tell me that no such figures exist in the Soviet Union, even more completely unknown outside (or for that matter inside) than the Czech heroes were, I shall tell you in return that it simply cannot be so. The odds against such an extraordinary aberration of the human spirit are so preposterously high that the chance can be ignored with impunity. They are there, all right, at this very moment, obeying orders, doing their duty, taking the official line against dissidents not only in public but in private. They do not conspire, they are not in touch with Western intelligence agencies, they commit no sabotage. They are in every respect model Soviet functionaries. Or rather, in every respect but one, they have admitted the truth about their country to themselves, and have vowed, also to themselves, to do something about it. That is how it will be done. There will be no gunfire in the streets, no barricades, no general strikes, no hanging of oppressors from lamp-posts, no sacking and burning of government offices, no seizure of radio-stations or mass defections among the military. But one day soon, some new faces will appear in the Politburo – I am sure they have already appeared in municipal and even regional administrative authorities – and gradually, very gradually, other similarly new faces will join them. Until one day they will look at each other and realize that there is no longer any need for concealment of the truth in their hearts. And the match will be lit. There is nothing romantic or fantastic about this prognosis; it is the most sober extrapolation from known facts, and tested evidence. That, or something like it, will happen. When it will happen it is neither possible nor useful to guess; but I am sure that it will be in the lifetime of people much older than I ... let us suppose, for neatness’ sake, on July 14, 1989’ (Levin, 1977b). Even Levin’s reluctantly predictive date – neatness coming from it being the bicentenary of the French revolution was about right, with the Washington Post on 14 July 1989 reporting that the priority of President Bush’s trip to Europe for the bicentennial celebrations was ‘economic aid for the nations of Eastern Europe now trying to develop a form of pluralist democracy and market-oriented economy under the benign gaze of Soviet leader Mikhail Gorbachev’. Fascinating though it may be to identify accurate predictions of Russia’s future, the apposite scholarly response is two-fold; scepticism, and methodological enquiry. So far as scepticism is concerned, assessments of broadly correct forecasts require an holistic approach in terms of the field of study and of the forecasts themselves. Writing in the late 1950s, Daniel Bell asserted that more had been written about the Russian revolution and the subsequent forty years of Soviet rule than about any comparable episode in history (Bell, 1958, 327). A decade later Merle Fainsod noted the ‘innumerable predictions of [the Soviet Union’s] imminent collapse’ over the previous half century (Fainsod, 1967). That being the case, then it is statistically unsurprising that a continuum of ‘incorrect’ to ‘correct’ predictions exists. As noted above, within the detailed literature explaining the apparent failure of the scholarly community to foresee the Soviet collapse (Cox, 1998; Lipset & Bence, 1994; Seliktar, 2004; The National Interest, 1993), there existed the category of ‘some who were right’ and had indeed forecast the eventual toppling of the Soviet regime (Lipset & Bence, 1994, 177). However, just because someone predicted that the regime would collapse, this does not necessarily mean that they were right in a broader sense. Take, for example, Andrei Amalrik’s 1970 essay ‘Will the Soviet Union Survive until 1984?’, cited as an example of one who forecast the end of the Soviet Union. Without undermining Amalrik’s astute discussion of declining standards of living and the growth of nationalism, both of which played their part in the failure of the Soviet state, more than half of Amalrik’s essay is taken up with detailed discussion of a projected future war between the Soviet Union and China which would be the main driver behind collapse (Amalrik, 1970). He was wrong about that war, but right about the collapse. In the absence of 20:20 foresight, all ‘correct’ forecasts have elements which are incorrect. Even Levin’s remarkably accurate and certain prediction was not entirely correct about the absence of bloodshed, defections from the armed forces, strikes, and barricades on the street, albeit that the scale on which these occurred was notably small. If we can agree that not all of those who ‘got it right’ in terms of collapse, got everything right, then can we also argue that not all of those who got it wrong, really got it so wrong? I think so. The difficulty with establishing prediction of the Soviet collapse as our marker for success or failure of Cold War forecasts is that it introduces a standard very near to a ‘point forecast’ (the forecasting of a specific event at a specific time, within a scale of, say, a year or two) into a situation where the subject of the forecast is a complex system. Forecasts concerning the future of an entire country, particularly a country of the size and opacity of the Soviet Union or even contemporary Russia, contain innumerable variables. The narrower and more focused the field of forecast, the fewer variables complicate matters. So, for example, the reliability of forecasts for gas production in a decade from now exceeds that of forecasts of the ‘whither Russia?’ type over the same period. In the broader debate surrounding the nature of social science, the ability to predict has repeatedly been cited by both interpretivists and positivists as an appropriate test of scientific theory. To the interpretivist, the general inability of theories to offer anything near a reliable ‘point forecast’ in the complex systems which make up the socio-political sphere demonstrates that human activity cannot be scientifically conceptualised in the same way as the natural world can be (Flyvbjerg, 2001, 39). The positivist response contends that, even in the natural sciences, theory-based prediction does not always require such precision, that probability estimates can suffice to demonstrate theoretical efficacy, and that social scientists have been able to forecast on the basis of theory in this way for many years (Laitin, 2005, 120). From this perspective, establishing one particular event – the collapse of the Soviet Union – as the qualifying standard for predictions about Russia is problematic. Such a qualifying standard promotes two types of error in complex systems, namely category and temporal errors. First, the category error. In the 1960s Robert Conquest predicted, with some accuracy, that ‘the USSR is a country where the political system is radically and dangerously inappropriate to its social and economic dynamics. This is a formula for change – change which may be sudden and catastrophic’ (Conquest, 1966, 37). In assessing the failure of ‘most Sovietologists’ to foresee the Soviet collapse, Lipset and Bence argue that many dismissed Conquest’s prediction because they thought that the Soviet economy was improving (Lipset & Bence, 1994, 177). The argument of these critics represents the sort of category error to be wary of in analysis of forecasts. Conquest made a political forecast, not an economic one. The collapse of the Soviet Union consisted of, by definition, the breaking away of the Union Republics from the control of the Soviet centre. This was primarily a political process, driven by nationalism, the decline of a belief in the imminence of global communism on the part of the Soviet leadership, and a subsequent lack of political legitimacy for the Communist regime. And of course economic decline too played its part in this loss of legitimacy. However, although closely connected, the economy and the polity are not the same thing. Economic performance represents a key variable in politics, but to forecast either economic growth or decline is not, in itself, to forecast political developments. Mark Harrison’s detailed account of the Soviet economy at the time of the collapse illustrates this point. Harrison argues that ‘the Soviet economy was stable until it collapsed ... The eventual collapse could not have been forecast on the basis of the command economy’s intrinsic properties alone’. In other words, criticising forecasts in one category for not predicting events in another category is problematic; ‘the first shock to which the Soviet economy was exposed was not economic but political, the dismantling of the command system’ (Harrison, 2001, 11). The second type of error is temporal. Using the same example, the temporal error would consist of mistaking the effects of the Soviet collapse – the dismantling of the command system – for its causes, which were primarily political. Forecasters in complex socio-political systems need to remain particularly alert to these potential errors. Although, as we will see when considering scenario development, it may be relatively straightforward to identify potential key drivers which will shape the future, these drivers interact in complex causal and temporal relationships with a huge number of potential outcomes. A reliable forecast of economic decline can too easily morph into a far less reliable forecast of its political effects, which can in turn be read backwards to imply that what occurred represents the inevitable outcome of the originally identified causal factor. I have argued that appropriate scepticism in the face of claims of correct and incorrect predictions enables a more holistic assessment of the predictions themselves, facilitates questioning of the criteria by which a forecast can be judged correct, and alerts us to the potential for category and temporal errors. Studying the predictive genre in relation to Russian and Soviet studies before the collapse of the Soviet Union also provides insights into the development of ‘future studies’ during these years. As noted above, predictive essays in relation to Russia in the early decades of the twentieth century tended to be written as narrative analysis, with little in the way of explicit theory and methodology. Re-reading them today, however, it is possible to identify implicit assumptions which would be familiar to later scholars. N.G.O. Pereira’s assessment of how the nineteenth century Russian intelligentsia conceptualised the future argues that the influence of Russian messianism was all but ubiquitous (Pereira, 1979). This focus on cultural and ideological factors represents a common strand in futurological studies, positing that the ideological views of the élite provide a key to future action. In the nineteenth century, such Russian élite conceptions consisted of the view that the future belonged to Russia, to fulfill its messianic role in the world. An 1877 translation of and commentary on the political will of Peter the Great in the Foreign and Commonwealth Office collection illustrates well how such an approach informed British understandings of Russia’s likely ambitions in the world, summed up by Peter’s prediction that Russia which he ‘found a river and left a flood, will become an ocean, and will spread over the continents to fertilize them with its mud’ (Obsta principiis pseud., 1878, 3-4). Whilst some authors of predictive essays in the late nineteenth and early twentieth century perceived the future from above, from the point of view of the elite, others, particularly after the revolutions of 1905 and 1917, set out to forecast from below, from an understanding of social factors. Landfield emphasised the urban-rural split which he argued would fatally undermine the Bolshevik project (Landfield, 1919, 309-10). A further approach focused on the nature of the Russian character, that to understand a country and how it might behave requires an understanding of its people’s mores, motivations, and way of life. Writing at the same time as Landfield, Story based his forecasts around the nature of ‘the true Russian spirit’ and Russia’s ‘great underlying unity of culture’ (Story, 1919, 8586). The fundamental approaches to forecasting seen, though not explicitly set out, in these examples represent contextual and intuitive interpretations. Neither grand theory nor the minutiae of methodology play a role. Although forecasting has moved on a good deal in theoretical and methodological terms from a century ago, many predictions since then have been and continue to be made on the basis of analysts’ identification of the key drivers and understanding of how they operate; some remarkably prescient, such as Levin’s forecast of a peaceful end to the Soviet system brought about by its own leadership, others foreseeing apocalyptic nuclear civil war as the Soviet Union fell apart (Levin, 1977a, 1977b, 1977c; Lourie, 1991, 83). During the Cold War years, however, a far more methodical, and methodologically conscious, approach to forecasting developed too. Daniel Bell’s 1958 article, subtitled ‘The Prediction of Soviet Behaviour in the Social Sciences’, represents a key, but largely forgotten, text in this process of formalising approaches. Bell surveyed the ‘new sophistication’ of Soviet studies since the end of the Second World War, noting the entry of new disciplines such as anthropology, psychiatry, and sociology, and providing the comprehensive classification of approaches set out in Table One (Bell, 1958, 327-38). [Table One about here] In terms of methodological development, Bell comprehensively covers the state of scholarly research in relation to predicting Soviet behaviour (Table One). With hindsight Bell’s overview can be seen both as a survey of disciplinary approaches to forecasting Soviet behaviour, and also as a picture of a unified field of study emerging from a range of disciplinary approaches. The process of classification misleads if viewed too rigidly, since in reality an iterative process occurred within the various epistemological communities, as they debated, read, and drew on elements of each other’s work. Bell’s classificatory system represents the most detailed overview of approaches to the study of the Soviet Union’s future in the Cold War years. Later accounts, both during the Cold War and retrospectively after the collapse of the Soviet Union, concluded that a straightforward bifurcation of approaches existed. Dziewanowski dubbed these approaches evolutionist and apocalpytic (Dziewanowski, 1972, 376), to Lipset and Bence they were pluralist and totalitarian (Lipset & Bence, 1994, 180-93), to Seliktar revisionist and totalitarian (Seliktar, 2004, 203), and to McNeill liberal and realist (McNeill,