CHAPTER 2 Virtual History
Many times more books and essays on what their authors, following Niall Ferguson’s lead in his 1997 edited volume Virtual History, now call counterfactual history, have appeared since 1990 than in the whole of real history before then. Counterfactual histories have now become so frequent that they need investigating as a genre in themselves. They no longer take the form of parlor games or jeux d’esprit; on the contrary, they take themselves very seriously indeed. In these collections the concept of the counterfactual is quite different from the concept employed by econometricians such as Fogel. These historians are not engaged in a notional statistical calculation, they are trying to put forward a series of serious arguments about possible, real alternatives to what actually happened.
What makes them think that what they are doing is more serious than the parlor games played by some of their predecessors or the wishful thinking indulged in by others? Their answer is that their declared purpose is to restore free will and contingency to history and to reenthrone the individual actor in a history too often studied in terms of impersonal forces. “History involving great people or pivotal events,” complains Robert Cowley in the Introduction to More What If?, “is out of fashion. Broad trends, those waves that swell, break, and recede, are everything these days. We are left with the impression that history is inevitable, that what happened could not have happened any other way, and that drama and contingency have no place in the general scheme of human existence.”1 Similarly Jeremy Black argues that the purpose of counterfactual history is to emphasize “the contingent, undetermined character of historical change and to undermine any sense of the inevitability of the actual historical outcome.”2 “Of course this line of thought,” Andrew Roberts declares in advocating the use of counterfactuals, “infuriates the Whigs, Marxists and Determinists and anyone who believes that some kind of preordained Destiny or Fate or Providence determines human existence.”3 Both Black and Roberts urge historians to liberate themselves from the tyranny of hindsight and to try and see the past as the people who lived in it saw it, full of open and undetermined possible futures. “By engaging in counterfactual thought experiments,” argues Benjamin Wurgaft in a recent article, “intellectual historians could restore an awareness of sheer contingency to the stories we tell about the major texts and debates of intellectual history”4
Similarly, Geoffrey Parker and Philip Tetlock, in their edited collection Unmaking the West, denounce the presence among historians of a “bias toward retrospective determinism” and declare that “shattering hindsight complacency is the best way to make us appreciate how uncertain everything seemed before everyone became contaminated by outcome knowledge.”5 Simon Kaye, in a recent article arguing for the usefulness of counterfactuals, has declared that their main purpose is to counter “an unfortunate contemporary mindset of assumed deterministic certainty”; by underlining “the importance of human agency” in history.6 None of this was particularly new: all previous essays in the genre have emphasized chance and contingency in history; in 1982 John Merriman, for example, declared that “chance has not always received its due from historians seeking to explain the flow of world events”; and he argued that “an appreciation of chance helps rescue professional historians from the self-indulgent temptation to try to explain everything.”7 What was new about the counterfactual history of the 1990s and 2000s was the frequency and vehemence with which a belief in chance was asserted.
Not surprisingly, most of the historians writing counterfactual history along these lines have been both politically and methodologically conservative. To a degree, they even form something of a group: John Adamson, for example, contributed both to Ferguson’s collection and to Roberts’s; Roberts himself contributed to Ferguson’s book and to Cowley’s More What If? as well as his own; Geoffrey Parker wrote for Cowley in What If? and More What If? as well as for his own book; while Cowley himself contributed to Roberts’s volume. As Jeremy Black says, since counterfactualism gives importance to “agency—the actions usually of a small number of individuals,” it is clearly more favored by right-wing historians because, he suggests, “the latter are much more prepared to embrace the concepts of individualism and free will” than left-wing historians are.8 There are indeed few, if any, counterfactuals written from a left-wing point of view. Whatever the eddies and countercurrents along the way, the Left has generally believed that the tides of history flow in its favor. Why should left-wing historians regret what did not happen in the past when the future is still theirs? E. H. Carr’s own speculations about what might have happened had Lenin lived into a ripe old age is one of the exceptions to this general rule; socialist speculations on how Hitler might not have come to power in 1933 if the Left had formed a united front in opposition to him would be another; G. M. Trevelyan’s thoughts on what might have happened had Napoleon won the Battle of Waterloo a third. But in general, left-wing speculations of this kind are few and far between. Moreover, there are political and methodological reasons for the relative absence of left-wing versions of the counterfactual genre. As Tristram Hunt has pointed out,
In practice, therefore, counterfactuals have been more or less a monopoly of the Right.
The counterfactualists routinely stress their belief in contingency and chance as a corollary of their championing of the rights of the individual. There may, however, be more to this than meets the eye. E. H. Carr thought that “in a group or a nation which is riding in the trough, not on the crest, of historical events, theories that stress the role of chance or accident in history will be found to prevail.”10 The same might be said a fortiori of imaginative constructions of what might have been, grounded as so many of them are in a sense of regret at what actually is. It may be, indeed, that British historians of a conservative inclination began to emphasize the role of chance and think about how things might—perhaps, should—have turned out differently in the mid-1990s, because a long period of Conservative Party dominance, beginning in 1979, was clearly coming to an end, with the replacement of Conservative heroine Margaret Thatcher by the colorless John Major, whose policies were increasingly disputed by a new generation on the Tory right, to which the counterfactualists clearly belonged, and whose government, weakened and divided by faction, was replaced by Tony Blair’s New Labour administration in 1997.
The principal target of the counterfactualists was clearly Marxism, or what they imagined to be Marxism. Naming three leading Marxist historians, Andrew Roberts, for example, declares that “anything that has been condemned by Carr, Thompson and Hobsbawm must have something to recommend it.”11 Carr in fact did not really belong in this short list. From Carr’s point of view, formed during his early career, when he was a Foreign Office mandarin, history that emphasized the role of chance or speculated on what might have been was useless because it could not form the basis for policy, which is why he dismissed the history of defeated alternatives to what actually happened as not worth pursuing (he was not a Marxist or even a socialist historian, as some, including Niall Ferguson, have wrongly claimed; he was in his own work not really interested in causes at all, only outcomes).12 This was a dangerous doctrine, however, that reduced historical investigation to the status of an adjunct to present-day politics, and it did not recommend itself to most historians, least of all at a time, in the 1960s, when the new social history was uncovering and celebrating the forgotten and downtrodden in the past, from Eric Hobsbawm’s studies of peasant millenarians and social bandits in the Mediterranean world to Edward Thompson’s resolve “to rescue the poor stockinger, the Luddite cropper, the ‘obsolete’ hand-loom weaver, the ‘utopian’ artisan, and even the deluded follower of Joanna Southcott, from the enormous condescension of posterity.”13 Carr’s disdain for what he regarded as failures in history caused him to lump together investigations of history’s victims and losers with speculations about history’s missed alternatives and frustrated developments in a single pot.
Although Carr was never really a Marxist, genuine Marxist historians such as Hobsbawm and Thompson were firmly in the counterfactualists’ firing line. This was all the more striking in view of the fact that Marxism was fast disappearing from the intellectual scene when they were writing, following the fall of the Berlin Wall in 1989. In a sense the counterfactualists were flogging an ideological dead horse, or perhaps shooting a straw man by criticizing something that none of these historians, nor indeed any Marxist not shackled to the dogmas of Stalinism, had ever actually said. The idea that history should be open to notions of contingency and chance was not news to Hobsbawm or Thompson, and even Carr’s History of Soviet Russia gives over considerable space to the personalities and policies of individuals such as Trotsky, Bukharin, and Zinoviev in the aftermath of Lenin’s death.14 Real, dogmatic insistence on historical inevitability at every stage can only be found in political texts like Stalin’s Short Course history of Soviet Communism, not in practicing Marxist historians; and even Stalin and his followers ended up, of course, by insisting on the relative autonomy of the “superstructure,” or in other words, the freedom of Communist leaders like Lenin and himself to override broader historical processes and structures, for example, pushing the Russian Revolution on from a “bourgeois revolution” in February 1917—a revolution that Marxist theory suggested should lead to a liberal-capitalist regime lasting decades—to a “socialist revolution” within a few months.
Ignoring such easy targets, Ferguson in particular directed his fire at Marx and Engels themselves, above all for their contempt for “free will,” which he says they expressed in statements such as “Men make their own history but they do not know that they are making it,” or in the question posed by Marx: “Are men free to choose this or that form of society for themselves?” leading to the answer “By no means.”15 Yet Marx himself did not treat human beings as helpless in the face of big historical forces in the way that these carefully selected quotations suggest. In a famous statement not quoted by Ferguson, he declared that “People make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly found to be already there, given and transmitted from the past.”16 Moreover, Marx never intended his general theory to explain specific events (although he did not always stick to this in practice, least of all in his journalism), rather, he meant it as an aid to understanding general trends. Ultimately, the argument here is not about history at all but about free will. As Allan Megill says, all historians assume “that human beings are both determined and free, both subordinate to external forces and capable of creating and exploiting such forces.” Ferguson, Roberts, and their fellow counterfactualists are basing their arguments on an unrealistic polarization of total freedom of will on the one hand, and complete subservience to impersonal historical forces on the other, which very few historians, or indeed very few people in their own daily lives, would in practice endorse: most human action takes place between these two imaginary poles. So the bold declarations by Roberts, Black, Cowley, and others are beside the point; each case has to be decided on its own merits by historical investigation.17
It was not merely Marx and Engels who emphasized the constraining influence on human affairs of factors beyond human control. The great French historian Fernand Braudel, in his epic history of The Mediterranean and the Mediterranean World in the Age of Philip II, posited three levels of history, using the sea as a metaphor: the deep, unchanging, still waters of the environment that framed peasant life in a preindustrial age; the slow-moving waves of economic and social change; and the froth and foam of political history on the surface. Braudel went much further than Marx in reducing the room for maneuver of individual human beings, perhaps reflecting his own predicament as he wrote the first draft of his book in a German prisoner-of-war camp in the early 1940s. For him, the overwhelming majority of people in the early modern period, living on the land and subsisting from its produce, were mere insects, helpless in the face of larger forces. Braudel did not satisfactorily link his different levels of history to one another and so never developed a convincing approach to questions of historical causation. He was unable to show how political history was shaped by larger factors, except in rather minor ways like the time it took to get a letter from Milan to Madrid in the sixteenth century. His political history operated more or less independently from his social, economic, and environmental history and did indeed, as the last sections of his great study showed in what was a relatively conventional narrative of political events, have a good deal of room for maneuver in making decisions, fighting battles on land and at sea, or engaging in faction fighting at court; the point his book as a whole was trying to make was that all of this political activity made little difference to the lives of ordinary working people on the land (a demonstrably flawed assertion, as peasants quickly discovered when marauding armies marched across their fields, devastating the crops and spreading epidemics).18
So Braudel was not really a determinist in the sense of reducing political events to epiphenomena of larger historical forces. Indeed, when it came to narrating political history, he was as enthusiastic about counterfactuals as anyone else: “It may be a bad habit to rewrite history as it never was,” he once said, “to alter the course of major events so as to imagine what might have happened. But although a sleight-of-hand like this may be an illusion, it is not pointless. In its own way, it measures the weight of events, episodes and actors, who were believed, or believed themselves, to be responsible for the entire course of history.” Among other things, Braudel imagined a France in which Protestantism had become the dominant religion without any major conflict by being imposed on the country by Francis I, contemporary of Henry VIII of England.19
Nevertheless, Niall Ferguson, in the introduction to his Virtual History collection, declares that his collection of counterfactual essays is a “necessary antidote” to historical determinism such as that which he accuses Braudel of practicing and Marx of preaching.20 What is determinism, and how can counterfactuals undermine it? In its normal dictionary sense “determinism” means the belief that events are, or can be, caused by forces external to the human will, that is, that things don’t only happen because people want them to happen, they can also happen as a result of factors beyond their control. Put in this way, determinism would seem to be a doctrine that few historians in practice could object to; you can’t always get what you want, as the Rolling Stones remind us, and what you get might not be the result of what someone else wants either; long ago, Sir Herbert Butterfield pointed out that the titanic religious struggles of the Reformation and post-Reformation period between Catholics and Protestants ended in an outcome that neither of them wanted, namely, the emergence of religious toleration and Enlightenment scepticism.21
This all seems unexceptionable enough. But what Ferguson appears to mean by determinism is not the idea that human will is circumscribed by impersonal forces, but a number of other things, which it is important to disentangle. Indeed, as Aviezer Tucker has pointed out, for Ferguson, determinism is “a conceptual blanket that covers several otherwise independent historiographical doctrines and methods of which he disapproves.”22 It is important, therefore, to look at these in turn. There are four main concepts covered by the label of determinism in Ferguson’s justification of counterfactuals. First, there is teleology, the writing of history in terms of how it moved toward a given end. Whig historians like Trevelyan saw the whole of British history, and by implication the history of all other nations as well, as leading up to the establishment of liberal democracy; Marx saw history as leading up to the creation of a socialist society. Viewed on a global scale, both of these teleologies rearranged past events and developments along a straight line leading to a predicted but as yet unrealized future. Tucker argues that the claim that history is predetermined in this way cannot be tested by evidence since we don’t and can’t know the future. A counterfactual might be used to cast doubt on a teleology at some point along its trajectory, but it can’t undermine it as a whole because we don’t know where that trajectory will end. It follows therefore that counterfactuals can only cast doubt on teleological theories of history at a metaphysical level; a counterfactual scenario positing some possible alternative course taken by events in the short or medium term can always be got around by anyone determined enough to assert the validity of a teleology in the long term.
Second, by determinism Ferguson means the idea that political events are determined by social and economic forces.23 This, of course, is something that Marx and Engels never said; all they said was that the broad long-term development of history from one social formation to another was relatively unaffected by individuals and discrete political and military events. They were not, in other words, saying anything about the immediate causes of specific historical happenings. Nevertheless, these can obviously get in the way of the broader trends whose overriding importance they were so insistent on emphasizing. Ferguson goes on to note how later Marxists, beginning with the early Russian theorist Georgii Plekhanov, attempted to reconcile obvious contingent factors such as military victories in battles that might easily have gone the other way, or the chance that one or other historical figure, like Robespierre or the first Bonaparte, might have died prematurely, before making his mark, with the Marxist insistence on the ultimate inevitability of political changes consequent on changes in the mode and forces of production, by declaring that individuals might influence how things happened, when and where, but not the general trend of history, which was determined by broader factors, namely, changes in the relations of production. Counterfactual speculations, Ferguson suggests, can undermine such arguments by positing alternative courses of history that are completely detached from social and economic forces. Yet this argument carries with it the danger of reducing everything to chance and eliminating all consideration of longer-term causes of major historical events.
In trying to answer the question of why Hitler came to power, for example, Ferguson complains that German historians “remain deeply committed to the idea that ‘the German catastrophe’ had deep roots.”24 Is he really saying it was all a matter of chance? A more fruitful way of approaching it, surely, would be to follow the lead taken by Plekhanov and say that factors such as the structure of German society, the nature of German political culture since Bismarck, the weaknesses of the Weimar Republic, the disastrous performance of the German economy in the 1920s and early 1930s, the humiliations imposed on Germany by the victorious Allies in the Peace Settlement of 1919, and other, similar factors, clearly help explain why democracy failed, but they do not explain precisely the timing or the nature of Nazism’s triumph or indeed lead to the claim that the Nazis and not, say, a different form of dictatorial regime with the German military in charge and the Nazis only lending support would have come to power in 1933. For this one has to adduce further, personal factors such as Hitler’s refusal to enter any coalition government except as its head, the machinations of the clique gathered around President Hindenburg, and so on. Adducing these longer-term causes does not commit one to the view that Hitler’s seizure of power was inevitable, but it does suggest strongly that Weimar democracy was doomed by the early 1930s. It does not require a counterfactual to establish the role of contingency in these events, simply an examination of the evidence. What is crucial for the historian is working out how chance and contingency operate in a context that constrains the extent to which they can have an impact.
Many counterfactualists take as their starting point familiar chance events that we already know occurred: the flight of Louis XVI and Marie-Antoinette from Paris that was intercepted at the town of Varennes, resulting in their capture and eventual execution by the revolutionaries; or the failure of the Gunpowder Plot to blow up the English king and Parliament in 1605; or the wrong turn taken by Archduke Franz Ferdinand’s driver in Sarajevo in 1914 that brought him into the path of the assassin; or the ill winds that scattered the Spanish Armada in 1588. Often, their speculations center on the chance survival of assassination attempts, the untimely deaths of kings, the unfortunate mortality of their offspring, the sudden reversal of fortunes in battles. Few historians would deny that such chance events had a major effect; yet for serious consequences to ensue, most historians would agree that other, larger factors had to come into play, of whatever kind. It would be easily possible to argue for instance that had Louis XVI escaped to head up an Austrian-led army in trying to suppress the revolution, the determination of the revolutionaries to protect their achievements would only have been redoubled; that the death of the political elite in England in 1605 would have radicalized English Protestants, who would have exterminated the Catholics in a bloodbath of biblical proportions; that the Bosnian Serb terrorists who failed to kill Franz Ferdinand in June 1914 would have tried again until they succeeded, with similarly cataclysmic results; or that a Spanish army landing in England in 1588 would have met with a crushing defeat by Elizabeth I and her armies and other supporters, so popular had she become by that time. Any of these events could easily have ended otherwise than it did, but larger changes in the historical context would probably have been necessary for this to have had the effects sometimes claimed by counterfactualists. Counterfactuals such as these can only be posited as having big effects by leaving out the historical context; though bringing it in does not mean the larger scheme of things would have been the same had the small event gone differently, it does reduce the scale of the change.
These examples have already taken us a long way from social and economic forces. The real question at issue here, however, is not the nature of the larger historical forces concerned, but how one prioritizes them. This brings us to the third type of determinism allegedly vulnerable to attack from counterfactuals, namely, the idea that history is determined by general laws of development such as those laid down for example by Arnold Toynbee in his Study of History.25 The debate on laws in history goes back a long way, but at no point have historians actually accepted that history is governed by laws operating in the sense of scientific laws; history is never rigidly predictable in the way that, say, chemical reactions are. We might emphasize some kinds of causal influences over others, or deploy a theory, say, a variant of Marxism or modernization theory, to help explain the past, but saying that some factors in history depend on others—say, cultural developments are dependent on social developments, or military strength is dependent on economic strength—does not commit us to a belief in predictability. And if we say that some factors are dependent on others, we are also saying that some are more independent than others. “Accident is relative to a system, not indeterminist.”26 Historians never operationalize theory in a rigid or undeviating way; the messiness of the past and the evidence it has left behind do not allow this. As Plekhanov suggested, Marxist historians commonly use “coincidental intervening variables and auxiliary theories to explain away the obvious deviation of history from what their favored theory would lead us to expect.”27 In Tucker’s words, “historiographical counterfactuals as such are not a challenge to privileging theories as such.” If determinism means that there are laws of historical development, then, according to Tucker, it is not clear “how counterfactuals can do a better job of refuting these theories than plain historical facts.”28
Fourth, following Hayden White, Ferguson argues that writing history as narrative trope—presenting a historical sequence as tragic or comic, for example—is necessarily deterministic because it presumes a specific outcome by virtue of historical style rather than historical content. Yet in my view it is not necessarily deterministic at all, because the trope does not precede the narrative but follows it. In other words, we put together a narrative then decide whether it is tragic or comic according to what seems fitting. Nor does this in any way commit the narrator to a particular view of discrete events or personalities; rather, it is a judgment on them, reached at the end of the presentation of the narrative. Hayden White, of course, argued that the framing of a narrative in one of these modes predetermined the selection of the constituent elements of the narrative. Any history beyond mere chronicle or antiquarianism was in his view a metahistory expressing a particular narrative trope created by the historian and not testable by empirical research. But of course, any practicing historian will tell you that this is not the case; we may devise an interpretative hypothesis but we are constantly revising it by examining the evidence for and against it, and our end product is likely to come up with something rather different than what we started out with.29
White’s claim that all history can be allocated to one or other of a small number of narrative tropes may hold good for the classic Victorian epics he studied like Macaulay’s History of England, but it does not apply to modern analytical history, which he does not discuss at all. And in practice there is always evidence that will not fit comfortably into a tragic or comic story, which we cannot ignore or explain away. Most historians who have told the life story of Mary, Queen of Scots, for example, have seen it as a tragedy with Mary as the victim of Tudor power politics: Mary was brought up in sixteenth-century France, expelled from Scotland as a Catholic by grim and censorious Protestants after a turbulent private and political life on the throne, imprisoned by Queen Elizabeth I of England as a threat to her position and that of the Protestant Reformation because she was next in line to the English throne, and finally executed after being implicated in a series of plots to seize the throne for herself. But nobody has sought to deny that she was herself in part to blame for her travails and her eventual death, and any historian who for instance simply failed to mention her involvement in the Babington Plot or the Throckmorton Plot, both of which aimed with Mary’s approval to kill Elizabeth and seize the throne, would not be taken seriously.
In practice, therefore, choosing a particular narrative mode is only a broad moral judgment, it does not impose any specific interpretation or moral judgment on the particularities of the evidence, and it does not deny the operation of contingency (for example, the exposure of Throckmorton when he was observed by the queen’s agents visiting the French embassy with suspicious frequency). In the end, too, White concedes that it is possible to ascertain the factuality or otherwise of particular events in history through the use of evidence (a concession he was forced to make when it was pointed out to him that scepticism on this point opened the door to Holocaust denial). Thus—as every practicing historian knows—factual evidence plays a significant role on its own account in shaping a narrative; we cannot simply select the facts that suit our narrative and ignore the ones that don’t. What Ferguson calls “narrative determinism” doesn’t therefore imply an arbitrary process of constructing a form of historical inevitability, nor, once more, is it easy to see how counterfactuals are necessary to undermine this process, since narrative by its very nature is going to mention chance events, things that might have turned out differently from the way they actually did.30
At times, Ferguson’s attack on determinism seems to verge on an attack not only on any notion of wider forces and deeper currents in history, but on any concept of causation at all. Calling the development of chaos theory and the scientific concept of indeterminacy as his witnesses, he suggests that human affairs are becoming more subject to individual free will, not less. For Ferguson, almost any attempt to argue that a large event or process like the rise and fall of the British Empire or the outbreak of the English Civil War, has large causes, is convicted of determinism because in listing a series or hierarchy of causes, a historian like—in these two cases, respectively, Paul Kennedy and Lawrence Stone31—strongly implies that the event was inevitable. Ferguson makes great play with chaos theory, symbolized by the fluttering of a butterfly’s wings in Japan leading to a hurricane in Bermuda—an example, incidentally, of the linkage of tiny causes to large effects so beloved of counterfactualist historians and so scorned in theory by Ferguson in his critique of Merriman’s volume. Those who argue that history is essentially chaotic—that is, predominantly the produce of chance—have always had to concede that larger systems of causation still exist, determining the overall pattern of events though not their precise nature or timing. Chaos theory is not susceptible to mathematical proof in its application to history, as it is in science; in fact, it is not applicable at all. There is no measurable evidence that human affairs are becoming more chaotic, whatever may be happening in the universe out there. “Believers in chaotic history,” as Tucker points out, “must assume counterfactual situations in which the initial conditions were slightly different but the outcome was radically other.”32 But this depends of course on isolating a particular cause or condition and identifying it as necessary rather than subsidiary; in other words, in order to create a counterfactual, we have to say that the condition or cause that we are altering was the decisive one, otherwise we are simply altering a rather trivial factor that most historians would agree made little difference to the outcome under consideration. This leads us back again to the very notion of a hierarchy of causes that Ferguson considers so deterministic.
Counterfactuals, then, do not do the job of undermining determinism in any of its possible meanings as well as simple factual evidence does. But this is not of course the only way in which their proponents claim that they are useful or even, in some cases, essential. They can show that particular events and decisions were not inevitable but governed by chance and contingency. By carefully scrutinizing the options confronting decision makers in an event such as, for example, the outbreak of World War I, we can come to understand better the decisions they eventually took. In doing so, says Ferguson, we should examine the possible alternatives to what happened, but of course, he adds, crucially, the number and variety of these possible alternatives is not infinite. On the contrary, it is obvious that the historian should only consider alternative outcomes that were at least plausible, or in other words, he says, possible alternatives that contemporaries themselves actually considered and wrote about (for if they did not write about them there is no evidence that they considered them). On the face of it, this seems a perfectly reasonable way of proceeding. This would, however, rule out factors such as impulsive behavior, human accident, unanticipated errors, and the like, and so reduces the role of contingency to a negligible quantity because only carefully considered and debated conditions could be taken into account.33 Moreover, by conceding that the number of possibilities was finite, Ferguson is to my mind actually agreeing with Marx’s dictum that people do not exercise their will with absolute freedom because the conditions under which they live and act are not of their own choosing and are often subject to forces beyond their control. It is after all such forces—economic, power-political, cultural, social, intellectual, geographical, or whatever—that constrain the operation of the human will, not any defects or weaknesses in the human will itself, as many a tyrant, from Napoleon to Hitler, has found to his cost. Still, the constraints are no more than constraints, and within them, decision makers were still able to choose from a number of options available to them.
Thus, for example, in July and August 1914, British foreign secretary Sir Edward Grey had the option of declaring war on Germany, or keeping Britain out of the conflict by remaining neutral; a declaration of war on, say, France or Russia was not an option in the context of the time. What can the counterfactual hypothesis of British neutrality do to help us understand why Grey did not go down that particular road? In his contribution to Virtual History—“The Kaiser’s European Union”—and in his book The Pity of War, Ferguson examines this question in detail. After the war, he points out, and ever since, participants and historians have concluded that Britain’s involvement in the war was inevitable, given the German violation of Belgian neutrality, which was guaranteed by the British, and the threat a German victory posed to the balance of power in Europe and, above all, Britain’s empire overseas. Neutrality seemed out of the question. Yet, says Ferguson,
In Ferguson’s view, what would have happened in the event of neutrality shows conclusively that Grey’s decision to enter the war was the wrong one. If Britain had stayed out, German war aims would have been more modest, and the Germans would have won.
Having achieved the hegemony over the rest of Europe that they had achieved by the end of the twentieth century anyway, through their creation and domination of the European Union, the Germans would not have felt frustrated and defeated, there would have been no Hitler and no Second World War, and there would have been no mass slaughter on the battlefields of Europe between 1939 and 1945, no gas chambers, no Holocaust. As Ferguson writes, “Had Britain stood aside even for a matter of weeks [in 1914], continental Europe could therefore have been transformed into something not wholly unlike the European Union we know today, but without the massive contraction in British overseas power entailed by the fighting of the two world wars.”35 The British Empire would have survived, and Britain would have continued to be a superpower throughout the twentieth century instead of declining to the level of a mere constituent part of a German-dominated united Europe.
Ferguson’s counterfactual speculations, written in 1997, dovetailed with those of Churchill: The End of Glory, a book published by another young British conservative historian, John Charmley, five years before. Charmley broke any number of taboos with this highly critical account of Winston Churchill’s leadership in the Second World War.36 Charmley accused Churchill, after he became prime minister in 1940, a few months following the outbreak of the Second World War, of making a fatal decision by insisting on continuing to fight Hitler. In doing so, he threw away vast sums of money, leaving Britain too weak to hold on to the British Empire after the war was over. Yet this was not just bad for Britain, it was bad for the colonies too, since they would have been much better off in every way had they not achieved independence but remained under the benign and civilizing control of London. Churchill surrendered British freedom of action to the United States, who lent money and equipment to Britain on a grand scale, and used the debt to blackmail the United Kingdom into surrendering the empire—one of the chief aims of U.S. president Franklin D. Roosevelt all along. Roosevelt ran rings round Churchill at the peace summits, laying the foundations for a postwar world order dominated by America rather than Britain. The war failed to protect Poland from tyranny, first that of Hitler then that of Stalin. At home, Churchill focused obsessively on the conduct of the war, allowing the Labour Party ministers in his coalition government to convince the electorate of their fitness to govern. This enabled their victory in the 1945 election and what Charmley sees as their disastrous creation of the modern welfare state between 1945 and 1951, a development that created a culture of dependency and complacency that led to a rapid decline of Britain until Mrs. Thatcher’s radical reforming Tory government began to rein back the state and create a culture of enterprise and initiative in the decade after 1979. Charmley’s conclusion was devastating: Churchill, he declared, stood for the British Empire, for British independence, and for an “anti-Socialist” vision of Britain; but by July 1945 the first of these was about to fall, the second was solely dependent on America, and the third had just vanished in a Labour election victory.
How different everything would have been, claimed the maverick right-wing Tory historian and politician Alan Clark when he reviewed Charmley’s book, had Churchill given way to the appeasers, led by Lord Halifax, in the spring of 1940, when they were urging a compromise peace with Germany following the catastrophic defeat at Dunkirk. Churchill foolishly rejected the “excellent terms” that Clark claimed the deputy leader of the Third Reich, Rudolf Hess, brought with him when he flew to Scotland a year later. If he had made peace with the Nazis, Clark argued, Churchill could have saved the empire by moving forces from the European theatre to the Far East to protect Malaya, Singapore, and Burma from the Japanese. With the British out of the war, the Americans would not have come in, and Hitler’s Third Reich and Stalin’s Soviet Union would have fought each other to a standstill on the eastern front, destroying each other in the process and weakening Russia to such an extent that it would not have been able to extend its grip on the rest of Eastern Europe after the war. Moreover, leaving the war in 1940 or even in May 1941 would not have meant abandoning the Jews of Europe to their fate, since the Nazi extermination program had not yet got under way, and in any case the British did little or nothing to rescue the Jews even when they did decide to carry on fighting the Nazis. At the same time, Charmley pointed out that in reality British governments had done nothing to prevent the mass murders and forced deportation of entire peoples in the Soviet Union during and after the war. Britain had no reason to be proud of its conduct in the Second World War, then. As both Clark and Charmley insisted, British involvement in the quarrels of continental European powers had been disastrous.37
These counterfactual speculations rest on an extremely shaky empirical basis. There are five main areas in which they fail to convince. First, as far as August 1914 is concerned, the argument that Germany was not a threat to British interests is contradicted by Ferguson himself, who in The Pity of War does indeed argue at one point that Germany was not powerful enough to be a serious challenge to Britain in 1914, but goes on to assert at another that the German war machine was more efficient and effective than the British because it killed more of the enemy at a lower financial cost to itself.38 Actually, if you take the principle of calculation behind it you will find that the most efficient nations in the war were Turkey and Serbia, which spent only a tiny proportion of their national wealth on the war but succeeded in killing a very large number of their enemies all the same. The calculation, in the end, doesn’t mean very much. In any case, there is no actual evidence that German war aims would have been more limited if Britain had stayed out of the war, or, to put it another way, there is no evidence that Britain’s entry into the war led to an immediate widening of German war aims beyond what they had originally been. Most of the aims of the notorious September Program of 1914 did not concern Britain anyway. What peace terms, in any case—to move on to the second set of empirical, evidential problems with the counterfactual of British neutrality—would Hitler have offered? In a number of speeches following the German conquest of western Europe, he made a point of holding out an olive branch to the British, but close examination of the speeches reveals them to have been entirely lacking in specificities of any kind. Actual negotiations never began, so we shall never know what conditions might have been laid down. Undoubtedly the spring of 1940 was the crucial moment, and it is entirely possible that Halifax’s peace party might have come out on top, though we do not need counterfactual speculation to tell us that. As for Rudolf Hess’s flight to Scotland, all the available German sources indicate that it was undertaken without Hitler’s approval, and that Hitler and the other Nazi leaders were shocked and dismayed when they learned about it. He did not bring any “terms,” excellent or otherwise. No convincing documentation exists of any attempt by British appeasers to invite him over.39
One might argue that had Britain not entered, or stayed in, either of the two wars, the German government of the day would eventually have declared war on Britain anyway, and with vastly greater chances of success had it succeeded in defeating its continental enemies in the meantime, or alternatively used a compromise peace to dismantle the British Empire bit by bit. This at least is what Churchill himself told his war cabinet on 28 May 1940:
Churchill’s views provided a better basis for counterfactual speculation than either Ferguson or Charmley was able to muster. Contrary “what-if” speculations can be proposed that are just as plausible as the ones put forward by those historians who think that Britain would have done better to have kept out of European wars and let the continentals slug it out among themselves. Take, for instance, what one might perhaps term a counter-counterfactual about the Second World War, by the left-liberal historian Paul Addison:
This is at least as plausible as the scenario painted by Charmley and Clark.
Indeed Andrew Roberts has endorsed the view taken by Addison. He argues that a separate peace with Britain would have led to a German victory over the Soviet Union because Hitler would not have needed to deploy troops in the Mediterranean, or at least North Africa, to support Mussolini. Hitler’s assurances to the British, contrary to what Charmley assumes, were patently insincere, and victory in Russia would have been followed by a full-scale invasion of the United Kingdom. In rejecting a separate peace, says Roberts, “Churchill was therefore right.”42 However, Roberts is stretching his counterfactual to breaking point and beyond, since it is extremely uncertain whether Hitler would actually have been able to defeat the Soviet Union even with Britain out of the reckoning. C. J. Sansom, author of the counterfactual novel Dominion, published in 2012, which takes a German-dominated Britain in the 1950s as its historical context, still thinks the war in the east was “always militarily unwinnable [for the Germans]; the country was just too vast, and the population totally hostile.”43 Contrary to what Roberts claims, the weakening of Operation Barbarossa by the deployment of German troops to rescue the failed Italians at least to the north of the Mediterranean would most likely have happened anyway, given the Italians’ lamentable performance against the Greek army in 1941. The “ifs” here are too large and too numerous to be plausible, especially given the vast depth and breadth of the Soviet Union’s resources. From the start of Operation Barbarossa, the Germans never committed less than two-thirds of their military resources to the Eastern Front, and a bit more would probably not have made that much difference.
Third, as far as the British Empire is concerned, nobody can prove that Britain would have kept the empire had it remained neutral in either 1914 or 1939–45. Had Germany won the First World War following a British refusal to fight, who is to say that the Kaiser and his generals would not have turned their envious eyes toward the British Empire, for which their High Seas battle fleet, under construction since Tirpitz’s plan to confront and cripple or destroy the Royal Navy in the North Sea, had been launched at the end of the nineteenth century? Historians are generally agreed that Germany’s longer-term foreign policy aims leading up to both 1914 and 1939 went much further than merely establishing economic hegemony over the European continent. In both cases there are clear contemporary indications that the German government was issuing a challenge, among other things, to the British Empire and British world power. Hitler’s professed admiration for the empire, cited by Ferguson as an indication that he might have wanted to preserve it, was intended in reality to hold it up as an example for Germany to follow. Yet another counterfactual scenario, developed by the German-American historian Holger Herwig, has the collapse of the British Empire actually preceding a separate peace, with a German victory in the east followed by the German conquest of the British-ruled Middle East and a threatened invasion of India.44
In reality, of course, there is a huge amount of evidence to indicate that the continued rise of American power on the one hand, and more importantly the changing and evolving nature of society and politics in the Indian, African, and other colonies on the other, were the real forces behind decolonization in the post–Second World War era, in other words, that this was a process that no amount of money and resources saved by Britain not spending billions of pounds on the war effort could have brought to a halt. The British Empire had been in long-term decline since the Boer War, and noninvolvement in a European war, Sansom has claimed, would only have accelerated this process, not only in 1940 but also, arguably, in 1914. Parts of the Empire, such as Australia or New Zealand, would not have accepted a separate peace in 1940, he speculates, and a link with Germany would have sparked uncontrollable unrest in India (itself already on the way to Dominion status by 1939). The weakness of a subjugated Britain would have been obvious to everyone across the globe, and would probably have led to an upsurge in national independence movements, above all in South Asia.45 Finally, the hypothesis that Britain might have been better off had the empire been retained ignores the enormous cost of running the empire in terms of manpower and resources as well as the increasingly negative impact of British rule in its later phases, from atrocities in Kenya and Malaya to famine and disease in India. Not every post-colonial state formerly ruled by the British has failed to develop or disintegrated into chaos and civil war. Moreover, it is just as arguable that the liberation of Britain from the vast financial burden of an excessive overseas commitment after the Second World War was essential for the general improvement of living standards and the economic boom of the 1950s and 1960s at home and so benefited Britain as well as her former colonies.
Fourth, it is empirically unsustainable to liken a Europe dominated by Germany in 1918 as a European Union avant la lettre or to describe the European Union of today as a vehicle for the German domination of Europe. During the 1990s, when Ferguson and Charmley were writing, Germany was turned inward, absorbed in the unexpectedly gargantuan task of bringing East Germany into the modern world after the fall of the Berlin Wall. Talk of a “Fourth Reich” was alarmist fantasy; Germany since 1945 has been afraid of power, not still wedded to it. And in any case, the European Union is simply too large and complex an entity to sustain sweeping generalizations about German domination. Even describing it slightly more accurately as resting on the joint hegemony of Germany and France, on the Franco-German partnership, does not do it justice, least of all in the expanded European Union of twenty-eight nations, today, and certainly ignores the way in which key decisions have to be taken, through the Council of Ministers, with a single-country veto on many crucial issues. German financial policy has, it is true, imposed stringent austerity on southern European economies within the Eurozone, exporting its traumatic fears of inflation, derived from the experience of 1923, and there is a case to be made for a stronger element of reflation to stimulate economic growth. However, it is clear that this bears little or no resemblance to German plans in 1914 for the creation of a central European economic area under German domination, still less to the brutal methods of plunder and exploitation imposed on occupied countries by Nazi Germany in the 1940s. The Kaiser’s European Union might well have been based on the territorially expanded Germany that the September Program and subsequent drafts of German war aims envisaged, not the much smaller Germany of today, and could have been bolstered by the imposition of an authoritarian and hierarchical system of rule, the denial of human rights to minorities, the curbing of trade unions and the enforced imposition of German customs and German institutions. We do not need to speculate about what the “New Order” in Europe meant under the Nazis: ruthless exploitation, mass murder, constant militarization. The one thing that can be said is that had the Nazis won World War II, it is highly unlikely that supposedly moderate and pragmatic elements like munitions minister Albert Speer would then have taken over and created a “New Order” that was more or less the same as the European Union later became; the EU was, and is, based on ideals of peace and compromise, both utterly foreign to Nazis, including Speer.46
Fifth, the counterfactual implicit in Charmley’s claim that had Labour not come to power in 1945 and inaugurated the welfare state, Britain would have been more successful economically because a culture of enterprise would have replaced what became a culture of dependency, is implausible on a number of grounds. The Beveridge Report that provided the foundations for policy in this area was accepted not only by the Labour Party but also by the Conservatives, who made no attempt to reverse its consequences when they came to power in 1951. The war had generated a consensus behind the idea of a welfare state that transcended party boundaries.47 Nor is there any evidence that this proved a hindrance to British economic performance, as postwar recovery was followed by the boom years of the late fifties and sixties. What changed the situation was the oil crisis of 1973, but while Mrs. Thatcher’s liberalization of the economy, privatization of industry and utilities, and deregulation of banks in the 1980s created the conditions for another upturn in the economy, the downside of her reforms was starkly revealed as the credit crunch of 2009 inaugurated a crisis that was made much worse by the failure of successive governments to introduce proper regulation of the banking system so hastily liberalized a quarter of a century before. Here again, the grounds for a counterfactual are empirically too weak to sustain it.
So the British neutrality counterfactual does not stand up to close examination. To the versions proposed by Ferguson and Charmley can be opposed quite different versions proposed by others, by Addison, Roberts, Herwig, or Sansom. Historians customarily disagree about interpretations of the past, of course, but here we have disagreements not about interpretations but about actual facts, or rather, of course, about imagined facts. And therein lies the problem. All these authors posit long-term consequences from the changing of a single event. Imagined British neutrality in World War I or World War II has imagined consequences extending for several decades, right up to the end of the twentieth century. The trouble with counterfactual arguments used in this way is that far from liberating history from an imaginary straitjacket of Marxist determinism they confine it in another that in reality is far more constricting. This is because the counterfactual proposes an alternative future, in the sense of “what if A had happened instead of B?” then assumes or posits a whole series of other things that would have inevitably followed: “If A had happened instead of B, then inevitably C, D, and E would have followed, instead of what actually happened, namely, X, Y, and Z.” But of course a thousand other things might, or would, have intervened to make this alternative course of events in practice completely unpredictable. If, for example, Britain had not entered the First World War, it is in practice completely inevitabilist to argue that this one change in the pattern of events would ineluctably have led to a whole chain of other events, from German victory in the First World War to the failure of Hitler to seize power and the retention of the British Empire all the way up to the end of the century. Chance and contingency are here eliminated altogether. Other factors extraneous to the limited question of relations between Britain and Germany are completely ignored, from the whole involvement and purposes of Germany in eastern Europe to the entire development of relations between the British colonizers and the Indian and African colonized. What would have occurred if other contingencies had intervened along the way? There is no way of knowing, but once you let the counterfactual genie out of the bottle, anything might happen.
Conditional statements of such hard-edged certainty are foreign to the historian’s way of going about explanation, which is almost invariably tentative and involves considerable use of the word “probably.” “Monocausal” explanations make historians uneasy; we prefer to pile up causes until events are overdetermined, that is, they have so many causes that if one did not operate the others would, and the event in question would still have occurred. The key of course lies in designating one cause as more operative than others. Historians usually construct hierarchies of causes—primary causes, secondary causes, main causes, subsidiary causes, and so on, which affect different parts of the explanation. If we are looking, for example, for causes of the First World War, we can say that the assassination of the heir to the Austro-Hungarian throne was a cause, but only of the Austrian ultimatum to Serbia (which had aided and abetted the assassins), and further, more important causal chains linking the Austro-Serbian conflict to wider European great-power rivalries needed to come into operation before the Austrian declaration of war on Serbia prompted a Russian declaration of war on Austria, a German declaration of war on Russia, and so on. British historians argued for some time that the principal cause of Britain’s involvement in the war was the German invasion of Belgium, whose neutrality Britain had guaranteed, but this simplistic view is no longer very widely accepted, and historians instead look to other factors, from Anglo-German naval rivalry to the British policy of maintaining the balance of power on the European continent. In none of these explanations is it very helpful to start thinking about what might have happened if things had been otherwise—for example, if the Austrian heir to the throne had not been assassinated, if Serbia had fully and unconditionally accepted the Austrian ultimatum, if the Russians had held back, and so on. Our main interest is in putting together the chain of events, not in unpicking it. We know that the British cabinet debated seriously whether or not to enter the war, and that Foreign Secretary Grey’s fears of a German-dominated Europe eventually won the day; we know this from the evidence of its deliberations, and we do not need any counterfactuals to help us understand how or why events took the course they did.
As Allan Megill has pointed out, the virtual historian usually has a starting point in real history, normally just before some momentous decision is made. Johannes Bulhof has helpfully noted that “counterfactuals are sentences that are of the form ‘if p then q’ where … the antecedent (the term following the ‘if’) is in fact false.”48 This moment of decision is conceptualized as a moment of contingency, when things might easily have gone in a different direction from the one they actually took, and what the counterfactual historian then does is to extrapolate history in one or other of those different directions. But this involves a huge range of assumptions about how history at that particular time, and in the following months, years, and sometimes centuries, operated. These assumptions necessarily eliminate contingency rather than underlining its importance and influence. As Megill notes, “contingency cuts two ways,” for if we have contingency at the outset of a counterfactual speculation, then we must also have it in the early middle, the middle, and the late stages, indeed all the way through. Thus “contingency is not a train one can get on or off at will,” so that counterfactual history in this sense “cannot follow any definable course at all. More precisely, it can follow a definable course only until the next contingency arises.” It is thus not history but fiction, or “imaginative history.”49 In a similar way, Martin Bunzl suggests that in a counterfactual chain of reasoning, the prime condition of plausibility is undermined further with each successive step taken. For example, the counterfactual claim that if Al Gore had been elected president of the United States, America would not have invaded Afghanistan, elides numerous intermediate links in the counterfactual chain of reasoning, ending in a counterfactual conditional that is so remote from the initial false claim—the entirely plausible hypothesis that Al Gore might have won the election—that it cannot in any way carry conviction. Thus what Bunzl calls the consequent of a historical counterfactual is always the product of an act of imagination, unprovable because of the lack of evidence. He can only carry plausibility insofar as the imagination is disciplined by historical knowledge.50
This point has been made in various ways by many analysts of counterfactualism, including some of its practitioners. Jon Elster, for example, has pointed out that the reality created by an alternative timeline would be totally different from the actual timeline because everything is in reality connected; it is entirely artificial to select one element and leave everything else unchanged while we trace its imagined consequences from the original point of departure.51 Because everything is connected, it is “arbitrary,” as Steven Lukes notes, “to propose that counterfactual reasoning should be confined to imagining (a subclass of) possible antecedents—namely, those which the theory we use to infer the consequent tells us are compatible with the elements of the world that are not assumed to vary.” Why should we not alter everything, he asks.52 Similarly, the French writer Emmanuel Carrère concludes his discussion of counterfactuals by noting that “the trajectory of the counterfactualist cannot be a single line. … It is … a sequence of innumerable points, and in departing from each of these points, a multitude of possibilities freely radiates.”53
Undermining his own contribution to Virtual History, Jonathan Clark points out:
A long-term counterfactual of the kind unfolded by Ferguson and Charmley, or even, in the case he himself discusses, namely, the proposition that if the American Revolution had not occurred in 1776 the French state would not have been overburdened by the financial cost of supporting it and so there would have been no French Revolution in 1789, is, as Clark notes, “so large, and so far removed from the actual outcome, that it loses touch with historical enquiry.”55 Statements such as the claim that had Britain stayed out of the First World War, then the German-dominated European Union would have been born much earlier than it was can never take the form of real retrospective predictions, since they do not specify the hundreds of intervening variables relevant to the prediction. In other words, the deployment of the what-if type of argument to posit a long-term alternative development assumes first, the nonexistence of any further contingencies and chances along the way, and second, the noninfluence of the initial alternative event, the first thing that didn’t happen, on subsequent events in ways that could not be foreseen or predicted. All of this removes chance and contingency from history almost totally. Instead of restoring open futures to the past, it closes them off.
This is a point that two of the most sophisticated advocates of counterfactual history, Geoffrey Parker and Philip E. Tetlock, perhaps surprisingly, concede. In the conclusion to their volume Unmaking the West, they note that
Thus in their view, anyone who emphasizes the role of chance and contingency in history will put little faith in the utility of counterfactuals; only those who believe in teleology and determinism will find them useful. Yet in both cases the historians concerned are caught in a paradox. Believers in chance and contingency as the key factors in history cannot on this showing accept the plausibility or usefulness of counterfactuals; believers in teleology and determinism will not in any event feel the need to employ them.
Ferguson does not in the end claim that the British neutrality counterfactual helps us understand the decisions taken in 1914 and 1940. All he says is that “only by attempting to understand what would have happened had Britain stood aside can we be sure the right decision was made.”57 But the fact that Addison’s counterfactual scenario, like Herwig’s or Sansom’s, is mainly a negative one points to the unavoidable conclusion that the positive counterfactual scenario painted by Ferguson, Charmley, and Clark is in the end little more than a rather obvious form of wishful thinking; rather than “what if,” it’s really little more than “if only.” In this form it contributes nothing to our understanding of what actually did happen, because its concern is not really with examining how and why people like Sir Edward Grey or Winston Churchill took the decisions they took. Instead it’s concerned with pointing up what are supposedly preferable alternatives and bemoaning the fact that they never came to pass. This kind of speculation is easy to do because, detached from the need to back arguments with actual, concrete evidence, it allows historians to rewrite history according to their present-day political purposes and prejudices. As Jonathan Clark warns: “Analysts of the counterfactual must beware of that easy escape which is offered by the argument that, but for some initial mistake, some tragic error, all would have been well, and mankind released from avoidable conflicts into a golden age of peaceful progress.”58 How that golden age might have looked is not a matter of informed counterfactual speculation but the product of purposive political motivation. Like the contributors to Daniel Snowman’s collection If I Had Been …, Niall Ferguson and John Charmley and Alan Clark spend their time telling people in the past what they should have done, not trying to find out what they did and explain it. Wishful thinking is everywhere in the world of historical counterfactual. In all these cases, “the need for consolation,” as Clark concludes, “overrides the desire for explanation.”59