Virtual History: Alternatives and Counterfactuals (4 page)

In the wake of this disaster Vincent imagines a period of ‘stability’ lasting well into the eighteenth century; but this ends with another calamity: ‘the collapse of state credit after defeat in the French war, and the concession to France of its “natural frontier” on the Thames’.

After this, things deteriorate rapidly, so that the nineteenth century becomes England’s nadir, rather than its zenith:

But the worst was still to come:

Unlike so many of the contributors to Squire and Merriman, neither Russell, Trevor-Roper nor Vincent can really be accused of wishful thinking. Nor are their assumptions reductive to the point of being merely humorous. In each case, a serious historical point is being made about the contingency of English ‘exceptionalism’. Yet their various contributions remain no more than suggestions, with only the sketchiest of supporting evidence. They are brilliantly formulated counterfactual
questions
, not answers.

A wholly different use of counterfactual argumentation has been made by exponents of the so-called New Economic History.
²⁷
The first serious venture into quantitative counterfactual argumentation, R. W. Fogel’s work on the contribution of railways to American economic growth, sought to construct a model of US economic development without railways in order to challenge the traditional assumption that they had been indispensable to American industrialisation. According to his calculations, if no railways had been built, US GNP would have been only slightly lower than it actually was in 1890, though the area of cultivated land would have been substantially smaller.
²⁸
Similar methods have been used by McCloskey and others in the debate on Britain’s relative economic decline after 1870.
²⁹

There is no wishful thinking here, and certainly no humour. However, there are serious objections to such ‘cliometric’ arguments. The most frequent is that the relatively narrow base of nineteenth-century statistics cannot sustain the edifice of extrapolation and calculation built upon it.
³⁰
In so far as this objection has been directed at Fogel’s work on the economics of slavery, it clearly has a political subtext: his argument that, but for the Civil War, slavery could have been sustained economically was naturally an unpopular one with many American liberals.
³¹
But it applies with considerable force to his work on railways too. Only by making fairly heroic assumptions about ’backward and forward linkages’ was Fogel able to conjure up - even if only on a computer print-out - an America without railways. A more serious objection to his approach is that the counterfactual scenarios in question lack historical plausibility - not because they are reductive or frivolous, but because they are anachronistic. Contemporary debates about railways were generally not about
whether
to build them but about
where
to build them. The best defence of Fogel is that the purpose of calculating the ‘social savings’ afforded by railways is not to conjure up a plausible alternative history but to test a hypothesis about the role of railways in economic growth. No one is in fact trying to ‘imagine’ nineteenth-century America without railways. Indeed, the ultimate effect of this kind of counterfactual is to show precisely why the railways
were
built, by quantifying their (quite considerable) contribution to the economy as a whole. In a similar way, the debate on economic policy options in the last years of the Weimar republic has tended to show that there were no politically viable alternatives to the deflationary measures implemented by Chancellor Brüning between 1930 and 1932.
³²

There are, in other words, two distinct kinds of counterfactual which have been used by historians: those which are essentially the products of imagination but (generally) lack an empirical basis; and those designed to test hypotheses by (supposedly) empirical means, which eschew imagination in favour of computation. In the case of the former, it is the tendency to rely for inspiration on hindsight, or to posit reductive explanations, which leads to implausibility. In the case of the latter, it is the tendency to make anachronistic assumptions. Just how hard it is to overcome these difficulties can be seen in the path-breaking attempt by Geoffrey Hawthorn to combine elements of both approaches.
³³
In one of his supposedly ‘plausible worlds’, he ‘subtracts’ the plague from French medieval history, imagining a consequent fall in rural fertility in France and a consequent acceleration in the pace of French economic and political modernisation in the eighteenth century. In another, he imagines the consequences of American non-intervention in Korea after the Second World War; and in a third he diverts the course of Italian art of the late Duecento and early Trecento away from the innovations which were the harbingers of the Renaissance. The second example has perhaps the greatest plausibility, rooted as it is in the American diplomatic documents.
³⁴
But Hawthorn’s other ‘worlds’ are less credible. The first involves an argument about the links between medieval demography and eighteenth-century economic and political development which even the boldest cliometrician would view with suspicion; while his vision of a ‘non-Renaissance’ in art depends almost entirely on questionable assumptions about the dynamics of stylistic change in art.
³⁵
As for his less detailed introductory sketches for a Labour Party renaissance in the 1980s and a Moorish superstate in the twentieth century (in fact, an extension of Guedalla’s essay of 1932), these would not look out of place in a new edition of Squire’s
If
...
³⁶

By themselves, the defects of all these attempts at explicit counterfactual analysis could almost explain the failure of counterfactualism to catch on. Whether by posing implausible questions or by providing implausible answers, counterfactual history has tended to discredit itself. Yet there are clearly other reasons why so few historians have attempted to argue in this way - or, when they have acknowledged the possibility of alternative outcomes, have left the counterfactual implicit, as a kind of subtext. Such veiled counterfactualism has been a striking feature of a great many recent ‘revisionist’ works of history - not altogether surprisingly, in that most revisionists tend to be challenging some form of deterministic interpretation. To take one example, R. F. Foster’s justly acclaimed
Modern Ireland
repeatedly calls into question the nationalist teleology of inevitable independence from ‘English’ rule. Yet at no point does Foster make the implicit alternatives (for instance, continued Irish membership of the Union, perhaps as a result of a successful passage of one of the early Home Rule Bills) explicit.
³⁷
Much the same can be said of John Charmley’s polemical critique of Churchill, which implies that the British empire could have been preserved after 1940 by means of alternative policies such as peace with Hitler, without spelling out how this might have worked.
³⁸
Clearly, something more than the defects of past attempts at counterfactual history has deterred such historians from spelling out the historical alternatives their books imply. A more profound suspicion of counterfactualism is at work - a suspicion which has the deepest of roots in the philosophy of history.

There was nothing inevitable about the triumph of historical determinism. As Herbert Butterfield suggested, the world in preliterate societies probably seemed anything but deterministic. Life was dominated by the effects of natural forces, some rhythmic and predictable (the seasons), others intelligible only with reference to supernatural forces:

Divine agency thus originated as a kind of explanation of last resort. In polytheistic religions, however, this was often merely a matter of giving names to conflicting natural forces. Indeed, the unsatisfactory nature of polytheism prompted the Epicureans’ rejection of any kind of divine agency: perhaps the earliest statement of an anti-determinist philosophy. Lucretius proclaimed the existence of an infinite universe composed of atoms with an essentially random dynamic:

The only remotely deterministic element in Lucretius’ thought was his primitive theory of entropy: ‘Everything is gradually decaying and going aground onto the rocks, worn out by old age.’
⁴¹

It was thus only slowly that the idea developed of an ultimate and purposeful supernatural arbiter. A good illustration of the evolving classical conception of ‘Fortune’ in this role can be found in Polybius’
Rise of the Roman Empire
(written in the second century BC):

Polybius’ suggestion that the ‘vicissitudes’ of Fortune in fact had a purpose - the triumph of Rome - was an important historiographical step towards a more deterministic notion of divine agency. A similar conception can be found in the work of Tacitus, though here it is Rome’s destruction which is the divine objective: ‘Rome’s unparalleled sufferings supplied ample proof that the gods are ... eager for our punishment.‘ For Tacitus, as for Polybius, ‘the outcome’ of ‘the actual course of events‘ was ’often dictated by chance‘; but events ’also had their underlying logic and causes’.
⁴³

An additional superhuman factor which Polybius acknowledged was the Stoic notion of historical cycles, culminating in periodic natural catastrophes:

The same idea of history as a cyclical process can, of course, be found in the Old Testament book of Ecclesiastes: ‘The thing that hath been, it is that which shall be; and that which is done is that which shall be done.’
⁴⁵
However, the divine Plan of the Hebrew God was rather more complex than that of the Graeco-Roman Fortune. In the Old Testament, Yahweh’s purpose unfolds itself in a complex historical narrative: the Creation, the Fall, the election of Israel, the prophets, the Exile and the rise of Rome. To this the early Christians’ New Testament added a revolutionary coda: the Incarnation, the Crucifixion and Resurrection. Jewish and Christian history thus had from an early stage a far more deterministic structure than classical historiography: ‘Not only did God direct the events of the world, but his intervention (and its underlying purpose) was for the early Christians
the only thing that gave any meaning to history
.’
⁴⁶
In the writing of Eusebius (c. AD 300), events and individuals are generally portrayed as either pro-Christian, therefore favoured by God, or anti-Christian, therefore doomed.
⁴⁷