The blurbs on the back hail Escape from Rome as “innovative and provocative”, “challenging and counter-intuitive”, and “iconoclastic”, and for once it is hard to disagree. Walter Scheidel argues that innovation, economic growth and human thriving are best promoted by large and highly centralized imperial states. Throughout history, as Scheidel shows, the most dramatic rates of technological innovation and economic growth have been achieved by authoritarian regimes: Ptolemaic Alexandria in the third century BC, early Abbasid Baghdad, the Soviet Union from the first Five Year Plan to Sputnik, the People’s Republic of China since Deng Xiaoping. Scheidel concludes that centralized and monopolistic policy-making tends to lead to high levels of resource efficiency and economic adaptability, while dispersed authority and market competition create wasteful inefficiencies and act as a brake on innovation. only thanks to its hegemonic globe-spanning empire was nineteenth-century Britain able to …
Nah, just kidding. In reality, Scheidel doesn’t argue any of those things; and for what it’s worth, I doubt the argument sketched out in the previous paragraph would stand up to five minutes’ serious interrogation. Still, it makes for a nice counterfactual thought experiment. Would a genuinely iconoclastic account of the origins of modernity have made it into the Princeton Economic History of the Western World series, or would Scheidel have been sent packing off to Verso? And would Francis Fukuyama have agreed to write a puff for the book? (I somehow doubt it.)
Along with his Stanford colleagues Ian Morris and Josiah Ober, Scheidel has over the past two decades been one of the most prominent advocates of a social-scientific approach to the history of the Greco-Roman world. No one has done more to illuminate the political economy of the Roman Empire by means of heroic quantification, model-building, and rigorous comparison with other premodern agrarian states (particularly China). His strengths are fully on display in the second chapter of Escape from Rome, a brilliant and compelling account of the rise of Rome to dominance in the Mediterranean and Western Europe. But the chief focus of this book is on the post-Roman world, and in particular the causes of the so-called “Great Divergence” – why it was that nineteenth-century Western Europe ended up pulling away so dramatically from the rest of the world in terms of per capita GDP, energy capture, and social development.
In Scheidel’s view, the answer lies in the “competitive fragmentation of power” which has characterized Europe (particularly northwestern Europe) since the early Middle Ages. After the fall of the Roman Empire, multiple small states emerged in Western Europe, which then spent the best part of a millennium and a half squabbling incessantly with one another. Within these small polities, authority and agency were widely distributed between different social groups: “weak kings, powerful lords, belligerent knights, the pope and his bishops and abbots, and autonomous capitalists all controlled different levers of social power”. It was this unruly polycentrism, both within and between states, which sent Western Europe bowling along the track that ended in the internal combustion engine, universal literacy and the maxim gun – or, as Scheidel puts it, “intensifying interstate competition and concurrent intrastate bargaining favoured certain performance-enhancing strategies”. (An easy read this isn”t.) That is why the breakthrough to modernity happened in Western Europe, not in (say) China or India; and why it was the plucky little kingdoms and republics of early modern Europe, not the monolithic European superstate of Roman times, that finally got things moving.
As will be clear, there is nothing here to frighten the horses. In its broad outlines, as Scheidel acknowledges, this kind of argument goes back at least to David Hume (“the divisions into small states are favourable to learning, by stopping the progress of authority as well as that of power”), if not to Montesquieu. Competition drives innovation, excessive state authority tends to curb economic growth: the basic argument of Escape from Rome is about as iconoclastic and counter-intuitive as the average issue of The Economist. The originality of the book lies not in its conclusions, but in its method. Scheidel wants to show that “The Great Escape” – the staggering transformations of human societies over the past two centuries – could only have begun in the special geographical, institutional and cultural circumstances of post-Roman Europe. In particular – and this is the thesis that gives the book its title – he wants to show that if another hegemonic empire had re-emerged in post-Roman Europe (a New Rome of some description), then the paths to modernity would necessarily have been blocked.
Scheidel’s method for demonstrating the truth of these two claims is a series of extended counterfactual thought experiments. Was it in fact even theoretically possible that a New Rome could have arisen in early modern Europe? If it had, could it have fostered the kind of competitive polycentrism that was, he argues, necessary for the world-transformative developments of the past 200 years? And could the institutional pre-conditions of modernity have emerged in any part of the world other than northwestern Europe?
The use of counterfactual scenarios is not original to Scheidel. Since the 1990s, beginning with Niall Ferguson’s Virtual History: Alternatives and counterfactuals (1997), a flourishing sub-genre of “what-if” history has emerged. As Richard J. Evans pointed out in his elegant 2014 take-down of the genre (Altered Pasts: Counterfactuals in history), a suspiciously large proportion of this work consists either of removing putatively key individuals at critical moments (James II executes Isaac Newton; Lenin misses his train from Zurich) or of reversing the outcome of particular battles (what if Napoleon had won at Waterloo?). Evans darkly suspects that the whole genre is a racket, designed to promote a great-man theory of history by re-enthroning individual agency as the chief motor of historical change.
Scheidel is far more sophisticated than this: he has little truck with historical parlour games of the what-if-Hitler-had-fallen-under-a-bus variety (though the old chestnut “what if Henry VIII had stayed Catholic” does rear its head at one point). The counterfactual scenarios that underpin the argument of Escape from Rome largely play out on a much more impersonal level: a putative east Roman reconquest of the western empire from the Visigoths and Franks in the mid-sixth century, successful Ottoman expansion into Latin Europe, and so forth. I know of no other major work of history which makes such extensive and critical use of modified pasts as a central part of its reasoning. Despite the banality of Scheidel’s historical conclusions, then, Escape from Rome deserves to be taken very seriously indeed. Does he vindicate the use of counterfactuals in history? Or is there something dodgy about the whole enterprise?
The case in favour of counterfactuals runs something like this. Historians are concerned with particular things which actually occurred in the past, and would like to know why they occurred. The trouble is that the past only happened (at most) once. As a result, when a historian claims to have identified the necessary and sufficient conditions for a particular event, it’s very tricky to know if she’s right. It would be helpful to be able to do a complete re-run of the history of Europe from 1400 to 1900 with, say, the Reformed churches or double-entry bookkeeping taken out, but that’s not so easy. In order to test the validity of any causal claim in history, we therefore have to resort to counterfactuals: we imagine taking out the Reformed churches, and infer what might have been different.
As Michael Oakeshott pointed out sixty-five years ago, in his delightfully annoying “On the activity of being an historian”, most historians do in fact engage surreptitiously in this kind of counterfactual reasoning all the time. Take a statement like “The Pope’s intervention changed the course of events”. Strictly speaking, the Pope’s intervention didn’t change the course of events; it was the course of events. The alleged course of events which was “changed” by the Pope’s intervention has no historical reality, and is really an implicit counterfactual. If a historian restricts herself to things that actually happened, she may (perhaps) be able to identify a set of converging choices and actions which make the outbreak of a war, the fall of an empire, or the onset of modernity minimally intelligible; but if she wants to show that the event was necessary or even likely, that can only be done through explicit or implicit counterfactual reasoning. Without counterfactuals, you can’t usefully explain anything in history at all.
This all sounds reasonable enough; but in practice things are a bit more complicated. Take one of the most famous of all historical counterfactuals: what if the Arabs had beaten the Franks at the battle of Tours in 732? In the Decline and Fall, Gibbon had a great deal of fun extrapolating from this non-event: “Perhaps the interpretation of the Koran would now be taught in the schools of Oxford, and her pulpits might demonstrate to a circumcised people the sanctity and truth of the revelation of Mahomet”. Ernest Gellner went one better: “No doubt we should all be admiring Ibn Weber’s The Kharejite Ethic and the Spirit of Capitalism which would conclusively demonstrate how the modern rational spirit and its ｅxpression in business and bureaucratic organisation could only have arisen in consequence of the sixteenth-century neo-Kharejite puritanism in northern Europe”. (That sentence, by the way, is itself a strong argument in favour of the heuristic usefulness of historical counterfactuals.)
Scheidel pours cold water on all this. Although he grudgingly accepts the counterfactual possibility of “a sustained Arab presence coupled with some degree of Islamization” in Northern Europe, he denies that this could conceivably have led to “the incorporation of large parts of Europe into a unified Arab empire”. Early Arab conquest society was divided and unstable, and the Umayyad caliphs at far-off Damascus never exercised any meaningful control over their western periphery: a single caliphate stretching from the Tigris to the Thames was therefore “never even on the table”. That may be right: but why are we only allowed to ask whether a unitary Umayyad or Abbasid empire was a plausible outcome of an Arab victory at Tours? A counterfactual series of Arab conquests in Francia in the 730s and 740s could perfectly well have led to an independent Arab–Berber splinter empire, with its capital at Paris or Aachen rather than Damascus or Baghdad, with vast and unknowable consequences for the history of the later Middle Ages and beyond.
Scheidel’s reply would be that he tries to follow a “minimal-rewrite rule”, which involves the “least amount of tweaking of actual history”. He therefore restricts himself to short-term counterfactuals, because “the farther we project ahead of actual history, the less we are able to control the thought experiment”. That is to say, once you have changed one variable, you aren’t allowed to think through what other knock-on variables might also have changed as a consequence: you can have an Arab victory at Tours, but you can’t have tenuous second-order effects like a resulting breakaway Berber empire in northwest Europe. But this is cheating. If you circumscribe the possible consequences of your counterfactual scenario in this way, then of course you end up concluding that what actually happened was overwhelmingly likely to happen. “Minimal rewrites”, he says at one point, “could not possibly lead to significantly different counterfactual developments.” That’s the whole problem.
Worse still, I ended up with the uncomfortable impression that Scheidel was making up the rules of “minimal rewrites” as he went along. A counterfactual Carolingian reunification of Europe? Not even worth contemplating, because of “the time-honoured [Frankish] practice of dividing the realm among a king’s sons”, which acted “as a built-in – almost homeostatic – constraint on durable state formation”. one can’t help suspecting that this might have been got around somehow or other. A Europe-wide Catholic Spanish Empire under Philip II? Ok, Philip clearly could have conquered England in 1588; but Scheidel then arbitrarily (and against his own self-imposed rules) drops in a second-order counterfactual to close off one set of possible consequences: “Spanish heavy-handedness in England might well have galvanised resistance elsewhere, discouraging Dutch proponents of peace and emboldening Germany’s Protestant princes.” Napoleon? No chance of a lasting empire in Europe, because a counterfactually less ambitious Napoleon “would not have created an empire of ancient Rome’s demographic pre-eminence, and would have kept Britain’s position (largely) intact. In this case, modern economic growth and industrialization could have proceeded unabated. A more ambitious Napoleon – the real-life version – by contrast, was bound to undermine his successes through unsustainable overreach”. That is to say, no Napoleon could have got the balance right because people like Napoleon are psychologically prone not to get the balance right. Finally, in order to argue that China could not conceivably have colonized the New World, Scheidel throws caution to the winds and sets up a counterfactual scenario which involves flipping the whole of Greater Afroeurasia around its axis at 63° E so that China faces the Atlantic and Europe the Pacific: in what sense this constitutes a “minimal rewrite” is, I have to say, completely beyond me.
My point is not that an imperial reunification of Europe would necessarily (or even probably) have happened if the Arabs had won at Tours, or if Louis the Pious had bumped off all his sons apart from Lothair I, or if William of Orange’s horse had tripped over a molehill in 1687 before he had had a chance to sail for England. My point is that I have not the faintest idea what the long-term consequences of any of these “altered pasts” would have been, and nor does Scheidel. As a wise man once put it, “time and chance happeneth to them all”. The moment you change one significant variable – whether it be the east-west orientation of Afroeurasia or the succession practices of Frankish monarchs – the course of subsequent events rapidly becomes completely uncontrollable. Scheidel’s answer is to close down long-term contingencies, focus only on the short term, and avoid second-order counter—factuals (if x changed, might y also have changed as a result?). The inevitable result of this “minimal-rewrite” rule is that the path of history is left looking more or less unchanged and unchangeable, because the rule itself prevents any meaningful exploration of alternative paths.
This is a fiercely intelligent, closely argued book. Yet in the end, what we are left with is a new Whig Interpretation of History quite as self-satisfied and triumphalist as Macaulay’s History of England from the Accession of James II. Things couldn’t have worked out differently from how they did in fact work out, and it is all for the best in this best – and only – of all possible worlds. The history of Europe since the fall of Rome is a series of inevitable steps towards (as Scheidel summarizes it) “Denmark”: liberty, representative institutions, Schumpeterian growth. Rome was always going to fall more or less when and how it did; no subsequent Rome could ever have emerged in Europe, and a very good thing too. The poor old Chinese, locked into their stagnant cycles of one empire after another, were never going to be at the cutting edge of history: too bad for them. Early in Escape from Rome, Scheidel notes with puzzlement that historians, on the whole, are not interested in the kinds of Very Big Questions he wants to answer (his capitals). There’s nothing wrong with very big questions; but I fear this isn’t the way to answer them.
Peter Thonemann teaches Greek and Roman history at Wadham College, Oxford