Capital in the Twenty-First Century
by Thomas Piketty (translated by Arthur Goldhammer)
Harvard University Press
Published March, 2014
Near the end of the introduction to Capital in the Twenty-First Century, Thomas Piketty writes: ‘In my mind, this book is as much a work of history as of economics.’ In the commentary that has surrounded the book’s meteoric appearance, this has emerged as a predominant theme. Piketty’s work heralds the return of political economy as the centerpiece of economic analysis and, along with it, the historical perspective such an analysis demands. The mathematical formulae he studied and produced in his professional youth have, in his view, led economics astray, condemning it to serve as an idle abstraction at best and an apology for predatory neoliberal practices at worst.
Capital in the Twenty-First Century begins with a recovery of names that are today more dear to historians than to most economists: Thomas Malthus, Adam Smith, David Ricardo and Karl Marx. Piketty finds valuable principles in these authors’ works – for example, Smith’s principle of scarcity or Marx’s idea that capital accumulation tends toward the infinite. According to Piketty, the error of these thinkers was not in their thinking, as such. Rather, it was that their debates were ‘debates without data’. Reflecting on the work of the American economist Simon Kuznets, Piketty’s most proximate forebear, Piketty writes, somewhat injudiciously:
Malthus, Ricardo, Marx and many others had been talking about inequalities for decades without citing any sources whatsoever or any methods for comparing one era with another … Now, for the first time, objective data were available. Although the information was not perfect, it had the merit of existing.
For Kuznets, data was not the problem. His data allowed him to see a compression of inequality in the period of the world wars that would not have been visible without access to taxation records. His problem was short-termism; his proximity to the wartime period prevented him from seeing how unnatural its shocks were, and that the tendency of the history of capital without such upheaval would be toward greater inequality. This tendency was the consequence of the higher rate of return on capital compared to growth, which is the central ‘contradiction’ discovered and elaborated in Piketty’s work.
The twin pillars of Piketty’s return to political economy are articulated as a pair of twin needs: the need for big data and for the long view. Zeitgeist or conjuncture, take your pick: Capital in the Twenty-First Century is central to it. But it is, of course, not alone. The 2008 financial crisis and the seemingly interminable effort to consign the period it inaugurated to the past has led to a new hearing for a variety of works not simply in political economy, but within an avowedly Marxist tradition. The widespread interest in David Harvey’s work is one example; and it is heartening to see books by Giovanni Arrighi and Robert Brenner, long respected among regular readers of the Verso catalogue, find a larger audience.
More suggestive still is the work of the anthropologist David Graeber, a leading instigator of the Occupy Wall Street movement and the source of the phrase ‘we are the 99%’. As the Occupy movement was getting underway, Graeber published his book Debt: The First 5000 Years (2011). More strictly communitarian than Marxist in its political convictions, Graeber’s book is a bravura performance, melding insights garnered from his extensive anthropological fieldwork in Madagascar and other lands into a new speculative history that displaces the history of money with a history of debt and credit. Graeber targets the myth of barter that has subtended the former history in a manner that truly puts the political in political economy. Yet for Graeber the political derives from the moral; political history is an anthropological history of moral conflict over the status of the community and its obligations to its members.
Though Piketty and Graeber are temperamentally antipodal, the thematic resonance connecting their works is clear enough. New data – economic in the first instance, anthropological in the latter – combined with a longer view results in a rewritten history, one that obeys as yet undiscovered laws that upend the conventional view of capitalism. They also engage in a similar sleight of hand, one that is by no means fatal for their accounts, but is illustrative of questions that attend the appeal to history on a grand scale. Though neither author is a historian, each finds his point of departure in the central gesture of modern historiography: the gesture of revisionism. In both cases, they target a mythic account, one that we can now see is ideological because it did not have access to the right evidence, and suggest a new account grounded in better, more comprehensive data.
At the same time, they are both explicit about the ideological moorings of their own accounts – and I am using ideological here in a non-pejorative sense to describe their political convictions and the way they are more or less explicitly encoded in their works. It is this very explicitness that is at once disarming and the site of the sleights-of-hand. In Piketty’s case, despite the celebration of new data sets, the heart of his analysis is, in fact, a reconfiguration of analytical categories. In other words, his antidote to previous abstractions is not recourse to the concrete, but a new, different set of abstractions.
Piketty is keenly aware of this. Regarding the central concept of his account, he writes: ‘Capital is not an immutable concept: it reflects the state of development and prevailing social relations of each society.’ It is this historical mutability of the concept that allows him to use it in a specific sense. For example, he excludes ‘human capital’ because he seeks to downplay its contemporary importance against the overwhelming force of inherited wealth. ‘The past devours the future’ is the book’s basic admonition. It is therefore ironic that by excluding ‘human capital’ he also excludes the phenomenon of slavery, which is only addressed for purposes of comparative periodisation and not as a factor in the contemporary accumulation of wealth. A comprehensive nineteenth-century account would have proceeded otherwise, but Capital in the Twenty-First Century is a product of its time and its available means. Piketty focuses on total inherited wealth, including land, because his sources allow him to do so. But his sources are themselves only the result of political developments. As he notes with a pride others have remarked upon, the French Revolution did not just beget political equality; it begat tax records.
Capital is just one of Piketty’s key concepts. Regarding the others – income, economic growth rate, the rate of return on capital – he explains that they emerged from his collating of historical data, and that he will try to use them sparingly. He doesn’t; they are the basic grammar of his book. He nevertheless concedes that these
are abstract concepts – theoretical constructs rather than mathematical certainties. Yet I will show that these concepts allow us to analyze the historical reality in interesting ways, provided that we remain clear-eyed and critical about the limited precision with which we can measure such things.
Piketty uses the phrases ‘historical sources’ and ‘historical data’ interchangeably. Attributing Marx’s ‘hasty pronouncements’ to his ‘political fervor’, he suggests that ‘economic theory needs to be rooted in historical sources that are as complete as possible, and in this respect Marx did not use all the sources available to him’. Six pages later he reiterates:
If the question of inequality is again to become central, we must begin by gathering as extensive as possible a set of historical data for the purpose of understanding past and present trends. For it is by patiently establishing facts and patterns and then comparing different countries that we can hope to identify the mechanisms at work and gain a clearer idea of the future.
There is a striking equivocation between these two passages in which ‘sources’ and ‘data’ collapse into one another, as equally given, found objects. (It is worth noting that the French word translated as ‘data’ is données, which also suggests simply ‘what’s given’ or ‘the givens’. The translation is perfectly correct, but the term is more banal in French, so its alignment with the ‘big data’ meme in English is in some respects fortuitous.) But if the concepts Piketty uses are admittedly mutable, then what constitutes the objects united by those concepts must be a mutable category too. In other words, for Piketty, data is a matter of technology – from revolutionary tax records, to number crunching hard drives – that is not historical precisely insofar as it is merely cumulative.
But what makes inequality visible is not simply that new evidence comes into view; it is the technological construction of a frame for viewing that evidence. In this regard, there is no essential formal difference between Piketty’s rate of return on capital and Marx’s mode of production. In both cases, you have a historical concept that emerged from a confined set of available data and political convictions, which is then used to render broader fields of data intelligible.
There is, however, a key difference between Piketty and Marx, and it is perhaps most clearly expressed in a recent review of Capital in the Twenty-First Century in the London Review of Books by Benjamin Kunkel. Notwithstanding the passage from Piketty’s book in which he speaks of a ‘hope to identify the mechanisms at work’ in the history of political economy, Kunkel’s critique of Piketty amounts to the observation that, in the end, his work is purely descriptive rather than explanatory. It demonstrates the historical fact of the rate of return on capital’s tendency to outpace growth, but it doesn’t tell us why.
At a glance, this might seem like the familiar complaint of a Marxist who asks for too much, betraying a need that no secular science can satisfy. Kunkel’s point is not so crass. His concern is that Piketty’s history of political economy is a history that leaves the politics out. The world wars intervene as contingent events, akin to natural disasters. Political pressure from organised labor is virtually discounted as a relevant variable. In a word, Kunkel makes clear the presumptive naturalism that underlies Piketty’s thoroughly historical and ostensibly political account. Perversely, Piketty’s problem is the same as Marx’s: the history of political economy becomes so naturalised that the only political solutions he offers seem plainly utopian. Here the relation between the antecedent and the consequent is almost syllogistic; once you establish a natural law, only a miracle can rewrite it.
Piketty has very good reasons for his restraint, but those reasons are as much political as they are disciplinary. Much has been made of his talk of being inoculated against Marxism or any kind of revolutionary radicalism. Nevertheless, many have found his calls for a global tax laughable. But such a call, however admittedly utopian, is inextricable from Piketty’s method and the way he conceives the relation between sources and data. For his ‘data’ are all state records, tax records mainly, and the various policy documents such records inform. This places the state at the heart of his analysis as the primary political mechanism for regulating the ‘natural’ tendencies of capital.
Piketty’s view is consistent with a brand of republicanism that is dear to many in France (and elsewhere) and which seeks to refashion the state as a site of civic engagement rather than a tool of privileged interest. The breadth of sympathy for this view no doubt partially accounts for the warm reception of Piketty’s work. But my point is that the contrast between Piketty’s view and the broadly Marxian one is not so much a question of ontologies or philosophies, much less a question of evidence. It is a question of politics. It is a question of how one construes, or where one locates, the political activity that lends political history its intelligibility.
Piketty’s ‘data’ is unproblematic as a historical source because state tax records express the very history that is his concern. To be clear, there is nothing wrong in this conflation. The division of labor dictates the parameters of his work, which is all well and good. But it is worth pausing to consider the seemingly trivial notion that the way he conceived his hypothesis – inequality as a function of income / capital ratios – told him where to look. And as any historian knows, you tend to find what you are looking for. Evidence can always be accommodated into an extant heuristic or interpretive frame.
But by the same token, historians know that the real historical work takes place when an encounter with evidence forces a rethinking of the frame. This is what makes it history, and not sociology or simply ideology. Bertrand Russell used to remark on a similar experience in philosophy; sometimes the logical object resists, just as the empirical object does. It resists in a way that forces a rethinking of the logical sequence, or that leads the sequence into an unforeseen conclusion. The same happens with the historical object. The challenge is that, unlike the empirical object or arguably the logical object, the historical object is never encountered immediately – it is always mediated by the fact that it is past, not present. The immediacy of the empirical object in the present has a kind of force that can alter our comportment. So if we are all Kantians to the degree that we concede that what we see is shaped by how we see it, we can admit that certain things can emerge with such force as to change our way of thinking. But the problem is more dramatic and intractable when it concerns our comportment toward the past, because that comportment determines, in an almost fundamental way, what objects will be allowed into view in the first place.
Cards on the table: I am totally sympathetic to Piketty’s work, even if it has the effect of empirically demonstrating something most of us already know – namely, that economic inequality is not a natural affliction but a historical consequence. And it is interesting to note how quickly his work has become orthodoxy for an ecumenical left. In his recent Alan Saunders Memorial Lecture devoted to the history of liberty, the philosopher Philip Pettit casually dropped an ‘as we now know from Piketty …’ into his talk, endorsing the view that inequality is an economic phenomenon that demands a political solution. What I am ambivalent about here is the rhetorical significance of this ‘we now know’. In other words, it is striking how quickly the political imperative can become the empirical told-you-so. And whether one is sympathetic to the imperative is irrelevant. It seems that one duty of the historian is to remain vigilant in maintaining that the relationship between the political imperative and the empirical told-you-so is fraught at best.
In certain respects, Piketty’s work resonates with broader calls for a return to the longue durée perspective, be it under the guise of ‘deep history’ or ‘big history’, or to grant the Anthropocene its status as a period that is as historical as it is geological. The difference is that much of this work seeks to incorporate natural history and the data of the natural sciences into the historian’s field of vision, whereas Piketty is content to remain within the parameters of human history and activity. One consequence of this is that environmental externalities do not figure in his picture of total wealth as a kind of global negative asset, a lacuna which curiously echoes the omission of human capital as a pertinent category in his analysis. To be sure, the incorporation of the naturalist view into history writing remains a contentious issue. It raises difficult questions about how historians handle evidence that is not textual in any kind of literal sense, and what levels of intelligibility or causality the historian can plausibly seek to correlate. In this regard, the narrow purview of Piketty’s work might be a virtue, not least because it allows us to consider how Capital in the Twenty-First Century and the peculiarities of its reception compare with similar bestsellers in the past.
Let’s just consider two books that retain a certain measure of renown: Francis Fukuyama’s The End of History and the Last Man (1992) and Paul Kennedy’s The Rise and Fall of the Great Powers (1987). The books shared a similar fate: they were met with an initial fanfare, but were quickly viewed as mistaken or obsolete. A lot of the derision was actually unfair in Fukuyama’s case. You only had to read a few pages – of either the book, or the original National Review article on which it was based – to appreciate the provocation of his title. His point was not that time had stopped, or that there would be no more conflict. But he did make an argument for the end of ideological conflict with the end of the Cold War, the claim being that a broadly liberal capitalism would face no further viable ideological challenges. Many saw the emergence of Samuel Huntington’s ‘clash of civilisations’ paradigm in the mid-1990s as the ruin of Fukuyama’s thesis, and the book fared poorly as a result. But more recently The End of History and the Last Man has acquired a kind of second life, not as a sound work of empirical history – there’s hardly an empirical page in it – but as an ideological bellwether that was perhaps wiser than it knew. This reading suggests that a world divided along ethnic or religious lines is, in fact, perfectly compatible with a neoliberal model of economic organisation in which inequality persists and is deemed not so much insoluble as irrelevant, or simply unfortunate.
I admit it is a stretch to compare Capital in the Twenty-First Century and The End of History and the Last Man, especially when the latter is so ideologically brazen where the former is so empirically rigorous. But my aim is to highlight a similarity in the way that each uses a scalar appeal to a historical tendency in order to buttress a particular view of political justice. It is also to suggest that such big history bestsellers can have surprising ramifications as they cease to be historical diagnostics and become, inevitably, historical objects themselves whose prescience demands re-evaluation.
This is even more pronounced in the case of Kennedy’s Rise and Fall of the Great Powers. This book was published in a moment of full Reaganism in the late-1980s. Steeped in a wealth of archival research and data mining, it argued that the modern history of great power relations formed a clear pattern, in which rising powers would enter a phase of inevitable decline once they embarked on a period of ‘imperial overstretch’. In such conditions, too much money would be spent on security and the cost would be reflected in impoverished infrastructure and conditions at home. Kennedy’s book made a splash because it suggested American imperium was a finite thing. Newsweek magazine remarked: ‘Kennedy gives epic meaning to the nation’s relative economic and industrial decline.’ But Kennedy looked like a foolish Cassandra a few years later when the Soviet Union collapsed and American power looked stronger than ever, the sole pole in a unipolar world.
The reputation of Kennedy’s long-view book suffered because it was seen from a short-term perspective. But if one takes the long view on the book itself, his conclusions actually – remarkably – look more sound today than they did when they were published. The nature of Kennedy’s work has changed in several ways since he wrote this book, with an interesting micro-military history of the Second World War and a less interesting apology for the UN appearing in the last decade. But Rise and Fall of the Great Powers was the culmination of a series of deeply researched works of diplomatic history. The Rise and Fall of British Naval Mastery (1976) and The Rise of the Anglo-German Antagonism, 1860-1914 (1980) are real masterpieces that had political economy at the heart of their analyses. Like Piketty, and like Fukuyama, Kennedy proceeded from a particular vision of the state and its place in history. In other words, his work was intimately bound up with the question of how politics works and what makes political history a coherent domain.
One reason for bringing up Kennedy and Fukuyama is to offer the uncontroversial suggestion that the virtues of taking the historical long view for shaping political action are not new, but a periodic if not endemic feature of modern historical writing. In this, the history of historiography can, as ever, show that there is very rarely anything new under the sun. But my main point is to use these previous cases to show, not simply that one can identify and isolate the abstractions at work in historical writing on a grand scale, but that one can then submit those very abstractions themselves to a kind of historical analysis. And in this regard we can strive to take a historical view of Piketty in the present in the same way we more readily and easily take a historical perspective on Kennedy in the past.
Many historians of widely divergent ideological persuasions have regarded the political use of abstraction as a historical object before; this might be the only thing Karl Marx and J. G. A. Pocock have in common. But I think that one of the main virtues, unintended or not, of the injunction to ‘go large’ is that it forces the abstractions of history into explicitness. And once those abstractions are explicit they become legible as figures of political investment and interpretation, which themselves have a history. This is why Kennedy’s and Fukuyama’s ideas can be as historically ramified as those of more august philosophers. Historians often talk of evidence, but the juridical or scientific frameworks that the metaphor calls to mind obscure the fact that historical evidence is never given, never determined by the trial of a particular crime or the terms of an experimental hypothesis. It requires abstraction to bring it into view, to create it as evidence.
So against the hopes of many, it seems to me that the virtue of taking the long view is not to forge a unified humanity or coerce a set of shared interests around unassailable empirical evidence. It is to make explicit the assumptions about value and purpose that inform all historical narratives, not as an opportunity to smooth them over in a common cause, but rather to make audible the articulation of genuine political disagreement. Whether this was Piketty’s intent is not entirely clear; it is nevertheless one of his book’s most salutary consequences.
This article was first presented on 11 July 2014 at the symposium ‘The Aesthetics of Scale: Emerging Trends in Historical Narrative’, hosted by the Centre for the History of European Discourses at the University of Queensland in conjunction with the Australian Historical Association.