Superstates: Empires of the 21st Century 
by Alasdair Roberts.
Polity, 235 pp., £17.99, December 2022, 978 1 5095 4448 6
Show More
Show More

Empires​ are supposed to be a thing of the past, yet in some ways the empires we knew are still with us. The great powers of the present were the great continental empires of the 18th and 19th centuries. The borders of Russia today are similar to those of the Russian empire in the 1750s. The territory of modern China largely resembles that of the Qing empire in 1760, the main difference being the loss of Mongolia. The United States approached its current mainland form in the 1880s. This may be an age of states, but some of them are so big that global politics is for the most part still a game for subcontinental powers.

Size, of course, is not a virtue in itself, and very large states have their own problems. Plato thought the ideal state would have a population of just 5040: ‘numbers enough for war and peace, and for all contracts and dealings’. Aristotle believed large states to be ‘almost incapable of constitutional government’, and that large populations carried the risk of foreigners blending in and acquiring the rights of citizens. The biggest states tend to have what George Kennan once called the ‘hubris of inordinate size’: an affliction that raises the question of ‘whether “bigness” in a body politic is not an evil in itself’. At the beginning of the 20th century there was a belief that smaller states would inevitably be consumed by the existing empires. Instead, small states survived and many flourished. One reason may be that advances in weapons technology increased the advantage of defence. International relations scholars invoke theories of great power settlements and ‘shatter belts’, while orthodox historians focus on 20th-century nationalisms. Whatever their orientation, historians have tended to be less than confident about whether the world is going in the direction of many small states or a few very large ones.

At the University of California in the late 1970s, the political scientist Rein Taagepera published a series of papers on the size and longevity of empires. Estimating the acreage of historical states is an inexact science; Taagepera had to use a special tool – a planimeter – to measure blocs on old atlases. But he found that the geographical size of states was growing. When you aggregate the extent of the three largest empires in the world over time, you find that the amount of territory under their control trends upwards, but that the gradient isn’t smooth. Taagepera argued that history was divisible into three distinct phases, separated by sudden increases in the maximum size of states. Each phase marked some kind of technological breakthrough. The first began around 2800 BCE with the emergence of the first cities in southern Mesopotamia. The second happened around 600 BCE in Persia with a ‘power delegation breakthrough’: the system of satrapies connected up by relay stations. The third phase started with the rapid expansion of the European empires. Taagepera was hazy on what caused this final stage – it’s a big question – and described it as a ‘communication speed revolution’ somehow connected to the industrial revolution. Since the telegraph wasn’t invented until the 1830s his dating seems somewhat questionable. Still, the pattern was there. After 2800 bcE there was never again a time without an empire of some sort, and after 600 bcE one or more of them always controlled an area of at least 2.5 million square kilometres. After 1600 CE the figure increased to at least ten million square kilometres: about the size of the US or China today.

Taagepera’s work challenged Eurocentric global histories. Rome barely featured on his lists of the world’s largest empires because for most of its existence Han China surpassed it in size and population. ‘The present Muslim world’ – Mesopotamia, Persia, Egypt and Anatolia – had been at the centre of the largest empire on earth for more than half of the last five thousand years. The Maurya empire, which covered most of the Indian subcontinent between 320 and 185 BCE, had been unjustly neglected by Western historians. And Taagepera had much to say about the outsized focus on Alexander the Great. Alexander’s invasion of Persia was culturally significant – not least for its wrecking – but his empire wasn’t greatly different in either size or duration to the Achaemenid empire that already existed on the same ground. It was as though ‘a historian with narrowly Germanic bias treated all of the Roman empire as a mere introductory footnote to some Germanic soldier-emperor of the Roman decay period.’ Taagepera’s semi-logarithmic graphs were perhaps too neat, but if the size of empires increases in great bursts does it follow that there will be another sudden expansion? Will the world eventually be consolidated into just a few states, and finally a world state?

According to Alasdair Roberts, the trend has been happening before our eyes. By 2050, almost 40 per cent of the world’s population will live in just four polities: India, China, the US and the EU. (By my calculations it’s already over forty per cent – India and China alone currently account for around 35 per cent of the world population.) In 19th-century Europe the ratio between the population of the greater and lesser states was about ten to one. Today the ratio between the population of India or China and the average small member of the United Nations is closer to forty or fifty to one.

Roberts acknowledges that there have always been empires that take in very large territories and lots of people. But he argues that the modern giants are different. Empires had undefined frontiers and were regularly invaded: barbarians were always showing up at the gates. And even when they weren’t, there were rebellious provinces or natural disasters that were hard to deal with at great distances. Modern states, by contrast, are obsessed with demarcating their borders, and have sophisticated tools of mass control at their disposal. The largest states in the world have the scale of empires but, Roberts suggests, they are a new breed. They are ‘superstates’, distinct from their neighbours ‘by expanse of territory, number and diversity of people, and social and economic complexity’. They aren’t quite empires and they aren’t all superpowers. Instead they represent a more recent development in a linear history: ‘The age of empires gave way to the age of states and next to the age of superstates.’

If we accept Roberts’s criteria – a superstate is defined by the extent of its territory, population and complexity – it isn’t always clear why some qualify and others don’t. It’s true that India, China and the US are structurally different from large nation-states such as Turkey and Egypt. But in other cases the difference isn’t so clear cut. Brazil is territorially very large, with a population of more than two hundred million. Current projections suggest that Nigeria will have a larger population than the US in the next thirty years. Indonesia, as usual, is overlooked. The US may by some definitions be more economically complex than Brazil – but socially? Is India socially more complex than Pakistan or Bangladesh? And even if it is, what difference does it make to the people who live there? The vast majority of EU citizens never leave their home state to live in another of the union’s member states. The majority of Americans live in the state in which they were born.

The main problem with the four superstate model is what to do about Russia. Roberts excludes what is territorially the biggest state in the world from his quadrumvirate because its population has collapsed and its economy is much smaller than that of the US, China or the EU. But India’s economy, too, is dwarfed by the others. One could reasonably include India and exclude Russia only if one were sure that India’s economy will grow close to the size of the others in the future, something that is by no means certain. Is Russia then a depopulated superstate? And if the superstate is a recent historical development, was the Soviet Union the holotype? Roberts doesn’t give a clear answer. At one point he suggests that the Soviet Union’s collapse marked the end of the age of empires and the decisive victory of Eastern European nationalism. At another he calls it a ‘failed superstate’. He puts the question aside by saying that the collapse of the Soviet Union may be a reminder that superstates, like empires, are fragile.

Classifications of this sort will always be contentious, not least because most states believe themselves to be sui generis. China is either the best example of the fusion of imperial sprawl and the modern surveillance state, or it transcends both. Reviled as an empire by American Cold Warriors (one Nixon aide described it as ‘the darkest, most mysterious part of the communist empire’), it is also described by its own leaders in imperial terms as the heir of millennia of Chinese civilisation. Is it the clearest example of the contemporary superstate? In China from Empire to Nation-State (2014), Wang Hui asked whether China was ‘not a true modern state, but an empire … yoked together by the force of its elite culture’. He argued that China came to define itself as a modern nation-state partly as a means of self-affirmation, in reaction to Western incursions before the revolution. But it also seeks to recover a national identity that stretches as far back as the Tang or even Han dynasties. China may be more unitary than India, but the question of whether it takes a residual imperial form will get a very different answer in Beijing from the one it would get in Baotou, Lhasa or Ürümqi. Roberts gives no historical account of the emergence of the contemporary Chinese state. But he does say the Chinese government under Xi is looking to apply 21st-century political management to avoid a Qing-style collapse. ‘China may show how a carefully designed authoritarian regime, augmented with new technologies of surveillance and control, can hold a superstate together for a very long time.’

As a candidate superstate, post-1947 India makes a good deal of sense. B.R. Ambedkar conceived of the new India as a dual polity, both union and state: an ‘indestructible’ federation without the right of secession. Perhaps India was made into a superstate by Nehru’s continuation of British means of imperial control, Indira Gandhi’s concentration of executive power, and the moderating influence of the coalition government period which began in the 1980s and ended in 2014. Roberts quotes Bal Krishna Sharma on the centrifugalities of Indian political life, but says these have now been largely solved with secessionist movements in Nagaland, Punjab, West Bengal and Kashmir outflanked or on the back foot. And then came the rise of Narendra Modi. The BJP victory in 2014 was a reaction to Congress’s dominance and corruption, but it has been followed by a thorough rejection of pluralist traditions in favour of Madhav Sadashiv Golwalkar’s vision of minorities ‘wholly subordinated to the Hindu Nation’. The Modi order is a return to the primacy of central authority achieved by the application of ‘Hindu superglue’. Along with virulent Hindu nationalism, the BJP represents the coming of a new elite. Many of the old industrial dynasties have been swept aside and in their place are the Adanis and Ambanis, the Mittals, Agarwals and Hindujas. During this oligarchic transition a Modi figure has proved useful. Both he and the new elite caste would probably like the idea of India as a ‘superstate’.

The EU has the dual distinction of being the only polity that is regularly referred to as a superstate and the only one that arguably shouldn’t be considered a state at all. Its economic power in the global system is undeniable, but in every other sense its power is much less certain. The EU’s formal structure, which seems to embody both postmodern technocratic management and premodern oligarchy, evades comparison. José Manuel Barroso once called it a ‘non-imperial empire’ (channelling the doyen of liberal international relations, G. John Ikenberry, who has used a similar phrase about the US). Roberts tends to see the EU more as a single highly decentralised state than a confederation. He justifies this on the basis that the ‘European regime as a whole’ exercises state-like control over European territory. The incoherence of the EU response to the financial crisis and the near permanent sense of disarray in EU institutions come in for criticism, and he offers some truly faint praise for EU leaders, who have become ‘more sophisticated in navigating the shoals of treaty reform’. Tom Nairn argued that the EU suffers from a condition that also characterised the pre-First World War multinational empires: uneven development generates subtle forms of nationalist resistance which frustrate supranational designs. The EU has a fearsome border control agency, Frontex, but no central security apparatus. Can you have a superstate without an army, or rather an army of one’s own? The only centralised pan-European army is that of the United States, which has around a hundred thousand military personnel deployed across Europe.

The US, for now at least, remains the only superpower. That it is also an amalgam of subnations is an old argument. Even American historians determined to ignore ubiquitous racial oppression and the dispossession of Indigenous Americans see a superficial geographical division of the country into regions: New England, the Midwest, the Great Plains and so on. Roberts argues that a distinct American national identity emerged only in the 1930s in reaction to the Depression and industrialisation. The New Deal entailed an expansion of central government and increases in federal social spending. The Second World War saw an expansion in American military power unprecedented in world history. Both developments revealed the concentration of power in the US presidency. The geographical division of the country remains an operative reality for alchemical electioneers, but talk of geographical divisions also serves as a convenient distraction from class divisions. Both Republican and Democratic administrations have talked of small government while increasing federal spending and the number of federal employees. But since the late 1970s they have succeeded in transferring trillions of dollars from the poor to the rich. One problem with calling the continental US a superstate is that it deliberately erases US overseas possessions – Guam, Puerto Rico etc. Another is that it distracts from the fact that so much of the energy of the US state apparatus is involved in its global posture.

None of the 21st-century empires is ruled by a hereditary emperor, but both India and the US have dynastic tendencies. If the superstates are empires in disguise, they are beset by the same sorts of dilemma. Too much central control leads to regional ire, too little means vulnerability to natural disasters and other mishaps – pandemics, for example. Communications technology has had an effect here. It no longer takes months for messages from the throne to reach the frontier. Many large states have independently come to the conclusion that governing a large, complex country necessitates autocratic leadership. The alternative is usually more powerful councils and courts, which bring with them the problem of ritualised bureaucracy. Roberts stresses that empires need to justify themselves with a noble creed, both to propagandise to conquered peoples and to bolster the self-delusion of elite cadres. It’s easier to assert yourself if you’re convinced you’re bringing either civilisation or salvation. But how zealous to be, and how tolerant of dissent? For Roberts, the dilemmas of empires and superstates are the same. Superstates exert tighter control than empires did, because technology allows it and because more is expected of them. And they animate their actions with stories of anti-imperialism, the virtue of the civilisational state and liberal democratic ideals.

Empires are sometimes said to be ephemeral things, subject to cyclical rises and falls. Even if it’s true that the average lifespan of an empire is just three or four generations, what does that tell us? The Han and Olmec empires survived at scale for centuries before declining. The occasional Napoleon can still build and lose an empire in a lifetime. The means of control afforded by mass surveillance certainly seem to work in favour of contemporary empires, but they are recent. Things appear robust until they don’t. Roberts is sceptical that his four superstates will all exist in their present form in a hundred years’ time. His advice for leaders engaged in the ‘relentless exercise of improvisation for the sake of survival’ is to avoid winner-takes-all institutions and rigid constitutionalism so as to allow for ‘tacking and jibing as winds shift’. He also advises them to transfer as much power as possible to technocrats, and not to pay too much attention to fancy ideas of transparency. But do we really need an apparatchik cookbook of this kind?

All today’s superstates exhibit a resurgence of something like nationalism. China has Xi Jinping thought, India has Hindutva, US presidents will make America great again and the EU, according to Josep Borrell, is a garden surrounded by perilous jungle. Nations with imperial ambitions do more than shape the internal configurations of giant composite polities. The war in Ukraine marks the present moment as one of both nationalist competition and a contest between the expansive world power of the US and the counter-assertion by Russia that it isn’t out of the game. The US itself is in a full moral panic about China, which now stretches in some US states to proposed bans on Chinese nationals owning land. The threat of inter-superstate violence no longer seems quite as latent as it did. Superstates or no, the five largest economies in the world are all either currently involved in a war, rearming, or making preparations.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 45 No. 11 · 1 June 2023

Tom Stevenson writes that, according to Alasdair Roberts, ‘by 2050 almost 40 per cent of the world’s population will live in just four polities: India, China, the US and the EU’ (LRB, 4 May). Stevenson adds that by his own reckoning, this mark has been reached already. Let’s call this group of polities the S4. In fact, the S4’s share of the world population has been greater than 40 per cent since at least 1960 (when the World Bank’s figures begin), but it has been declining year-on-year since 1973 (when the UK, Denmark and Ireland acceded to the EU), except in the years when the EU was enlarged: 1981, 1986, 1995, 2004, 2007, 2013. A pairwise comparison of these years shows that the sole case in which an enlargement made up for the decline since the preceding one was 2004, when the EU admitted most of its Eastern European membership. Even that increase had been reversed by 2012.

What could increase the S4’s share of the world population? Not births: each of the S4 has a lower birthrate than the world average. Not territorial expansion either: the EU is running out of plausible future members, and the legacy of the one-child policy in China can’t be made good by swallowing Taiwan or Arunachal Pradesh. That leaves immigration. But the relatively liberal policies of the EU and US haven’t reversed the decline in their share of the world’s population, and in any case it’s unlikely that India and China will permit immigration on a similar scale.

J.P. Loo
Oriel College, Oxford

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences