History Books » World History

The best books on Empires

recommended by Peter Fibiger Bang

The Oxford World History of Empire: The Imperial Experience (Volume 1) by C.A. Bayly, Peter Fibiger Bang & Walter Scheidel

The Oxford World History of Empire: The Imperial Experience (Volume 1)
by C.A. Bayly, Peter Fibiger Bang & Walter Scheidel


Empires are a reflection of the fact some states are stronger than others and are by no means just a relic of the past, says Peter Fibiger Bang, historian of empire and world history at the University of Copenhagen. Here, he recommends books on a variety of empires, from the ancient Romans to the Mughal, Qing and Russian empires and explains what it is that made some empires so durable and resilient across the centuries.

Interview by Benedict King

The Oxford World History of Empire: The Imperial Experience (Volume 1) by C.A. Bayly, Peter Fibiger Bang & Walter Scheidel

The Oxford World History of Empire: The Imperial Experience (Volume 1)
by C.A. Bayly, Peter Fibiger Bang & Walter Scheidel


We’re here to talk about books on empires. You’ve just completed a two-volume history on the subject: what have you learned about empires as a result?

Many things, of course, because it’s so  vast a topic. But one thing I have learned is that when you describe the histories of individual empires, you don’t get a sense that the world of state-making, or state formation, is expanding. In fact, the world history of empire basically traces the expanding world of state formation. In the beginning, you can’t have many empires because the world of state formation is very small. Then it expands and spreads out across Afro-Eurasia. This happens continuously until the period of European colonialism, which continues that process. But for some mysterious reason, which we can then come up with explanations for, the strongest European powers failed to conquer their neighbours.

Most other empires have always formed around the conquest of their strongest neighbours. You create a stable core from which you can then dominate an area for a number of centuries. But in Europe, the biggest and strongest nations or powers never conquer each other. Instead, expansive ambitions begin to spill over into territory outside Europe, in areas that are less intensely contested. This is what becomes, eventually, European colonialism. The overseas empire is a kind of compensation for lack of expansion in Europe.

And did they fail in Europe because they tried and they failed, or did they, for whatever reason, simply not try?

For a period, Europe was the most belligerent continent on the planet. There was no shortage or lack of trying whatsoever. They were doing their very best, and they just failed spectacularly, again and again. We talk about realpolitik in this context, with the British always very good at allying with the powers most threatened by the most powerful on the continent. But there’s actually another empire in that story that is very important for understanding this failure, and that is the Ottoman Empire.

“For good and bad, empire has been a formative force in human history for the last five millennia”

We always say that in Europe there is no new Roman Empire. That is, of course, true, but only up to a degree, because we actually do have a new Roman Empire, we just normally describe it as the Ottoman Empire. The Ottoman Empire looks very much like Justinian’s Empire. They didn’t get Italy. On the other hand, they got Mesopotamia. I think that Justinian would have preferred Italy because of the connection with ancient Rome. But if he had been offered the consolation prize of Mesopotamia, I think he would have been pretty happy.

That’s fascinating because Sultan Mehmed, when he conquered Constantinople, entered Hagia Sophia and, as everyone knows, went in to pray and converted it to a mosque. But the other thing he did was declare himself the Roman Emperor. And yet, in the West, we’ve always said, ‘1453—that’s when the Roman Empire finally ended.’

And that is an absolute mistake. One of the closest structural parallels to the Ottoman Empire is the late Roman Empire. Of course, it is not the same. And there are important differences—like the existence of gunpowder. It’s not like history has just been standing still. That’s very important to say. The Ottomans do not exemplify a static Orient, they are a very, very successful, modernised version of the late Roman Empire in many respects.

And did they adopt Byzantine institutions?

Much of that had already happened earlier. The late Roman or Byzantine Empire, which they conquer, is basically a city-state. There’s very little empire left. But if we go back to late antiquity, what is Islam? Islam emerges out of a late Roman competition of religions, about who can formulate the best monotheistic creed. The Quran is full of debates with Jews and Christians. In that respect, even the religion is a project of late antiquity.

Then the Arab Caliphate grows out of the competition between Rome and Sasanian Persia and builds on its institutions. The Ottoman Empire is not identical, but it belongs to a family of state forms that grows out of the ancient world. And it succeeds in becoming a New Rome more than whatever feeble attempts we see in Europe at the same time.

Let’s move on to the books you’ve chosen about empires. The first is The Roman Empire: Economy, Society and Culture by Garnsey and Saller. Why have you chosen this book?

I’ve always liked Garnsey and Saller’s book on the Roman Empire because they come up with this concept of ‘government without bureaucracy’. Discussions of empires had always talked about these super mighty structures. There will be no end to the number of history books that tell of how empires petrify under bureaucratic overgrowth. But most empires in history have had tiny, tiny bureaucracies. Compared to modern states, empires are virtually not states at all. The administrations are so small that explanations of decline that point to empires somehow collapsing under the weight of their over-bloated bureaucracies are, we have to say, just myths. In all the empires we were studying, the bureaucracies are so small that they would never have been able to impose the kind of bureaucratic domination that is implied by that myth.

Get the weekly Five Books newsletter

Garnsey and Saller’s book is so good on trying to ask, ‘how we can understand these empires as government without bureaucracy?’ In the Roman Empire, most provinces have tiny governing groups. If a Roman governor goes out to a province the size of Syria, or Gaul, he would bring with him perhaps a handful of other high-ranking people and maybe 20 or 30 servants and secretaries, and some soldiers to assist him, too. This is not a super-strong bureaucratic structure with which you can count and register everything in the most minute detail and then shuffle things around.

What did hold it together if it wasn’t bureaucracy?

First of all, it was the army. The army was by far the biggest institution. And that goes for all these empires. One of the reasons the Roman army functions so well and is so strong is not that there are never rebellions in the Roman Empire. Most emperors had to face a rebellion somewhere in the Empire. But these rebellions are only locally organized, so they are relatively restricted in scope. For instance, the famous Jewish rebellion which led to the destruction of the Temple—which is why we have the Wailing Wall today—is basically confined to Judea and the districts around it. So, the Roman army can lose once or twice, but you can send in the army from all over the place, and carry on.

The zealots of the Jewish Revolt took refuge at Masada. There are a few hundred Jewish diehards who decided to take refuge in Herod’s desert fortress. And the Romans sent an entire legion, which could stay there for  a year or so.  They build a huge ramp and take the fortress. Those kind of resources most local rebels can’t really match in the long run. They can win one or two battles, but in the long run they are simply ground down.

The other thing is that because you have these big territories, the army is then big enough to be able to hold off outside rivals, at least for long periods of time. It creates an enormous stability. You can easily challenge it from inside, because it’s very decentralised, but it’s too big to be unseated. On the other hand, it’s strong enough to keep competitors away. Therefore, you have these relatively stable empires, the Ottoman Empire lasts half a millennium, the Roman Empire even longer. In the Chinese Empire, you have a number of dynasties, but for very long periods it’s very, very stable.

“Most empires in history have had tiny, tiny bureaucracies”

Stephen Howe wrote a small introduction to empires 20 years ago. It’s very, very good, but he  argued that ancient empires came and went and were very unstable, whereas modern empires were stable structures. There he may been a little misled by a modernist perspective. Modern empires were certainly strong, but to insist on the ephemeral nature of ancient empires is a mistake. That is probably an illusion that you get from looking at it from a modern perspective where you assume that because there were, say, a number of dynasties it looks as if everything changed very quickly. But actually, we are talking about hundreds of years for each one.

In a world history perspective, European colonialism is actually surprisingly short-lived. It has shaped how the modern world looks, and also many of the conflicts of the modern world. So it’s certainly not ephemeral. It’s a very, very intrusive phenomenon. But in terms of stability, it was surprisingly unstable. That was one of the other things which really came out of this project.

Let’s go on to the next book you’ve chosen, which is about the Mughal Empire. This is Muzzafar Alam’s Crisis of Empire. Why have you chosen this book?

I came to it while I was doing comparisons between Rome and the Mughal Empire. If you are studying these structures of government without bureaucracy, you can mistakenly see them as being very ephemeral because of their limited superstructure. A lot of imperial history is about how empires fragment, about their decline. But what was interesting about Muzzafar Alam’s book on the crisis of the empire of the Mughals in north India was that the period of what we normally see as decline is also a period of deeper penetration of imperial statecraft. The form of statecraft that the Empire represents becomes more deeply entrenched in provincial society. So, you could say that the fragmentation is, in a way, the price of success. It’s not failure.

It boils down to this: if you do government without bureaucracy, you’re basically relying on local forces. And empire tends, in the long run, to strengthen local forces because they thrive under it.

If you go to the Mediterranean as a tourist in the summer, when you get tired of lying on a beach, you can go and see a Greek and Roman ruin nearby for entertainment, and you can go back to the beach afterwards, less bored. All these ruins were basically built by local elites under the Romans, who then managed the empire for the Romans, and thrived under it, because the Romans allowed them to skim a large part of the agricultural surplus. And this is what happened in Mughal India too.

The Mughals enabled the local zamindars to get a stronger hold on agricultural society. But that also means that they become stronger and able to drive a harder bargain with the central authorities. So then you have to build up strength in order to control local society more. All of a sudden, the governors and nobles of the Mughal Empire don’t want to change their postings and revenue assignments frequently, because that means they are without power because they have to build up power on a local basis. This is when the Empire fragments.

If you look at late Roman history, from one perspective—the old-fashioned one—historians have identified the development of a very strong bureaucratic machine. It certainly did become stronger bureaucratically than the early Roman Empire. But it’s certainly not a ‘strong bureaucratic machine’. What is really happening is that local elites are getting imperial offices. So they rise in local society. What looks as if you’re getting a bigger state apparatus does not actually quite transform into the much increased power that you would expect. It is the local power holders that are getting more power, so they can actually renegotiate their deal with the state to prevent it from extracting too much. The Roman central government is, in other words, having to run faster to stand still. And, for that reason, as it strikes deeper roots in Roman provincial society, you also see that the Empire begins to fragment into regional courts. In that respect, late Roman history and late Mughal history have surprising structural resemblances.

“The successful empire is characterised by being reconquered constantly from within itself”

Of course every historian has learned that everything is always unique. They will tell you, ‘Ah, but it wasn’t quite like that.’ No, it wasn’t. Nothing ever is. Historians often resemble Heraclitus, who said you can never step into the same river twice, which is of course true. On the other hand, it is useful to be able to observe that there are some banks of the river and that they stay relatively stable and make the water run  in something which is recognisable as the same course.

One can easily recognise that there are a lot of specificities about how these forces play out. But there are also some very important, general characteristics that show that these empires are created out of the same impulses, and that they also revert to some of the same solutions.

When you start to look into it, these imperial administrative techniques and cultures are rarely simply reinvented from scratch every time. They actually build on previous experiences and traditions. Muslim imperial statecraft grows out of late antiquity, so it belongs to the same family as its Perso-Hellenic-Roman predecessors. It’s not just that everything is created uniquely anew every time. Of course there are lots of inventions and variations, but they often belong within a family of solutions.

The next of your books on empire focuses on what is now China. This is Pamela Crossley’s A Translucent Mirror: History and Identity in Qing Imperial Ideology. What’s this book about?

Another problem of understanding these empires is that they are poly-ethnic or multicultural. We always talk about the Roman Empire. Of course it is the Roman Empire in a sense, but there are all sorts of other cultures within it. For instance, there’s a huge Greek speaking elite situated around the coasts of the eastern Mediterranean. They maintain their Greekness under Roman rule. It’s reshaped by Roman rule, but it’s certainly still Greek, and they’re very proud of being Greek—or Hellenes as they called themselves.

Pamela Crossley is looking at the Qing Dynasty (1644-1911). Traditionally we just refer to them as Chinese emperors. But Crossley persuasively argues that we have to take their background as Manchus seriously. The Qing dynasts were very, very keen on cultivating their Manchu-ness. In fact, they claim multiple roles. In addition to being Manchus, they also insist that they are Mongol overlords in the traditions of the Steppe. To this they then added the title of a Confucian ‘Huangdi’ who governed the rice and wheat fields of the main areas of what we understand as China.

She formulates a notion of emperorship as being a sort of parallel rulership. I started off as a Roman historian, so much of my comparative thinking has developed through me trying to learn from other historiographies about Rome. And then I’ve seen patterns that are bigger than Rome emerge and I have broadened out.

One of the other books I could have mentioned, would be Fergus Miller’s book, The Emperor in the Roman World from the 1970s, which was a kind of precursor for the notion of government without bureaucracy. One of the things in that book, which emerges clearly, is that, for many people in the Roman Empire, the Roman emperor was treated, fashioned and constructed as a Greek, Alexander-like King. They engaged with him as a Greek monarch. That phenomenon is what Pamela Crossley describes in a much more articulate way; that most rulers of these big, vast territorial empires had to preside over such diverse populations—and especially diverse elites—that they had to cultivate different formats, in order to embrace their different constituencies. They had to be like chameleons that could change their appearance, depending on whom they were going to talk to.

She was very keen on the multiple versions of the Chinese Emperor, and the Roman Emperor would certainly also both be a Roman republican magistrate in some respects, but he would also have to be a Greek ‘basileus’ in another respect. And, although he doesn’t normally present himself like this, in Egyptian temples built still in Pharaonic style under the Romans, the Roman Emperor is appropriated and depicted as a pharaoh because he had to serve in the role of the pharaoh in these temple cults.

That’s almost like the different localities taking over the centre.

Exactly. And a successful empire is actually one that is eventually taken over by the localities.

In my book I’ve  tried, a little provocatively, to say that Roman history could be written as a history of waves of reconquest. First you have the Romans conquering the Mediterranean. Then, in the third century, you’ll normally have the civil war when the Roman Empire threatens to break up. How is it saved? It’s saved by provincial elites that reconquer the Empire. One of the most famous rivalries from that period is between the Palmyrean queen, Zenobia, who is very integrated into the Roman elite, and Aurelian, backed by an army from the Danube, and Aurelian ends up conquering Zenobia. But there was no strong Italian general. They were coming from the provinces with provincial soldiers, and they were reconquering the Empire.

“In a world history perspective, European colonisation is actually surprisingly short-lived”

At the end, we have Diocletian and Constantine, both of them provincials who reconquered the Empire. And again, with the break-up phase of the Empire, what really happens is that the recruitment patterns of the imperial army changes. There is a tendency to recruit from the frontier regions, it is actually that frontier army that reconquers the Empire. And in some parts the pre-existing structure holds, and in other parts you see a break-up into provincial monarchies, but still visibly shaped by the late Roman world with Roman Christianity and Latin used, albeit to a lesser degree. The successful empire is characterised by being reconquered constantly from within itself.

Let’s move on to a book about the Russian Empire next. Tell me about Dominic Lieven’s Empire: The Russian Empire and its Rivals from the 16th Century to the Present.

A lot of the things that we already talked about are present in Lieven’s book. What I really like about it is the willingness to take this global, long-term view. He actually goes back to antiquity in his first chapter. The thesis is that Russia represents the last big land empire on the planet. And he has some very illuminating analyses of the significance of European state formation, and also the competition between the Ottoman Empire and the Russian Empire.

I learned something about what happens when European colonialism emerges but also about the strength of these older, territorial empires. They were actually pretty durable. The myth is that these empires were basically moribund Leviathans that were degenerate. In 19th century paintings one will see degenerate despots living inside the hareem or the palace, spending their time in debauchery, rather than governing the empire—all this kind of thing. But these so-called degenerate empires were pretty successful actually and continued to develop for a very long time. That was something I took away from Lieven’s book, plus the ambition to see empire in the very long term, instead of just seeing it within a very narrow span of pre-configured world-historical time periods.

Let’s move on to the last of your books about empire which is C. A. Bayly’s The Birth of the Modern World.

This book is not simply about empire. It’s a very successful attempt at writing a global history in the phase where we turned from a pre-industrial age to an industrial age, while also being aware of the dynamics in the non-industrialised portions of the planet. Bayly taught me that I needed to embrace Hegel somehow, because what is characteristic of Hegel’s philosophy is his capacity for combining opposites, to see how they can coexist and maybe even to synthesise them at a higher level.

Support Five Books

Five Books interviews are expensive to produce. If you're enjoying this interview, please support us by .

No one would like to write history as Hegel writes philosophy. But Chris taught me something very important for the historian: the capacity to bridge contradictions—because human society is full of contradiction. It is, in fact, not a logically coherent, Kantian universe. It is much more a universe of very different forces, and often mutually conflicting, and contradicting, and yet forming patterns together. And this is an ideal which I have learned from Bayly and I’m trying to learn from it, but it is a very high art. One can only aspire to it. Chris was also a good friend, so I’m probably biased, but I’m not the only one saying this.

Are there particular conflicts, or contradictions that he identifies that produced the modern world out of the 19th century empires?

The basic claim is that classic modernisation theory will tend to see modernity simply as an outgrowth of either England or maybe Europe while the rest of the world is then left  just standing there, passively on the sidelines.

Bayly was able to show in this book that actually Europe and Britain might have been at the centre of modern development, but part of the energy was driven by the capacity of this centre to enter into collaborations with, and tap the energies of, societies around the planet that might not necessarily have been on course for modernity, but nevertheless had strong dynamics of their own. Modernity arose from this merger, rather than simply something internal to ‘the West’. The ability to tap the energies of India was very, very important in shaping the modern world. So, some of the things that we would tend to think of as purely internal European developments were already connected with the rest the world. Bayly overcame that old-fashioned way of thinking about the West and the rest. He has a brilliant discussion of the Indian liberal thinker Ram Mohan Roy, who was formulating liberal principles for India, long before they came to many European countries.

When was that?

Around the 1820s. The idea that Europe was just confronted with this totally static inert ‘rest of the world’ does not hold water. He shows how this period of colonialism actually also meant an energising of other groups in other parts of the world and produced huge transformations, which also contributed to the transformations that were taking place in Europe. It was a constantly dialogical feedback process.

But of course—which is very important also to remember, in these days, where we actually like to hear about how countries are super nicely interconnected with each other—it was a hierarchical process. Power was exercised and Europe was the main beneficiary of that power, and also able to exercise that power for a period. It wasn’t a cosy, nice, multicultural society where everyone was happily engaging with each other. There was lots of engagement and mutual learning, but it went together with power, oppression and exploitation. It was the combination of these things that was so brutally efficient that it left human society utterly transformed.

Do you think the current discussions of imperialism in all areas of life and culture are helpful? Or do you think they detract from a richer understanding of imperialism, some of which is positive, broadly speaking?

I suppose the current debates are a reflection that empire, both current imperial powers and the legacies of past ones, continue to inform and shape our world in important ways. After decolonisation there was perhaps a period when people tended to think of empire as a thing of the past, a stage in human history which had been left behind. Eric Hobsbawn, the great historian, published a book in the 2000s arguing from this position. But the wars in the Middle East, the rise of China, and our discussion of lingering inequalities and racism show that empire is anything but a thing of the past. As an ancient historian I might be considered quite a radical. I always remind my Roman colleagues who wax lyrical about the blessings of the Roman peace, that the Romans were also pretty good at raping and plundering. Slavery belongs as much to the picture as the prosperous growth of provincial cities. For good and bad, empire has been a formative force in human history for the last five millennia.

I believe that thinking about Rome in a purely European way is often too limiting. We need a world history framework for thinking about Rome. We should perhaps shed some of the baggage from the age of European empires in order to understand Rome better, and take a less Eurocentric view. As an empire, Rome wasn’t simply European, it spanned three continents and depended on tapping and exploiting the energies of the entire greater Mediterranean region.

One of the things that the current debate about empire sometimes misses is that we thought that decolonisation meant the end of empire. But it actually led to the emergence or re-emergence and re-birth of two or three big imperial powers. This is so political because empire is a bad word, but what is America if not an empire? It basically took over and expanded the network of bases that the British Empire built up.

Get the weekly Five Books newsletter

And the Soviet Union—wasn’t that an empire? Of course it was. That is one of the things which Lieven is very good at showing. And, when we think of decolonisation as the dissolution of empire, for the Chinese decolonisation was the salvation of the Qing Dynasty’s empire. Right now, it looks as if there is a pretty strong empire there, which has weathered the storms of the 20th century, and is going to be a very strong force in shaping the future of our planet. That is not surprising when you think of it, because the long-term view is that states and empires go together. States are never equally strong and that’s why empires form. I know it sounds as if nothing ever changes, but empire is simply a reflection of the fact that some states are stronger than others and so they exercise their power.

The other imperial project in the contemporary world at the moment is the European Union.

Yes. I’ve actually written extensively about this. I don’t think that the European Union is in fact an empire.

How would you describe it?

This kind of multinational, polyethnic construct has imperial features. It often ends up with imperial problems. But it has not been created on the basis of a strong army. It has all the weaknesses of empire, but it does not have most of the oppressive strengths of empire.

This is something that the British should remember in talking about the bureaucratic monster of Brussels. The great German 17th century student of Thomas Hobbes and his ideas of sovereignty, Pufendorf, remarked that the Holy Roman Empire was a somewhat irregular body politic, much like a monstrosity. But the monstrous appearance of Europe was precisely not an overwhelming Leviathan, it was a weak, jumble of arrangements, the thing that British elites have always thought it in their interest to maintain.

Interview by Benedict King

Five Books aims to keep its book recommendations and interviews up to date. If you are the interviewee and would like to update your choice of books (or even just what you say about them) please email us at editor@fivebooks.com

Peter Fibiger Bang

Peter Fibiger Bang

Peter Fibiger Bang is a historian at the Saxo Institute, part of the University of Copenhagen. His work combines Roman, comparative imperial and world history.