Jump to ContentJump to Main Navigation
The Prospect of Global History$

James Belich, John Darwin, Margret Frenz, and Chris Wickham

Print publication date: 2016

Print ISBN-13: 9780198732259

Published to Oxford Scholarship Online: March 2016

DOI: 10.1093/acprof:oso/9780198732259.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy).date: 13 December 2018



The Prospect of Global History

(p.3) Introduction
The Prospect of Global History

James Belich

John Darwin

Chris Wickham

Oxford University Press

Abstract and Keywords

Global history may be boundless, but global historians are not. Global history cannot usefully mean the history of everything, everywhere, all the time. Three approaches seem to have real promise. One is global history as the pursuit of significant historical problems across time, space, and specialism. This can sometimes be characterized as ‘comparative’ history. Another is connectedness, including transnational relationships. The third approach is the study of globalization. Globalization is a term that needs to be rescued from the present, and salvaged for the past. To define it as always encompassing the whole planet is to mistake the current outcome for a very ancient process.

Keywords:   global history, globalization, approaches, comparisons, connectedness


Global history may be boundless, but global historians are not. Global history cannot usefully mean the history of everything, everywhere, all the time. There are various ways of usefully focussing the global approach, and the eminent contributors assembled in this volume innovatively deploy some of them. Here, the editors focus on three approaches that seem to us to have real promise. One is global history as the pursuit of significant historical problems across time, space, and specialism. This can sometimes be characterized as ‘comparative’ history, but it can be seen differently as well; it seeks answers to the same question in multiple sites. It is, of course, no easy art, and is discussed in the second section of this Introduction. Another is connectedness, including transnational relationships; these will be discussed in the third section. The third approach is the study of globalization, with which we begin. Globalization is a term that needs to be rescued from the present, and salvaged for the past. To define it as always encompassing the whole planet is to mistake the current outcome for a very ancient process.


Globalizing can be seen as the formation of trans-regional entities, extending beyond any single culture or polity, a process that began many millennia ago. Some simple typologies measuring the space, intensity, and vectors of globalization, may be useful. Globalization connected three categories of space: sub-global, semi-global, and pan-global. There have been only two cases of the last: the original spread of homo sapiens to all six habitable continents, and modern globalization, even if current scholarship unhelpfully dates this to 1492, 1800, the 1940s, and the 1970s. The ‘semi-global’ category stretches across a whole hemisphere. One remarkable example is the spread of Austronesian or ‘Malayo-Polynesian’ cultures, which in and around the first millennium AD spanned the whole Pacific and Indian Oceans, reaching East Africa and very probably the Americas. Another is the Arctic ‘world system’, circling the north of the planet, with specialized bio-technologies, (p.4) ranging from domesticated reindeer and sled dogs to ice houses, toggle harpoons, and whale-hunting umiak boats. In the early centuries of the second millennium AD, a suite of these technologies enabled the Thule Inuit to migrate from Siberia to Greenland, via Canada, displacing the Paleo-Eskimo, Amerindian, and European (Viking) cultures that stood in their way. A third example, on which we concentrate below, is the semi-global system that stretched across Afro-Eurasia, the ‘Old World’, from the fourth millennium BC.

The sub-global category has sneaked into common usage in the form of sub-global ‘worlds’ as in ‘Atlantic World’, or ‘Arab World’. Such worlds can emerge through the meshing of smaller systems, or can spin off a single mega-empire, but stretch beyond it in time and space. Japan and Ireland were part of Chinese and Roman worlds, but not empires. Hellenic and Mongol worlds survived the fragmentation of the relevant empires.

Europe was not a sub-global world in itself, but part of one. Its world also included the Middle East and North Africa. This macro-region’s natural unity stemmed not from similar climate or terrain, but from shared boundaries of ocean, steppe, and desert, and a shared internal mix of land and water. It featured an unusual number of inland seas, enclosed or semi-closed by land. Historians have brilliantly demonstrated how the great Mediterranean linked its littorals and their histories.1 But we have neglected the possibility that other seas did likewise, and that a whole constellation of seas could be connected. The Mediterranean was the flagship of a fleet that also included the Red, Black, Caspian, North and Baltic Seas, the Persian Gulf, and the Bay of Biscay. Straits connected the Baltic and North Seas, and also connected the Mediterranean to the Atlantic Ocean and the Black Sea. Rivers link, or almost link, the other seas. Interestingly, this ‘world’ has no accepted name—West Eurasia, though unfair to North Africa, is the best of a bad job. Whatever its name, it means that Europe is the wrong space in which to understand its own history.

The advent of sea-going boats, about ten thousand years ago, and the domestication of horses and camels, four or five thousand years ago, activated West Eurasia’s connective potential, and while great differences persisted, some common history emerged. The whole region shared the Neolithic Levantine package of domesticated biota, the spread of Indo-European and Semitic languages, and the influence of the ancient Fertile Crescent civilizations. From 500 BC it also shared the influence of a series of tri-continental empires, each of which claimed the legacies of its predecessors: Persian, Hellenic, Roman, Arab, and Turkish. One can almost imagine a Chinese-style situation in which these were seen as successive dynasties of the same empire. A propensity to monotheism was a West Eurasian peculiarity: the Abrahamic religions of Judaism, Christianity, and Islam, to which one could (p.5) arguably add Zoroastrianism and Manichaeism. Finally, the ‘One-God’ world shared two great plague pandemics, further proof of its cohesion.2

This West Eurasian world was Europe’s sub-global context; its semi-global context also included the Chinese, Indian Ocean, and steppe nomad ‘worlds’—together the four ‘Old Worlds’. To measure the intensity of connection within and between these worlds, we need our second typology. Mere contact is occasional and indirect, though it can be important in transferring biota and techniques. The second category, interaction, involves regular and on-going contact, exemplified by luxury trade. The next step up is circulation, exemplified by bulk trade. Finally, integration makes at least sections of a global system mutually dependent parts of a single economic or cultural whole.

The four old worlds, and the system linking them, were created by several vectors of connectivity. Here we deploy the last of our overarching typologies. Vector One is diffusion, where indirect long-range links are achieved by a series of transfers from neighbour to neighbour. Vector Two is outreach, a catch-all category including trade, religious evangelism, and long-range hunting or extraction, an important but neglected sub-type. Vector Three is dispersal, long-range migrations that reproduce the source society but do not necessarily remain connected to it. Austronesian and Thule migrations were of this kind, as were Bulgar and Viking conquest migrations in the late first millennium AD. Both of the latter pair formed several powerful but far-flung and largely unconnected polities. With Vector Four, expansion, the expanding entity remains connected to its point of origin. Great empires are the classic case, but not the only one. The empires of Alexander, Mahomet, and Chingis Khan quickly fell apart, but the parts remained expansive and connected for some centuries. Classical Greek, Phoenician, and modern European expansions were never unified into a single empire, but remained connected to their metropolis. Dispersal and expansion are the two forms of spread, in which the spreading culture dominates local folk, if any. They can be likened to a stretched rubber band, which either breaks into fragments (dispersal) or remains intact (expansion).

A fifth globalizing technique is to globalize through attraction, bringing things and thoughts to you via intermediaries rather than going to get them yourself. Finally, nodal places, regions, cultures, or networks linked two or more expansions or outreaches. An interesting form of nodality is the neutral resource region, unpopulated but used by multiple tribes or nations who have to negotiate shared access. An ancient example is the obsidian island, from Melos in the Eastern Mediterranean to Tuhua in the New Zealand archipelago. Both names mean ‘obsidian’, and were places where tribes met and mixed. A more modern example is the Newfoundland Bank, which from the sixteenth century hosted the cod fishing fleets of several European nations. Nodes are the lynchpins of globalization, and in turn depend on it (see section, ‘Connectedness’).

In this context, the question of divergence becomes something more than an international beauty contest in which the winner always looks rather like the judge. (p.6) If one part of a global system hits upon some development considered advantageous by the rest, they will seek to adopt it, or raid or trade for it. If the advantage facilitates dispersal or expansion, they may have little choice. Eventually the divergent advantage will disseminate to at least the urban clusters within the global system. Thus divergence will lead to some convergence until another divergence arises. Convergence is not intended to imply homogenization. What converged was a growing menu of bio-technological and cultural (e.g. religious) options, mixed and matched in locally very different ways. Divergences could also create or expand globalized zones. Europe’s ‘great divergence’ was in fact the last of at least four stemming from the Old Worlds. One could easily add a fifth: the massive spread, through peaceful outreach, of Buddhism from India, mediated by the Kushan Empire from the first century BC, and stretching from Japan to Samarkand, but that will not be discussed here.3

Ivory from an Asian elephant has recently been found in a Neolithic workshop in Spain, dated to 4,500 years ago,4 and cloves from Southeast Asia were reaching the Euphrates less than a thousand years later.5 These valuables were probably transferred by diffusion involving a sequence of land and sea transport. The first clear-cut divergence connecting and so creating the Old Worlds system did not come from towns or trades but from the domestication of the horse by the nomads of the Eurasian steppes. At first, from about 4,000 years ago, horses were used in twos and threes to pull chariots: from 3,000 years ago men rode them singly using compound bows. In the second millennium BC, horse nomads conquered and migrated their way deep into West Eurasia, Northern India, and Northern China, and their technology migrated even further to enable others to stop them, so linking and converging the four old worlds. Pulses of dispersal from the steppes, such as that of the Huns in the fourth century AD continued to afflict all three urbanized old worlds until the eighteenth century.

It is easy to underestimate the full historic importance of the horse-human alliance. Horses tripled human power, speed, and mobility. Equipped with Chinese horse collars, they helped revolutionize agriculture on heavy Northern European soils around 1000 AD. As late as 1850, horses supplied over half of all work energy in the United States.6 Simply as transport, they provided a huge military advantage. They made possible long-range raiding, which reduced the risk of retaliation. They delivered fresh troops to the battlefield to face exhausted foot soldiers. When warriors learned to fight from chariots and horseback, the equestrian advantage (p.7) multiplied. Chariots were quite a complex technology, a basket on metal-rimmed wheels weighing as little as 35 kilograms, whose horse team and charioteers required complex training. With compound bows and mass-produced cast metal arrowheads, trained archers could fire ten arrows a minute from horseback, even when retreating—the Parthian shot.

Both chariot and mounted archer bio-technologies were state-of-the-art, and sedentary states were forced to emulate them if they could. This was an enduring problem for China and India, whose ecologies did not suit the breeding of large horses. Here, sedentary states were compelled to acquire horses regularly from nomads through trade or alliance. Unlike most Sino-Indian imports, fresh horses were a necessity, not a disposable luxury. The advent of guns around 1300, in China then Europe, did not eliminate the horse archer advantage. Horse nomads would not wait around to be shot by cannon, and hand-held firearms were too short-range and cumbersome to be much of a threat to them until the later nineteenth century. Qing China crushed the last Mongol empire, that of the Zhungars in Central Asia, a century earlier, but only by a sustained, massive, and hugely expensive overland projection of force. Europeans in the Americas had to wait until the advent of repeating firearms in the 1870s to overcome the horse nomad empires of the Sioux, Comanche, and the Araucanians of the Argentine Pampas. Until then, European expansion had transferred its own antidote—horses.

The horse nomads connected the old worlds, at first through contact and dispersal, and then, with various Turkic and Mongol empires from the third century BC, through expansion and interaction. A second great divergence was anchored in China and India and based on the export of fine coloured textiles of silk and cotton. Beginning with diffusion and sporadic contact before 1000 BC, it moved into regular and substantial interaction in the last centuries BC, via the Silk Road, with nomad assistance, or via a monsoon-driven maritime route centred on the Indian Ocean.7 By the early centuries AD, Chinese silk and Indian cotton had reached the Atlantic coast of West Eurasia.

Immanuel Wallerstein might deny the label ‘world system’ to this vast interaction on the grounds that it lacked an axial division of labour.8 Manuel Castells might contest the label ‘globalization’ because a shared and self-perpetuating body of information was absent.9 Yet there was a fundamental and resonant division of labour. China and India did the manufacturing, and the rest of the world paid in non-manufactures, such as spices, dyestuffs, horses, furs, and, especially, bullion. And there was a complementary set of semi-globally shared and self-perpetuating information and valuations. All cultures within the system agreed that bright silks and cottons, later joined by Chinese porcelain, were the most desirable of manufactures. ‘For over a thousand years, Chinese porcelain was the most universally (p.8) admired and most widely imitated product in the world.’10 It was also internationally agreed that silver and gold, intrinsically the most useless of metals, were similarly valuable. From Pliny the Elder in the first century AD, through Armenian and Ottoman commentators in the fourteenth and seventeenth centuries, to the governors of the British East India Company in the early nineteenth century, we hear the same complaints about the one-way flow of bullion from West Eurasia to China and India.11

In the debate on ‘the’ Great Divergence—of Europe from the rest of the planet—those who argue that China and India had an economic edge as late as 1800 have been accused of over-correcting the Eurocentric pendulum, even of ‘Euro-envy’, in Eric Jones’ pungent term.12 Yet the evidence of bullion flows, like the reverse flow of Chinese techniques, seems impartial, and it survived 1,500 years of West Eurasian efforts at emulation. This Sino-Indian Great Divergence was not a matter of some intrinsic Asian dexterity. Rice, a highly productive super-crop, permitted dense populations and regional specialization in growing and weaving silk and cotton. Social complexity attuned textile producers to multiple and changing markets. Caste, clan, and lineage encouraged hereditary occupations in which children learned knack as well as skill. The Sino-Indian manufacturing divergence was also part of the answer to the question of why China and India did not expand overseas, though there were times when they could clearly have done so. The world came to them, bearing its valuables. They had no need to go to it, and globalized more by attraction than expansion.

From about 500 BC, West Eurasians could regularly buy Sino-Indian luxury textiles. Within a few centuries they could weave silks and cottons themselves from imported yarn. From about 500 AD, they could farm cotton bushes, mulberry trees, and silkworms and create the whole product themselves, though not, until about 1800, to quite the Sino-Indian benchmarks in quality and price. Tiny volumes of bright luxury cloth may seem unimportant, but this is deceptive. It could be used to express anything from individuality to uniformity, as well as status. The trade itself was light, but not small. Thirty square metres of fine cotton fitted within a coconut shell.13 What economists would call its linkage effects was considerable—ten tons a year of silk or cotton fibre employed a thousand weavers in Genoa or Cologne. Above all, it drove emulation. The Lancashire cotton industry is widely agreed to have been crucial to English industrialization, and its main game was matching Bengal.

(p.9) West Eurasia probably achieved the shift from interaction to circulation around 100 BC, thanks largely to Roman expansion. The archetypal circulation was the grain trade, in which Egypt, Tunisia, and parts of the Black Sea coast fed Rome itself, its army, and eventually the second capital of Constantinople. There were other bulk trades: timber, marble, salt, wine, olive oil, and stranger trades too, notably the commerce in wild beasts, destined to be killed in arenas for the entertainment of Roman crowds. This last trade pushed out the edges of the Roman world, as Roman hunters or their agents searched for big beasts along the shores of the Baltic and deeper and deeper into Africa.14 We can document at least two post-Roman periods of West Eurasian circulation corresponding to the two plague pandemics, 540s–760s and 1346–c.1800. A third, modern, pandemic, 1890s–1920s, covered much more space but killed fewer people. All three pandemics required bulk trade or circulation to spread and re-spread. Pandemics and intensifying globalization are good proxies for each other.

In the early seventh century, the third old world ‘Great Divergence’ took place: Islamic expansion. The astonishing initial conquests of the Arabs, unified by Islam, stretching from Spain to Sind in less than a century, were matched and extended by Islam’s own equally astonishing outreach, through trade and conversion. The former reached India and China by sea in the eighth century AD; the latter voluntarily converted steppe nomads, including two Mongol khanates, and many trading partners. Muslim merchants in China were sufficiently numerous and rich to provoke massacres in 760 and 878.15 But they kept coming. The famous Ming admiral Zheng He, whose great fleets reached East Africa and the Red Sea in the early fifteenth century, was Muslim. Where documents are lacking, Islam’s penetration of South East Asia is mapped by the sudden disappearance of pork bones from middens.16

Though the Mongol legacy can be underestimated, Islamic expansion was more durable. The same is true of the fourth old world ‘Great Divergence’, that of Christian Europe, from 1400. As various historians have observed, Islam forced Christian expansion east and west, away from the more favoured south—the ‘Great Diversion’ of Europe. It is intriguing to note that the terrible twins of West Eurasia came from the same sub-global world and shared essentially the same god. A global approach makes it hard to see how their histories can continue to avoid each other.

(p.10) Comparisons

Kenneth Pomeranz may be right or wrong in dating ‘the’ Great Divergence, the fourth by our count, between (north-west) Europe and (central-southern) China to as late as 180017—our authors themselves disagree about that—but the message of the foregoing is a different one. The range of these varied divergences, as already stressed, decentres the standard ‘we-are-best’ international grand narrative which concentrates on pan-global globalization, and on the fourth divergence listed above. The teleologies of power relationships, when they are posed as always having to lead to the Western conquest of most of the known world in the century and a half after 1750, or to the Western-dominated world economy of 1870–1970, are false. Global history must extend beyond them, and beyond the issue of divergence, to such issues as the globally focussed analysis of difference which underlies comparative history, and the globally focussed analysis of connectedness which underlies transnational history. Global history-writing also does not have to be restricted to the modern or modernizing world; there have always been globally directed developments of different types. We will come to connectedness in a moment, but let us look here at how to analyse what distinguished the various regions of the globe which did (eventually) connect with, or diverge from, each other, at how such differences can be compared, and at what such comparison is for.

Comparison can of course be done using any number of axes; what is essential is that it is done rigorously. And we would add reciprocally, using Pomeranz’s terminology, that one side of the comparison should not be considered the ‘norm’ and the other the ‘deviation’. This point is indeed widely accepted; it underlies the several discussions of comparison in the book which most closely parallels this one, Writing the History of the Global, edited by Maxine Berg.18 It also underpins Max Weber’s still-essential discussions of ideal types, which are invented constructs developed specifically for the purpose of neutral comparison, in that different real societies can be usefully analysed and compared according to how they fit, or fail to fit, the different elements in an ideal type, as long as it is set up properly.19 Comparison can, however, be developed at various levels. Victor Lieberman’s massive comparative volume sets whole political systems against each other, from Thailand to France, across a millennium of history up to 1800, and links state structures, economics, and culture in each as he does so. He also argues that much structural change in each of his case studies had a (very) broadly similar chronological phasing. He develops a precise argument from that similarity: that Eurasia as a whole tended to react in parallel ways to common stimuli, such as climatic shifts, and also, later, came together more and more in its development as a result (p.11) of the early globalization trends discussed above.20 Alternatively, comparison can be more specific, as in Pomeranz’s comparison, not of the whole of China, but of the lower Yangzi, with England in a single century, 1750–1850.

Similarly, at a more abstract level, Garry Runciman’s major work of historical sociology, even though it looks to the whole world and to every historical period for its typologies, focusses on separable modes of power (production, coercion, and persuasion) when he makes his comparisons, and above all on ‘the competitive selection of practices’, such as patron–client relations as opposed to mass political parties, which led individual polities in history in one socio-political direction or another.21 It must be added, and stressed, that comparison can be more focussed still: it can be restricted to individual features, which appear, in different contexts in each case, in a wide range of different societies, with the differences between the contexts then used to unpick and explain wider differences in development. In principle, almost anything could be used here as a way into such comparative analysis: the political role of fortifications, the training of officials, the extent of wage labour, the elements which go together to define political élites, the economic autonomy of religious organizations, the social role of literacy, the level of violence considered to be necessary to protect honour, the perceived immediacy of divine intervention, are all axes along which comparison could usefully be pursued in any period and across any geographical space.

Each of these levels of comparison can be defended for its usefulness; we certainly do not advocate one over another. But each contains risks. Lieberman’s style of large-scale holistic comparison runs the risk of mixing incompatible elements together, or else, by focussing on whole political systems (China rather than the lower Yangzi), of implicitly privileging the political over the economic or cultural. But non-holistic comparison carries a risk too, of failing to understand properly the exact way in which the axis of comparison chosen might allow us to understand wider economic, social, political, cultural structures in each chosen region, without which the exercise is effectively useless. These risks need to be recognized explicitly, and constantly corrected for, in any global (or, indeed, non-global) comparison. There are other risks too: for example, failure to understand differences in our evidence-base—in both what it makes clear or ignores, and in the constraints of its genres—from place to place. That in itself tends to be a spin-off of the most substantial risk of all when undertaking comparative work: too great a reliance on prior work in the secondary literature. Such work will seldom have been done with a comparative eye, so will often fall into the traps which characteristically appear in non-comparative work, such as considering one’s own field of study as normative, or else failing to confront (or even notice) the dominant grand narratives in (p.12) the region one is focussing on. Scholars of ‘world systems’, and of divergence and convergence, must guard against this danger. Global yet focussed comparative history may be the best way to test, and if necessary undermine, grand narratives. But it will find it difficult to do so unless it recognizes their existence in most secondary work written on the area of study. The solution to this particular risk is hard, but usually necessary; it is to understand and rely on the primary sources above all. In global comparison, they will usually be in an uncomfortable variety of languages. So a properly conducted global comparative study will have to confront linguistic variety (and will thus, in practice, usually be collective): we may skitter over the surface of understanding if we do not.

Why do we do comparative history, given these risks? We would say, for three reasons. First, as has just been implied, to undermine—if possible, to destroy—inaccurate and self-serving national or international clichés, such as that ‘Western feudalism’ is a better preparation than ‘oriental despotism’ for the capitalist breakthrough (this is a terminology which no-one uses now, but it underpins a surprising array of more apparently scientific works), or, more specifically—and this claim is still made—that the security of property was too weak in Asian systems for capital accumulation to take place.22 Second, to solve specific problems by looking at them comparatively, testing explanations against different examples of the same process. Print culture, for example, is often stated to be one of the elements which fuelled the spread of Protestantism in northern Europe, but it never had that effect in Italy, although Italy was a focus of early European printing and of a long tradition of autonomous lay piety; and print had no special links with dissent in China.23 We look at Protestantism differently once we recognize that. Third, to get an idea of (and explanations for) the real inner structures of diversity, including in contexts of change and convergence. Thus, taking the example of a superficially convergent transnational structure, we need to recognize that the social and political role of the Christian church was very different in 1000 between Ireland, Italy, Byzantium, and Chinese Turkestan, as also is that of Islam today (even if it usually has no structure equivalent to a church) between Britain, Egypt, Iran, and Indonesia; and we will need to analyse both sets of differences comparatively if we wish to see why. In this context, it may indeed often be better to start with regions, not political systems, except when one is studying state structures; the political role of the church in Italy in 1000 was different between Rome and Milan too. It is also sometimes better to start with elements rather than wholes, as Linda Colley does in this book with the writing and reception of constitutions, for the different structurations of individual elements, again in a context of transnational connectivity, will give a more immediate idea of difference.

(p.13) Let us take a brief example of a possible comparative study, one with little discernible evidence of transnational influences associated with it, the relation between military elites and civilian elites in pre-modern governments. (Here, the object of comparative attention has to be whole political systems.) All developed polities rely on organized military defence, and thus have armies which tend to dominate coercive power; military leaders are thus normally major political players. But all developed polities also need trained and expert officials to run them, whose careers are not typically focussed on the battlefield. Furthermore, all developed polities have various levels of landed aristocracy, which, particularly (but not only) in the pre-capitalist world, tend to have their own direct access to political power, thanks to their private wealth. How did all these interrelate in the past?

The Roman empire had two levels of landed aristocracy, focussed on the cities of the empire and on the capitals, which were both characterized for the most part by a very strong civilian culture, based on a literary education; both furnished the personnel for the civilian bureaucracy. The army, although perfectly capable of replacing emperors by coup, was relatively marginal to the standard career-structure of the other elites; although it was, as a result, rather more meritocratic too, wider social status depended more on wealth and education than on arms. Rome’s Byzantine successor, however, although maintaining much the same form of state, after c.800 developed a military rather than a civilian aristocracy; as a result, given that aristocrats could and did gain education and civilian training as well, there was a much closer link between civil and military careers, and military expertise had a much higher status—even emperors composed manuals of military strategy. In China, often compared to Rome, the aristocracy was more military under the Han and Tang dynasties than it was later, under the Song and Ming. In China career bureaucrats were more often called to arms than under Rome, but a separation of roles is clear by 1000, in the Song period, with local aristocratic elites more directed to the bureaucratic career-structure, and civilian/educated (‘mandarin’) values, than to the military one, which was relatively culturally marginal—Song China being in this respect more similar to Rome than was the Han dynasty with which it is most often compared. The Arab caliphate and its medieval successors in the Middle East from Egypt to Iran, at least after c.860, was dominated politically by military leaders and their clients—themselves often future leaders—who had great power but rarely any landed base or élite origin; here the educated landed elites either provided bureaucrats who were subjected politically to army leaders, or restricted their political ambition to provincial-level politics. This did not lead to a hegemony of military values, however, but rather to two cultural and political worlds which had relatively little to do with each other, except at the interface between politico-military strategy and governmental expertise (the level at which military and civil were linked everywhere, that is to say). Finally, in the smaller and simpler polities of medieval Western Europe, where landed aristocracies were fully militarized and bureaucracies fairly sketchy until the thirteenth century, military values reigned supreme as secular status markers; even after c.1250, when states and thus their officials were more powerful, any secular leader needed to have a military training, and only a specifically ecclesiastical career (which usually (p.14) excluded people from the highest levels of political leadership) could exempt the ambitious from it.

These five examples were all long-lasting and stable political systems (stability being here measured by the survival of the system, not of its individual leaderships—all these political systems in some periods and places, third-century Rome, tenth-century China, fourteenth-century Egypt, fifteenth-century England, had very short-term leaders, who died by violence). They had in many respects similar political structures—the most distinct was probably the medieval west, which had much simpler structures until the fourteenth century or so. The military vs. civilian balance was not, then, itself of structural significance for their durability: an observation which already runs counter to some of the local historiographies for each of these large regions. It should also be clear, even from these highly summary descriptions, that the involvement of the landed aristocracy in military activity, or its absence, was fundamental for determining the parameters of socio-cultural status in each of these systems. Only in the Islamic world, with its unusually total separation between the military and the civilian worlds (in some Islamic polities, ruling military elites were exclusively recruited from abroad), were the military values of non-landed groups not marginal to the rest of society; indeed, only in the Islamic world did landed aristocrats not dominate political practice and most political power. But we can also see that the varying association of much the same elements led to wider political practices whose dynamics were quite distinct. To push this further, we would of course have to add more parameters: the role of tax-raising in each system, or the level of centralization (how far did people look to the capital and the supreme ruler for advancement and cultural leadership in each system, and how important were provincial leaderships). We could of course also add more examples; Japan, the Guptas, the Khmers, the Aztecs, would all add variants on these basic patterns, to nuance our hypotheses and our explanations. But we can get a long way in posing, and sometimes answering, questions about how political systems worked in the past by developing just the examples presented here, in a comparative analysis.

Looked at with a comparative eye, every society of the past, across the globe, gives us a new set of questions to pose of other societies, and a new set of alternatives. The global becomes, indeed, an array of possibilities. Every society has paths not taken; which, and why? Comparison allows us to see which they might be, and in which ways the ‘normal’ can be refigured as the atypical. The grand narratives which dominate our historiographies will often dissolve as a result. Not always; some of them are strong enough to resist. But the proper testing of each of the storylines which historians like to construct for the past is a testing which encompasses all the possibilities which global comparison can bring.


The history of connectedness, strictly, perhaps, of uneven or differential connectedness, is, finally, the most distinctive terrain of the global historian. It follows that (p.15) global history must be centrally concerned with the history of mobility—the movement across space of people, goods, and ideas. Indeed, it might well be argued that one of the most compelling insights that global history has to offer is to insist that the systematic investigation of the forms of mobility is an indispensable part of historical inquiry in every place and period. If l’histoire immobile was the call to arms of one great school of history, for global historians the reverse must hold true: history is movement.

That this should be so may be explained by two exceptionally durable traits of human behaviour. It was Adam Smith who remarked on the human propensity to ‘truck, barter and exchange’. Smith was thinking perhaps chiefly of goods. But a moment’s thought will suggest how easy it is to extend that insight into the realm of ideas, information, beliefs, and practices—since few material goods travel far without some cultural freight. The long-distance traffic in objects that were thought to confer either cultural prestige or magical powers was probably the oldest kind of transcontinental trade in Afro-Eurasia. Indeed, much of that traffic must have assumed prior knowledge of the status of the goods sought out by their would-be consumers. Thus it was their alertness to the importance that the rulers of Middle Eurasia attached to the conspicuous display of gemstones that created the ‘market’ for rubies and sapphires in the courts of China.24 Much the same might be said of the trade in relics and sacred texts that connected the Buddhist world, as it did the Christian. Belief and knowledge preceded ‘demand’. Both were the product of the thirst for novelty—in equipment, possessions, or cosmological power—and for the advantages it might bring, that might be thought characteristic of most human societies, unless restrained by isolation, poverty, or prohibition.

The second great trait has been the urge to migrate, from the original migration ‘out of Africa’ some 70,000 years ago up to the vast flows of migrants, migrant labour, and refugees that make up the ‘demographic globalization’ of our own day. Migration in search of safety, survival, freedom, prosperity, or simply loot, has surely been one of the great constants of world history. It was the variations in the volume, velocity, direction, and motive of this ‘world on the move’, and in the regimes that have sought to contain or constrain it, that helped shape the (shifting) patterns of global connectedness as much, perhaps, if not more than those that arose from the exchange of goods and ideas.25 Indeed, the two are not easy to separate, since both exchange and migration required human vectors; and while both contained the potential to reward those who entered the traffic, they could also subvert, demoralize, or even demolish the rulers and communities that they enveloped. We might note in passing that states and empires were invariably parasitic upon these forms of connectedness, and sought to extend and exploit them. But they were also fearful of their unwanted effects, and much of their administrative apparatus sprang up to surveille and proscribe ‘undesirable’ kinds of mobility. We will return to this point later on.

(p.16) If exchange and migration required human vectors they also depended upon, and necessarily created, a more or less dense set of networks, the infrastructure of transmission. That infrastructure was determined in part by the constraints of the physical world, since networks need routes over land or sea (or through the air). Hence no historian of mobility can neglect the changing geography of roads, railways, sea-lanes, telegraph-lines or wireless transmissions, and the influence it exerted. Roads and railways disfavoured mountain, marsh, and jungle and sought out instead ridges, rivers, plateaux, and steppe.26 The sea might have been a ‘wide common over which men may pass in all directions’. But it was also one ‘on which some well-worn paths show that controlling reasons have led them to choose certain lines of travel rather than others’.27 In the sailing ship age, those paths were dictated in part by the seasonal pattern of winds and currents, and the constrictions imposed by navigational hazards.28 On both land and sea, it was the supply of energy that ruled the volume of traffic. Draught animals needed grass (or a substitute); steamships and railways needed wood or coal; sailing ships needed wind. The result was that communications, and the networks that depended on them, have been channelled into corridors, and sucked towards junctions the alternatives to which were costly, inconvenient or dangerous. Human infrastructure was erected on these physical foundations. All networks required agents as well as principals, and some level of trust between them. They required a ‘currency’, whether of money, reputation, or faith, to reward their participants and discourage defection. Their institutional needs were governed by the size and shape of the traffic they served: thus (to name only three kinds) chambers of commerce and commercial exchanges; the Singapore doss-houses in which Chinese migrants were held while their indentures were sold; the mission-stations, monasteries, or sufi lodges (zawiyas) where the reservoirs of faith were filled up to irrigate the converted or swamp the infidel. Since institutions need resources, the capacity of a network would also depend upon its ‘earning power’ or its access to wealth or the wealthy.

For the global historian, the attractions of the network are obvious. They permit an escape from the historiographical domination of states and empires and their overbearing archives.29 A network may evoke a world of artisans, dealers, and traders whose archival footprint is usually quite modest. Amongst elites, it may provide a social and cultural context that helps to explain their decisions and preferences.30 Networks remind us that non-hierarchical, negotiated worlds coexist with those governed by regulation and rank. They can help us to grasp the scale of subaltern or (p.17) subversive movements and the reach of their influence.31 They reveal, as is now almost commonplace, the permeability of borders, and the multiple (sometimes conflicting) allegiances of historical actors. It is easy to re-imagine a borderless globe populated not by states but by a pullulating mass of networks: administrative, commercial, financial, religious, migrational, scientific, educational, and military—all bound together by investments, remittances, bills of exchange, reports and despatches, telegrams, letters, parcels, newspapers and books, as well as by soldiers and sailors, merchants and salesmen, indentured labourers and slaves, peripatetic officials, wandering scholars, journalists, travellers, and tourists. But as the more rigorous practitioners of network analysis warn us, before we can judge the significance of a network (or even concede its existence) some hard questions must be asked. How extensive is the network? How frequent are the exchanges (of information, personnel, money, or goods) between the different parts of the network? How large is their volume and on what scale of intensity—that is, how important to its members are the exchanges between them? But this is only the start.

Perhaps the crudest distinction drawn between different kinds of network is between those that are ‘weak’ and those that are ‘strong’. A ‘weak’ network might link a large number of ‘nodes’ or members, but as with those who lay claim to hundreds if not thousands of virtual ‘friends’, the bonds and solidarities between them might be fragile at best. We might be tempted to dismiss large ‘weak’ networks as useless or dysfunctional, and, where they have existed historically, as of little significance. That might be mistaken. Some network analysis suggests that large weak networks can be highly functional if the object is to maximize flows of information (a classic case might be our use of the Web). Those bound together by the circulation of newspapers might have had little else in common. Yet by their use and recirculation of the information received they might (dramatically) strengthen the cohesion of their other connections and networks. Moreover, there is no reason to assume that the conditions in which a network emerges remain constant. An event that promises political or social upheaval can galvanize a weak network, and endow it, if only temporarily, with an intense solidarity. Indeed, a weak network may have the capacity to transform by circulation, repetition, and ‘translation’ the significance of the event itself. The weak networks that helped mobilize the first crusade32 or which ignited the ‘Arab Spring’ as a mass movement might serve as examples.

They might also suggest the importance of distinguishing networks by function, where different rules of connection might apply. Perhaps the obvious distinction lies between networks designed to inform and those intended to coordinate—for example, political, commercial, religious, or migrational action. In the ‘real’ world, the difference might seem more theoretical than real, since many—maybe most—networks combine both these functions in varying proportions. How effectively they do so (and how they do so) is likely to depend upon the wider ‘ecology’ of (p.18) which a given network is part. The transmission of information, goods or people was always conditioned by at least three larger environmental constraints. The first was geopolitical: the ease or difficulty with which borders could be crossed; the safety or otherwise of long-distance travel; the reception of aliens or immigrants in a foreign jurisdiction; the extent to which movement was facilitated by the protection of large states and empires—the Pax Britannica or the Pax Mongolica. The second was technological: the ease and cost of movement, and (as we saw earlier) the influence that the technologies of movement exerted upon the circuitry of travel (which places were accessible, in what sequence, and in what season), and the realities of distance reckoned as time. The third was cultural, or perhaps primarily linguistic: the availability of a lingua franca and the status of linguistic intermediaries to manage the transition between language zones; the existence, or otherwise, of formal prohibitions such as those that long discouraged the development of printed publications and newspapers in the Ottoman Empire or prevented the circulation of information except that approved as ‘Dutch knowledge’ in pre-Meiji Japan; the extent to which a scribal elite could exercise a monopoly over the diffusion of information and knowledge.

We might want to ask what it is about a set of connections that would justify describing them as a ‘global network’. In an influential article, S. M. Sindbaek contrasted the shape of early medieval networks of communication and exchange with their ‘global’ successors. The former, he argued, were characterized by the ‘thinness’ of their connections, and hence their vulnerability; the small number of ‘hubs’; the paucity of travellers and the tendency for the same few persons to become ‘specialized’ travellers.33 It is a useful reverse template for the global network, but perhaps one that flatters. A modern global network, like a missionary organization, might well depend upon little more than a handful of specialized travellers, some well known to each other. But a more important issue might be how many ‘global’ networks were or are actually global, in the sense of spanning the world. The changing constraints of geopolitics, technology, and culture were always likely to make them at best ‘sub-global’, capable of wide but not unlimited reach. That should point to the importance of ‘nodes’ in a network, and especially of those that functioned as ‘choke points’, ‘bridgeheads’, or ‘gateways’. These were the locations on a network where it was forced to ‘change gear’. They might offer a range of services: the storage and defence of valuable goods as in Saharan oases;34 the negotiation between widely differing conventions of commercial practice including the forms of currency; the special transport expertise (on land or sea) required to enter unfamiliar or demanding environments; the provision of expertise in language or cultural lore needed for onward connection; or simply market information about regions with which contact was irregular. This kind of node served as a ‘transformer’, changing the ‘voltage’ that passed through the network, changing the medium through which its content was transmitted, and perhaps changing its message and substance as well.

(p.19) We can see these functions most vividly at work in the history of port cities. Port cities existed to collect and distribute flows of goods, information, and people. They grew up at locations arranged along different maritime circuits and throve where long-distance seaborne traffic was most easily transhipped into smaller craft or forms of landward transport—caravan, ox-cart, railway, or lorry. Because they usually linked several different ‘interiors’ as well as the way stations on the sea-lanes that served them, they attracted a variety of different mercantile groups. Indeed, a striking feature of port cities was how often their merchant elites were drawn from distant locations and had few cultural ties with the local community. Since they traded in knowledge (including religious knowledge) as well as in goods, alongside their banks and exchanges, go-downs, and counting-houses, could be found centres of learning or religious instruction, as well as (in more modern times), post offices, telegraph offices, museums, geographical societies, and botanical gardens. They buzzed with rumour, correspondence, and the scribblings of news-writers. By the mid-nineteenth century, few lacked a printing press, churning out books and newspapers, some intended for faraway readerships strung out along their sea-lanes, river routes, railway lines, and roads.35 Here was the setting in which multiple networks conducted their business, recruited new partners, and spied out the land.

But for those networks to flourish something more was required than the mutual attraction of their members and nodes. Their fate depended in part upon a larger ‘ecology’. Among the most successful port cities were those that could profit from the ‘overlap’ between very large networks. In the nineteenth-century world, for example, the ‘British world’ (a zone made up of colonial possessions and the informal spheres where British influence predominated) shared large tracts of the globe with the ‘Islamic world’ and with the large maritime region occupied by settlers and sojourners from South China, the so-called Nanyang. Port cities like Aden, Bombay, or Singapore were outposts of British authority, and hubs in their system of steamships and mail-routes. But they were also key staging posts for the movement of Muslims whether on pilgrimage or in search of employment or learning. Singapore and Hong Kong formed the twin poles of the Nanyang: between them flowed a wide stream of migrants and remittances, commerce and philanthropy. It was ‘overseas Chinese’ in Singapore who helped to endow Hong Kong University and who dreamt of the modernization of China under ‘Anglo-Chinese’ auspices.36 As these examples suggest, crucial to the success of port-city networks, and perhaps to global networks more generally, was the geopolitical regime to which they were subject. Those networks might flourish in the absence of empires or predatory states: perhaps the Indian Ocean before 1498 provides an example of this. But (as the same Indian Ocean after 1498 reminds us), they were vulnerable to the intrusion of armed interlopers on land or sea and the rivalries they brought with them. The richest merchant in Bengal before 1757 was part of (p.20) the extensive Armenian mercantile network. Plassey ended his reign and he died in prison.37 By contrast, a regime sympathetic to free movement and free trade, and with the power to enforce them (a parsimonious description of British imperialism) enabled the proliferation of innumerable networks and might even offer them legal protection. It was a Bombay colonial court that upheld the claim of the Aga Khan to the hereditary leadership of the far-flung Ismaili sect, and to the vast tithe income that came with it.38

Port cities thus functioned in part as ‘vertical nodes’: that is, they linked networks at different levels or scales. They connected a mass of ‘unofficial’ networks to the apparatus of authority and military power—the networks of empire and of the global order of which empires were part. They also serve to remind us that states and empires have many of the characteristics of a ‘super-network’. Realistically viewed, empires (whether land-based or sea-borne) were typically a patchwork of tracts, precariously linked by a skein of routes and corridors, beyond which control was much less complete. ‘[A] l’origine des grand pays, des grand corps nationaux, il y a toujours une “route”  ’, said Lucien Febvre.39 Securing the corridors was the first priority of imperial rule: it was hardly surprising that the earliest forms of imperial map-making were so-called ‘route-surveys’ or itineraries.40 Empires existed to exploit the opportunities created by mobility and to control its consequences—so far as they could. Unless they could do both, their time would be short.

Much of the historiography that we have inherited was fashioned for a world preoccupied with nation-making. It saw history as the rise of discrete and durable structures: constitutions, races, languages, literatures, ‘advanced’ economies full of railways, bridges, and banks, and, above all, states. The shock of the First World War began to unravel this historical fabric and set off a brief era of exceptional creativity in historical writing. But the uncertainty principle on which J. M. Keynes laid such emphasis in the mid 1930s—that flux and instability were the norm in human affairs—was checked by the outcome of the Second World War which licensed new histories of progress: Marxist, nationalist, liberal, and social democratic. Our own times have revived the sense of uncertainty. They encourage a view of world history as one swept by endless tsunamis of unpredictable change and irresistible force: climatic, epidemic, environmental, ideological, religious, technological, demographic, and geopolitical. The vehicle for many if not most of these is the incessant mobility of human societies. Tracking that mobility across the world through its complex circuits and networks may be the main contribution that (p.21) ‘global’ history can make to the wider discipline. But those circuits and networks should also remind us that for historians of every period and specialism understanding mobility is foundational. Why, asks a Saharan historian, should we give precedence to place over movement?41 Why indeed?

The Essays in this book

The essays which follow are designed to develop the problematic of global history in different ways. We start with two general methodological discussions. Jürgen Osterhammel and Kevin O’Rourke look at the interrelationship between history and two of the social sciences in turn, historical sociology and economics, as guides for constructing a rigorous analysis of the global; both of them stress the weaknesses as well as the strengths of the contribution of the two social science disciplines in this field, but each shows how the combination with global historical practice strengthens both sides in the research process. Osterhammel sets out six alternative models for a theoretical global history, arguing, in effect, that none of them can work unless they are seen through a sociological lens, and that historical sociology is the best interlocutor: not only because it has a historical dimension by definition, and—importantly—not necessarily a teleological one, but also because it is less locked inside the ‘arid idiom’ of other strands of sociology. Sociology can add rigour and explanatory power to global history, while the latter can add fluidity and source-based problem-solving to sociology. O’Rourke is impatient both with the lack of interest in history in economics and with the parochialism of much economic history; he sees the advantage of global history precisely as being its capacity to undermine the current trend to national-focussed economic history, in that it recognizes that all states are part of larger-scale, and, already by 1600 (to an extent) global, economic systems. Conversely, economic analysis (he gives the example of price convergencies) can add rigour to historians’ vaguer analyses. These two are genuine challenges to historians’ assumptions; they also remind us that a key, essential, role for social theory of all kinds, in any analysis done by a historian, is its focus on the parameters of proper—again, rigorous—comparison.

From here, we move in a more ‘historical’ direction, and chronologically. Nicholas Purcell, Bob Moore, and James Belich discuss ancient and medieval global history. Purcell takes the ancient incense trade, which connected the Mediterranean, east Africa, and south Arabia (with links eastward from there) from at least the sixth century BC, as a model for what he calls ‘soft globalization’, following de Vries: as a set of more localized interrelationships through nodal points, with few if any traders moving across the whole trade route. He also stresses the degree to which this trade depended on culturally contingent assessments of value and rarity, and thus that economic and cultural history are less separable than many assume, not least when seen with a global perspective. Moore focusses on the importance of the urban revolution of the period 1000–1250 and, above all, the intensification of socio-political (p.22) relationships and socio-economic practices across the whole of Eurasia in the period 500–1500; he does not believe in the usefulness of the concept of ‘medieval’ when seen in global terms, but he pinpoints some common features of the ‘medieval’ centuries nevertheless. Belich offers a transnational case study of the effect of the Black Death on European socio-economic activity. He argues that the plague led to a surge in European commerce, and stresses the importance of multinational migrant groups such as sailors and mercenaries, in what he calls an ‘expansion kit’—ships, cannon, smallpox, racism, and risk-taking—which all facilitated European (including Russian, Ottoman, and Moroccan) advances outside the home continent after 1500.

Matthew Mosca and Francis Robinson discuss the early modern world. Both of them look at the issue of the global from non-European perspectives, Qing China and the Islamic world, arguing that each are pivots for understanding global relationships, and globalization in general, perhaps the principal ones until 1800. Mosca makes a strong case for the centrality of Qing China to the globe, given that it had the largest population in the world by far, given that it was the main centre of east Asian trade up to at least 1800 (and with extensions outwards from that), and given that it was the cultural focus for surrounding societies as well. He parallels some of Kenneth Pomeranz’s arguments here, although his focus and evidence-base is entirely different. Robinson stresses how the Muslim expansion between the eighth and the eighteenth century was the largest-scale—indeed, the most ‘global’ if one excludes the Americas—in human history before the eighteenth century; Islamic polities were the pivot of all long-distance economic, religious/legal, and scientific transnational connectivity in those centuries. After 1800, indeed, western conquest helped extend and entrench Islam (and non-religious cultural practices associated with it) still further, in India, Africa, Indonesia, and even America. These two contributions are excellent examples of the decentring of European-focussed historical metanarratives which has been one of the stated aims of global history from the outset; they show us the directions it can take in the future too.

The final two papers offer case studies in how to approach later modern global issues. Antony Hopkins argues against US exceptionalism in its imperial experience, and shows how the United States, particularly after 1898, in fact had just the same sort of empire and colonial experience that the European powers did. For her part, Linda Colley takes on the issue of constitution-making after 1750. Constitutions spread by ‘contagion’, largely to impress others, initially in the context of the American and French revolutions, then as an image of early nineteenth-century liberalism, and then, dramatically and competitively, everywhere in the world after 1860 or so. She shows clearly how not just economic history, but also political and institutional history, can and must be seen in transnational and global terms as well.

We entirely agree. As the foregoing makes clear, everything can be seen in global terms; and the possibilities and advantages of a global approach are as great in the medieval and early modern periods as they are in the nineteenth and twentieth centuries. The essays offered here show that the prospect of a complex, many-sided, global history of every period is bright.


(1) Fernand Braudel, The Mediterranean and the Mediterranean World in the Age of Philip II, 3 vols. trans. Siân Reynolds (Berkeley: University of California Press, 1995); Peregrine Horden and Nicholas Purcell, The Corrupting Sea: A Study of Mediterranean History (Oxford: Blackwell Publishers, 2000); David Abulafia, The Great Sea: A Human History of the Mediterranean (Oxford and New York: Oxford University Press, 2011); Cyprian Broodbank, The Making of the Middle Sea (London: Thames and Hudson, 2013).

(2) See ch. 6.

(3) Liu Xinru, The Silk Road in World History (Oxford: Oxford University Press, 2010), ch. 4; Liu Xinru, ‘Silks and Religions in Eurasia, c.AD 600–1200’, Journal of World History 6, 1 (1995), pp. 25–48.

(4) F. Nocete et al., ‘The Ivory Workshop of Valencina de la Concepcion (Seville, Spain) and the Identification of Ivory from Asian Elephant on the Iberian Peninsula in the First Half of the 3rd Millennium BC’, Journal of Archaeological Science 40, 3 (2013), pp. 1579–92.

(5) Richard L. Smith, Premodern Trade in World History (London and New York: Routledge, 2009), p. 85.

(6) Joel A. Tarr, ‘A Note on the Horse as an Urban Power Source’, Journal of Urban History 25, 3 (1999), pp. 434–48

(7) David Christian, ‘Silk Roads or Steppe Roads? The Silk Roads in World History’, Journal of World History 11, 1 (2000), pp. 1–26.

(8) Immanuel Wallerstein, The Modern World-System, 3 vols (Berkeley, Los Angeles, and London: University of California Press, 2011).

(9) Manuel Castells, The Information Age: Economy, Society, and Culture, 3 vols, (Oxford: Blackwell, 1996–8).

(10) Robert Finlay, ‘The Pilgrim Art: The Culture of Porcelain in World History’, Journal of World History 9, 2 (1998), pp. 141–87.

(11) Premodern Trade in World HistoryArmenian ReviewAmita Satyal, ‘The Mughal Empire, Overland Trade, and Merchants of Northern India, 1526–1707’ (unpublished PhD thesis: University of California, Berkeley, 2008).

(12) Eric Jones, The European Miracle: Environments, Economies and Geopolitics in the History of Europe and Asia, 3rd edn (Cambridge and New York: Cambridge University Press, 2003), p. 247.

(13) Prasannan Parthasarathi, Why Europe Grew Rich and Asia Did Not: Global Economic Divergence, 1600–1850 (New York: Cambridge University Press, 2011), p. 33.

(14) Michael MacKinnon, ‘Supplying Exotic Animals for the Roman Amphitheatre Games: New Reconstructions Combining Archaeological, Ancient Textual and Ethnographic Data’, Mouseion III, 6 (2006), pp. 137–61; Christopher Epplett, ‘The Capture of Animals by the Roman Military’, Greece && Rome 48, 2 (2001), pp. 210–22; Fik Meijer, The Gladiators: History’s Most Deadly Sport (London: Souvenir Press, 2004 [2003]); George Jennison, Animals for Show and Pleasure in Ancient Rome (Philadelphia: University of Pennsylvania, 2005 [1937]).

(15) Hugh R. Clark, ‘Frontier Discourse and China’s Maritime Frontier: China’s Frontiers and the Encounter with the Sea through Early Imperial History’, Journal of World History 20, 1 (2009), pp. 1–33.

(16) Peter Vanderford Lape, ‘Contact and Conflict in the Banda Islands, Eastern Indonesia 11th–17th Centuries’ (unpublished PhD thesis: Brown University, 2000).

(17) Kenneth Pomeranz, The Great Divergence: China, Europe, and the Making of the Modern World Economy (Princeton: Princeton University Press, 2000).

(18) Maxine Berg (ed.), Writing the History of the Global: Challenges for the Twenty-First Century (Oxford: Oxford University Press, 2013),

(19) Max Weber, ‘Die Objektivität sozialwissenschaftlicher und sozialpolitischer Erkenntnis’ (1904), in Max Weber, Gesammelte Aufsätze zur Wissenschaftslehre, 2nd edn (Tübingen, 1951), pp. 146–214, esp. pp. 190–200.

(20) Victor Lieberman, Strange Parallels: South-East Asia in Global Context, c.800–1830, vol. 2: Mainland Mirrors: Europe, Japan, China, South Asia, and the Islands (Cambridge: Cambridge University Press, 2009);Journal of Global History

(21) Walter G. Runciman, A Treatise on Social Theory, 3 vols (Cambridge: Cambridge University Press, 1983–97); for a detailed reaction, see Chris Wickham, ‘Systactic Structures’, Past and Present 132 (1991), pp. 188–203.

(22) This claim is efficaciously disposed of in a few pages in, for example, Pomeranz, Great Divergence, pp. 168–70.

(23) Kai-Wing Chow, ‘Writing for Success: Printing, Examinations and Intellectual Change in Late Ming China’, Late Imperial China 17, 1 (1996), pp. 120–57

(24) Craig Clunas, Jessica Harrison-Hall, and Yu-Ping Luk (eds), Ming China: Courts and Contacts 1400–1450 (London: British Museum Press, forthcoming 2016)

(25) Jan Lucassen and Leo Lucassen (eds), Globalising Migration History: The Eurasian Experience (16th–21st Centuries) (Leiden: Brill, 2014).

(26) Fernand Braudel, The Mediterranean and the Mediterranean World in the Age of Philip II, 2 vols, trans. Siân Reynolds (Berkeley: University of California Press, 1995), pp. 53–5; Jos Gommans, Mughal Warfare: Indian Frontiers and High Roads to Empire, 1500–1700 (London and New York: Routledge, 2002), ch.1.

(27) Alfred T. Mahan, The Influence of Sea Power upon History (London: S. Low 1890), p. 25.

(28) James Horsburgh, India Directory, or, Directions for Sailing to and from the East Indies, China, New Holland, Cape of Good Hope, Brazil, and the Interjacent Ports, 3rd edn, (London: Parbury, Allen and Co., 1827).

(29) Gagan Sood, ‘Circulation and Exchange in Islamicate Eurasia’, Past && Present 212 (2011), pp. 113–62

(30) John F. Padgett and Christopher K. Ansell, ‘Robust Action and the Rise of the Medici’, American Journal of Sociology 98, 6 (1993), pp. 1259–319.

(31) Tim N. Harper, ‘Singapore, 1915 and the Birth of the Asian Underground’ Modern Asian Studies 47, 6 (2013), pp. 1782–811.

(32) Marcus Bull, ‘Origins’, in Jonathan S. C. Riley-Smith (ed.), Oxford History of the Crusades (Oxford: Oxford University Press, 1999), pp. 15–34.

(33) Søren M. Sindbaek, ‘The Small World of the Vikings: Networks in Early Medieval Communications and Exchange’, Norwegian Archaeological Review 40, 1 (2007), pp. 59–74.

(34) Judith Scheele, ‘Traders, Saints and Irrigation: Reflections on Saharan Connectivity’, Journal of African History 51, 3 (2010), pp. 281–300

(35) James Gelvin and Nile Green (eds), Global Muslims in the Age of Steam and Print (Berkeley: University of California Press, 2014), p. 13.

(36) Michael R. Godley, The Mandarin-Capitalists from Nanyang: Overseas Chinese Enterprise in the Modernization of China 1893–1911 (Cambridge: Cambridge University Press, 1981), p. 13.

(37) Sushil Chaudhury, ‘Trading Networks in a Traditional Diaspora: Armenians in India c.1600–1800’ in Ina B. McCabe, Gelina Harlaftis, and Ioanna P. Minoglou (eds), Diaspora Entrepreneurial Networks: Four Centuries of History (Oxford: Berg, 2005).

(38) Nile Green, Bombay Islam: The Religious Economy of the Western Indian Ocean 1840–1915 (Cambridge: Cambridge University Press, 2011), p. 173.

(39) Lucien Febvre, La terre et l’evolution humaine (Paris: Renaissance du livre, 1922), p. 409.

(40) Matthew H. Edney, Mapping an Empire: The Geographical Construction of British India 1765–1843 (Chicago and London: University of Chicago Press, 1997), pp. 91–6; Ian J. Barrow, Surveying and Mapping in Colonial Sri Lanka (New Delhi: Oxford University Press, 2008), pp. 42, 81, 191.

(41) Scheele, ‘Saharan Connectivity’, p. 284.