‘A few years back, this people would have refrained, with superstitious horror and dread, from the infringement of a single Taboo ceremony, and never approached their morais or temples except with feelings of mingled terror and veneration. They now entertain sentiments as strong the other way, and despise the very idols which they till lately had adored … It may be fairly questioned whether a similar change might not be brought about, by the operation of similar causes, in other Pagan countries, – namely, a despotic monarch’s will! How feeble must be the tie where the ceremonial and not the doctrine constitutes the essential point of religious obligation!’Footnote ¹

This is the comment of a very young man, staying in Hawaii less than two months of 1822, which seems to mingle naïve Christian hope and rank prejudice. Yet Gilbert Farquhar Mathison, en route to the opium trade in China, could not have known that so many Pacific chiefs would prove him right, converting one after the other. Tahiti, it is true, had already fallen to the new faith from 1812, while in Hawaii itself, Mathison was here able to observe the effects of the remarkable abandonment of the traditional cultic system in 1819. But it was not until 1825 that the baptism of the high chiefs of Hawaii was achieved. Hawaii was followed by crucial chiefly conversions in Tonga (1830–1831), Samoa (1832) and Fiji (1854), all of them issuing in mass conversions among the broader populace. More remarkable than Mathison’s fortunate prediction is that his explanation for it touches upon something real. High chiefs and monarchs, even where they were regarded as akin to the gods, might suddenly turn their authority towards changing the very nature of the religious system that seemed to hold them in place.Footnote ² And this seems to have been connected with the kind of thing their religious system was, a matter of ceremony (or ritual, as a means to obtain worldly flourishing) as opposed to a matter of doctrine (or teachings, as a means of obtaining salvation).

The first of these gestures to a type of religiosity that may be described as ‘immanentism’ and the second to ‘transcendentalism’, both of which are explained in Chapter 1. A major dynamic of world history of the past two thousand years has been the way that the former has lost ground to the latter. Thus, we live in a world dominated by the salvific religions that may by defined by the transcendentalist element at their heart: especially, Christianity, Islam, Buddhism, and some forms of Hinduism. And a principal mechanism by which this was accomplished was the conversion of rulers whose subjects were – however swiftly or gradually – converted in turn. After military conquest and the sheer expansion of empires, this may be the most important reason why the religious map of the world today looks the way it does.

How could this be? Everywhere in the premodern world, societies conceived of rulership as a sacred office. We might imagine, then, that rulers would always jeopardize their legitimacy if they tried to convert before their subjects had. ‘Conversion’ is understood here as the public announcement of a new religious commitment, most emblematically through baptism, although this was in fact just one part of a much longer process starting with mere toleration or patronage of missionaries.Footnote ³ Indeed, the most acute risks for jeopardising legitimacy tended to occur not at the point of baptism but when rulers sought to abandon the patronage and performance of the traditional rites. All this is quite intuitive: Given that we have long understood – at least since Durkheim, and in many ways long before – that religion in the premodern world was a vital means by which social groups were formed and political orders were cohered, the prospect of a ruler stepping outside the religious framework by which their subjects understood themselves and the world around them seems somehow … absurd.

I have already addressed this question in a companion volume, Unearthly Powers (UP), which proposed a range of conceptual tools for helping us think about religion and its relationship with politics over the long term of world history. The bulk of the present book adopts a very different kind of comparative approach by taking four case studies and analysing them in detail. The first is the conversion of the kings of Kongo following their encounter with the Portuguese: Nzinga a Nkuwu received baptism in 1491 and then apostatised but was followed by his son Afonso, who succeeded to the throne in 1506. This was the first major royal conversion secured by Iberians as they began their seaborne expansion and helped encourage the notion that other princes and potentates in the world beyond would follow suit. Afonso established an enduringly Christian dynasty in Central Africa, and the religion would be a significant element of the regional culture for a long time to come.

The second case is that of certain warlords in Japan in the later sixteenth century, at a time when the country was reaching the climax of the sengoku or ‘warring states’ period and such lords enjoyed substantial authority over their territories – they were referred to by the Portuguese as ‘kings’. The three earliest to convert were located on the southern island of Kyushu, which was particularly open to the newly intensive commercial and cultural exchanges of the early modern period. The most important convert, indeed one of the most powerful players in Kyushu, was Ōtomo Sōrin of Bungo (1530–1587), who finally received baptism in 1578. This was the most important breakthrough for Christianity among the ruling elites of major Asian societies. Yet it is no less important to understand why these Christian daimyo only ever formed a distinct minority among their peers, and why, when the sengoku was drawn to a close by the unifying rulers of Japan towards the end of the century, Christianity was reprehended and then persecuted.

The third case concerns the attempts of the French court to convert King Narai of Siam (r. 1656–1688), then at the height of his power as the ruler of the large and cosmopolitan kingdom of Ayutthaya and held aloft by an elaborate system of royal ceremonialisation. The French chased his soul through a series of extravagant embassies dispatched and received in the 1680s only to find that their efforts precipitated a ‘revolution’ in 1688 that saw the French ejected, Christians briefly persecuted and a Buddhist order trenchantly reaffirmed.

The last case is that of Hawaii, or the Sandwich Islands as the British termed the archipelago, in the first decades of the nineteenth century. Here, the whole of the cultic system seemed designed to promote the power of the ruling chiefs, for Hawaii under the conquering leader Kamehameha was fast making the transition from chiefly competition to paramount kingship, and its leading chiefs or ali‘i were treated as divine beings. But when Kamehameha died in 1819, his son and successor Liholiho consigned that system to oblivion through a drunken but highly deliberate act of desecration: He broke the eating tabu by sitting to eat with women – a group of high-ranking female chiefs – during the rituals of his installation. He then ordered the destruction of the temples. The real agency here, however, belonged to the widow of the dead king: Ka‘ahumanu. She considers us confidently from the front cover, painted a few years before the revolution began.Footnote ⁴ After the collapse of the eating tabu, Ka‘ahumanu toured the islands, pulling down ‘idols’, and then began moving towards Christianity, culminating in her baptism in 1825. The Christianisation of Hawaii was thereby secured.Footnote ⁵

The final chapter of this book moves away from these detailed investigations into a quite different genre of scholarship in order to make sense of patterns of ruler conversion across the rest of the world. Gathering together all this material, the following three factors stand out as the most consistently significant in inducing rulers to convert; they may also be thought of as stages to the extent that they often followed chronologically.Footnote ⁶ (1) Religious diplomacy. Rulers tolerate and favour Christianity because of the enticements of diplomatic advantage, military assistance and commercial profit. (2) Accessing immanent power. The breakthrough to a more substantial incorporation of the new cult is often connected to its perceived ability to provide supernatural assistance. (3) Enhancing authority. The Christianisation of the realm may then be seized upon as a means of enhancing the authority of the ruling dynasty.

These factors tended to be present regardless of the social or cultural setting. This observation does not preclude us attending to a variety of other dynamics as we move from case to case, and, indeed, two further factors, intellectual appeal and cultural glamour, are briefly considered in Chapter 1 and from time to time in the chapters. But, by itself, this model does not help us get much further in explaining global variations in the spread of the world religions. To do this, we need to return to the question of what kind of thing ‘religion’ was in the societies that received proselytisers. Thus, a range of more particular questions opens up. Why is it that in some societies displays of immanent power by an incoming cult seem to threaten the pre-existing traditions more gravely than in others? Why is it that the prospect of conversion risks fundamental damage to the ruler’s political legitimacy in some cases, while, in others, taking this step might even enhance their status?

It turns out that examining moments of ruler conversion allows us to turn a spotlight upon questions that are intrinsically difficult for the historian to investigate: How does religion structure political possibility, exactly – not as matter of law or explicit theory but as a sociological reality? How different are the forms that ‘sacred kingship’ may take? Was legitimation a significant concern for pre-modern rulers? If so, did the sensibilities of the wider population matter or only those of the ruling class and potential challengers? How able were contemporaries to articulate such matters explicitly? In turning upon such questions, the analyses offered here cast light on broader aspects of each of the societies examined, as well as forming the basis upon which the larger picture of global religious transformation may be assembled.

The Comparative Approach

A range of considerations has governed the selection of cases. In order to ensure that each one sheds light on local religious–political dynamics, they had to turn on the calculations of an essentially independent ruler endowed with genuine agency as to whether or not to convert. They were selected in order to consider two ‘immanentist’ regions (Kongo, Hawaii) and two with established ‘transcendentalist’ traditions (Japan, Thailand), while each pair contains different kinds of success and failure. Kongo (Chapter 2) is a striking success story for Christianity following surprisingly slight contact with the Portuguese but was not followed by many equivalent cases across sub-Saharan Africa in this period.Footnote ⁷ The conversion of the high chiefs of Hawaii (Chapters 7 and 8), by contrast, is representative of the wider region. One of the earliest acts in the drama of chiefly conversion spreading across the Pacific, it is particularly intriguing for the unique way that the abandonment of the pre-existing cultic system was disaggregated from the conversion to Christianity – allowing us to speculate on certain vulnerabilities of immanentism. The warlords of Kyushu in Japan (Chapters 3 and 4) present us with the only exceptions to the otherwise uniform intransigence of major Confucian and Buddhist rulers towards monotheism, while the subsequent repression offers clues as to how those broader regional obstacles may be conceived. Meanwhile, doomed attempts by French Catholic (and also Persian Muslim) missionaries to convert Narai of Siam (Chapters 5 and 6) represent an unusually source-rich illustration of the ways in which monotheism met a brick wall in the form of Theravada Buddhism. In this way, each case has also been selected in order to speak to larger regional patterns. Even this brief description indicates that religious systems are granted a kind of structural force in this analysis.Footnote ⁸ Yet this is evident only as one of many interlocking factors, without diminishing the role played by local, cosmopolitan and evanescent forms – which the specific focus of each chapter indeed affords. If, in Chapter 10, there is an attempt to explore how far transcendentalist religions were more likely to obstruct ruler conversions, the major case studies are not set up to merely confirm that hypothesis but to challenge it in various ways.Footnote ⁹

The different nature of the source material for each case, and the rather specific questions they each throw up, meant that it proved more efficient to consider them one by one. But I hope readers will perceive that, while the narrative unfolds differently in each case, the discussion deploys the same conceptual framework in order to test the same set of propositions. As with any methodology, comparative analysis has its problems, but the opportunities it affords are receiving somewhat more attention of late.Footnote ¹⁰ We are, for example, pushed into a more critical deployment of our theoretical apparatus by the process of ‘reciprocal comparison’, according to which starting concepts are transformed as they are forced to incorporate the stubborn particularity of each case in turn.Footnote ¹¹ It is due to this imperative to be self-reflexive and explicit about our analytical terms – which cannot reflect the emic register of any one case – that UP coined some new terms and gave specific definitions to more familiar-looking ones. Chapter 1 and the Glossary summarise these conceptual interventions.

The chief advantage of the comparative method is that it also allows us to make progress in identifying the most significant lines of causation.Footnote ¹² To be sure, pretensions towards scientism must be held in check; we are not, of course, in a position to test any one variable by holding the rest constant. The terminology of variables is itself somewhat artificial; in the study of history, independent variables are a chimera and we are compelled to deal with what John Llewis Gaddis referred to as interdependent variables.Footnote ¹³ It turns out, indeed, that the Buddhist theory of causation as developed by the fifth-century exegete of the Theravada tradition Buddhagosa had already articulated something like this quite exquisitely. Causation is to be understood as ‘a concatenation of several potentially causative factors which are activated as causes only when they enter into a specific and opportune relationship with each other’.Footnote ¹⁴ What we may choose to refer to as ‘causes’ are not determinative in themselves but rather function as what Philippe Buc calls ‘conditions of possibility’.Footnote ¹⁵

From one angle, this condemns the comparative method to inevitable inexactitude, an inability to produce truly robust and parsimonious principles. From another angle, it rescues it – because we are compelled to appreciate that any one ‘condition of possibility’ or ‘variable’ may not lose analytical relevance merely because its causal outcomes diverge somewhat from case to case.Footnote ¹⁶ This sounds both abstract and messy, but the truth is that we can watch how conditions or forces interact in a more concrete way in any given case study. If we use the Great Divergence debate in economic history by way of an example, we might want to identify the fact that coal was relatively accessible to existing population centres in Britain as a major reason why the Industrial Revolution had its origin there.Footnote ¹⁷ But this was not an independent cause: It was rather a structure that was only ‘triggered’ into critical relevance once technological and capitalist development allowed this coal to be profitably exploited. The presence of carbon-rich deposits underground was of little significance until it suddenly became the axis on which the world turned.

There are, then, no laws of human history.Footnote ¹⁸ But, surprisingly often, recognising the futility of identifying laws seems to slide into denying the possibility of spotting patterns or tendencies (which is the best we can expect if we are dealing with interdependent variables). If one or two counter-examples will puncture a law, assessing the utility of a pattern requires a more holistic perspective. Still, comparative historians must always face the question of how representative a small number of cases may be taken to be. It is partly in order to ameliorate this that Chapter 10 seeks to identify patterns across much of the rest of the world. Naturally, this cannot involve the kind of primary research and close investigation devoted to the case studies, and thus the relevant mechanisms involved can only be far more generally and speculatively evoked. But scaling up to the global level allows us to see how the basic propositions might work in contending with a much broader array of instances and problems. In order to assist the global perspective, the question of how Hinduism and Confucianism relate to the construct of transcendentalism is considered in Chapter 9. The original conception of this project would have involved the spread of Islam as well as Christianity, which would have allowed for a more complete view of the victory march of the monotheisms. While I have attended to Islamisation in a few other publications, simple practicalities militated against substantial consideration in this book.Footnote ¹⁹

The Sources

The case study chapters all concern important events in their respective national histories, and many generations of scholars have written about them.Footnote ²⁰ I set out to use the vast majority of sources that region specialists had already identified; this is not, then, primarily an archival project. As ever, asking new questions of sources allowed them to speak new things. Still, the source basis remains starkly imbalanced, heavily dominated by European voices and largely missionary ones at that. In the case of Kongo, there are simply no relevant sources in local languages, forcing us to rely on chronicles and letters written by the Portuguese, or by later Italian Capuchins. However, we do have some long and highly significant letters written by the principal Kongolese protagonist, King Afonso, drafted in Portuguese with the help of a trusted scribe, João Teixeira, a local nobleman who had studied in Portugal and returned to the court.Footnote ²¹ These are vital for conveying a perspective that often differed markedly from that of the Portuguese on the ground. Indeed, John K. Thornton has remarked that this correspondence ‘makes Kongo in this period one of a very few African kingdoms … whose history is documented by its own people’.Footnote ²² Hawaii was another society without a pre-existing tradition of literacy, although in this case we have vital texts containing rich ethnological material written by Hawaiians such as David Malo (1795–1853), John Papa ʻĪʻī (1800–1870) and Samuel Mānaiakalani Kamakau (1815–1876).Footnote ²³ These texts bring significant problems of source criticism in their train, as their authors were writing with some considerable hindsight and after they had become Christian – but they had been alive during the events in question and even involved in them.Footnote ²⁴ The majority of sources are by American and British visitors and missionaries, although they do not, by any means, speak with one voice.

Japan and Ayutthaya represent quite a different scenario, given their long histories of literacy. But evidently missionaries were far more interested in recording the details of the daimyo conversions than were the Japanese, to judge by the preponderance of evidence: The extensive correspondence left by the Jesuits and the particularly voluminous chronicle of Luís Fróis forms the mainstay for any work in this area.Footnote ²⁵ Nevertheless, these may be supplemented by a number of letters and other sources written in Japanese that provide more than context; they yield independent confirmation or contestation of certain points emerging from the Jesuit evidence. They have been used here through the assistance of several translators. Turning to Thailand, the affairs of the royal court of Ayutthaya in the 1680s produced an extraordinary wealth of European letters, treatises, and travel accounts written largely in French, but the destruction of the archives of Ayutthaya after they were sacked by the Burmese in 1767 has led to a dearth of corresponding evidence in Thai. Still, some important Thai sources have remained that help us to flesh out certain aspects of Narai’s court and its worldview, including some royal decrees, poetry, temple murals, and the royal chronicle of Ayutthaya (although the section dealing with Narai’s reign from 1663 contains multiple recensions all deriving from the late Ayutthaya or Bangkok era and driven by the dynastic politics of that time).Footnote ²⁶ Thus, for all the cases, it was a working method to seek out non-European sources as much as possible in order to mitigate the fundamental source imbalance; to work closely with translators and regional specialists (who also read drafts of each chapter); to use, occasionally, later ethnographic evidence where it might help (with due regard for the serious methodological problems entailed); and to cultivate an awareness of the predispositions and blinkers of the European evidence.

It should go without saying that missionary depictions of the societies they encountered and the conversion processes they (sometimes) stimulated must be approached in a highly critical spirit. Their epistles often had an exhortatory and edificatory function, with many consequences for their representation of events. Almost as a rule, wishful thinking and providential reasoning combined to overestimate how close local rulers were to making a commitment to Christianity; frequently, what was essentially the play of religious diplomacy was taken as a sign of spiritual inclination. Accounts of the role of the individual’s heart and mind were liable to be so overdetermined by standard expectations of what should happen that the reader is left feeling rather unenlightened about what really did happen. On the other hand, claims about the eruptions of supernatural power into the mundane world (miracles, magic, divine interventions), which are frequently set to one side by historians, are taken seriously here as indications of a consequential logic.

Ethnologically, our sources are hobbled in many ways. Some examples would include the use of a homogenising language of kingship to describe diverse political forms; an inability to grasp hold of non-monotheistic configurations of religious identity; and a patriarchal rendering of female agency. The blanket category of paganism frequently served as a bar to more subtle appreciations of local religion – even if it is revealing as an expression of transcendentalist alienation from the sensibility of immanentism.Footnote ²⁷ At the same time, we must take note of the whole range of cross-cutting agendas and personal proclivities that might structure missionary sources, and which ensure that light may be cast on events from various angles. Protestant missionaries in the early nineteenth century Pacific were entirely predictable in their antipathy to the ‘savagery’ of the paganisms they encountered, but their dialogic anxieties could be as much stimulated by inter-missionary debates as any desire to advertise pagan iniquity. One of our most important sources for Hawaii is the American missionary Hiram Bingham, a puritan and sombre writer if ever there was one, but whose emotions seem to be even more exercised by the spectre of atheism than by the tyrannies of the local tabu system – for which, indeed, he develops a strange sympathy. Bingham’s particular concerns means that he is antagonistic towards a range of propositions concerning the conversion of the Hawaiian ruling elite – that it set off a direct process of top-down conversion, that missionaries were critical elements in political reconfiguration, that autonomous developments in Hawaiian society were ultimately responsible, and that a decisive role was played by the arrival of a group of Tahitians accompanying William Ellis – which his own evidence nevertheless provides reason enough to endorse.