A Multitude of Indigenous Domains

To investigate this hypothesis, I consider the history of the eastern side of the island of Timor, a colonial possession of the Portuguese Crown between the 1500s and 1974, and the independent nation of Timor-Leste since 2000. Portuguese traders and missionaries first visited the island in the sixteenth century, while in the seventeenth century, through the mediation of a powerful mestizo class known as the Topasses (or “Black Portuguese”), the Portuguese Crown expanded its military power and influence from Flores and the Solor archipelago to most of the island. Portuguese expansionism was soon limited, however, by competition with the Dutch, who had been firmly established in Kupang since the 1650s, and who subsequently gained control of the western half of the island. Established in Dili since 1769, Portugal’s domain in Timor then became reduced to East Timor and the region of Oecussi. Throughout the 1800s, therefore, the focus of successive governors was on strengthening the colonial grip on the eastern side of the island by seeking to extend Portuguese military power at the same time as improving the capacity of Dili to concentrate the flows of value to be extracted from indigenous land and peoples, while a multitude of indigenous domains outside Portuguese control sought to block these colonizing dreams.

While Timor-Leste is currently presented as a country marked by great cultural, ethnic, and linguistic diversity, nineteenth-century conceptions of East Timorese diversity revolved mainly around a patchwork of many highly autonomous domains or polities of various sizes, referred to by the Portuguese term reinos (kingdoms), and headed by what were known as Timorese reis and rainhas (kings and queens).⁴ The colony was widely perceived by Portuguese governors and military as a crowd of many apparently disjointed and uncontrollable reinos, and as a hostile web of indigenous collectives, saturated with internal conflict, viciously inclined to enmity, and eager to take over the governor’s position in Dili. This threatening and untamed multitude of socio-political others-largely independent, but also ambivalently linked-was ordinarily perceived as traditional and characteristically “native”; a universe over which the governors felt they had little or no control politically, militarily, or epistemically. This notion hindered colonial dreams of sovereignty, control, and exploitation of people and land, with these imageries of an unruly indigenous polity being ingrained in the self-perceived vulnerability of the small community of white colonizers, most of whom spent their entire appointments in the Dili district (^{Roque 2010}b). Thus, nineteenth-century Portuguese were captivated by, though also fearful of, the multitude of reinos that surrounded Dili, the seat of rule.

The emphasis on plurality and alterity was accompanied by a belief that this multitude also formed one distinctive and coherent reality. Hence the Portuguese engaged with the indigenous polity as an entity of colonial thought and action distinctive for its radical display of hazardous diversity. This perception, in turn, with its accompanying feelings of fear and desire, was a catalyst of intellectual labor. The questions of how, or even whether, this multitude could be known, governed, and tamed for the empire’s sake was a matter of administrative concern. Making the ethno-political multiple comprehensible to and governable by the colonial state required a suitable work of knowledge, to which end various Portuguese colonial writers penned reports, letters, book sections, articles, maps, lists, and tables about Timorese reis and Timorese reinos.

Reinos as Units of Colonial Knowledge

The main materials of my study are Portuguese writings on ethno-political diversity. The relative presence or absence of this subject in early modern Portuguese knowledge is still difficult to ascertain. Yet, the Portuguese accounts of Timor in the sixteenth century seem to refer mainly to ports of call, coastal regions, and potential trading partners or allies, while missionary records tend to focus on conversion efforts aimed at the perceived members of the ruling classes. Historian Hans ^{Hägerdal observes (2012}: 24-25, 51-52) that the early modern European references to the “traditional” organization of Timorese powers look incidental and fragmentary, even though an awareness of a multitude of indigenous polities is occasionally identified. Hägerdal (2012: 52) specifically notes a scarcity of information about Timorese polities in Dutch records, and concludes that the Dutch interest in indigenous “kingdoms” declined from the late eighteenth century onwards.⁵ Portuguese nineteenth-century writings suggest, by contrast, that the reinos and their kings and queens were seen as the main units for governance and for organizing information about Timor from the time when the Portuguese took firm steps to establish a governor appointed from Goa in the early 1700s. From the late 1700s onwards, in particular, an increase also seems to have been seen in the amount of information produced on Timorese polities and specifically the eastern part of the island. This probably reflects the Portuguese government’s decision in the late eighteenth century to shift the colonizing focus from west to east.

In 1769, unable to cope with Dutch and Topasse opposition in West Timor, the Portuguese governor moved the seat of rule from Lifau to Dili, a port on the northeast coast of Timor, after reaching agreement with the rulers of the Motael domain, where Dili was situated. This may have led to interest in producing knowledge specifically about the eastern Timorese reinos, at the same time as the issues of control and management of the East Timorese polities were becoming increasingly important. Portuguese investigations of indigenous polities subsequently gained momentum with the period that started with the Dili settlement being marked, I believe, by a kind of reinos- and reis/rainhas-centered government activity in East Timor. This largely continued until 1911-12, when Portuguese government forces confronted and defeated a massive anti-Portuguese coalition of East Timorese reinos headed by the King of Manufahi in what became known as the Manufahi War or Manufahi Rebellion. This event marked a turning point in the local balance of power between Dili and the kingdoms and was followed by administrative changes that significantly reduced, without completely eliminating, the importance of the reinos and royal lineages. The chronology framing my analysis-the long nineteenth century between circa 1769 and 1912, the Portuguese foundation of Dili, and the Manufahi War-is aimed at exploring this hypothesis. I suggest that the theme of indigenous ethno-political diversity was central to the production of colonial knowledge during this long period. The East Timorese world was principally addressed by colonial administrators as an ethno-political multiple and as an object of both political and epistemic concern, materializing in a variety of enumerative studies as well as in works of historical and ethnographic tenor.

An almost structural tendency of indigenous Southeast Asian states in the region to form volatile, fluid, competitive, and plural polities fragmented into petty domains has been widely identified and discussed by recent scholarship sensitive to indigenous cultural idioms (^{Colombijn 2003}; ^{Bentley 1986}; ^{Christie 1995}; ^{Hägerdal 2016}). Centrifugal tendencies towards fragmentation in Southeast Asian polities developed in a tensional relationship with centripetal tendencies towards the formation of broader state realms, in which ritual centers exerted symbolic and ceremonial sovereignty over smaller polities, based on ritual, tribute, myth, and spiritual power (^{Tambiah 1977} and ^{Geertz 1980}). The question of whether the traditional political systems in the Sunda Islands, including Timor, tended towards unification or fragmentation before the arrival of Europeans has long concerned historians perusing colonial records in search of traces of pristine indigenous pre-colonial domains (^{Forman 1977}; ^{Schulte-Nordholt 1971}: 165-85; ^{Fox 1982}; ^{Hägerdal 2016}, ²⁰¹²: chapter III). But these domains hardly come in such pristine form in the archives. For instance, it has to be acknowledged that European outsiders were not simply external to the Southeast Asian dynamics registered in the archival records themselves. Instead, and over the course of a few centuries, Portuguese and Dutch agents alike became active players in the indigenous political field, thus themselves becoming enmeshed in the tensions inherent to the centrifugal and centripetal forces of the Southeast Asian polities, and, in several instances, transforming these polities with their presence along the way. Timor offers an exemplary instance of this hypothesis as Portuguese and Dutch rivalry and exchanges with indigenous domains acted upon the shape and content of the local polity, resulting in complex entanglements between European and indigenous formations coming into being (^{Roque 2011}, ²⁰¹⁰a; ^{Hägerdal 2012}).

Historical and anthropological scholarship has thus approached the problem of the traditional order of indigenous states through the lens of diversity and fragmentation, with colonial texts being mined as sources of data about this diversity for the purposes of constructing an ethno-history of the pre-colonial order. Yet, this orientation runs the risk of overlooking the fact that indigenous ethno-political diversity itself is a construct of European texts, which, from the outset, framed this diversity as a problem of both power and knowledge. It may thus be worth recalling that this subject began its intellectual career as a privileged subject of investigation in the context of European problems of colonization. In other words, there is a genealogy of current ethno-historical debates about “traditional” political organizations with roots in the power- and vulnerability-obsessed investigations of European colonizers, such as those of Portuguese colonizers in East Timor, and it is these debates that I propose to examine further.

Enumerative, Historiographic, Ethnographic

I argue that the double resolve to both seek protection from and attain control over ethno-political diversity was critical to Portuguese colonial administrators in Timor. This, in turn, translated into a set of colonial studies that took on the diversity of reinos as the principal focus of their inquiry and governance. The primary units and subjects of this ethno-political government were “kingdoms” and royal lineages, rather than “tribes,” “races,” “populations,” or individuals. Managing a diversity of human polities, rather than just managing a diversity of human bodies, thus became a fundamental dimension of administrative thought and action. This included consequential attempts made since the 1860s by governors such as Afonso de Castro and Celestino da Silva to set up a labor extractive coffee-plantation system in the reinos with the involvement of indigenous royal classes (^{Shepherd and McWilliam 2013}). In this context, colonial governmentality assumed a strong ethno-political orientation, one that perhaps combined with bio-political possibilities concerning the government and discipline of living bodies (^{Foucault 2010}). Hence one may speak of a colonial will to generate a kind of ethno-political power-that is, a power able to sustain, control, and regulate the unfathomable diversity of the indigenous polity or polities, to command its multiplicity, and to put it in order. I am led to think of a colonial will to exercising power over the indigenous polity that was at the same time overrun by fear, anxiety, and disquiet. This will to ethno-political power, in other words, was a function of the will to protect oneself from the perceived dangers of a multiple polity.

Nineteenth-century Portuguese documents shed special light on these knowledge efforts in East Timor by allowing for an analysis of the main distinctive investigative modalities characteristic of this historical process.⁶ Without dismissing the significance of knowledge efforts made during earlier periods (some evidence of which I discuss, albeit only briefly, in this article), my analysis focuses primarily on the long nineteenth century (1769-1911) and especially the period beginning in 1850. This decade seems to have marked another local turning point in the administrative reorientation towards East Timor, including events such as Governor Lopes de Lima’s controversial decision in 1851 to give away the centuries-old Portuguese establishment of Flores to the Dutch (without Lisbon’s approval) in exchange for monetary compensation. This event was met with indignation in Portugal and followed by a new wave of intensity regarding the administrative focus on the eastern half of Timor. The 1859 appointment of Afonso de Castro, a naval officer who had been involved in the intricate diplomatic negotiations that followed the sale of Flores to the Dutch, to the position of governor with considerable autonomy was a manifestation of this moment of imperialist impetuosity. As we will see, Castro employed a research style of governing ethno-political diversity in East Timor that later governors saw as referential for their own practices. In effect, and as discussed below, Castro’s governorship not only intensified colonial concerns about managing ethno-political diversity, but also prompted the development of strongly historiographic and ethnographic styles of studying and writing about the reinos besides the hitherto conventional enumerative investigations.

My purpose here is to analyze three principal ways of investigating the ethno-political multiple in East Timor that, I believe, developed during that longer timespan. I refer to what I conceptualize as the enumerative, historiographic, and ethnographic “investigative modalities” of colonial knowledge, to use a term from Bernard S.^{Cohn (1996}: 5). Each of these modalities, Cohn suggests, implies distinct bodies of information, modes of intellectual inquiry, techniques, and written formats and genres, and this article examines the historical unfolding of these knowledge approaches in the governance of Timor during the 1800s. Enumerative routines seem to dominate in early documents, while a historiographic orientation pervades much of nineteenth-century writing and ethnographic accounts of scientific intent gain a strong presence at the turn of the twentieth century. Yet, as I clarify, these are analytical concepts that refer, in concrete artifacts and historical situations, to changing practices that sometimes coexisted or appeared to be blended, even if their emergence was necessarily contextual and may have followed a certain diachronic sequence.

The first section of the article looks at what I refer to as the enumerative or synoptic modality. This mode of researching ethno-political diversity was censitary in approach and expressed in tables, counts, charts, lists, and other literary and numerical abbreviations. I then examine a mode of investigation that was historiographical in orientation and style, and that focused on considering the historical past of the indigenous polity, especially in relation to the Portuguese colonial presence. Finally, I explore an ethnographic investigative modality that gained momentum at the turn of the twentieth century and that focused predominantly on what was seen as the traditional realm of uses and customs in the kingdoms.

Putting Reinos on a List

Ethno-political fluidity, independence, and diversity were frequent subjects of an early mode of censitary studies in the eighteenth and nineteenth centuries. Portuguese governors produced lists, maps, tables, and charts in which certain relevant information on every reino (or a lack of such information) was enumerated, abbreviated, and condensed. These writings seemed to configure an early mode of colonial census and statistics focused on ascertaining the indigenous polity as a diverse unit. Evidence of how these knowledge technologies were used in Timor can be traced back to at least the early eighteenth century, when governors routinely collected data and organized tentative official inventories and lists (called listas or relações), as well as synoptic tables (designated mapas), that systematically enumerated the names of reinos and their respective kings or queens, their vassal situation and their degree of loyalty or enmity, the potential number of men at arms and, lastly, their estimated wealth expressed in terms of potential agricultural produce, land resources, or taxes and tributes in kind, or in the form of warriors or laborers, owed as an obligation to the Portuguese governor, as the king’s representative, under the vassalage treaties.

Imagined land wealth, military potency, and political diversity: this combination of elements in the imageries of Timor can already be discerned in one of the earliest accounts of the region, provided by the famous Luso-Malayan scholar Manuel Godinho de Eredia in the early 1600s. Eredia (1930 [1613]: 254) referred to Timor Island as an “Empire” divided into “two parts” and mentioned that “besides the Empires … many powerful Kings” existed “who have amassed a great deal of Gold” and sandalwood. The concern with self-survival when faced with a multitude of competing kings, and the desire to extract indigenous resources in the form of men and land through a tribute system based on vassalage ties, traverses the enumerative writings produced by Portuguese administrators in the eighteenth century. In the early 1700s, for example, attempts to reform the tribute system known as finta (under which vassal kings were obliged to pay an annual tribute to the Portuguese) were accompanied by lists made “of the kingdoms, kings and [finta: tax] amounts owed” to the state both in the past and in the present (^{Macedo 1737}; ^{Castro 1867}: 225-27, 260-63). In 1811, in instructions issued to the new governor of Timor, the Viceroy of India, the Count of Sarzedas (1811: 200), gave the official count of forty-six reinos in the Belos province (East Timor), stating they were “all free and independent from each other” and adding, in reference to earlier “lists ordered in 1722 and 1725,” that the reinos altogether comprised “forty-six thousand armed men, three thousand rifles and more men with sword, shield, spear and bow and arrows.”

This suggests that the practice of enumerating ethno-political diversity on paper lists and synoptic maps was a routine of eighteenth-century Portuguese governors in Timor. Tables, counts, inventories, and lists became the lasting and favored administrative formats for representing the ethno-political multiple. This genre persisted throughout the nineteenth century, when various newly produced tables and inventories of reinos and reis continued to reflect two related concerns: the urgency of mapping the distribution of indigenous rulers’ hostility and friendliness, and the fixation on the real or imaginary wealth of the reinos. The promise of future colonial prosperity, contrasting with the chronically impoverished present of colonial finances, drove governors incessantly to seek to enumerate factual or supposed riches in the reinos. As an army officer affirmed, Timor was “an island fertile with riches” (ilha prenhe de riquezas), but inaccessible to the Portuguese except for the “insignificant” fintas paid by “different kingdoms” (^{Garcia 1870}). Colonial imaginaries of the dangers of the ethno-political multiple were accompanied by colonial fantasies of prosperity. If the wild and plural polity could only be brought under control as the Portuguese governors wished, taxes and land resources would in future flow abundantly from the reinos into the colonial coffers.

Portuguese Ignorance, Timorese Mediations

Governor Afonso de Castro’s studies, though predominantly historiographical, also bear traces of this enumerative approach. In As Possessões (1867), for instance, we find colonial tables of reinos from distinct epochs, as well as a rare cartographic attempt at visualizing the diversity of Timorese polities in the form of a geographical chart (carta geográfica) roughly displaying the location of the reinos in East Timor.⁷ The manuscript entitled Map of the Kingdoms in the Timor District (^{Anonymous ca. 1870s}), produced around 1878-1880, is another revealing example of the continuing drive throughout the nineteenth century to produce enumerative knowledge (^{França 1897}: 236; Dores 1903). This “map” (in reality, a synoptic table) counted fifty-one reinos and listed the name, status title, and military rank of every current king or queen at the head of every named reino, while also classifying these rulers as “vassals” or “rebels,” and lastly enumerating the mineral and agricultural “products” of each reino, including reference to its spoken “dialect.” This table is valuable also because it makes manifest the colonial officials’ difficulties in gathering reliable and systematic information for these lists, with several sections of the table deliberately left blank or containing dots to indicate a lack of information. In this table, the categories to which eight reinos should be attributed were evidently unknown as only their names were specified, while five “vassal” kings were left nameless. The author of the map explained the missing data as follows: “The kingdoms with just the dots are those of which the names are ignored; they are neither rebels nor vassals, and very distant from the seat of government [praça] in Dilly.” This acknowledgment, the blank spaces displayed, and even the anonymous authorship of this manuscript begs the important question of how the governors gathered their enumerative (let alone any other type of) intelligence about reinos in practice. Of course, Portuguese governors and officials tended to assume authorship of these texts, and consequently any possible agency of local Timorese agents, is hidden or left unmentioned. Sadly, then, the documentary evidence available is very sparse, especially with regard to the possible role of Timorese agents as either authors or intermediaries in the making of these knowledge artifacts. However, a very few instances, to which I refer below, hint at the mediation of such agents and suggest a significant role being played by certain members of the Timorese higher classes or royalty as providers of information that was included in lists and tables, and also of accounts of more historical and ethnographic importance about the kingdoms.

As well as being a problem for Portuguese colonizers fantasizing about full domination and extraction of value from the indigenous world, the intractable “polity of many” could also be of concern for activities relating to religious conversion. In 1883, for example, Father João Gomes Ferreira attempted to count the number of people living in the kingdoms. The result was a summary censitary table, listing the “approximate number of [pagan] souls” and the “number of Christians” according to “reino or locality” and “language spoken” (1902: 131). Ferreira’s table is also relevant for another reason as, unlike most enumerative lists and synoptic works (which are usually silent on the processes of data-gathering), his text bears traces of a significant role for Timorese agents as mediators and informants. Ferreira claimed, however, that his count was “unreliable” “because when one interrogates the régulos [Timorese kings] in order to know the population in their reinos, they hesitate and give us lower [figures]” (1902: 131). Reliance on Timorese kings, nobles, and persons of status as intermediaries may have been a routine procedure in synoptic inquiries as the same type of Timorese royals and nobles accused by the Portuguese of acting treacherously, or independently of the government, often became the sources of data about the reinos themselves. This being the case, even the labor invested in knowing the ethno-political multiple was caught in the indigenous webs from which the colonizers wished to escape.

Writing Luso-Centric Pasts for the Reinos

Enumerative writings tend to offer portraits of the ethno-political multiple that are frozen in time, as if the reinos, though constantly in flux, were indifferent to historical change. However, it would be wrong to claim that colonial inquiries were simply oblivious of temporal dimensions of this polity: issues regarding the past of the reinos were also of the essence. The same authors who prepared lists could also philosophize about the past of the Timorese polity, conjecturing, in particular, about the configuration of this polity prior and in relation to the Portuguese presence. Historiographical reflections were thus an intrinsic part of colonial attempts to get a grip on the reinos. As a rule, this colonial historiography was not concerned with collecting indigenous accounts. Indeed, until the 1910s, it was largely indifferent to or unaware of how the past was reported and understood from a perspective of Timorese cultural idioms.⁸ In effect, the subject matter of historiographical speculation was not the history or ethno-history of reinos in Timorese terms, but instead the history of reinos as a subsidiary narrative to the self-referential master narrative of the Portuguese colonial establishment in Timor and Solor. Generally, therefore, the past of the ethno-political multiple was addressed and conveyed in Luso-centric narratives of empire and colonization.

Until the 1900s, three main issues motivated this Luso-centric colonial historiographical tradition. Firstly, there was the pattern of the pre-Portuguese indigenous polity, with Portuguese administrators speculating about how the indigenous polity on Timor Island was organized and how hegemonies of power were distributed before the arrival of Portuguese agents in the 1500s. Secondly, colonial writers addressed the problem of the survival of this ancient indigenous polity in the present. They were worried by those aspects of the old indigenous polity that continued to hinder and interfere with Portuguese claims to power. Lastly, conjectures developed around the Portuguese role in the genealogies of the ethno-political multiple. This question concerned the extent to which the centrifugal tendencies of the polity either sprang from pure indigenous sources or were ultimately Portuguese creations. These three intersecting themes are examined below.

Indigenous “Empires?”

By 1860, it had become difficult to discern what the multiple polity would have looked like before the Portuguese intrusions. It seems plausible, however, that after the Portuguese started intervening in the local polity in the 1500s, the situation developed into an effervescent plurality, diversity, and rivalry that had probably been absent in more ancient times. The Portuguese eagerness to have dominion over Timor-first through Catholic missionaries and the mediation of Topasse lineages, and later through formally appointed governors from outside the region-was accompanied by fragmentation of the ethno-political multiple. One question asked by the colonizers, therefore, was whether this plurality of reinos had ever been (and could ever be in the future) subject to forms of centralized power analogous to those the Portuguese wanted to realize for themselves.

After the move to Dili in 1769, the Portuguese governor’s claims over eastern Timor intensified. Subsequent conjectures about the past thus concerned especially the effects of this past on the governance of the eastern half of the island. Historical conjectures commonly posited that the Timorese polity-either on the whole island, or the eastern side alone-had once been structured in wider domains and centers of rule. The Portuguese term império and the title of imperador could be used to refer to these domains and their rulers. Vestiges of these once great indigenous realms and centers were seen as still being active in the present, thus rivaling Portuguese ambitions to hegemony. While Godinho de Eredia’s brief statement of a country divided into two parts or “empires” in the 1600s, as mentioned above, offers an early example of this, perhaps the document that most influenced the later development of this historical concept was the previously mentioned instructions given by the Viceroy of the Estado da Índia, the Count of Sarzedas, to the upcoming governor of Timor in 1811. Sarzedas then wrote a lengthy manuscript containing numerous data, recommendations, and guidelines on the past and present situation of the country. This document was sent from Goa to Dili, where it was archived for reference by successive local governors until Afonso de ^{Castro gave the manuscript wider diffusion by analyzing it and publishing it in his book of 1867}.⁹ Afonso de Castro, however, was not the first nineteenth-century governor to write extensively about the history of Portuguese colonization in Solor and Timor.¹⁰ In effect, Castro’s investment in research, transcription, interpretation, and publication of older Portuguese documents is an original development of a colonial historiographical modality of investigation that found fertile ground on Timor to flourish.

According to the Sarzedas document, as transcribed by Afonso de Castro, the indigenous polity of the island of Timor in the remote past was composed of “so to speak two provinces and two nations”-the eastern Belos and the western Vaiquenos or Servião-“each one divided into numerous reinos […] all free and independent from each other.” The Servião kingdoms acknowledged one king, referred to as “emperor” of Senobay, as “superior,” while the eastern Belos were split into two or perhaps three “higher” kingdoms or kings, the “Suray [Liurai] of Riquitta, the king of Luca”;¹¹ the “Suray of Uzalle” [Wehale]; and “the house of Camenasse” (Sarzedas 1811: 200). Thus, the instructions by Sarzedas present a portrait of an indigenous order of broad realms as perceived by the author in the early 1800s. Although Sarzedas did not speculate about its deeper past, Afonso de Castro used Sarzedas’s description to formulate his own interpretation of the ancient past of the reinos before the arrival of the Portuguese.

In his work of 1867, Castro eventually projected Sarzedas’s words retrospectively back to a pre-colonial past and articulated a historical narrative of the Timorese polity that was to gain wider currency in the coming decades. Castro suggested the reinos once formed two large compounds dominated by two indigenous “imperial” sovereigns. “It seems,” Castro speculated, “that before our [Portuguese] domination there was in Timor two formidable potentates that made almost all peoples in the island their vassals. One of them was the emperor Sonebay [Sonbay], the other the Behale [Wehale]. The first governed over the western side of the island, the second the eastern side” (1867: 314). In this view, an island divided into two great indigenous domains, which also corresponded to a division into eastern and western lands (the Belos and Vaiquenos provinces), preceded the Europeans. Castro also devised a possible origin story for the wild ethno-political multiple. He hypothesized (1867: 314) that, at some point in time, the large indigenous domains had exploded into a galaxy of competing and independent polities. The two potentates had lost control of their vassals who became “independent, every one governing his kingdom despotically and arbitrarily.” Lastly, Castro also conjectured that maybe “in former times,” back in a deep and distant past (possibly prior to the existence of two empires), Timor was not divided into reinos, but into sucos; a mere guess that he inferred from the influence that an ancient Tetum nobility, the datós, still held in the 1860s (^{Castro 1867}: 18, 313-14; ^{França 1897}: 236). Castro’s influential studies seemed, therefore, to express ambivalence as to whether the radical multiplicity of the 1850s-60s resulted from growing fragmentation of an originally bicephalous polity or was instead a variant of yet another dispersive form of indigenous organization-a polity broken into even smaller units, the sucos. Other governors addressed this dilemma in Castro’s wake, with the governors Celestino da ^{Silva (1901}) and Cipriano ^{Forjaz (1893}) considering the East Timorese polity to have been less fragmented in the pre-Portuguese past than in the present. In contrast, however, to Castro’s conjectures on ancient empires and primordial sucos, these two governors speculated that Timor as a whole had once been divided not into two “empires” (Sonbay and Wehale), but rather into “two important houses” or “families”-the house of Wehale (Portuguese: Veále or Beale) and the house of Likusaen or Liquiçá-and that these two “houses” eventually stood in constant rivalry and continued to exert their influence, to the Portuguese disadvantage, into the late 1800s. Silva and Forjaz further hypothesized that two single rulers, bearing the title of loro, headed these two ancient and larger polities of the past. Prior to the establishment of diarchic authority by the dató nobility, the loros mediated major disagreements between domains and concentrated both the “religious and civil powers.” Yet, while their influence was seen in the 1890s as the dying vestige of a powerful primitive aristocracy, Governor Celestino da Silva (1901) sensed that the loro of Wehale in particular could still exert considerable influence.

The Indigenous Polity as a Portuguese Creation

The ethno-political multiple was ordinarily perceived as traditional and characteristically Timorese. The reinos, in this view, were above all autochthonous creations originating purely from indigenous sources. Nevertheless, it was also historical common sense that the role played by the Portuguese in shaping this Timorese world over the centuries was not entirely innocent. Some governors were more skeptical about indigenous origin “purity,” and even writers, such as Castro, who endorsed that viewpoint, also accepted and practiced various old-style Portuguese institutions that produced or reproduced some crucial aspects of the polity. As even Afonso de Castro acknowledged, following Sarzedas (1811: 198), the titles of rei and coronel, for instance, had been granted to Timorese individuals and groups as part of a political strategy followed by the early colonizers in the seventeenth and eighteenth centuries. Indeed, many of the status signs, symbols, titles, and institutions that configured jural and military authority in the indigenous states by 1860s had a Portuguese mark. Timorese kings and nobility seemed to rely on several external inputs of Portuguese creation in order to ensure their own kinds of sovereign authority and governing power in the communities. Changes introduced by the Portuguese in former times had become included in the Timorese socio-political order-to the extent that, for some observers, it was unclear which aspects of reinos actually belonged to a distinctively indigenous past.

In 1901, sensing something of this intricate process, Governor Celestino da ^{Silva (1901}) advocated a strong and self-serving interpretation of the existence of reinos. In a report on “Timorese uses and customs” submitted to the Minister in Lisbon (though never published), this governor put forward an argument about the origins of reinos that centered on the commanding figure of the Portuguese governor. Observing the wide diffusion of Portuguese power-charged status signs, rites, and political objects among Timorese kings, he argued that Portuguese actions were the genesis of the profuse distribution and multiplication of kingship authority. To this purpose, he engaged critically with ^{Castro’s well-known work of 1867}, specifically criticizing Castro’s approaches and evidence concerning the “uses and customs, estilos or consuetudinary legislation of the Timorese” reinos. One aspect of Silva’s criticism was of a methodological nature. He claimed Castro’s work was “most deficient, frequently romantic, in what regards the uses and customs” because, Silva (1901) estimated, Afonso de Castro was not fluent in any “native language” and he relied on Timorese informants who “could hardly speak Portuguese.” Celestino da Silva did not specify the sources of this criticism. In an unpublished letter to Lisbon forwarding his extensive essay about Timorese uses and customs, however, Afonso de Castro (1860) indeed acknowledged that information for this work was provided to him by a Timorese interlocutor, a certain king he described as “Timor king Alferes-mór of the District of Belchior, an aged man and very intelligent,” whose name he did not disclose. By contrast, Silva (1901) claimed a direct kind of empirical authority, largely unmediated by Timorese informants, based on his acquaintance with several “native languages” and on seven years of practical experience of Timorese uses and customs.

Celestino da Silva specifically challenged Castro’s interpretation of the “election and appointment” of Timorese kings and, consequently, of the origins of the latter’s authority as heads of the reinos. According to Castro, the independent authority of Timorese kings stemmed from purely native processes of election. Silva contested this view, claiming that Castro had failed to recognize the decisive actions of Portuguese governors in empowering Timorese kings through a series of ceremonial actions of investiture and donation of the insignia of kingship power. Contradicting Castro, Silva proposed that the Portuguese governor in particular was the source of origin of the constitution of reinos, and the center from which jural and political authority emanated throughout the indigenous realm (^{Roque 2012}b). In other words, for this governor, the Portuguese themselves were the creators of the ethno-political multiple. Here, it is worth quoting Celestino da Silva at length (1901):

The chiefs of the native states have a variety of names according to the language of the kingdom, thus the chiefs, and not only the régulo, are called “TETUM” LIU-RAI, from “LIU” more and “RAI” land. When the people want to name the kingdom’s chief they always say-the colonel [coronel]-and never “Liurai,” and so it seems it has always been since our governments started, in a line of thought much to regret, to grant the rank of colonel with inherent honors to the chiefs of native states, or régulos, as we ordinarily call them. . . Was this always like this? I don’t think so, because before we set foot in Timor there were no native colonels, lieutenant colonels, majors, etc.; the chiefs were “Liurai,” “Dato,” “Aran,” “Rata,” “Najufa,” etc. depending on the kingdom’s language; but such times are pre-historic, and the customs and ways according to a natural evolution gave new forms to consuetudinary legislation, and so little is left now. . . the régulo [king] was imposed upon the minor states subdued by the governor, and from the latter’s authority all power [of the régulo] is derived. . . Before we landed in Timor, the régulos were not what they are today.

Portuguese governors had an upper hand in generating hierarchy and diversity in the indigenous polity by granting kingship titles and honors to myriad indigenous classes. Thus, in this self-serving theory of colonial power, reis and reinos were above all a Portuguese creation. This governor-centric view implied a historical narrative of the origins of reinos that emphasized the Portuguese imprint upon political processes to the complete exclusion of indigenous agency. However, as even this governor acknowledged, such Portuguese mastery was overstated. It was also clear that the polity had gotten out of hand and become alienated from its European creators. Portuguese actions caused an explosion of kings, colonels, brigadiers, and so forth, who could turn against their alleged creators. As a result, the polity of reinos, although interweaved with the Portuguese, was also outside Portuguese control. This independence, furthermore, followed separate patterns and seemed to work in accordance with principles of governance and society-referred to as usos e costumes-that the Portuguese did not recognize as their own.

The reinos, it was believed, relied largely on the existence of a separate governance system based upon institutions and norms that were extraneous to and independent of Portuguese authority and European-styled legislation. According to this hypothesis, it was possible, beyond and underneath the great diversity of reinos, to unveil an ordered realm of distinctively Timorese institutions and norms, a set of static “traditional” parameters and designated uses and customs, or estilos. The polity of many could thus be efficiently ruled as one unified polity, it was believed, providing the Portuguese administrators were able to grasp the unique lawful uses and customs that were key to a successful colonization. In the 1800s, therefore, alongside investigations of historical import, efforts were made to isolate a distinctively indigenous Timorese socio-political nature, a kind of static essence of indigenous society that could be studied and governed for colonizing purposes. To this end, ethnography rather than historiography was to become the main investigative modality mobilized for understanding the indigenous kingdoms as a multiple of uses and customs.

Surveying Timorese “Uses and Customs”

The focus on justice, governance, and customary norms and practices was paramount to investigations of uses and customs. In colonial writings, uses and customs resembled a wide and plastic colonial concept that covered a wide variety of themes, not limited to strictly legal and juridical aspects. Topics described under this heading spanned from native “religion” to “marriage,” and from “penal punishments,” warfare, social stratification, slavery, funerary rites, and witchcraft to the election of kings. The study of uses and customs thus formed a broad sociological field of ethnographic investigations of the indigenous polity. In particular, knowing uses and customs was seen as the pathway to developing a mimetic style of governing the ethno-political multiple, and one dictated less by European norms and laws than by a selective incorporation of what was perceived to be the “Timorese” forms of governance and law. It expressed an emerging form of mimetic governmentality, as I have argued elsewhere (^{Roque 2015}), whose underlying principle of action was the incorporation and selective reproduction of the perceived otherness of “primitive” populations, with a view to ruling and governing their existence. Thus, rather than ascertaining land produce, taxes, or political loyalty, the issue here was to determine the indigenous institutions through which Timorese socio-political alterity could simultaneously be understood, imitated, governed, colonized, and “civilized.” In this context, the science of ethnography became the knowledge endeavor suited to this type of colonial governmentality and aimed at exercising justice and command over the “natives” from the viewpoint of their “uncivilized” customary norms and traditions.

Governors Write about Uses and Customs

As Afonso de Castro’s government style expressed this mimetic principle, his writings devote close attention to customs and manners that he saw as defining traits of the otherness of Timorese societies (^{Roque 2018}). As explained above, Castro’s works stand out as attempts at grounding the colonial governance in Portuguese-centered historical inquiries. Simultaneously, they also stand out as seminal tentative descriptions of Timorese “uses and customs.” Indeed, in this mid-nineteenth-century endeavor, colonial historiography and colonial ethnography came into being as an intimate pair, almost a blend perhaps, of colonial knowledge forms. The will to penetrate the obscurity of indigenous governance and justice though a research focus on what were seen as indigenous uses and customs was at the core of Castro’s controlling anxieties regarding the ethno-political multiple: “We allow the kings to administer justice,” a vexed Castro wrote, “and the people . . . are tormented by one thousand small tyrants over which the Governor has little authority” (1860). In addition, the Timorese polity was seen as following utterly distinct and independent “native” norms, external to European law and authority. “The kingdoms are governed by what one calls estilos,” he concluded, “and there our mild legislation is unknown” and “[they] are not ruled by written laws, they are ruled by Timorese estilos, which are preserved unchangeable in tradition” (¹⁸⁶⁷: XV, 315; ^{Castro 1863}). For this reason, in Castro’s phrasing, a critical aspect of a modernizing colonial government was to know “how the reinos of Timor are governed” (1867: 321).

Further steps in this same direction were taken during Celestino da Silva’s fourteen-year tenancy as governor of Timor (1894-1908). For this governor, too, the need to know the diversity of Timorese customs and lawful ways of governance in the reinos was at the core of a good administration. Above all, army officers were tasked with studying “the character of peoples and their history” (^{Silva, 1896}: 5) and acquiring knowledge about the many and diverse uses and customs of the reinos under their jurisdiction, which they were intended to take into account in their judicial sentences and in the relations with the Timorese aristocracy. He also encouraged the “study” of indigenous customs by colonial officers as a step towards the idealized goal of “codification,” or the production of a formal written code of native law (Silva 1897: 42). These views aligned with the anti-assimilationist (also “specialization”) juridical theories and colonization doctrines then in vogue in scientific and legal circles in Europe (^{Silva 2004-2005}). These were overtly racist and social evolutionist theories that presupposed European “civilized” law was superior and could not be transferred to inferior “primitive” races. Instead, different laws, or special “codes,” had to be locally designed in harmony with each “primitive” world, a work implying rigorous studies of the different uses and customs, and leading to their codification. “Colonial legislation,” the judge of law in Dili, Albano de ^{Magalhães (1907}: 31) defended, “by means of a perfect knowledge of the peoples, should proceed from a meticulous study of the regions, from initiatives adapted to the social scale of that people.”

The Emergence of Ethnographic Questionnaires

Celestino da Silva’s successor in office, army officer Eduardo Augusto Marques, took these views forward by launching the first ethnographic survey on uses and customs in the colony. This initiative in Timor was also one of the first to be launched in the Portuguese colonies.¹³ During his intense one-year tenancy as district governor (1908-09), Marques put research on uses and customs at the heart of his reforms of the colonial state. In 1908, he created in Dili a “Committee on Uses and Customs” that aimed to “collect and codify with the greatest urgency the native juridical uses and customs, organize a special and swift process in the district native courts and to study a penal system applicable to the natives in the territory” (^{Marques 1908}: 176-77). Creating such a system meant, primarily, getting to know indigenous laws and traditions, and collecting “native customs” across the country. Accordingly, Marques determined that officers and missionaries stationed in the reinos should use an “ethnographic questionnaire” for collecting the data.

Alberto Osório de Castro, the then judge of law in Dili, as well as a Coimbra graduate, an accomplished poet, and a passionate amateur scientist, was put in charge of designing this questionnaire. Arriving in Dili in 1908, the judge allowed his vocation as a colonial official to be followed by his ethnographic fervors as, while in Timor, he developed an avid interest in the sciences of ethnography and ethnology, to this purpose reading widely and cultivating an international network of correspondents, most notably including the French ethnographer Arnold van Gennep. Like Van Gennep, Osório de Castro (1909: 199) saw ethnography as a science of “practical scope” in the service of modern colonization and based on scientific work, and enthusiastically embraced the governor’s initiative of a “Questionnaire about Timorese uses and customs for the organization of a native law code” (^{Castro 1909}: 206). “It is to be expected,” Castro commented, “that the survey in all military commands will also show ethnographic data of the highest importance for the government of the colony” (1909: 155). At a distance, from France, Van Gennep eventually mentored and somehow gave orientation to the project in Timor. Osório de Castro discussed the questionnaire with the French scholar by letter and, in the course of these exchanges, sent him a draft copy for appreciation and approval, which was subsequently received. “In the Mercure de France of 15 January,” as the judge in Dili wrote proudly to his father in Coimbra (1910), “is already kindly announced my questionnaire on Timorese uses and customs, classified as excellent.” Van Gennep continued to guide and encourage the colonial ethnographic questionnaire in further private letters: “Mr. van Gennep tells me in his last postcard,” the judge informed (1909: 244), and “L’initiative de Mr. Le gouverner est excellente … C’est excellent pour le développement de nous études.”

Under Van Gennep’s guidance, Osório de Castro divided the questionnaire into three main sections: Domestic Organization, Social Organization, and Political Organization, each of which included five to seven sub-sections on the topics. The questionnaire was then quickly put into circulation in 1909. While Osório de Castro (1909: 241, 245) discussed the “first” responses from missionaries received in Dili in 1909, it was not until 1916-17 that the partial results of this survey (concerning some “districts”; note the unit used here was districts, not reinos) were published in the Official Bulletin of the Timor Province. But although only partially published, the questionnaire left a strong mark on the local society, with new terms and administrative districts being used in the drive to grasp Timorese uses and customs in the ethno-political multiple. This desire to know the Timorese uses and customs continued after 1909, with later officers and administrators also making similar attempts to launch ethnographic questionnaires, even if they seem to have had little or no effect (^{Correia 1935}: 254; ^{Carmo 1965}: 10). In this way, the functional dimension of ethnography as a science of colonial administration was reinforced as, alongside recurrent questionnaires, colonial officers produced several ethnographic studies. Some of these studies circulated as articles and monographs, whereas many did not leave the manuscript form and the local administrative circuits, including several “Reports on Uses and Customs” produced by colonial officers in 1926 and 1933. But although these reports were never published, they were incorporated into one of the most prominent monographs in this colonial tradition, being Captain Armando Pinto Correia’s Gentio de Timor (1935)-a work that an elderly Van Gennep (1936), recalling Osório de Castro’s pioneering efforts, eventually praised as a “descriptive monograph” among the “best in the world.”

However, the rise and continuance of ethnographic surveys in early twentieth-century Timor did not simply supplant the prior enumerative and historiographical approaches, which continued to be practiced and reformulated by colonial agents. Although I will not elaborate on developments beyond the 1910s in this article, colonial historiography remained a relevant intellectual activity of several twentieth-century colonial officials and administrators (^{Martinho 1943}). Hence, and in contrast to the analysis of scholars such as Nicholas ^{Dirks (2001}: 44), who, in the case of British India, argues straightforwardly that “Anthropology supplanted history as the principal colonial modality of knowledge and rule” in the late nineteenth century, these activities should in no way be placed in an evolutional line of successions and replacements. Instead, in the decades preceding the tipping point date of 1911-12, all the former colonial investigative modalities of ethno-political multiplicity in East Timor seemed to have been simultaneously available and practiced. And in the years to come, all of them were able to be put to governmental use, coexist, blend, and mutate, rather than simply replace or supplant one another.

Glimpses of the Ethno-Political Multiple after 1912

By 1910-12, a vast coalition of Timorese kingdoms had joined a major anti-Portuguese uprising known as the Manufahi Rebellion. In the aftermath of the subsequent Portuguese victory, a profound administrative reform was undertaken to introduce new structures and divides in the territory that significantly interfered with (but did not eliminate entirely) the reinos and their royal aristocracies. The repression of this rebellion was followed by attempts to reduce or even eradicate the influence of these upper Timorese lineages. Many reis and reinos were eventually extinguished, replaced, somehow emptied of honors and titles, or perhaps forced into a socio-political life in the shadows. In the 1910s, a new colonial administrative division into sucos (Tetum: sukus) was imposed and took over the role of the former reinos in official dealings. In the three subsequent decades, “district” units and “ethnic” or linguistic groups tended to replace reinos as investigative units of colonial ethnography and anthropology. In this way, during the twentieth century a kind of ethno-racial or ethno-linguistic notion of diversity was replacing or perhaps being juxtaposed to the ethno-political multiple as knowledge object. The possibility of juxtaposing the geography of reinos with a racial topography was for a moment conjectured by the leading figure of Portuguese race science and colonial anthropology, António Mendes ^{Correia. In 1944}, imagining possible equivalences between ethnic types and reinos, Mendes Correia briefly considered mapping the ethnic groups of the colony in accordance with the “traditional division into reinos” (1944: 77). However, this procedure was regarded as too complex and so was quickly abandoned by colonial racial anthropologists in favor of ethno-linguistic categories and the more recent administrative divisions created after 1910 (^{Roque 2022}).

Although the reinos were officially side-lined as administrative units around 1913, neither the reinos nor the investigative modalities with which they were associated simply vanished, either from policy imaginaries or from everyday colonial life. In Portugal, the topic of the effective management of the organismes politiques indigènes (native political organisms) featured in wider comparative discussions of past and present colonization policies in Portuguese Africa and Asia (^{Negreiros 1910}), while in Timor some of the former world of multiple kings and reinos seems to have endured, albeit at a different pace, partially visible and partially hidden, and either passively or actively behind the scenes. In 1927, for example, the officer Almeida Pinheiro referred to the Timorese as living in villages “subordinated directly to diverse kingdoms whose superior chief is the king or chief of the kingdom” and counted “around 30 reinos subject to Portugal” (1927; ^{Sherlock 1983}). Indeed, until at least 1952, over thirty-nine Timorese “liurais or kings” continued to be counted and acknowledged as royal figures in official acts and lists by Portuguese authorities. Thus, in 1944, the army officer and once interim governor Gonçalo Pimenta de Castro could lament the enduring existence of kings: “There are still kings and queens who live with relative pomp, in brick-built houses, wearing silks and dressed up in European ways, and traveling to Dili” (1944: 13). Although, therefore, the ethno-political multiple of reinos seemed to have moved off the front stage of colonial investigations, it remained a significant political actor, and a haunting presence in the minds and activities of colonial researchers and administrators.¹⁴