sign in sign up
<<
Home , Dutch East India Company, East India Company, Jacob Mossel, Josiah Child, Pirates in Batavia

Principals and Agents, Colonialists and Company Men: The Decay of Colonial Control in the Dutch Author(s): Julia Adams Source: American Sociological Review, Vol. 61, No. 1 (Feb., 1996), pp. 12-28 Published by: American Sociological Association Stable URL: http://www.jstor.org/stable/2096404 Accessed: 30/03/2010 08:56

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=asa.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to American Sociological Review.

http://www.jstor.org PRINCIPALS AND AGENTS, COLONIALISTS AND COMPANY MEN: THE DECAY OF COLONIAL CONTROL IN THE *

Julia Adams University of Michigan Patrimonial states and their chartered East companies propelled the first wave of European in Asia during the seventeenth and eigh- teenth centuries. The metropolitan principals of these organizations faced special problems in monitoring and controlling their own colonial agents. Focusing primarily on the Dutch United East Indies Company and second- arily on its English counterpart, I argue that the network structure of each organization affected the degree to which relationships between patrimonial principals and their agents could serve as a disciplinary device. Dutch de- cline was imminent when alternative opportunities for private gain, avail- able via the ascending English Company, allowed Dutch colonial servants to evade their own patrimonial chain and encouraged its organiza- tional breakdown. Features of network structure determined whether colo- nial agents saw better alternatives to the official patrimonial hierarchy, when they could act on them, and whether principals could respond.

"O! when degree is shak'd, petitors were central challenges, of course, Which is the ladder to all high designs, although they could be taken too far, even from the colonialists' perspective. For, as The enterprise is sick. " Coen's metropolitan critics reminded him, -Shakespeare, Troilus and Cressida "there is no profit at all in an empty sea, empty countries, and dead people" (quoted in Meilink-Roelofz 1962:232). Within these The famous, or infamous, pioneers of the not unimportantlimits, however, colonialists first wave of European colonialism also faced dilemmas inherent in their own or- never tired of asserting that the project of co- ganizational structures; it is this complex of lonial domination was a difficult and precari- issues that is highlighted here. ous one, requiring the "strongest power on I argue that network structures mediated water and land," in the words of Jan Pieters- principal/agent relationships among early zoon Coen, an early -Generalof the modern European colonialists. The capacity Dutch East Indies (Coen [1620] 1919:554). of principals in Europe to control their agents Subjugating indigenous populations and in the depended on specific struc- overcoming metropolitan and colonial com- tural relationships-simultaneously political and economic-that bound them together. In *Direct correspondence to Julia Adams, De- the Dutch case, the principals first disposed partment of Sociology, University of Michigan, of resources that the agents required, and Ann Arbor, MI, 48109 (Internet: jpadams@ agents lacked viable alternatives to the net- umich.edu). My thanks to Fred Cooper, Femme work channels that linked them to the Neth- Gaastra, Mark Gould, Mark Mizruchi, several erlands. But a seismic shift in that opportu- anonymous ASR reviewers, the ASR Editor and nity structureopened the way for heightened Managing Editor, and Deputy Editor Charles Tilly for their helpful comments on the manu- principal/agent problems and undermined script. The research was partially supported by a group discipline, contributing to the demise summer grant from the Fulbright Foundation. of Dutch hegemony and the rise of the En- [Reviewers acknowledged by the author include glish empire in the eighteenth century. This Peter Evans, and Edgar Kiser. -ED.] was clearly an outcome of global historical

12 American Sociological Review, 1996, Vol. 61 (February:12-28) PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 13 importance and one that illustrates the im- Dutch case is indispensable for sociolo- portance of network structuresin epochal so- gists-a key to our understandingthe forma- cial change. The story of early modern Euro- tion of the global colonial system in the sev- pean colonial enterprise should interest so- enteenth century and clarifying the causal cial theorists as well as students of the past factors that made for organizational success on other grounds as well: It reveals both the in that system. potential fruitfulness of principal/agentmod- We can also investigate factors that led to els for comparative historical sociology and failure and systemic transformation. By the the need to better specify these models sys- end of the eighteenth century, the close of the temically and historically. Dutch ancien regime, the patrimonial struc- ture was severely strained. It soon gave way altogether. In the metropole, the European WHY THE ? center of the global colonial system, it was SETTING THE SCENE replaced by differentiated profit-making en- In its first, triumphant, phase, the seven- terprises, twinned with a power-wielding teenth-century Golden Age (Gouden Eeuw), state. In the colonies, the developmental the Netherlands established an unpreceden- story is not so neat. A cursory glance at the ted position of world power. Dutch develop- post-Company situation in , the ments during this period illuminate the gen- central pillar of the , reveals eral character of the first wave of European new and unstable modes of colonial domina- colonial enterprise. The basic structure of tion. But there also, ancien regime styles of early modern European colonialism was cre- accumulation and rule were displaced, and ated when merchant capitalists and their Dutch colonialism moved away from Com- home states joined together to charter large- pany rule and toward a more bureaucratic, scale companies aimed at global socially interventionist systemr.2 commercial and imperial dominance. The Clearly, exogenous shocks played a part in Dutch pioneered key aspects of the chartered the sagging performance and ultimate col- company form with the foundation of the lapse of the partnership between the Dutch United East Indies Company (Vereenigde state and the VOC. Particularly salient were Oost-Indische Compagnie, or VOC) in 1602. the growing economic and military power of The VOC merged individuals' assets into a other metropolitan state-company duos and single permanent ongoing enterprise, and the collapse of social formations indigenous was invested with sovereign rights over for- to so-called colonial target areas. Metropoli- eign territory and vassals. Charteredcompa- tan Dutch developments were also important. nies were quintessentially patrimonialforms, Here, I treat these processes as external pres- conjoining economic and sovereign political sures and opportunities, bracketing their goals at the behest of the ruler's personal dis- causes analytically, in a kind of "thought ex- cretion. l periment."The empirical questions I address Once launched, the VOC soon became an are: What endogenous developments under- organizational template for other metropoli- mined the Dutch colonial system? To what tan merchants and rulers, inspiring, among extent, in particular, was the troubled rela- others, the English and tionship among colonial rulers themselves a the many French Compagnies des Indes problem? How did these internal processes Orientales. It remained one of the most suc- cessful of the many hybrid colonial enter- 2 I follow conventional historical practice by prises whose licensed mercantile ambitions designating 1795 as the end of the Dutch ancien and fields of operation spanned the globe, regime. In that year invaded the Nether- ranging from the and cloth of lands and set up a client state that lasted until Indonesia and India, to the Brazilian 1815. Schama (1977) provides the comprehensive English-language account of this period in the industry and the African slave . Thus the metropole. In the East Indies, the remnants of the VOC limped along until 1806, the onset of a 10- I On the concept of patrimonialism, see Weber year, mainly English, (1806-1816). ([1922] 1968, especially pp. 1006-1007, 1010- From 1816, the Dutch state assumed sovereignty 15, 1022-23, 1028-31). over the VOC's Indies territories. 14 AMERICAN SOCIOLOGICAL REVIEW link up with external constraints and oppor- an attractiveopportunity, they will pursue it, tunity structures, especially the appearance even if they must contemplate guile. "Prom- of alternative (non-Dutch) channels of patri- ises to behave responsibly that are unsup- monial authority? ported by credible commitments will not, That these questions remain genuine therefore, be reliably discharged" (William- puzzles is partly due to gaps in the available son 1991:92). Whence the principal's prob- historiography. The period of decay and re- lems with its agents, and the need for incen- construction of colonial power, particularly tives and sanctions. I am provisionally work- the late eighteenth and early nineteenth cen- ing, then, from within that branch of utilitar- turies, is one of the least researched eras in ian theory that views information gaps and the history of the "GreaterNetherlands," per- enforcement problems as consequential to haps because it pales in comparison to the the analysis. Later in this paper, I will pro- glittering Golden Age. Yet it spans the cru- blematize these essentially rational-actor as- cial transformation from a fundamentally sumptions. But for the moment, the discus- commercial empire, flowering under the pat- sion assumes that they hold.4 rimonial protection of the Dutch East Indies Certain organizational arrangements can Company, to the less extensive but vastly help minimize the problem. Principals will more penetrative combination of colonial be drawn to any arrangement that makes state and system of forced cultivations and agents easier to check up on, that brings extractions that replaced Company rule in agents' aims more closely into line with Indonesia.3 theirs, that increases agents' dependence on them, that decreases collusion among agents, and that enables the principal to mete out re- PRINCIPALS AND AGENTS IN wards and punishments to maximum effect. PATRIMONIAL SYSTEMS On one end of the spectrumof organizational The organization of early modern European options lie command structures, or hierar- colonialism in general, and Dutch colonial- chies, which tend to lend continuity and sta- ism in particular, was an ambitious effort to bility to ongoing enterprises; on the other, accumulate capital and project power abroad. ideal-typical contracts, which promise As such, it involved several levels of agency greater flexibility. The two types are best relationships, ties that may be said to exist seen as poles bracketing a continuum that in- "when a principal delegates some rights ... cludes many forms of contractualhierarchies to an agent who is bound by a (formal or in- or hierarchicalcontracts.5 Furthermore, these formal) contract to represent the principal's interests" (Eggertsson 1990:40-41). Expan- 4 Major texts in the rational-actor tradition that sion by means of extending and multi- deal with early modern Europeansocieties include plying agency relationships was a two-edged Ekelund and Tollison (1981), Kiser and Tong sword. While it promised gains in the (1992), Levi (1988), North (1981), and Root principal's efficacy and reach, it also created (1987). Greif's (1994) and Thomson's (1994) im- problems of monitoring agents' activities and portant contributions to these discussions were published when this article in enforcing compliance through sanctions. was press. See Gould (1992) for a compelling criticism of the Following Simon ([1947] 1961) and Wil- utilitariantheoretical foundations of recent socio- liamson (1975), I begin by assuming that logical appropriationsof economic theory. both principals and agents tend to act in in- 5 Starting with Coase (1937), who initiated tendedly rational fashion, and opportunisti- these debates, a vast literature has developed, cally, to advance their own individual gains largely founded on a stark contrast between mar- (exogenously specified). If social actors see kets and hierarchies (on this and related issues, also see Eggertsson 1990; Stiglitz [1987] 1989; 3 I use the term "Indonesia" anachronistically, and Williamson 1975). I define a contract as an since no single nation-state existed during the pe- agreement between persons or firms that governs riod under examination. Nor did "India" or "the an exchange, and a hierarchy as a structure em- Indies" have exact geographical referents; from bodying relations of authority and subordination. the western European vantage point, these words White (1985) and Stinchcombe (1990) argue that referred vaguely to the lands east of the Cape of these two ideal types should be treated as end- Good Hope and west of the Azores. points on a conceptual continuum. PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 15 arrangements may be complexly and imper- structurallyunstable, as monarch and corpo- fectly articulated: One of the arguments in rate elites strove to subordinate one another. this paper is that vulnerable command struc- When one side was relatively successful, pat- tures may be strengthened when they are rimonial governance veered toward a unitary, complemented by, and rest upon, certain one-headed system of rule. But the practice types of tacit contractual supports-and of consolidating and extending rule by del- weakened when they do not. egating sovereignty ensured that patrimonial Patrimonial systems embody several spe- systems continually threatened to parcellize cial twists to the relevant principal/agent re- into multiple, segmented headships. Colonial lationships-two are important here. First, systems, in which the capacity to exercise the substantive content of roles and ties is military force devolved down to agents, were multivocal. That is, key agents were del- particularly liable to fragmentation because egated multiple and interrelated organiza- agents could more easily generate the power tional goals, including what we think of as to act as principals and thus as competitors economic ends (extracting surplus resources to their own principals. This tendency gave and making profits), as well as the coercive anotherturn of the screw to the daunting task goals of maintaining and extending sovereign of managing patrimonial agents. reach. The top Dutch colonial agents were, Like other Europeanpatrimonial states, the by virtue of their singular positions, simulta- early modern Netherlands exemplified these neously traders and rulers. The metropolitan two endemic characteristics:multivocal roles principal therefore faced the ticklish task of and parcellized power. A few remarks are in strengthening the power of these agents to order regardingthe latter feature, particularly insist that surplus pass through the restricted with respect to the relationship among the set of nodes leading back home, while mak- metropolitanprincipals. Elsewhere I have ar- ing sure that wily agents did not evade that gued that potential governance problems patrimonial chain for their own ends. Thus were minimized at the outset of the first patrimonial principal/agent ties present more wave of European colonialism, circa 1600, complexities than the employer/employee or by the Netherlands' special brand of estatist manager/worker relationships in capitalist patrimonialism, which helped make the firms with which most contemporaryprinci- Dutch the most successful players in the pal/agent models are preoccupied. Managing Asian mercantilist game (Adams 1994). The patrimonial agents was always a political and Dutch metropolitan state and the VOC were economic quandary,requiring multivocal or- segmented in corporate bodies dispersed ganizational innovations if colonial rule were throughoutlocalities and provinces, but these to remain in place. bodies were controlled by members of a he- The second twist derives from the multiple reditary patriciate-a merchant-regent elite. headship of patrimonial systems and the (The patriciate, particularly those based in varying degrees to which the principals' , also controlled the navy, the goals are integrated with one another. In the "last instance" military backup for Company broadest sense, patrimonial systems in early ventures.) The stadholders were margin- modern Europe were dual ones, in which a alized throughout the era of Dutch commer- ruler (or rulers) and state-sponsored corpo- cial hegemony (1602 to 1672). Thus the re- rations jointly carried out systemic political lationship between the corporate state and tasks and shared the prerogatives of sover- the small VOC directorate, the Heren XVII eignty. Policy was severally steered by mon- (the Seventeen Gentlemen) was relatively archs or (in the Dutch case) stadholders and harmonious. Both drew their members from by the corporate elites of the urban, provin- the same regent patriciate, and in fact many cial, and national estates.6 This duality was the late sixteenth century. Rowen (1988) de- 6 Each of the Dutch provincial States appointed scribes the early modern stadholderate, which a stadholder (often, however, the same indi- quickly became the de facto patrimonial posses- vidual). Dutch stadholders were at one time like sion of the House of Orange-Nassau and assumed provincial , but the stadholderate be- some of the substance and trappings of monarchy. came more powerful after the Netherlands had There were two stadholderless periods in the asserted independence from Habsburg in , 1650 to 1672 and 1702 to 1747. 16 AMERICAN SOCIOLOGICAL REVIEW men held positions in both bodies. The VOC Two- or multi-headed organizations can was basically a one-headed organization dur- perform well under certain circumstances. ing this period, a signal advantage when the partners in a competitive capitalist Netherlands' leading European competitor, firm residual claims on the firm's prof- , was saddled with a political system its and are thus constrained to cooperate be- torn between and the patrimonial cause of their shared bottom-line goal. They elite in Parliament. In England, this split en- may also be disciplined by the economic en- couraged conflict among contending princi- vironment: All else being equal, a capitalist pals in the metropolitan wing of the English firm will presumably go bankrupt if it fails East India Company (EIC). to perform efficiently, since it is competing A quick juxtaposition with the then two- with other firms.8 But patrimonial principals headed EIC should clarify the relative advan- in early modern Europe enjoyed much more tages of the Dutch situation. The English social latitude for nonoptimizing behavior. Company merchants saw crown patronage Even when engaged in joint enterprises, they and exercize of royal prerogative as essential were wont to aim at goals that were contin- but dangerously unpredictable.And for good gently compatible at best (the monarch seek- reason: James I licensed rival traders (for a ing territorial aggrandizement and members fee) to the East Indies in 1604 and 1617, of the estates pursuing mercantile profit, to against the charterthe crown itself had given put one common situation crudely). Call it the EIC in 1600. The Company was forced to the "Hydra ": The multiple heads or compensate the new contenders, and thus in- principals lacked institutional mechanisms to directly the crown, to resecure its monopoly. resolve the resulting uncertainties and in- Charles I similarly authorized Sir William fighting. Courteen to set up a rival company in 1632 to This factor influenced the comparatively trade to , Malabar,China, and Japan.The gradual of the EIC. Although struggle between the EIC and the Courteen the English and Dutch companies were company lasted for several years, severely launched within two years of one another, weakening the EIC abroad.Thus, in England, the English company was much slower to monarchs routinely violated Company trust, develop an organizational identity and per- while the Parliaments of the day suspected manent capital. The leisurely rhythm of the EIC of royalism and doubted the legality company development was also imposed by of privileged monopoly companies tout court the EIC's inability to best the Dutch in the (Hill 1961:37-42; Brenner 1993:170-81). Spice Islands. The EIC was pushed back The EIC stayed afloat and navigated the onto the at a very early Scylla and Charybdis of ruler and estates stage, where it expanded more slowly. From (unlike some of the French companies, for 1600 to 1610, when the VOC sent 76 ships example), but only with difficulty. Episodic to Asia, the EIC sent only 17; from 1610 to contention between the Crown and Parlia- 1620 the respective figures are 117 and 77; ment, the two putative heads of international from 1620 to 1630, 141 and 58; from 1630 commercial/colonial policy, continued to un- to 1640, 157 and 59; and from 1640 to dercut the Company's ability to enforce its 1650, 164 and 75 (Gaastra and Bruijn monopoly and to hold together its own mer- 1993:182). These figures register the lower chant sponsors. The EIC's prospects im- initial capitalization of the EIC and the po- proved under Cromwell, whose 1657 charter, modeled on the VOC's, endowed the EIC further confusion of accounts and authority (see with its first permanentjoint organiza- Chaudhuri 1965:40 and Brenner 1993). Conflict tion. This arrangement was reaffirmed by between metropolitan principals reverberatedun- Charles II after the Restoration. But the til 1708, when a rival English East India Com- backwash of the metropolitan struggle be- pany (created during one particularly heated pe- tween monarch and estates continued even riod of struggle) merged with the original EIC, healing the split and ending the conflict. beyond the of 1688.7 I Firms may have market niches that enable them to avoid direct competition, but they are 7 Prior to 1657, when the EIC's joint stock was nonetheless vulnerable to flights of venture capi- made permanent, overlapping syndicates caused tal. See Williamson (1975:185, 201). PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 17 litical troubles that beset it during the sharp- THE DUTCH EAST INDIES est disaccord between Crown and Parlia- COLONIAL HIERARCHY ment.9 In contrast, the Dutch Company thrived At the outset of the colonial project, the nuts from early on. During the era of Dutch he- and bolts of the Dutch East Indies system gemony (1600 to 1672), the VOC turned a worked as follows. The Seventeen Gentle- profit without receiving direct economic in- men would send two to three Company fleets puts from the States-General or other metro- a year to their main outpost in Indonesia- politan state bodies. The Company ploughed (present-day Jakartain ). These profits back into its own coffers, while typi- fleets carried goods and precious metals for cally paying a 10-percent-plus annual divi- trading, men to replenish the colonial ser- dend to after 1630 and passing vants (who were continually dying off), and along money to the state, such as the 1.5 instructionsto the colonial government. Each million it paid the States-General in trip took between seven and nine months 1647 for the renewal of its charter until each way, sailing via the 1672 (De Korte 1984:6). In return, the to Batavia and back. Once unloaded in Indo- States-General staunchly supported the nesia, the materials, men, and instructions VOC's monopoly privileges (Algemeen fell under the of Batavia's top Rijksarchief 1.01.07 #1235, #1237, #1244; colonial agents: the members of the Indies Israel 1989:103; Elias 1923:39-44). In addi- Council and its chairman,the Governor-Gen- tion, the VOC proved able to offer its home eral, together known as the High Indies Gov- state a backup naval force, which pressed ernment (Hoge Regering). These agents, or the Iberian empire in the East and helped company servants, as they were called, force Spain to negotiate a truce, and eventu- would then dispense the cargos to the lower ally a permanent peace, with the Nether- rungs of the colonial hierarchy.10 These lands. lower levels included the VOC's roving mer- So far I have focused on the metropolitan chants and its settlements and "factories" pole of colonial rule, concentrating on the (factorijen). period in which its patrimonial organization Of particularinterest here are the topmost was launched. As the colonial system got agents in Asia-approximately 100 Com- underway and state-sponsored corporate pany merchants, heads, and men in bodies like the chartered companies took the highest positions in the High Indies Gov- off, the relationship among metropolitan ernment-that VOC records counted as the principals became more -adversarial. The decision-making elect in the late seventeenth VOC in particularcould act as a state within century.11 Their numbers were a drop in the a state-so various Dutch regents com- bucket compared to the 13,000 Company ser- plained, or exulted, depending on whether vants in Asia at the end of the or the they controlled one of the coveted director- 25,000 a century later, but most of these ships. But the Seventeen Gentlemen were lesser servants were sailors and soldiers who also the principals of the Dutch East Indies had virtually no chance of rising in the hier- Company, and in that role they confronted the intractable independence of their own l' Gaastra (1991) offers the best overall review colonial agents. Governor-Generalvan Rie- of VOC politico-economic organization. For gen- beek, head of the VOC's East Indies opera- eral discussions in English, see Arasaratnam tion, put this in a nutshell when he chided (1986), Boxer (1965, 1979), Furber (1976), and his metropolitan superiors in a 1714 letter: Israel (1989). My discussion of the trade cycle draws on all these texts. "The Gentlemen in the Fatherland decide I things as they see fit, but we do things here, Contemporaries estimated the number of top as we best understand and decide them" Dutch servants at 1-15 in 1688 (Gaastra 1991:94). The number of English servants was also small. (quoted in Gaastra 1991:68). "By 1799, the total number of the English comp- any's civil servants had only reached 748-Ben- 9 For details on the Dutch elimination of the gal had 351, 202, Bombay 101, Benkulen English in the Archipelago in the , except 55, St. Helena 19, China 20" (Furber 1976, note on VOC sufferance, see Furber (1976:38-42). 13:374). 18 AMERICAN SOCIOLOGICAL REVIEW archy to occupy major decision-making po- the Netherlands on one of the large, armed, sitions, notwithstanding popular Horatio return fleets. On arrival, the return cargoes Alger-style tales of humble cabin boys as- would be held in VOC warehouses until the cending through VOC ranks to the post of time came for the Company's periodic and Governor-General.12 highly profitable auctions, releasing the From Indonesia, VOC merchants and their goods to the European market. Some of the assistants transported some of the incoming monies realized would then be allotted for specie and goods to India, Japan, Persia, and organizationalrunning costs, some for share- eventually China. Since there was little de- holder and directors' cuts, while mand for European commodities at this early the rest was ploughed back into expanding stage of , precious metals, both the Company capital. The whole trade cycle, in bar form and in specie (), played an from the original decisions to the important role in this trade. In China, for ex- final realization of profits, stretched over two ample, the VOC sold silver imported from years-more if one takes into account the ul- Europe and invested the profits in Chinese timate source of the specie, which was not in . The silk was then shipped to Japan and the Netherlands, but derived from the Ameri- traded for gold and copper. The gold and cas and Japan.13 copper were exchanged for textiles in India. The trade cycle was also fraught with un- Buying cheap and selling dear was a funda- certainty. The genius of the VOC lay in the mental source of VOC profits, not simply at novel steps it took to reduce . On the one the endpoint of the European staple market, hand, the unprecedented scale of its capital but at every stage along the way. base and operations enabled it to manage The VOC had fully 30 factories in the East supply and demand, and to cut out traditional Indies by the end of the seventeenth century. middle markets that had raised a whole se- These were not factories in the modern ries of intermediary (Musgrave 1981). sense, of course; they were points of collec- One striking aspect of the Dutch East Indies tion for goods from specific local trades and system was the VOC's ambitious defense of organizing nodes for exercizing control over a triple monopoly. That the Company laid prevailing production relations. On Ceylon claim to a world monopoly on certain and the Moluccan islands, VOC factors orga- is well known, as are its state-sanctioned nized extraction within coerced tributaryre- claims to all Dutch trade east of the Cape of lations. There spices were the commodity in Good Hope. But the VOC also asserted do- question; , , , and minion over what was dubbed the "country mace composed the heart of the monopoly. trade," intra-Asian commerce in key com- The VOC later applied this schema to cof- modities, and tried to impose this on its own fee-growing, which it engrossed in Batavia's servants as well as on indigenous and Euro- hinterlands, after they were "pacified" in the pean merchants.14 . In other factories, the Company did not function in this feudal-lordly fashion, but 13 Van der Wee (1981), and Prakash (1985:11- instead ordered goods on consignment from 13) discuss the role of bullion in the VOC's Asian trade. Prakash calculates that between 1651 and indigenous merchant/brokers,to be delivered 1657, goods made up 45 percent of the total of at piece rates. This was the case in the facto- bullion and commercial goods received at ries in India, where the VOC bought local Batavia; the figure for 1700 to 1750 drops to textiles. Finally, some goods, like Chinese about 33 percent. Chaudhuri(1978, app. 5) shows , were delivered by means of inter-mer- that the situation was similar for the EIC. chant contacts in Batavia. 14 In this it differed markedly from the EIC, Once appropriated,the Indies goods would whose early aspirations to interport Asian mo- pass through Batavia, to be shipped back to nopoly were thwarted by inadequate capitaliza- tion and by infighting among metropolitan prin- cipals that strengthened metropolitan interlopers. 12Taylor (1983:5) lists the few Governors-Gen- The EIC permitted its own servants to trade in eral who rose from lowly positions as ship's boy Asia, a practice that was initially a symptom and or soldier. Actually, as Schutte (1974:32-35) source of organizational weakness. Furber (1976, shows, mid-level Company servants stood a chap. 6) compares the "countrytrade" patterns for slightly better chance of being promoted. the Dutch and English empires. PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 19

As we would expect of a multivocal patri- cal contracts with Asian agents. This is con- monial structure, the Dutch East Indies sistent with Thompson's (1967) claim that Company's capacity to muster massive force economic uncertainty may engender vertical was key to its success. The VOC undertook integration, and with Stinchcombe's (1990) some extraordinary coercive interventions, argument that environmental uncertainty is like Governor-General Jan Pieterszoon the driving force influencing the direction of Coen's selective extermination of the Band- growth of formal structures.Yet the very ex- anese Islanders, and the military attacks that pansion of organization and the internaliza- ousted the English East India Company from tion of middle markets and armed force into the Archipelago; but the Company also de- its hierarchy that made the VOC a vaunted pended on dull daily compulsion to carry on model for other European colonialists also business as usual. At every stage, military generated new uncertainties, especially new pressure helped the VOC maintain the inter- problems of internal discipline. The Seven- locking institutional arrangements for ex- teen Gentlemen expected their colonial tracting and realizing surplus. Force was ap- agents to do as they were told. But given the plied at the point of production (most impor- multi-step organizational character of the tantly in the spice monopoly process) and at VOC and the length of time it took informa- the point of exchange, in dealings with in- tion and orders from the metropole to arrive digenous merchants and brokers. As the in the Indies, thorny problems of communi- VOC's holdings expanded, so did its need for cation and control were bound to arise. effective coercion. And in all cases, the cred- ible threat of force guaranteed the general rules of the game, from the bogus treaties CENTRALITY, DEPENDENCY, AND mandating relations of vassalage with local DISCIPLINE potentates to the rule of law that secured the Because Batavia was the central economic, VOC's monopoly in the Netherlands. Rela- military, and communication node connect- tions of force and fraud were not external to, ing the VOC's Indies outposts, the link be- but were constitutive of, the Dutch patrimo- tween the metropolitan directorate and the nial colonial chain.15 Batavian leadership was critically important The VOC managed to militarily enforce its to the functioning of the whole colonial sys- world monopoly on cloves, near monopoly tem. In the social network literature, central- on nutmeg, mace, and cinnamon, and to con- ity is understood in a number of ways. One trol the pepper trade against a formidable ar- dimension, "betweenness," is a useful start- ray of contenders, both European and Asian, ing point for my purposes. If betweenness is in Dutch-Indonesian trade and intra-Indone- taken to indicate the extent to which a unit sian commerce. By the mid-seventeenth cen- must traverse the unit of reference in order tury, the VOC had consolidated its hold and to reach other units, then units (or actors) was trading widely, not only in spices, but that are central, in the sense of high between- also in Chinese, Persian, and Indian , ness, are the most likely to be situated on Japanese copper, Indian sugar, and other unique chains joining peripheral actors commodities. (Freeman 1979). These units or actors are The VOC extended its organizational ca- strategically situated, and strategic position pacity by substituting Company hierarchies in a network is one ingredient of power ad- for middle markets and negotiating hierarchi- vantage. Figure 1 renders the colonial situation in the early to mid-1600s as starkly as possible, 15The VOC induced Asian rulers to sign trea- while preserving the essential logic of the ties with discriminatory provisions against other organizational structure. No Company actor Asian and European powers, a practice that holds a truly central position. The Seventeen transformed indigenous rulers into VOC vassals Gentlemen and the Batavian elite both occu- (Alexandrowicz 1960:267-70). If necessary, mili- tary backup was available from Batavia, where pied positions of intermediate centrality on VOC troops were garrisoned, standing ready to all paths to realizing their economic gains; be deployed on the orders of the Governor and they were both brokers, that is, intermediary the Indies Council. actors, who could "facilitate transactions be- 20 AMERICAN SOCIOLOGICAL REVIEW

THE DUTCHEAST INDIESCOMPANY (VOC) THEENGLISH EAST INDIACOMPANY (EIC)

Metropolitanand non-East Indiescolonial Metropolitanand non-East Indiescolonial producers and consumers producers and consumers

Merchant-regents \11 Crown Parliament SEVENTEENGENTLEMEN COURTOF DIRECTORS

Batavia

VOC factory VOC factory EICfactory EICfactory EIC factory Colonial merchants Colonial producers Colonial producers Colonial and consumers and consumers merchants

Figure 1. Schematic Representation of Company Organizations: Early to Mid-Seventeenth Century Note: Lines indicate any type of structuralconnection between organizationalnodes. tween other actors lacking access to or trust ing our plain and direct orders to you, as if in one another" (Marsden 1982:202). you were not a subordinate but a coordinate Equally important, they depended on each power with us" (quoted in Chaudhuri 1978: other for brokerage services. The mutual and 77). But control was a matter of degree (pun symmetrical dependency inscribed in the intended). Unlike the VOC, the EIC Court of heart of the VOC's hierarchy undercut the Directors' central position allowed it to ma- potential power advantage of the metropole nipulate similarly situated dependent nodes over Batavia. and play them off against one another, for Figure 1 also contrasts the VOC's organi- example by procuring reports about corrup- zation with that of its archrival, the English tion in one colonial settlement from another East India Company. The EIC had multiple one not beholden to it.16 Thus for all its prob- East Indies headquarters-Bombay, Madras, lems with multiple metropolitan principals Calcutta, and others. Each main settlement jostling for supremacy, the EIC had fewer was headed by a Governor General or Presi- difficulties dictating specific courses of ac- dent and a Council composed of senior mer- tion to its top colonial servants than did the chants. There was no English colonial equi- VOC. Recall, however, that the EIC was also valent to Batavia, in India or elsewhere. In a feebler organization than the VOC in the the English case, the metropolitan directors early seventeenth century, because of the and the multiple East Indies headquarters two-headed structureof English patrimonial- were brokers, but the metropole alone occu- ism prior to the Glorious Revolution. So the pied the central position in the overall orga- EIC did not-yet-offer Dutch Indies ser- nization and gained a potential power advan- vants serious competition or a possible alter- tage as the less dependent member of a lop- native employer. sided exchange relation. It was not that opportunism, shirking, and 16 free-riding magically disappeared from the The EIC did use such information to nega- tively sanction agents. Most spectacular, perhaps, EIC's ranks. Far from it: The metropolitan was the 1732 decision to dismiss the entire com- director Sir could still charge mercial council of Calcutta. For this and related the EIC's Madras servants with "perverting events, see Chaudhuri (1978:76; 1986:101, 117- or misconstruing, procrastinatingor neglect- 18). PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 21

VOC servants used their middleman posi- gal bureaucraticone), they did not do so by tions to capture some of the abundantsurplus means of sharply graded monetary incentives for personal advantage. In fact, such eco- keyed to effort or output. The Gentlemen nomic opportunism among Company ser- paid their servants an ungentlemanly pit- vants was a persistent feature of the Asian tance, even at the highest levels, and counted establishments. Practices like diluting the on their making money on the side. Nor did precious metals that arrivedin Batavia, skim- the Company insist that its servants post for- ming off a percentage of goods bound for feitable bonds, which might have provided a VOC warehouses, taking a brokerage per- potential negative sanction on malfeasance.18 centage from indigenous suppliers, or de- Instead, I argue, the network of brokerage frauding the Company on its return cargo, relations enabled the VOC principals to were widespread. They were also very lucra- maintain some measure of cartel discipline. tive. Gerard Demmer, for example, whose This structuralfeature was a crucial element salary was set at 350 fl. (guilders) a month, in the Seventeen's measure of disciplinary sent over 165,000 fl. to the Netherlands in success in the early years of colonialism. At 1652, and 57,000 in 1654. Pieter this early stage, stylized in Figure 1, the elite Sterthemius, director of the factory, at all colonial nodes still depended on the made 200 fl. a month, but carried 52,000 fl. Dutch home base for key economic broker- with him at his repatriation (Gaastra age services, as well as for the political pro- 1991:95). Particularfactories were popularly tection they needed to conduct their semipri- known as the best sites for graft, and direct- vate trade. There was no plausible "alterna- ing one of these outposts was viewed as a tive employer" or outlet for the fruits of the plum job. These practices extended all the VOC servants' clandestine commerce. Some way up the colonial hierarchy to the Gover- profits or goods, such as precious stones and nor-General himself.17 jewels, could be smuggled back home on Even more important for VOC servants' VOC ships, but sending anything beyond a prospects of gain was what historians have limited amount was risky, and required adroit designated the "privatetrade." The term "pri- collusion with ships' crews and a whole vate" is a misnomer, however: This trade ac- chain of metropolitan officials, arranged tually revolved aroundVOC employees who from a long way away.'9 The employees in had access to Company monopoly goods be- Asia who wanted to transfer larger amounts cause of their privileged intermediary posi- of funds to the Netherlands could do so tion, and who traded in these goods on the legally, by depositing their money in the side. This semiprivate contraband trade un- VOC's Batavian treasury or in the treasuries folded at the direct expense of Company of Ceylon and the Cape of Good Hope (and commerce, since servants skimmed off mo- other VOC after the mid-eighteenth nopoly and monopsony products and were century). In return, employees could draw loath to enforce Company political dictates bills of exchange, which were then shipped when such restrictions interfered with the sideline trade. Private trade also took place 18 Salaries paid by the EIC to its employees at all points in the colonial network, although were also stingy: ?. 5 a year for an entry-level it was specially concentrated in some spots, writer, ?. 200 for a factory president (Keay such as the Bengal-Batavia link (Prakash 1991:234-35). These salaries were not enough to 1985). How did the Seventeen Gentlemen live on. Later on, the EIC introduced the practice keep the problem in bounds? Surprisingly, to of bonding. It is not clear whether the bonds were a "modern" eye (or at least to a rational-le- more than a symbolic counterweight to tempta- tion. For more on bonding (a type of incentive 17 Prakash (1985:84) discusses VOC servants' contract keyed to output), see Shapiro and Stiglitz illegal disposal of Company goods and their pro- (1984:442) and Lazear ([1987] 1989). curement of export goods. Gaastra (1994) de- '9 See, for example, Bluss6's (1986:212-14) scribes employees' smuggling of silver into story of the diamond smuggling engineered by Batavia and their illegal attempts to make the one Batavian employee, Johan Bitter. In 1677, most of discrepancies in exchange rates between Bitter was finally betrayed to the metropolitan di- the colonies and the metropole. For analogous rectors by "a character named Jan Hay. (Jack practices in the EIC, see Watson (1980, chap. 4). Shark)," the mate of the homebound ship. 22 AMERICAN SOCIOLOGICAL REVIEW to Europe on the Company's returnfleet and elite fought so hard against the Seventeen cashed by relatives, agents, bankers, or the Gentlemen's attempts to establish analogues repatriatingemployee himself in Amsterdam to their services elsewhere in the East Indies. or one of the five other equipped with Such innovations would have underminedthe a VOC chamber (De Korte 1984). Batavian High Government's capacity to The agents' dependence on these mecha- squeeze out commissions, such as those no- nisms for transferring their profits enabled toriously characteristicof the Bengal-Batavia the principals, the VOC directors, to enforce shipping link, and would have given the Sev- conventional limits to profiteering. Besides enteen more political leverage, including the policing their home harbors, the Seventeen capacity to force intermediate nodes to com- Gentlemen also sought to hold down the pete with each other.21 overall amount of transfers. In 1636, they Batavia's privileged position meant that decided that Batavia would only be allowed the Seventeen Gentlemen lost some of their to accept a fixed amount of money for bills precious control over the more peripheral of exchange, an amount far below what VOC nodes below Batavia in the hierarchy, even servants demanded. Although that edict and at this early stage of the Dutch colonial en- subsequent proclamations were often evaded terprise. The metropolitan directorate did and the directors increasingly acknowledged have the final say in promoting men from that "a strict observance of these rules would middle-level Company ranks, and peripheral stimulate their servants to seek other ways- colonial outposts, to Batavia, the higher and for instance via the English company-to more central node.22 The Seventeen Gentle- transfer their money" (Gaastra 1994:3), the men sought to use this leverage to strengthen Seventeen continued monitoring individual the link between promotion and doing well servants. They would generally agree to for the Company as a whole, not just for one- in a servant's chips, but would refuse to pay self and one's family. But given the immense up when they thought the amount requested problems of monitoring performance, it was was exorbitant.20Thus what counted as cor- lucky for the Seventeen that corruption was ruption or "free-riding"on the Company or- restrained by tacitly negotiated deals that ganization was both contested and conven- rested on the lack of established alternative tional: It was a matter of tacit, elastic, con- outlets for servants' gains. tractual limits that the actors themselves ne- gotiated after the fact within a patrimonial context. SHIFTING PRINCIPAL/AGENT TIES: CARTEL DISCIPLINE BREAKS Precisely because of Batavia's centrality, DOWN the Batavian elite benefited disproportion- The metropolitan directorate tacitly depend- ately, in an economic sense, from this lim- ed on the absence of alternative employers ited tolerance of Asian agents' opportunistic to keep graft under control and semiprivate ventures. For a time, therefore, the Batavia 21 elite gained leverage over both the metropole The Ceylon node was one such potential and the subordinate settlements. Marsden competitor and a great (albeit -lived) irritant to Batavia. From 1662 to 1675, the Seventeen (1982) argues that an actor's gains in power Gentlemen collaborated with Rijklof van Goens, stemming from brokerage behavior are due the VOC's Governor of Ceylon, deliberately by- to net excesses of inflows of resources over passing Batavia in their plans for expansion. outflows. Betweenness is particularly impor- When Van Goens himself was promoted to Gov- tant with respect to the inflows, since an ac- ernor General in 1678, he ordered the new tor that is centrally positioned in this sense Ceylonese governor to stick to the chain of com- can take commissions-cuts whose fractions mand and not to go over Batavia's (now his) head can go up as the number of workable indi- (see Arasaratnam1958). rect ties goes down-just as Batavia did. 22 Wijnaendts van Resandt's (1944) study of This feature helps clarify why the Batavian the men who served as supervisors of VOC colo- nial factories outside the Malay Archipelago shows that, in the VOC's early years, their pro- 201 Within limits, this practice could also help motions were more subject to metropolitan dic- the VOC finance its return goods on the cheap tates than to Batavian influence. This situation re- (see Gaastra 1994). versed itself in the eighteenth century. PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 23 trade within acceptable bounds. In the early to obtain the necessary quota of trade goods stages of the empire, the Batavia High Gov- to support its position in Europe.23 It also ernment and the lower-level company ser- presented a more subtle, but equally damag- vants in the Indies needed the VOC's metro- ing threat, by providing new and appetizing politan organization economically (as the opportunities that sabotaged the control that pipeline to the consumer market in Europe the Seventeen Gentlemen had over their where most of the Company's profits were Asian agents. The rise of the EIC and of En- realized and as the only reliable channel for glish private traders offered new opportuni- converting resources into currencies good in ties for Dutch agents to make a on Europe) and politically (as the military guar- the side. Private trade was easier to carry on, antor of unmolested private trade). But as particularlyfor the top servants at peripheral English private traders and the English East outposts close to English turf. These prac- India Company established themselves, this tices extended all the way up the hierarchy dependency diminished. to the Governor-General.For example, Gov- English competition escalated sharply in ernor-GeneralJacob Mossel was heavily tied the quarter century after 1660, just after the up with British country traders(Furber 1976: EIC was endowed with its first permanent 281). Conveying the profits back home was organization and capital. Chaudhuri's(1978) also simplified. VOC servants could now quantitativehistory of EIC trade attests to the convert their gains into cash and bank them Company's improving fortunes during this in the metropole by shipping them through period, showing that the value of the EIC's either the English hierarchy, the VOC, or total annual imports rose to ?. 800,000 in both.24 And although pursuing the English 1684. But imports fell to only ?. 80,000 by route involved additional steps and risks of 1691 in the wake of English interloping and apprehension, these inconveniences were the punishing Mughal War (1686-1689). counterbalanced by the greater amounts of Two events that set the stage for the next re- clandestine goods that could be converted. vival were the Glorious Revolution of 1688, These new opportunities affected the for- which ushered in a unitary regime in En- mal hierarchy in a twofold way. On the one gland, and the acquisition of a new farman, hand, the Batavian elite used its privileged or imperial license to trade, in 1690 under brokerageposition to tighten its grip on what the auspices of , Emperor of the was becoming the central position in the Mughals. The EIC's auspicious position was Dutch colonial network by the late seven- consolidated when its contending companies teenth century. The Seventeen now held an finally merged in 1708. and the English intermediate slot, structurally similar to the joined other battered European belligerents EIC's Court of Directors. On the other hand, in calling a halt to the War of the Spanish VOC officials situated at peripheral colonial Succession. In both periods, 1660 to 1683 nodes found that they could dodge the more era with and 1710 to 1760 (the latter ending 23 the EIC's assumption of territorial power in By the end of the , the English domi- nated the Indian piece-goods market so thor- Bengal), the happy conjunction of new op- oughly that "the Dutch had to travel to portunities and renewed capacity to make the and purchase samples to send back to India for most of them enhanced the EIC's commer- reproduction" (Furber 1976:249). From 1717 to cial performance. After 1710, the EIC (in 1727, boosted by another generous farman, the conjunction with private traders) specialized EIC's trade in Bengal more than doubled: From in reexporting India piece goods, and contributing 40 percent to the Company's total later, spectacularly, trade in tea, of which imports, it rose to 70 percent (Keay 1991:234; England's annual consumption per head rose also see Marshall 1976, chap. 1). At the same some six- or seven-fold between 1725 and time, Furber notes, the tonnage of English coun- more than doubled between 1724 and 1760 (Wilson 1965:308). try trade 1742. For more on the growth of the tea trade, For now, my argument treats the ascent of see Furber (1976:127, 244) and Nightingale the EIC and its corollary, the rise of sideline (1970). traders, as an exogenous variable. As such, it 24 As did the VOC servants in even the heart of posed a direct threat to the VOC by cutting the Dutch empire, on Java and (see into its profits and undermining its capacity Young 1969). 24 AMERICAN SOCIOLOGICAL REVIEW

THEDUTCH EAST INDIES COMPANY (VOC) THEENGLISH EAST INDIA COMPANY (EIC) Merchant-regents \ King-in-Parliament SEVENTEENGENTLEMEN

Privateand COURTOF DIRECTORS semiprivate Batavia traders

VOC factory VOC factory E IC factory EICfactory EICfactory VOC factory

Figure 2. Schematic Representations of Company Organizations: Emergent Links Provided by Pri- vate and Semiprivate Traders Beginning in the Late Seventeenth Century Note: Lines indicate any type of structuralconnection between organizationalnodes.

central nodes in the Company hierarchy, in- der Batavia in the hierarchy were simulta- cluding Batavia itself, relying on the English neously breaking away from Batavia's influ- instead, to their own (and their families') ence. This twofold structuralshift heralded a economic advantage. "If we allow the En- general breakdown of VOC cartel, or group glish freely to conduct this back and forth discipline. trade,"warned one prescient VOC officer, "it To counter the breakdown, the Seventeen would be an inwardly-eating cancer [for the Gentlemen had more than one logical . Company]" (quoted in Winius and Vink They could have proffered the carrot,by rais- 1991:58). The VOC was becoming a Hydra, ing wages. A variant of this strategy would a many-headed politico-economic organiza- have involved acknowledging agents' rights tion, with one principal at home, another in to private trade and then taking a cut of their Batavia, and still others in the periphery of agents' new-found fortunes, which was basi- the organization. Where was the new Her- cally the English solution. Or they could cules who could cut a few heads down to size have increased situational sanctions-in without slaying the beast? Figure 2 schema- other words, brandished the stick. A helpful tizes the emergent situation. analogy is to the reservation utility, the next- We have seen that betweenness made a dif- best package of resources for workers that ference in a node's capacity, in this case becomes available when alternative employ- Batavia's, to control resource inflows. When ment opportunities appear. When the reser- it comes to brokering outflows, the flip side vation utility increases, incentives (such as of how network position contributes to in- wages) or monitoring levels must also im- creased power, what matters more than prove if employee performance is to be sus- betweenness is closeness-the number of tained.25 nodes that a unit must traverse to reach other The Seventeen Gentlemen were not will- nodes (Marsden 1982; Freeman 1979). To be ing to increase incentives, and they experi- close to most other actors in a system is to enced heightened problems applying nega- have relatively little need for any particular tive sanctions. The Seventeen certainly had node's brokerage services. This dimension of the de jure right to fire agents whose malfea- centrality is correlated with autonomy, or in- sance reached levels that commanded their dependence: relative freedom from the con- straints imposed by having to depend on any 25 The concepts of reservation utility and reser- particular actor. In this sense, the Batavian vation wage are used in more than one way in the officials were growing more autonomous economics literature. This version derives from from their home base, while the officials un- Gintis and Ishikawa (1987). PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 25 attention. De facto power was another story. and influence with the Seventeen Gentlemen, The principals had always had problems as high officers linked to factions in power gaining accurate information because of the in the Netherlands used their positions to great distances involved and agents' years- dodge charges of corruption (Blusse 1986, long delays in filing reports; these monitor- chap. 8). Both stories are probably true. The ing problems were now exacerbated by the point is that the Company's metropolitan breakdown in cartel discipline. Periodic re- principals had never considered offering the ports of rampant corruption did provoke the carrot; now they were failing to wield the Seventeen to send representatives to deal stick as well. with the problem. What could be more un- At this stage, strategically situated VOC derstandable than sending new brokers to agents had generated sufficient power to act discipline old, errantones; and what less sur- as principals, and thus as potential competi- prising than the novices' speedy fall from tors of their principals. Given the opportu- grace? The first serious disciplinary effort nity, such agents could divert organizational was made in the 1680s, when the Seventeen resources through an alternative set of nodes sent the allegedly incorruptible Hendrik for their own benefit rather than that of their Adriaan van Rheede tot Drakestein to inves- principal. As some VOC agents began to do tigate the Indian factories and vested him this, articulating their economic ends inde- with extraordinary powers of punishment. pendently of their metropolitan principal's Van Rheede estimated that during a few short politico-economic goals, their actions years, from 1678 to 1686, graft and private forged new links in the colonial chain and trade by the VOC's Bengal servants had cost began to unravel the several strands of patri- the Company as much as fl. 3.8 million. The monial position. The agents' autonomous upshot was that a few highly-placed employ- economic activity was still directly secured ees lost their posts. Nevertheless, "as soon as by force, to be sure, but in some areas that van Rheede left Bengal, everything returned force was increasingly delivered by the En- to the old footing," as one highly-placed met- glish Company. The VOC was witnessing ropolitan VOC servant sourly observed.26 the genesis of new contractual relations Furthermore,the High Indies Government within its hierarchy,but the tacit contract fa- was able to insist that Van Rheede's investi- vored by some of its agents had been negoti- gations give a wide berth to Batavia. This ated with another metropolitan principal. success can be interpreted in two ways. It Depending on how one looks at it, Simmel's may have been a dramatic sign of Batavia's (1950) tertius gardens, or laughing third politico-economic muscle. vis-a'-vis the met- party, arrived in English garb, or in the form ropole, now solidly entrenched. In support of of the VOC's own agents. In any case, the this interpretation,note that Batavia also re- joke was on the VOC. Formal hierarchical sisted the Seventeen's renewed attempts to authority was still nominally in place circa establish direct connections that bypassed 1795, in the waning years of the Dutch an- the High Government's broker role between cien regime, but it was no longer an effec- Indies factories or merchants and the met- tive disciplinary device. ropole. The High Government squeezed out abolished the direct Amster- and, by 1718, CONCLUSION dam-Ceylon-Bengal link (Winius and Vink 1991:60). Alternatively, Batavia's success I have focused on a network of principal/ may testify to top colonial officers' links to agent ties linking the Dutch metropole-the merchant-regents and the VOC's Seventeen Gentlemen-to Batavia, its Indonesian out- 26 That servant was Company Advocate Pieter post that was, in turn, linked to many East Van Dam, cited by Prakash (1985:88). A later Indian factories. Metropolitan control was De Roo, came to seventeenth-century reformer, from the outset. virtue of their grief more dramatically. While on his tour of insecure By duty, he was apparently poisoned by resentful multivocal positions, the VOC's colonial VOC colonial servants. Gaastra (1991:95-97) re- agents tried to enrich and empower them- views attempts by the Seventeen to find emissar- selves, but at first the potential impact of ies who could discipline the colonial servants. their opportunistic actions was limited. The 26 AMERICAN SOCIOLOGICAL REVIEW agents could not exit the Dutch colonial sys- is a context for ties that cast shadows of com- tem, and they were constrained by metropoli- mitment" (p. 189). Dutch regents possessed tan control over cashboxes and promotions. lineage property in patrimonial state offices The rise of the English Companyput an end and their spinoff :VOC director- to VOC dominance. In part, straightforward ships passed from generation to generation commercial competition infringed on the as part of the political privilege controlled by Dutch company's bailiwick. But, as I have urban regent patrilines. And because family tried to show, the presence of the EIC also and kin relations are privileged conduits for opened new opportunities for Dutch colonial the strong sentiments on which normative servants to evade their own patrimonialchain. commitment is founded, metropolitan elites In particular, it nourished Batavia's central- tried desperately to keep their top colonial ity and autonomy, and allowed agents at pe- agents from pulling away and founding In- ripheral colonial nodes to step up their pri- donesian political patrilineages of their own. vate commercial and financial transactions, They failed, and power parcellized along now via the British and the Dutch hierarchies. family lines. Had they succeeded, it seems This heralded the breakdown of VOC cartel likely that agents' politically secured private discipline. In theory, such structuralalterna- accumulation would have been limited. tives could be engendered endogenously or The full implications of the principal/agent exogenously to any given network. problems for the Dutch colonial system are When any alternative path raises agents' too complex to explore here. For the metro- reservation utilities, all else being equal, pole, which has been my focus, the implica- agents can be expected to pursue the alterna- tions included lower profits, loss of military tive opportunities, in the absence of an effec- capacity and organizational flexibility, all of tive response on the part of metropolitan which undermined Dutch commercial hege- principals. Diverse responses were possible mony. These dynamics and implications within the framework of patrimonial princi- should be furtherexplored, for they were de- pal/agent relations. In contrast to the VOC, velopmentally consequential in the widest as we have seen, the English Company low- sense. The rise of trade conducted by Com- ered its resource dependence on any single pany men as well as interlopers-in fact, all colonial node, thereby augmenting its capac- private traderswho operated within the shell ity to keep any one node in line, while al- of patrimonialcorporations but who were in- lowing its agents to take a higher cut of the creasingly independentof their hierarchies- surplus, effectively raising agents' wages. paralleled and reinforced similar trends and The EIC stumbled on this two-pronged ap- tensions in the metropole. Clio is a known proach, which could only be dubbed a strat- ironist, and she delights in unintended egy in retrospect, and was relatively success- consequences. The world colonial network, ful for a time. formed by competing but unwittingly linked A crucial qualification concerns the scope charteredcompanies, was the womb in which conditions within which the above argument, revolutionary European challenges to patri- or any fundamentally utilitarian approach, is monialism were nurtured. valid. Agency problems are most likely to emerge when agents are not normatively Julia Adams is AssistantProfessor of Sociology committed to the organization. Contrary to at the Universityof Michigan.She is currently the thrust of the work of Simon ([1947] completinga bookon stateformation in Western 1961) and Williamson (1975, 1991), there- Europe,The FamilialState: Ruling Families and Merchant in the Modern should never be assumed to Early fore, uncertainty Netherlands,France and England(forthcoming, entail opportunism. This qualification is par- WilderHouse Series, CornellUniversity Press). ticularly important in the historical context of patrimonial systems, which depended not only on material incentives and force, but REFERENCES also on principals' and agents' pursuit of Adams, Julia. 1994. "Trading States, Trading family honor and position. "Agency is more Places: The Role of Patrimonialism in Early than a tie" in such situations: To borrow Modern Dutch Development." Comparative HarrisonWhite's (1985) evocative phrase, "it Studies in Society and History 36:319-55. PRINCIPAL/AGENT PROBLEMS IN EARLY MODERN COLONIALISM 27

Alexandrowicz, C. H. 1960. Treaty and Diplo- Networks 1:215-39. matic Relations between European and South Furber, Holden. 1976. Rival Empires of Trade in Asian Powers in the Seventeenth and Eigh- the , 1600-1800. Minneapolis, MN: Uni- teenth Centuries. Leiden, The Netherlands: A. versity of Minnesota Press. W. Sythoff. Gaastra, Femme. 1991. De Geschiedenis van de Algemeen Rijksarchief den Haag [National Ar- VOC. Zutphen, The Netherlands: De Walburg chives, ] 1.01.07:#1235, #1237, Pers. #1 244. . 1994. "Private Money for Company Arasaratnam, Sinnappah. 1958. Dutch Power in Trade. The Role of the Bills of Exchange in Ceylon, 1658-1687. Amsterdam, The Nether- Financing the Return Cargoes of the VOC." lands: Netherlands Institute of International Itinerario 18:65-76. Cultural Relations. Gaastra, Femme and J. R. Bruijn. 1993. "The . 1986. Merchants, Companies and Com- 's Shipping, 1602- merce on the , 1650-1740. 1795, in a Comparative Perspective." Pp. 177- , India: Oxford University Press. 208 in Ships, Sailors and Spices: East India Blusse, Leonard. 1986. Strange Company: Chi- Companies and their Shipping in the 16th, 17th nese Settlers, Mestizo Womenand the Dutch in and 18th Centuries, edited by J. R. Bruijn and VOC Batavia. , The Netherlands: F. S. Gaastra. Amsterdam, The Netherlands: Foris. NEHA. Boxer, C. R. 1965. The Dutch Seaborne Empire, Gintis, Herbert and Tsuneo Ishikawa. 1987. 1600-1800. London, England: Penguin. "Wages, Work Intensity, and Unemployment." 1979. Jan Compagnie in War and Peace, Journal of the Japanese and International 1602-1799: A Short History of the Dutch East Economy 1:195-228. India Company. Kong: Heinemann Asia. Gould, Mark. 1992. "Law and Sociology: Some Brenner, . 1993. Merchants and Revolu- Consequences for the Law of Employment Dis- tion. Commercial Change, Political Conflict, crimination Deriving from Sociological Recon- and London's Overseas Traders, 1550-1653. struction of Economic Theory." Cardozo Law Princeton, NJ: Princeton University Press. Review 13:1517-78. Chaudhuri, K. N. 1965. The English East India Greif, Avner. 1994. "CulturalBeliefs and the Or- Company. London, England: F. Cass. ganization of Society: A Historical and Theo- . 1978. The Trading World of Asia and the retical Reflection on Collectivist and Individu- English East India Company, 1660-1760. New alist Societies." Journal of Political Economy York: Cambridge University Press. 102:912-50. . 1986. "The English East India Company Hill, Christopher. 1961. The Century of Revolu- and Its Decision-Making," Pp. 97-121 in East tion: 1603-1714. : W. W. Norton. India Company Studies: Papers Presented Israel, Jonathan I. 1989. Dutch Primacy in World to Professor Sir Cyril , edited by K. Trade, 1585-1740. Oxford, England: Claren- Ballhatchet and J. Harrison. Hong Kong: Asian don. Research Service. Keay, John. 1991. The Honourable Company: A Coase, Ronald H. 1937. "The Nature of the History of the English East India Company. Firm." Economica 4:386-405. London, England: Harper Collins. Coen, Jan Pieterszoon. 1919. Bescheiden Omtrent Kiser, Edgar and Xiaoxi Tong. 1992. "Determi- zijn Bedrijf in Indie. Vol. 1. Edited by H. T. nants of the Amount and Type of Corruption in Colenbrander. 's-Gravenhage, The Nether- State Fiscal : An Analysis of lands: Martinus Nijhoff. Late Imperial China." Comparative Political De Korte, J. P. 1984. De Jaarlijkse Financiele Studies 25:300-31. Verantwoording in de VOC. Leiden, The Neth- Lazear, Edward P. [1987] 1989. "Incentive Con- erlands: Martinus Nijhoff. tracts." Pp. 152-62 in The New Palgrave: Al- Eggertsson, Thrainn. 1990. Economic Behavior location, Information, and Markets, edited by and Institutions. New York: Cambridge Uni- J. Eatwell, M. Milgate, and P. Newman. New versity Press. York: Norton. Ekelund, Robert Jr. and Robert D. Tollison. 1981. Levi, Margaret. 1988. Of Rule and Revenue. Ber- as a Rent-seeking Society. Col- keley/Los Angeles, CA: University of Califor- lege Station, TX: Texas A & M University nia Press. Press. Marsden, Peter V. 1982. "Brokerage Behavior in Elias, Johan E. 1923. Geschiedenis van het Restricted Exchange Networks." Pp. 201-18 in Amsterdamsche Regentpatriciaat. Den Haag, Social Structure and Network Analysis, edited The Netherlands: Martinus Nijhoff. by P. V. Marsden and N. Lin. Beverly Hills, Freeman, Linton C. 1979. "Centrality in Social CA: Sage. Networks: A Conceptual Clarification." Social .1976. East Indian Fortunes: The British 28 AMERICAN SOCIOLOGICAL REVIEW

in Bengal in the Eighteenth Century. Oxford, ganizations. Berkeley/Los Angeles, CA: Uni- England: Clarendon. versity of California Press. Marshall, P. J. 1976. East Indian Fortunes: The Taylor, Jean Gelman. 1983. The Social World of British in Bengal in the Eighteenth Century. Batavia. European and Eurasian in Dutch Oxford, England: Clarendon. Asia. Madison, WI: University of Wisconsin Meilink-Roelofsz, M. A. P. 1962. Asian Trade Press. and European Influence in the Indonesian Ar- Thompson, James D. 1967. Organizations in Ac- chipelago between 1500 and about 1630. The tion: Social Science Bases of Administration. Hague, The Netherlands: Martinus Nijhoff. New York: McGraw-Hill. Musgrave, Peter. 1981. "The Economics of Un- Thomson, Janice E. 1994. , Pirates, certainty: The Structural Revolution in the and Sovereigns. State-Building and Extraterri- , 1480-1640." Pp. 9-21 in Ship- torial Violence in Early Modern Europe. ping, Trade and Commerce: Essays in Memory Princeton, NJ: Princeton University Press. of Ralph Davis, edited by P. L. Cottrell and D. Van der Wee, Herman. 1981. "World Production H. Aldcroft. Leicester, England: Leicester Uni- and Trade in Gold, Silver, and Copper in the versity Press. , 1450-1700." Pp. 79-86 in Pre- Nightingale, Pamela. 1970. Trade and Empire in cious Metals in the Age of Expansion, edited Western India, 1784-1806. Cambridge, En- by H. Kellenbenz. Stuttgart, : Klett- gland: Cambridge University Press. Cotta. North, Douglass C. 1981. Structure and Change Watson, Ian Bruce. 1980. Foundationfor Empire. in Economic History. New York: W. W. English Private Trade in India 1659-1760. Norton. , India: Vikas. Prakash, Om. 1985. The Dutch East India Com- Weber, Max. [1922] 1968. Economy and Society. pany and the Economy of Bengal, 1630-1720. 2 vols. Edited by G. Roth and C. Wittich. Ber- Princeton, NJ: Princeton University Press. keley/Los Angeles, CA: University of Califor- Root, Hilton L. 1987. Peasants and King in Bur- nia Press. gundy: Agrarian Foundations of French Abso- White, Harrison C. 1985. "Agency as Control." lutism. Berkeley and Los Angeles, CA: Univer- Pp. 187-212 in Principals and Agents: The sity of California Press. Structure of Business, edited by J. W. Pratt and Rowen, Herbert H. 1988. The Princes of Orange. R. J. Zeckhauser. , MA: Harvard Busi- The Stadholders in the Dutch Republic. New ness School Press. York: Cambridge University Press. Wijnaendts van Resandt, W. 1944. De Geza- Schama, Simon. 1977. Patriots and Liberators. ghebbers der Oost-Indische Compagnie op Revolution in the Netherlands 1780-1813. New hare Buitencomptoiren in Azie. Amsterdam, York: Vintage. The Netherlands: Liebaert. Schutte, G. J. 1974. De Nederlandse Patriotten Williamson, Oliver E. 1975. Markets and Hier- en de kolonien. Een onderzoek naar hun archies: Analysis and Antitrust Implications. denkbeelden en optreden, 1770-1800. New York: Free Press. , The Netherlands: H. D. Tjeenk . 1991. "The Logic of Economic Organi- Willink. zation." Pp. 90-116 in The Nature of the Firm. Shapiro, Carl and Joseph E. Stiglitz. 1984. "Equi- Origins, Evolution, and Development, edited librium Unemployment as a Worker Discipline by 0. E. Williamson and S. G. Winter. New Device." The American Economic Review York: Oxford University Press. 74:433-44. Wilson, Charles. 1965. England's Apprentice- Simmel, Georg. 1950. The Sociology of Georg ship, 1603-1763. London, England: Longman. Simmel. Edited by Kurt H. Wolff. Glencoe, IL: Winius, George D. and Marcus P. M. Vink. 1991. The Free Press. The Merchant-Warrior Pacified. The VOC Simon, Herbert. [1947] 1961. Administrative Be- (The Dutch East India Company) and its havior. 2nd ed. New York: Macmillan. Changing Political Economy in India. Delhi, Stiglitz, Joseph E. [1987] 1989. "Principal and India: Oxford University Press. Agent." Pp. 241-53 in The New Palgrave: Al- Young, Robert. 1969. "The British East India location, Information, and Markets, edited by Company's Pepper Trade at Sumatra, 1730- J. Eatwell, M. Milgate and P. Newman. New 1760." Ph.D. dissertation, Department of His- York: Norton. tory, University of Pennsylvania, Philadelphia, Stinchcombe, Arthur. 1990. Information and Or- PA.