Abstract
Prominent accounts of the transition to capitalism have a far too limited understanding of pre-capitalist agrarian economies’ potential for dynamism. Recent research shows that conditions earlier accounts identify as triggers for a transition to capitalism could be present without a transition occurring. I expand on implications of these cases of “missing transitions” for theorizing the dynamics of pre-capitalist agrarian economies. I present a theoretical framework that shows how phenomena previously associated with the transition to capitalism—such as flexible property rights in land and labor, extensive markets, and accumulation of industrial and mercantile wealth—emerged in pre-capitalist societies without leading to capitalist development. I illustrate the analytic upshot of this framework by considering the case study of Japan in the Tokugawa era (1603 to 1868). For historical sociologists, early modern Japan has long been seen as an anomalous case, puzzlingly mixing developments thought to represent early signs of capitalism with evidence of the durable survival of the feudal social order. In light of a more accurate account of the forces for and limits of dynamism in pre-capitalist agrarian economies, Tokugawa-era Japan is no longer a puzzling anomaly.
One of the enduring theoretical projects of historical sociology has been the attempt to describe and explain the emergence of modern capitalist economies. Generations of scholars have developed, criticized, and refined accounts of why and how capitalism appeared when and where it did. These accounts of the origins of capitalism include, at least implicitly, understandings of the kinds of societies out of which it could emerge. I contend that recent developments in research on the transition to capitalism call for a corresponding revision to understandings of pre-capitalist society. An accumulation of comparative case studies undertaken by historical sociologists since the 1990s have built the foundations of a revised understanding of the dynamics within pre-capitalist societies, including societies that failed to make an early transition to capitalism. What has not yet been undertaken is an attempt to synthesize the implications for a systematic theory of pre-capitalist agrarian states and economies. This article shows how this existing research can be recombined into such a theory and contributes to its further development by analyzing the case of early modern Japan, which previous comparative accounts have been forced to treat as anomalous. The upshot of this new theory is to show what processes brought about economic dynamism in pre-capitalist economies but also what limited it—because, as recent studies show, dynamism frequently appeared without being part of a transition to capitalism.
A complete theory of the transition to modern capitalism logically involves accounts of three elements: the starting point or points from which the transition begins, the economic ending point that needs to be explained, and the mechanisms that bring about the transition generally or in particular cases. The salient features of the ending point of modern capitalism that require explanation are uncontroversial: “an economic system based on private ownership of means of production and self-sustained, self-transforming growth” (Hung 2008:569). The “great debates” on the transition to capitalism surveyed by Emigh (2005) largely focused on the mechanisms that initially produced the recognizable features of this economic system. Competing theories offer alternative accounts of the distinctive conditions of early, successful cases of capitalist transition that drove their exceptional trajectories of economic development. Yet, despite pronounced disagreement over the crucial mechanisms of transition, there has been remarkably little debate about how to characterize the pre-capitalist economy.
This absence is important because a claim about the mechanisms that drive a transition to capitalism implies that a pre-capitalist economy in which those mechanisms are present will soon cease being a pre-capitalist economy at all. Thus, an upshot of the clutch of recent studies identifying cases of missing transitions—that is, what Emigh (1997) calls “negative cases,” in which conditions hypothesized to lead to capitalism appeared without a subsequent transition—is that historical sociology’s established image of pre-capitalist economies is far too limited. Pre-capitalist agrarian societies could be and were characterized by free labor, widespread markets, and substantial concentrations of industrial and mercantile wealth without entering a capitalist trajectory of growth. Recent studies amply demonstrate this fact and suggest revisions to existing accounts of the mechanisms of capitalist transition. In this article, I consider the implications of these findings for historical sociology’s understanding of pre-capitalist economic life.
Through a synthesis of the conclusions of recent research on the transition to capitalism as well as state formation and revolutions, I offer a systematic account of the internal political and economic dynamics and limits of pre-capitalist agrarian societies. The research I draw on focuses primarily on Europe and secondarily China. To demonstrate the analytic insights of this theory beyond these previously studied cases, I consider the case of early modern Japan.
Japan during the Tokugawa period (1603 to 1868) is striking because it fits so poorly into existing understandings of pre-capitalist economies and the transition to capitalism. Viewed from the perspective of political and social institutions and overall economic dynamics, Japan cannot be said to have begun a capitalist trajectory of development until after the political upheavals of the late nineteenth century. However, looking to the commercial economy, which increasingly extended into rural areas in the eighteenth and nineteenth centuries, it is equally hard to say that Japan was a feudal regime that successfully crushed any kind of economic dynamism. Previous comparatively informed interpretations of Tokugawa-era Japan tend to cast it as an exceptional case—either a genuine transition to capitalism that remained hidden underneath a feudal exterior, or a case in which tendencies toward capitalist development were derailed by tenacious feudal institutions. Either way, anomalous facts—whether the survival of apparently pre-capitalist institutions or the high level of commercialization—that would seem to contradict the conclusion offered are explained away. In contrast, put in the context of the research synthesized into a revised theory here, it becomes clear that Tokugawa-era Japan’s durably feudal political institutions, as well as its flourishing urban and rural commerce, were actually well within the typical bounds of variation of pre-capitalist agrarian economies.
Theories of the Transition to Capitalism
One can slice up and categorize accounts of the transition to capitalism in various ways, for instance, based on Marxist, Weberian, or institutionalist theoretical traditions. In fact, these traditions have a great deal of overlap. For the purposes of the argument here, the most salient point is the different conditions that each account identifies as crucial moments in the initiation of a transition to capitalism. I suggest that the conditions emphasized by the most prominent theories can be organized into three categories: property rights including free labor, markets, and the concentration of industrial and mercantile wealth.
The first argument holds that the removal of “inefficient” property rights, such as serfdom, and their replacement by flexible and enforceable individual rights to labor and the disposal of land, made it possible for profit to become the guiding force for the appropriation and allocation of resources. Scholars frequently argue that in at least some parts of Western Europe, the “right” kinds of property rights were established in the late-medieval and early modern periods (North and Thomas 1973). Lords’ capacity to own serfs’ labor eroded amid the demographic crisis of the Black Death and shifting fiscal strategies of expanding monarchies (Anderson 1974). Moreover, states, motivated by their own geopolitical and fiscal reasons, increasingly established predictable legal institutions that enforced property rights consistent with capitalist economic relations (Collins 1986).
Second, the importance attributed to market institutions is not so much whether trade happens, but the extent to which competitive relations of exchange, rather than authoritative regulation by a political organization, serve as the sinew holding a region together economically. There were many attempts to unify Europe into one “imperial” framework, but these plans failed decisively by the middle of the sixteenth century (Wallerstein 1974). Instead, the survival of multiple political centers encouraged a dynamic of political competition generating innovation and emulation (Collins 1990; Mann 1986). Indeed, this competitive political dynamic pushed states into cooperation with merchant classes (Tilly 1992) and, more generally, to develop “organic” ties with societies (Hall 1986).
Third, although frequently in connection with property rights and markets, scholars argue that the concentration of industrial and mercantile wealth in particular social and geographic spaces generated political and economic forces that pushed forward the development of capitalism. Cities are often presented as “the favored sites of capitalists,” produced by “the processes that accumulate and concentrate capital” (Tilly 1992:17). By this view, urban concentrations of capital represented the leading edge of economic development in medieval and early modern Europe, inflecting the economic and political trajectories of feudal states (Anderson 1974). Indeed, leadership passed from one center of accumulation to another in the long process of capitalist transition in Europe (Arrighi [1994] 2010).
The various theories of the transition to capitalism focus on different combinations of these conditions and sketch out different kinds of relationships among them, and the debates over these theories have contributed to a deeper understanding of the processes of capitalist development in Europe. However, as Emigh (1997) argues, there are methodological limits to an overwhelming focus on positive cases like Western Europe in the development and refinement of theory. When Western Europe is viewed in isolation or in terms of an exaggerated contrast with cases of “failed” development, it is all too easy to assume that conditions that happened to precede or accompany European economic development were indeed theoretically necessary and sufficient for the development of capitalism. To counteract this bias, Emigh calls for a heightened attention to “negative cases,” in which conditions that previous theories link to a “positive” outcome are in fact accompanied by the absence of a transition to capitalism. Emigh’s prescription entails a concerted effort to seek out cases that exhibit the various preconditions of capitalist development—the “right” property rights, markets, and the concentration of industrial and mercantile wealth—to confirm, or not, that the predicted outcome of capitalist transition occurred.
Emigh’s own research, as well as several other studies published in recent years, pursue this line of inquiry, and these studies have proved deeply problematic for many established theories of the transition to capitalism. This research has discovered the accumulation of mercantile wealth within flourishing markets, not constrained by inefficient feudal property rights in land or labor, in late-medieval Tuscany (Emigh 2008), post-Golden Age Holland (Lachmann 2003), precolonial South India (Chibber 1998), and eighteenth-century China (Hung 2008). Yet, in all these cases, commercial prosperity did not continually flow back into ever-expanding markets and increasing production. It is increasingly clear that this is a very widespread historical pattern. Goldstone (2002:342) notes the occurrence in many places and times in preindustrial economic history of “periodic ‘efflorescences’ leading to intensive rounds of per capita income growth, urbanization, extensive regional and international trade, and considerable population growth accompanied by stable or rising incomes and economic, political, and cultural specialization and complexity.” Yet, these “efflorescences,” despite being accompanied by conditions conventionally identified as direct antecedents of the beginning of modern, capitalist economic growth, were not followed by sustained economic breakthroughs.
These studies identify recurrent anomalies for the theories of capitalist development surveyed above. Based on their findings, these authors justly offered revisions that seek to correct earlier theories’ inaccurate predictions for their cases. These revisions expanded our understanding of the conditions for transition, but they paid less attention to the implications of these findings for the question of how to conceptualize the pre-capitalist starting point for the theory of capitalist development. Indeed, contributions to the debate on the transition to capitalism are often quite schematic about just what it is that capitalism replaced. The economies that existed before the emergence of capitalism are all too often defined negatively, as the absence of whatever is taken to be the defining features of capitalism (Emigh 2005). However, insofar as recent studies cast doubt on the conditions that earlier theories identified as sufficient for a transition to capitalism, they also suggest that these theories’ accounts of pre-capitalism are too narrow. Far from being a distinctive harbinger of capitalist development, flourishing commercial activity emerged in a number of cases that were neither capitalist nor developing toward capitalism. Yet, how did such economic dynamism emerge, if only temporarily? I now pull together the findings of a growing body of research on premodern economic relations and political institutions to develop a framework that accounts for the limited dynamism of pre-capitalist economies.
The Limited Dynamism of Pre-Capitalist Agrarian Societies
A Definition: Elite Extraction and Peasant Production
This section presents a revised theoretical framework describing the dynamics and bounds of variation for pre-capitalist agrarian societies. This framework does not overturn the large quantity of high-quality research on the processes of economic and political transformation prior to the consolidation of industrial capitalism and national states in the nineteenth century. Instead, because this research tends to proceed by an accumulation of case studies that challenge and revise existing theories in different subfields—notably the transition to capitalism, state formation, and revolutions—I aim to synthesize the findings of this research into a revised, systematic theory of the political economy of pre-capitalist agrarian societies.
The basic identifying features of such societies can be summarized by two simple premises: (1) majorities of economic actors and activity are involved with agricultural production undertaken by peasants who directly provide for most of their own subsistence needs, often within the framework of communal organizations, and (2) through political means, elites derive their income from claiming a share of the product of economic activity (Brenner 2007). To borrow an alternative but complementary characterization from Polanyi (2001), the typical pre-capitalist agrarian society combines redistribution from producers to elites with patterns of reciprocity and householding among the peasantry (cf. Polanyi 1957). 1 To say that elites use “political means” to “redistribute” income means elites make demands, ultimately backed by the threat of coercive force, on the labor services, goods, or money held by others. As a corollary, elites’ ability to realize their claim to income does not depend primarily on the market sale of commodities at competitive prices. Because peasants make up the majority of the productive population, they are most often the objects, or should we say victims, of elites’ political claims to income. 2
One could imagine an ideal-typical feudal manor, in which a lord forces peasants to spend a substantial amount of time working the lord’s demesne and performing other kinds of labor service. It should be obvious that this lord would find it immensely difficult, acting individually, to maintain and enforce claims over the peasants. At a bare minimum, the lord would need lackeys to threaten or carry out violence against peasants—just as a mafia boss needs enforcers. Yet, even with lackeys, the lord would face a number of problems. The peasants are not totally helpless: they can flee or try to organize passive or active resistance. The lord must also worry about betrayal by lackeys and the threat of invasion by other, perhaps larger and better-trained or better-equipped groups of thugs. The upshot of this thought experiment is that organization is key: first, organization of elites for the extraction of income facing off against organizations of peasants (and others) seeking to minimize that extraction; second, organization of a group of elites to adjudicate and enforce their competing claims to political power; and third, organization of elites of one territory to defend their control or extend it at others’ expense.
Existing research has done quite a lot to illuminate variations of the three axes of the political organization of pre-capitalist societies as well as their interrelation. “Predatory” monarchical states are one relatively common form of extractive organization; their fiscal strategies have been analyzed as a kind of “bargaining” with subject populations (Kiser 1987; Levi 1981) or with the administrative staff on which rulers rely (Kiser 1994; Kiser and Baer 2005). Of course, feudal or imperial monarchies are by no means the only state-like organizations for elite extraction in pre-capitalist societies, in the sense of wielding political authority or coercive capacity (Lachmann 1989), and exigencies of the struggle between extractive claims and producers’ resistance can influence the form and relations among such organizations (Anderson 1974) and vice versa (Lachmann 1990). Moreover, even if a particular state organization emerges as the predominant, if not monopolistic, arbiter of political claims and holder of coercive capacity, pre-capitalist states are, in practice, infested with elites vying to patrimonially appropriate pieces of its political capacity for themselves and their families (Adams 2005; Lachmann 2009).
Previous accounts that take a wide-lens view of premodern societies largely agree that their basic pattern can be sketched in terms of variable configurations of various strata of elites drawing resources from, but not deeply penetrating, a largely subsistence-oriented, communally organized peasantry (e.g., Anderson 1974; Hall 1986). The framework presented here departs from earlier accounts in the potential for flexibility and dynamism within pre-capitalist societies. Signs of economic dynamism in medieval and early modern Europe were assumed to be unique forebears of the subsequent emergence of capitalism, and scholars often focus on the question of which features of European feudalism account for this (Emigh 2005). However, more recent historical research shows that the mercantile development of early modern Europe was by no means exceptional (Goldstone [2002] surveys this literature). Instead, significant commercialization and other kinds of economic dynamism were recurrent features of pre-capitalist economies. The rest of this section lays out the mechanisms that typically produced, but also limited, dynamism.
Sources of Economic Dynamism and Political Change
Pre-capitalist societies are not static. Even if the political relations among elites and between elites and subjects coalesce into a relatively stable form, the interacting strategies of states, other elites, and direct producers generate dynamics that, in turn, put pressure on temporarily stable institutions. This subsection lays out four major forces of dynamisn in pre-capitalist agrarian economies.
The upside of the demographic cycle
For elites and the state, more population and more output simply mean a larger pool of resources from which to extract income, and so they are usually quite willing to support growth of this kind. This means, especially, the encouragement of land reclamation. Peasants are often eager to participate in new settlement to increase their own holdings or to set up children on their own land; this is often the only “old-age insurance” peasants have access to (Brenner 2007). Thus, in periods of political stability and favorable ecological conditions, pre-capitalist agrarian economies tend to increase in population, area under cultivation, and total economic output (Goldstone 2002). The thirteenth and sixteenth centuries in Europe were notable peaks in this type of extensive growth (North and Thomas 1973). Wong (1997) describes eighteenth-century China as flourishing in a similar way.
Yet, the precariousness of survival in preindustrial economies and the absence of means to control epidemics periodically results in large-scale decreases in population (Goldstone 1991). Depopulation and the subsequent disruption of economic and political institutions typically leads, in turn, to resettlement and renewed extensive growth. As is often emphasized with reference to the Black Death in late-medieval Europe, the initial period of regrowth, characterized by relatively higher wages and lower rents, is also an opportunity to introduce more productive economic organizations and more effective fiscal institutions (North and Thomas 1973). As a concrete instance, Hopcroft and Emigh (2000) find that despite their very different subsequent economic histories, both Tuscany and eastern England saw consolidation of holdings and improvements in cultivation flowing from the drastically reduced population to land ratio in the late fourteenth century. 3
States, cities, and commerce
Commerce can not only coexist with the political and coercive extraction that is the basis of pre-capitalist elites’ positions, but it in fact fills a key role in the logic of elites’ strategies of reproduction. In some cases, peasants and artisans directly provide elites with food and other goods (Weber [1922] 1978). The more common pattern, however, is for elite incomes to be received in either the staple agricultural product alone or in cash; in either case, elites must exchange this income for the other goods they need—the weapons and luxury consumption goods that Makki (2011:164) calls “the means of coercion and persuasion.” The concentration of purchasing power in the hands of elites thus supports the emergence of specialized merchants and artisans to supply them and their entourages, and the degree of state centralization tends to increase the depth of commercialization—more populous capital and provincial cities and larger armies, administrations, and courts draw resources from a larger hinterland and can support more specialized markets. 4 Increasingly centralized states with larger armies and more urbanized elites in turn become more politically and financially dependent on merchants (Anderson 1974; Ertman 1997; Tilly 1992). Rural areas are rarely completely spared from commerce, because they are the sources of food for urban populations and raw materials for artisans. And as land-labor ratios decrease, surplus peasant labor can be cheaply diverted to by-employments—that is, part-time work in cottage industry and commerce undertaken alongside agricultural cultivation (Brenner and Isett 2002).
The politics of trade networks
Commercial flows themselves become the object of political competition among elites and states. Trading and finance often concentrate on particular hubs due to what economists call “economies of scale” and “reduced transaction costs” (North and Thomas 1973). Such concentrations represent a lucrative source of taxes and finance for states, especially when compared to the logistical challenges of gathering revenue from peasants dispersed over an extended territory. For their part, merchants (and landed elites who are often not averse to a little commercial investment on the side) are more than willing to pay taxes or offer credit in return for state support in pursuing the lucrative status of being a mercantile and financial hub (Tilly 1992). In medieval and early modern Europe, the leading position in trade networks shifted over time, from northern Italy to the Low Countries and eventually to England (Arrighi [1994] 2010; Lachmann 2003; Wallerstein 1974). Although this process of migrating economic and political hegemony is sometimes linked to the long-term emergence of capitalism in Europe, the pattern is in fact far more general. It can be seen also, for instance, in the special relationship between monarchies and Jewish populations in the Middle Ages (Barkey and Katznelson 2011) or the “symbiotic relationship” of “warrior nobility and trading merchants” in successive polities in northeast Africa (Makki 2011:179).
Adaptation and cooptation
Extensive growth and shifting patterns of trade often result in the accumulation of wealth in the hands of new or previously marginal groups. States, especially when faced with the acute fiscal demands of war, are usually eager to give political and status recognition to changes in the distribution of economic resources. Merchants of commoner origin who want to reorient themselves to the state are given opportunities to do so (Hung 2008). States hungry for cash will often trade privileges and titles to get it (Ertman 1997)—the massive apparatus of venal offices in Absolutist France is a particularly well-known example (Bien 1987). In general, states try to co-opt potentially lucrative supporters or useful servants into existing elite relations, even as incumbent elites retain their own status and privileges and assert their distinction from newcomers.
Limits to Dynamism and Pathways of Crisis
These four forces of dynamism run up against characteristic limits and counteracting forces. Even if the upswing of the demographic cycle often generates “efflorescences,” the dynamic of extensive growth is not benign in the long-run. In pre-capitalist agrarian economies, increases in population sooner or later push up against the availability of arable land and the limitations of existing production techniques, leading to rising grain prices that make it harder for states and elites to provision their cities and soldiers (Goldstone 2002). Peasants unable to support themselves on small plots turn to migration and by-employments for extra income, creating a population of the semi-employed floating poor and adding extra pressure on the wages of established artisans (Goldstone 1991). Moreover, pre-capitalist commerce and industry are not immune to the increasing economic pressure associated with the cresting of the demographic cycle; urban merchants’ and artisans’ fortunes tend to rise and fall with those of their elite consumers and peasant suppliers (Lachmann 2000). Population growth tends to strain the grain supply, raising the cost of food for urban residents and suppressing the share of income left over for the purchase of manufactured goods (Brenner 2007).
The history of modern capitalism encourages an expectation that markets will continually expand to include new actors, products, and regions (Sewell 2008). However, the economic life of pre-capitalist societies provides few signs of such an “inherent tendency of markets toward excrescence” (Polanyi 2001:60). Neither the peasants who make up the majority of the productive population, nor the elites who hold political power, rely primarily on market exchange to secure their livelihoods. Elites in pre-capitalist agrarian societies depend on their politically mediated capacity to claim income; as a result, elites almost always prioritize the stability of their positions within political institutions, even at the expense of commercial competitiveness and economic productivity (Lachmann 2003). Moreover, given the opportunity, elites typically attempt to convert commercial success into political position in the form of status privileges or state office (Hung 2008; Lachmann 2000). As for peasants, although proto-industrial by-employments and some cash crop cultivation are often compatible with asserting and defending their rights to the means of their subsistence, they are highly unlikely to attempt to increase their market involvement and competitiveness if this entails a substantial risk of disruption to their continued subsistence (Brenner 2007). Even when peasants are drawn economically closer to highly commercialized urban centers, the immense imbalance of economic power between them and elite merchant-landlords can paradoxically drive peasants out of land and product markets (Emigh 2008). In summary, markets in pre-capitalist economies fail to continually expand because actors have the option, and often the incentive, to withdraw from exchange.
These demographic and economic tendencies have important political consequences. If demographic expansion eventually reduces per capita agricultural output and generates destabilizing marginal populations, and the commercial economy also stagnates, then the pool of resources over which elites struggle stops growing or even shrinks. Incumbents of powerful political and economic institutions—from privileged nobles to the patrician merchants of established entrepôts—will likely clutch more tightly to their positions, contributing to an ossification of political and economic institutions that further strangles economic growth (Goldstone 2002). This conflict is further exacerbated insofar as the number of claimants to elite privileges increases, due to natural growth and the cooptation of educated officials and new wealth. These conflicts can develop in any number of patterns, including central state versus provincial or local elites (Barkey 1994), overlapping factions of officials (Lachmann 1989), mutually exclusive religious camps (Gorski 2003), or, perhaps most explosively, marginal elites versus well-entrenched incumbents (Goldstone 1991). The emergence and intensification of such battle lines increases the likelihood of the breakdown of the state into rebellion or civil war.
Ultimately, there are limits to the flexibility of pre-capitalist political and economic institutions to adapt to changing conditions. The entrenchment of competing elites narrows the margin of adaptation, but geopolitical conflict plays an important role in exposing the transition from flexibility to brittle rigidity. Military competition among neighboring states pressures states to develop their fiscal capacities and internal organizations (Tilly 1992). However, these pressures are mediated through previous institutional legacies, with the result that new fiscal and military challenges are frequently met by adding to an increasingly complex institutional assemblage (Ertman 1997). This can lead elites to become increasingly sluggish in responding to new threats, resulting in a state’s fall in relative geopolitical or economic position (Lachmann 2003, 2009). An alternative, perhaps even more dangerous, possibility is that states provoke acute resistance from elites and broader populations as they attempt to respond to the pressure of military competition by reorganizing their fiscal, administrative, and military apparatuses (Skocpol 1979). In short, alongside economic problems and domestic elite conflict, an additional consequence of the limits to dynamism in pre-capitalist agrarian societies is increasing vulnerability to geopolitical and military challenges.
To summarize, societies in which elites obtain income by extracting resources by political means from a predominately agrarian economy contain typical forces for dynamism but also countervailing tendencies. This section outlined four. First, demographic patterns are a two-edged sword: the upswing of the cycle is often a period of incredible dynamism, but population growth typically outstrips the productivity of agriculture, generating mounting economic and political pressure. Second, elites’ and states’ extractive strategies can encourage extensive growth and an expansion of commerce, but the interplay of their coercive claims and producers’ defensive strategies constrains intensive economic growth and binds the commercial economy to the fluctuations of the agricultural sector. Third, far from shunning the market, ostensibly feudal states can proactively compete over control of trade routes, but limits to the expansion of the commercial economy make this an intensifying struggle over a static pie. Fourth, political institutions adapt to changing conditions to some extent, but tendencies toward economic stagnation and deepening elite competition, potentially exacerbated by geopolitical threats, place tensions on political institutions that they cannot endure. Thus, while pre-capitalist agrarian societies contain internal mechanisms that generate the kinds of dynamism that scholars conventionally associate with the beginning of a transition to capitalism, they are also characterized by mechanisms that limit that dynamism, generating the apparent puzzle of “missing transitions.”
The Puzzle of Early Modern Japan
Japan is often invoked as a key case of relatively successful, late capitalist industrialization outside Western Europe and its settler-colonial offshoots. Japan’s experience of political upheaval, economic development, and geopolitical ambitions beginning in the latter half of the nineteenth century has been analyzed in comparison with the earlier capitalist industrial powers of Western Europe and the United States as well as other late developers in Asia (Eisenstadt 1996; Goldstone 1991; Mann 2012; Moore 1966; Skocpol 1979; Trimberger 1978). While many of these studies include some analysis of the Tokugawa-era background in discussing Japan’s modern development, a relatively smaller number of scholars look directly at the pre-1853 period in a comparative lens with early modern Europe (Anderson 1974; Collins 1997; Ikegami 1995, 2005; Ikegami and Tilly 1994). Whether considering the Tokugawa era in its own right or primarily as background to the transformations set off by the opening of ports in the 1850s, historical sociologists who examine this era are confronted with a seemingly paradoxical mix of phenomena. On the one hand, Japan in the eighteenth and early-nineteenth centuries seemed to fulfill many of the preconditions for capitalist development as identified by predominant theories of the transition. On the other hand, the overall character of political and economic institutions as late as the mid-nineteenth century was unmistakably feudal.
To spell the puzzle out in more detail, late Tokugawa-era Japan was characterized by (1) de facto freedom of peasant labor and agricultural land from restrictive feudal property rights, (2) widespread market institutions linking rural areas with the central cities, and (3) substantial accumulation of wealth in the hands of merchants, first in the cities and later in rural areas. These characteristics would seem to fulfill the conditions for a transition to capitalism, as identified in the theories summarized earlier. However, just as in other recent case studies of missing transitions, these conditions in Japan did not set off a sustained process of capitalist development prior to the opening of the ports and subsequent political transformations. First, although tenancy and commercial by-employments increased in frequency among the peasantry—despite attempts by the nobility to forbid both—peasant households retained a primary orientation toward subsistence agriculture, without major technical changes increasing per capita productivity, and peasants were supported by communal institutions that regulated cultivation techniques and tenancy relations. Second, even in the nineteenth century, the anchor for the market remained the tax-supported consumption demand of the feudal nobility. Indeed, rural areas’ deepening involvement in this market in the latter half of the period was achieved at the expense of many existing urban commercial centers. Third, merchants certainly accumulated wealth even as feudal rulers struggled to maintain, let alone increase, their tax revenue, but increased merchant wealth did almost nothing to challenge the political ascendancy of the feudal order. Indeed, the feudal elite recaptured some of the wealth it failed to extract as agricultural taxes through irregular taxes and fees, forced borrowing, and manipulation of the currency.
This apparently paradoxical combination of phenomena has confronted historical sociologists with an unenviable theoretical dilemma. Did Tokugawa-era Japan achieve a transition to capitalism or not? The most common response, which follows the consensus of postwar Japanese historiography, is to insist it did not and to explain away developments that are acknowledged as genuine sprouts of capitalist development (Anderson 1974; Furushima 1963; Moore 1966; Norman [1940] 1975; Takahashi [1953] 2001; Trimberger 1977). By this view, commercial activity was diverted into a symbiotic relationship with the feudal order, and thus the potentially disruptive force of commercial wealth was suppressed (Moore 1966). An alternative interpretation, drawing on revisionist economic historiography, claims that Tokugawa-era Japan was fundamentally capitalist and the tenacity of feudal relations was illusory (Collins 1997; Francks 2006; Hanley and Yamamura 1977; Howell 1992; Sanderson 1994). Indeed, according to this perspective, it “was a substantially modern society[, and] its troubles were largely those of a market-dominated economy” (Collins 1997:861). However, this view requires effectively ignoring the unchallenged political supremacy of the warrior nobility and its continued ability to claim upward of one-third of the country’s agricultural output as well as demand contributions from and cancel debts to merchants and commoner landlords with impunity. On both prongs of the dilemma, one set of phenomena or another—either the developments that are agreed to be preconditions to capitalist development or the evidence that pre-capitalist economic and political institutions remained strong at the end of the period—becomes theoretically inconvenient and needs to be explained away. In light of the predominant theories of the transition to capitalism, there is no escaping the appearance that Tokugawa-era Japan was an anomaly.
However, this appearance is an artifact of existing theories’ too-narrow conception of the dynamics of pre-capitalist economies. The weakening of feudal controls on agricultural labor and land, the geographic spread of markets, and the accumulation of industrial and mercantile wealth are not necessarily signs of a coming capitalist transition. Instead, as outlined in the theoretical framework presented in the previous section, mechanisms typical of pre-capitalist agrarian societies produce these developments but also limit the process of commercialization and economic growth. The result is the pattern of episodes of economic “advance” followed by periods of “stagnation” or “retreat” that is so puzzling for many theories of the transition to capitalism. This section uses this revised theory of pre-capitalist economic and political dynamics to resolve the apparent paradox of Japan’s missing transition.
State Formation in Seventeenth-Century Japan
From 1603 until 1868, the Tokugawa shogunate was the dominant political authority in Japan. What is a shogunate? In theory, shogun was a title granted by the emperor in Kyoto to his chief military commander. 5 In practice, the Tokugawa shogunate was a feudal monarchy. The shogun’s authority over the rest of the ruling warrior nobility was based on the other lords swearing fealty to him, in return for the shogun confirming their authority. The result was a patchwork political map. The Tokugawa house—that is to say, the feudal monarchy—directly controlled an immense territory, including land assessed at about one-quarter of the agricultural productivity of the country, as well as the three great cities of Edo (now Tokyo), Osaka, and Kyoto and the southwestern port of Nagasaki. The rest was divvied up as fiefs among almost 300 magnate lords (called daimyo, “great names,” in Japanese). Each magnate lord ruled his domain—collecting taxes, deciding and enforcing laws, and so on—with little interference from the shogunate in Edo, although the monarchy periodically demanded contributions from the lords for projects ranging from riparian works to maintenance of the Tokugawa house shrine. Yet, Tokugawa rule did impose important restrictions on the lords: limits on their military capacity, prohibitions on private relations among lords, and control over the succession of each lordly house. The latter included the requirement that lords’ wives and heirs live at all times in the monarchy’s capital of Edo and the lords themselves spend every other year there (Hall 1991). The administrative staff and coercive manpower by which lords ruled and taxed their domains was provided by samurai—hereditary retainers to each lord, marked by the privilege of wearing two swords. The Tokugawa house itself had its own band of retainers and ruled its territories in essentially the same way as the lords did theirs (Totman 1967).
The monarchy, lords, and samurai of the Tokugawa-era political order lived off taxes collected from the peasantry, mostly as a share of the rice harvest. In the late-sixteenth and early-seventeenth centuries, the monarchy and the various lords conducted cadastral surveys that listed the productivity of the landholdings of each peasant household. The cadastral surveys became the basis for tax assessments, but responsibility for payment fell collectively on the village, so the monarchical or domain authorities did not care how the tax burden was distributed within the village as long as the total bill was paid (Smith 1968). Indeed, as long as peasants paid their taxes and did not become restive, the monarchy, lords, and their samurai subordinates kept themselves at arms’ length from village life (Befu 1968). Moreover, during the first century of the era, the lords progressively weakened the individual ties of samurai to rural areas by requiring them to live in the lords’ castle-towns, heavily regulating the administration of their fiefs, and ultimately, in most cases, converting fiefs into stipends paid out of domains’ general tax revenue (Brown 1993; Hall [1966] 1999).
Although samurai continued to be organized into military units (Totman 1967), the Tokugawa-era political order was if anything a contraption for the degradation of military capacity. To be sure, the monarchy and the lords maintained forces for putting down peasant rebellions and were never really at risk of overthrow on that front (White 1988). However, the Tokugawa regime discouraged lords from developing military capacities beyond this basic level; for instance, the regime issued regulations strictly limiting lords’ building of fortifications (Hall 1991). On paper, samurai were supposed to use their stipends to maintain arms and a fixed number of troops, to be called up at the lord’s command, but by the middle of the Tokugawa period, the entirety of samurai income (and then some) was taken up by consumption expenses (Yamamura 1974). As early as the suppression of the last great armed rebellion at Shimabara in the 1630s, commanders were lamenting the woefully poor equipment and skills of samurai “warriors” (Odo 1975).
The key point is that the institutions of Tokugawa rule successfully avoided military conflict within the class of lords and with other states. Beginning in the mid-seventeenth century, the Tokugawa shogunate monopolized and restricted Japan’s trade and diplomatic relations with the outside world (Hall [1966] 1999). From the perspective of Tokugawa rule, this isolation policy removed the potentially destabilizing threats of individual lords gaining diplomatic support or military resources through contact with foreign powers and of foreign trade disrupting domestic commodity supplies and prices. Indeed, these threats were realized in the 1860s after the opening of the ports and the erosion of Tokugawa hegemony.
Developmental Dynamics of the Tokugawa-Era Political Order
The last echo of the civil war of the sixteenth century was silenced in the battle of Osaka in 1615, and the last armed rebellion involving members of the warrior nobility was put down in 1638. From this point until the 1860s, Tokugawa hegemony kept elite conflict in check, and as long as it did so, the political order was fundamentally stable and secure. However, Japan was not politically and economically stagnant for these nearly 250 years.
The upside of the demographic cycle
The first half of the Tokugawa era was an immense boom driven by extensive growth and the stabilization of political institutions. Even in the later stages of the sixteenth-century civil wars, as warlords began to consolidate larger, more centrally controlled territories, they invested—directly and by providing tax incentives to wealthy commoners—in irrigation works and land reclamation to increase agricultural output and thus tax revenue (Yamamura 1981). This greatly accelerated once the pax-Tokugawa began in the seventeenth century, although population grew considerably faster than the area under cultivation and agricultural output. From roughly 1600 to 1721, the population grew 261 percent, from 12 million to 31.28 million people, the area of arable land grew 142 percent, and agricultural output grew 162 percent (Miyamoto 2004). 6
The warrior nobility, cities, and commerce
Under the Tokugawa-era political order, lords generally required their samurai vassals to live exclusively in the lords’ castle-towns, rather than on rural fiefs, and the Tokugawa shogunate required magnate lords to live every other year in the shogunate’s capital Edo, leaving their families there as hostages when the lords returned to their domains (Hall [1966] 1999). The concentration of elites regionally in magnate lords’ castle-towns, and nationally in the shogunate’s capital of Edo, created large centers of consumer demand. 7 Because taxes were predominately paid in kind by peasant villages, the monarchy and lords turned to merchants to sell rice collected as taxes and to buy other commodities. The monarchy and the lords offered various privileges, including exemption from taxes, to the merchants and artisans who gathered in their capital cities and provisioned them and their retainers (Nakai and McClain 1991). On the national level, rice and other products flowed out of the regions where they were produced and into the great population centers of central Japan, providing consumption goods and raw materials for the urban merchants and artisans and financing the lords’ consumption in Edo (Miyamoto 2004). For much of the period, the merchants of Osaka managed this circulation of commodities and money, and the most successful among them accumulated massive fortunes as the primary financiers of the lordly class (Crawcour 1968).
Growth of this commercial circulation, both within domains and in the country as a whole, did not leave rural areas untouched. Urban merchants developed rural hinterlands as sources of raw materials—cotton and indigo for textiles, soybeans for soy sauce and tofu, and so on. In the first half of the Tokugawa-era, the areas immediately surrounding the castle-towns and great cities—particularly the Kinai region around Kyoto and Osaka—provided many of these materials, which urban artisans worked into finished products (Hauser 1974). In addition, beginning in the eighteenth century, because of the cheapness of surplus peasant labor in the context of decreasing land-labor ratios, merchants increasingly encouraged rural cottage industries to substitute for more expensive urban labor (Kwon 2002). The result was a proliferation of commercial agricultural and cottage industry as by-employments for peasant households that remained, on the whole, focused on self-sufficiency (Saitō 2005a). Commoner landlords, often the same wealthy families that served as local officials, took a leading role in this process of commercialization, and over time, their role in superintending the circuit leading from raw material production to finished consumer goods destined for castle-towns and Edo expanded at the expense of the urban merchants (Pratt 1999).
Politics of the trade network
The fundamental fiscal problem faced by a lord and his officials was that their tax income came from agriculture in their domain, but a large, even predominant, source of their expenses came from their commodity consumption in Edo. What they needed, rather than more tax-rice, was purchasing power in the Japan-wide commercial network that provisioned Edo. Thus, improving the position of their domain’s subjects in that commercial network by the promotion of cash crops or handicrafts that could be sold to other parts of the country, such as cotton cultivation or textiles, would improve the domain’s fiscal situation. Even in the seventeenth century, many lords taxed the specialized commodity production that existed in their domains, but from roughly the mid-eighteenth century onward, more and more domain administrations experimented with policies to promote the development of new cash crops and handicrafts and to manage the marketing of domain exports (Nishikawa and Amano 2004; Ravina 1999; Roberts 1998).
Adaptation and cooptation
Political institutions and elite relations adapted in two main ways to these developments. First, to implement fiscal reforms, including commercial policies, lords increasingly promoted relatively low-ranked retainers into powerful positions in domain administrations. For the retainers, especially those with little status or wealth, education in domain schools—founded in growing numbers by lords for just this purpose—or study with prestigious teachers was the mark by which they tried to distinguish themselves as deserving of promotion (Dore [1964] 1984). This produced a competitive dynamic in domain administrations that would have important consequences for the political process of the Tokugawa regime’s terminal crisis in the 1860s (Cohen 2014).
Second, the alliance between domain administrations and rural commoner elites was strengthened. From the beginning of the Tokugawa era, the feudal administrations named wealthier landowners as village officials, responsible above all for ensuring the tax bill was paid each year (Befu 1968). In the eighteenth and nineteenth centuries, when domains sought to promote new export industries, they turned once again to wealthy commoners, especially landlords. After all, they were the rural economic actors who had the resources and ties with the peasantry to experiment with new products, and they were sometimes even the initial proponents of domains’ commercial policies (Roberts 1998). Moreover, recognizing the growing wealth of landlords and rural merchants, lords realized they could be a lucrative source of “loans” and “gifts”—that is, irregular taxation. 8 In return, domain administrations granted them status privileges (or confirmed privileges they already claimed to have), such as having family names or carrying swords (Pratt 1999), and authorities generally turned a blind eye to the transfer of land from smallholding peasants to their wealthier neighbors through foreclosure on loans, which was technically illegal (Kwon 2002). 9
Tensions of Late Tokugawa-Era Japan
The boom in population, cultivated land, and production—not to mention lords’ tax revenues—lasted to the early eighteenth century. However, even during the seventeenth-century expansion, population growth outstripped arable land and output growth, and when the latter all but stopped in the early decades of the eighteenth century, serious problems emerged for the agrarian economy and, by extension, lords’ fiscal situations. By the time population growth leveled off around 1720, agricultural output per person had dropped over 30 percent from its level in 1600 (Miyamoto 2004). 10 This situation was exacerbated by several periods of poor weather and natural disasters in the middle and latter half of the century, leading to sharp drops in population in some regions (Sippel 1998). However, there was no general demographic crisis. Other regions did not lose population, and even some relatively hard-hit areas eventually recovered from eighteenth-century losses (Roberts 1998).
This was possible because peasant households found ways to work more intensely to make up for the decreased availability of land per person. One element of this was more meticulous methods of cultivation that raised yields per area of land (Saitō 2010:258; Smith 1959); a second tactic was planting new crops, including cash crops (Saitō 2005a). By-employments in commerce and cottage industry was a third and crucial element (Howell 1989; Nishikawa 1987; Smith 1969). However, that peasant households adopted more “industrious” labor patterns “never implies they were willing to substitute the purchase of commodities at markets for the production of those goods at home” (Saitō 2005a:38). Studies of local village records show that households with less land tended to plant more cash crops and engage in more by-employments; moreover, when these households were able to secure more land, they tended to withdraw from commercial activities (Hauser 1974; Kwon 2002). All in all, the commercial by-employments and cash crop cultivation of peasant households merely supplemented, rather than significantly supplanting, their focus on subsistence cultivation.
From the perspective of the overall economy, the increase in the number of peasant households whose members worked part-time producing textiles and other consumer goods came largely at the expense of the urban mercantile centers that had previously produced those goods (Saitō 2005b). Production and trading, as well as population, shifted from cities to rural areas, in both the merchant metropolis of Osaka and the castle-towns of individual domains (Hanley and Yamamura 1977; Hauser 1974). In the nineteenth century, even leading areas of rural commercialization, such as the Kinai region around Osaka, began to lose out to other provincial areas (Saitō 1978). More generally, rather than particular regions progressively developing industrially or commercially, different regions cyclically rose and fell in commercial prominence. The expanding rural commerce of the latter half of the Tokugawa period reflected domain administrations’ and local elites’ vying with each other and with the privileged urban merchants to secure a place in the commercial network that had emerged in the first century of the Tokugawa era (Miyamoto 2004). However, the basic character of that network did not change, and lordly demand, concentrated in the castle-towns and above all Edo, remained the anchor of the whole system.
Even as the majority of peasants struggled in the face of reduced availability of land, wealthy commoner landlords and rural merchants prospered in the eighteenth and nineteenth centuries. Indeed, the narrowed margin of peasant subsistence cut in their favor, because they made loans backed by poorer peasants’ land to cover subsistence and tax shortfalls in bad years; when borrowers with already overstretched budgets could not pay back their loans, the land became the de facto property of the lender. In this way, land ownership grew increasingly concentrated in the latter half of the Tokugawa era (Kwon 2002; Smith 1959). At the same time, rural elites’ mediating position between feudal rulers and the peasantry gave them the leverage to thwart attempts by the feudal monarchy to enforce the nominal prohibition on alienation of peasant holdings (Kwon 2002) and to deflect increases in taxes away from their expanded holdings onto the peasantry’s already burdened shoulders (Ōguchi 2004).
Yet the accumulation of land by rural commoner elites, even when accompanied by increasing mercantile wealth, did not generate a progressive deepening of market production in the countryside. The expansion of tenancy in the eighteenth and nineteenth centuries usually buttressed the subsistence agrarian economy by giving poorer peasants access to an adequate base of land (Saitō 2010). Even as land ownership concentrated, units of cultivation remained highly fragmented, because landlords rented land in small strips to households with insufficient plots (Araki 1990). Likewise, even when wealthy landlord-merchants hired workers to cultivate their fields or to weave cotton or press oil, these enterprises remained basically auxiliary to—and used the same techniques as—the subsistence production of peasant households (Furushima 1963). Moreover, just as peasants planted cash crops or worked in cottage industry as by-employments, proto-industrial investment remained a sideline to rural elites’ main enterprises of collecting rent and serving as local officials. The success of a given landlord household in any one of its ventures—even in the most dynamic commercial sector—was often transient, and maintenance of various sources of income was a necessary form of insurance (Pratt 1999).
Underlying these phenomena was the fact that landlords were constrained by the continuing importance of the peasant community. “There seems to have been some institutionally determined controls of village sentiment which obliged landlords to keep rent low in order to uphold tenant reproduction” (Kwon 2002:71; cf. Saitō 2009:173). Early Meiji-era surveys of tenancy relations confirm that many places had customary limitations on eviction and expectations of rent adjustment in response to poor harvests (Waswo 1977). Tokugawa-era peasants’ expectations for the economic behavior of local elites were expressed through various forms of contention—including highly organized punitive “house-smashings”—that grew in frequency in parallel with peasants’ decreasing economic security and local elites’ increasing involvement in lending and commerce (White 1995). The institutional basis of peasants’ non-market subsistence strategies remained strong through the end of the Tokugawa era, and lords, merchants, and landlords accommodated rather than challenged these strategies.
Even if the intensification of peasant labor prevented an outright economic crisis, the strained economic conditions of the eighteenth and nineteenth centuries fueled tensions among elites. The exhaustion of extensive growth in the early eighteenth century meant that if lords wanted more income, they would have to increase tax rates, but they were trying to siphon more out of a shallower pool. Many attempts to do so were met with peasant protests large enough that lords were forced to settle for only relatively small increases in revenue (White 1988). However, it was generally not the lords who suffered most from the fiscal problems of their domain administrations. Instead, lords transferred much of the burden to their retainers, who from the early eighteenth century increasingly had to “lend” a sizable portion of their stipends back to their lords (Bolitho 1991). Thus, as lords increasingly appointed relatively low-ranked retainers into powerful offices to implement reforms, these promotions represented the only outlet for low-status warrior nobles to move out of rather difficult economic circumstances. Domain administrations in the nineteenth century alternated between contending cliques of officials, and although this competition usually did not threaten lords’ authority, it would come to provide the organizational basis for the movement that ultimately overthrew the entire Tokugawa-era political system in the 1860s (Cohen 2014).
Domains’ reform programs, especially their attempts to implement commercial policies, also brought them into conflict with the feudal monarchy. Domain commercial policies threatened the monopoly privileges granted by the feudal monarchy to the merchants of Osaka, which had been expanded in the eighteenth century in return for monetary contributions and promises to cooperate with the monarchy in controlling prices of consumer goods in Edo (Hauser 1974). In this latent conflict between the commercial interests of the Tokugawa and the various lords, the advantage largely fell to the latter. By the early nineteenth century, the monarchy backed away from maintaining the privileges of its favored Osaka merchants, and the lords had clearly established the precedent for essentially unlimited autonomy in commercial policies (Bolitho 1974). By the 1840s, the balance of power had shifted to the point that an outcry by lords easily scuttled a plan to move several lords to different domains, even though such moves had been common occurrences in the early seventeenth century. The most activist lords began to agitate for formal approval to build up their own military capacities and to have more direct input on national policy (Bolitho 1974).
Nonetheless, it was not until the foreign crisis of the 1850s and 1860s that the mounting economic and political tensions exploded into overt conflict. As noted earlier, the isolation of Tokugawa-era Japan from geopolitical relations (and conflict) removed an important source of pressure on the regime. However, if isolation contributed to the stability of the Tokugawa-era political order, it had the side effect that when the United States, and then the major European powers, demanded that Japan open itself up to diplomacy and commerce, the feudal elites who ruled Japan were utterly unprepared to respond. In the sixteenth century, Japanese warlords had been near the worldwide cutting edge in technology and tactics (Ikegami 1995), but the rulers of nineteenth-century Japan were ignorant of the improvements in firearms, artillery, and drilling that developed over the subsequent three centuries of European warfare. Japan needed large-scale expenditures and military reorganization to even begin to catch up with the Western powers, and this was severely constrained by the feudal organization of political institutions. The shares of tax revenue granted to samurai retainers by the various lords, including the Tokugawa house itself, were supposedly tied to those retainers being able to muster troops at the lords’ request, but by the late Tokugawa era, this scheme of feudal military levies was an utter fiction (Totman 1980).
Summary
This analysis shows how and why the apparently puzzling juxtapositions of the late Tokugawa era came about. First, the impressive degree of commercialization, even in rural areas, is the central focus of claims to identify tendencies toward capitalist development in Tokugawa-era Japan. Observing the regional variations in commercial activity, scholars interpret more commercialized areas as the leading edge of market expansion, in contrast with the continued survival of feudal political and social institutions (e.g., Hanley and Yamamura 1977). However, changing patterns of commerce do not necessarily presage a progressive conquest of the market, as long as producers and elites avoid market dependence. Indeed, Japanese peasants broadly limited their market activity to supplementing their communally defended access to subsistence. The expansion of rural commerce in Tokugawa-era Japan was generated by the confluence of strategies adopted by various groups, all of whom were acting to defend and advance their positions within the existing class structure.
Turning to the other side of the puzzle of early modern Japan, scholars frequently suggest that in the eighteenth and nineteenth centuries, the forces of feudal rule and capitalist development came into increasing conflict, generating the fiscal problems and social unrest characteristic of the period. However, the analysis here shows that the problems faced by the feudal regime from the early-eighteenth to mid-nineteenth centuries were not the result of the emergence of new economic actors and relations. Instead, the end of extensive agricultural growth and the high plateau of the population-land ratio that followed from it were the root causes of the fiscal problems that dogged the feudal elite throughout the latter half of the Tokugawa era. The conflict that accompanied these problems was not one between feudal and capitalist economic relations, but instead between different factions of elites over the distribution of the agrarian surplus extracted by political means. Magnate lords cut their vassals’ stipends. Rural elites used their influence in the distribution of taxation within villages to shield their expanded landholdings from increased taxation. Lords and the feudal monarchy struggled over whose preferred merchants and producers would have privileged positions in the commercial network, and hence who would have the authority to siphon off rents from the flow of goods in it. In general, the winners of these battles were the magnate lords and rural elites, and the result was a decentralization of economic and political power at the expense of the Tokugawa and established mercantile centers. Yet, this did not free commerce from feudal shackles; it only shifted the key sites of the symbiotic relationship between merchants and feudal elites from the Osaka region to provincial domains and rural areas.
Conclusions
Until the relatively recent past, agricultural production organized around household subsistence and communal reciprocity was the predominant form of economic activity in human societies, and elites used political means to redistribute the fruits of this production. In the past few hundred years, a very different set of economic relations—market-centric, capitalist, and industrial—has proliferated throughout most of the world. Nonetheless, modern capitalism should not be seen as the teleological endpoint of earlier economic activity. Historical sociologists need to understand the dynamics of pre-capitalist economies on their own terms.
This article laid out a theoretical framework identifying four major dynamic forces in pre-capitalist societies as well as the counteracting forces that limit them. All four dynamics and their limits are apparent in the case of Japan in the Tokugawa era. First, seventeenth-century Japan enjoyed a dynamic upswing of the demographic cycle, but both peasant livelihoods and elite incomes were put under sustained pressure when the cycle crested in the eighteenth century. Second, the Tokugawa era brought impressive commercial flourishing, most pronounced at first in the urban centers but increasingly shifting into the countryside. Indeed, rural commercialization augmented peasant subsistence as land became scarce. However, increasing rural commerce largely came at the expense of the urban centers, and it remained primarily an adjunct to peasants’ subsistence strategies. Third, the feudal rulers of Tokugawa-era Japan responded to and encouraged the commercialization trend by competing over the control of trade flows. Yet, this became a zero-sum conflict, in particular between the feudal monarchy of the Tokugawa shogunate and the magnate lords, establishing political fault-lines that would rupture in the 1850s and 1860s. Fourth, political institutions adapted to the changed environment, as lower-ranking samurai were promoted within domain administrations and the relationship between commoner landlords and the feudal elite strengthened. However, these adaptations were merely tacked onto an unreformed feudal hierarchy, the brittleness of which was dramatically revealed when the forced opening of Japan created a need for major military reforms.
In short, the analysis here shows how Tokugawa-era Japan’s combination of commercialization and persistent feudal institutions, which puzzled previous accounts, was not abnormal but instead reflected processes seen in many other pre-capitalist societies. This study complements recent research on the transition to capitalism that shows conditions previously thought to be clear signals of capitalist development—such as flexible property rights in labor and land, widespread markets, and accumulation of industrial and mercantile wealth—can in fact be identified in many cases that did not experience an early transition to capitalism. Whereas these recent studies focus on revising our understanding of the mechanisms that lead to modern capitalist growth, I focus instead on the question of how pre-capitalist economies repeatedly produced phenomena that, according to the previously dominant theories, should only have appeared in economies on the verge of capitalism.
The framework presented here is applicable to a wide range of historical cases. However, I do not claim that every pre-capitalist agrarian economy develops according to a fixed template. Instead, the forces for dynamism and their limits can appear in variable combinations. An interesting limit case that has received much attention from scholars over the years concerns regions specializing in luxury manufacturing and long-distance trade that are politically independent from the territorial states whose elites are their main customers—such as the merchant republics of medieval and early modern Europe, including Florence and the Dutch United Provinces (Adams 2005; Emigh 2008). Elites of these republics did not draw their income primarily through the political domination of their cities’ hinterlands. For instance, as Emigh argues, Florentine merchants leveraged their substantial monetary wealth, derived from the high profits of long-distance trade, to economically dominate Tuscan peasants through the land market. Because it was this dominance combined with merchants’ pursuit of the higher profits available in their urban enterprises that forestalled the development of a domestic market, Emigh (2008:210) claims it was a “capitalist, not a feudal, dynamic” that led to the eventual stagnation of the Tuscan economy.
From the perspective of the empirical and theoretical analysis of this article, one can accept Emigh’s account of the Tuscan case while also asserting the background role of pre-capitalist dynamics on its trajectory of development. Enterprises of the groups who specialized in trade in pre-capitalist societies were quite often, to borrow a phrase from Weber ([1905] 2001:29), “of a purely business character.” This was as true of the merchants of Tokugawa-era Osaka as it was of Renaissance Florence. Conversely, Florentine merchants were constrained by the wider pre-capitalist economy in similar ways as their Japanese counterparts were centuries later. As long as they held an advantageous position in the trade flows that provided consumer goods to elites in return for the money (or grain) elites extracted through taxation and rents, merchants could earn impressive returns whether or not consumer and factor markets were thriving in their own localities. However, as argued earlier, precisely because trade was so lucrative, other groups of merchants sought to break in, often with the support of their local political elites, and because of constraints on the growth of the wider economy, merchants competed over a market of relatively fixed size. The merchants of Florence and Osaka faced and lost out to this kind of competition. When this happened, even if they had been adventurous profit-seekers in their heyday, they retreated to secure what economic position they could as comparatively passive rentiers. Thus, even when profits in the enterprises that had previously distracted them from the local economy shrunk, the result was not a redirection of investment into, for instance, agriculture, but instead overall economic stagnation or decline, as seen in Tuscany by the end of the fifteenth century, the Netherlands beginning in the late seventeenth century, and Osaka in the early nineteenth century (Adams 2005; Emigh 2008). 11 In this way, even apparently capitalistic merchant republics are constrained by the dynamics of pre-capitalist agrarian economies—no less than states dominated by landed elites, albeit in somewhat different ways.
Finally, although my analysis is angled primarily toward providing a fuller picture of pre-capitalist agrarian economies on their own terms, a better understanding of pre-capitalist dynamics is also important for making sense of the transition to modern capitalism, especially beyond its initial emergence. Lagging economies were confronted with capitalism as a world-historical fact manifest in the superior economic and fiscal-military capacity of the leading capitalist states. France faced this problem in the late eighteenth century, struggling to remain competitive with Britain while experiencing increasing political and economic stress from a lagging agrarian sector and a tangled web of fiscal institutions infested by patrimonial elites—ultimately producing the crisis that led to the French Revolution. 12 The dilemma of the French monarchy attempting to hold its own geopolitically against Britain was widely repeated, and if anything only intensified as the gap between the leading industrializing and lagging agrarian economies grew over time.
The collapse of the Tokugawa regime and its replacement by a centralizing state following the Meiji Restoration of 1868 was one such later example. As argued earlier, Japan by the mid-nineteenth century was experiencing an intensification of conflict that was quite typical of pre-capitalist agrarian economies. Yet, due to the intervention of the Western powers that forced the opening of Japan, the specific political problems faced by Japanese elites in the midst of their quite traditional crisis were unavoidably modern in their content. Imperialist nation-states with capitalist economies speak the military language of pre-capitalist political orders—indeed, more loudly and fluently than pre-capitalist states themselves. A pre-capitalist state—such as the feudal monarchy of the Tokugawa shogunate—threatened by the Western capitalist powers in the nineteenth century felt a competitive pressure to meet its adversaries’ military and fiscal capacity. Meeting this competitive imperative was not just a matter of adopting new weapons, tactics, or fiscal-military organizations; successful adoption of those things hinged on a much deeper transformation of political and economic structures.
The forces that limit dynamism in pre-capitalist agrarian economies are also, by implication, blocks that must be circumvented for the beginning of modern capitalist development. Achievement of sustained per capita productivity growth in the rural sector, including agriculture, is crucial (cf. Senghaas 1985). However, getting to this outcome requires cutting the Gordian knot of peasant subsistence strategies and elite extraction. As long as elites remain entrenched in—and competing over—political institutions as the basis of their income, the economic impasse is unlikely to be crossed. Following the Meiji Restoration, the Japanese state’s key contribution to economic growth was not so much its ability to divert agrarian income into industrial investment and to nurture the emergence of an “urban entrepreneurial elite” (Hung 2008:582). Instead, it is more significant that the Meiji state, in the process of neutralizing potential competitors to its newly established rule, liquidated the institutions through which the warrior nobility had maintained its rule over rural areas and in their place constructed a far less patrimonial relationship with the rural population, especially with the commoner landlords. Not necessarily intentionally, this facilitated an impressive acceleration of rural economic growth, in which landlords played a driving role, alongside renewed urbanization (Francks 2006; Waswo 1977). In contrast, as Hung (2008) shows, the gradual weakening of the Qing regime in China was accompanied by an intensification of the close but economically counterproductive ties between landed gentry, merchants, and state officials that were the basis of the imperial political system. A similar contrast can be made with the typical policy adopted by the colonial regimes in Africa of relying on “customary authority,” which had crippling long-term consequences (Mamdani 1996).
The spread of capitalism has been highly uneven over time and space and variable in terms of institutional form. These uneven and variable processes involved complex interactions with existing political institutions and economic trends, which were themselves not static but the product of histories of development and crisis. More research is needed to flesh out the bounds of variation of pre-capitalist economies and political institutions and to investigate the impact of these variations on later processes of capitalist development. The study of pre-capitalist economies can and should aspire to the sophistication of research in comparative political economy that traces the alternative configurations of key institutions in different contemporary capitalist economies under the shared constraints of firm profitability and world trade and analyzes how different institutional set-ups lead to different trajectories of growth. 13 The theoretical framework presented in this article is meant as a step, and hopefully an inducement for further work, toward this goal.
Footnotes
Acknowledgements
The author would like to thank Vivek Chibber, Jeremy Cohan, Ihsan Ercan Sadi, Erik Van Deventer, and the anonymous reviewers for their comments on previous drafts.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
