Patterns of Manumission in Lima's Hinterland

Amid the enormous variety of connections between countryside and city that we have seen, two questions remain. The first relates to the frequency of slave manumission on the hacienda and in the city; and the second concerns the significance and the trajectory of the changes.

The notarial record books and the cartas de libertad illustrate the frequency and methods of manumission (Table 8).^[70] If we take 1830 and 1850 as parameters, we observe an increase in the percentage of manumissions of slaves coming from haciendas: it went from 13.1 percent in 1830 to 21.0 percent in 1850 of all recorded manumissions in Lima. Rural manumissions were on the rise but urban manumissions

predominated and explain slaves' desire to abandon the hacienda and their strategies to relocate. Around 1840, in the years in which slave owners tried to reverse the trend toward manumission, the relative percentage of manumissions dropped; only about 5 percent of total manumissions were recorded in the rural area.

Self-purchase was the most common method of manumission on the haciendas and increased over the decades (from 41.2 percent in 1830 to 50.0 percent in 1850), and other methods (a master's grant of freedom—often conditional—and the intervention of relatives) declined. Among these, the mother intervened more and more often, as we have seen in the cases examined.

In the twenty years before slavery was abolished, the number of cartas de libertad registered in the notarial record books dropped in absolute terms, from 130 in 1830 to 106 in 1850 (the change was even greater for 1840: only 77 appear). But the speed of manumission was almost the same in 1830 and 1850, rising from 2.2 to 2.3, as the number of slaves decreased.^[71] The figures from the censuses of 1836 and 1845 show that Lima's slave population decreased from 10.4 percent to 6.9 percent of its total population. If we project the speed of manumissions over the percentage decrease of slaves, we see that the records of the cartas de libertad reflect the rhythm of decrease of the slave population in almost the same fashion. In other words, they not only verify the reliability of the figures but also show that slaves, through their own volition or ability to maneuver, were the agents central to this process. In 1850, acts by slaves—the total number of manumissions by self-purchase or purchase by a relative—account for 90.9 percent of all rural manumissions.^[72]

Even if the figures for Lima are above the average, it is not surprising that in broader terms, the conditions in Lima coincide with tendencies recorded elsewhere in Latin America:

A surprisingly high percentage of manumitted slaves were Africans in urban manumissions from Mexico City to Buenos Aires.... From 40 to 60% of the ex-slaves purchased their freedom, and one-third have been granted theirs free and unconditionally by their masters. The remaining 10 to 20% of the manumitted slaves had been granted conditional freedom, mostly having to do with continued demand for familial service. All recent studies have found that approximately 2/3 of the manumitted were women (from 60 to 67%) and few were found to be 45 years of age or older. (Klein 1986, 227)

Although maroonage and banditry were ways that some slaves used to free themselves from labor on the haciendas, day-to-day opportunities and attempts at permanent negotiation were the core responses of slaves on the road to freedom; these were the slave population's long-term options.

Maroonage and brigandage were considered crimes, subject to intermittent massive repression by governments—between 1760 and 1809 and between 1836 and 1839, coinciding toward the end of the colonial period with final attempts to strengthen ethnic segregation (Moerner 1969, 225) and in the 1830S to reinforce order on the haciendas—until slavery was abolished in 1854.^[73] Yet many times what the authorities termed an escape was nothing more than a search for protection (by a godfather, for example), a change of jobs (from rural work to day labor), or a protest against excessive or unlawful punishment. Slaves were attempting to deal with the conditions of rural slavery in legally defensible ways that owed as much to their involvement in the independence struggles as to their awareness of marches and countermarches against laws that were always imprecise and vague enough to permit the extreme alternative of escape.

The likelihood that slaves would use the methods of day-to-day and permanent negotiation we have described depended on various factors: occupational situation in Lima, contacts with persons (free or slave) who had urban experience or were willing to help, and circumstances on the hacienda. The actions of the Defensor de Menores, who represented the slaves in several of the cases examined, and—even more important—their chances of arrival in the parish of San Lázaro, where both social and working contacts (even with maroons) were established, illustrate that slaves quickly grasped opportunities for moving from countryside to city life.

Like their counterparts elsewhere in Latin America, female slaves were the earliest and most fortunate travelers from the hacienda to freedom in the city. They owed their success to individual strengths and to a family strategy intended to benefit all its members. As we have noted, the bond between mother and children could survive the painful setbacks and obstacles of the journey; on the hacienda female slaves were judged less productive and more costly (Haitin 1983, 167). Less useful on the hacienda, women had better possibilities for work and friendship in the city.

On the whole, countryside-city connections indicate a blurry image of the rural-urban boundaries and may help explain those "tricky boundaries" in the census information. At the same time, this haziness perpetuated mechanisms of vertical ascent (the slave population's internal differentiation) and geographic-horizontal mobility. In slave patterns of mobility, agents appear (family members in Lima, relatives of masters, state officials, bandits and maroons, owners of tambos ) who interceded or acted unwittingly in favor of slaves. Thus, given the vast array of conditions and actors, we cannot think of rural slavery as the defining factor of the prevailing slave relations. The rural sphere without the urban, particularly in the context we have seen, is a partial truth.^[74]

Finally there is room for reflection on the significance of the success of slaves in arranging rural-urban relations and on what the over-arching implications of this dynamic implied with respect to slave relations in limeño society.

We have seen that, given the resistance of individual slaves and couples and the support proffered by authorities upon satisfaction of the conque —as well as slaves' preference for the city—masters could not simply transfer their slaves from the urban to the rural realm. Only in cases judged as extreme could the punishment of relocation be ordered. Apart from the recorded opposition, an owner's greater or lesser control depended on the conditions on the hacienda (which as we have seen were varied) and on the relations slaves established with the city. All forces were not equal: what prevailed was movement from the countryside to the city.^[75]

In the rural sphere we record a great variety of productive units, from units where relative control over the labor force existed, to cases of an owner's depravity or utter neglect (interpreted by the contemporaries as alarming). The margins of "rebellion" diminish if we evaluate the priorities of slaves in terms of their access to freedom. The contemporaries' rendition seems exaggerated. A related issue is to determine if the transfer of slaves to the urban realm could be considered part of this "chaos," which contemporaries spoke of so readily. If it was masters who assigned slaves to earn daily wages, they passed the costs of upkeep to the slaves themselves or to another master, by putting the slaves into a socioeconomic framework better suited to exploit (and control) their work. Along with Scott (1988, 31), I believe that the appearance of "chaos" did not necessarily coin-

cide with the system's collapse. In the short term, at least, slaves' options could bring about new contradictions, which could even revive the slave system. In addition, from a logic of urban expansion—which occurred at a still relatively slow pace in Lima—change could generate an increasing demand for services (Algranti 1988, 23), especially if other ethnic groups sold more expensive services or pursued other circuits of consumption, and if the ethnic-occupational structure were very rigid. Consequently, slave owners had incentives to hire out slaves' labor.

All such expansion and change eventually involved a certain amount of chaos and entailed above all the rationalization of a system, with greater individual exploitation, and not its eradication. The question is, thus, if the transfer from a hacienda to earn daily wages in Lima was the result of a greater rationalization of the system. And it is at this point that we must restate an apparent paradox, between more daily wages and less control (rather than more slavery), which we can resolve if we evaluate the logic of the system not from the interests of the owners but from the viewpoint of the slaves.

We have no evidence to show that masters transferred their slaves to the urban area. More likely the slaves were the ones who took the initiative, although a slave's advancing years might make this alternative the most rational option from an owner's perspective. Our assertion challenges the hypothesis that a decline in productivity in Lima's rural zones caused a relocation of slaves to the city and that the transformations of the slave system reflected a greater rationalization by slaveholders to safeguard a threatened system. In actuality, what took place (at least in the first two decades of the nineteenth century) was an increase in the number of slaves on big haciendas. Furthermore, the generalization of rural stagnation in the Lima valley at the end of the eighteenth century and the beginning of the nineteenth—as Haitin (1983, 145) points out—is also questionable.^[76] Another fact that contradicts the view that it was masters who initiated the hiring-out system is the gender composition inside the haciendas. The ratio of men to women documents a conscious policy of retention of the work force and of an aversion to its transfer to Lima. A complementary argument along the same lines is what has been termed in other regions as the "peasant gap," the cultivation of basic foodstuffs that simultaneously solved three problems for owners: it gave slaves food, secured the labor force, and served as a counterweight to the expansion of

commercial crops for the international market (Cardoso 1988). We have seen that in Lima, and beyond Lima (in Huaito, for example), slaves used this peasant gap not only to solve problems for hacendados and to meet a rising urban demand but also converted it into a base of accumulation within the hacienda through the supply of local and regional markets.

What owners did in a marginal fashion (for example, by paying their slaves) and slaves did in a substantive manner (by filling the "subsistence gap," earning wages, buying freedom) was to introduce a component of monetary interdependence into the relations on the hacienda. This interdependence eroded the basis of the slaveholding authority, practically at the level of interpersonal relations. Perhaps the most dramatic expressions of this relation between money, slave price, and freedom were payments on credit and reductions of a slave's value, which were considered a posteriori payments of outstanding daily wages. All things considered, these are also very specific aspects of the lengthy processes of formation of both the rural and urban labor markets. In the following chapter we use the same perspective—that of the slaves—to observe how distinctly urban processes unfolded in both a similar and a contrasting manner.