Chapter Three
In the City

We have watched many members of the black population—free or slave—travel from the country to the city. For individuals and families alike success or failure depended on a multiplicity of strategies and links. The Lasmanuelos family gave us an example that many other slaves traced; now we turn to their experience in the city to compare their aims and strategies on the way toward freedom with those they used in Lima. For men and women the experiences of journey and arrival were distinct. Economic and racial heterogeneity, as well as the slaves' ties to the haciendas, the city, and their owners, underlie the differences. What circumstances awaited them and how did they live?

In 1792 the city of Lima had 56,627 inhabitants and 3,641 houses, of which 2,797 belonged to private individuals and the rest to religious entities, including hospitals and cofradías (mutual-aid societies or sodalities established by the Church and dedicated to the cult of specific saints). Life revolved around the small marketplaces and seven parishes (Bromley and Barbagelata 1945, 76 ff.). The census taken by Egaña in 1790 (cited in Romero 1980, 31), which considered only those living in the precinct of the city of Lima, found that of a total of 3,287 persons in monasteries and similar residences 1,060 were slaves.^[1] In 1828 the reported number of houses was 3,380 and of entrances, 10,605 (Haitin 1983, 102). ^[2] An earthquake that occurred in 1828 probably caused the number of houses to decrease, even though the number of entrances increased. In 1857, two years after public gas lighting first illuminated Lima (Fuentes 1866, 506), the population had almost doubled, with a concomitant increase in demand for goods and services. Lima's black population represented 44.7 percent of the total inhabitants in 1792; by 1818 this number had decreased to 38 percent. And more than a century later, in the 1940 census, the number of blacks recorded on the national level would reach only the figure of 29,054—a statistic that equals less than 0.5 percent of the total population (Labarthe 1955, 15). Intermarriage and aspirations to

whiter skin explain the long-term diminution of the black population. During our period, however, other factors account for the decrease in this population.

In 1792 slaves represented 25.6 percent of Lima's residents and only 15.8 percent in 1818, dropping to 10.5 percent in 1836, and to 6.9 percent in 1845. The gradual decline resembled that recorded for the viceroyalty and republic: in 1792 slaves were 3.7 percent of the total population, in 1854 only 1.1 percent (Jacobsen 1974, 82–84). Although the traffic in slaves began in the sixteenth century and increased throughout the colonial period, the slave population's natural growth did not compensate for the high mortality and low fertility rates or for self-manumission. The demographic shifts suggest that Peruvian slavery was a system highly accessible to social mobility. If we consider the reduction of the black population in Lima's rural sector after the wars of independence and also remember that a substantial part of this population relocated to the city of Lima, we recognize that the rate of manumission after 1821 continued to grow.

The Complexion of Black Society

Urban society offers more fertile ground for social conflict than the countryside ever can. Its internal divisions result equally from authorities' strategies of divide and conquer and from the beliefs of the varying actors involved in this process. The city offers more opportunities for internal differentiation than the countryside, both because of its complex social relations and because of its organization of labor.

A society is never homogeneous in a racial, cultural, social, or economic sense. As the criteria to determine and shape segregation develop, they define not only how persons will interact but also what form society will take. The black population found in segregation by skin color a means of conscious opposition to the dominant white sector and a mechanism of individual and collective survival. Slave and free black populations developed hierarchies and perceptions that differentiated darker-skinned blacks from lighter individuals, poorer from richer, and free blacks from slaves. In Lima, as in other places, ethnic diversity among blacks and differences in social status explain the absence of slave revolts (Karasch 1987, 325), or of revolts by all blacks against the domination of whites.

At the bottom of the black social pyramid were the bozales , newly arrived from Africa. They were the group least touched by the criollo experience; white owners preferred them for their docility. Even the criollo blacks who had lived in the colonial territory for some time described bozales as less refined. General Miller, who wrote his memoirs in 1829, noted that "the black criollo believes himself superior to his brother brought from Africa." At the apex of this pyramid were the quarterones or quinterones , who thought of themselves as white and had the greatest chances of obtaining freedom. Lighter-skinned women were the ones most likely to marry mestizos or whites.

The lot of bozales was not enviable. Those who survived the high mortality rate of the Atlantic trade to be unloaded at their port of destination were handed over to the representative of slave asentistas (who held long-term contracts from the Crown to buy or sell goods or services) or, barring their presence, to the boat's captain.^[3] Eventually, slaves were transferred to the port's barracones , where the conditions were often more horrifying than those on the boat; many slaves took their last breath here. The survivors were taken from quarantine in the barracones to a central location where they were grouped according to size, physical condition, and gender. They underwent a medical examination so that a report about their health and skills could be written up; the objective was to appraise them and classify their respective duty payments. This process, referred to as palmeo —literally, measuring by hands—culminated with the sear of an iron brand that served as permanent proof that all slaves had been brought into the country legally. Sometimes this procedure was repeated with the asentista 's emblem in order to prevent the theft of slaves before they were sold. Once these formalities were concluded, blacks were ready for sale. Petty merchants from near and far came to the marketplace; slaves would accompany them on yet another journey to their final destination: house, hacienda, trade, mine, or textile mill (Rout 1977, 69 ff.).

This process fragmented the identity of black inhabitants. First, violent removal from the natal environment and subsequent encounters with men and women belonging to different African cultures weakened a slave's chance for communication with other human beings. Next, dispersal of the slaves to different units of production in the colonial territory brought them face to face with other Africans and with an additional network of laborers from distinct and equally dis-

parate origins: indigenous and casta . The dual rupture required, as a response, the recreation of lost social spaces. The more fortunate bozales were those who remained in the urban environment. The city, more than the countryside, offered a realm where slaves might keep any social contacts established during the passage, and whatever they still held of their African identity and culture. A particular tribal group might predominate on the transatlantic journey; the preservation of these ties would ease the reordering of slaves' social space in the new setting. Reorganization along tribal lines could take a long time. During this process many slaves joined cofradías , where members could worship a particular saint within their cultural and ethnic traditions. The cofradías fulfilled a broader social function in helping blacks reconstruct their identity. But even within the refuge of family relations or cofradías , conflicts always beset the process, intrinsic conflicts that hampered collective endeavors. Through an examination of incidents within the cofradías and (in chapter 4) of matrimonial strife, we observe the different levels of conflict that simultaneously expressed notions of hierarchy—in class, ethnicity, and gender—within black society at large.

In 1615 there were fifteen cofradías in Lima, while in the rest of the viceregal territory, the black presence in cities and towns was so low that the creation of other cofradías was not warranted. Until approximately 1630, cofradías received any and all blacks; twenty years later we find that three groups existed, differentiated by skin color. The first rupture occurred between criollo ladinos and bozales . The former had learned Spanish and were familiar with Spanish customs. They insisted on breaking away from the black bozales . The second rupture arose between mulatos and criollo ladinos and resulted from the pressures of a third black generation that had established sexual and matrimonial alliances with women from the indigenous population. The monopolization of power within the cofradías by whiter mulatos paralleled their increasingly superior placement in other arenas of society. These ruptures slowly transformed the cofradía into an institution that housed a black elite (Bowser 1977, 307–311). Almost always the most visible manifestations of internal rupture were racial; racial tensions accompanied any competition over jobs within black society.^[4]

The social and individual function of the cofradías , as well as the relations between owners and cofradías , are illustrated in the account of the traveler Stevenson:

In the suburbs of San Lázaro are cofradías or clubs belonging to the different castes or nations of the Africans, where they hold their meetings in a very orderly manner, generally on a Sunday afternoon; and if any of the royal family belonging to the respective nation is to be found in the city, he or she is called the King or Queen of the cofradía , and treated with every mark of respect. I was well acquainted with a family in Lima, in which there was an old female slave who had lived with them for upwards of fifty years, and who was the acknowledged Queen of the Mandingos, she being, according to their statement, a princess. On particular days she was conducted from the house of her master by a number of black people, to the cofradía , dressed as gaudily as possible; for this purpose her young mistresses would lend her jewels to a considerable amount, besides which the poor old woman was bedizened with a profusion of artificial flowers, feathers and other ornaments. Her master had provided her a silver sceptre, and this necessary appendage of royalty was on such occasions always carried by her. It has often gratified my best feelings, when Mama Rosa was seated on the porch of her master's house to see her subjects come and kneel before her, to ask her blessing and kiss her hand. I have followed them to the cofradía and have seen her majesty seated on her throne, and go through the ceremony of royalty without a blush....

The walls of the cofradías are ornamented with likenesses in fresco of the different royal personages who have belonged to them. The purpose of the institution is to help those to good masters, who have been so unfortunate as to meet with bad ones; but as a master can object to selling his slave, unless he proves by law that he has been cruelly treated, which is very difficult or next to impossible, the cofradías raise a fund by contributions, and free the slave, to which the master cannot object; but this slave now becomes tacitly the slave of the cofradía and must return by installments the money paid for his manumission.^[5]

Indoctrination and religion certainly did not cease to be the central objectives of the cofradía , but shortly after their diffusion into the New World cofradías also assumed a social role and were helping to liberate slaves from their owners, advancing money for manumission or mediating as the slaves searched for a new owner, extending credit for the most abused and discouraged members of the slave population, and serving as a temporary refuge. At the same time cofradías became a cultural emporium for the black population's most privileged sector and imitated the larger society's dominant layers, which in turn often supported these associations. This support often led owners to loan clothes and jewels and to attend the ceremonies as spectators. But many other slaves chose not to turn to the cofradías . An almost end-

less litigation filed by slaves in their quest for freedom did not mention the intermediation of cofradías or note the case of any liemeño hacienda slave who knocked at the door of a cofradía .

Despite the attractions of their elite status within black society, cofradías also reproduced within their ranks the divisions of black society. In theory each cofradía corresponded to one African ethnic group and excluded the rest but in practice racial justification often meant that the whitest among the blacks occupied the highest posts in a cofradía 's internal hierarchy, or economic motive made the richest into creditors for its needier members. Above all, cofradías accommodated the master-slave relation. They came to function as a parallel judicial system; owners could turn to the associations not only to purchase new slaves but also to request that their slaves be punished for robbery, idleness, cursing, or running away (Bastide 1969, 90). For owners, an additional advantage of their intermediation was that owners could occasionally avoid the lengthy petitions and high costs of official court litigation.

Cofradías exist today and still have an essentially black complexion. We find cofradías described by travelers at the beginning of the nineteenth century, by which time racial splits seem to have subsided, perhaps because slaves had more possibilities of moving up the social ladder without becoming white. At this point different criteria began to determine the occupant of a cofradía post: the preoccupation with pigmentation gave way to a more intense interest in legal status. As these positions carried power both in ecclesiastical and social realms—age and legal status, or rather, experience and seniority came to be perceived as central characteristics for the leaders of cofradías .

Even the mulato elite, which had the resources to manumit a slave so that he or she could escape from a bad owner or to finance the festivities that the cofradía life-style implied, also tried—albeit timidly—to secure positions of control. The hierarchies within the cofradía were expressed visually in the physical placement of the brothers and sisters in reunions (Fuentes 1866, 83). Members avidly maintained these internal hierarchies, even at the cost of having to resort to white judicial intermediation. A sequence of events that began in 1812 with the death of the queen of the Congos-Mondongos cofradía illustrates the shift from criteria based primarily on color or tribal affiliation to ones based on legal standing.

In accordance with the traditions and customs of the Congos-Mondongos cofradía , the successor to the queen should have already occupied the post of the queen's captain-assistant. Being the captainassistant implied considerable expenditures, as it was this person's charge to finance the cofradía 's festivities and events. In fact, these payments had more weight than putative or bona fide descent from an African king or prince (which Stevenson suggests), and the privilege of becoming queen or king could be bought. However, in 1812 the Congos-Mondongos cofradía (consisting of twenty-nine persons), in opposition to this unwritten law, divided over the question of the throne's successor. The majority opted not to elect the present captain-assistant but another queen who had one decisive attribute: she was free. In response to this decision on the part of the cofradía 's members, María Santos Puente, the dethroned, presented an appeal before the civil court in order to protest what she perceived to be a divestiture of her authority:

On the grounds of being a member of the Congos-Mondongos cofradía , I declare that I have been the queen's captain and assistant for twenty-five years with the hope of being successor to her post, because the deceased queen promised me as much, in front of witnesses ... and additionally I paid the funeral costs, without any other motive than anticipation of the promised recompense. In the meantime I supplicate that in the name of justice the post be given to me, or that the money I spent be returned to me.

A few days later, María requested that several of her witnesses be interrogated in accordance with a questionnaire she had presented. The witnesses, seven in all, responded affirmatively to the following question: "State if it is true that examples exist of others of the nation who being slaves just as I, have been elected queen; that due to an unjust mishap, the queenship was rashly taken from me and given to Manuela who is free."

María Santos was aware of the reason for her removal and was relying upon the cofradía 's traditions in order to reclaim the post of queen. At this point there was a long abeyance in the trial. The public prosecutor stated:

This silence, for the interested party as much as for the other individuals in the cofradía and group, who are inventing the farce of the election of a queen (which might perhaps be of importance to them) causes the responding minister to think that this is not a case of law and private interests but pertains

rather to the public order and the preservation of the customs and practices that these blacks are permitted, and in whose possession they should remain provided the public peace not be offended.^[6]

Since the minister no longer knew if what he had in his hands was a farce or a case concerning the public order, he appealed to the cofradía 's own traditions that had until now guaranteed the public peace. Nonetheless, the votes of the cofradía 's members went against tradition. The result of the elections, after this run-around in the official courts, was twenty-one votes for and seven against the free queen. No one returned the invested money to María, the deposed captainassistant. The dream of becoming queen had left her penniless. Instead of purchasing freedom with her money, she had financed the cofradía and kept its hierarchy and structure intact. This defeat was not only a legal loss. Settling this controversy of succession cost the cofradía approximately half of the contributions that María claimed she had made over the past twenty-five years. According to her, her expenditures had reached ninety-two pesos—at a rate of four pesos a year—an amount that the elected queen was supposed to return to her, in addition to sixteen pesos "for taking to the streets the standard of the nation, with nine more pesos for the price and release of a barrow."

The breach of long-standing traditions at once reveals more intense yearnings and greater chances for freedom that existed. Substantial support for these desires was slaves' ability to earn daily wages and to save a portion of them. By 1859 the Congos-Mondongos cofradía had been abandoned, and the Merced convent, which had declared itself proprietor, was embroiled in a judicial dispute over the site. Moreover, there was no queen.^[7] In 1866 sixteen cofradías , each affiliated with an African tribe, were still recorded (Fuentes 1866, 83), but over the course of time the cofradías —as the institution of slavery declined—lost their social function for slaves and their disciplinary usefulness for owners. Long before slavery was abolished, the black population's core institution had shifted its priorities.

Seen in this manner, the changing visions within the cofradías reflected one facet of the black population's perception of itself, a facet that escaped Stevenson: he depicted poor creatures with pretensions of royalty. In a context in which blacks could obtain freedom relatively easily, there was no reason to choose a slave queen. Consistent with

broader political and social changes, Lima's blacks were willing to throw aside long-held customs. Perhaps the increasing presence of free blacks also made cofradía members feel more represented by a free queen. Times had changed and everyone had observed the events in the Congo-Mondongos cofradía . More concretely, the value that participants in the election put on the status of freedom undoubtedly influenced slaves' decisions to accumulate and to try out their own variants on the association's message.

Similar antagonisms would express themselves in another type of organization: the guild, in which the black and casta populations predominated. Some guilds were formed exclusively by people of color. When the mayoral guild election of carriage-makers and handbarrowcarriers took place in Lima in 1774, the bozales made it known that they did not wish to be controlled by an "enemy," which was their label for the criollo who was its leader. In retribution, the criollos accused the bozales of "a callous spirit and lack of obedience." Additionally, the criollos portrayed the bozales as inept for labor because of their "clumsiness and lack of reason and conduct." As in other guilds where similar quarrels had occurred, the solution to this concrete problem was to name two mayors: one a bozal , the other a criollo (Romero 1980, 22). In contrast to what happened in the cofradías , which excluded entire groups of blacks from official positions, the guild kept its occupational exclusion by dividing internal governance along racial lines.

In a more general sense, the criteria used to establish social relations were ethnicity, wealth, and legal status. Beyond the events within the Congos-Mondongos cofradía and the guilds, smaller-scale incidents embodied the tensions in society. Scuffles and street brawls between blacks, confrontations between neighbors, small purchases in corner shops, or broken street lamps could all bring about strings of epithets. And the worst insult was one that invoked a racial or economic condition inferior to that purported. In one such altercation, a woman named La Camacho called an acquaintance (a man who was perhaps more than a simple acquaintance) "a big black mule, and fists were drawn." Immediately, "La Camacho, [who] was a zamba by self-description, quarreled with the black mule's lover." Matters ended when the lover bit a finger off La Camacho's hand.^[8] More than a few defamation trials were litigated and more than one angry black was presented with a hospital invoice for an amputated finger. However,

what underlay all these seemingly insignificant quarrels were the deeply entrenched notions of hierarchy that prevailed among blacks. They were more than skin-deep.

On a different plane, social distances and hierarchies were also expressed in the city's geography and its slaves' work there.

Distribution of the Slave Population

Lima was and is a city of barrios with easily detectable social and economic characteristics. The barrios, districts, and parishes of wealthy people were easily distinguishable from the less white and less wealthy areas. Slaves moved about in all these areas because many lived outside the owner's household—depending on their trade and the work arrangements between them and their masters. The San Lázaro parish, for example, was considered to be an area of black residence, and we know that the links between slaves on the haciendas of Lima's hinterland and the residents of this parish were crucial elements of the trajectory between countryside and city. This parish appears to have been a place of arrival, residence, and meeting for blacks from all walks of life. Santa Ana, by contrast, located within the walls of the city of Lima, was a parish composed of whites but had a large number of domestic slaves.

A residential census of the Santa Ana parish, dated 1808, indicates that this parish housed 11,432 inhabitants of principally Spanish descent in 1813.^[9] This census includes the number of slaves per household, which permits us to measure the distribution of slaves per residence and per owner in Lima or, in other words, the relative concentration of each group. Unfortunately, the census recorded only 3,460 persons, 30.0 percent of the parish's population, and 6.4 percent of the total urban population. The slave population makes up 22 percent of the total in this sample and might indicate that each fifth resident of Santa Ana owned a slave. The 3,460 inhabitants were grouped into 898 dwellings of which 187 housed 763 slaves (Table 9). Of the various types, houses and cottages (small houses) were large enough to hold domestic slaves (74.8 percent), whereas apartments, interiores (usually smaller rooms within a house, sublet to third parties), and single rooms (11.2 percent) were smaller, crowded residential units where slaves could hardly live with a master. The presence of market shops and stores indicates the existence of some type of mer-

cantile activity involving slaves, (13.3 percent); orchard houses were the parish's more rural element (0.7 percent).

Ascertaining the number of slaves per owner helps us clarify the dispersion and concentration of the black population in Santa Ana. Among the parish's households, 108 (57.8 percent) had one to two slaves, 44 (28.9 percent) three to six slaves, 10 (9.1 percent) had from seven to 12 slaves, 7 (3.7 percent) from 13 to 37, and one, Lima's mint, had 60 slaves (0.5 Percent).^[10] Thus it works out that 57.8 percent of the residential units counted in the 1808 census contained 19 percent of the slaves; 28.7 percent (small to medium units) had 31.3 percent; 9.1 percent (medium to large units) had 20.2 percent; and 3.7 percent (large units) contained 21.6 percent of the slaves (Figure 2). A high percentage of households in Santa Ana reported one or two slaves. This percentage increases if we add the medium-sized ones. Our bet is that the fewer slaves per owner, the greater the tendency to assign slaves to daily-wage labor, and the greater the dependence on slaves' earnings. Lima's nobility could most likely be found in the middle and tipper categories of slave owners; although they too occasionally hired their slaves out in exchange for a daily sum, they usually considered their slaves part of their retinue and social status. Their slaves, therefore, would be placed in domestic service. If we extrapolate on Santa Ana's figures for the city as a whole, we would estimate that 20 Percent of Lima's households had slaves, and that approximately x o percent of Lima's households lived off the income earned by slaves.

[Full Size]

Figure 2.
Distribution of slaves per household in Santa Ana: 1808.

Thus urban society included a high number of individuals who depended on slaves' wages for survival itself—even to the point of selling their only slave to avoid the anticipated disgrace of a pauper's burial.^[11]

From the viewpoint of slaveholders—discreetly poor or prosperous—the hiring-out system was lucrative, though contemporaries condemned its "idleness." An owner's profits depended on the number of slaves, their age and health, the arrangement the owner made with those who were interested in hiring his or her slave, and profits hinged on the slave's "willingness" and aptitude for the given work. Despite the belief that a slave was a good investment (sometimes the only one possible), an owner knew its risks. Calculations for regions other than Lima have shown that the hiring-out system began to be profitable only when an owner possessed at least twenty slaves (da Silva 1988, 109). For Lima, the Defensor de Menores articulated the reasoning as he evaluated a father's hopes to invest money he had received to buy a slave whose daily wages would assure the rearing of the prospective owner's son in the capital. The Defensor de Menores evinced skepticism for several reasons:

The first is that when such daily wages reach no more than four or six pesos a month, it is not possible to meet the costs of food, shoes, clothing, and schooling [for the child]: The second that the daily wages of slaves tend to have the effect of causing debt, and much more so if women are involved. Given the uncertain outcome one cannot even consider that scant income: The third is that if it is necessary to pay for the treatment of the sick slave or if death accrues, everything will go to pot.^[12]

With great assurance, the judge was dismissing the father's proposal. But can we take the judge's evaluations at face value? Mendiburu records a daily-wage arrangement that would yield annual interest—which fluctuated between 5 and 20 percent (1987, 39)—on an owner's investment equal to a slave's purchase price and comparable to earnings from other investments (i.e., in land, mercantile activity, factories). Several documents confirm that at the close of the eighteenth century the minimum daily wages given to owners by apprenticed slaves amounted to three reales. It has also been noted that the slave day laborers tended to work during the day—at jobs assigned by their owners or perhaps searching for employment—and returned at night to the homes of their masters, "to be of service to them in whatever capacity they might offer." The capital slaves acquired during holidays, which exceeded that earned during weekdays, was for the slaves' use and for the purchase of clothing. This pecuniary source allowed slaves some freedom; however, it also represented a way in which owners transferred a portion of their obligations to the slave. During the workday the slave ate in the home of his or her current employer. This meant that, especially when the day-labor contract was temporary, that whoever was exploiting the slave's labor power would have little interest in giving the slave an adequate diet; contracts of the time stated that "all of this is observed as firmly established practice and custom in all these kingdoms."^[13]

In 1828 a daily-wage guarantee recorded before the notary Manuel Suárez noted that "According to established practice slaves are commonly obligated to pay their owners for the right of the daily wage only 1 real for each 100 [of their value]."^[14] Thus, if a slave's conque listed his or her value as 300 pesos, then his or her daily wage would amount to 3 reales (see also Fuentes 1867, 191). And in this case 5 to 20 percent of a slave's annual earnings (as calculated by Mendiburu) represented 30–110 pesos each year—based on a daily wage of 3 reales and annual work of 300 days—which in turn equaled a slave's high aver-

age price of between 550 and 600 pesos; the average price for slaves in Lima was 280 pesos (Aguirre 1993, l02). The calculation suggests that in real terms slaves were paying more than 1 real for each 100 of their value. The earnings noted by María Baraona (from 4 to 6 pesos a month) would yield an annual income between 48 and 72 pesos.^[15] Averaging both, we obtain an annual income of 65 pesos (70 pesos as calculated by Mendiburu and 60 as declared by Baraona). The proximity of the two values indicates their reliability and suggests that an owner could recuperate his or her initial investment—barring disturbances along the way—in seven or eight years. With the passing of the centuries, the recuperation period of this investment seems to have lengthened from the one Bowser (1977, 189) notes for the seventeenth century, when the owner of a specialized slave could—with a little luck—recover the value of his or her slave in two or three years. We might interpret the extension of the slaves' mortgage period as an increase in slaves' potential for accumulation, assuming that slaves kept a greater part of the earnings and the price of slaves had dropped. This outcome might, in turn, result from a real decrease in the profitability of the slave system except that other circumstances, such as the general state of the economy, explain similar outcomes.

We will never know precisely how many slaves earned daily wages for their masters, since we cannot measure the variations in daily wages over the course of the decades under consideration. Wages were determined by the arrangements between owners, or in their absence, between slaves and owners, and also by the slave's qualifications. In addition, slaves were not subject—despite contrary assertions—to regulations, but rather to the practices established in each particular case.

In spite of these limitations, we must hazard an approximation in order to understand the significance of daily wages in the lives of urban slaves and to illustrate—through the preponderance of wage labor—the exhaustion of the slave system. Three additional forms of evidence suggested by our data—of varying validity and weight—further corroborate the importance of daily wages to slaves. The first is the existence of San Lázaro as a parish in which freed and enslaved blacks lived together. They lived there because they had somehow found ways to leave the homes of their masters at night. As we will see, this option depended on the slaves' use of daily wages and of arguments about their rights to matrimonial and family life (such argu-

ments before ecclesiastical courts gave Manolo Lasmanuelos the chance to marry Manola, a free zamba ). The second hypothesis involves an analysis of the mechanisms through which slaves obtained freedom. Daily wages, along with robbery and eventually the lottery, were the only mechanism of accumulation to which slaves had access, especially in cases in which a slave was free of the owner's domestic control and the provided daily wages were both fixed and relatively low. The figures of the notarial record books entitled cartas de libertad (see Tables 1 and 8) help elucidate slaves' ability to accumulate and eventually manumit themselves.

If we base our calculations of the slave population's decrease on the censuses (1820, 1836, 1845, and 1850), we discover that from census to census the slave population fell at a rate that fluctuated between - 2.4 and - 2.8 percent annually (in absolute terms this rate is equal to an annual decrease of 143 slaves between 1836 and 1845, and 175 slaves between 1820 and 1836). If we project our statistical information for 1830 and 1840 onto these figures, we see that the manumissions recorded in the notarial record books represent 97 percent of the total slaves who disappeared from the statistical sampling between the first two censuses (1836–1845), and 74 percent between the second and third (1820–1836). This exercise reveals the reliability of the censuses and of the notarial record books and also demonstrates that slaves increasingly turned to notaries to record their freedom. Throughout all the years examined, the total percentages of manumissions through self-purchase or purchase by a relative (63 percent, 66 percent, and 68 percent respectively) were significantly greater than manumissions through the intervention of a third party or through a master's grant of freedom (37 percent, 45 percent, 41 percent), which—as we have seen—was not always easy to obtain (an assertion documented and valid for both rural and urban areas).

Beyond statistics, an examination of many slaves' daily life yields a third hypothesis about the prevalence and importance of daily wages amid the vicissitudes of urban existence.

Daily Wages: Part of Everyday Life

In 1819 a deceased master's testamentary wishes handed over a slave to his new owner, a priest.^[16] From the convent he wrote to the archbishop, stating that despite the fact of his marriage he was forced to

perform arduous labor and furthermore had been assigned on repeated occasions to work on the Monterrico hacienda close to Lima, and thus to abandon his wife. The response to the slave's first petition was that he should not be separated from his wife and children. The priest's defense revealed the range of negotiations in the case of—we should remember—a married slave: "He was placed in the kitchen to prepare the daily meals, however the difficulties that he caused owing to the vice of drunkenness were so frequent that the situation became intolerable."

It was the slave who immediately proposed to apply himself to "work, and contribute a daily wage so that the convent can replace the post with another hand." And, the priest replied, "I complied in order not to contravene the slave's wishes, and after living with free rein, because this class of people does not respect the sacred duties to which marriage is bound, his wife barely furnished me with a few reales, and he now owes me a considerable sum of money."^[17] To understand how negotiations with masters were carried out, we must consider the implications of this brief dialogue.

What the slave argued was that since he was married, a transfer to another hacienda would interrupt his marital relationship. But the interference of alcohol sabotaged his kitchen labor; the priest's response illustrates that he did not wish to override the slave's wish. In short, a convenient solution for both sides was to use the slave's daily wages to pay for a replacement. In this way, the slave left the convent, sought day labor in the city, and demonstrated by the subterfuge of subverting his kitchen duties and the appeal to the Church's sacrosanct image of marriage that he could transform himself from a domestic into a day-labor slave. However, he did not keep his side of the bargain. As we see, the slave's initial negotiations were embedded in the context of the moral conditions sanctioned by the Church, a strategy that frequently led to success. Later, when a master's control had lessened, little could be done to make a slave comply with his promises. The priest argued that the wife—as he defined marital duties among slaves—was obliged to assume her husband's debts. But she did not: the male slave was deceitful, and the female slave was immoral since she did not understand her sacred duties. Yet both had succeeded on their own terms, something the priest, of course, failed to admit.

Others paid their daily wages and used their compliance to restrict an owner's arbitrary decisions. The court listened to these arguments

and was inclined to favor the slaves. Manuela, a free black, lived with her husband in Bellavista. She claimed that she had punctually complied with the payment of daily wages (ten pesos a month) to her husband's owner, for which the latter allowed the couple to live together outside his household. One fine day, the owner decided to sell the husband in Pisco. Manuela would use the two-sided argument (of marriage and compliance with the payment of daily wages) in order to prevent this occurrence. And Manuela was successful in this endeavor.^[18]

Marriage, family life, and daily wages appear together, whether to arrange a change of ownership, or to maintain a place in the day-labor market. Within the hundreds of records preserved in Lima's Archivo General de la Nación and Archivo Arzobispal are very few that do not explicitly refer to slaves' daily wages and marriage. This fact alone indicates their enormous reach and importance. Many of these cases were filed by owners demanding outstanding daily wages, and the owners' action tacitly gave slaves juridical personality: their response to owners' claims took place in the courtrooms.

Shrugging aside such cases, one owner declared, "It should be stated that there is no law that grants slaves the civil personality to make contracts or gives any consequent value to their contracts."^[19] This statement, made at the beginning of the nineteenth century, underscores the fact that legal authority did not entirely cover the operations of day-to-day life. Both sides defended their interests in the civil and ecclesiastical courts. Slaves coveted their daily wages and sought to enlarge the degrees of relative freedom; owners tried to recover capital that the new arrangement forced them to collect.

Another gauge of the prevalence and significance of day labor were the cases that illustrate the amount of accumulation that daily wages made possible. As slaves became able to generate savings more quickly, a growing number of day-labor slaves were able to pay owners their purchase price. The slaves who encountered the greatest problems in negotiating margins of freedom, and thus in obtaining the sums needed for manumission, were those who we would generally suspect had the worst chances of earning and saving money: the bozales . This group of newcomers lacked the normal contacts and connections within the ranks of the slave and black population and could not make similar appeal to nonblack padrinos . Moreover, women—except wet nurses—also represented weaker links in the new economic and social network because they had only a remote chance of learning a

trade. Although female labor was always needed, even in the midst of economic crises that contracted the male labor market, women's concentration in "female tasks" reduced their market flexibility and options, particularly because more free black women competed for the same tasks. Consequently, over the long run and the closer we come to abolition, slave women's potential for accumulation was reduced. Women and bozales , thus, represented fragile links of the slaveholding structure. And if it is true that both groups were successful at the urban process of accumulation, other groups with more experience and more contacts, and with varied and versatile occupational options, were even more effective in procuring their purchase price.

The tribulations of Luciana, born on the Bocanegra hacienda in 1753, are representative of the experience of slave women and their earning capacity. After five decades of work for two owners, Luciana had ended up in the hands of Doña Paula Almogera. For a long time she had contributed her daily wages to the maintenance of her owner, with whom she shared her small house. In 1810, at the age of fifty-seven, Luciana pleaded before the judicial courts that she be exonerated from the obligation of supplying her daily wages. In her appeal she enumerated her activities and sufferings:

During the years when I was at my owner's side, serving her in her own house, I gave birth to sixteen children of whom four remain, and for three of whom my said owner, Doña Paula, has collected 400 pesos for one, 350 for another and 300 for the third, renting out the fourth, who is a girl, for whom she would no doubt demand 500 pesos if she were to attempt to sell her. In addition to the hardship that the upbringing of my children has naturally caused me, I have had to undertake that of my owner's grandchildren, all of whom were put in my hands, and also the individual care for an orphan whom I nursed with my own milk, and for which my master received six pesos a month; furthermore I sell bread dough in the streets, from which I gain one real and a half per day, and finally I gave her fifty pesos last year toward my own purchase price.^[20]

In the litigation pursued by Luciana, she enumerated the multiple ways in which her presence had been exploited—what we might call an amortization of self. Luciana stated that she had worked for her owner for sixteen years, and that practically each year she had given birth to a child. Of these, four had lived to reach working age and had doubled the owner's collected income. We can break down the slave's annual contribution to the household:

As we see without counting the slave's hours of additional domestic work, each year Luciana brought in a sum that equaled her purchase price. Luciana represents the case of a slave who went out into the neighborhood every day but remained subject to her owner's close control. The breakdown of the income supplied by the slave illustrates a degree of profitability beyond that of daily wages. Luciana worked as a day laborer on two levels: she was a bread-dough vendor and a wet nurse. Both activities amounted to 45.8 percent of the total 299 pesos. Additional sums derived from the children (the daily wages supplied by one daughter and the sale of the other children) represented 33.4 percent of the total. The remaining amount was at least equal to the value of the domestic labor. Even if the daily wage was the most significant percentage of the slave's total payments, it alone cannot account for all the characteristics of the slave system's profitability. Thus, what we have examined raises two fundamental issues: first, the profit possible in an urban hiring-out system beyond strict calculations of daily-wage contributions; and second, a slave's capacity to generate mechanisms of accumulation and surplus in the context of slaveholding. In spite of all the work and the beings—literally—handed over to the owner, in 1809 Luciana could still pay her owner fifty pesos toward her purchase price, which inevitably had to come from one of the listed activities: the sale of bread dough or wet nursing outside the house. Her capacity to accumulate was reduced by her specific circumstances, her status as a slave. In other words, a slave's earning power was quite impressive, although restricted by the master's control. As long as control was effective, it guaranteed the profitability of the hiring-out system, especially linked to other forms of slave exploitation. The lessening of authority, which the slaves fought for in

myriad ways, would make the slave system useless from the viewpoint of slaveholders.

As Luciana's case reveals, slaves recognized that their owners rapidly amortized their initial investment through the great diversity of work slaves performed and through other benefits that accrued. Slave women's reproductive capacity, along with their work inside and outside an owner's household, made slave women more successful at bargaining for freedom. Sometimes, but not always, these arguments supported a ruling in the slave's favor, in the name of ill-defined but incipient notions of social justice within the slave system.

The greater the distance from a master's control, the greater the likelihood of channeling substantial sums of money into a slave's own pocket. A case of marital conflict between two slaves allows us to underscore this assertion and further illustrates the differences between men and women, the dynamics of strife over daily wages, and—as we are examining a bozal couple—the problem of the rate of accumulation even for those recently immersed in the turbulence of urban life.

In a petition dated 1806 Catalina, married to Miguel, related how—thanks to daily wages—she had managed to obtain the freedom of both:

Last year, in 1791, being subject to the slave servitude of Doña Sipriana Palacios, I married Miguel Geronimo de Teruz, a black bozal , born and raised in the Portuguese Indies and then after we married and when he began to work with me, we reaped the fruits of our labor and he was freed first, and afterward we proceeded to liberate me. However, Your Most Honorable Sir should understand the conditions in which I did this: he being a slave and house servant, and I being a day laborer, who would be the one who worked to liberate both?^[21]

Catalina and her contemporaries took for granted that they could purchase freedom with sufficient daily wages but were unlikely to reach their goal as long as they remained in domestic service. With her labor, Catalina managed over the course of approximately five years to free her husband and herself. Along the way, she had to hand over to the owner part of her daily wages; once one spouse was free and resided outside the master's household, the couple would assume maintenance costs for the freed slave, a minimum monthly expenditure of twelve pesos that decreased the couple's real income.^[22] Even so, the total of slaves' daily wages given to the master, the costs of

maintenance, and the additional accumulated amounts were by far more important than what the master contributed to the slaves' upkeep (food, shelter, and clothing). Thus bozales quickly learned to loosen their ties to the master's house and go out into the street to earn wages, and even for them the period of accumulation could be relatively short.

If we compare Luciana's capacity for accumulation with that of Catalina, we see what statistical information on methods of manumussion (Table 1) meant in real life. Luciana was able to save only fifty pesos over the course of thirty-nine years, or one or two pesos per year, whereas Catalina (with her husband's help) managed to save approximately six hundred pesos in fifteen years, or about forty pesos a year.^[23] Catalina's rate of accumulation is very close to the seven or eight years previously recorded as the time in which masters could amortize a slave purchase. Thus in the best of cases, slaves needed twice the time to do what slaveholders did; in the worst of cases they died before they could finish payment of their purchase price.

As a consequence, the scale was tilted in favor of accumulation and freedom through the daily wage; it gave slaves increasing leverage against the slave system. In this sense the Defensor de Menores's skepticism was justified, though perhaps not for the reasons he gave to a father trying to recover the costs of his son's education. The purchase of freedom through a daily-wage arrangement could take an entire lifetime because daily wages came first and savings toward freedom followed. Yet this mechanism alone permitted accumulation, unless slaves could devise other strategies to lower their purchase price or obtain money outside the hiring-out system. The extreme alternative was to hide away and not pay the daily wages. But such an option could carry with it judicial persecution and possibly the enforcement of retroactive payment before delinquent slaves could purchase freedom.

If slaves had any incentive to reproduce, it was the fact that they could use children, as they could use marriage, to lower their own price—as Luciana did—or advance moral claims that would oblige masters to free their illegitimate offspring.

Children in Slaves' Lives

The situation of slave children and, more specifically, the political, legal, and philosophical discussions it generated serve as additional in-

dicators of the importance of daily wages and of the possibilities for negotiation and accumulation that daily wages offered Lima's slaves. Slave children became a subject of contention among masters and slaves, specifically over the respective obligations of maintenance and the rights and privileges of the slave family. Children, like marriage itself, were part of a moral package dear to the Church.

In the residential census of the Santa Ana parish (mentioned earlier), of a total of 187 households containing slaves, only nine (4.8 percent) recorded the presence of slave children, an indicator at least as low as the one recorded for Lima's haciendas. The number of children ranged from one to twelve. Along with high mortality—we should not forget that the slave Luciana had given birth to sixteen children of whom only four reached adulthood (a fairly typical survival rate)—in the urban context, the type of residence outside the master's household was the key explanation for the low presence of children. Slaveholders were convinced that their slaves did not want to bear or raise slave children and that pregnant slaves would abort if they had the choice or the opportunity. On more than one occasion this resistance to procreate prompted contemporaries to describe slaves as simply clumsy, uneducated, and primitive. But matters were more intricate than masters perceived.

María del Carmen Breña, slave of a Lima merchant, purchased her freedom soon after giving birth. She left her owner's house and her child behind. In 1815 the owner demanded that she return to his house to nurse the child; however, María wished to move the child to her own house. The owner stated his conviction:

The [proposed] relocation exposes the infant to whatever might anger her mother, otherwise the slave will let the infant die through hatred of her masters, and to liberate the minor from slavery, with neither religion nor humanity to stop her, because owing to the uncouthness and ignorance of such people no rational consideration comes to their mind.... Being pregnant with another baby ... in another master's house ... she fiercely threw herself against a wall with the aim of aborting herself, and so that the fetus would not reach slavery alive. The black did this in the house of some masters who looked upon her with affection; what will she do when she comes up against the intentions of those whom she detests, and against whom she has litigated?^[24]

What the owner argued here was that hatred of masters and desire for freedom were stronger than maternal instincts. To this perception

we can add the fact that slave women who ceased being slaves often had a dangerous road ahead, particularly if they were to survive as single mothers. An infant, whom the master would safeguard in order not to lose a slave, at the very least had food and shelter. Moreover, it was no coincidence that freedom was granted to the woman after her child was born. A child born after the purchase of freedom would be free as well. María del Carmen's owner stated that in order to subsist, the slave mother "will have to spend all day in the street selling, leaving her daughter abandoned and exposed to all the accidents of her age. If she carries her on her back, the sun will be enough to kill her; and above all she is a bozal woman, so ignorant of religion, who was never convinced to go to church ... to learn the Christian doctrine. Causing death to her child would be such an ordinary thing, like going to sleep, or something similar."

While the mother sought a livelihood and a new way of life, the child would have the owner's help and maintenance 'allowance precisely because the child was a slave. Occasionally—as in this case—such assistance would lead the owner's wife to nurse the slave's child. María del Carmen had left her child behind as she was unable to carry it with her as a free child; at the same time she was saving her child by making very literal use of the master's shelter and concern.

Apparently a very widespread practice, which helps explain the low presence of children, was the concealment of children. More than one female slave managed to run away from her master's house, give birth, and return. When asked what had happened to the baby, she would claim that it had died or been stillborn.^[25] This alternative was possible when external links existed: loved ones or friends who would take responsibility for the newborn. Furthermore, if the slave mother lived with her masters, she would have to figure out how to leave in order to nurse her child or eventually find somebody to take her place. Claiming that children had died prevented an owner not only from designating them as his or her property but also from transferring children to another owner and separating them from the mother—a separation much more difficult to control because the slave mother would have a harder time tracing where her children were. The possible sale of children always loomed despite the Church's disapproval of the practice. Against the allegations of "clumsiness" that slaveholders imputed to their slaves we note accusations of "inhumanity" that slaves raised against owners.^[26] Even if masters later realized that their female

slaves had hidden children, time had passed during which the owners had not paid for the children's upkeep and thus the owners' property demands were weakened. Placement of children in a convent was a unique method of concealment. If an owner made a demand for the child in such circumstances, there was an entire institution behind the defense of the slave. The additional argument that the convent had assumed the child's maintenance would greatly decrease the child's price and facilitate manumission.^[27] To give birth outside the master's control, to look elsewhere for day labor, or to abandon a child was to act well within the interests of slaves, given the impositions of the slave system. In such contexts as the ones described, and others—as we will soon see—it is difficult to find one sole rationale for slaves to express parental or filial love.

For slave women who did bear children in slavery, various options could at least guarantee that the slave family would not be arbitrarily scattered. Urban society offered nonmarried slave women with children alternatives that reflected its moral principles and sought to keep the slave family together without threatening the slaveholding system. A common practice among proprietary families was to make a prenatal partition of the slave children among their own sons and daughters. Contemporaries noted that "there is nothing more frequent in slaveholding families than parents distributing among their children the offspring of slaves."^[28]

This distribution was seen as a portion of the inheritance, one realized "by womb" that increased the likelihood of greater equity among an owner's children. If the assigned slave died, it would simply be a matter of misfortune. But with luck, slave children would survive and as children would be assigned to serve specific members of the owner's family. Thus, while the master's children grew up they would have companions who would play with them, then wait on them, and much later accompany them in their departure from their parent's house to begin this cycle anew. Such mechanisms internalized slavery from infancy and partly explain the loyalty of slaves toward their masters.

For slaves who did not abandon their children in order to scratch out a bare living in freedom or give up babies assigned before birth to serve members of the owner's family, a vast array of negotiations existed to keep slaves' families intact and obtain their children's freedom. Slaves who lived outside the owner's household faced urgent questions of legal responsibility for the slave children: who was obli-

gated to feed slave children—the couple, or the free spouse, or perhaps the owner? On the one hand, when slaves chose to live outside the household, owners would allege that they had to do without the services slave children usually provided—playing with their own children, doing minor chores of fetching and carrying—and that when slaves decided to live outside, they took on the parental duties and were obligated to nourish the children. On the other hand, slaves would say that the children were property that belonged to the owner who was therefore required to maintain the children. Maintenance costs, and the way in which they were provided, were at the heart of this dialogue. Both lines of argument furnished extensive matter for debate between two of the most distinguished lawyers of Lima's Audiencia Real at the start of the nineteenth century: Isidro Vilca who defended a female slave, and Pablo Ramírez de Arellano who represented this slave's female owner.

Between 1810 and 1814, the slave María Andrea had fought hard but failed to free two of her children from slavery. Five years later, Isidro Vilca would adduce in her name that the earlier negative judgment "also gives me the right to demand food for someone else's slaves [this was how the mother now referred to her two children] owing to the fact that I was not obligated to feed them. This petition is the most legal, just, and sacred of so many that may be filed in the courts of justice." The petition did not stop there. The owner's debt was specified: "The fairest sum that can be ruled for the sustenance of an individual is that of two reales per day , which makes eight pesos per month" (emphasis added). This sum is very close to that enunciated by hacienda owners intent on establishing their expenses, but it occurs here in a claim by slaves demanding retroactive payment.

María Andrea's oldest daughter was fourteen years old; her son eleven. Vilca's appeal continued:

In every court it is unquestionable and generally agreed that the owner should feed them.... No one is unaware that two reales per day is not sufficient even for the bare essentials, and even more so if one contemplates that we live in a country where in the last twenty years the cash supply has decreased and the price of the essential items has increased.

The total sum resulting from the petition amounted to 3,456 pesos. Announcement of the sum set off a cavalcade of arguments between the two lawyers. Ramirez de Arellano asked:

Who is the person filing a suit, and who is the one who has nourished these slaves? The same María Andrea who was owned by Don Eugenio, the natural mother of those children and who has benefited by her master's testament, obtained her freedom as well as that of her mother, María Jesús, and that of another daughter named Asención, a godchild to my party. With whose orders or capacity did she decide to assume charge of rearing her children? She did it all by herself because she wanted to do so as a mother, culpably and criminally swindling her master of his slaves.... After the violent extraction [of the children] from the owner's house, in which she would have never lacked food, with nothing to feed those two boys now, today she dares to demand food without honesty, owing to her rancor at her sons' declared slavery. This case is nothing more than a scandal.... She has done this because it pleased her, because she wanted to cheat her master, hide his slaves from him, feed them as if she had fed them as a mother. A popular saying tells us—and such attitudes are quite common among such people—that whoever gives bread to another's dog loses the bread and loses the dog.... [Furthermore,] it is indubitable that both boys at their respective ages, would have been capable of menial labor: in the house, accompanying their young masters to play and to school, taking care of the little tasks in their master's retail store for their own subsistence and for that of their master's children and family or of other jobs and tasks of this kind. From all these services my party has not benefited the least, but María Andrea has had them all to herself while the children helped her in the profession of preparing food.... Ultimately it must be said that obligations between masters and slaves are reciprocal.

In 1819 the owner agreed to sell the two boys and signed their cartas de libertad .^[29]

In accordance with the legislation in force, an owner was required to feed his or her slaves. A slave remained a slave whether or not he or she lived in the master's household. If a slave couple had been given permission to leave along with their children, the slave family regained not only family unity but also—as in the cited case—the benefit of the children's labor; in effect the family deprived the owner of their labor. The defender of the owner's interests based his argument on this deprivation: the slave stole the children from the owner's house. Thus, the owner was freed of the responsibility of feeding them, and the mother, María, had to assume it. The fact that the lawyer revealed that María Andrea's demand was commonplace among "such people" corroborates not only our assertion that slaves could live with their children outside the owner's household, but also that conflicts over the

obligations of parents and of owners were common and that these contradictions were frequently utilized by the slave population.

The amount María Andrea claimed was not small. Nonetheless, it tallies with what was minimally needed at that time for daily maintenance. Furthermore, it represents a decision by the slave family to feed itself and bear the costs of distance from the owners' house through daily wages. Tacitly, owners allowed familial slave life outside their dominion with the intent, once the children reached working age, of resuming their proprietary rights. When an owner did so, the loss for slave parents was double: they lost the granted maintenance allowance and were menaced by the owner's potential sale of the children. Lawyers for one such owner argued, "As the control of slave children belongs to owners, it is further just that during infancy these latter also furnish food and appropriate clothing until they [the children] are able to be passed into the hands of other owners."^[30]

Law and reality clashed. While legislation stated that maintenance was an owner's obligation, many slave families chose to live outside the owner's household. Such a decision on the part of slaves presupposed the existence of mechanisms decisive to the attainment of freedom that could stop owners from selling their children. An extraordinary mechanism in the context of slavery—which relied on the apt intervention of the lawyers of the Audiencia Real—was a concrete calculation of the amount that the slave parents had invested in rearing their children, an investment that owners were responsible for repaying to the slave parents. The calculation simply followed the law. It safeguarded family unity even against owners' well-entrenched notions of private property. And in the case of María Andrea, it resulted in her son's accession of freedom.

Even so, however, the space claimed by the slave population remained—as inscribed by the definition of slavery itself—a transitional context. No slave's success was ever completely assured, and it is likely—as has been asserted for other areas as well—that this insecurity itself imparted particular traits to filial relations.

The supreme decree of 24 November 1821 reasserted that owners were responsible for feeding slave children, a ruling that gave slave petitions an even firmer basis. If the slave could prove that he or she had assumed these costs, it would be difficult for owners to recuperate their rights of property over the children. Thus the decree set up conditions in which slave children could live with their parents outside the

owner's household.^[31] The Defensor de Menores was the official commended to enforce this decree, and it is through his agency that slaves and slave children expanded their civil and legal representation.

For widowed and single slaves, and for those in situations such as the ones they have described for us, struggling to pay daily wages and at the same time to maintain one or several children, existence was a doubly demanding and toilsome task. The weight of this two-sided burden was also probably a key reason for marriages between slaves and arrangements in which, from the onset, one freed spouse lived outside the owner's household (see chapter 5). We observe the relative weakness of a single woman in the case of María del Milagro Solórzano, who was represented by the Defensor de Menores. María filed a suit against Doña Juana Murga over the freedom of her daughter, Micaela Bartola. The owner had placed María in the San Bartolomé hospital, where she remained for many years until she was well. In the hospital she was required to work; in other words, she paid for her own treatment with her daily wages. While in the hospital María gave birth to a daughter. When the daughter was only a few years old, the owner snatched Micaela from her mother's side and sold her, even though the owner had never made the slightest gesture to support the mother or the child. María set before the court the arguments of her use of daily wages to survive and her owner's failure to provide food or pay for her treatment, and attempted to take her daughter out of slavery and patronage. The judge denied the child freedom, despite the fact that María could even demonstrate that her daughter had been registered as free at her baptism.^[32] María—so ran the counterargument—had not delivered daily wages to her master, which would have allowed the owner to provide food and medical expenses. Thus, the litigation turned the reasoning from wages against the slave. María was not married when she gave birth to Micaela, an additional fact that the judge might have viewed with disdain. Furthermore, the father of María's daughter did not admit his paternity or was at least unwilling or unable to help María out. Still, considering the decisions made in most cases, this judgment appears exceedingly arbitrary, and the Defensor de Menores must have appealed it.

What is certain is that slave mothers who had little contact with other persons, or who were not married, even if they might have managed to reside outside their owner's household, encountered the

most difficulties in liberating their children. There were many women in similar situations who, with their "incessant work, industry, and economy," managed to amass enough money to manumit themselves and eventually their children.^[33] In every case the slaves needed external support to strengthen and legitimize their demands. Sometimes the cogency of demands even led to physical confrontations between owners and slaves. At times the Defensor de Menores appeared; at others it was the slave couple who acted to free the children, especially when one of the spouses was free. Godmothers and godfathers (either black or white) also were decisive agents who intervened on the road to freedom.

In 1839 Marcos Esquivel, a disabled first sergeant of the Colombia battalion, filed a suit against Doña Isabel Espinoza, the owner of his mother and of his wife, Justa Torres. Justa and Marcos had given birth to two daughters, one eighteen years old, and the other nine. Marcos claimed his military salary had enabled him during all the past years to contribute a sum of twelve pesos each month to Justa, who lived with their daughters in the owner's house: "whom I have fed and educated since birth, paying for baptismal fees, clothes, and whatever they needed for subsistence."^[34] Marcos alleged that because the owner had failed to provide for the children she had lost her patronage of them; he based his claim on the November 1821 decree. In December 1839 Marcos was imprisoned. The owner had appeared at his house in order to take away the youngest daughter.

And because I insisted that my daughter did not have to leave so soon, she raised the parasol that she was carrying in her hands, she hit me repeatedly, I prudently told her not to abuse me. Her furor increased and she continued to beat me with no respite; upon which I took out some small scissors that I was sharpening and gave her a small cut: the grocer Don José del Carmen Seco, the immediate neighbor of my workshop, was witness to all this.

The owner told the municipal governor what had happened, and Marcos, a free black, was sent to prison. The April 1841 judgment ruled that since Marcos had not convincingly proved that he had fed his children, the owner could recuperate the right of patronage. This episode shows the violence that could ensue over negotiations involving children: from scissor cuts to umbrella beatings and owners' visits to slaves' homes. All participants dreaded such actions and escalations but they did occur, despite attempts to mark spatial segregation. Vio-

lent encounters occurred and expressed changing social claims from Lima's black population.

Throughout the range of possible arrangements for slaves—from continuity within a fixed time and space to various demands for upkeep by slaves and ex-slaves who had gained a measure of independence—the elements of the situations we have depicted were the daily wage and the gradual distancing from an owner's control. The situations varied in rates of accumulation and in living arrangements, particularly those involving children. In a situation of womb assignment, children and the daily wages they produced were the property of the owner, even if a margin of accumulation (in Luciana's case, fifty pesos in fifty-seven years) existed. But a slave who had managed to purchase freedom but still had enslaved children had two options. Slaves might leave the children in the owner's house, thereby transferring the maintenance costs to the owner and risking the possibility of the later sale of the children. Or, gathering together 'all their funds, they might ask for the children back and consolidate the family unit. Conflict arose when both possibilities intersected: when an owner imposed his or her property rights and a slave demanded the ordained maintenance allowance. A successful outcome depended on slaves' skill in negotiation and accumulation.

As the price of slaves dropped and that of clothing and food rose, to leave the children under the temporary care of the owner was a way to accelerate accumulation. Rather oddly, the slaveholding system worked to the slaves' advantage in this situation because the value of slave children was, according to age, less than that of adult slaves. The calculations made by Isidoro Vilca, the lawyer of the Audiencia Real, represented the minimum amount necessary for survival—ninety-six pesos a year for the upkeep of a child. Thus, the cost of a child's maintenance for four years might equal the purchase price of an adult slave. Within the slave family's budget maintenance of sons and daughters was a considerable expenditure for slaves whose first necessity was to pay the daily wages they owed to owners. Because many slaves opted for family life, we conclude that they were willing to pay these expenses. Their decision to maintain a family outside the master's household delayed the attainment of their own freedom in order to negotiate the freedom of their children through the argument of maintenance. In calculating the profitability of the hiring-out system, we must add sums from daily wages to the maintenance costs of chil-

dren. For owners, the strategy of leaving slave children with their parents meant reducing costs; yet for slaves, leaving their children with owners was a choice with the same objective. It was more profitable to wait until children grew up (and in the meantime avoid relocation outside Lima), and pay their purchase price when they reached working age. María Andrea's eleven- and fourteen-year-old sons assisted her with the chores of food preparation and strengthened the likelihood of their continued freedom. Freedom was a double-edged sword; for slaves, the final outcome depended on their place in the labor market. And success was skewed. After all, every nuance of any potential negotiation—including the payment of daily wages—was an expression of a slave system.

The circumstances of the slave Antonio described by his wife, Juana Pedreros, a free black originally from Angola, remind us of the brutal dimensions of slavery. When Antonio and Juana married, the master allowed the couple to live outside his household in exchange for the payment of six reales a day (twice what was currently stipulated). The couple had religiously complied with the payments for many years. One morning in 1812, however, Antonio hanged himself from a tree in the Alameda del Pino. Eloquent in its description of the reasons for this act was the declaration of Antonio's wife:

For some years she has been married to the black, Antonio, the slave of Don Ignacio Meléndez. Together they had five children. Her husband earned money as a water carrier, for which reason his master obligated him to pay the daily wage of six reales even if he did not earn them, and because of this contribution the declarant and her children had to pass many days begging in order to sustain themselves. Yesterday morning the black, Antonio, awoke at five to go to work and left, as he usually did, and a little while later she was notified that he was found hanged in one of the trees of the Alameda ... that the reason that the black, Antonio, had hanged himself is the very obligation to his master who required from him the payment of six reales a day.... When he could not come up with the money he had to resort to borrowing. He mentioned this situation to his wife on several occasions. Added to all this distress was the illness of one of their children, and often they simply had no medicine to treat her, for which reason her illness worsened. Two months' rent was also owed to the owner for the room where the declarant is staying.^[35]

Although the image of success overshadows the hardships on the journey toward freedom, some individuals could not face the dual

exigencies of daily wages for an owner and bare necessities for a family. This compounded burden crushed their plans and desires for freedom and demonstrates both the fragility of the liberation process and the merits of those who crossed the threshold. For this same reason, it suggests that we must avoid excessive optimism about the possibilities of self-manumission.

The situations we have viewed lead us to three general points. First, the daily wage gave impetus to the development of an artisanal market for black urban labor. Second, owners benefited from their slaves' daily wages as the slaves gradually accumulated their purchase price. Third, the slaves' capacity to earn and save depended on their owners' circumstances and their own. The more slaves owners had, the less their dependence on slaves' earnings; the fewer slaves they had, the greater flexibility for slaves to negotiate spaces of relative freedom—circumstances that fit a significant percentage of urban slaves. The slaves' varying degrees of urban experience and real or fictive kinship ties shaped their opportunities and strategies for urban life. Over the long term and despite the enormous disparity in wages, the daily wage had not only the greatest influence on the organization of the masterslave relationship; it spurred the dissolution of the slaveholding system and the differentiation among slaves and the black population in general. For the slaves—witness Antonio's case—this transition was risky and hazardous. Yet slaves were ready and willing to take this risk and exploit all the small openings Lima's slave system provided.

We can confirm many of the tendencies we have analyzed by examining Peruvian slavery from another angle—from the perspective of the marriages and family life of slaves. Considering the debates about the vitality of the slave family and its importance in the construction of black culture, we must ask how slave family life began and flourished. Did an urban slave family often start with the birth of an owner's illegitimate child to a slave woman in domestic service? What disappointments and triumphs emerged from marriages between slaves and free persons and from other relationships that involved slaves; and how did all these factors influence conceptions of family life and slaves' marriages? Here are the questions we will explore next.