3
The Growth of a Culture of Opposition 1850–1947

We can say that, after the fight of Tupac Amaru, it was the Community of Huasicancha which was the second in the Nation, in the Department of Junin, in the Province of Huancayo, to begin the revindication of our lands.
—Huasicanchino ambulant fruitseller, Lima, 1981

The land recuperation campaign of 1963–1972 is the contemporary expression of a culture of opposition that had been developing through history.^[1] On the more distant past can now be superimposed the contours of a more recent culture of opposition, for what we encounter now are variations in the ways contention is expressed as a function of the national political and economic contexts that effectively open and close the spaces for resistance. The first national political context was the Peruvian-Chilean War of 1879 to 1884. The Huasicanchinos, mobilized as guerrilla fighters, obtained a vivid political experience of conflict expressed openly and sustained over a number of years. Moreover, in turning to this more recent history we are dealing now with the not-too-distant memory of today's Huasicanchinos. Old and middle-aged people often refer to the accounts their parents gave of these historically significant events—not just for Huasicanchinos—but for the highland pastoralists of the central Andes as well. And the relevance of this past is not confined to storytelling; it is a part of past experience that provides just one of a myriad of sources for addressing contemporary and daily problems.

In coming to this century, the context changes and with it the arena of conflict and daily survival. An increasingly commercialized regional economy effectively put pressure on both the hacienda and the highland pastoralists to operate competitive enterprises, which provided the pressure for changes in the respective relationships between the two: Tucle and Huasicancha. A new set of contradictions came into play, thus providing a somewhat different space for the expression of contention. To go back for a moment to the metaphor of a river flowing over limestone to suggest different stages of resistance,

what we now witness is at first the very visible flow of resistance along the surface, as the Huasicanchino guerrillas take on an open expression of class resistance, and then the more sporadic and less easily visible subterranean flow of daily conflict.

In the first stage the centerpiece of the drama is the montonera resistance to the Chileans and to Peruvian collaborators with the Chileans in the war of 1879–1884, which subsequently developed into an independent guerrilla campaign directed generally against a much-fractured regional power bloc.^[2] The montoneras were crucial in Huasicancha's political development. But then in the early 1890s the hacendados used the dispersion of the montoneras and the consolidation of their holdings as the basis for a drive toward the rationalization of production. Yet in the case of Hacienda Tucle, this process was very much conditioned by the immediately previous experience of open mobilization on the part of the labor force; these attempts to modify existing social relations of production in the area effectively changed the dimensions and shape of contention. The relevant national context was no longer the political one of a foreign and then civil war but the economic one of sporadic but, in the long run, persistent commercialization.

The period we are dealing with here provides the conditions for the most recent recuperation campaigns. Broadly speaking these were threefold. First, the period witnessed a final shattering of what little was left of assumptions regarding the responsibilities of powerfully placed patrons, still referred to in the terms of kinship. In the first stage of open mobilization this change could be seen as guerrillas were first mobilized as the followers of their hacendada's cousin, General Caceres, then saw their leaders executed by him, and finally found themselves taking on the entire regional power bloc. In the second stage a shift occurred from personalized to commodified relations, as the hacienda administration pressed forward toward the rationalization of farm operations.

Third, where the earlier links between the hacienda and the community might be seen in terms of a benign cancer, the effect of increased hacienda use of communal land after the montonera decade, combined with the growing reliance of comuneros on sources of livelihood beyond the village, reduced the monopoly community authorities had over crucial moments in social reproduction. Such crucial practices as the designating of pasture to certain households lay now in the hands of the hacienda administrator. So while many of the instrumental uses of the community remained, some of its more immediately visible manifestations were seen to be disappearing as better-off pastoralists, finding no immediate economic advantage to be gained, lost interest in political office. In the midst of these changes there emerged new forms of interdependence among Huasicanchino enterprises at the precise moment when their heterogeneity increased as the result of their insertion into ever more widely dispersed sectors of the Peruvian economy. Thus, the hundred years I am about to cover represent two major experi-

ences in the Huasicanchinos' past: one that occurred at the level of history writ large, that of open and sustained political confrontation; the other just as important although far less visible, occurring as it does within the gray monotony of daily living: the commodification of increasing areas of their livelihoods.

From Mobilization to Rebellion, 1850–1899

At the beginning of the period, the relation of Huasicancha (and indeed the surrounding highland area generally) to the national political crisis took on a quite personal character, for there were a number of powerful families who played important roles both in Huasicancha's history on the one hand an in national developments on the other.^[3] General Caceres, heroic opponent of the Chileans, fought much of his campaign in the Mantaro area. He was a cousin of the owner of Hacienda Tucle and one of his reserve officers was the hacienda administrator. The Valladares family, owners of neighboring Hacienda Laive, however, were collaborators with the Chileans and members of the civilista party to which Caceres was opposed. But as the period developed, there was a shift from a conflict reflecting, "at bottom, the irreconcilable interests of individual chieftains and patrons, and their dependent followers" (Bonilla 1978:100), where great names within the Peruvian power-bloc mattered, to one where great names gave way to broader social forces in which guerrillas—originally recruited through personal ties of loyalty between patrón and peasant—became a force of their own pitted against both the large hacienda owners and the valley-based capitalist entrepreneurs.

The montonera campaign started as an expression of resistance to foreigners and those who collaborated with them and ended as a class struggle, a struggle that brought to the surface and set the stage for the ineluctible collisions between classes formed on the basis of the newly commercialized regional economy.

The Pre-War Situation

In the middle of the nineteenth century the Mantaro Valley was a fertile agricultural area surrounded by highland grazing lands and with poor communications over the Andes to the more commercialized coast. The conjuncture of classes, besides reflecting the changes that were occurring in the regional economy, was also spatially situated. The area was dominated by three towns on the left bank of the river: Jauja, Concepción, and Huancayo. Jauja and Concepción retained the character of their colonial past and were

still the seats of the largest hacendado families, who found markets for their produce on the coast (Manrique 1978). Huancayo, however, became increasingly the economic nexus of the valley, sharing its vigor with the prosperous smaller towns of Sicaya, Chupaca, and Chongos Bajo situated on the western side of the valley, up against the hillsides that reached up to the highland pastures in which Huasicancha was located (see map 1). Huancayo, the department capital, and these smaller towns (distrito capitals) were populated by a few smaller hacendados, independent farmers, manufacturers and comerciantes .^[4] All of these came to benefit increasingly from the growing mining economy, which, in the 1850s, was scattered through the area, but by the end of the century became concentrated to the north, in Cerro de Pasco. They were, then, distinct from the larger landowners in their reliance on the regional economy or, put another way, in their relative lack of reliance on the coastal markets.

Surrounding the valley was the high pampa best suited to llama and sheep grazing but where arable crops were grown in sheltered valleys for subsistence, and both the larger hacendados and the comerciantes of Sicaya, Chupaca, and Chongos Bajo (which, as distrito capitals and old ayllu seats, laid claim to highland pasture) had interests in these highlands. As we have seen in the previous chapter, the relationship between these groups and the highland pastoralists was one of uneasy tension; nevertheless, at midcentury there was still sufficient pasture to provide for a certain degree of coexistence.

The degree of independence of the pastoralists varied according to their links to the haciendas on the one hand and to the lower distrito capitals on the other. Around 1850 the labor requirements of the haciendas were not great enough to involve all pastoralists for very much of the time. The haciendas' labor force was made up to a small extent of permanent staff located on the estates and to a larger extent of people from the surrounding pastoralist communities. But this still left the majority of the community at any one time independent of the hacienda. For some communities however there were ties to the lowland distrito capitals. Though relatively independent economically, the highland community was, administratively, an anexo (annex) required to perform services for the lower town as a whole, as a means of acknowledging the de jure claim of the capital town to this highland pasture. Moreover, within these anexos poorer people acted as shepherds for the livestock owned by the comerciantes resident in the lowland towns and hence had direct ties of dependency to them. Hacendados and lowland comerciantes, then, both had claims on highland pasture.

Huasicancha was on the very edge of the Province of Huancayo, however, and this remoteness diminished the Huasicanchinos' subordination to lowland communities while putting Hacienda Tucle in an especially exposed position vis-à-vis highland pastoralists. In this sense then, Huasicancha was similar to other pastoral communities in the area but with a distinctiveness that de-

rived partly from sheer remoteness and partly from an historical tradition of resisting intrusions by families from the lowland towns. The combination of Hacienda Tucle's indirect link through its owner and administrator to General Caceres and his army of resistance and Huasicancha's relative independence from the claims of the increasingly commercialized valley towns and their comerciante inhabitants, gave to the Huasicanchinos a distinctive history in the formation of oppositional consciousness.

And it was the Peruvian-Chilean War from 1879 to 1884 and the years following it, which provided the quantam leap in the development of a vigorous culture of opposition born out of the role of the Huasicanchinos in guerrilla fighting. The events of the period vividly reveal how the particulars of local culture can become extended, through the outcome of historical events, to represent more universal forces; the pastoralists around Huasicancha began by fighting against the Chileans, then against specific landlord families whom they adjudged colaboracionistas , and finally against landlords as a class.

At the beginning the personalities seemed to stand paramount: Bernarda Pielago, the self-educated hacendada and businesswoman; Tomas Laimes, the ex-corporal and rebel general; and Andres Caceres, the Indios' taita (uncle) and eventual betrayer. At the end, these same personalities were lost in the greater sweep of forces represented by confrontation between commercial, predatory haciendas and highland pastoralists. As a result Huasicancha's culture of opposition, once juxtaposed against the names and personalities of powerful families, became increasingly set within the context of more impersonal forces of which personalities became merely the representatives.

In other words, at the outset what was occurring was not just a struggle between collaborators and patriots, nor between landlords and peasants, nor highland pastoralists and valley comerciantes, but struggles between particular families of landlords and comerciantes and particular groups of pastoralists, each pursuing their own goals and policies. By the end of the period, experience of events as they unfolded gave rise to the beginnings of clearly perceived conflicting forces. The effect then of this political engagement was to clarify in the consciousness of all participants a sense of the irresistable force of the commercializing haciendas and the immovable block of the resisting pastoralists—regardless of the personalities and families involved.

This observation is well illustrated by comparing two similar highland communities whose historical experience led to the emergence of differences in the nature of their political consciousness, specifically in the extent to which their local culture is indeed one of opposition to a prevailing hegemony. One community, Yanacancha (see map 1) used pasture claimed also by both a hacienda (Laive) and lowland comerciantes (from Chupaca). The owners of the hacienda, the Valladares, were a powerful civilista family who had supported the civilista claimant to the presidency, Echenique, and were later opposed to the war, collaborating with the Chileans. The other commu-

nity, Huasicancha, had successfully fought off claims to its pasture by Chupaca families. The primary link between hostility and interdependence for Huasicancha was with Hacienda Tucle. In contrast to Laive, the owner of this hacienda (Bernarda Pielago) had been a staunch supporter of President Castilla against the civilista President Echenique and then later supported General Caceres, who was Castilla's protegé and the hacendada's cousin. Caceres was the personification of opposition to Chile and actually mobilized the highlanders around Huasicancha to fight as nationalist montoneras. These interpersonal alliances and strategies prior to the war and then during it were to provide distinct experiences for Yanacancha and Huasicancha.

One of the three most powerful families in the region was the Valladares clan who lived in Jauja. Their fortune originated from mining, the profits from which were used to buy up haciendas in the area and restock them. This vast dynasty of interlocking family interests was, up to the outbreak of war, the most dynamic force in the region, owning in 1879 at least twenty haciendas covering over 300,000 hectares (Manrique 1981). By the 1870s it was headed by three men, the two brothers, Juan Enrique and Manuel Fernando, and the brother-in-law, Luis Milon Duarte, who was the political leader of the clan. All three were directly involved in the administration of their haciendas which they aggressively restocked immediately following purchase, with often devastating consequences for neighboring small herders. In 1848, for example, they had bought Hacienda Laive entirely denuded of livestock; by 1878, just before the war that was to shatter their fortunes, they had restocked Laive to the extent of 38,000 head of sheep.

In addition to the large families like Valladares, there were a group of smaller hacendados who often combined the ownership of just one or possibly two ranches with commercial activities in Huancayo. Among them was Bernarda Pielago, who owned Hacienda Tucle. Another of this group was Mariano Giraldez, who owned Hacienda Antapongo, which lay between Laive and Tucle and whose land Huasicancha also claimed as their own. Giraldez, the civilista mayor of Huancayo in 1877, was a political ally of the Valladares.

But the Valley also experienced the growth of local textile mills, a wide variety of craft workshops, and, above all, trade. The distrito capitals of Sicaya, Chupaca, and Chongos Bajo, old ayllu centers of the past, were now thriving trade centers with an important class of people who combined small-scale commercial farming with trading ventures. These towns all had claims to the pasture of their anexos in the higher pasturelands, but their interest in these so-called "dependent" communities was to control the commodity trade with them.

Once the Valladares family bought Laive and began restocking it, they were quickly brought into a potential conflict with Laive's neighboring highland community of Yanacancha, an anexo of Chupaca. Pastoralists had been

using land in the area relatively freely because, as a result of the previous owner's depleted stock, many estancias were left unused. The Valladares now pressed these pastoralists either to move back off the land or to remain with their animals, paying a rent in labor service as shepherds for the hacienda and accepting thereby the "protection" of the Valladares family (Samaniego 1974).

But this set off an old conflict between Yanacancha and its distrito capital, Chupaca, which had always claimed legal title to the land. The growing commercialization of the Mantaro Valley farms meant that these highland communities no longer bartered with them but grew equivalent root crops in sheltered valleys instead, while specializing in sheep farming for trade. The Chupacans, for their part, were prepared to leave the highlanders to their sheep farming as long as they sold their produce to Chupacan intermediaries in what was becoming a growing commodity market.

Wool prices remained strong from 1866 to 1876, which meant that the intermediaries became increasingly committed to this trade. But an economic collapse in 1876 led merchants to attempt more direct control of production and hence a greater share of the surplus on what they saw to be "their" highland pasture; it also led the pastoralists, themselves faced with falling prices, to seek out the best bargains by breaking the monopsony of the merchants in Chupaca. In August 1877 Chupacans invaded the pasture used by the Yanacancha pastoralists around Hacienda Laive and demanded both a rent and a commitment to sell livestock through them. The Valladares family, following the strategy of civilistas throughout the department (see Piel 1970:122), sought to mobilize the Yanacanchinos against Chupaca. The Chupacan comerciantes were, after all, far more dangerous and powerful claimants to the pasture that the Valladares family coveted, and, in this manner, the family was able to appear as the benevolent protector of the highland pastoralists. The Chupacans did have the support of a national political party, but it was through the influence of Mariano Giraldez, civilista ally of the Valladares who was mayor of Huancayo, that the Chupacans were removed from the pasture by a cavalry contingent (Samaniego 1974).

At this crucial moment then, just prior to the outbreak of war, when highlanders would be offered the opportunity of being armed and mobilized, the pastoralists around Hacienda Laive were being protected from the aggression of petty capitalist entrepreneurs in the valley by the local hacendado with the help of the armed might of the state, in the form of a cavalry contingent. It is important to contrast this situation with that of the Huasicanchinos, whose subsequent political mobilization was to be so different.

To begin with, Huasicanchino pastoralists had established a far stronger independence from the valley communities than had Yancancha. Indeed this independence and physical remoteness meant that Huasicancha was by no means as integrated into commodity relations as Yanacancha. One result was

that herd size continued to be constrained by the availability of "family" labor rather than by the more commercial relationships growing up in Yanacancha. Yanacancha's conflict was a direct result of these commercialized relationships: there were no pastoralists in Huasicancha who could herd over 800 sheep as was the case for a dozen or more families in Yanacancha.

By the same token, Hacienda Tucle lacked the commercial development of Hacienda Laive, and through its remoteness was more exposed to the whims of the local pastoralists. Nevertheless the relationship between Bernarda Pielago and the pastoralists of Huasicancha—one of suspicion and interdependence—was the crucial link between the community and national politics. Bernarda Pielago, with her partner Faustino Chavez, bought Hacienda Tucle from the Valladareses in 1846. This she owned with her two shops in Huancayo. As the most relevant "significant other" in the Huasicanchinos' experience at the time, it is worth dwelling for a moment on the kind of person she was.^[5] In her will she writes: "Born poor, I lived under the protection of my parents and when they died I was left an orphan with no more support than that of Providence nor more hope than that of my labours" (Archivo del Juzgado de Tierra, Huancayo ). Despite her poor beginnings, a local historian wrote of her, "Like women of the period, she lacked any formal instruction, which handicap she overcame by her great intuition; she played the piano, played chess, knew numbers and history, managed her haciendas and was rich" (Tello 1971:28). Married twice ("Of the first I have nothing to say"), she brought to her second marriage (to Fernando Fano) 30,000 pesos. He brought no capital whatever and died ten years later, leaving her childless. Indeed Bernarda had a poor opinion of marriage altogether; it gave her "nothing but pain and suffering."

Once having bought Hacienda Tucle, Bernarda gradually gathered a considerable fortune. Her partner soon died, leaving nothing but debts and, in paying these off on behalf of his heirs, she gained full control of the hacienda. The fundo itself remained a miserable building that she rarely visited, but this contrasted with her Huancayo residence, still inhabited by her heir in present-day Huancayo. This "mansion" was described as,

Very pretty and charming. Passing from the wide entrance to the house, one arrives at a beautiful garden surrounded by wrought iron railings which enclose vast corridors, from which one gains access to salons, bedrooms and other quarters. Within, the rear door is a special entrance where the llama trains are received, carrying produce from the haciendas. (Moreno de Caceres 1974:78)

The intricacies of personal ties and national politics are well illustrated if we pick up the story on 31 October 1854, when Bernarda Pielago, returning to this mansion from Hacienda Tucle on a mule, encountered the exhausted troops of Ramon Castillo and offered the ex-president hospitality in her man-

sion. Now in rebellion against President Echenique, Castilla was given Bernarda's full support. She put all her wealth at his disposal, turned the town into an arsenal, and organized the town's myriad tool shops and small manufacturers for the production of the materiel of war. Castilla's first term in office had coincided with the beginnings of economic expansion in the Mantaro Valley, and he, therefore, found support from Bernarda Pielago, whose wealth likewise had increased, as well as from many of the comerciantes of Chupaca and Sicaya.

In raising funds for Castilla, Bernarda turned to her cousin, Domingo Caceres, to raise financial support in Ayacucho. In return for these favors Castillo agreed to become the patron of the son of Bernarda's cousin, Andres Avelino, later to become General and then President Caceres. And when Castilla's rebellion led him to a second term as president, Avelino Caceres became committed to the same kind of professional militarism as his patron.

What we have then, in 1854, is the establishment of a firm alliance between the owner of Hacienda Tucle and national political figures with strong nationalistic and militaristic inclinations. Bernarda's support of Castilla eventually was to provide her with the patronage of General Caceres (himself later to become a president of the republic). And, in turn, it was because of Bernarda, the patron of Hacienda Tucle and its neighboring villages that, in the war with Chile, General Caceres was able to mobilize montoneras. But the experience of mobilization itself gave the pastoralists a new perspective on the world. It is a sign of changing political attitudes that at the outset of hostilities these montoneras referred to Bernarda's cousin, General Caceres, as taita; at the end their leader was sending him a message: "Tell Caceres I am as much a general as he is and will be dealt with as equal to equal."

The War and its Aftermath

The war with Chile (1879 to 1884) marked a turning point in the relations between hacienda and highland pastoralists: from this period on Tucle and Huasicancha were set on a path of perpetual conflict. During the war, highland pastoralists, acting initially as guerrillas in the patriotic cause against the Chileans, became aware of their own potential strength. With the war's ending, however, a political climate adverse to the pastoralists provided the hacienda with an opportunity to expand its boundaries as it had never done before. The war thus gave birth to a fatal combination—a self-confident peasantry and an expansionist landlord.

But the war had a far greater effect than this, for it acted to position the highland pastoralists—politically—in a wider set of class relations. Hacendados, valley merchants, and commercial farmers as well as highland pastoralists all saw the war affecting their interests in different ways. The

montoneras around Huasicancha, once mobilized, remained so. But the composition of their enemy shifted. At the beginning of hostilities these montoneras were fighting the foreign invaders; at the end they fought alone against a wide range of opponents—landlords, the commercial classes of the valley, and the agents of the state. Such an experience made a profound impression on their culture of opposition, coloring their attitude toward political confrontation for the century that followed. The events of the war then form a vital part of Huasicancha's contemporary politics.

Peru's war with Chile was a disaster in which successive presidents in Lima sued for peace, while a recalcitrant General Caceres waged his own campaign against the enemy in the central Andes (Bonilla 1978). This enemy included both invading Chileans and Peruvians who, in line with the presidents in Lima, were sympathetic to an early peace. Then, for two years after peace had been signed and the Chileans had left, Caceres continued to fight in the region, this time for the presidency, which he eventually gained in 1886.

To fight the Chileans with a meager force of less than 1,500 professional troops, Caceres embarked on what he called una guerra en pequeña o de guerrillas ("a small or guerrilla war"). This meant a sacrifice of control over the campaign in favor of fomenting resistance from independent, local militias of "patriots." By far the most enduring of these forces were the so-called montoneras recruited in the high and inaccessible mountains to the southwest of the valley, around the area of Huasicancha. These forces were originally recruited for a confrontation with the Chileans to be fought at Sicaya, in the valley itself.

After a small victory in the south of the valley in February 1882, Caceres retreated south to Ayacucho, leaving the enemy to occupy the valley. The latter quickly alienated the comerciantes through the administration of a head tax and food levies and then alienated the poorer peasants by their ostentatious cruelty toward "Indios." When spontaneous opposition broke out, it was among the highland pastoralists rather than the valley residents, but not near Huasicancha; rather it was on the opposite side of the valley, where Chilean troops were ambushed at Comas. This appears to have been the catalyst for confrontation. Systematic, organized resistance followed, with Caceres's assistance being focused on the towns of Sicaya, Chupaca, and Chongos Bajo, where the local comerciantes enthusiastically began the task of recruiting militia.

Based in Chupaca, a local comerciante, Jose Gabin Esponda, was commissioned to encourage support from the pastoralists in the highlands behind these towns. He was quickly impressed by the enthusiasm with which he was aided by Tomas Laimes, an ex-corporal, native of Huanta in Ayacucho and veteran of the Miraflores campaign (Tello 1971). Laimes mobilized guerri-

lleros from Colca, Huasicancha, and Chongos Alto. Soon a stream of peasants was pouring into Chongos Bajo and Chupaca to volunteer:

Chongos Bajo started up and swelled its forces with volunteers from the highland zone of Chongos Alto, Huasicancha and Colca. In charge of their preparation was Colonel Ceferino Aliaga of Chongos (Bajo). The patriot fighters from these pueblos arrived in Chongos Bajo infected with the courage, fervour and 'voz vanguardia ' of the subsequently famous bandolero (sic ) Tomas Laimes, acting 'cabo ' in the battle of Miraflores. (Samaniego 1972:17)

Troops were organized around the elected authorities of each village, and then battalions were arranged as confederations of neighboring villages. In this way, Tomas Laimes came to head the battalion that revolved around Chongos Alto, Huasicancha, and Colca.

The troops that originally faced the Chileans so disastrously at Miraflores two years earlier had been largely an army of amateurs scorned by the professional soldier, Caceres, and it seems that, on a smaller scale, the comerciantes of Sicaya envisaged a similar field battle: facing the enemy head on with their faithful followers. They were quickly and ruthlessly defeated on 19 April 1882. The Chileans then followed up their victory by driving into the highlands behind Chupaca and Sicaya and sacking Hacienda Laive to provision their troops with meat, an example not lost on the highland montoneras, as events were to show.

Thenceforth, the only effective campaigns were the guerrilla tactics of the montoneras. As a historian of the war describes it:

There always existed the right conditions for a retreat into the high ground around Chongos Alto, Ninalanya, Colca, San Juan de Jarpa, Potaca, etc.; it is worth noting that there existed a broad mountainous territory where the enemy's advantages were rendered null in large part, as much because in ravines artillery and cavalry lost much of their efficiency, as because the soldiers were less well adapted than the guerrilleros to operate in territory with these conditions. Peasants with immense endurance capabilities, accustomed to implausible marches who, accompanied by their inevitable bola de coca (wad of coca leaves), knew no fatigue, were hence dreaded rivals in a war in the mountains. They were able to count, moreover, on a perfect knowledge of the terrain to compensate for the overall disadvantage which resulted from their meagre and primitive arms (mostly pikes, spears, slings and rocks). (Manrique 1981:171–172)

It is important at this stage to remind ourselves of the alliances and conflicts that provide the context for the formation of the montonera fighters of this period. The Chilean soldiers, by no means familiar with the Andean Indios, treated them as scarcely human and this at a time when landlords in the

valley were apparently hosting Chilean officers in their houses. Nevertheless the initiation of the montonera campaign followed closely the vertical ties of allegiance associated with patronage. Caceres was in a sense the last of the caudillos , commanding a personal following of fighters. His aunt was the patron of Hacienda Tucle, and it was around this hacienda that the montoneras were most successfully recruited. When Caceres advanced back into the valley in July of 1882, these guerrilleros swamped Bernarda's Huancayo house. Against Bernarda's advice, the general let them into the reception rooms where they came forward and kissed his hand, calling him "Taita."

The montoneras' opponents formed around these kinds of allegiances. For the pastoralists around Huasicancha and Colca the enemy were not landlords generally at this stage but specific landlords sympathetic to the Chileans—and anyone who followed their leadership.^[6] What this meant for the consolidation of all the pastoralists in the highlands as a body united against a class enemy can be seen by reference once more to Yanacancha. Though pastoralists from all parts were recruited for the Battle of Sicaya, it is worth remembering that those who did the recruiting were the comerciantes of the lowland distrito capitals. These were the same comerciantes who had been intent on maintaining their dominance over the highland pastoralists of Yancancha (and its neighboring communities). In this struggle it had been the hacendado himself, Luis Milon Duarte, head of the Valladares clan, who had "protected" the Yanacanchinos. So, after the defeat of Sicaya, and with Duarte energetically recruiting supporters for the Chileans (Manrique 1981:233–234), the montoneras were unlikely to be focused around Yanacancha.

Meanwhile, enthusiasm for the national cause among medium-sized hacendados and comerciantes began to wane after the defeat of Sicaya, and Caceres, on a brief return to the valley late in 1882, issued orders that patriots were to identify traitors and execute them on their own initiative. When the Chileans reoccupied the valley in May 1883, they were met by a far more docile population than on their previous encounter, and this redefined the objectives of the montoneras around Huasicancha, who were now the only organized body of opposition to the Chilean army that confined itself to the valley. Guerrilla attacks therefore involved forays into the valley and its communications routes south to Huancavelica and Ayacucho, followed by retreats to the highland basecamps.

The campaign then involved two interdependent activities: offenses away from the basecamps and subsistence during the long periods of retreat. The war economy of subsistence (of necessity involving plunder) is simply the other side of the coin in which the more spectacular attacks on the enemy occur. Unless this is acknowledged, we are forced to decide whether the montoneras were either attacking the national enemy (as they maintained) or merely plundering the citizenry indiscriminately like bandits (as landlords and comerciantes in Huancayo later argued).

The first attacks on the Chileans began in July 1883 and doused the last embers of any possible mutual understanding between the commercial classes of Huancayo and the montoneras. As a Chilean battalion moved north toward Jauja, the montoneras attacked its rear from the south. They also used the opportunity to exercise their hostility toward those Huancaínos who appeared to be hosting the occupying army. Whatever their behavior at this time, their arrogant appearance in the town, armed with 200 Remington rifles, was enough for the city's burghers, and a call arose, repeated thereafter, to each and every passing commander to disarm the highland peasants.^[7] Such a call, involving as it did the disarming of the only effective resistance to the Chileans (Manrique 1981:308) reveals the sentiments now brought to the surface by the fortunes of this supposedly national war. To the propertied classes of Huancayo, nationalist fears of the Chileans were nothing to the class fears aroused by the sight of an armed and self-confident peasantry swamping the streets of the provincial capital. And as such sentiments of class hatred hardened in Huancayo and were reciprocated in the highlands; the montoneras now turned to the large haciendas to provide the resources for their war economy.

In late 1883 a Chilean advance on Ayacucho, followed shortly by retreat back to Huancayo, provided the montoneras with further opportunities of attacking Urriola's 1,500 men as they passed through the deep chasms leading out of the Mantaro Valley. A reporter from the Valparaiso newspaper, El Mercurio , referred to cholos , perfectly placed in carefully constructed fortifications, from which they fired down on the troops. A part of the battalion that tried to escape attack passed close to Huasicancha but was ambushed at Macchu, not a mile from the present village.

On the return journey the reporter (certainly with exaggeration) referred to 10,000 Indios molesting the troops from their highly-placed redoubts. Hacendados from Huancavelica provided the Chileans with a militia (Favre 1966:6), but the troops returned to Huancayo exhausted after two weeks of fighting. Leaving a small detachment in Huancayo, Urriola now left for Lima where his troops arrived "worn out and in rags" according to the El Mercurio correspondent. This marks the end of the montoneras' strictly nationalist campaign. On 29 November of that year Caceres wrote of the montoneras: "The pueblos' resistence to recoup the integrity and honour of the nation in these last days will merit a special place in the illustrious pages of Peruvian history" (Manrique 1981:343). But within three months he was issuing instructions to sow discord among these same montoneras (Favre 1975:64). From February 1884 the survival of the guerrillas as a consolidated and defensible force therefore began to surpass in importance their forays out of the mountains to attack an invading enemy. And this campaign of subsistence took on a momentum of its own that set the guerrillas against any and all newcomers.

In fact President Iglesias's signing of a peace treaty with Chile the previous October signified, for most of the valley residents, that the war was essentially

over. (Around Tarma to the north of the valley, Luis Milon Duarte had been appointing local authorities on the basis of their support for the occupying forces.) But the montoneras remained recalcitrant and began to direct their attacks against all who did not actively support them. They had already expropriated the livestock of Tucle's neighboring haciendas (owned by the colaboracionistas, Duarte and Giraldez), and they now turned against all outsiders who threatened their survival. Early in 1884 both Giraldez and Duarte were executed, and terror was used against inhabitants of Chacapampa (Huasicancha's neighboring village) when a number of men had their ears cut off by the guerrilleros. Discipline among the montoneras themselves was also increased against any changes of heart.

When the remaining Chilean forces retired from Huancayo in May, 1884, and Caceres began to move back north toward Huancayo, the montoneras appeared on the outskirts 5,000 strong and threatened to occupy the city.^[8] There appears to have been a general panic in Huancayo, but after the event Laimes withdrew back to the highlands before Caceres's arrival. Caceres's own strategy at this time greatly influenced this new stage of the montoneras' campaign. By the spring of 1884 he was prepared to agree to a peace with the Chileans so that he could turn attention to unseating Iglesias as president of the Republic. To do this he had to reestablish a broad base of support in the valley itself. But what he encountered on his return was an economically distraught group of hacendados and comerciantes united in their demand for the dissolution of the peasant forces in the highlands.

Why had this change occurred? It is necessary to look at the guerrilla campaign which required that between 1,500 and 3,000 people be led and kept in fighting condition, for it was this element of guerrilla warfare that most antagonized the dominant classes of the region. At the outset this part of the guerrilla campaign focused on securing materiel —food as well as goods that could be sold for arms. It was therefore directed against haciendas and not personnel (hacendados or administrators). But after Caceres's orders of February 1884 to encourage discord among the montoneras, a campaign to maintain an undivided fighting force (among guerrilleros from neighboring and hence often rival communities) meant increased discipline directed against fifth columnists.

Although the highland pastoralists' mixed economy of herding in the mountains and cultivating in the sheltered valleys was not significantly interrupted by the war (in contrast to the agriculture of the valley which had to feed the occupying troops), 1881 and 1882 were years of drought that severely affected subsistence crops. The attractions of using the large haciendas as the basis for a war economy of plunder, moreover, was made apparent when the Chileans themselves drove off animals from Laive early in the war. Eventually the montoneras occupied all the haciendas of the region and began, in line with Caceres's orders, to confiscate the property of colaboracionistas

(i.e., the herds of Laive and Antapongo). By the time these resources became exhausted (by the end of 1883 and early 1884) Caceres had ceased to be an ally and his aunt's hacienda—Tucle—previously occupied by the guerrillas, was now plundered for livestock.^[9]

There is no doubt that the pastoralists of Huasicancha did accumulate significant numbers of sheep during this period, and the desire to keep their booty lent great ferocity to their subsequent resistance. But it is also likely that livestock were driven westward to the coast and sold for the purchase of arms, which is what the guerrillas of Comas had done early in the war. Between the montoneras' first appearance outside Huancayo in 1882 and Laimes's capture in 1884, their rifles had increased from 200 to 750. After the war, when the hacendados tried to seek reparations from the peasants as well as turn national opinion against them, they were obviously at pains to exaggerate the extent of their own economic collapse due to the guerrilla campaign. Even so, the extent of the highland pastoralists' destruction of the haciendas can be seen, however partially, from the figures in Table 1.^[10]

There was another element in the survival campaign of the guerrillas—the maintenance of solidarity in extremely harsh conditions. This is a problem that inevitably increases over time, especially when large numbers of compatriots are living relatively peacefully by collaborating with the enemy. The montonera army was organized around existing village institutions of the Personero and his elected council and the varayoc who meted out sanctions within the village (see chap. 1). Both institutions, however, existed only within the bounds of each community and could not be extended across villages. Moreover the relationships between villages had traditionally been those of rivalry over pasture rather than cooperation against the haciendas. For example, Huasicancha's claims to land in Hacienda Antapongo were matched by similar claims made by the community of Chongos Alto, and their claims to Hacienda Tucle were disputed by the neighboring communities of Chacapampa and Moya.

Laimes and his right-hand man, Briseño, however, were not from these villages and their presence certainly helped to overcome local xenophobia. When, however, Laimes shared the spoils from the expropriated haciendas, he acted to promote unity. But when this resource ran out, he turned to other methods. Although Laimes and his followers were often later accused by local historians (residents of Huancayo and members of the professional classes) of assassination (as he was at his trial) very few specific cases are mentioned, given the length of the war and the number of guerrilleros. Those that are, Laimes freely admitted to. They were the executions of Giraldez, Wheelock (the Guatamalan Consul and probably with Giraldez at the time) and the gobernador of Moya. The first two he called "spies" and the last he justified in terms of his refusal to cooperate.^[11] All three assassinations fall within the bounds of Caceres's earlier orders to identiy and execute traitors. But more

significantly, they all took place after Caceres wished to sow discord among the montoneras. In other words, terror, directed both outward and inward (against the fainthearted), was introduced to maintain solidarity.

So the success of the montoneras' campaign was dependent not only on their formal and legitimate goal of molesting the occupying army and discouraging collaboration, it was dependent too on a war economy to provision not just a meager band of twenty to thirty professional guerrilleros but between 1,500 and 3,000 men. And it was also dependent on some form of war discipline. These requirements are interlinked, and as a national war turned into a civil war with elements of a class war emerging too (as was the case in 1884) so these facets ran over into one another.

For this Laimes was held responsible. For the dominant classes it was not at all inconceivable that the stone-faced Indios who kissed Caceres's hand in 1882 could become savage and vicious fighters soon thereafter. But it was, so it seems, inconceivable that they could actually plan and organize such strategies. And so within three weeks of his arrival back in Huancayo, Caceres set about the task of capturing Laimes and his three aides, Brisefio, Vilchez, and Santisteban.

It was a surprisingly easy task. Caceres had some 1,200 thoroughly worn-out regular soldiers, but no campaign was necessary, no siege, not even threats. It is a fascinating reflection on the complexity of their political awareness that these fighters who knew that they had molested the nation's enemies and that they were feared throughout the area, expected in consequence to be respected and treated on equal terms by the professional army. It is dif-

ficult otherwise to explain how Caceres so easily pulled Laimes into his web, there to be executed. Indeed on receiving Caceres's orders to present himself and his troops in Huancayo, Laimes replied from his mountain stronghold: "Tell Caceres that I am as much a general as he is and if he wants me to go to Huancayo he must treat me as equal to equal" (Tello 1971:77). And with a naive faith in Caceres's ultimate justice and fairness the montoneras descended into the valley and proceeded to Huancayo presumably with a view to striking a bargain with respect to the land under their control.

Once there, however, Laimes and his lieutenants were quickly separated from their troops, who were quartered in a schoolhouse (previously requisitioned as the Chileans' barracks) and surrounded. On 2 July 1884 a Council of War tried the four men, found them guilty of assassination, robbery, flagellation, and mutilation and executed them by firing squad. Laimes's statement, as recorded in the national newspaper, El Comercio (19 July 1884) is worth quoting at length:

After cross-examination he confessed having sacked the Haciendas Tucle, Laive and Ingahuasi whose produce had been divided up among his men. With reference to the death of the Gobernador of Mayo, he was dead as a result of his hostile reception (of the troops); and to the deaths of Señores Wheelock (Consul of Guatemala) and Giraldez which were done by his aide, Briseño, who assassinated them for having been believed to be spies for the Chileans, en el trayacto de Izcuchaca donde se les conducía presos; that he had accepted the honorific tribute of an Inca emperor was because he had been drunk (beodo ), to which other crimes could be ascribed; that it was certain that he had himself assassinated, or had assassinated, all who were accused of being traitors to la Patria; that thus it was that some had been mutilated who were believed to be enemies and guerrilleros who had incurred 'faltas.'

The montoneras, now supposedly rendered powerless without leadership and anyway in no mood to be left around Huancayo, had to be encouraged to retire peacefully. On 29 July a supreme decree was issued: "offering total guarantees to a comunero of Huasicancha in lands known as Analanya, Ananhuanca, Huaculpuquio and Patapata for having been servidor como guerrillero en la causa de la Nacion" (A.C.).^[12] The montoneras retired back to the hills, but their control of the area did not cease. Manrique, concluding his study of the Chilean-Peruvian War in the Mantaro Valley, writes "The postwar pacification process was extraordinarily complex and merits a separate study on which I am engaged" (1981:366), and the evidence of Huasicancha's position during this period is scant. But two years after Laimes's execution, the municipal proceedings of Huancayo still recorded discussion of the return of the four haciendas, and Bernarda Pielago's will of 1887 refers to the peasants' control of the hacienda and, more importantly, the hacienda's

livestock. Manrique estimates (1981:368) that as late as 1888, 45 haciendas from the south of the region up to Cerro de Pasco were still in the hands of guerrillas.

These experiences are a vital element in the Huasicanchinos' culture of opposition. As recently as 1981 villagers were pointing out bluffs and caves in the highlands and saying, "This is where we [sic] ambushed the Chileans" and "Here is where we hid during the guerrilla war." The war is talked about as a period in which villagers showed great tactical skills and cunning but were constantly betrayed by outsiders. During periods of repression when quiescence is forced upon them, oral histories of this period serve as important means for establishing their interpretation against the more widely accepted, prevailing "official version."

Thus, in 1937, when Huasicancha was engaged in its first major confrontation with Hacienda Tucle since the "montonera decade" (that period from 1882 to 1892) a prominent figure in Huancayo, later to become an APRA senator (Alianza Popular Revolucionaria de America [Popular Revolutionary Alliance of America])^[13] wrote of the guerrilleros: "Laimes with the pretext of combatting the Chileans only dedicated himself to committing crimes. He and his compañeros , far from confronting the enemy, killed defenseless compatriots. He went to his death a palid cholo of small build, with all the signs of a bad man, more dead than alive, weeping and in a cold sweat" (Tello 1971:75). This is far from the memory shared by the Huasicanchinos. Among them, the violence remembered is that perpetrated by Laimes's enemies. According to the popular history of Laimes's end, on his way to a conference with his cofighter Caceres, Laimes, unarmed because he trusted his companion-in-arms, was ambushed and then savagely and brutally executed.^[14]

The "montonera decade" was a drama which began with all the intrigue of the heroes in a Greek play and ended with the plebeian chorus taking center stage and refusing to let the curtain fall. It might be possible for a historian to uncover the economic forces and social structure that gave rise to the conjuncture of events around the Mantaro Valley, but I am more concerned here with the results of the political mobilization of Huasicanchinos than the cause. I think in this respect it is important to stress two elements in the unfolding of the campaign. The first is that it isolated the highland pastoralists around Huasicancha and broke their ties to the personal figures of patronage that opened this period. Despite the propensity to dwell on the character of leaders such as Laimes or the political ambitions of Caceres, the evidence suggests that these personalities became less and less decisive in the course of events. The highlands were not pacified, for example, until at least five years after Laimes's death.

And the second element offers something of an explanation of this. At a certain point political mobilization took on a momentum of its own. Too often, in our concern to get at social and economic causes of political move-

ments, we tend to forget that the movements themselves create their own institutions and ideas. For the purposes of understanding contemporary Huasicanchino culture, the important point to remember is that these do not entirely disappear with the end of hostilities. The institutions and ideas through which daily life is expressed in periods of apparent quiescence contain also a perpetual resonance of this oppositional culture, as is suggested by the statement of the Huasicanchino migrant, which opened this chapter.

The Response to Hacienda Expansion 1900–1947

The overt and concerted resistance of the previous two decades set the conditions for the Huasicanchinos' ongoing struggles as they entered the twentieth century. But many of the tendencies were contradictory. The montonera decade established the reputation of the Huasicanchinos as recalcitrant fighters, but the hacienda now laid claim to much larger stretches of land than ever before and hence made the pastoralists in fact weaker. As the century progressed the commercialization of the Mantaro Valley encouraged the expanded hacienda to rationalize production, which in turn reinforced the decline in the role of a patrón as the protector of her or his peasants, a decline that had already occurred in the bitter wake of the guerrilla campaign that had turned Caceres from patrón to enemy. But, ironically, the expansion of hacienda control of pasture effectively made the hacienda administrator the distributor of pasture, thereby preempting the ability of the community authorities to do so and hence undercutting their role in the reproduction of the pastoral households. Finally, though the hacienda's expansion in effect encouraged the Huasicanchinos to migrate away, depopulating the village and temporarily shattering the old ties of interdependency within and between households, the long-term effect of the Huasicanchinos' involvement in the highly volatile marginal economy of Lima and Huancayo was to reformulate those ties and eventually to put pressure once more on the land.

It is now to these more daily aspects of struggle and interdependency, that I turn, focusing first on the hacienda and then on the Huasicanchinos themselves.

The Social Relations of Hacienda Production: Tucle 1900–1920

Under successive governments in the period following the 1880s the land of the peasant communities of the central Andes was swallowed up by expanding haciendas. The rebellion of the montoneras had not been an isolated

incident. To the north of the area the Atusparia rebellion had gained a brief success (Stein 1976) and to the south the rebellion in Huanta from 1890 to 1896 had resulted in Ayacucho being taken over entirely by the rebels. But then a reaction set in. An upswing in agricultural prices beginning in 1896 gave impetus to unprecedented offensives on the part of hacendados, aided by a sympathetic central government. This political and economic climate remained favorable to the hacendados until the early 1920s. Wool prices steadily increased from 1914 until 1921, when they collapsed. Meanwhile the persistence of adverse conditions suffered by peasants eventually led to further outbreaks of rebellion from 1918 onward, chiefly in the south of the country, which led to a softening of attitudes with the coming of Leguia's government in the following year. From 1919 to 1923 there was a brief period in which communities were able to reassert their claims to land, but again this was followed by repression.

The owners of Hacienda Tucle responded to the pacification of the guerrilleros by making sweeping claims to land. At the time of Bernarda Pielago's will in 1886 the hacienda claimed 12,000 hectares of land. In 1906, her heirs registered the hacienda in Huancayo's registry of properties as covering an area of 103,274 hectares, over eight times as much land and an area greater than the combined total of all three major haciendas—Laive, Antapongo, and Tucle—at the time of the Reforma Agraria of 1970!

Huasicancha, as a community, made claims against the hacienda in 1889 and 1902, but they had little effect. However, the new constitution of 1919 provided for the registration of all Indian community lands as well as the setting up of a Dirección de Asuntos Indigenas under the direction of the founder of Peru's Socialist Party, Castro Pozo. In November of that year Huasicancha registered its claims to land in the Huancayo registry office. It claimed all the land of Hacienda Tucle and Rio de la Virgen, much of the land of Hacienda Antapongo, as well as the pastures of the neighboring communities of Palaco and Chongos Alto. Such claims were of little practical significance at the time, of course: nine other communities laid claim to parts of Hacienda Tucle, and Chongos Alto claimed most of Hacienda Antapongo. Although the registration of deeds in 1919 was later to become crucial in proving legal claims to land, its effect at the time was minimal. By 1923 Castro Pozo was driven into exile, and the cause of Indian communities was left in the hands of a variety of nongovernmental pressure groups in the rising tide of 'Indigenismo' in Lima.

In turning to the relations of production on Hacienda Tucle, I shall refer in turn to the hacienda's control over land and its mobilization of labor. The two are crucially tied together. Tucle's claim to land was not less legitimate than that of most other highland haciendas, but Huasicancha's persistent counterclaims and the pastoralists' insistence on their own rights to pasture their animals were an endemic hindrance to the hacienda's attempts to ra-

tionalize relations of production along capitalistic lines, that is to say, along lines based upon contractual rights to property on the one hand and contractual transactions based on the payment of a wage for labor on the other.

What then was the basis of Hacienda Tucle's claim to ownership of its lands? In fact there were, eventually, two haciendas involved, Tucle and Rio de la Virgen. Bernarda Pielago's claim to Hacienda Tucle was based on two forms of evidence appropriate to two concepts of possession. Vis-à-vis her fellow landlords she had her deed of sale from the Valladares: vis-à-vis the neighboring peasants there was not a contract but a performance, mediated by a judge of the province of Jauja and witnessed by Don Mariano Tacona, representative of the community of Huasicancha, together with other comuneros from the village:

All these being gathered here at this place Doña Bernarda de Pielago proceeded to take the judge by the hand and thus took possession by the authority of the nation and in recognition thereof she did throw stones, pull up plants and perform other demonstrative acts of taking real and corporal possession, saying in a loud voice, "Possession, possession, possession." And under the same procedure we went on to Huaichan-puquio and from there to Yutempampa where there is a round corral of stones. . . . Taking all these places in this form, encountering complete quiet and no opposition or contradiction in any way. (Signed by the judge and all witnesses; F.H.T.)^[15]

Ownership of Hacienda Tucle was never based on any other evidence. Throughout all the litigation that followed in the hundred or so years after this, the owners of Hacienda Tucle never produced this nor any other evidence of the legitimacy of their claim to the hacienda (Juzgado de Tierra, Huancayo).

Bernarda left the hacienda to the three illegitimate children of her housekeeper. That they shared the name of Bernarda Pielago's onetime partner, the now defunct Faustino Chavez, allows us to speculate that these were his offspring. In any event, Maria Luisa Chavez, who eventually bought out the shares of her two brothers, was the daughter of a woman of Huasicancha where she continued to have cousins throughout the struggles that followed. (There was not much of the "radically distinctive" Spanish landlord left among the owners of Tucle.)

Although Maria Luisa Chavez was not herself especially concerned with the efficient running of the hacienda (in contrast to Bernarda Pielago), which she inherited thoroughly denuded of livestock, she was always intent on consolidating her landholdings. It was she who expanded Tucle's claim eightfold in 1906, and soon thereafter she married Manuel Duarte whose kinship ties to the Valladareses would have allowed their son to consolidate Haciendas Laive and Tucle under one owner. So aware of this were the Huasicanchinos that, when Manuel died leaving one son and Maria Luisa married her tenant

(of Hacienda Tucle), villagers believe that she allowed the two subsequent offspring to die so as to ensure the unpartitioned succession of her son by Duarte.

In 1920 then, Maria Luisa Chavez rented 50 percent of Hacienda Tucle to Manuel Pielago, whom she was later to marry, in 1937. Her attempts to restock Tucle had been only partially successful, and the increment in wool prices after 1914 had for her the adverse effect of driving up the price of the stud animals she needed to buy. Pielago, however, was an aggressive member of Huancayo's commercial class whose interests included the growing and milling of flax, a hide processing business, and firms connected with mining supplies (see Roberts 1974b ). From his tenancy onward there was a concerted effort to rationalize production on Hacienda Tucle.

To this end, Pielago became interested in the small neighboring hacienda of Rio de la Virgen whose irrigated land made possible the growing of crops that Pielago intended to use for fodder in order to raise dairy cattle. This hacienda had been bought by a certain Juan Bazo Velarde from the Parochial Council of Huancayo in 1926. The title to this land (that Huasicancha claimed in its entirety) was based simply on the council's claim to the land "from time immemorial and at least the past forty years (sic!)," but there were no documents and the council sold Velarde much more land than he was ever able to find (see documents of Juzgado de Tierra, Huancayo). Pielago was able to buy the hacienda because Velarde, on his death, left 54 percent of it to his wife and the rest to his eleven children. Under such circumstances most heirs chose to liquidate their holding and split the cash. Such were the bases upon which the landlords of Haciendas Tucle and Rio de la Virgen "owned" their land.

But how did they meet the need for labor on the hacienda? The annual demands of the enterprise required three types of labor: permanent shepherds; seasonal team labor (chiefly for sowing and harvesting at Rio de la Virgen), and occasional day labor. All three were met, in principle, without cash changing hands.

The most important work required by the hacienda was shepherding, which required the services of a family of two or three adults and a number of children. These were the huacchilleros discussed earlier. After 1900 Hacienda Tucle had acquired virtually all the pasture suitable for sheep grazing, while the Huasicanchinos, for their part, had as a result of the guerrilla campaign of the 1880s acquired a considerable number of sheep (as well as llamas and alpacas). A family undertook to care for the hacienda's animals in return for being allowed to "enter" the hacienda territory with a designated number of their own animals: the huacchas. Substantial numbers of animals were pastured in this way. Of the roughly 250 households in Huasicancha, Tucle employed about thirty at any one time as shepherds. Agreements between the hacienda administration and the shepherds allowed the latter to bring often

as many as 200 head of sheep with them. While a few families held these jobs for extensive periods of ten to fifteen years, most were dismissed after a year or two or themselves chose to leave, so the position of huacchillero moved through the community.

The administrator, of course, sought reliable and skilled shepherds, but rarely if ever did Huasicanchinos graze just those animals agreed upon in the contract. Apart from the natural increase in flocks through the years, most shepherds were also grazing the animals of neighbors from the village and, with Tucle still understocked, it is likely that huaccha sheep accounted for at least 50 percent of the animals using hacienda pasture.^[16] In the early part of the century the administration was less concerned about the huacchas' use of pasture than the fact that shepherds devoted much of their time and energies to their own animals and polluted the hacienda flocks by interbreeding with their own chuscos ("mongrels").^[17]

The hacienda's requirement for team labor was met by the community institution of the faena. Correspondence between the hacienda and the community authorities in the first two decades of the twentieth century indicates that faenas were mobilized for the hacienda on the basis of the hacendado being an ersatz member of the community who, like the curaca in days past, had rights to the occasional labor of the community as a whole. Faena work was taken by both parties as an obligation of the community to the patrón, repaid only by the traditional gifts of coca, chicha (maize beer), and a meal on the day plus gifts to the village on feast days.

Day labor was the least important requirement for a livestock hacienda but even so there were always jobs that had to be done on the basis of the jornal (i.e., by the day). This demand was met chiefly by the use of pasaderos , that is, those villagers who were caught trespassing on the hacienda pasture with their animals. In these circumstances, the animals were driven off to the hacienda's "coso" (described by the villagers as "the animal prison"), where they were kept until their owners had performed sufficient day labor to gain their return.

Bearing in mind that the owners of Hacienda Tucle were laying claim to lands that they could not possibly stock with sheep, historians must ask the question: What was the relationship between Tucle's insistent claims on territory (despite evidently shaky foundations) and its mobilization of labor? If the question is asked slightly differently—What was the mechanism that induced peasants to leave work on their farms and in the community to labor for the hacienda?—the answer to both questions seems at first obvious. Haciendas expanded their claims to land (even when they did not use it) in order to create a scarcity of land for Indians and hence induce them to work for the hacendados, that is, the so-called primitive accumulation (Marx 1976: 873–895).

This is a nice argument, although care must be taken not to confuse cause

with effect. It is undeniable that the effect of hacienda claims was to make community lands scarce, leading peasants into relationships with the hacienda, but haciendas may nonetheless have expanded for a variety of reasons to give this consequence, intended or unintended. By claiming community land without stocking it, the hacienda was, I shall argue, not creating land scarcity for pastoralists per se but rather undercutting a resource on which the politics of the community depended, in favor of the patron of the hacienda. Regardless of the motive of the hacendados, the effect of hacienda expansion at the beginning of this century was to channel important community institutions through the medium of the hacienda patrón, rather than the community authorities.

Obviously the social relationships between hacendados and neighboring comuneros varied considerably from one place to another,^[18] but even in Huasicancha where a long history of confrontation and independence existed, it would be incautious to underestimate an equally long history in which Andean peasants have been expected and, in their turn therefore expected, to offer services to superiors who often went under kin terms or pseudo-kin terms such as taita or padrino/patrón (godparent/patron) (see chap. 2). This is not to deny that such an ideology could not have been maintained without at least a minimum of political might and, as ideas lost their intensity, so the political means might have gained theirs. In the aftermath of the montonera decade, for example, Tucle had the political backing of the state (the army, the police, the legislature), but the ideology of hierarchy had taken a severe beating during the guerrilla campaign.

The mechanism inducing peasants to work for the hacienda then, could have been of three kinds. It may have been, to all intents and purposes, a tradition in which villagers accepted as the order of things that they should offer labor to certain superiors: curaca, Inca, patrón. It may have been that an element of physical force was used to set the process in motion and then maintain it. Or it may have been that peasants were driven, out of necessity, to work for the hacienda to gain access thereby to resources insufficiently available on their farms and in the community. While one or other of these ideological, political, or economic mechanisms may at times have been paramount, the important fact is that all three played a role in the social relations of hacienda production.

In this initial period of hacienda expansion, Tucle continued to mobilize labor through the manipulation of traditional community institutions whose functions they thereby transformed. The direct result of this was to deprive the formal community of Huasicancha of its political component and to substitute the hacienda administrator for the village elders (essentially the better-off family heads in Huasicancha). Because Tucle was unable to restock, its claims to territory did not immediately create land scarcity for comuneros of

Huasicancha, nor did it deprive better-off families of extrahousehold labor since the hacienda's demands were not great. Indeed expansion, if anything, increased the availability of impoverished households willing to work for better-off families. What it did do, however, was to reduce the role of the community authorities in the organization of pastoral production:

1. By depriving the community of its lands, control over pasture—specifically allotting highland estancias to shepherds—was taken away from the community authorities, thus depriving them of the basis for their power.

2. Because community lands were so decimated, the role of the faena was proportionately reduced. Insofar as I argued in the previous chapter that this was a form of rent-by-performance in payment for household plots distributed by the authorities in direct proportion to work done at the faena, so reduction of faenas on community land undercut the reciprocal tie between the formal "community" and access to household plots, so that these plots came to be seen as private property whose distribution was removed from the hands of community authorities to those of the family head. Arable landholdings were thus effectively privatized once access to them was no longer a function of the obligation to work on the community lands, which in turn resulted from the decimation of those lands by the hacienda.

The effect of hacienda expansion at the beginning of the century was not to deprive better-off pastoralists of land nor of labor but to reduce the power they derived from holding political office as community authorities. This could no longer be used either as a means of securing the best pasture sites nor of steering community labor in directions best suited to their interests. I shall argue in a moment, after we have looked at the Huasicanchino petty production side of this, that despite this reduction in the political component of community, the social and economic reproduction of only partially commoditized petty production continued to require the essential features of community social relations of production.

Hacienda expansion at the beginning of the century then did not mean a clean sweep, a revolution in the social relations of production in which ideological and political mechanisms were replaced by the iron hand of economic necessity. As a result, the conversion of the hacienda into a capitalistic firm was delayed. Day labor was only available when peasants obliged, by trespassing and then getting caught. Faenas were the occasion for much socializing and rowdiness. And shepherds were, in effect, sheep farmers competing with their employer for pasture. With the arrival on the scene of an aggressive, entrepreneurial tenant in 1920, therefore, attempts were made to ratio-

nalize production so that, through the use of cash relationships, labor would become more responsive to the needs of capital. Of major concern is how this process affected petty production in Huasicancha.

The Social Relations of Petty Production: Huasicancha 1900–1947

A description of the social relations of production in Huasicancha must deal with the same two dimensions discussed for the hacienda: the relationship between control over the physical resources needed for production and the organization of the labor process. The nature of petty production changed over this period, and these changes provide the basis for the heterogeneity among Huasicanchinos. The route along which these production relations developed was to give rise to structural incompatibilities between the community and the hacienda, which, in turn, became the source for political unrest.

At the beginning of the century, the people of Huasicancha maintained tenancy relations with the hacienda, and despite their remote location, also worked in a local mine (Cercapuquio), and engaged in widespread trading. These links in turn served to increase the heterogeneity among the different units of production. Changes in the form of petty production, moreover, must be placed within the context of national and regional developments. For the past eighty years this has predominantly been a reflection of the growth of different sectors of the Peruvian economy such as cotton production, mining, smelting, banking, wool production, and manufacturing.

The most immediate and oldest of these relationships has been the one between petty production units and Hacienda Tucle, which, at the beginning of this century, controlled at least 30,000 hectares of land.^[19] For a long time changes that occurred in petty production in Huasicancha were closely related to the operations of Tucle. The fortunes of Tucle, in turn, were closely intertwined with changes in the Peruvian economy. At the beginning of the century, Tucle was not participating in the expansion of wool production that occurred among other highland haciendas. It claimed large amounts of the community's pasture, but the real effort to turn those claims into effective control, as also the attempts to rationalize production along cash lines, were delayed until after the First World War, twenty years later. This means that two periods' in the history of the Huasicanchinos' petty production between 1900 and 1947 can be distinguished.

1900 to 1920 . The hacienda's operations were diversified. Wool production was the mainstay, but arable crops were also produced. A limited amount of labor from Huasicancha was required and claims were made on the community's pasture.

1920 to 1947 . The hacienda specialized in livestock production, both sheep and cattle. This led to real expansion onto the available pastures and an attempt to rationalize production. Huasicanchinos began to enter other areas of the labor market so that extrafarm income became increasingly important, and ties to the hacienda were made less important.

1900 to 1920.   The nature of petty production during this period was greatly influenced by the decision of Hacienda Tucle not to become heavily involved in the expanding wool production that took place in Peru after 1890. (Over this period, both the neighboring haciendas, Antapongo and Laive-Ingahuasi, were taken over by a banking consortium which turned them into part of the Sociedad Ganadera del Centro , and sold shares to Lima businessmen.)

Huasicanchinos were involved in a variety of different activities in addition to farming. A certain amount of specialization occurred from one household to another, and this had three important implications for social relations. First, complementary ties occurred between households of differing specialization. Second, these at first minor differences in "occupational mix" gave rise to differences in wealth. And third, specialization and wealth differences in turn gave rise to distinct social relationships both within and between households. The forms of specialization were herding, agriculture and trading.

Pastoralism.   While none of Huasicancha's flock owners was fully committed to herding activities, the dictates of llama and sheep farming dominated the organization of production. Llama and sheep husbandry at high altitudes has three important characteristics: it is transhuman; its success is greatly dependent upon the skill and conscientiousness of a small number of shepherds; and it demands a heavy input of relatively skilled labor at certain periods of the year. Each of these characteristics has important implications for organization within the production unit, as well as for the nature of its ties to other units.

The year-round labor of household members varied seasonally. During the rainy season (October to April) sheep could be pastured within easy reach of the village, at lower altitudes. Shepherding during this season was therefore undertaken by all members of the household from the age of ten or twelve and older. Llamas, however, were only brought down for short periods during the season, so that shepherding at high altitudes was still necessary. During the dry season (May to September) sheep and llamas alike were moved up to seek out the scarce grasses of the high zone. Access to these pastures was of great importance and highland estancias (the shepherds' dwelling and corral) were usually shared among a number of households. The task of shepherding at these altitudes was a far more arduous and vigilant job than lower

down. Flocks were most vulnerable during the dry season, and the strategic location of high altitude estancias was, therefore, an important variable contributing to the success of the farm.

Much of this pasture was held by Hacienda Tucle, so that hacienda shepherds became the nodes of sets of relationships involving ownership of sheep concentrated at one particular estancia. (In other words a hacienda shepherd's huacchas were not necessarily all, or even principally, his own.) Much pasture, however, was held in such remote areas that, claimed by the hacienda or not, it was "no man's land." Though in principle this pasture was not held privately, in practice, as pasture became scarce, estanceros began to guard with some jealousy what they regarded as their grazing area. Fights to establish grazing rights over pasture often involved conflicts between villagers and between communities (as well as against the hacienda), so that it was important to have a concentration of mature men at the estancias.

The organization of the estancia gave rise to a complex set of interhousehold ties, as households' flocks were intermingled on the estancia. Although this meant that for herding, households did operate in complementary clusters, the corporate nature of these clusters was weakened because of the multiplex ties of each household to a number of estancias and the short-lived nature of each of these ties. Since a newly-formed household received animals from parents on both sides, its partnership in a particular estancia had to be negotiated from the start. In fact households did not necessarily concentrate all the animals to which they had some claim in any one estancia flock. Rather they were scattered through a variety of estancias, reflecting both inheritance claims and the various prestations of households over the years. The composition of estancias was far less a function of kinship than of instrumental decisions by livestock owners in the entire operation of their production unit. Apart from the various obligations to others emerging in the spheres of, say, arable farming and intervillage trading, a household's ties into an estancia reflected both the hazards of animal husbandry itself and the changing labor composition of the household over its life cycle.

But the given nature of household composition could be adjusted to the needs of livestock farming through two kinds of reciprocal arrangement. The first of these was the huacchilla. Relationships among Huasicanchinos (as opposed to relationships between them and the hacienda) functioned much as they had in the past: when the household lacked young men for high-altitude shepherding, members of households deficient in livestock were recruited. Such people became shepherds for a household with sheep, being rewarded with a share of the lambs and llama calves. The accumulated animals grazing alongside those of the larger flock owner came to be known as the huaccha animals. The second institution was that of michipa. Here the flock owner took in and cared for the animals of another household. A multitude of exchanges of goods and services were taken care of in this way, such as labor

on the flock owner's arable land, transportation services on the part of a muleteer, and so forth.

These two institutions allowed a flexibility in the operations of the herding sector of the production unit. A household with a large number of women, for example, found it convenient to maintain its livestock as michipas in the estancia of another, while performing arable farming tasks in the vicinity of the village, in return. The huacchilla, on the other hand, allowed for the accumulation of sheep beyond the constraints deriving from the composition of the household. This arrangement was used to advantage by some households insofar as livestock owners were in the habit of delaying the handing over of huacchas to the shepherd when, for example, he chose to marry and set up house. Pressures were then brought to bear by the community or the relatives of the huacchillero or his wife, but in many cases this led the better-endowed household to adopt the huacchillero as a son rather than release him with all the sheep to which he could lay claim.

In fact, much of the hacienda shepherd's flock consisted of michipas, that is to say, many of the animals grazing on the hacienda were in fact owned by households having no direct ties to the hacienda but having instead relations of reciprocity with the households employed by the hacienda because the demands made by the administration on its shepherds were such that the job required all the time of all the active members of the household. As a result, hacienda shepherds became increasingly indebted to other households who fulfilled tasks for them in the village, especially on their arable plots. While these arrangements did give rise to some socioeconomic differentiation (see Smith 1979a ), they also involved essential links of interdependence between heterogeneous domestic enterprises.

Pasturage for sheep and alpacas was available to Huasicanchinos, therefore, either through seeking out high puna unattractive to the hacienda or through taking huacchas onto hacienda land while acting as employees. If all else failed there was always trespassing. Since the boundaries of Tucle were ceaselessly in dispute, this was anyway an occupational hazard for all herders. But the most consistent pasaderos were the small herders with ten to fifteen sheep—the majority of Huasicanchinos. The hardest time for these farmers was the dry season, when grasses were most scarce, and this was the period when they were most consistently caught by the hacienda's caporales (foremen). Offenders had to pay a fine in labor-service, but this was the season when there was little work to be done on the arable plots, so that the household suffered little in loss of labor. So astute were the villagers in gearing their trespassing to the seasonal demands of their farm operations that the hacienda administration had to adjust its operations in order to take advantage of the excess of labor which thus arose as a function of the farming rationale of the petty producers.

Although Tucle's real control of pasture was limited, its expansion at the

beginning of the century was an overall constraint on comuneros' accumulation of livestock. But, if anything, this gave the edge to the few households with well-located estancias. At the same time those with larger flocks were encouraged to find means for pasturing them that tied them neither to the hacienda nor to the community. In the latter part of this period two brothers (Pedro and Silvestre Hinostrosa: see chap. 4) actually began renting pasture from communities in Huancavelica for an annual cash fee. And the wool boom from 1914 to 1921 (in which international wool prices doubled) gave opportunities for a few others to secure jobs as caporales and mayordomos on those haciendas a little removed from Huasicancha that were becoming more commercialized and paying good wages for such jobs.^[20]

Nevertheless, there were constraints on the long-term and systematic accumulation of livestock, and these were connected to the personal skills and conscientiousness so valued by highland pastoralists everywhere. As far as year-round shepherding was concerned, the necessary personal commitment was encouraged because inheritance was partible—including adopted workers. But this in turn split up the large herds in each generation. Meanwhile the demand for skilled seasonal labor led to crises for the overexpanded during those seasons of the year, such as shearing and parturition, when such skills were required.

Agriculture.   Arable farming was by no means as important to the Huasicanchinos as livestock husbandry. Many of the household needs in this sector had been traditionally obtained through bartering livestock products for goods grown more easily at lower altitudes. However, households became increasingly involved in arable farming over this period, firstly because Hacienda Tucle's expansion onto grazing land reduced Huasicancha's bartering power, and secondly, because farmers in the Mantaro Valley, who had hitherto bartered their crops for livestock products, were increasingly producing commercial crops for the national market. Thus during this period, the majority of households witnessed a reduction of herding in favor of arable farming.

The effect was to reduce the importance of the communal organization of production while giving greater dominance to reciprocal ties negotiated directly between households. In the past, arable land was divided into two areas: "communal fields" farmed by the community as a whole, and fields farmed by each household. These latter took the form of strips of land (faldas) running down the hillsides and controlled by the village's three barrios. This situation changed with the growing need for arable land. The effect of increased demand for this land was to reduce the overall land area devoted to the communal arable fields and of practically ending the barrios' planting decisions and redistribution rights. Plots were not returned to the village authorities when heads of households died but were handed on directly to offspring on a partible basis. A patchwork quilt of small arable plots began to obscure the old logic of the barrios.

As with herding, the fixed nature of household labor in any one year could be overcome through various reciprocal arrangements. Extrahousehold labor was necessary especially at ploughing and harvest times, but, as with herding, its amount and form were a function of the relationship between the composition of the household and the amount of property worked. Harvesting could be undertaken with the aid of uyay, which involved the reciprocal exchange of labor or minka, which further involved the payment of part of the harvest for labor. There was as much heterogeneity in arable farming as there was in raising livestock. Apart from division of labor resulting from differences in crops produced from the microclimates and soil differences of each household's plot, ploughing with oxen involved clusters of households forming around the rotation of one pair of oxen, and in some cases ploughing teams offered their services out to those unequipped for the job, payment being through minka or michipa arrangements.

Trade.   Trade beyond the village also involved some specialization among households, basically of two kinds: trueque (bartering) with households in the high jungle and muleteering, with some trading, among the small mines of the area. While the latter had always been the affair of a few specialists, trueque had originally been an activity shared by all households, but no longer. As bartering relationships with households in the Mantaro Valley dried up, trueque continued with households in the high jungle, which was at a lower ecological niche than the Mantaro. But, as arable land became more important, the demands for labor at harvest time increased. Since harvest and jungle trueque occurred at similar times, the effect of increasing arable work was to put great pressure on the household's labor. So, where originally each household was involved in the trueque journey, now a small group of young men were assigned the task on behalf of a number of households.

Trueque's contribution to differentiation derived from this growing, if minimal, degree of specialization. It was not limited to the exchange of equivalents between direct producers whose differing ecological locations led to complementary demands for different use-values: a Huasicanchino herder exchanging fleece for a jungle-dweller's corn, for example. Those who undertook the trueque journey obtained rock salt and dried meat from the higher-altitude community of Acobambilla in Huancavelica. Raw wool, spun yarn, hides and animals on the hoof were provided from various households in Huasicancha itself. All these were exchanged for corn and coca leaf in the high jungle. But part of the return consignment was then used for obtaining commodities in the cash market of the Mantaro Valley on the return journey. These commodities were in turn exchanged with other households (in Acobambilla and Huasicancha), leaving the trader with a profit.

Hacienda Tucle's expansion at the beginning of the century, then, by no means shattered the preexisting social relations of production, driving land-

starved peasants into a purely economic relationship to the hacienda. The effects were somewhat more complex. Membership in a community of Huasicanchinos continued to be important for the functioning of petty production through the social relations upon which it relied—that is to say, the bundle of institutions such as huacchilla, michipa, minka, and uyay. But these were now far less mediated or controlled through the community officeholders' abilities to exercise authority. The allotting of estancias and of arable plots was now out of their hands.

The better-off pastoralists were, through the hacienda's decimation of community resources in land, deprived of the political means for securing their positions through office in the community at a time when their economic well-being was increasingly linked to relationships beyond the community—to renting land, shepherding for the hacienda, and earning a salary as caporales. Nevertheless, insofar as many of the relationships on which petty production depended were still a feature of the community, Huasicanchinos continued to have a shared interest in its maintenance.

1920 to 1947.   The overall trend in this period was the growth of the hacienda as a commercially oriented firm and a severe reduction in the operation of the Huasicanchinos as small wool and meat producers. During the latter part of the previous period, both the hacienda and the larger village herders sold some livestock to the commercial market; from 1920 onward the larger-scale firm was progressively to dominate in supplying this market. This process was a direct result of changes that were occurring in the structure of the Peruvian economy.

As the resource base for the Huasicancha's farms was eroded, social relations became increasingly dominated by the dictates of nonfarm, petty commodity production in the urban centers and of wage labor. Insofar as increasing amounts of labor power were shifted from within to beyond Huasicancha, the parameters for the expropriation of surplus labor were expanded to include other sectors of the economy. Here relations of production and distribution were far more thoroughly expressed in terms of commodities than had hitherto occurred within Huasicancha. This increased commoditization of social relationships and the extension of these relationships to a variety of nonagricultural pursuits provided a new set of forces for heterogeneity among Huasicanchino enterprises.

In Huancayo too, the balance of social forces changed. Nationally a commercial bourgeoisie gradually replaced the previously dominant rentier class, giving a new direction to the development of the Peruvian political economy, and this transformation was especially noticeable in Huancayo. Wool, Peru's fifth largest export at that time (Thorp and Bertram 1978), doubled its price on the United States and British markets during the First World War (Mitchell

1962; U.S. Department of Commerce 1960). Such a market attracted a Huancayo businessman, Manuel Pielago, to rent 50 percent of Hacienda Tucle from Maria Luisa.^[21] But no sooner had favorable wool prices on the international market attracted a capitalist farmer to Tucle, than the economic scene changed once more. The hacienda never supplied raw wool directly to the international market because the domestic demand for livestock products picked up in the 1920s while international raw wool prices declined.

The effects of all this on Hacienda Tucle were slow. In the early 1920s the American-owned Cerro de Pasco Corporation set up its smelting operations to the northwest of the Mantaro Valley. Much of the city's commercial growth was tied to provision of goods and services for the mines and smelter (Roberts 1974b ). Huancayo was also the site of two medium-sized textile mills (Roberts 1974b ), and there were mills in Lima too that were supplied by highland haciendas (Thorp and Bertram 1978). The railway now made it possible for highland cattle ranchers to supply dairy produce to Lima, and meat was also being transported down to the national capital from the haciendas of the central Andes (Miller 1974). The combination of Hacienda Tucle's remote location and the sustained resistance of the local peasantry, however, meant that it still lagged behind its neighbors. During the 1920s the hacienda increased its emphasis on wool production, but it was not until near the end of this period that its cattle stock was significantly improved.

The effects of Peru's changing economic structure on Huasicancha were threefold: in the transformation that occurred in the village's relations to Hacienda Tucle; in the growing influence of the Huancayo nexus on the economy of the Huasicanchinos; and in the demands that occurred in the labor market first on the part of coastal cotton plantations, and second in the construction industry.

Where the Mantaro Valley had hitherto been just one of the economic areas in which Huasicanchinos were involved, it now became increasingly influential. The operations of the smelter as far to the north of the valley as Huasicancha was to the south, the penetration of the railway again from the north and ending midway down the valley, and the termination of the central highway to Huancayo by the same route in 1939, were all to change the center of gravity for the village. Even so, the effects of the opening of the smelter were by no means as direct or as thorough (in terms of labor recruitment or the opening up of demand for farm produce) as they were for the communities situated to the immediate north of Huancayo (cf., Laite 1981). Rather it was the cotton plantations of the coastal region to the west of Huasicancha which now came to modify social relations within the village.

During the 1920s and 1930s, under the entrepreneurial influence of Manuel Pielago, Tucle began actually to use the pasture it had hitherto only claimed. It increased its own stock of animals, and it increased its demand for skilled and conscientious shepherds. This had profound effects on the Huasican-

chinos: by reducing the local resource base it also reduced the strength of the interhousehold ties described above; it drove a few herders ever higher into the mountains; and it made trespassing onto the hacienda land a way of life. In retrospect it can be seen as a watershed for petty commodity production in which the linkages between units were reduced to be reformulated once more during the 1960s.

Pielago employed a Scottish administrator and commissioned an agronomist to do a thorough report on the hacienda. It was found that hacienda stock were greatly at a disadvantage as a result of the practice of paying shepherds in the form of the huacchilla. The administrator pressed for the upgrading of the sheep to Corriedale purebreds, and the agronomist pressed for the introduction of wages in lieu of huacchas. If shepherds became entirely dependent on a wage their first commitment would be to the hacienda, whereas huaccheros were in fact small-scale farmers whose first commitment was to the maintenance of their own herds (F.H.T. and A.R.A.). Time and again shepherds, obeying the seasonal demands of their own animals, simply deserted the hacienda sheep in their care and went off in search of other estancias. Agreements with Antapongo and Laive not to employ such shepherds were hard to enforce, and anyway those that did not so leave were frequently found to have increased their huaccha herds inordinately. Moreover the hacienda was not only providing the shepherds with pasture but with sheep as well: the agronomist's report found that 73 percent of newborn lambs were reported to have been lost in birth or carried off by "foxes" (the foxes, of course, being the Huasicanchinos). However, if hacienda sheep could be upgraded then they would produce more wool from a given amount of pasture than the shepherds' chuscos, so that the wages paid to shepherds could be worth more than the value of their sheep but less than the value of the upgraded purebreds that replaced them, thus benefitting both parties. Similar arguments were made for the use of team labor that could either be paid per head or in a lump sum. As for trespassers, they would now be required to pay a fine in cash and thus release the hacienda from the inflexibility inherent in the present system.

These proposals were a direct threat to the social relations of petty commodity production among Huasicanchinos. I have already shown that through michipa, it was not the hacienda shepherds alone who benefitted from the practice of huacchilla, now to be replaced by wages. The huacchilla was not merely important for the direct cash equivalent of the livestock involved but because it fitted into a particular slot in the operation of a small enterprise run with extremely scarce, unstable, and varied resources. For the large flock owners too, their mountain estancias were further depleted by the renewed vigor of the hacienda administration. This drove them still further toward renting land from other communities and even from a number of small haciendas in Huancavelica. Others who had taken up jobs as caporales on other haciendas, while maintaining their houses and small plots in Huasicancha,

moved their households away from the village and began to send their sons to Huancayo for schooling or apprenticeships that would prepare them for nonagricultural pursuits. These, however, were the better-off households.

The poorest households were the hardest hit and the most isolated from the old interhousehold economic relationships. Their abilities to pasture sheep—for themselves or on behalf of others in return for reciprocal services—were forestalled because they could not find the cash for pasadero fines. Some were able to take advantage of the labor shortages in better-off households who sent their children to school in Huancayo. Now these households needed huacchilleros. Others were maintained by specializing in skills that had hitherto been undertaken by all households as part of the peasant farm, such as pottery, weaving, and making harnesses, so that the division of labor between households became more emphatic.

Between these two groups, the remaining households became involved in the seasonal migration of selected members. With the exception of the poorest, virtually all households sent members to work on the coastal cotton plantations at some time during the period 1920 to 1939. The demand for labor was greatest from March to June (thus absenting men from the village during the beginning of the harvest), but in fact many people tried to find other casual labor for longer periods. During the second presidency of Leguia (1920–1930) the government devoted more of the national budget than ever before to construction work, so that the most usual job was labor on construction projects, some on the coast, some in the Mantaro Valley. Virtually all those who eventually worked under the contratos viales (the road construction program of Leguia and, later, Benavides) after 1936 were recruited while on the coast.

Although remittances were insufficient during this period, by the end of the period the scene was set for increasing the role of extrafarm income through migration, firstly to Huancayo and later to Lima. From the 1930s, households that had sent members to Huancayo began to move their residence to the town. What commitments they still had in Huasicancha were now maintained through members of the immediate household returning during important periods of the year and through huacchilla arrangements in the case of livestock farming. Then, in 1938, the bottom fell out of the cotton market, while those who had been working on road-building gangs were released. A number of these people journeyed up the coast to Lima. Few of them remained permanently in the city, but by the end of this period there was a small colony of Huasicanchinos established in an inner-city slum.

In the previous period there was considerable heterogeneity in the form of the petty production unit. Most were involved in a variety of different occupations, but the various permutations of these occupations within each unit led to significant differences between one and another. These internal properties of the units of production had an effect on the relations between them. With the exception of those involved directly in providing labor for the ha-

cienda, the bulk of productive relations were concentrated within the community itself. A household possessing a certain ratio of personnel to types of available resources at a particular period in its development cycle sought out others to complement its own productive operations.

During the second period, however, changes in the configuration of the regional economy—especially Huancayo's increased economic expansion as a result primarily of the Cerro de Pasco's operations—led to the involvement of dynamic commercial interests in the running of Hacienda Tucle, leading to its increased use of pasture. Reductions in livestock production by small farmers followed, and thus changed the internal configuration of each unit. These changes varied from one unit to another, depending on the particular commitments and resources of each during the early years of this period. In a few cases larger herders reduced their livestock commitment, bought property—like a small corner shop—in Huancayo and finally moved their entire household there. If they kept a few sheep, they were cared for by poorer households within the village, while members of the wealthier household returned during crucial periods to keep up minimal arable production.

But for the majority, the transformation took the form of maintaining the units' multioccupational character while fragmenting it as a nucleated household. Internally, the increase in nonfarm occupations did not have the effect of reducing the importance of producing subsistence items on the farm. On the contrary, great emphasis was now placed on the ability of the unit to produce subsistence items, which in turn modified the role of women. Such was especially the case for livestock farming. Women increasingly took over the running of the estancias for long periods during the year. This was then reflected in inheritance patterns; many newly-formed households received the larger share of their livestock through the wife, and the location of their estancias thus tended to be determined by reference to the wife and her mother.

With respect to relations external to the production unit, the new occupational combinations meant that the unit, as an integrated functioning entity, had far fewer sets of ties with other such units. Production relations were no longer confined to the community of Huasicanchinos: households, straddling uncomfortably across wide geographical regions for long periods of the year, had few resources to commit to the continued reproduction of such relations. So the configuration of a household economy came to depend increasingly on nonfarm occupations. The diverse directions of their subsequent development were now dependent upon such factors as: (a) the character of the particular sector of the national economy to which members were tied, be it agriculture, mining, or urban commerce; (b) the nature of those ties, for example, whether they involved the sale of labor or the sale of products; and (c) the degree to which they were committed to that sector, in terms of capital invested in nonfarm operations on the one hand, and the amount of farm resources to fall back on, on the other.

By the end of the period a system of closely interrelated petty production

units whose reproduction had been a function of the unmediated social relations of the community was being threatened by a process of commodification. In fact the same development of the Peruvian economy which, by stimulating the hacienda to rationalize production, was pushing these people off their land, was simultaneously creating the interstices in the cash economy that they were to occupy. But whether drawn into the cash economy as commodity producers, whether drawn in as wage laborers, whether left behind with the drudgery of subsistence, or whether a combination of all of these, each Huasicanchino household had reasons for resisting this process or at the least trying to control its direction. Against the steadily mounting pressures to erode Huasicancha as a community, following the montonera period, this common ground within heterogeneity coalesced once more around the (much modified) institutions of (a now geographically dispersed) community.

And, insofar as the imperatives of capitalist development that were enveloping the highland pastoralists in a variety of ways were most immediately visible in the form of the neighboring commercializing hacienda, Huasicanchinos concentrated their resistance in that direction. In effect this meant what it had always meant, in one way or another: a question of control over resources. But the parameters were different, for this now represented an attempt to replace the pasture hitherto made available through being hacienda huaccheros and pasaderos, with pasture gained by invading the hacienda. The hacienda was therefore faced with Hobson's choice: the "legitimate" encroachment on their pasture by huacchilleros contracted by the hacienda or the less controllable "invasion" of pasture by land-hungry herders (cf., Barona 1963; Martinez-Alier 1973, 1977).

Throughout the 1930s and 1940s this stalemate continued, simultaneously preventing the hacienda from developing into a streamlined capitalist enterprise, and also restricting the development of Huasicanchino petty commodity production. And it is from here that we shall take things up in the next chapter. But what occurred between 1947 and 1948 encapsulates the contradictory processes that have been described. A concerted campaign to recapture land on the part of the Huasicanchinos met with a disappointing settlement in 1948, resulting in increasing out-migration throughout the repressive Odria years that followed. Ironically though, the configuration of the Huasicanchino economy that resulted did not, over the long run, reduce the number of their livestock. Although increasing engagement in nonagricultural occupations temporarily reduced the number of livestock from the 1920–1947 figures, village households as a whole began to hold ever greater numbers of sheep from 1950 to 1972, thus providing the basis for the major mobilization that ended in the later years. To understand why this was so and why, albeit in much modified form, the institutions of community remained the means of its expression, I turn now to the participants in that mobilization, in order to get to know them better.