Population Data for Indian Peru: Sixteenth and Seventeenth Centuries

Accurate counting of the aboriginal populations of the Andean region was not a priority of the early Spanish conquerors. It was only after the vast amount of precious metals that had been amassed by the natives over centuries began to run out that the Europeans turned their attention to the true source of wealth: Indian workers. At this chronological juncture, in Peru in the middle of the sixteenth century, the number of inhabitants became an object for investigation by colonial bureaucrats. By the late 1560s and early 1570s, the framework for the systematic collection of demographic information had been elaborated, and for the next few decades officials made periodic head counts. The purpose of the counts was fiscal: to provide the statistical evidence for an equitable tribute assessment. Census taking continued until the population declined to the level where the cost of making the enumeration outweighed the benefits. Following catastrophic depopulation, those dependent on Indian tribute began to delay new counts in order to keep tribute collection high. Few assessments were made between the second decade of the seventeenth century, and the administration of Viceroy Palata in the 1680s.

My purpose here is to make available population data that I have collected during the past decade from various archives in Europe and America, as well as from published materials.¹ The collection of demographic material is at best a slow task, for counts are often reported in a variety of types of documents: mercedes of encomiendas, general administrative reports, legal suits over boundaries of Indian grants, tribute records, and so forth.

Most early population counts made in Peru are of little use for systematic analysis. The numbers from the 1530s are notoriously imprecise. Reports of the size of the population of Chincha range from about 25,000 to 50,000, and illustrate the difficulty in using figures from the first years of the colony.² Do the numbers refer to all Indians, or male heads of household, or perhaps tribute-paying adult males? And, further, how was the number generated: by an actual Inca census, or by the impressionistic estimate of an early European eyewitness? In addition, there is the question of the size of the territory that was covered, as in the Chincha case, where the figures may refer to the entire valley population, not to the later repartimiento unit, which held only one part of the area’s total inhabitants. Special care should be exercised when attempting to analyze the figures of the 1530s that I have included in the table.

By the time of the end of the Indian uprisings of the late 1530s, and the seizure and distribution of the liquid wealth that could be found during the first years of the conquest, the Spanish began to devote closer attention to the potential size and value of the Indian grants (repartimientos) that were being made by Francisco Pizarro. Pizarro set out to organize inspections of several of the Indian grants that had been made to the conqueror-turned-settler. He hoped to provide information on Spanish holdings, agricultural products, mining, and the treatment of Indians by their encomenderos. His inspectors were required to record individual inhabitants, house by house, and to ascertain the number of tributaries. One purpose of the census was to end the disputes over Indian grants that had begun to plague colonial authorities. Pizarro also planned to collect the statistical data needed to effect a general distribution of repartimientos among the Spanish who wished to remain in America.³ Few of Pizarro’s visitas have been located, and there are problems in using those that have been found. For example, Pizarro adopted an age definition of tributaries different from that used later; repartimiento boundaries changed between the 1540s and 1570s; there is no report of the total population in the 1540s; and an undetermined number of males of tribute-paying age were exempt.⁴

Two sets of population data originated during the administration of Pedro de la Gasca. The first set is based on the Huaynarima distribution of Indians (August 17, 1548) by Gasca to Spaniards who ultimately joined the crown during the Gonzalo Pizarro revolt. The second set derives from the general inspection and tribute assessment of repartimientos ordered by Gasca. The inspection was still in progress when the administrator returned to Spain early in 1550. The allotment, made in Huaynarima, a small village outside Cuzco, was made by President Gasca with the help of Archbishop of Lima Jerónimo de Loaysa and the notary Pedro López. The men carried with them files containing various reports and a register of previous repartimientos. Unfortunately, the exact location of the grants in this report is vague; even major villages are often not mentioned. Second, the number of tributaries is approximate, frequently being rounded off to the nearest ten or hundred. Third, the population figures are not dated, and could refer to tributaries as early as the middle 1530s, or as late as 1541; by 1548 the population was substantially different. Fourth, no figures for the total population are given.⁵

Pedro de la Gasca, like Pizarro before him, decided that a complete inspection and a tribute assessment were necessary, and entrusted Archbishop Loaysa with the task. The archbishop was assisted by Friar Tomás de San Martín, Friar Domingo de Santo Tomás, and Hernando de Santillán. Visitas of repartimientos were filed in quadruplicate; one copy was intended for the local Indian kuraka, one for the audiencia, another for the tribute collector, and a final one for the Council of the Indies in Spain. Few of the several hundred copies originally filed have been located. The extant copies of the tribute assessments of individual repartimientos are rich in economic data, and are of exceptional value for the historian. Each item of tribute is meticulously listed. The demographic information in the Gasca tribute assessment, however, is disappointing. The inspectors must at least have estimated the number of tributaries in each repartimiento, but rarely are tribute figures included in the final report.⁶

Subsequent assessments were made in the 1550s. The demographic value of these is similar to that of President Gasca’s census. In 1561 Pedro de Avendaño, secretary of the audiencia, compiled a new list of repartimientos. He remitted the summary to Spain, along with a more complete list that included the location, the current encomendero, the name of the official who made the grant, and the amount of tribute.⁷ Unfortunately, Avendaño did not list the number of Indian tributaries for each grant, although he did provide tributary totals by region within the viceroyalty. His reports appear to have been the primary source for the population data included in a geographical compilation prepared by Juan López de Velasco.⁸

Important regional inspections of Peru were made in the 1560s: the 1562 Huánuco visita and that of 1567 for Chucuito are good examples. Copies of both documents have been published, and both are excellent sources for demographic studies.⁹

The most important source for a study of the historical demography of late sixteenth-century Peru is Viceroy Toledo’s general census and tribute assessment of the 1570s. On December 28, 1568, Philip II ordered Toledo to compile a tax register (libro de tasa) for repartimientos. The register was to include the names of encomenderos, the amount and type of tribute, and the number of tributaries. Information on the history of the local assessment was also requested. Toledo decided to undertake a detailed and complete inspection of the viceroyalty, naming sixty-three secular and ecclesiastical inspectors, one or more for each administrative district. These visitadores were ordered to ascertain how many kurakas (“chieftains ”) and parcialidades (“divisions,” often ayllus, and sometimes moieties) were in each repartimiento. The register was to include married tributaries and children (legitimate and illegitimate), orphans, and blind, ill, crippled, or otherwise incapacitated males of tribute age. The number of old and unmarried women was to be reported. Males not present during the investigation had to be included on the register, with notation of the length of their absence and current residence. The quantity, type, frequency, and place of tribute payment were to be written down. The visitadores were ordered to inspect local parish records of baptisms, marriages, and deaths to check if kurakas were attempting to conceal the true number of residents. There exist several summary copies of the Toledo inspection based on the aggregate population and tribute records.¹⁰

Equally good counts followed in the late sixteenth and early seventeenth centuries. The revisitas were made because of catastrophic depopulation, and the consequent inability of Indians to remit tribute at earlier assessment levels. European epidemics devastated Indian communities. Toledo’s resettlement policy increased population densities, and thus contributed to higher disease losses. Severe droughts and floods frequently destroyed coastal valley agricultural systems. As a result of a series of particularly serious epidemics between 1589 and 1591, most repartimientos had to be reassessed.¹¹ The findings of these revisitas were compiled for Viceroy Garcia Hurtado de Mendoza, second Marqués de Cañete, by the audiencia secretary, Luis de Morales Figueroa, who had kept on file all tribute documents. It was relatively simple for Morales Figueroa to prepare a list of the number of tributaries and the amount of tribute for each repartimiento. He reported that the compilation was based on the “Libro de las tasas de la visita generai” of Viceroy Toledo, and on following reassessments of repartimientos (before 1591). Many new quotas were made under Viceroy Cañete because of disastrous population losses. Documents of the period reveal that in the areas where disease mortality was the highest, the shrunken Indian population was incapable of paying tribute at the earlier rate. In many areas the corregidors jailed Indian officials for failing to collect tribute and confiscated land and property. In severely affected regions, Viceroy Cañete excused payment by tributaries for one year.

Several historians have consulted published copies of the Morales Figueroa document to study various aspects of the social and economic condition of Peru in the late sixteenth century. Unfortunately, the published versions are faulty, and a series of errors in detail and generalization has followed. The marred edition was published several times between 1867 and 1935 by Luis Torres de Mendoza, Sebastián Lorente, Manuel de Mendiburu, Enrique Torres Saldamando, and Silvio Zavala. The publications are based on a late copy of the document compiled by the famous eighteenth-century historian Juan Bautista Muñoz. Muñoz often transposed the numbers 4 and 6, did not report the name of the encomendero, and gave no indication of the census date for each repartimiento. A more reliable early version of the Morales Figueroa document was collected in Peru by Juan Luis López in the late seventeenth century. Luis López, the Marqués del Risco, was secretary to Viceroy Palata, and also corregidor of the mercury mining center of Huancavelica. In Pern, Luis López put together several volumes of important documents and prepared a history of Peru’s famous mercury mine. One of the documents the Marqués gathered was a copy of the Morales Figueroa summary of repartimientos.¹² This copy has not yet been published.

Another major source of population data for Peru was compiled by Antonio Vázquez de Espinosa, a Spanish Carmelite, born in Jerez de la Frontera sometime in the last half of the sixteenth century. He resided in Peru from 1615 to 1619. In 1622 he returned to Spain and there probably wrote the Compendio y descripción de las Indias Occidentales. He died in Seville in 1630. In the Compendio, Vázquez lists each repartimiento in the viceroyalty, with the number of tributaries, boys under eighteen, old and disabled men, and women. The value of the tribute is also given. Vázquez de Espinosa probably collected many of the figures during residence in Peru. He generally used the most recent available census totals for each repartimiento. His Compendio contains a wealth of population information, but has no chronological meaning. The inhabitants of one repartimiento could have been counted in 1610. In another repartimiento a census might have been taken in 1591. And in many cases, Vázquez de Espinosa did not have access to population data other than that from the 1570s of Viceroy Toledo. Unfortunately, he did not record the census dates along with his figures. The result is a collection of figures that initially appears accurate, yet on close examination is highly misleading.

The Vázquez de Espinosa figures can often be dated by use of other documents, especially original inspections, which the compiler might have had at hand when he prepared his manuscript in the early seventeenth century. I have attempted to do this in the present list. Particularly good sources for the population counts are to be found in the Archivo Nacional del Perú in Lima. There, the Residencias section includes several original census counts, or copies that contain the number of tributaries of repartimientos. The residencias of the corregidores de los indios provide a detailed view of the working of colonial Indian rural society. The Derecho Indígena y Encomiendas section often gives similar information. Indeed, it is in this section that the published inspection of the Indians of Chupachos of 1562 was discovered. The category of Real Hacienda also at times indicates census dates and tribute payments to the royal authorities. Similar material is found in the Archivo General de Indias in Seville, in the sections of Audiencia de Lima, Contaduría, and Indiferente General, as well as Justicia. The best data are to be found in Peru itself, however, and not in Spain.

Local parish archives in Peru, as in the case of Yanque in the Coica Valley, north of the Spanish colonial center of Arequipa, appear to contain not only the regular parish registers of marriages, births, and deaths, but also occasionally copies of sixteenth-century population counts. Unfortunately, the process of cataloging and microfilming parish registers in Peru has not progressed as rapidly as in other Latin American republics, so we may not know the full extent of the richness of local Peruvian church records for some time. Naturally, the present compilation of population data does not include the parish returns. Parish and repartimiento boundaries are not necessarily the same. Indian repartimientos could be made up of several distinct communities, each with its own parish. Consequently, the figures provided in parish counts are usually not the same as those of repartimientos.

Some Indians may have escaped the census-taker in sixteenth- and seventeenth-century Peru. The Cañaris were exempt from tribute payment altogether, and were rarely counted. The yanaconas, or “special servants,” also may have escaped enumeration, although strenuous efforts were made to get them counted and included in the normal tribute system in the 1570s and 1580s. After Viceroy Toledo’s establishment of the special mining mitas for Potosí and Huancavelica, many Indians found it advantageous to migrate out of their native provinces. As forasteros (those who had left their home communities), they lost access to land and ayllu, but were exempted from the mita. Increasingly in the seventeenth century, emigration to avoid mita service negatively influenced the number of tributaries (originarios) in many corregimientos, particularly in southern Peru where mine labor was especially arduous. While the total of originarios was decreasing, the number of forasteros was increasing.

Few population counts were made in the half century following 1620, for several reasons. By delaying counts as long as possible, those who benefited most by high recorded numbers of tributaries maximized their revenues. In some cases, Indians paid the Toledo tribute assessment up to a century later, even though their number had dwindled to a fraction of its earlier size. Furthermore, by the 1680s, it often cost more to send census-takers to small isolated communities in the Andes than could be collected from the local annual tribute payment.

No population list will be totally accurate. Even contemporary censuses of modern industrialized nations err. Underenumeration in major cities is a weakness of the United States census of 1980. Yet one cannot help being impressed as one reads the census manuscripts prepared by sixteenth- and seventeenth-century Spanish bureaucrats. In the Lima Indian count of 1613, for example, the enumerator took care to record those absent, and checked residents with marriage, birth, and death registers of the parish priest. Similar care was exercised in isolated Indian communities, with kurakas and corregidors, encomenderos and clergymen contributing to the elaboration of the count. Fraud may have existed, since the basis of the count was fiscal, but the fact that the aims of the interest groups involved in preparing the census did not always coincide quite possibly kept the level of fraud within limits.

Author notes