The Pacajes-Inka phase settlement data clearly suggest a policy of
forced resettlement on the part of the Inka empire. Large numbers
of people were relocated from more inland areas such as the Tiwanaku
Valley to areas closer to the lakeshore - including the Taraco Peninsula
and various islands in the lake - which are much more agriculturally
productive. This reversed a very long-term trend of population movement
away from the lakeshore to inland locations. This inland migration
began in the LF1 and continued through the end of the LIP. This suggests
a complete abandonment of raised field agriculture![[*]](file:/usr/share/latex2html/icons/footnote.png){kind=icon}
as well as either a reduction in the importance of pastoralism or
a reorganization of pastoral production. It may be that under the
umbrella of a pax incaica groups were able to live at a distance
from their herds. This would not really have been possible in the
more bellicose and dangerous environment of the LIP. Pacajes-Inka
phase groups and families may have had large camelid herds in inland
areas which were tended by a rotating detail of herders.
This removal of the necessity for propinquity between a household and its animals also removed what I have suggested was the original reason for the shift from nucleated to dispersed habitation in the Early Pacajes phase (see discussion in Chapter 9). Indeed, in the Pacajes-Inka phase some villages were founded and a site size hierarchy appeared. I hypothesize that these small villages represent the households of local leaders who participated in the Inka administrative hierarchy together with those of their dependants and retainers. Thus, the Inka went some way toward reestablishing the system of lakeshore agricultural villages which had existed prior to the Tiwanaku collapse. However, the local population still resided predominantly (probably more than 70%) in dispersed farmsteads.
On the whole, then, the Pacajes-Inka settlement pattern represents the reorganization of the local population in order to increase agricultural production. This increase in agricultural production may or may not have come at the expense of pastoral production, depending on whether and in what ways management of camelid herds was modified. This question goes well beyond the scope of the present discussion, however.
The Late Pacajes settlement pattern is in most ways a straightforward continuation of the Pacajes-Inka pattern. The population continued to reside in dispersed households, with a few small villages, and one substantial one (Sonaji [T-271]). All of the village locations changed however. That is, all of the Pacajes-Inka phase villages were abandoned or reduced to hamlet status, and new villages were founded. This seems to me to be further evidence that the Pacajes-Inka ``villages'' were not in fact true villages but rather an agglomeration of dependent households around a single wealthy or influential one. All such configurations would have depended for their stability on continued privileged access to the workings of the Inka administrative hierarchy. With the collapse of the Inka Empire these ``villages'' would have dissolved into their component households. Seen in this light, Sonaji was probably not a true Spanish ``town'' either, but rather a concentration of dependent and attached households about a representative of the crown. That is, it was the same kind of phenomenon as the Pacajes-Inka ``villages''.
However, the Late Pacajes settlement pattern does display one peculiarity which was absent from the Pacajes-Inka pattern, and which indicates a somewhat more heterogeneous political landscape. This peculiarity is a radical change in site and population density evident in the settlement location (Figure 10.4) and density (Figure 10.6b) maps. These figures show an area of very low site and population density between the 520000E and 522000E meridians. The 3 km stretch has very few sites in the Late Pacajes period, and the ones which are present are very small. Figures 10.3 and 10.6a clearly show that this area was well-populated in the Pacajes-Inka phase.
Alert readers will recall that this line is quite near to one I identified as a political boundary in the Late Formative and Tiwanaku periods (see discussions in Chapters 7 and 8). The earlier boundary was in fact located at 519000E, more or less, several kilometers to the west of the line I am discussing. It is sufficiently nearby, however, to suppose that the Late Pacajes settlement discontinuity may represent a political boundary of some kind as well.
|
In order to investigate this possibility, I compared population index growth rates on either side of this boundary, which I fixed at 521000E. These data are presented in Table 10.3. This table clearly shows that growth rates were roughly similar on both sides of the line during the Early Pacajes and Pacajes-Inka phases. Thus it seems not to have constituted a boundary at all in these phase as far as demographic processes are concerned. However, in the Late Pacajes phase growth rates diverge drastically. Population grew by 0.81% annually on the western side of the line, a very rapid growth rate indeed, undoubtedly implying a considerable population influx. On the eastern side of the line, however, population decreased at an annual rate if -0.54%, again a rapid rate of decline.
These data almost certainly indicate that population was moving out of the Katari Basin and onto the Taraco Peninsula during the Late Pacajes phase, and at a rapid rate. What could case this kind of population displacement?
Spanish administration of the Titicaca Basin was probably regularized around the same time as the discovery of the Potosí silver deposits in southern Bolivia in 1545. At this time, the Titicaca Basin became essentially a sustaining area of the Potosí mines. Its population was reorganized in order to provide food, coca, other supplies, and, critically, manpower to the mining operations. The founding of the city of La Paz in 1548 was part of this process. Titicaca Basin groups were subjected to various administrative strategies - encomienda, repartimiento, corregimiento - which served to enrich their conquerors and to impoverish themselves.
Later in the colonial period - the mid-seventeenth century to be exact - Taraco was a marka (see [Albarracín-Jordan 1996a] and [Albarracín-Jordan 1996b] for more detailed discussion of the concept) or town with eight ayllu or communities under its jurisdiction. In the Lower Tiwanaku Valley it included the ayllu of Iwawe, Chivo, Pillapi, and Jawira Pampa ([Mamani Condori 1991]: 22). Pillapi and Jawira Pampa are still communities today. This allows us to locate the seventeenth century borders of Taraco as follows:
Of course, there is no guarantee that these were the boundaries of sixteenth century (Late Pacajes) Taraco also. In fact, the town of Taraco did not yet exist in the sixteenth century as far as I have been able to determine. Sonaji (T-271) was the only town on the peninsula at this time. Further research will be necessary to confirm the sixteenth century borders of what we now call Taraco. However, given the settlement data already discussed it is not unreasonable to assume for the moment that it was the same as the seventeenth century borders just defined.
Assuming that the borders were the same, the entirety of Late Pacajes
Taraco has been archaeologically surveyed, together with portions
of neighboring administrative units. The area of the Lower Tiwanaku
Valley contained territory pertaining to both the encomiendas
(later repartimientos) of Tiwanaku and Guaqui. And now an interesting question emerges. If Taraco's boundary with
Huacullani was characterized by such a marked difference in population
growth rates, might not its boundaries with Guaqui and Tiwanaku display
the same discontinuity? To answer this question, I entered the Lower
Tiwanaku Valley settlement data collected by Albarracín-Jordan into the same database I used for my own sites (see Appendix F).
I corrected his site sizes and derived population measures in the
manner described in Chapter 4. Thus, the
resulting data were entirely comparable to my own from the Taraco
Peninsula. I then calculated population growth rates for the area
of the Lower Tiwanaku Valley which was a part of Taraco in the Late
Pacajes phase, as opposed to the area which pertained to Tiwanaku
or Guaqui at that time. The results are presented in Table 10.4.
|
Remarkably, though perhaps not surprisingly, the Lower Tiwanaku Valley
data display exactly the same patterning as the Taraco Peninsula data.
Both areas grew at about the same rate during the Early Pacajes phase,
and the population of both declined strongly in the Pacajes-Inka phase.
In other words, this boundary, like the Taraco-Huacullani boundary,
was not demographically significant during the LIP or the Late Horizon.
However, in the Late Pacajes phase the Taraco side of the boundary
grew at the very rapid rate of 0.64% annually, which the population
of the rest of the Lower Tiwanaku Valley declined even more rapidly,
at -1.29% annually.
It appears, then, that in the Late Pacajes phase people were deserting the encomiendas of Tiwanaku, Guaqui and Huacullani at a rapid rate, and were relocating to the territory of Taraco. Albarracín-Jordan recognized that a general population decline took place in the Late Pacajes phase in the Lower Tiwanaku Valley.
... no se puede ignorar la depressión demográfica que se presentó hacia comienzos del siglo XVII. La minería de plata, en Potosí, se convirtió en principal factor de desiquilibrio poblacional en la región. El Valle Bajo de Tiwanaku fue incorporado a la mit'a. Un gran número de tributarios fueron reclutados para trabajar las minas de Potosí ... La versión española de la mit'a hizo que de 868 tributarios, enlistados en el Repartimiento de Tiwanaku en 1583, solamente resten nueve, en 1658... [[Albarracín-Jordan 1996a]: 313]Albarracín-Jordan attributes the population decline to mit'a obligations in the Potosí mines, and other onerous exactions of the colonial government and its representatives. These processes are well-documented (cf. [Choque Canqui 1993,Sánchez-Albornoz 1979,Spalding 1984,Stern 1982]) and relatively well-understood. Throughout the seventeenth century extremely onerous labor exactions, as well as very poor working conditions and a series of catastrophic epidemics, decimated indigenous populations throughout the Andes.
How, though, are we to understand the meteoric population growth on the Taraco Peninsula in this context? A clue is provided by the fact that Taraco was not, in the Early Colonial period, given in encomienda to any private individual. Rather, the peninsula was tierra de la corona, crown territory (Waldemar Espinoza Soriano, pers. comm.), and was ruled directly by the viceregal government. Though documentation for this period from Taraco has never to my knowledge been published, it may be that conditions were more comfortable under crown rule in the Late Pacajes phase than in any of the neighboring encomiendas.
As Roberto Choque Canqui puts it, ``[l]os pueblos de Waki y Tiwanaku estaban `tan despoblados y perdidos sin gente por la mita' porque muchos indios tributarios se han huido ... `` ([Choque Canqui 1993]: 83). If they were fleeing from Tiwanaku and Guaqui, perhaps some at least were fleeing to Taraco. The anomalously high Late Pacajes population growth rates may therefore be explained by an administrative anomaly: Taraco was ruled by the crown, and surrounding areas by encomenderos. The possibility is certainly an intriguing one and merits further research.