Several aspects of the Tiwanaku Period settlement system are worthy
of note. First of all, after the radical population decline in the
LF2, population growth returned to normal in the Middle Horizon. The
annual rate of population growth for the Taraco Peninsula as a whole
was 0.08% during the Tiwanaku period (Table 8.1),
close to what is an expected growth rate for preindustrial village
societies. This means that the dramatic population movement from the
Taraco Peninsula into the urbanizing center of Tiwanaku had ceased.
This is not to say, of course, that Tiwanaku ceased to grow. It continued
to increase in size, though at a slower rate than in the LF1.![[*]](file:/usr/share/latex2html/icons/footnote.png){kind=icon}
It is only to observe that the population moving into Tiwanaku was
coming from somewhere other than the Taraco Peninsula.
|
As far as the site size hierarchy is concerned, the Tiwanaku Period
is not so different from the Late Formative. The Taraco Peninsula
settlement system still displays a three-tier site size hierarchy
(Figure 8.3), just as it had in the LF1 and
LF2. In fact, the same sites, roughly, occupy the same positions in
the settlement hierarchy. The Santa Rosa group (T-271/T-272/T-322)
continues to be by far the largest population concentration on the
peninsula, with a total of more than 33 ha of occupation area. Chiripa
(T-1), Yanapata (T-123/T-130), Titicachi (T-394/T-415)and Chiripa Pata (T-4) continue to occupy the second tier. In addition,
Tiwanaku itself became a primate center sometime in the LF2, alone
comprising a fourth, higher-level tier in the settlement hierarchy.
However, if we look carefully at the relative population growth rates of the various levels in the Tiwanaku Period site size hierarchy (Table 8.2), it becomes clear that in fact an important change has taken place in relation to patterns observed in the Late Formative. It will be recalled that in both phases of the LF, as discussed in Chapter 7, an asymmetry was noted in the growth rates of the various settlements on the Taraco Peninsula. Throughout the Late Formative, one group of sites, comprising the political and demographic center of the peninsula, grew relatively very rapidly. In the early LF1 this was Kala Uyuni, and in the later LF1 and the LF2 it was the Santa Rosa group. The remainder of the sites on the peninsula grew more slowly. Additionally, a trend was noted to the effect that smaller sites (hamlets and smaller villages; third-tier settlements) increased in terms of their collective size, while the middle range sites, those of the second-tier, grew slowest of all. This fact was interpreted as reflecting the political and economic dominance of the leaders of the first-tier community, and the decreasing importance and influence of the leaders of the second-tier communities. This configuration was argued to be consistent with the formation of a multicommunity polity, and the usurpation the functions of the second-tier leaders by those of the first tier.
Tier 1: T-271, T-272, T-322 Tier 2: T-1, T-4, T-123, T-130, T-394, T-415 Tier 3: The remainder.
|
The Tiwanaku Period settlement data present exactly the opposite picture (Table 8.2). in the Tiwanaku Period the first- and third-tier sites grew at a rate below the phase average (0.05% and 0.06% annually, respectively), while the second-tier sites grew more rapidly, at an annual rate of 0.13%. This fact is extremely interesting. I interpret it as suggestive of a change in strategies of Tiwanaku administration on the Taraco Peninsula. If in fact Tiwanaku politically dominated the peninsula during the LF2, it did so through the intermediation of the leaders of the Santa Rosa group. These leaders, by virtue of their intermediate position, retained access to resources and to labor sources which were not available to leaders of the second-tier communities. In the Tiwanaku Period, by contrast, it appears that the Tiwanaku state circumvented the Santa Rosa group to some degree. The disproportionate growth rate of the second-tier sites at this time suggests that the leaders of these communities had increased in importance and in access to resources (by this time less exotic trade goods and more special craft goods produced and distributed by the state apparatus). On the Taraco Peninsula, at least. it seems that the mature Tiwanaku state was administered on a community by community basis, and that the importance of the regional capital (the Santa Rosa group) had begun to wane. This scenario is entirely consistent with my earlier suggestions regarding the near-universal distribution of decorated pottery in the Tiwanaku period. The elites of the old Taraco Peninsula Polity were no longer capable of monopolizing decorated ceramics. I take this to be another indication of their increasing irrelevance.
Of course, there is no reason to suppose that this pattern characterized Tiwanaku administration in all of the heartland provinces. In the Katari Basin, at least, there is evidence that the role of the regional capital - Lukurmata - increased vastly in importance during the Tiwanaku Period. Janusek has estimated that the area of the metropolitan cluster of Lukurmata in the Tiwanaku Period was close to 200 ha, some 87% (200/229 ha) of the total Tiwanaku Period occupation area recorded in his Pampa Koani survey area ([Janusek and Kolata 2002]). We may suppose, therefore, that the settlement determinants of the Pampa Koani settlement system were radically different from those of the Taraco Peninsula, and that these settlement determinants were largely formed by Tiwanaku administrative strategies in the two regions. On the Taraco Peninsula, an existing primate settlement distribution was diluted by simultaneous stagnation of the old center and disproportionate growth of the second-tier sites. In the Pampa Koani, by contrast, the old center (Lukurmata) grew at a tremendous rate, while other sites were either abandoned or grew at a much slower rate.
In fact, it is of the utmost interest that the sites in the easternmost portion of the Taraco Peninsula seem to have grown at a much slower rate than those to the west (Table 8.3). This is the precise opposite of the situation noted for the LF1 (see Chapter 7). The reader will recall that I presented an argument for locating the boundary of the Taraco Peninsula polity between the sites of Chiripa (T-1) and Chiripa Pata (T-4), since sites to the West of this line seem to have been impacted by the expansion of the Taraco Peninsula capital, while sites to the East seem not to have been so impacted. In fact, this line (around 519000 East) seems to function in exactly the same way in the Tiwanaku Period. That is, sites to the West of the line seem to have grown at a more or less normal rate, as a whole, while sites to the East of the line experienced very slow population growth or decline. I propose that the slower population growth East of the 519000E line reflects a loss of population to the urban center of Lukurmata, and that the normal growth of sites to the West of this line indicates that they did not participate in this process.
|
In other words, I believe I have located the boundary between the
Taraco Peninsula and the Katari Basin administrative areas of the
Tiwanaku state. It is very interesting that this administrative boundary
seems to coincide exactly with a political boundary dating at least
to the LF1, or no less than 700 years prior to the Tiwanaku Period.
This fact serves to emphasize the extreme durability of this type
of boundary. In fact, this boundary is no more than a few kilometers
removed from the modern border between Canton Taraco and Canton Huacullani,
and between the Provincias Ingavi and Los Andes.
This border is also interesting in that it emphasizes the heterogeneity
of Tiwanaku administrative practice, even within the heartland region.
Janusek has emphasized that ``... [t]he Tiwanaku core, considered
here [as] the central three-basin network, was not simply a homogeneous administrative unit of the Tiwanaku
state'' ([Janusek 1994]: 73). Specifically, ``settlement organization
and strategies of agricultural exploitation in the Tiwanaku Valley
were entirely distinct from those in the Katari Basin'' ([Janusek 1994]:
69). On the Taraco Peninsula, Tiwanaku administrative strategies were
entirely distinct from those in either of the other two study areas.
Settlement in the Tiwanaku Valley and the Katari Basin in the Tiwanaku
period was oriented to the construction, maintenance and production
of large-scale raised field agricultural areas (cf. [Kolata 1986,Kolata 1991,Kolata and Ortloff 1996b]).
The Taraco Peninsula, by contrast, was and is entirely unsuited to
large-scale raised field agriculture. In the Taraco Peninsula survey,
only 12.47 ha of raised fields were recorded which were certainly
pre-modern and probably prehistoric. Of these, 10.51 ha (84%) were
located to the east of the 519000E meridian (Table 8.4),
and therefore within what I have argued to be the Pampa Koani administrative
area in the Tiwanaku period. Looked at another way, the density of
raised fields to the east of the 519000E line - within the hypothetical
administrative orbit of Lukurmata - was ten times greater than that
of the area to the west of the line, within the territory of the old
Taraco Peninsula polity. The administrative boundary identified above
does in fact seem to correspond to the boundary of raised large-scale
field agricultural production.
|
