A Reassessment of Collagua and Provincial Inka Ceramic Styles of Arequipa, Peru
Steve Wernke, University of Wisconsin-Madison
Paper presented at the 66th Annual Meeting of the Society for American Archaeology
April 18-22, 2001, New Orleans, Louisiana
during the symposium, "Recent Research on Local and Supra-Local Trajectories and Intersections in the South-Central Andes", Steve Wernke and Michael Malpass, co-organizers; Michael Malpass, chair.
Many figures below are in Adobe Acrobat .pdf format. If you don't already have it, Acrobat Reader can be downloaded free here.

Explication of local ceramic styles remains a critical research frontier in many Pacific slope drainages of southwestern Peru. This paper presents a sequence of the Collagua style of the Colca Valley and compares it with Chuquibamba and other styles in the region, while exploring its relationship to subsequent provincial Late Horizon ceramics. Analysis of data from full-coverage survey and documentary sources from the nuclear area of the late prehispanic Collagua ethnic señorío provides a preliminary view of Inka imperial control of local ceramic production, while the possible significance of the macro-regional distribution of these provincial Inka styles is also discussed.

Introduction. In the valleys of Arequipa of southern Peru, as in other areas of the Andes, the expanding Inka empire confronted a balkanized political landscape composed of autonomous polities that filled the power void left by the disintegration of Tiwanaku and Wari. This region has only recently attracted enough scholarly attention to begin delineating how this ethnic mosaic was reified or redefined in the process of Inka imperial expansion and consolidation. Household and regional-scale research in Moquegua and other areas makes good headway on this problem by illustrating the varied changes in domestic and political economy behind the rapid expansion of the Inka empire (Bürgi 1993, Covey 2000, D'Altroy 1992, Stanish 1989, Stanish 1992, Van Buren 1993, inter alia). The research I report on here today complements this work by providing a preliminary view of the local changes that occur under Inka expansion at the interface between domestic and political economy.

In this paper, I begin to define such community-scale changes in subject-state relationships within one of the regional ethnic polities of Arequipa: the Collaguas of the Colca valley. First I provide a brief overview of settlement patterning during the autonomous Collagua occupation during the Late Intermediate Period and the subsequent Inka occupation of the valley, using data recovered from my survey of the central Colca valley and surrounding uplands. Second, I present a preliminary sequence of the local Collagua ceramic style using artifacts recovered from the survey. Finally, I analyze contextual and documentary data that together provide insight into how the production of these local ceramics was organized. These data, I suggest, yield a view of craft specialist households (to use John Janusek's term) embedded within hybrid state and local political structures, in turn providing a glimpse of the refractory effects of imperial expansion at the community-scale interface between domestic and political economy (Janusek 1999).
Field Methodology. Community-scale studies generally use a mixed regional and intensive survey strategy, allowing for greater spatial coverage than site-specific intensive survey or excavation, while providing higher resolution field data than traditional regional survey methods (Gaffney and Gaffney 1988, Kolb and Snead 1997, Yaeger and Canuto 2000). The archaeological portion of my dissertation field study employed a mixed-strategy of regional and site-specific survey in the nuclear area of the Collagua ethnic polity. I surveyed 100 square kilometers with archaeologists Willy Yépez and Erika Simborth in the Colca valley and surrounding puna (dry steppe) (Figs. 1, 2-3). We recorded 161 archaeological sites with 248 field-dated occupational components during the survey. The sites range from preceramic lithic scatters, rockshelters and open air sites, through large late prehispanic and early colonial habitational sites with well-preserved architectural remains. We mapped and described all sites using standardized field forms, and collected a representative sample of surface artifacts. We also created large-scale maps and collected more detailed data using additional architectural forms and more systematic collections at selected village sites with occupations spanning the development of the Collagua ethnic polity: the LIP, Late Horizon, and Early Colonial (AD 1532-1570) periods.
LIP and Late Horizon Period settlement patterns. The survey data indicate that during the Late Intermediate Period, habitational sites of three to seven hectares in size developed in two principal contexts: first, on the valley floor among massive agricultural field wall complexes with defensive, storage and mortuary features (Figs. 4-6), and second, on the promontories of the surrounding valley sides in association with the valley's extensive irrigated agricultural terrace complexes (Figs. 7-8). Settlement patterning during the Late Horizon Inka occupation of the valley is characterized by some growth with overall locational and organizational continuity. During the Late Horizon, no site dominates the settlement pattern in terms of size, and all of the major LIP settlements with elite Collagua architecture also have rustic forms of Inka architecture. Neither this heterarchical relationship between sites, nor the elaborate agricultural infrastructure already established during the Late Intermediate Period (Brooks and Olivares Ayala 1998) changed drastically under Inka occupation. However, the appearance of single-occupation Inka sites associated with relict corrals, artificial lagoons and other hydraulic features on the puna slopes above the north side of the valley suggests an intensification of pastoral production by the Inka.

The settlement pattern data (Fig. 9) therefore indicate a strategy of imperial control that is neither clearly "direct" nor "indirect". On the one hand, there was no major change in settlement patterning between the LIP and the Late Horizon, and there is no single site that could be considered a dominant administrative center. On the other hand, the presence of cut stone masonry and large, multi-door kallanka structures diagnostic of Inka architecture (Hyslop 1990), suggest considerable investment in a decentralized, but coordinated, Inka presence in this part of the valley (Fig. 9).
Differentiation of Local LIP and LH styles. Similar to the settlement pattern data, Late Horizon ceramics found in the Colca valley present a mixed picture of Inka administration that does not appear to result from either clearly "independent" or "attached" specialization. I employed both type/variant and data matrix approaches in my analysis of the 3847 ceramic fragments recovered from the survey. The preliminary results of my type/variant analysis reinforce the findings of Sarah Osgood Brooks who differentiated the local Collagua style from Chuquibamba and other regional styles (Brooks 1998). As opposed to, for example, the Churajón style of the valleys of southern Arequipa, both the Chuquibamba and Collagua styles emerge from a common Wari, rather than Tiwanaku-influenced substrate. However, Chuquibamba more directly incorporates some of the common motifs of the Wari Q'osqopa style just outlined in the paper by Willy Yépez and colleagues, as well as coastal influences in the form of specific design elements such as the eight-pointed star. Also, the painted design elements common to both Collagua and Chuquibamba are executed in distinct ways, with Chuquibamba motifs generally executed with thinner lines in a lustrous (and sometimes dull) greyish paint, as opposed to the thicker, more pronounced execution of Collagua designs in heavy black mineral paint.
Apart from differentiating Collagua from Chuquibamba, my type/variant analysis also produced a preliminary sequence that begins to address the perennial problem of differentiating local Late Intermediate and Late Horizon wares. This is a stylistic sequence of four overlapping phases: Collagua I (early LIP)-Collagua II (middle-late LIP)-Collagua III (transitional-LH)-Collagua Inka (Figs. 10, 11). The beginning of the sequence is positioned by reference to radiocarbon dates associated with Wari-influenced ceramics from a domestic terrace at the site of Chijra, excavated by Michael Malpass and Pablo de la Vera Cruz (Malpass and de la Vera Cruz Chávez 1988, Malpass and de la Vera Cruz Chávez 1990). This site was also recorded during my survey. The phases generally parallel the three phase Chuquibamba sequence developed by de la Vera and Augusto Cardona, and are based principally on marked changes in form and decoration (Cardona Rosas 1993, de la Vera Cruz Chávez 1988, de la Vera Cruz Chávez 1989). Today I will focus exclusively on the bowl and plate forms. The general trend over the sequence in terms of form is a change from more globular bowls, perhaps overlapping with the terminal Middle Horizon as noted by Willy Yépez and colleagues, to more open, flat-bottomed bowls and shallow plates of the Late Horizon (Table 1). Accompanying this change is a shift in the placement of decoration from the external to internal surface. Of course, this sequence must be considered provisional in the absence of a more robust suite of radiocarbon dates and stratigraphic excavations.


Everted lip bowls

Cumbrous bowls

Open bowls


 Collagua I

11% (N=10)

50% (N=47)

39% (N=37)

0% (N=0)

100% (N=94)

 Collagua II

 0% (N=0)

12% (N=22)

76% (N=135)

11% (N=20)

100% (N=177)

 Collagua III

 0% (N=0)

1% (N=2)

62% (N=105)

37% (N=63)

100% (N=170)

 Collagua Inka

 0% (N=0)

0% (N=0)

0% (N=2)

100% (N=563)

100% (N=565)

Table 1: Collagua I - Collagua Inka bowl and plate forms (N=1006)
There are three principal Collagua I bowl forms: a cumbrous, slightly restricted form, a more open form with straighter sides, and an open form with a slightly everted lip (Figs. 11, 12, 13-14). The horizontal design fields of Collagua I bowls are executed in black and occasionally white on a red slip, and include curvilinear and geometric designs such as serpentine curves, X's, S's, pendant triangles, as well as chained diamonds and rhomboids with parallel and reticulated line fills. Collagua I bowls also have black ticks at the rim, generally interpreted as a Wari-influenced trait in the region. Collagua II bowls include more open forms with rounded bases, as well as vessels with more vertical sides (Figs. 11, 12). Designs on Collagua II bowls, also executed in black on red, include zoomorphic motifs, mainly representations of birds, and curvilinear designs such as concentric arcs draping from the rim (Figs. 11, 15). Small button-like protuberances below the rim are also common on the exterior of Collagua II bowls (Fig. 16). Collagua III bowls, which appear to be transitional between the LIP and Late Horizon, are open, generally with flatter bottoms, and are decorated with curvilinear designs such as paired wavy lines just below the rim, accompanied by designs at the base of the interior of the plate, a pattern which continues through the Late Horizon (Fig. 11, 17, 18).
The diagnostic material from the Late Horizon Period is dominated by shallow open plates and long-necked amphora (aríbalos), with lesser quantities of cups, beakers (k'eros), and one-handled pitchers typical of Late Horizon assemblages throughout the south-central Andes (Brooks 1998, Bürgi 1993, de la Vera Cruz Chávez 1988, de la Vera Cruz Chávez 1989, Julien 1983, Lumbreras 1974a, Lumbreras 1974b, Malpass and de la Vera Cruz Chávez 1988, Malpass and de la Vera Cruz Chávez 1990, Pärssinen and Siirriäinen 1997, Stanish 1991, Tschopik 1946, Van Buren 1993, Wernke 2001). The plates can be divided into two general categories: black on red bichromes, and polychromes. The few published reports, along with theses and the extensive gray literature from the region report similar plates from the Chuquibamba and Majes/Camaná valleys to the southwest (de la Vera Cruz Chávez 1989, Márquez and Montoro 1990); as well as the Siguas (Yépez Alvarez, pers. comm. 1999), Yura (Linares Delgado 1989), and Chili/Vitor/Quilca drainages to the south (Yépez Alvarez 1991, Cardona Rosas, pers. comm. 1999). Throughout the region, the bichromes are much more common.

The more elaborately decorated polychrome plates have been preliminarily identified as altiplano imports by previous investigators (Linares Delgado 1989, Malpass as cited in Stanish 1991:15). This identification was made in part because they are less common than the bichromes, but also because they share broad similarities in the colors and decorative motifs employed in the Sillustani Polychrome style of the northwestern Titicaca basin, as well as the Pacajes style of southern Titicaca and the highland drainages of extreme southwestern Peru and northern Chile (Julien 1983, Tschopik 1946). However, I argue that these bichromes and polychromes together represent a provincial Inka style that I call Collagua Inka, produced in the Colca valley and other locales throughout the Department of Arequipa.
Amongst the Collagua Inka bichrome plates, the most common design elements are thin concentric lines inside the rim, some with geometric designs or representational motifs on the plate interior (Figs. 11, 19). These plates also show stylistic continuities from external design fields on Collagua style bowls, such as bird motifs, also common to other southern provincial Inka styles (Fig. 20). Other bichrome plates display geometric designs common to the Cuzco Polychrome style, such as dentate patterns and chained diamonds (Fig. 21), as well as representational motifs such as the footplow (or chakitaclla) (Fig. 22), fish, cultigens such as ají, and camelids, all of which also appear in polychrome (Fig. 23) (D'Altroy and Bishop 1990, D'Altroy, et al. 1994, Fernandez Baca 1971, Pärssinen and Siirriäinen 1997, Rowe 1944).

I have recovered direct evidence for local ceramic production in the form of potter's wheels, called muyuchiku, or falso tornos (Hagstrum 1989) (Fig. 24-25). Five of the six fragments of potter's wheels from the survey were recovered from a single domestic terrace within an elite sector of Uyu Uyu, one of the LIP to early colonial period sites selected for more intensive study during my project (Figs. 26-27). This area of the site is composed of a U-shaped patio group around a large central structure with a worked stone façade typical of high status Collagua architecture. The close association between this patio group and the large (29 x 9 m) Inka kallanka dominating the central plaza directly to the east also suggests a high-status residential area (Fig. 26-27). This contextual information suggests that during LIP and/or Late Horizon times, ceramics were produced at large settlements such as Uyu Uyu, perhaps associated with elite households.
Documentary data relating to local ceramic production. While these archaeological data at the scales of artifact, site, and settlement pattern point toward a generally "indirect" but coordinated Inka administrative strategy, my analysis of detailed Spanish censuses (or visitas) of the Colca Valley provides independent evidence for the organization of local ceramic production, in turn rendering a view of one aspect of the community-level mechanics of Inka expansionism. These mechanics involved the reordering of linkages between individual households and the nested hierarchy of kin-groups, or ayllus, which in turn articulated with the state bureaucracy of the Inka empire.

The visitas in question-from 1591, 1604, and 1617-are from an area of the Colca valley that overlaps my archaeological survey. The visitas record the members and landholdings of all households within their ayllus, each in turn headed by a kuraka (leader) and mandón (or segunda persona-second in charge). These ayllus are organized by low, middle, and high status designations within the Hanansaya (upper) and Hurinsaya (lower) ranked moieties of Yanquecollaguas (Fig. 28). On the one hand, this hierarchical tripartite ranking suggests that local kindreds were significantly reworked to mirror the structuring of the royal ayllus (panacas) of Cuzco (Zuidema 1964). On the other hand, the pataca designation of the ayllus, indicating a unit of 100 tributary households within Inka decimal administration (Julien 1988), appears to have been adapted to the extant number of ayllu households. This suggests that these seemingly rigid Inka bureaucratic units also yielded to local organization. Within this structure, the members and landholdings of three craft specialist groups are listed: silversmiths, weavers of sumptuary cumpi cloth, and potters.
For example, in the 1604 visita for the Hurinsaya half of the village of Coporaque (one of the villages in my survey area), pottery specialist households are listed in two ayllus: five tributary households in Collana ayllu (here referred to as "Indian potters"), and six tributary households in Cayao ayllu (here referred to as "official potters"). The fact that these potters are embedded within the ayllu structure I just described, and the listing of the second group as "official" potters when no such colonial period position existed, strongly suggest that they are descendents of Inka period pottery specialists. This arrangement appears to reflect a level of coordination of manufacture beyond casual household production or "independent specialization". But yet it does not appear to represent an attached specialist production regime, because the potters are not listed as retainers (or yanacona)--a category which does occur in the visitas. Furthermore, large-scale nucleated workshop production appears unlikely since these potters represent less than 1% of the tributary population (168 tributaries).

Rather, I suggest that these data, together with the archaeological data I have briefly reviewed here today, illustrate an example of "embedded" craft specialization (Janusek 1999), in which artisans produce neither for a specific patron, as in "attached specialization", nor independently for a general demand population, as in "independent specialization", but rather as part of the broader relations of reciprocity and redistribution that connect hierarchies of ayllus.

In this paper, I have briefly sketched my research that combines archaeological and documentary data in order to better understand the community level politics of Inka expansionism within the Colca valley. These combined data provide a more detailed view of the cultural context of Inka rule than would be possible with either data set alone. Ongoing research will continue to integrate these data sets to further clarify the workings of Inka expansionist politics at this critical community-level interface between domestic and political economy.

Warm thanks to the Wenner-Gren Foundation for Anthropological Research for supporting the research presented here through a dissertation research grant and a Lita Osmundson fellowship. Logistical support from the Center for Archaeological Research in Arequipa (CIARQ) also greatly facilitated my field research.

References Cited

Brooks, S. and M. Olivares Ayala
1998 The pre-Inca Huarancante canal, Colca Valley, Peru. Paper presented at the 26th Annual Midwest Conference on Andean and Amazonian Archaeology, Anthropology, and Ethnohistory, University of Illinois, Champaign-Urbana.

Brooks, S. O.
1998 Prehistoric agricultural terraces in the río Japo basin, Colca Valley, Peru. PhD Dissertation, University of Wisconsin-Madison.

Bürgi, P. T.
1993 The Inka empire's expansion into the coastal sierra region west of Lake Titicaca. PhD Dissertation, University of Chicago.

Cardona Rosas, A.
1993 Características geográficas del patrón de asentamiento para el valle de Chuquibamba - Arequipa, durante el período comprendido entre el Horizonte Medio, y el Horizonte Tardío. Licenciado en Arqueología, Universidad Católica Santa María.

Cock, G.
1977-78 Los kurakas de los Collaguas: poder político y poder económico. Historia y cultura (10):95-118.

Covey, R. A.
2000 Inka administration of the far south coast of Peru. Latin american antiquity 11(2):119-138.

D'Altroy, T. N.
1992 Provincial power in the Inka empire. Smithsonian Institution Press, Washington.

D'Altroy, T. N. and R. L. Bishop
1990 The provincial organization of Inka ceramic production. American antiquity 55(1):120-138.

D'Altroy, T. N., A. M. Lorandi and V. Williams
1994 Producción y uso de cerámica en la economía política Inka. In Tecnología y organización de la producción cerámica prehispánica en los Andes, edited by I. Shimada, pp. 395-441. Pontifica Universidad Católica del Perú Fondo Editorial 1994, Lima.

de la Vera Cruz Chávez, P.
1988 Estudio arqueológico en el Valle de Cabanaconde, Arequipa. BA Thesis, Universidad Católica Santa María.
de la Vera Cruz Chávez, P.
1989 Cronología y corología de la cuenca del río Camaná - Majes - Colca - Arequipa. Licentiature Thesis, Universidad Católica Santa María.

Espinoza Soriano, W.
1987 Migraciones internas en el reino Colla. Tejedores, plumeros, y alfareros del estado imperial Inca. Chungará 19:243-289.

Fernandez Baca, C.
1971 Motivos de ornamentación de la cerámica Inca-Cuzco, Tomo I. Libreria Studium, S.A., Lima.

Gaffney, C. F. and V. L. Gaffney
1988 Some quantitative approaches to site territory and land use from the surface record. In Conceptual issues in environmental archaeology, edited by J. L. Bintliff, D. A. Davidson and E. G. Grant, pp. 82-90. Edinburgh University Press, Edinburgh.

Hagstrum, M.
1989 Technological continuity and change: Ceramic ethnoarchaeology in the Peruvian Andes. PhD dissertation, University of California, Los Angeles.

Helmer, M.
1955-1956 La visitación de los Chupachos: Inca et encomendero. Travaux de l'Institut Français d'Etudes Andines 5:3-50.

Janusek, J. W.
1999 Craft and local power: Embedded specialization in Tiwanaku cities. Latin american antiquity 10(2):107-131.

Julien, C.
1993 Finding a fit: Archaeology and ethnohistory of the Incas. In Provincial Inca, edited by M. Malpass, pp. 177-233. University of Iowa Press, Iowa City.

Julien, C. J.
1983 Hatunqolla, a view of Inca rule from the Lake Titicaca region. University of California publications in anthropology ; v. 15. University of California Press, Berkeley.

Julien, C. J.
1988 How Inca Decimal Administration Worked. Ethnohistory 35(3):257-277.

Kolb, M. J. and J. E. Snead
1997 It's a small world after all: Comparative analyses of community organization in archaeology. American antiquity 62(4):609-628.

LeVine, T. Y.
1987 Inka labor service at the regional level: The functional reality. Ethnohistory 34:14-46.

Linares Delgado, L. R.
1989 Cronología y relaciones culturales pre-hispánicas del Valle del Chili - Arequipa. Licenciado, Universidad Católica Santa María.

Lumbreras, L. G.
1974a Los reinos post-Tiwanaku en el area altiplanica. Revista del museo nacional 40:55-85.

Lumbreras, L. G.
1974b The peoples and cultures of ancient Peru. Smithsonian Institution Press, Washington, DC.

Malpass, M. A. and P. de la Vera Cruz Chávez
1988 Ceramic Sequence from Chijra, Colca Valley, Peru. In The cultural ecology, archaeology, and history of terracing and terrace abandonment in the Colca valley of southern Peru. Technical report to the National Science Foundation and the National Geographic Society, vol. 2, edited by W. Denevan, pp. 204-233. Department of Geography, Madison.

Malpass, M. A. and P. de la Vera Cruz Chávez
1990 Cronología y secuencia de la cerámica de Chijra, valle del Colca. Gaceta Arqueológica Andina 18/19:41-57.

Márquez, M. G. and R. B. Montoro
1990 Arqueología del valle de Majes. Gaceta arqueológica andina V(18/19):25-40.

Murra, J. V.
1978 Los olleros del Inka: Hacia una historia y arqueología del Qollasuyu. In Historia, problema, y promesa: Homenaje a Jorge Basadre, edited by F. Miro Quesada, F. Pease G.Y. and D. Sobrevilla, pp. 415-423. Pontificia Universidad Católica del Perú, Lima.

Pärssinen, M. and Siirriäinen
1997 Inka-style ceramics and their chronological relationship to the Inka expansion in the southern Lake Titicaca area (Bolivia). Latin american antiquity 8(3):255-271.

Rowe, J. H.
1944 An introduction to the archaeology of Cuzco. The Museum, Cambridge, Mass.

Stanish, C.
1989 An archaeological evaluation of an ethnohistorical model. In Ecology, settlement, and history in the Osmore drainage, edited by D. Rice, C. Stanish and P. Scarr, pp. 303-320. vol. 545. British Archaeological Reports, International Series, Oxford.

Stanish, C.
1991 A late pre-hispanic ceramic chronology for the upper Moquegua valley, Peru. Fieldiana 16 (New Series). Field Museum of Natural History, Chicago.

Stanish, C.
1992 Ancient Andean political economy. University of Texas Press, Austin.

Tschopik, M. H.
1946 Some notes on the archaeology of the department of Puno, Peru. Papers of the Peabody Museum of American Archaeology and Ethnology XXVII, no. 3. Peabody Museum of American Archaeology and Ethnology, Cambridge, MA.

Van Buren, M.
1993 Community and empire in southern Peru: The site of Torata Alta under Spanish rule. PhD dissertation, University of Arizona.

Wernke, S.
2001 (In prep.) El Proyecto Prospección Regional Valle del Colca. Report to the Instituto Nacional de Cultura del Peru, Lima, Peru.

Yaeger, J. and M. A. Canuto
2000 Introducing an archaeology of communities. In The archaeology of communities, edited by M. A. Canuto and J. Yaeger, pp. 1-15. Routledge, New York.

Yépez Alvarez, W.
1991 Estudio de la cerámica de las fases Churajón Tardío y Chuquibamba Tardío en el sitio arqueológico de Umacollo, valle del Chili Arequipa. Universidad Católica de Santa María. Submitted to Technical report.

Zuidema, R. T.
1964 The ceque system of Cuzco : The social organization of the capital of the Inca. E.J. Brill, Leiden.