Producing legibility through ritual: The Inka expansion in Huarochirí (Lima,Peru)

Introduction

Between the 15th and 16th centuries, the Inka empire, or “Tawantinsuyu,” extended over 4000 km of western South America, occupying territories in the modern countries of Peru, Bolivia, Chile, Argentina, Ecuador, and Colombia (D’Altroy, 2015). One of the most impressive characteristics of the Empire was its rapid expansion while incorporating a multitude of different ethnic groups. Material correlates of imperial expansion (e.g., formalized roads linking a network of administrative sites, large storage complexes, standardized administrative buildings implanted throughout different indigenous settlements) reflect a set of regulatory and administrative practices geared towards the construction of a unified government. However, in the Central Andes, the Inka were the culmination of an ongoing experiment with statecraft that actively mediated imperial institutions with local beliefs, cultural practices, and material culture.

In this article, I elaborate on a framework for the analysis of how state institutions became materialized in regional settings through the mediation of local material culture and practices. I build upon and expand the concept of legibility popularized in James Scott’s (1998) book, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. I propose that this model is relevant for the study of ancient states because it holds the potential to investigate not only top-down state policies but also bottom-up forms of resistance, as well as their points of interaction. According to Scott, legibility from the state is an overarching and simplifying view that erases diversity and variability among its subjects in order to establish successful policies. However, subjected polities also aim to make the state legible. Consequently, I contend that familiarity with cultural norms between conqueror and conquered, their negotiation and adaptation, are a central aspect in the process of building legibility.

I illustrate this point through the discussion of the Inka engagement with local ritual sites in Huarochirí (Lima, Peru). My excavations in the site of Canchaje demonstrate that the Inka actively sought to attach their public infrastructure (plazas) to local ritual emplacements (rock outcrops). By doing so, the Inka built upon existing commonalities in Andean understandings of the power of place in the creation of social relationships among communities. Conversely, by grafting themselves into local lore, Inka imperialism created spaces—unexpected or not—for the survival of local practices. I argue that the use of familiar cultural practices by the Inka to control their expanding empire also created the social spaces for local polities to maintain, formalize, and, at times, expand their own cultural practices and traditions. Through this analysis, I present a framework for exploring the role of local traditions within state-sanctioned spaces of political authority.

Legibility as a framework for state and local interaction

As initially defined, the concept of legibility highlights the need for states and empires to have a simplified view of their subjects; in other words, through standardization of local diversity, state policies could be uniformly imposed. The narrower the view, the more in-depth knowledge states have (Scott, 1998: 11). Scott’s work poses a paradox. If states engaged with the diverse know-how of the local groups under their rule, the overload of information would make it impossible for their bureaucracy to develop an informed policy applicable throughout its territory. As a consequence, modern states develop and impose an ideology of standardization, geared towards the imposition of specific worldviews, and ultimately, gain political and economic control. However, in doing so, states ignore how local societies organize themselves (in Scott’s text, “mētis” or “practical knowledge”). While this is a necessary transaction, it also limits the capacity of the state to act upon potential crises that may arise at a local level (i.e., managing droughts, conflicts over lands) through the mechanisms already in place, which makes the state’s organization and these schemes susceptible to failure.

Following this model, states are fragile contingencies always on the verge of collapse because of their intrinsic mandates to simplify what they do not fully understand (Scott, 2017). Scott’s model received criticism for its tendency to portray states as following similar patterns and having similar imperatives. More pointedly for this essay, his portrayal of states as intrinsically different from the civilizations from whence they rose leaves little common ground connecting standardization and mētis (for further discussion see: Dennis, 1998; Magnusson, 2011; Scholtz, 2010; Wong, 1999; Woost, 2000). Scott (1998: 2) intentionally excludes premodern states from his analysis, since he considers that they lacked the knowledge of their subjects to implement successful interventions in conquered territories. My analysis not only refutes this interpretation but shows that ancient states could potentially have tools for understanding their subjects that were partly unavailable in modern forms of colonialism. Not considered in Scott’s discussion is whether the different degrees of cultural familiarity between conquerors and subjects mediate standardization and thus impact the construction of legibility. From an archaeological standpoint, this requires investigating whether material evidence that predated the state’s imposition was repurposed within state-sanctioned institutions, and conversely, how the state’s material culture was used in local contexts.

Building from this definition, I argue that Scott’s concept can be expanded to include premodern states and a broader understanding of standardization by questioning the local roots of provincial imperial institutions. Through research in Huarochirí, I contend that some communities conquered by the Inka Empire managed to inform state policies through the negotiation of local ritual spaces and practices into the state’s institutional ones. In other words, familiarity facilitates legibility, creating a middle ground between top-down mandates and bottom-up agency through a mutually legible negotiation—and even resignification—of existing practices and materiality. The concept’s potential for bringing into the same analytical scope the different goals and actions of states and subjected communities makes it a valuable tool for the archaeological study of states’ expansion.

In order to illustrate how the incorporation of premodern states expands the concept, I discuss Scott’s (1998: 4) four coexisting elements in the collapse of state plans geared towards simplification through the lens of Inka imperial policies. Archaeological research shows that while the same variables existed within ancient states, collapse was hindered by building upon familiarity. Scott’s first element is an administrative ordering of nature and society through state transformations. The Inka produced ordered landscapes through the carving of rock outcrops, interconnecting them through official narratives of mountain-deities (apus and wak'as) and fictive kinship. By carving the natural landscape across the empire, the Inka built upon a shared ritual system that interconnected all the groups under their political control (see: Bray, 2014; Christie, 2016; Dean, 2010; Meddens et al., 2014). The principles of ancestrality and belonging inscribed in the natural landscape had a political connotation of unity (see: Salomon, 1998). Local landmarks and understandings of place were incorporated and elevated into imperial ideology, thus creating a flexible view of the empire that was both global and local (Acuto, 2005; Astuhuamán, 2008; van de Guchte, 1999). In other words, shared understandings of sacred landscapes rather than state control could lead to order.

The second element is a “high-modernist” ideology. Scott (1998: 93) explicitly states that “First and foremost, high modernism implies a truly radical break with history and tradition” by embracing new “scientific” ideas and forms of knowledge. However, there is no obligation for a drastic break with traditional knowledge. This point was argued by Valverde (2011), who demonstrated that in recent cases of North American zoning practices, the high modernist ideology of “governing space” rests heavily on leaving room for “exemptions” of local practices to avoid contestation. That is, existing practices can become “new” when they are attached to new meanings produced from either the top or the bottom (Thomas, 1992). Most ancient empires built upon existing technologies; they learned what they could and should from the communities they conquered and then reshaped such technologies within their schemes (Berdan et al., 1996; Kolata, 2013). While empires may create radical breaks when facing military resistance, most evidence of Inka imperialism shows the appropriation of previous cultural forms. Chase (2018: 521) argues this point through his model of “local-imperial landscapes” as “ever-shifting combinations of elements that were geographically local and pre-Inca, and those introduced or imposed by the Inca state, united through the invention of tradition.” Conklin’s (1982) work on Wari quipus and their later appropriation by the Inka is another example. The ideal of scientific advancement can be built upon previous knowledge incorporated into state discourse.

The third element is the states’ willingness to use coercive power to bring their worldviews into new territories. The threat of coercion consistently marks ancient states’ expansion (Sinopoli, 1994). Societies such as the Aztec or the Inka identified their leaders as deities in a high degree because of their military prowess (Berdan, 2014; Burger et al., 2007). While armies were not necessarily a standing institution, early empires could create significant military strike forces when needed. In the case of the Inka, we know they limited warfare to cases where negotiation and alliance were impossible (Murra et al., 1986). The Inka relied on their allies’ cooperation to extinguish rebellions (Stern, 1982). Traditionally, Andean communities built cooperation through kinship, which then created social obligations such as reciprocity. Consequently, even the military prowess of the Inka was reliant on engaging with familiar forms of interaction and alliance-building with other Andean communities.

Finally, the fourth element is a dormant civil society that would not contest the state’s standardization plan. Research on the Inka expansion and the early colonial period demonstrates that indigenous communities were apt in negotiating indigenous symbols and practices when new political orders were established (see: Stein, 2005; Thomas, 1992; Wernke, 2010). Scott (2009) himself argues that subjected people learn to re-establish their own identities to their benefit within imperial institutions. Some groups within the Inka Empire may have considered imperial imposition as bearable, or as less taxing, than contestation. I argue that subjected societies in the ancient Andes, rather than being dormant, aptly used common cultural practices as a means to keep their agency in the construction and maintenance of new political systems. In this essay, I investigate this element through the redefinition of community-building practices. By the same token, the Inka may have used these common practices as part of their official policy.

In reviewing these four points, I show that ancient empires like the Inka may have grafted their demands for political transformation upon local forms. As a consequence, recent scholarship on the Inka Empire shows potential “blind spots” from the imperial view and the incorporation of “unexpected consequences” into the system (see: Garrido, 2016; Wilkinson, 2019). Legibility in the Inka Empire was a negotiation between the need for standardization and the acceptance that local forms were better suited to mediate the interaction between the state and its subjects. Even if building legibility was a state’s tool of political subjugation, it also may have reinforced shared idioms in ways that were leveraged by local communities for their own interests (cf. Gosden, 2004). Legibility may be the ultimate goal of the state. However, the process of producing legibility may reinforce local practices, new interpretations of cultural forms, or the outright appropriation of state-sponsored institutions through the filter of local know-how. Therefore, I argue that legibility as a conceptual tool enables our understanding of coexisting processes of interaction and political subjugation within ancient states in general and the Inka Empire in particular.

The Inka in Huarochirí

Huarochirí before the Inka

Huarochirí is located in the easternmost highlands of the modern province of Lima (Figure 1). According to historical sources, Huarochirí was once part of the Yauyos province, a name taken from one of the many dispersed groups inhabiting the region and generalized after the Inka conquest to create a single tributary unit (Dávila Briceño, 1965 [1586]). Huarochirí is the land of origin of the most singular piece of colonial-period text preserved for the Andes, the Huarochirí Manuscript (hereafter HM). The HM was compiled in 1608 under the auspices of a Catholic curate whose indigenous assistants set out to conduct interviews with town elders and collected the histories and myths of the people called Yauyos. Designed as a tool of colonial conversion, the HM is a unique example of history-making, as the indigenous assistants sought to create a formal history for their people on par with that of the Spanish (Salomon, 2008). The text also provides a unique insight into the local organization before the Inka.

OPEN IN VIEWER

Figure 1.

Location of Huarochirí and Canchaje. The waranqas represent the different communities that composed the people of Huarochirí.

Through a close reading of the HM, I have argued that the Yauyos people were not politically centralized, but instead loosely tied by principles of kinship, community, and sacred ancestry (Hernández Garavito, 2019). The central motif of the HM is the birth, history, and veneration of the snow-capped mountain Pariacaca who was the mythical ancestor of the people of Huarochirí. Mountain deities in the Andes are collectively known as wak'as, sacred beings that also are sacred places and agents in a community (Bray, 2014). It was through Pariacaca’s veneration and degree of kinship that the communities organized themselves and determined hierarchies among one another. The HM also notes that by the time of the Inka, rituals mostly moved to mountains closer to each community’s (ayllu’s) territory. This shift corresponds with my field observations that rock outcrops were consistently the central element of both residential and public sites. Chase (2014) has developed the idea that the outcrops represented material manifestations of potential “children” of Pariacaca as mnemonic elements through the natural landscape. I interpreted the outcrops as avatars of Pariacaca through which the ties that brought communities together could be experienced in everyday rituals (Hernández Garavito and Osores Mendives, 2019).

Legibility, rock outcrops, and plazas: Excavations in Canchaje

The site of Canchaje overlooks one of the quebradas through the course of the upper Lurín River valley, in the district of Lahuaytambo, Huarochirí. Canchaje is approximately 3 hectares in size and rises on a small mountaintop accessible by a 30-minute walk from the contemporary town. Two sectors comprise the site: to the north, the architectural features associated with a rock outcrop, and to the south, a large structure dominated by two plazas (Figure 2). AMS radiocarbon dating of organic remains confirmed that the activity areas associated with the rock outcrop were utilized previous to the Inka incorporation, and the plazas built afterwards. Plazas were canonical spaces for the performance of Inka provincial policies (Morris, 2013). It is in plazas that the Inka established social relationships of reciprocity with subjected groups through feasting (Bray, 2003; Makowski et al., 2012) and the selective display and gift-giving of Inka-style artifacts (Bray, 2018; Moore, 1996).

OPEN IN VIEWER

Figure 2.

Map of Canchaje. Excavation units indicated with fuchsia.

Through my analysis of the HM, I contend that Canchaje was a local-level wak'a used by one of the Huarochirí aggregated communities or waranqas, which contained several smaller communities within, for the performance of Pariacaca’s festivities. According to the HM, “The residents of Santa Ana, those who live in San Juan, and all the ones called Chauca Rimac, during Paria Caca’s season, reportedly worship from the mountain called Acu Sica, the one we descend on our way to the Apar Huayqui River” (Salomon and Urioste, 1991: 74–75). Following the topographic references, I suggest that Canchaje corresponds to Acu Sica’s location and shows the necessary infrastructure to have served as one of Pariacaca’s regional veneration centers (Hernández Garavito, 2019). In this section, I present material evidence of how the wak'a body and the Inka plazas exemplify negotiation through cultural familiarity. I discuss how the process of abandonment of the rock outcrop and the formal characteristics of the plazas demonstrate how the Inka grafted state-sponsored rituals upon the veneration practices directly associated with Pariacaca.

Excavations in the Rock Outcrop

The Rock Outcrop measures 56 × 62 m, and it incorporates a series of retainer walls and rooms on its peak. Horizontal steps lead to a poorly defined small plaza. Excavations in the outcrop sector confirmed its use as the location of specific rituals described in the HM as part of Pariacaca’s veneration (Figure 3). Three AMS radiocarbon dates position the use period of the outcrop between 1287 and 1404 CE (calibrated with SHCal 13 atmospheric curve, 95% confidence), with a single date between 1420 and 1450 CE. Overall, the primary use of the outcrop seems to be before the Inka arrival to Huarochirí.

OPEN IN VIEWER

Figure 3.

Excavations in the rock outcrop. (a) Location of the excavation units. (b) Drawing of Room 2. (c) Detail of the mochadero in Room 1. (d) Photograph of Trench 1.

We placed Trench 1 (6 × 3.5 m) in the front of the outcrop, where we traced stone alignments, defining broad steps. To the west of the wall, we found a semi-subterranean circular structure with a diameter of 50 cm. We recorded similar structures atop the outcrop, which, as I argue later, had a funerary function. To the east of the wall, we found a small semi-rectangular structure with an approximate length of 80 cm per side. This section was built on top of small platforms that constitute the façade to the outcrop. Once we reached the floor, we found evidence of small burning events.

We placed Trench 2 (7 × 3 m) to encompass both sides of the apex of the outcrop, characterized by large standing boulders. From the surface, we observed at least three rock alignments suggesting broad steps heading towards the boulders from the north (the opposing and harsher face of the outcrop). After cleaning the vegetation and loose rocks, we found a copper tupu or pin commonly associated with female shawls. Although the lack of good association for the artifact limits interpretation, we did not find similar artifacts outside of the outcrop. The surface was poorly preserved, with gravel perforating the soil matrix and broken in some sections. Under this level, we found another broken grayish surface, possibly showing repairment. The steps were well-defined and carefully built; unlike those in the façade in Trench 1, the alignments created small two-face walls joined with mortar for each of the steps leading to the rocks. These walls contained a leveling fill over the bedrock that created the surface for each step. Under the boulders and to the south, we found a small room built against the rock. From this room, we recovered coarse ceramic sherds and evidence of reddish and yellow fine soil that could suggest the removal of organic materials.

The first point of access to the outcrop after climbing from the southern façade was Room 1. The main wall was two-faced, 7 m in length; it was oriented roughly to the northeast, with a rounded corner that shifted into the southeast, which led into four steps leading to the room’s access. The stairway access was delimited by a transversal low-rise wall, defining a platform associated with a large boulder, which was likely a place of offering. This feature, which I interpret as mochadero or place of libation, had a triangular form (approximately 1.40 × 0.80 m) and was carved in the middle, thus creating a recess that looked like a seat where it faced the platform. The room’s floor had many features suggesting intensive use. In the southwest corner, we found a large millstone surrounded by several large slabs. To the north, right below the platform corner, we found several burning events covered by boulders and white ash right by the foot of the stone feature. Under the floor, we found a considerable accumulation of ash, part of a leveling fill. This layer had a high count of artifacts associated, including lithics, ceramic sherds, and animal bones. This fill covered another floor, which, in turn, covered another leveling, this time over the bedrock. Despite its location, the animal bones recorded on the ash had no evidence of burning, which could suggest they were part of a ritual for covering the original surface in order to lay the second floor. Similarly, the other ash lenses had no burnt material and therefore no association with cooking or household activities.

Excavations in Room 2 revealed highly ritualized closing events. We found an arrangement of three levels of circular steps heading out of the room and directly to the base of a large standing stone that looked similar to a huanca, a stone slab considered as a ritual marker or a lithic representation of a wak'a. Several features at the floor level demonstrate intentional and careful abandonment of the room. In the northwestern and southeastern corners, we found distinct features that seem to, in at least one case, have broken the wall and floor to position a corner feature where another smaller huanca was hidden. This latter feature was unusual compared to other artifact caches recovered during the excavation, and I will elaborate on its importance in the next section.

From the surface, we observed several looted semi-subterranean cysts and excavated the best-preserved one, Room 3. Inside the shaft, we identified two post-depositional levels associated with different events of rockfall covering two sections of a floor. The opening of the cyst was defined by a wall that originally surrounded the whole opening and was made with selected small and medium stones, joined through mortar; this wall enclosed most of the opening of the cyst initially. During the excavation, we found remains of human hand and foot bones, demonstrating its funerary function. Funerary events (inferred by the semi-subterranean cysts) were a central association of the outcrops, suggesting that the individuals buried in these cysts likely were lineage heads of the communities that were part of Canchaje’s ritual performances.

Finally, throughout the excavation of the outcrop, we recovered a set of lithic projectile points of different shapes and colors (Figure 4). We did not find evidence of projectile points through the excavations in a nearby domestic settlement, nor did we find evidence of hunting in the faunal remains recovered. Most of the offerings found in the different contexts were of local style and widely available before the Inka period. We did not find Inka-style ceramics directly associated with the use of the outcrop.

OPEN IN VIEWER

Figure 4.

Collection of lithic points recovered during the excavations in the rock outcrop.

Excavations in the plazas

After the Inka conquest of Huarochirí, Canchaje shifts from Acu Sica, the rock outcrop where the peoples from nearby communities enacted rituals honoring Pariacaca, to the current layout of the site. AMS radiocarbon dates suggest the construction of a large structure dominated by two plazas and subsidiary residential space between 1450 and 1626 CE (calibrated with SHCal 13 atmospheric curve, 95% confidence). From the south, access to the structure leads to a trapezoidal plaza with an area of 800 m². The plaza is not closed-off to the south, yet the other three walls are well-defined. Trapezoidal architectural elements are characteristic of Inka spatial configurations throughout the Andes (Gasparini and Margolies, 1980); this is the only plaza we have recorded of its characteristics throughout our survey of the region.

To the east, we registered a rectangular enclosure with large internal patios surrounding rectangular rooms. Our excavations showed that these rooms served as temporary residential spaces, including storage areas for large jars. It was in one such storage room that we found an Inka-style decorated sherd, further confirming its use period. The second plaza is north of the trapezoidal one. This plaza is semi-circular in shape and encompasses an approximate area of 700 m². Three curving walls delimit the plaza to the north. The innermost curve wall was incomplete and built using a distinct pattern: six standing stone pillars sectioned the wall (two were unattached). Each pillar differed; yet, on average, the bases measured between 1 and 1.5 m per side and had a 3 m maximal height. The pillars are unique to both local and Inka-style architecture and have no known architectural reference to date.

My team and I excavated two units and a trench in the plazas (Figure 5). We placed the first excavation unit (3 × 3 m) adjacent to the northern wall of the trapezoidal plaza. Excavation of this unit showed that the use-surface was very close to the modern surface and confirmed cleaning by the limited number of ceramic sherds recovered. Under two compacted surfaces, we found a prepared clay floor that was likely the original surface of the plaza, which covered the bedrock. We found the same pattern in the trench (7 × 3 m), where we also uncovered at least three other surfaces. The low density of materials recovered (less than 50 g of diagnostic sherds and few unidentifiable fragments of faunal remains), considering the succession of floors, was informative. I interpret these findings as evidence of continuous cleaning of the floors.

OPEN IN VIEWER

Figure 5.

Excavations in the plazas. (a) Excavation unit in the trapezoidal plaza. (b) Trench in the trapezoidal plaza. (c) Photograph of Canchaje with location of the excavation units.

Finally, we excavated another unit (3 × 3 m) at the foot of the last pillar of the semi-circular plaza attached to the curve wall (Figure 6). Excavations showed the same pattern as in the trapezoidal plaza, with a succession of four clean floors. However, from this unit, we recovered several ceramic fragments (count of 34 diagnostic sherds) already significant in comparison to the very low association in the trapezoidal plaza. We found sherds that ranged from a decorated style of local brown wares associated with early periods to Spanish-period Loza. The most remarkable of the local-style decorated sherds are those with a mixture of incision, pinched, and digit-pressed linear sequence. We found five fragments with this decoration: two came from the rock outcrop and three from the plazas. The only reference to similar ceramics I found came from the unpublished results from Thomas Patterson’s 1968 expedition up the Mala River. Drawings of similar sherds come from sites dated to the Early Horizon (900–200 BCE) or Early Intermediate Period (200 BCE–600 CE). Building on this comparison, it is possible that these sherds were part of a ceramic tradition that precedes the occupation of the site. The sherds could potentially be stylistic archaisms or an example of a continued ritual style. Future research and systematic excavation in more sites are necessary to confirm the temporality of the style.

OPEN IN VIEWER

Figure 6.

Semi-circular plaza. (a) Ortophotograph of the semi-circular plaza with excavation units. (b) Detail of one stone pillar. (c) Excavation unit with sequence of floors. (d) Photograph of wall and stone pillars.

There are two significant observations from the results of the excavations in the plazas. First, the continued cleaning of floors and the storing and resting spaces associated with them suggest a cyclical use that is significantly different from the results from the rock outcrop, supporting that the Inka-period buildings manifest a different type of ritual intention and practice. And second, the curved wall and stone pillars are an anomaly when compared to other Inka regional plazas (Moore, 1996).

Discussion: Producing legibility in Canchaje

Through the descriptions of Pariacaca’s rituals in the HM, I suggest that the excavation results from Canchaje show specific examples of how mutual legibility between the Inka and the Yauyos was built through ritual spaces. Two descriptions of ritual ceremonies matched the layout and materials recovered from the outcrop. The first ceremony was the Auquisna, the yearly pilgrimage to Pariacaca, which encompassed astronomical observations, ritual dances, uphill races, feasting with ancestors, sacrifices of llamas, and offerings placed at the center of rock outcrops. Races to the mountaintop could explain the broad steps uncovered in Trench 2:

People run a race on their way to this mountain in accordance with the yanca’s instructions, driving their llama bucks. The strongest ones even shoulder small llamas. They scramble upward, each thinking, “I mean to get to summit first!” <margin, in Spanish:> [Which is where <Paria Caca> can be seen from.] The first llama to arrive to the mountain top was much loved by Paria Caca. (Salomon and Urioste, 1991: 72)

The physical spaces of Canchaje are well-suited for the activities described in the text. Surrounding communities might have run towards the site during Pariacaca’s festival, making it to the top. Possibly, people attempted to reach the top of the outcrop through the ravine and not the façade of the plaza following the quasi-circular layout of the steps to the top. The finding of camelid bones in ash lenses and the mochadero in Room 1 could be associated with the offerings made to the wak'a described in the text. The prominence of funerary contexts in the outcrop substantiates the emphasis on feasting with the ancestors in the HM. Ceremonies in the rock outcrops strengthen community bonds among themselves and with their tutelary deity. The performances created spaces for different peoples to come together and reaffirmed their belonging to a broader community.

The second ceremony was the Machua, a ceremony for a specific community and likely celebrated in the outcrop avatars of Pariacaca. This ceremony also encompassed dancing and music, as well as similar journeys with llamas to mountain tops. A distinctive feature of the Machua involved a spear-throwing contest:

Braiding some straw called chupa, as they say, and tying together a lot of wooden slats, they’d bind two effigy bundles around with the straw. The height of these bundles was seven and a half armlengths. (…) As soon as they erected the two chutas they began to throw spears. They say that while the people threw spears, while they entered into competition hurling spears ayllu by ayllu, the women would dance without drums and chant these words: “Receive your poor forlorn children!” (Salomon and Urioste, 1991: 121–122)

This description could explain the unexpected presence of projectile points in the outcrop. The altitude of Canchaje limits its potential as a hunting ground, and the evidence for the consumption of non-domesticated camelids or deer is minimal. Furthermore, there is no evidence of defensive infrastructure or artifacts in the outcrop. The concentration of the points in the outcrop generally supports their use in similar ceremonies.

Plazas were a centerpiece of Inka provincial sites. Here, Inka administrators would honor local leaders with food, drink, and gifts, creating fictive reciprocity between them that masked the fact that local lords had become dependent on the Inka administration to access luxury goods (D’Altroy et al., 1985, 2000). Archaeological research in wak'as centers on the idea of Inka appropriation of local ritual space in order to materialize their political control and communicate their power to their subjected communities (Acuto, 2005; Acuto et al., 2012; Kosiba, 2012). While Chase’s (2018) work shows Inka reinvention of local ritual landscapes in Huarochirí, my research also suggests the simultaneous transformation of Inka ritual space through local tradition.

In Canchaje, while there is a direct relationship between the closure of the outcrop and the implementation of the plazas, they were never physically disconnected. My excavations recovered direct evidence of Inka-standard feasting in the plazas, yet local practices and artifacts became part of the whole design. While results from the trapezoidal plaza show evidence of a traditional attachment between new and old sacred places in a way that signals the Inka replacing previous ritual practices, excavations in the semi-circular plaza show a more complex interaction. In particular, the presence of the stone pillars suggests an alternative reading.

Sanhueza (2012) recorded similar stone pillars in the Inka road, passing through the desert of Atacama, Chile. She argues these features were markers for symbolic and geographic boundaries on the roads. In her interpretation, the pillars were sayhuas or markers expressively built by the Inka surveyors (Guaman Poma de Ayala, 1993 [1610?]: Drawing 139). However, I know of no mention in either archaeological research or Spanish-period chronicles of the Inka Empire of pillars joined as a wall. I argue that a possible explanation also comes from the HM, which describes the inaugural ritual of Pariacaca’s festival:

To first become a huacsa, people in fact perform a certain ritual. It’s like this: a man of the Caca Sica ayllu functions as officiant for these ceremonies. From early times these officiants were only one or two people, and, as for their title, it was yanca. <margin, in Spanish:> [The master is called yanca.] The same title is used in all the villages. This man observes the course of the sun <margin, in Spanish:> [That is, from the shadow that the wall casts in the sun.] from a wall constructed with perfect alignment. When the rays of the sun touch this calibrated wall, he [proclaimed] to the people, ‘Now we must go’; or if they don’t, he’d say, ‘Tomorrow is the time.’ Following this command, people go to Paria Caca in order to worship. (Salomon and Urioste, 1991: 72)

From the argument above, I contend that we can see how specific aspects of Pariacaca’s worship became part of the most sanctified spaces of Inka provincial politics. In doing so, plazas became spaces for the strengthening of macro-communal bonds among the children of Pariacaca.

Moreover, the plaza ratified an idealized narrative of the Inka as kin with the people of Huarochirí and subjected to their ancestral deity. The formal similarities between the pillars from the HM and the sayhuas from Guaman Poma’s account further support the idea of the negotiation of the meaning and function of these material markers of state and local ritual, and their potential capacity to serve as a middle ground between the state and their subjects. Furthermore, the finding of early-style local ceramics associated with the pillars suggests that the wall may have predated the plazas and be contemporary with the use of the outcrop. The sequence of abutting walls within the layout of the plazas supports this reading. Incorporation of the wall into Inka plazas could be read two ways, as the Inka erasing the specific function of the pillars, or as the Inka reshaping rather than erasing a feature that was central to Pariacaca’s veneration. The fact that the ritual was performed early into the colonial period makes me more partial to the idea of incorporation rather than erasure.

Conclusions

I argued that the concept of legibility holds the potential to incorporate both the top-down state mandate of standardization and the bottom-up demand for agency and the maintenance of tradition. While Scott’s analysis excludes premodern states, I contend that their inclusion provides an example of how to avoid the “catastrophic” events that are the consequence of state projects lacking understanding of local forms. The excavations of Canchaje show how the Inka selectively incorporated specific aspects of local ideology within the central scenarios of provincial politics, the plazas. At the same time, local traditions were negotiated and expanded through this incorporation. In other words, mutual legibility was achieved through the attempts between a state that aims to avoid “violent imposition” and an active rather than dormant subjected society. As a consequence, a new world order did not need a break with previous traditional knowledge, but rather its incorporation into state ideology.

In Canchaje, architectural integration of ritual spaces suggests that the Inka and the people of Huarochirí successfully produced mutual legibility based on the reinvention of sacred spaces, allowing for their continued use as places for building communal solidarities. For the Inka, it was beneficial that there was already a local system through which people came together, and that could be incorporated into state practices. For the people of Huarochirí, the subjugation of their ritual practices within Inka standard plazas was offset by the opportunity to preserve local traditions. Specific forms that could be translated from one worldview to others, such as the pillars, were potentially material spaces for building mutual legibility.

By expanding the concept of legibility to consider that familiarity with cultural norms between conqueror and conquered may have played a significant role in standardization, the application of this framework puts at the center the need to consider both top-down administrative mandates and the ability of bottom-up practices to reinvent themselves and affect colonial systems. Ancient states, in particular, needed to level the demand for expansion and taxation with their ability to control a diverse population and territory. Building upon what is familiar and minimizing the need to engage in outright coercion may have curtailed the state’s need for absolute control. At the same time, subjected groups could use the states’ needs as a means to develop new ways of political and social agency. I present the idea of producing legibility as an ongoing process through which ancient states, like the Inka, could build upon familiar cultural practices with their subjects to facilitate their political control. In doing so, my research shows how local agency is both required and enabled. For archaeology, this means focusing on tradition on par with disruption. Rather than a disregard for local practical knowledge, the Inka Empire built legibility upon shared cultural forms that had direct correlates in ritual architecture.