Ritual Geography, Settlement Patterns and the Characterization of the Provinces of the Inka

Heartland
Author(s): I. S. Farrington
Source: World Archaeology, Vol. 23, No. 3, Archaeology of Empires (Feb., 1992), pp. 368-385
Published by: Taylor & Francis, Ltd.
Stable URL: http://www.jstor.org/stable/124769 .
Accessed: 21/10/2011 03:40

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to World
Archaeology.

http://www.jstor.org
Ritual geography, settlement
patterns and the characterization
of the provinces of the Inka
heartland

. S. Farrington

Introduction

The settlement archaeology of the Inka heartland (within c. 70 km of the capital, Cuzco) is
extremely varied, ranging from areas where there is virtually no evidence, to landscapes
which have been substantially modified. It is insufficient to explain such variability solely
with reference to the agricultural productivity of a district, because adjacent areas often
display quite different patterns of Inka activity. In order to understand such differences
more fully, it is perhaps more appropriate to examine the underlying cultural patterning,
to view the landscape as sacred for the enactment of its rituals and legends for the cultural
establishment of its power and authority. A state, such as the Inka, had the opportunity to
impose at a broader scale its own perceptions of a ritual landscape which justify and
legitimate its own gods, history, social order and authority. It was empowered to transform
that landscape by moving people around within it and, indeed, by moving around the land
itself in the form of terraces, river and spring modifications, and rock carvings to create the
required benign settings for legends to be told and rituals enacted, that the well-being of
the empire might thus be preserved.
In Cuzco, the rich ethnographies and legendary histories recorded by the Spanish
chroniclers from Inka oral tradition have largely been ignored by archaeologists, save for
the wringing out of them dynasties and chronologies which may serve as explanations for
the stylistic variation of both pottery (Rowe 1944) and architecture (Kendall 1985). The
literal translation of one account into archaeological meaning is notoriously difficult,
particularly when there are several, often conflicting, stories of the same event or
phenomenon. It is perhaps more prudent to consider the range of documentation available
and to develop modes of reasoning which might serve to explain the variation of
archaeological settlement patterning, isolating some districts for more detailed investi-
gation. There has been a wealth of anthropological enquiry, mainly structuralist in nature,
which has examined the Chronicles in order to describe the social, political and economic
organization of Cuzco and its valley and to evaluate the role that mytho-history and ritual
practices played in these characterizations (e.g. Zuidema 1964; 1990; Sherbondy 1982;
Urton 1990). Such interpretations do indeed give an insight into Inka perceptions of their
landscape - for example, there was an underlying trait of mountain top (apu) worship
throughout the Andes (Reinhard 1985; Heffernan 1988; Sallnow 1987). Validation of Inka

World Archaeology Volume 23 No. 3 Archaeology of Empires

? Routledge 1992 0043-8243/92/2303/368 $3.00/1
 The characterization of the provinces of the Inka heartland 369

rule came about through the acceptance and performance of commonly understood myths
and ceremony.

The physical geography of the Cuzco region

The imperial capital of Cuzco is situated at 3,450 m above sea level at the head of the
Huatanay Valley on an alluvial fan formed by its tributaries emerging from steep gorges
which surround it. To the north and north-west, the upper catchment comprises a rolling
limestone plateau, with many springs and deep ravines, stretching up to the limits of
cultivation around 4,000 m. To the south, the landscape is more steeply dissected with a
series of tributary streams flowing steeply through granite hills, but yielding a limited
amount of good agricultural land. The Huatanay flows through a broad flood plain from
west to east and, approximately 20km from the city, passes through a gorge (La
Angostura), which effectively delimits the upper basin. The valley opens into a second
basin (Lucre), before joining the Urubamba river at Huambutio.
The physical geography of the Cuzco region is delimited and dominated in the north and
west by the Vilcanota Range. It is drained by two major rivers, the Urubamba and the
Apurimac. The former is divided into basins comprising fairly broad flood plains separated
by narrow gorges, and is characterized by a temperate climate with a valley floor altitude of
3,000 m at Pisaq and 2,750 m at Ollantaytambo; about 10 km below Ollantaytambo, it cuts
through the cordillera in a deep gorge and into the tropical montane forest. The Apurimac
valley is a deep and narrow gorge and its tributaries are steep, with only limited amounts of
good agricultural land. To the west of Cuzco, there is an area of fairly gentle, rolling
country, dominated in the south by a large, marshy area known as the Pampa de Anta
(3,600 m), and to the north by a number of streams and lakes of the limestone pampas. The
Huatanay valley, the Pampa de Anta and the Apurimac valley are surrounded by
mountain ranges, rarely higher than 4,500 m.

Social organization of the centre

Cuzco was divided into two moieties, all those descended from an Inka father and Inka
mother belonging to Hanan (upper), whilst those with a non-Inka mother were Hurin
(lower) (Zuidema 1989a: 257-61). Each was divided into five principal descent groups,
ayllus or panaqas, eight of which were associated with a mythical, royal Inka ancestor.
According to Betanzos (1551 [1987]), Cuzco had been replanned by the king, Pachacuti,
who reformed its social organization by adding the two principal ayllus: Capac Ayllu (the
royal ayllu), with his legitimate son and heir, Tupa Inka Yupanqui, as its head; and Hatun
Ayllu/ Ifiaca panaca (the great ayllu) for his other legitimate descendents, with himself as
its ancestor.
The king list therefore comprises ten rulers, each the ancestral head of an ayllu.
However, there was at least one other - Huayna Capac, the son of Tupa Inka - who was
not included in it. Since it is more appropriate to view the system as ahistorical, there was
probably little need for him to be accommodated as an ancestral head in a revised scheme,
particularly since all group leaders and members performed specific functions as the
370 I. S. Farrington

descendants of mythical ancestors. He would have been head of Capac Ayllu during his
reign and subsumed into Hatun Ayllu after his death. Conrad and Demarest (1984) regard
dynastic succession and the need to accumulate private estates as a driving force for the
expansion of the empire. But they fail to understand the inherent stability in the social and
political organization of Cuzco, and the concept and role of the mythical ancestors as
binding agents rather than forces of expansion.
Zuidema (1989a: 256) indicates that the city and its moieties form a central area separate
from the upper Huatanay basin. Pachacuti assigned land to all Inkas in the valley from
both moieties, to its original inhabitants (pre-Inkas) and to foreigners (non-Inkas), who
had been brought in to assist the Inkas cultivate the land, build the city and channelize the
Huatanay as far as the Angostura. Each moiety was divided into quarters socially and
spatially: the most important was Chinchaysuyo in north-west Hanan, the others being
Antisuyo in the north-east and the Hurin ones of Collasuyo (south-east) and Cuntisuyo
(south-west). These also form the main organizational and spatial division of the whole of
the empire. Indeed, according to Guaman Poma (1615 [1980]: 337-8), the four quarters,
Tawantinsuyo, is the Inka name for the empire.

The ceque system and the definition of the inner heartland

The ceque system comprised 328 wakas of the Cuzco region arranged along 41 lines called
ceques, which radiated from Korikancha, the principal temple (Cobo 1653 [1990]). For
each line it contains such information as its category, the name of the social group which
had responsibility for it, as well as the waka's name, its type, what was sacrificed,
occasionally a geographical description of its location, and even information linking it with
an astronomical observation and calendrical event. Korikancha was the axis mundi of
Cuzco around which the ritual cycle and spatial organization of the centre was arranged.
There were nine ceques, arranged in groups of three, in each of Chinchaysuyo, Antisuyo
and Collasuyo, and fourteen in Cuntisuyo.
Zuidema (e.g. 1964; 1990) has utilized a structural analysis of the ceque system to
describe the social and political organization of Cuzco, in particular the ranking of royal
ayllus and non-Inka ones, and the marriage and political relationships between them. He
has also determined its use as a calendrical device to define a ritual year of 328 days
followed by an extra period of 37 days, and co-ordinated by six important astronomical
observations from various wakas to others on the horizon during the year (Zuidema 1988;
Aveni 1981).
Importantly, in contrast to the work of Niles (1987: 178), Zuidema and Aveni (Aveni
1981: 313) have established by field verification and toponymy that the lines are straight,
even over distances of up to 20km; about thirty of them have been mapped with reasonable
precision. The four quarters can be fairly well delimited and the outer boundary of the
inner heartland defined as the upper Huatanay basin as far as the Angostura (Fig. 1). In
other words, the area that Pachacuti is alleged to have reorganized is delimited by the
ceque system. A few ceques in each quarter have one or two wakas beyond the basin,
suggesting that the system may well have extended into the outer heartland.
Both Zuidema (1986; 1990) and Sherbondy (1979; 1982; 1986) have determined that the
lines can also be used to identify the irrigation districts of Hanan Cuzco, because the main
 The characterization of the provinces of the Inka heartland 371

A Mountainsummit // uzco - Suyo boundary K = Koricancha

___PsvilgCeSque
line S = Sacsahuaman
-- Pass Native village e
H Haukaypata
. Inkasite - Innerheartland
boundary

Figure 1 The inner Inka heartland.

Figure 2 Hypothetical distribution of lands and water around Cuzco (after Sherbondy 1982).
372 I. S. Farrington

ceques often cross a canal intake, spring or important point along the canal system.
Sherbondy (1982: 85-92) has laid down principles by which the lands of each ayllu can be
identified by mapping its first and last ceques, e.g. in defining the lands of Sucsu and
Aucaille panaqas in Antisuyo (Fig. 2).
The ceque system, therefore, is a multi-purpose device which defined and divided social
space in the upper Huatanay basin, the inner heartland, in a web of daily ritual practices
radiating from the principal temple, Korikancha. Each social group's rights to land and
water were confirmed by the rituals performed at the various wakas belonging to them.

Centre and boundary: further rituals and legends

The bounded nature of this space and the links between its edge and centre are confirmed
by a number of other ceremonies which took place at various times of the year. These
include: the situa ceremony; initiation rites to the mountains of the southern and western
boundaries; and the first day of the year, the harvest procession, and the coronation
procession of Huayna Capac - each following ceque lines to the centre from the eastern
boundary. The first phase of situa took place during August when four groups of 100,
representing the social groups of each quarter, assembled at the usno in Haukaypata
(Molina 1552 [1943]: 29-32). After suitable libations, the groups ran in straight lines in the
principal direction of their quarter, shouting in order to eliminate evil from the city. At the
boundary of the inner heartland, they handed over their weapons and shouts to the
non-Inka groups of the outer heartland.
Legends may also be interpreted to confirm these boundaries. For example, the origin
myth symbolically displays the importance of both the boundary at Wanakauri, the sacred
mountain, and the centre at Wanaypata, the original name for Korikancha, with the death
and petrification of one of Manco Capac's brothers (Urton 1990: 37-40, 120-4). Indeed,
the very journey from the Wanakauri to the centre followed the line of ceque forming the
boundary between Collasuyo and Cuntisuyo. The symbolic act of possessing the centre for
the Inkas was conducted by a sister, Mama Wako, by plunging a golden rod into the earth
at Sausero near to Korikancha. A different kind of story is that told about the Chanka wars
and the assumption of power by Pachacuti (Murua 1611-18 [1964]). In this tale a red giant,
thought to be the king Viracocha, appeared at Chitacaca on the northern edge of the basin
and threatened to destroy the city; Pachacuti asked him not to play his trumpet for fear of
destroying the world; he did not, and Pachacuti became king. This symbolizes the
importance of the new king who, by defeating the threat from the deposed at the boundary
of the central area, protected it from defilement and destruction, enabling him to validate
the reorganization of the city and of that space.

Archaeology of the inner Inka heartland

The settlement archaeology of the upper Huatanay basin (c. 270 sq km) is dominated by
Cuzco itself. The city was small, with an orthogonal plan of elite compounds and plazas,
lying between two channelized streams. There were two central foci: the Korikancha,
which was the religious centre or axis mundi, and the usno, a stone pillar in the plaza,
Haukaypata, about which state ceremony occurred (Fig. 1). Its overall plan has been
Table 1 Settlements and other sites of the inner Inka heartland.

Chinchaysuyo Antisuyo

Capac A Hatun A + Kenko, Callachaca Pumamarka, Val

Pukara and and Huaccoto and
Inquiltambo Yuncaypampa San Jeronimo

Single house - 1 1 - - 3
Small residential group 1 1 - 1 - 3
Large residential group - - 1 5 - 3
Planned town - - - - 4 1
Storerooms - - - - 3
Elaborate terrace - - 1 - - -
Palace - - -- .-
Carved rock - 20 25 3 - -
Hilltop terrace - 5 1 2 1 2
Ritual complex - 1 1 3 1 -
Agricultural terracing 1 31 25 8 5 4
with wall-steps - - - 1 1 -
with recessed parallel
staircases - 1 1 - - -
Channelization - * * - * *
with flood plain drainage - - * - *
Salt pans - --
Spring/reservoir - 7 8 4 - -

*=
present
374 I. S. Farrington

likened to a puma (Rowe 1967); but a more appropriate analysis of the Chronicle evidence
considers that the concept of the puma referred to the Inka body politic, with the king as its
head (Zuidema 1983).
The city was surrounded on its alluvial fan and hillsides by agricultural terracing,
storerooms and a number of small suburban villages (Hyslop 1990), although few surface
remains survive and much has been looted for building stone in the immediate environs of
the modern city. Nevertheless, there are clearly marked differences between the moieties
and suyos which are a function of their cultural importance, with those of Hanan displaying
the greatest amount of influence (Table 1). For example, there are over forty carved rocks
in Hanan (the majority arranged in at least six major groupings), but none at all in Hurin,
despite the fact that stones occur as wakas in both with similar frequency; the carving of so
many limestone boulders in Hanan was thus a major cultural statement. Similarly, there
was an almost equal frequency of springs in all suyos, yet most Inka reservoirs have been
recorded in Chinchaysuyo and Antisuyo. And despite the fact that more hill tops were
worshipped in Hurin, there is hilltop terracing in Hanan.
In Hanan, close to Cuzco, Inka settlement is of a ritual nature. Even Sacsahuaman,
overlooking the city, is now regarded as a ritual location for ceremonial fighting
(Rostworowski 1988: 79). It comprises three monumental zig-zag walls; buildings,
including a round tower; a plaza; and a whole array of carved rocks and waterworks.
Excavations have also revealed many prestigious burials. The site lies on the main ceque of
Hatun Ayllu, I 2b. Elsewhere, many Inka building groups are associated with ritual sites,
such as carved rocks, baths, and hilltop' terracing (e.g. Tambomachay, Puka Pukara,
Callachaca and sites on the flank of Wanakauri; cf. Fig. 1).
Residential settlement is highly localized, with virtually none in Chinchaysuyo, a
number of planned towns towards the Angostura in both Antisuyo and Collasuyo, and
only one nearer to the city in Cuntisuyo (Niles 1982). Some other groups of houses, both
large and small, also exist in the same general areas. There is historical information
concerning the villages of Sanu and Oma (Fig. 1); but in neither San Sebastian nor San
Jeronimo, their suspected locations, have any substantial remains been found. There are
at least three groups of long, narrow, multi-doored buildings, which are claimed to be
storerooms, on slopes overlooking the river in Collasuyo (Morris 1967).
There are important differences between suyos in terms of their agricultural landscapes.
In Chinchaysuyo, the main Chacan canals provide irrigation to large areas of Hatun Ayllu,
particularly at Llaullipata and Fortaleza (Sherbondy 1982). There is also one small area of
Inka terracing called Machaybamba, on the west bank of the Chacan, containing recessed
parallel staircases (RPS), which are correlated elsewhere with known areas of private
royal land (Farrington 1988; Heffernan 1989); it also lies adjacent to ceque I 2b in Hatun
Ayllu. Indeed, there is little private land claimed by noble Inkas in the inner heartland
during the later sixteenth century, and most claims were for fields and pastures elsewhere
within Hatun Ayllu and for salt pans near Callachaca in Antisuyo (Farrington and
Heffernan in prep.).
In contrast, the lands of Capac Ayllu, south of Saphi, were on relatively poor and
unstable slopes watered by a canal from a spring on ceque I Ic, itself augmented by a longer
one from outside the basin on the end of ceque I 2b of Hatun Ayllu (Sherbondy 1986: 43).
Its wakas were natural features and there is only one group of structures and terraces
 The characterization of the provinces of the Inka heartland 375

surviving at Killke. In fact the settlement pattern and geography in this sector is more
characteristic of Cuntisuyo. Nevertheless, the most important terracing in this sector must
have been the large area to the west of the city, between ceques I la and IVA la.
The difference in quality between these lands probably reflects the relative size and
importance of the two panaqas. Capac Ayllu, the king's descent group, would have been
smaller than Hatun Ayllu, that of the ruler's father, and would not have required such an
extensive tract of land with good water supplies as did Hatun Ayllu. However, the king
probably did require the terraces next to the city.
The agricultural landscape of Antisuyo contains some extensive, but separate, terracing
schemes between San Sebastian and San Jeronimo, whereas in Hurin, although
subdivided and assigned by the Inka in the same manner, there is virtually nothing. The
best land in the inner heartland lies on the floor of the Huatanay valley and Betanzos (1551
[1987]: 55-63) indicated that the river was to be walled for 20km to the Angostura and the
remains of some channelization have in fact been found (Farrington 1983). Yet these were
not royal private lands, nor did they belong to the main ayllus but to some non-royal ones
in both Antisuyo and Collasuyo.
The inner heartland was clearly a ritual landscape with carved rocks, springs, fountains,
reservoirs, channelized streams and terraced hilltops. It was also a functioning agricultural
landscape with irrigation canals, reservoirs, terraces and even drained fields. Its
archaeological variation was a function of the importance of the ayllus, with Hatun Ayllu
and, to a lesser extent, those of Antisuyo showing the most development. In contrast,
Capac Ayllu, the king's group, had a good water supply, but was poorly endowed, save for
an area of terracing alongside the city.

The boundary of the outer Inka heartland

The area surrounding the upper basin was occupied by a number of separate non-Inka
ethnic groups, collectively known as Inkas-by-privilege, because they were loyal allies who
performed certain tasks for the benefit of the Inkas. For example, they provided labour for
initial projects in the reconstruction of Cuzco, supplied tribute and served as loyal mitmaq
in distant provinces. They engaged in marriage alliances with the Inka and were linked to
the centre by a number of legends and rituals which underscored their subordinate, yet
important position, within the state. The use of their distribution to map the outer
heartland is fraught with danger, since many had been moved and replaced by Inka ayllu
representatives and foreign mitmaq from elsewhere in the empire, but Zuidema and Poole
(1982) have used it to demonstrate that the boundaries of the four suyos continued straight
through what will be defined as the outer heartland.
Several rituals and legends serve to define the boundary of the outer heartland (Fig. 3).
The second phase of the situa ritual began at the four points on the edge of the inner
heartland where Inka warriors handed over their weapons and shouts to the local ethnic
groups of the four quarters, who then ran in relay in a straight line through each suyo to
throw the evil into a river and wash their weapons. The ending points were near Tilka, near
Pisaq, near Urcos, and at Cusibamba (Fig. 3). Zuidema (1986: 188) has equated the
running lines with extensions of the ceques hierarchically, first in Chinchaysuyo and
Antisuyo, and last in Collasuyo and Cuntisuyo; yet they also seem merely to be convenient
376 I. S. Farrington

Figure3 The outerInkaheartland.

ceques within the most important group. In Chinchaysuyo, the line went between its first
ceque and the last of Cuntisuyo, along the watershed between the Anta and the Apurimac,
to reach the latter river near the important hilltop shrine of Tilka (Heffernan 1989). This
line was also utilized during the return of the capaccocha sacrifices to the provinces of
Chinchaysuyo (Molina 1552 [1943]: 70). The implication is that the lands of Capac Ayllu
must extend between these two ceques to include Limatambo. Further confirmation of this
place's role as a boundary is found in the symbolic death of a loyal captain, Villca Quiri,
near the waka of Apurimac during the Chanka threat (Santacruz Pachacuti 1613 [1950]).
Another boundary point is discovered in the Mayucati ritual held in January each year to
eliminate sickness from Cuzco (Molina 1943: 64-6; Zuidema 1983: 76-9). At the most
propitious moment, ashes were thrown into the Huatanay at Pumaqchupan in Cuzco and
flood waters carried them beyond the empire to the ocean; a group of noble runners
followed them down river and then along the Urubamba to Ollantaytambo, from where
they returned in a straight line to Cuzco, collecting salt icons from the vicinity of Maras
(Fig. 3). Zuidema (1983) contends that this ritual serves to describe the political
relationship of the centre with its outer heartland as the tail of the puma.
 The characterization of the provinces of the Inka heartland 377

The Ollantaytambo boundary location can be further refined with reference to the
legendary death of Inka Urqo, the chosen heir of Viracocha. He had led an uprising
against Pachacuti which had been put down in Yucay. In the struggle he was thrown from a
cliff into the Urubamba river and managed to drag himself out at Choquellusca, where he
was killed by a blow to the head with a stone (Sarmiento 1572 [1942]). The latter location,
12km downstream from Ollantaytambo, lies in the Urubamba gorge where the river
plunges down towards the jungle. It has an elaborate Inka wall and a painting of the sun on
a cliff. This must be another place of symbolic transformation at the edge of the outer
heartland. The earlier struggle probably demarcates the ethnohistorical boundary
between Yucay and Ollantaytambo (Farrington 1988).
Delimitation of the boundary is complicated by a long distance line to Villcanota, the
house where the sun is born, at the head of the Urubamba valley (Zuidema 1982: 439-45).
Priests, who undertook this pilgrimage, stopped at eight wakas in a straight line on the way
there, and at twelve on the return along the river; it occurred twice, on the June solstice
and again in the weeks prior to Capac Raymi, the December solstice. According to
Santacruz Pachacuti (1613 [1950]) this line marked the border of the region administered
by Cuzco, whilst Cobo (1653 [1980]: 139) suggested that the boundary of Collasuyo lay
near there at Lurucachi, the first station on the return. However, I do not believe that this
defined the outer heartland per se, whose boundary lay near Urcos, a hill-top shrine and
the turning point on the return, as well as the end, of the Collasuyo situa line. Certainly,
this line to Villcanota signified an area under Inka control, but beyond the outer heartland:
indeed, in its broader context the line continued to Lake Titicaca and Tiwanaku, a cult
centre of a millennium earlier, and is associated with the origin myth of the god Viracocha.
Santillan (1563 [1950]: 9, 47) defined the outer heartland by noting the places where the
four imperial quarters began: Vilcaconga, Abisca, Urcos and Cuzco. Of these, only Urcos,
the Collasuyo situa terminus, and Vilcaconga, which lies at the head of the Limatambo
valley on the main road to Chinchaysuyo (Heffernan 1989), are useful here. According to
Cobo (1653 [1990]: 166), during the war of succession following the death of Huayna
Capac, the army of the Cuzco king Waskar fought a battle at Vilcaconga - symbolically on
the boundary of the outer heartland - but it was defeated by that of his rival, Atawallpa. Of
the other places noted by Santillan, Abisca lies along an extension of the conquest line to
Antisuyo beyond the waka Cafiachuay (Zuidema 1976), whilst he has either mistaken
Cuzco for the Cuntisuyo situa station of Cusibamba, or has mixed the spatial concepts of
suyos within the different concentric zones of the heartland because all would have begun
at Korikancha in Cuzco.
The boundary of the outer heartland can therefore be drawn by linking together the
various rituals with legends of conquest, battle and death (Fig. 3). Beginning at Urcos, it
crossed the divide via a mountain-top shrine above Rondocan (station 5, Villcabamba
ritual) to the Apurimac river at Cusibamba. The southern boundary was the deep gorge of
the river as far as the Inka bridge below the tambo of Apurimac and Cerro Tilka. From
here it ran around the Limatambo basin, passing across Salkantay and turning east towards
Vilcaconga, and then northwards to the Urubamba at Choquellusca. The northern
boundary was probably along the watershed of the Urubamba tributaries (including the
peaks, and apus such as Chicon and Pitusiray), and thus back again to Urcos.
378 I. S. Farrington

Social geography

It is possible that ceque lines could have been extended into this area (Fig. 3). Some of
them reach prominent peaks, on the boundary of the outer heartland, which feature in
Inka myth, for example I la to Salkantay and I 3a to Pitusiray. Inka sites also are to be
found on, or close to, the end-points of ceque extensions (e.g. Mawka Llacta (IVB Ic) and
Pisaq (III 2a)). Intermediate points along or near these may also be marked by mountain
tops (e.g. Antaquilka (I 2c), Huaypo Wanakauri (I Ic), or by Inka sites, (e.g. Cuper (I 2b),
Huchuy Cuzco (III la)). Therefore, coupled with the known Situa, Mayucati and Capac
Raymi lines, the outer heartland can be regarded as a ritual landscape of shrines, mountain
top apus and Inka sites.
Using Sherbondy's principles, the extended ceque lines can also be used to distinguish
the lands which could possibly have been associated with Cuzco ayllus. The majority of
land remained in the hands of Inkas-by-privilege, but the Inkas did expropriate lands for
royal estates, including maize and potato fields, pastures, woodlands, salt pans and
hunting grounds, and these were scattered throughout the zone (Rostworowski 1962).
They were concentrated in Chinchaysuyo, and in particular in the Valley of Yucay, an area
which would have been associated with Hatun Ayllu. Evidence presented in a 1574 court
case noted which kings possessed which estates (Rostworowski 1970: 253). For example,
Viracocha's estate at Caquia Jaquijahuana and Paucartica clearly lies in the direction
associated with himself in Antisuyo, and indeed within the territory defined by the ceques
III la-b of his panaqa.
Pachacuti had estates at Pisaq and Tambo, but their analysis is less straightforward. The
former lies on III lc, an area associated with Viracocha'spanaqa, and which, according to
one chronicler, was burned after the coup against his father. The latter at Ollantaytambo
was in the sector associated with his own panaqa, Hatun Ayllu, between I 2a and I Ic.
There is some suggestion, in a document recently researched by Rowe (1987), that
Pachacuti was also associated with the area of Picchu further down the Urubamba, beyond
the boundary at Choquellusca, in the Cusichaca and Machu Picchu areas. Indeed, ceque I
Ic can be further extended beyond the outer heartland as far as Patallacta in Cusichaca and
Machu Picchu. However, such estates were not claimed in any extensive manner in the
early Colonial period by his descendants and their exact status as private lands must
remain in question.
The estates of Huayna Capac were at Quispeguanca in the Valley of Yucay, and
adjacent to them those of Topa Inka at Chinchero and Huayllabamba. They are in the
sector defined by ceques I 2a and I 2c, and it is possible that I 2b marks the boundary
between them. A detailed document of 1551 locates these lands and confirms their royal
ownership. It describes their layout within the remaining local communities and also land
devoted to the cult (Villanueva 1970; Farrington 1988; Farrington and Heffernan, in
prep.). These lands are in the sector nominally associated with Hatun Ayllu, yet
conventionally neither of these two rulers were. Indeed, in the ceque system, as it is
understood as a social model, the lands of Topa Inka's descent group, Capac Ayllu, should
have been in the western parts of the Outer Heartland between ceques IVA la and I lc.
Private royal estates are known through early colonial documents in the Limatambo and
 The characterization of the provinces of the Inka heartland 379

Chinchaypukio valleys and at Zurite on the Pampa de Anta, all in the area of Capac Ayllu,
and which are claims by sons and grandsons of Huayna Capac or are associated with his
panaqa (Heffernan 1989).
Similarly, in the ceque system there was no space for the panaqa of Huayna Capac and
no lands in Cuzco could be attributed to him. Zuidema (1989b) views his position in the
system as a king of the outside, of Ecuador, with few ties to Cuzco. The recency of Huayna
Capac's death and the civil war may have masked the true ownership of private land and it
could be argued that the royal lands of Yucay and Chinchero within Hatun Ayllu's sphere
are correctly assigned, because these rulers were dead, and therefore 'fathers', who would
belong to that descent group as ancestors. The lands in the western districts claimed by
Huayna Capac's descendants may also be legitimate assertions of former ownership to
land which - as a result of the civil war and Spanish invasion - had failed to be reassigned to
the new king. The area of Capac Ayllu could then be regarded as containing the estates
belonging only to the incumbent king. In the Valley of Yucay it was lands belonging to
Topa Inka that were made available to the Spaniards in 1551, not those of Huayna Capac;
this may suggest that the importance of Topa's estate was diminished with the death of
Huayna Capac (i.e. the previous king's personal lands only remained his for one
generation).
The lands of Waskar in Calca on the boundary between Chinchaysuyo and Antisuyo,
and at Muyna on the boundary between Antisuyo and Collasuyo, indicate that he was not
fully sponsored by Hatun Ayllu or Capac Ayllu in his abortive attempt to remain king.

Inka settlement archaeology of the outer heartland

There has been no comprehensive survey of the whole outer heartland, although there
have been a number of important regional studies, including: Heffernan (1989) in
Limatambo and adjacent valleys; Bauer (1990) in Paruro and Pacariqtambo; MacEwan
(1987) in Lucre; and Farrington (in prep.) from Pisaq to Ollantaytambo. These, together
with less systematic work in the Chinchero, Maras and Pampa de Anta regions, provide
the basis for analysis. The areas of Cusichaca (Kendall 1984; 1985), Cedrobamba (Fejos
1944) and Aobamba/Santa Teresa (Drew 1984), outside the heartland, may also be
considered.
The imperial imprint in the outer heartland, although made up of the same building
forms and landscape ideals of shrines, terraces and channelized rivers, is varied. This must
be related to the distribution of Inkas-by-privilege, ritual geography, the association of
particular sectors with various royal ayllus, and the agricultural importance of a district. It
is not possible to utilize standard procedures for the analysis of settlement hierarchy,
because the dominant Inka impact in a district is not always an urban settlement, but could
equally well be a palace, a ritual site or an agricultural landscape. Indeed, in the outer
heartland, and possibly throughout the empire as well, the nature of the Inka city was one
of a central place pursuing certain ritual and administrative functions, and whose
population was largely drawn from the nobility or the Inkas-by-privilege. The Inkas
certainly did not completely transform the settlement pattern into one dominated by their
Table 2 Settlements and other sites of the outer Inka heartland.

Cuntisuyo

Yaurisqui Paccari- Paruro Maras Yucay Chin- Calca Lima- Ollanta

tambo chero tambo tambo

Sherd scatter 22 31 40 some - some some 7 -

Single house - - - - - - 2 - 3
Small residential group - - - 1 1 2 - 2 14
Large residential group - - - - - - - 5
Planned town - 1 - - 9 - 1 - 1
Storerooms - - - - some - 2 - many
Elaborate terrace - - - - - I - 3 1
Palace - - - - 3 1 1 - -
Carved rock - - - - 3 1
Hilltop terrace 1 - - - 1 1 - 2 2
Ritual complex -- - - -_- -
Agricultural terracing - * * * * * * *
with wall-steps -* - 1 4 most most some most
with recessed parallel
staircases - - most some 2 most 1
Channelization - ? - - * * * * *
with flood plain drainage - - - _ * * * ** *
Saltpans - - - *

= present
 The characterization of the provinces of the Inka heartland 381

occupation: the local population and economy was allowed to continue. Three types of
settlement pattern can be observed (Table 2):
1. The first is that in which the level of Inka presence is minimal, with continuity of
pre-existing settlements; it is characteristic of Cuntisuyo (Bauer 1990), where in three
survey districts Inka occupation was marked by pottery scatters of mixed provenance.
There were a few small terracing schemes along the Taray, Huaynacancha and Molle
Molle rivers and an urban centre, Mawka Llacta, alongside the main Inka origin shrine,
Puma Orqo. This area had loyal and important Inkas-by-privilege (Poole 1984). It is
similar to the area from Maras to east of Anta.
2. The second pattern is illustrated by Yucay, whose landscape is dominated by extensive
Inka terracing and canalization schemes. There is no town, and hierarchy is dominated by
three palace complexes with their attendant ritual rocks and elaborate terraces. There are
some isolated buildings and storerooms. Recently, Farrington (1988) and Heffernan
(1989: 415-35) have argued that such terraced lands can be distinguished architecturally,
by reference to the mode of access between levels. The private estates of Huayna Capac
and of Tupa Inka and their relatives are characterized by recessed parallel staircases
(RPS), lands of the sun by wall-steps, whilst the lands of the local population were
unterraced. Lands reclaimed by river channelization also belonged to these kings. The
distribution of RPS and major channelization schemes may reveal the pattern of royal
private lands throughout the outer heartland. It is interesting that in the Urubamba valley
not only Yucay, but also Chinchero and Calca, contain more than one such scheme
(Fig. 3). Elsewhere in Chinchaysuyo, the major terracing at Zurite, Limatambo and
Chinchaypukio all have RPS, as does one flight in the small Pantipata valley (Heffernan
1989). This confirms the royal and private nature of the lands of Capac Ayllu.
3. The third pattern, characterized by Ollantaytambo, is a well-transformed landscape,
focusing on an Inka town and ritual area with carved rocks, fountains, and sun temple with
many outlying elite settlements and ritual complexes. River channelization and terracing
with wall-steps extend for about 10km either side of the town along the Urubamba valley
and up its tributary to around 3,300m, just below the potato growing zone. There is only
one small flight of RPS style terracing near the town. For this pattern of development,
there had to have been an almost total removal of the local population and the land
expropriated for terracing. Only a few areas of steep hill country remained in the hands of
local groups and therefore unterraced. Ollantaytambo is understood to be Pachacuti's
estate, but its pattern is not like that of Yucay where royal estates have left a particular
archaeological signature. Whilst it is possible that there were some small-scale private
lands there, this district's pattern should be interpreted as a state development associated
with Pachacuti's own panaqa, Hatun Ayllu, not that of his private estate.
This pattern is very similar to Cusichaca and Cedrobamba/Machu Picchu, down valley
and beyond the boundary of the outer heartland. Each contains a town, many ritual sites,
isolated smaller Inka groups, and much terracing with wall-steps. In the upper Cusichaca
drainage, there are two channelization schemes, and in the lower one a relatively small
scheme associated with wall-stepped terracing. An RPS is reported from near Torontoy
(Orellana, pers. comm.). Despite the argument that these form Pachacuti's estate (Rowe
1987), private land is not archaeologically manifest in either district. The similarity to
Ollantaytambo suggests that these were also developed in his name.
382 I. S. Farrington

Conclusion

The two zones of the Inka heartland are defined by the analysis of rituals, shrines and
legendary history. The relationships between the Inka centre and these zones are a
complex web of social, political and economic factors which are underscored and
elaborated by a commonly understood culture. The explanation of archaeological
variability of the inner provinces lies in the full understanding of their ritual geography and
its social implications, with most development in Hanan moiety - in particular the zones
attributed to Hatun Ayllu in Chinchaysuyo. Three basic settlement patterns characterize
the outer heartland: one of continued local settlement with an Inka veneer; a second of
private royal estates; and the third a totally transformed landscape. The correlation
between these archaeological patterns, the use of ceques to indicate land associated with
the noble ayllus, and other sixteenth-century documentation suggests that Capac Ayllu
remained for all time the incumbent king's ayllu with a few active estates, and on death that
king became incorporated into Hatun Ayllu for at least a couple of reigns, with a different
set of lands. The association of Pachacuti with estates in Ollantaytambo and Picchu is not
one of ownership, but rather of state development in the sector conventionally attributed
to him.

Acknowledgements

I am grateful to the Australian Research Committee and the Australian National

University Faculties Research Fund for several research grants for field work in the
Urubamba valley, and to Ken Heffernan for numerous discussions.
23.v.91 Department of Prehistory and Anthropology
The Australian National University
GPO Box 4
CanberraACT 2601
Australia

References

Aveni, A. F. 1981. Horizon astronomy in Incaic Cuzco. In Archaeoastronomy in the Americas (ed.
R. A. Williamson). Los Altos: Ballena Press.
Bauer, B. S. 1990. State development in the Cusco Region: archaeological research on the Incas in
the province of Paruro. Doctoral dissertation, University of Chicago.
Betanzos, J. de 1551. Suma y Narracion de los Incas (1987) (ed. M. del Carmen Martin Rubio).
Madrid: Atlas.
Cobo, B. 1653. History of the Inca Empire (1980) (transl. and ed. R. Hamilton). Austin: University
of Texas Press.
Cobo, B. 1653. Inca Religion and Customs (1990) (transl. and ed. R. Hamilton). Austin: University
of Texas Press.
 The characterization of the provinces of the Inka heartland 383

Conrad, G. W. and Demarest, A. A. 1984. Religion and Empire. The Dynamics of Aztec and Inca
Expansion. Cambridge: Cambridge University Press.
Drew, D. 1984. The Cusichaca Project: aspects of archaeological reconnaissance - the Lucumayo
and Santa Teresa valleys. In CurrentArchaeological Projects in the CentralAndes: Some Approaches
and Results (ed. A. Kendall). Oxford: British Archaeological Reports International Series 210.
Farrington, I. S. 1983. Preliminary notes on river canalization in the Sacred Valley of the Incas. In
Drained Field Agriculture in Central and South America (ed. J. P Darch). Oxford: British
Archaeological Reports International Series 189, pp. 221-35.
Farrington, I. S. 1988. The sixteenth century landscape of Yucay. Paper read at the 46th
International Congress of Americanists, Amsterdam.
Farrington, I. S. In prep. Inka settlement in the Sacred Valley.
Farrington, I. S. and Heffernan, K. J. In prep. Inka land tenure.
Fejos, P. 1944. Archaeological Explorations in the Cordillera Vilcabamba. New York: Viking Fund
Publications in Anthropology, 3.
Heffernan, K. J. 1988. Inca sites in high places near Cuzco. Paper presented at the 46th International
Congress on Americanists, Amsterdam.
Heffernan, K. J. 1989. Limatambo in late prehistory: landscape archaeology and documentary
images of Inca presence in the periphery of Cusco. Doctoral dissertation, Australian National
University.
Guaman Poma de Ayala, F. 1615. El Primer Nueva Coronica y Buen Gobierno (1980). M6xico D F:
Siglo Veintiuno.
Hyslop, J. 1990. Inka Settlement Planning. Austin: University of Texas Press.
Kendall, A. 1984. Archaeological investigations of Late Intermediate Period and Late Horizon
Period at Cusichaca, Peru. In Current Archaeological Projects in the Central Andes: Some
Approaches and Results (ed. A. Kendall). Oxford: British Archaeological Reports International
Series 210: pp. 247-84.
Kendall, A. 1985. Aspects of Inca Architecture: Description, Function and Chronology. Oxford:
British Archaeological Reports International Series 242.
McEwan, G. F. 1987. The Middle Horizon in the Valley of Cuzco, Peru: The Impact of the Wari
Occupation of the Lucre Basin. Oxford: British Archaeological Reports International Series 372.
Molina, C. de 1552 [1943]. Fdbulas y Ritos de los Incas. Lima: Los Pequefios Grandes Libros de
Historia Americana, Serie 1.
Morris, C. 1967. Storage in Tahuantinsuyo. Doctoral dissertation. Cornell University.
Mur6a, Fray M. de 1611-18. Historia General del Peru (1964). Madrid: Instituto Gonzalo Fernandez
de Oviedo.
Niles, S. A. 1982. Civic and social engineers: Inca planning in the Cuzco region. Doctoral
dissertation. University of California, Berkeley.
Niles, S. A. 1987. Callachaca: Style and Status in an Inca Community. Iowa City: University of Iowa
Press.
Poole, D. A. 1984. Ritual-economic calendars in Paruro: the structure of representation in Andean
ethnography. Doctoral dissertation, University of Illinois at Urbana-Champaign.
Reinhard, J. 1985. Sacred mountains: an ethno-archaeological study of high Andean ruins.
Mountain Research and Development, 5: 299-317.
Rostworowski de Diez Canseco, M. 1962. Nuevos datos sobre tenencia de tierras reales en el
Incario. Revista del Museo Nacional, 31: 130-59.
384 I. S. Farrington

Rostworowski de Diez Canseco, M. 1970. El repartimiento de Dofia Beatriz Coya en el Valle de

Yucay. Historia y Cultura, 4: 153-267.
Rostworowski de Diez Canseco, M. 1988. Historia del Tahuantinsuyo. Lima: Instituto de Estudios
Peruanos.
Rowe, J. H. 1944. An Introduction to the Archaeology of Cuzco. Papers of the Peabody Museum of
American Archaeology and Ethnology, vol. 22. Cambridge, MA: Harvard University.
Rowe, J. H. 1967. What kind of settlement was Inca Cuzco? Nawpa Pacha, 5: 59-76.
Rowe, J. H. 1987. Machu Pijchu a la luz de los documentos del siglo XVI. Kuntur, 4: 12-20.
Santacruz Pachacuti Yamqui Salcamaygua, J. de 1613 Relaci6n de Antigliedades deste Reyno del Pirt
(1950). Asunci6n: Editorial Guaranfa.
Sallnow, M. J. 1987. Pilgrims of the Andes: Regional Cults in Cusco. Washington DC: Smithsonian
Institution Press.
Santillan, F. de 1563. Relaci6n del Origen, Descendencia, Politica y Gobierno de los Incas (1950).
Asuncio6n:Editorial Guarania.
Sarmiento de Gamboa, P. 1572. Historia de los Incas (1942). Buenos Aires: Emece.
Sherbondy, J. E. 1979. Les r6seaux d'irrigation dans la geographie politique de Cuzco. Journal de la
Societe des Americanistes, 66: 45-66.
Sherbondy, J. E. 1982. The canal systems of Hanan Cuzco. Doctoral dissertation. University of
Illinois at Urbana-Champaign.
Sherbondy, J. E. 1986. Los ceques: c6digo de canales en el Cusco Incaico. Allpanchis, 27: 39-74.
Urton, G. 1990. The History of a Myth: Pacariqtambo and the Origin of the Inkas. Austin: University
of Texas Press.
Villanueva, H. 1970. Documentos sobre Yucay en el siglo XVI. Revista del Archivo Hist6rico del
Cuzco, 13: 1-54.
Zuidema, R. T. 1964. The Ceque System of Cuzco: The Social Organization of the Capital of the
Incas. Leiden: E. J. Brill.
Zuidema, R. T. 1976. Mito, rito, calendario y geograffa en el Antiguo Peru. Actes du 42 Congres
International des Americanistes: 347-57.
Zuidema, R. T. 1982. Bureaucracy and systematic knowledge in Andean civilization. In The Inca
and Aztec States 1400-1800: Anthropology and History (eds G. A. Collier, R. I. Rosaldo and J. D.
Wirth). Cambridge: Cambridge University Press.
Zuidema, R. T. 1983. The lion in the city. Journal of Latin American Lore, 9: 39-100.
Zuidema, R. T. 1986. Inka dynasty and irrigation: another look at Andean concepts of history. In
Anthropological History of Andean Polities (eds J. V. Murra, N. Wachtel and J. Revel). Cambridge:
Cambridge University Press.
Zuidema, R. T. 1988. The pillars of Cuzco: which two dates of sunset did they define? In New
Directions in American Archaeoastronomy (ed. A. F. Aveni). Oxford: British Archaeological
Reports International Series 454, pp. 143-69.
Zuidema, R. T. 1989a. The moieties of Cuzco. In The Attraction of Opposites: Thought and Society
in the Dualistic Mode (eds D. Maybury-Lewis and U. Amagor). Ann Arbor: University of Michigan
Press.
Zuidema, R. T. 1989b. At the King's table: Inca concepts of sacred kingship in Cuzco. History and
Anthropology, 4: 249-74.
Zuidema, R. T. 1990. Inca Civilization in Cuzco. Austin: University of Texas Press.
 The characterization of the provinces of the Inka heartland 385

Zuidema, R. T. and Poole, D. 1982. Los limites de los cuatro suyos incaicos en el Cuzco. Bulletin de
l'Institut Francaise d'Etudes Andines, 11: 83-9.

Abstract

Farrington, I. S.
Ritual geography, settlement patterns and the characterization of the
provinces of the Inka heartland
This paper defines two zones of the Inka heartland around its capital, Cuzco, using both rituals and
legendary history to describe boundaries and identify territories associated with the noble descent
groups. The variability of archaeological settlement patterns in both zones is explained by this ritual
geography. The most important sectors of both are associated with Chinchaysuyo and Hatun Ayllu
in particular. Three basic patterns are described which have implications for the understanding of
land tenure and succession to the throne.