COCA LEAVES IN THE CONTEXT OF
THE CENTRAL ANDEAN WARI STATE
Lidio M Valdez & Juan Taboada
A paper delivered at the 46th. Annual Chacmool Archaeology Conference,
Trading Spaces: the Archaeology of Interaction, Migration and Exchange.
University of Calgary, November 07 – 11, 2013.
The Middle Horizon (ca. 550 – 1100 CE) was a period during which the Wari State
emerged in the Peruvian central highland valley of Ayacucho and expanded
beyond its heartland to exercise political and economic control over most of the
Central Andes. One of the probable reasons for the expansion of Wari was to
establish direct access over the resources of other regions, including the tropical
forest region. In addition to the unprecedented finding of the burial of an elite
Wari leader at Vilcabamba, east of the Apurimac Valley, archaeological research
carried out in the Apurimac Valley has shown that the Wari State successfully
colonized the region. In the particular case of Apurimac Valley, it has been
suggested that the establishment of Wari settlements in the tropical forest region
was in order to access local products, in particular coca leaves. Until recently, only
indirect evidence indicated the use of coca leaves by the inhabitants of the Wari
State. New evidence coming from a Wari settlement in the highlands, consisting
in the fortuitous and unprecedented finding of coca leaves in association with
Wari material culture, confirms that the Wari State produced and consumed coca
leaves and that this product was likely transported from the Apurimac Valley.
Correspondence: L. M. Valdez, firstname.lastname@example.org
Within Central Andean archaeology, the Middle Horizon (ca. 550 – 1100 CE) was a period during
which the Wari State emerged and from its center in the Peruvian central highland valley of
Ayacucho expanded over most of the Central Andes to establish political and economic control.
Wari-style artifacts and several highly uniform architectural complexes – identified as Wari
administrative centers – are found throughout much of the Central Andes signaling the presence
of the Wari State. The inhabitants of the newly incorporated regions experienced significant
changes when their settlements were relocated, often near maize growing ecological zones. Such
changes likely mirror the social and economic reorganizations imposed on the local conquered
peoples by the Wari administration. Available archaeological evidence indicates Wari’s expansion
beyond the Ayacucho Valley likely was to establish direct access over the resources of other
One such region incorporated into the Wari domain was the Apurimac Valley, a tropical
forest region east of the Ayacucho Valley and only 3 days walking distance from the Wari capital.
In contrast to the highlands and the western lowlands, the tropical forest region offered a whole
range of resources such as colourful feathers and other exotic resources (Raymond 1988:290)
that could not be found elsewhere. However, it has long been suggested that one of the most
valuable tropical forest resources was coca leaves (Erythroxylum coca). This suggestion assumes
that coca leaves were consumed at the time the Wari State flourished in the Central Andes.
Coca is a tropical rain forest plant and non-frost resistant, therefore it could not be
cultivated in the highlands, such as in the Ayacucho Valley (Allen 1988:221; Raymond 1992:23).
For the Inca State coca leaves were a highly valued resource, but in order to cultivate them the
state colonized the tropical rain forest region (Gade 1999:139). However, colonization of a totally
different ecological zone, such as the tropical rain forest region, was nothing new since several
centuries before the rise of the Inca State, the Wari State already had established lowland
outposts in the Apurimac Valley.
It is also important to point out that at the time of the Spanish conquest coca use in the
Andean region of South America was widespread (Burchard 1974). Cieza de León (cited by
Mortimer 1974:151) pointed out that ‘throughout Peru the Indians carry this coca in their
mouths; from morning until they lie down to sleep they never take it out. When I asked some of
these Indians why they carried these leaves in their mouths, which they do not eat, but merely
hold between their teeth, they replied that it prevents them from feeling hungry, and gives them
great vigor and strength’ (see also Cieza de León 1973:221). In more recent times, coca leaves
continue to play an important role in the traditional highland economy, such as the exchange of
goods (Valdez 1997:68). Ann Kendall (1973:92) also notes that besides its action as a stimulant
or a painkiller, ‘coca leaves stop diarrhoea and its juices dry up ulcers.’
The idea that not only coca chewing was practiced during the time the Wari State
flourished, but also that the Wari administration cultivated their own coca supply is further
supported by the finding of Wari settlements in the tropical forest valley of Apurimac. In 1968
and 1970 Scott Raymond carried out an archaeological survey of the lower Apurimac Valley,
locating several archaeological sites, some of which were identified as Wari outposts (Raymond
1985:42, 1988, 1992). On the basis of those findings, Raymond (1992:30) has convincingly argued
that Wari colonization of the Apurimac Valley likely was in order to secure coca leaves. Raymond
also mentions that all known Wari settlements in the Apurimac Valley were found at elevations
suitable for coca cultivation and, with a single exception, all established along the western side
of the valley.
On the basis of available archaeological evidence it can be argued that the tropical forest
region was not alien to the Wari State. Confirming Wari’s familiarity with the region, (8) there is
a Wari ceramic drinking cup molded in the shape of a tapir foot. The drinking cup was found at a
highland Wari settlement in the Pampas Valley, south of the Wari capital (Raymond 1979:81).
The tapir (Tapirus terrestris) is one of the largest mammals of the tropical forest of South America
Furthermore, recent archaeological research at Vilcabamba, east of the Apurimac Valley
and deep in the tropical forest region, has uncovered the burial of an elite Wari leader (Fonseca
Santa Cruz 2011; Valdez 2011). This finding is unprecedented and demonstrates that Wari
colonization of the tropical forest region was successful. In addition, it appears that the Wari elite
took an active role in such a process, as the finding from Vilcabamba suggests. If Wari’s presence
in the lowlands was indeed aimed at coca cultivation, the participation of the Wari elite may
indicate that coca leaves played a critical role within the political economy of the Wari State.
Additional indirect archaeological evidence further supports the idea that coca chewing
was indeed practiced during Wari times. An offering pit excavated at Pikillaqta, a Wari provincial
center in the Cuzco region, reveals – among other items – the presence of miniature Wari
warriors (Arriola Suni 2008:34, 36; Arriola Tuni and Tesar 2011: Figures 16 & 19). One interesting
aspect of the miniature warriors is their bulging cheeks. Ethnographically it is known that coca
chewers held the coca leaves between the mouth and gums, resulting in bulging cheeks (Allen
1988). Therefore, it is possible that the bulging cheek of the Pikillaqta miniature warriors indicate
If this observation is correct, it is reasonable to argue that the personnel of the Wari army
were provided with supplies of coca leaves. As already pointed out, coca leaves are an excellent
stimulant to overcome hunger, thirst, fatigue, and altitude sickness (Allen 1988:221-222;
Flannery, Marcus & Reynolds 1989:43; Cieza de León 1973:221; see also Mortimer 1974). An
armed contingent on the constant move and facing the challenges of hunger, thirst, fatigue and
altitude sickness likely benefited from the use of coca leaves. Therefore, it would not be an
overstatement to argue that – to some extent – the success of Wari’s military expansion was
perhaps due to the use of coca leaves; ultimately, such an association perhaps was one the
reasons that transformed the value of this resource. Other uses of coca leaves in the distant past,
such as its linkage with ritual activities, may have been other reasons.
Recently, at the site of Convento, in the locality of Puerto San Antonio (3450 m asl), in
Tayacaja (Huancavelica), about 90 km North of the Wari capital, school construction workers
unearthed archaeological remains, including ceramics. The fortuitous presence of the second
author made it possible to rescue some of the archaeological remains, including a burial offering.
As the workers opened trenches in order to establish the foundation of the new school building,
they came across several stone walled burial cists, similar to those uncovered at other Wari sites
in the Ayacucho Valley, such as Posoqoypata (Valdez, Bettcher & Valdez 2002:400).
Unfortunately, the archaeological remains, in most cases were smashed, while others had been
removed and taken by the working party.
The offerings in consideration were partially uncovered when the second author arrived
at the location and he was allowed to clear, photograph and recover the findings so that the
school construction work would continue. In addition to a large plain ceramic vessel already
smashed, there were fragments of what appeared to be pieces of an anthropomorphic vessel,
unfortunately smashed. However, a second vessel was recovered intact and consists of a small
bottle decorated in the Cruz Pata style that according to Lumbreras (1974:137) dates to the end
of the Early Intermediate period. Here it is important to stress that according to Knobloch
(1991:248) Cruz Pata ceramics are found at Wari in Middle Horizon 1A contexts. For the specific
case of the bottle from Convento, Knobloch identifies it as Cruz Pata white, again placed in Middle
Horizon 1A (Menzel 1964). This stylistic identification is crucial in the absence of absolute dates.
In association with the Cruz Pata bottle were also found two metal tupus pins one placed
over the other. Most noteworthy is that between the two tupus noted here there were coca
leaves. In a region such as the central Andean highlands, organic remains – such as coca leaves –
seldom survive; consequently this finding is extraordinary. We should also mention that tupus
are artifacts with a strong gender association, as they were used to fasten women’s clothing
(Kendall 1973:36-37, 170; Malpass 2009:78). Therefore, the artifacts noted here likely are the
grave goods of a female individual.
The association of the artifacts noted here and the stylistic placement of the bottle are of
critical value in order to elucidate the significance of this finding. On the basis of the stylistic
identification provided by Knobloch, it becomes evident that as early as the Middle Horizon 1A -
if not earlier – coca leaves were already used in the central highland region. In addition, the
evidence in consideration leaves open the possibility that coca leaves already played a magic-
religious function. At the same time, the association of the tupus with the coca leaves allows
arguing strongly that in the ritual activities that involved the use of coca leaves women played a
Previous research and indirect archaeological evidence suggested that coca leaves were
already in use during Wari times; the evidence presented here largely supports the involvement
of the Wari State in coca production. However, the early chronological placement of the Cruz
Pata ceramics suggests that by the time the Wari State emerged, coca leaves were already
transported and distributed across the central highland region. Therefore, it appears very likely
that prior to Wari, highland populations were already familiar with this tropical forest resource,
and it is equally possible that the first highland outposts in the Apurimac Valley were established
sometime before the emergence of the Wari State.
Here, it is timely to point out that according to Raymond (1992:25-26), the Simariba
ceramic tradition from the Apurimac Valley shares some formal features with “those found in the
neighbouring highlands from the latter part of the Early Intermediate period.” Such a similarity
may well be because the Simariba ceramic tradition is contemporaneous with the highland late
Early Intermediate period; alternatively, the Simariba ceramic tradition may actually represent
original highland occupation established prior to the emergence of the Wari State. In order to
clarify this issue, it is evident that further research is needed in the Apurimac Valley.
On the basis of the present discussion, and to conclude, it is plausible that Early
Intermediate period highland communities were the first to establish the earliest lowland
outposts in the Apurimac Valley; during the Middle Horizon, the Wari State continued with this
already established pattern. It is certainly possible that with Wari coca production may have
intensified, resulting not only in a major occupation in the Apurimac Valley, but also in the
widespread use of coca leaves across the central highlands. Ultimately, the Wari State may be
the political organization that made this tropical resource of widespread use, which continues to
Allen, Catherine J.
1988 The Hold Life Has: Coca and cultural identity in an Andean Community. Smithsonian
Institution Press, Washington, D.C.
Arriola Suni, Carlos
2008 Excavaciones arqueológicas en Pikillaqta. Saqsaywaman 7:15-61.
Arriola Suni, Carlos & Louis D. Tesar
2011 The Pikillaqta 2004 eastern gate offering pit. Ñawpa Pacha 31 (1): 1-44.
Burchard, Roderick E.
1974 Coca y trueque de alimentos. In, Reciprocidad e Intercambio en los Andes peruanos,
edited by G. Aliberti & E. Mayer, pp. 209-251. Instituto de Estudios Peruanos, Lima.
Cieza de León, Pedro
1973 La Crónica del Perú. Ediciones PEISA, Lima.
Flannery, Kent V., Joyce Marcus & Robert G. Reynolds
1989 The Flocks of the Wamani: a study of llama herders on the punas of Ayacucho, Peru.
Academic Press, New York.
Fonseca Santa Cruz, Javier
2011 El rostro oculto de Espíritu Pampa, Vilcabamba, Cusco. Arqueología Iberoamericana 10:
Gade, Daniel W.
1999 Nature and Culture in the Andes. The University of Wisconsin Press, Madison.
1973 Everyday life of the Incas. Dorset Press, New York.
Knobloch, Patricia J.
1991 Stylistic date of ceramics from the Huari centers. In Huari Administrative Structure:
prehistoric monumental architecture and state government, edited by W. H. Isbell & G. F.
MacEwan, pp. 247-258. Dumbarton Oaks Research Library and Collection, Washington,
Lumbreras, Luis G.
1974 The Peoples and Cultures of Ancient Peru. Smithsonian Institution Press, Washington, D.C.
Malpass, Michael A.
2009 Daily life in the Inca Empire (2nd. Edition). Greenwood Press, Wesport, Connecticut &
1964 Style and Time in the Middle Horizon. Ňawpa Pacha 2:1-106.
Mortimer, W. Golden,
1974 History of coca: ‘the divine’ plant of the Incas. Fitz Hugh Ludlow memorial Library Edition,
Raymond, J. Scott
1979 A Huari ceramic tapir foot? Ñawpa Pacha 17:81-86.
1985 Quechuas and Chunchos: ethnic boundaries in eastern Peru. In, Status, Structure and
Stratification, edited by M. Thompson, M. T. García & F. J. Kense, pp. 39-50. The
Archaeological Association of the University of Calgary, Calgary.
1988 A view from the tropical forest. In, Peruvian Prehistory, edited by R. W. Keatinge, pp. 279-
300. Cambridge University Press, Cambridge.
1992 Highland colonization of the Peruvian montaña in relation to the political economy of
the Huari Empire. Journal of the Steward Anthropological Society 20 (1-2):17-36.
Valdez, Lidio M.
1997 Ecology and ceramic production in an Andean community: a reconsideration of the
evidence. Journal of Anthropological Research 53:65-85.
2011 Wari e Inka: el significado de Vilcabamba. Arqueología Iberoamericana 10: 3-5.
Valdez, Lidio M., Katrina J. Bettcher & J. Ernesto Valdez
2002 New Wari mortuary structures in the Ayacucho Valley, Peru. Journal of Anthropological