An offprint from

VARIATIONS IN THE EXPRESSION OF

INKA POWER

Edited by Richard L. Burger, Craig Morris, and Ramiro Matos Mendieta

Variations in the Expression of Inka Power
A Symposium at Dumbarton Oaks
18 and 19 October 1997

Richard L. Burger, Craig Morris, and Ramiro Matos Mendieta, Editors

Joanne Pillsbury and Jeffrey Quilter, General Editors

Published by Dumbarton Oaks Research Library and Collection, Washington, D.C.

Distributed by Harvard University Press, 2007
Toward a Reconceptualization of the Late Horizon
and the Inka Period: Perspectives from Cochasquí,
Ecuador, and Samaipata, Bolivia

Albert Meyers

INKA ARCHAEOLOGY, LIKE THAT OF THE OTHER GREAT EMPIRES OF LATE

Pre-Columbian America, suffers from the paradoxical situation that on
the one hand we possess much more data than for earlier cultures and, on
the other hand, this wealth of data has led us to neglect the study of mate-
rial culture and its interpretation.1 One reason is that archaeology has
become an addendum to the interpretation of the written sources. This
is problematic when dealing with nonliterate societies. In Mexico, where
some original native textual documents exist, archaeological evidence has
not only complemented, but also corrected ethnohistorical interpreta-
tions. One of the principal postulates is the strict separation of the anal-
ysis of the material remains from that of the textual material, combining
them only when comparing the results (Meyers 1998a [1976]; Smith 1992;
Pärsinnen and Siiriäinen 1997: 256).
Another misunderstanding in the treatment of the two different
source materials is the tendency to verify through archaeology the
historical data, such as the change of rulers, the conquest of the prov-
inces, or the exchange of people throughout the empire. Such events are
rarely reflected in the archaeological material and, in general, as has been
stated (Sinopoli 1994), an archaeology of empire that concentrates too
much on military conquests will run the risk of overlooking other factors
such as exchange relations. Consequently, the distribution of imperial
ceramic styles in time and space may not correlate exactly with military
or political expansion.
In this chapter, I present two different loci of power of the Inka state,
one in the northern and one in the southeastern periphery. In order to
evaluate the style of Inka power, I shall briefly refer to the archaeology
of Cusco. After a discussion of some manifestations of Inka power in the
two cases, I shall draw some conclusions in terms of different superim-
posed religious power systems and try to interpret the evidence within the
framework of an archaeological chronology for the Inka culture.

Cusco Archaeology: The “Bottling Up” of the Inka Style

In 1993, John Hyslop (1993: 351) posed a classic question: How did the
Inka do so much in such a short time? The standard version of Inka history
for more than sixty years, is that roughly in the decade between 1460 and
1470 all major provinces between Ecuador and Chile and Argentina were
conquered (e.g., Rowe 1945). Here is not the place to discuss the method-
ological problems of this short version: the tendency of the informants
to attribute the great deeds to their closest relatives; the logical, logistical,
and organizational problems of such a rapid expansion and stabilization of
the empire (see Meyers 1975, 1998a [1976]: 31–42). Rather, I maintain that
an archaeological treatment of this phenomenon, separate from a study
of colonial textual sources, can lead to different conclusions. From the
perspective of art history, the supposed creation of such a sophisticated
and stylistically well-balanced style as the Cusco Inka style can rarely be
developed by one ruler and promoted in such a short time. This seems
even more obvious if we consider the limitations of the proposed anteced-
ents that have more in common with contemporaneous neighboring styles
than with the Imperial Inka style. While I cannot agree with Hyslop’s
acceptance of the short chronology, I find more value in his statement
that the “Inka horizon must be viewed as the mobilization of pan-Andean
resources, not just Inka ones from Cusco” (Hyslop 1993: 351).
Whereas this statement has been applied to the imperial expansion
only, it can be expanded to include the initial phase of the empire, which
would begin earlier than the proposed date of 1438, as well as the forma-
tive phase of the Inka style. Geographically speaking, isolated develop-
ment is unlikely not only for theoretical reasons, but also because of the
various parallels that can be found in surrounding areas close to Cusco
(Bauer 1992a, b) and more to the south, as Max Uhle (1912) and Philip
A. Means (1931: 199) suggested long ago. Possible candidates to “compete”
with Killke as precedent of the Inka style are Churajón, Chiribaya, San
Miguel, Gentilar, and others in the southwest, and also the altiplano style
called Kolla, and the Mollo style, east of Lake Titicaca (Ponce 1957). It
is this latter style that shows the closest stylistic connection between
Tiwanaku and Inka. This position represents a serious alternative to the
“bottling up” of the Inka style in the Cusco area.
The same argument could be made for the architecture. Despite all
differences, there exist clear reasons for postulating a stylistic line from
Tiwanaku to the Inka, probably via the chullpa (burial structures) complex

224 Albert Meyers

or the Mollo architecture as documented, for example, in Iskanwaya (Arel-
lano 1985).

The Inka in the Province: The Style of Power, the Power of Style
It can be assumed that the Inka state exercised its powers in a variety of
manifestations. What we know about the ruling principles shaping polit-
ical, economic, social, and religious reorganization is primarily based on
written evidence from the Spanish Colonists. Also, ethnohistorically
defined principles of state dominance such as ethnic relocations and their
replacement by Cusco people or Inkaized groups, not to mention general
models such as economic complementarity (e.g., Hyslop 1993: 348), have
not been convincingly verified by the archaeological evidence. The same
is true of the military expansion of the Cusco people, headed by certain
kings and generals, as assumed by the “standard version.” No Inka fortress,
civil or religious construction element, or any material object, especially
in the provinces, can yet be convincingly attributed to a specific Inka
emperor such as Tupac Yupanqui or Huayna Capac.
What we have is a general set of elements such as fortifications, storage
and road systems, “Imperial Inka” architecture, and transportable cultural
remains, whose careful examination over the last decades has advanced our
knowledge considerably and has changed our view from a rather standard-
ized to a much more differentiated one. By classifying all these features as
manifestations of state power, however, we again run the risk of oversim-
plifying. What we can learn from the ethnohistorical record is that the
Inka style of power was apparently a flexible one, with ruling principles
changing from region to region. Consequently, one of the tasks of the
archaeologist is to consider every element of the material culture sepa-
rately in terms of style, geographical distribution, and chronology before
drawing upon generalizations mostly based on ethnohistorical concepts.
In the case of ceramic styles, I would prefer not to speak of a symbol
of state power (Morris 1995) or even that of a single ruler (Rowe 1996),
but reformulate the problem as one of the power of the style itself. Given
the high mobility of all sorts of goods in the Andes (staple and wealth),
it would not be surprising to find the expansion and imposition of the
Imperial Inka style around the Andes unsupported by military conquest
or other sorts of political pressure. And, from another perspective, why
should a Cusco Coya not prefer fine Chimú pottery in her household?
Were there official prescriptions determining the forms and decoration of
Inka ceramics and were they controlled by the state? Does the evidence
of the existence of state potters in some places mean that there was no
production for the “free market”?
Given a degree of self-regulation in the field of production and
distribution, variety of use in functional sense, availability, and prestige

Toward a Reconceptualization of the Late Horizon and the Inka Period 225
 Inka Sites
and fashion of certain wares, it is difficult to
Modern Cities distinguish between Cusco Inka pottery, state
Areas with Caranqui pottery, provincial pottery, and Inka pottery as
Inka Materials
Cochasquí Morris (1995: 426) has suggested. Drawing an
Quito analogy to Roman provincial archaeology, in an
earlier article (Meyers 1975), I underlined the
widespread phenomenon of a clear Inka influ-
ence on local wares, especially of domestic use.
I called a second group “Inka imitation ware,” a
more or less crude imitation of the most typical
forms and motives. The group of “mixed styles”
is the privileged field for the study of power
Guayaquil balance between all the elements of both styles
Cashaloma Ingapirca (Meyers 1975). Their existence throughout the
Inka Empire, not only in Peruvian territory, is
Cuenca
Pacific evidence of the free contact between styles and
Ocean probably also potters, and an argument against
state regulation.
0 100 km
CASE I: THE RAMP PYRAMIDS FROM COCHASQUÍ
Inka culture in Ecuador is represented by a
FIG. 1 Distribution of variety of features, including the different ceramic styles (influenced,
imperial Inka materials in imitation, and mixed) mentioned above particularly in the Cashaloma-
Ecuador (drawing by Karen
Rasmussen, Archeographics, Inka complex in the southern highlands (Fresco 1984). The distribution
after Meyers 1976: map 2) map of several elements of the Imperial Inka styles shows ten geograph-
ical units of findings, with a slight decrease in intensity toward the north
(fig. 1). In contrast to Hyslop’s interpretation (1993, citing Salomon
and Fresco), however, the Inka frontier north of Quito clearly demon-
strates a strong stylistic influence on local ceramics, as I have demon-
strated elsewhere, based on an analysis of the ceramics from Cochasquí
(Meyers 1989). Apart from the well-documented concentration of Inka
fortresses in that area, a thin-section analysis by Tamara Bray proved that
Inka-style pottery from this region is indistinguishable from the local
Caranqui ware (Bray n.d.: 140, 1992: 227), suggesting that local pottery
was produced along Inka lines. In the ceramics of Cochasquí’s second
phase, Inka influence is evidenced by the use of black as an additional
color, and by a characteristic rim treatment (Meyers 1989).2 Among the
whole vessels found at Cochasquí, the prevailing vessel shapes are Form 1
(“aryballus”), mostly in imitation style; Form 10 (foot bowl); and cooking
pots (ollas) with tripod legs instead of the common ring base.3 The most
curious specimen is the “fern pattern” on an Inka-Form 1-imitation vessel
applied using the technique of resist painting (Oberem and Wurster 1989:
figs. 96, 97).

226 Albert Meyers

 Cochasquí, in the Sierra Norte de Quito, consists of a complex of FIG. 2 View of Pyramid E at

fifteen stepped earthen pyramids with up to 200 m long ramps that lead Cochasquí during the 1964–
65 excavations (photograph
to the platforms (fig. 2). The same cultural and chronological complex by Udo Oberem)
includes a series of burial mounds and a habitation site. Excavations at the
largest construction, Pyramid E, in 1964 and 1965, uncovered the lower
section of the stepped structure, which was made up of walls constructed
of big blocks of volcanic tuff (cangahua) with a pillow-like outside surface,
resembling Cusco Inka stone walls (fig. 3). This same observation was
made by Max Uhle in the 1930s, when he watched the work of grave
robbers at the same building. He interpreted this huge building as an Inka
sun temple (Uhle 1933).
The platform of Pyramid E as well as those of other similar excavated
structures was consolidated with brick-hard burned clay. These structures
contained a series of postholes that, together with small circular ditches,
suggested the existence of round houses with conical roofs (fig. 4). Rec-
tangular cavities up to 6 m long and 50 cm wide molded in clay were posi-
tioned symmetrically in relation to the central axis (fig. 5). Every cavity
formed a rectangular area into which another rectangular area had been
deepened in a concentric way with the sides a little bit inclined. On the
platform of Pyramid E the excavators found a series of stone cones stuck
in the cavities, always in groups of three, forming a triangle that could hold
a cooking pot. The excavators interpreted the pyramids with the house

Toward a Reconceptualization of the Late Horizon and the Inka Period 227
FIG. 3 Detail showing the
blocks of cangahua,
Pyramid E (photograph by
Udo Oberem)

FIG. 4 Reconstruction of
Pyramid E at Cochasquí
(after Oberem and Wurster
1989: fi g. 41)

228 Albert Meyers

constructions as seats of the Caranqui lords of
the region, famous for their resistance to the
troops of Huayna Capac. They assumed that
the cavities served as domestic kitchens and did
not consider a ritual function for these features,
although they did not deny the possibility of
other explanations (Wurster 1989: 58–60).
The documentation of an analogous cavity
from the carved rock in Samaipata, eastern
Bolivia, has led me to reconsider the function
of this fairly rare element at Cochasquí. The
feature at Samaipata has a striking similarity to
those of Cochasquí, the major difference being
that it is carved in the northwestern corner of
the huge sandstone rock and is among hundreds of carved elements at that FIG. 5 Cavities on the

site (fig. 6). It lacks the stone cones, but it has the same dimensions (6 m platform of Pyramid E
at Cochasquí (photograph
long and 50 cm wide), running in an east-west direction. There is sufficient by Udo Oberem)
evidence to assume that they are the products of a common culture, while FIG. 6 Rectangular cavity
not necessarily assuming a contemporaneous origin for these features near feline relief on the
encountered across such a vast geographical distance. These have not yet carved rock at Samaipata

been documented in other literature, so far as I know. My suggestion is

that they served as recipients for liquids during libation and/or burning

Toward a Reconceptualization of the Late Horizon and the Inka Period 229
FIG. 7
Lake
Location of Samaipata and Titicaca BOLIVIA
other sites in the eastern
Andes (drawing by Karen Low
La Paz l an
Rasmussen, Archeographics, ds
after Meyers and Ulbert
1998b)
PERU Cochabamba
opposite page Santa Cruz
FIG. 8 View of Samaipata
Oruro Samaipata

Al
from the southeast, with

ti
pla
the carved rock outcrop CHILE

no
visible above the semitropi-
cal vegetation. Sucre

FIG. 9 Sector 3 (Spanish

Potosí
House) of Samaipata, after Pacific
consolidation Ocean
Coa
stal
inPla

Tarija

Inka Site
ARGENTINA
Modern Cities

rituals. In contrast to Cochasquí, the carved rock of Samaipata can be

dated to the Late Horizon. In the case of the Cochasquí cavities, my expla-
nation would only be plausible if the house was burned and the cavities
added after that event. This is technically possible. I therefore hypothesize
a change at Cochasquí of the civil and perhaps ritual residence of the local
lord into a place for the execution of state rituals. Practicing Inka public
rituals on top of a former native elite residence not only provides a clear
demonstration of power but also may have been intended as humiliation
or punishment of those who resisted conquest by the Inka.

CASE II: THE CARVED ROCK FROM SAMAIPATA

The so-called Samaipata Fort (“Fuerte de Samaipata”) is situated on one
of the last flanks of the eastern Bolivian Andes, at an altitude of about
1,900 m, a little more than 100 km to the southwest of the modern city of
Santa Cruz de la Sierra (figs. 7, 8). The site was visited by Thadäus Haenke
at the end of the eighteenth century and described by the French trav-
eler Alcide d’Orbigny some forty years later (d’Orbigny 1835–1847). His
interpretation as a “lavage d’or des Incas” was rejected at the beginning
of the twentieth century by the Swedish scholar Erland Nordenskiöld,
who considered it the last outpost of the Inka Empire toward the eastern
lowlands. He also placed more emphasis on its ceremonial character
(Nordenskiöld 1924), as did Hermann Trimborn (1967).

230 Albert Meyers

Toward a Reconceptualization of the Late Horizon and the Inka Period 231
 1 2 3 4 5 6 7 8 9 10

1825
S1 S7
A

1875
S6
S16 S9
S19
B S20 S3
Plateforme 1
1 92
5 S5
S4 S15
S14 2
rme
tefo
C Pla

S10 S2 S8
S12
S13
D
Kallanka
S17 S18
S11

E
1875
0
185
F

G
N

FIG. 10 Plan of the Since then, Samaipata has been generally interpreted as one of the
archaeological complex great fortresses that protected the eastern flanks of the empire against
of Samaipata indicating
the excavation sectors the continuous attacks of the savage forest peoples (“Chunchos salvajes”)
(drawing by Peter Pahlen, of the lowlands. An examination of the ethnohistorical material reveals,
Cornelius Ulbert, and however, that these writings have more to do with a projection of the
Rolando Marulanda)
various armed conflicts between the Chiriguanos and the Spaniards during
the viceroyalty of Toledo ( Julien 1997), than a rejection of the time of the
Inka. This vision is still manifested today by the common denomination
of the site as “El Fuerte de Samaipata.” Indeed, the Arabic-Andalusian
style “patio-house” situated on top of the Inka ruins, almost attached to
the carved rock (fig. 9), can be interpreted as the throne/headquarters of
a Spanish captain from the time of these conflicts (Meyers n.d.). Further-
more, we identified at least one Inka fortress some 60 km southeast of
Samaipata (Meyers and Ulbert 1998a, b), which would confirm the recent
suggestions of a more eastern expansion of the Inka Empire in that region
(Pärsinnen 1992; Saignes 1985).
D’Orbigny’s map indicates a “village des Incas” to the south of the
famous carved rock, and during our excavations between 1992 and 1994
we documented an extended complex of terraces, structures, and build-
ings of various sizes at the site (fig. 10; Meyers 1993, 1998b; Meyers and
Ulbert 1998a, b).4 The great hall (kallanka), with an extension of 68 m x

232 Albert Meyers

 FIG. 11 Imperial Inka
ceramics from the
excavations in Sector
2 at Samaipata
FIG. 12 Ceramics of
low-land origin (known
locally as Chiriguano
ware) found between the
two Inka occupations in
Sector 2 at Samaipata

16 m, that flanks the great plaza to the south is described by my colleague

and codirector of the last field season, María de los Angeles Muñoz (see
her chapter, this volume).
In four different excavation sectors, we documented two construction
phases of Inka buildings, the later one evidencing two occupation floors
separated by layers of destruction. This is not the place to describe the
occupation sequence in detail or the diversity of the material recovered.
The pottery assemblage includes at least three ceramic wares of Inka style
(fig. 11) as well as local wares (fig. 12). The local wares show an affinity with
a stylistic complex distributed along the eastern flank of the Andes and
with wares found in the Amazonian region to the north and the Chaco

Toward a Reconceptualization of the Late Horizon and the Inka Period 233
FIG. 13View of the carved
rock outcrop at Samaipata,
from the west

N
0 25 m

234 Albert Meyers

region to the south. Rather than focus on these artifactual remains, I want
to consider the large rock-carving complex that reveals a heterogeneous
combination of diverse motifs, multiple processes of carving, and stylis-
tically different features, which could serve as a unique “quarry” of Pre-
Columbian Andean and perhaps non-Andean ritual complexes (fig. 13).
The carved rock of Samaipata consists of a huge sandstone ridge with
a 250 m east-west and 50 m north-south extension. It is covered by figu-
rative representations of felines, serpents, and geometric figures such as
rhombuses, meanders, triangles, rectangles, and circles (fig. 14). Several
carved basins are distributed around the rock. A special unit appears to be
formed by the steps and rectangular seats ordered in rows on the southern
part (fig. 15). These give the whole complex the appearance of a sort of
amphitheater. Other features on the top of the rock have been interpreted
and named in the earlier literature, especially in the work of Leo Pucher
(1945). For example, he interpreted the grooves of rhomboids paralleled
by two canals (fig. 16) as the representation of a serpent (el dorso de serpi-
ente). He called the complex of seats, carved in the form of a circle, “the
choir of the priests” (el coro de los sacerdotes). Five niches were carved on
the northern flank (fig. 17), part of a large temple building. Portions of the
wall are still in situ (fig. 18). While not considering those elements clearly

FIG. 14 Plan of the carved

rock at Samaipata; arrows
indicate super positions
(drawing by Peter Pahlen,
Cornelius Ulbert, and
Rolando Marulanda)

Toward a Reconceptualization of the Late Horizon and the Inka Period 235
15 16

17

236 Albert Meyers

 opposite page
added in a later phase of utilization and construction, I want to point out
FIG. 15Aerial view of the
two cases of a clear superposition of Inka elements whose interpretation southern portion of the
could not only provide a new vision of Inka presence in the provinces, but rock outcrop at Samaipata
also shed new light on Inka development in the core area. FIG. 16 Aerial view of the
two canals on the rock
outcrop at Samaipata
Two Religious Systems Superimposed in Samaipata?
FIG. 17 Aerial view of
In this section, we are concerned with two complexes of worked sandstone the temple structure with
walls erected on top of the rock at Samaipata. On the western side, there niches at Samaipata, from
is a transversal wall oriented in north-south direction and still conserved the northwest

to a height of 80 cm (fig. 10, S16; see also fig. 13). It contains three double- this page
FIG. 18 Detail of the temple
jambed niches facing west and three single-jambed niches facing east
structure, with stone wall
toward the two long carved canals. Along its southern part, it cuts a 40 cm in foreground
deep basin into two parts. The other complex lies at the eastern end of the
carved rock and consists of two walls positioned in the form of an L (fig.
10, S9; see also fig. 14, arrows). The transversal wall contains six double-
jambed niches arranged in a meandering form whereby three niches face
alternately to the east and the west. It is separated by a small passage from
the longitudinal wall, which has seven double-jambed niches facing south
and six single-jambed niches facing north. The eastern end enters into
a semicircle carved into the rock, which contains several triangular and
quadrangular seats. The wall ends in front of one of these seats, leaving the
semicircle structure dysfunctional.

Toward a Reconceptualization of the Late Horizon and the Inka Period 237
FIG. 19 Carved stone In both of the above cases, we are dealing not with buildings, but
outcrop (Th rone of the rather with niched galleries, which were superimposed on top of the rock
Inka) on Rodadero Hill,
Sacsayhuaman (photograph carvings. The western gallery was probably connected to the two canals to
© Edward Ranney) form a ritual complex, especially considering that the canals are inclined
FIG. 20 The upper stone toward the gallery. On the eastern end of the canals there is a basin posi-
of Sahuite (photograph tioned asymmetrically, which seems to be rendered dysfunctional in the
© Edward Ranney)
same way as the eastern niche gallery.
What has attracted our attention at Samaipata is the apparent nonin-
tegration of the previous system into the new ritual complex. Obviously,
we have two different ritual complexes. Moreover, we interpret the two
systems as opposed to each other, with the later one not respecting and
perhaps even explicitly rejecting the earlier one. This impression could
be extended to the majority of the other carvings, many of which have
been interrupted and dismembered. For example, transversal staircases or
niches are carved at the edges of the rock. These belonged to large roofed
buildings built on the flank of the carved monument, whose reconstruc-
tion by computer simulation reveals that the elements of the carved rocks
could no longer be seen from the adjacent platform or from the grand
plaza with the kallanka down to the south. Thus, superposition of the late
complex on top of the former one can be interpreted as a demonstration
of power similar to that at Cochasquí at the other extreme of the empire.
Several conclusions can be drawn from the evidence. Three of them
are presented here as hypotheses. Because of the methodological consid-
erations explained above, only archaeological arguments based primarily
on the traditional methods of stylistic comparison and stratigraphical
superposition will be employed.

HYPOTHESIS 1: ALL ELEMENTS ARE OF INKA ORIGIN

All the complexes described at Samaipata are hypothesized to be of Inka
origin, and the stylistic differences and superpositions related to different
phases of occupation and use by the Inka. The style and technique of
rock carving and of almost all the structural elements have parallels in
the Cusco region. Analogues can be drawn to various carved boulders
and rock outcrops, including the carved steps of Rodadero Hill at Sacsay-
huaman (fig. 19) and the rock of Sahuite in Apurimac (fig. 20). These, and
many other cases of such features, have until now mostly been interpreted
as Inkaic in the archaeological literature. Consequently, at Samaipata, we
have to define two stylistically and stratigraphically separate complexes,
both created by the Inka.
In the phase Inka I, the rock would have been carved with canals,
seats, and steps, as an aesthetic expression of Inka masonry art and for
ritual use in making liquid and/or burnt offerings or other ceremonies.
In the phase Inka II, the rock would have been cut at the edges to add

238 Albert Meyers

Toward a Reconceptualization of the Late Horizon and the Inka Period 239
the niched temples. The niched galleries were likely constructed at the
same time, along with other elements such as the two canals. Cult activi-
ties would have been concentrated on the niches, which are carved on
a smaller, more human scale. The result is a fundamental change in the
orientation of the two complexes. The first is directed against the rock
or toward the earth. This change is reflected at the level of action as well
as at the site of contemplation. In the first phase, as seen from above, the
rock appears as a giant sculpture of the earth, a sort of “sculptarchitec-
ture” (Paternosto 1996: 60–61 and passim), so the whole rock looks like
a single great temple reminiscent of other architectonic examples such as
Malinalco in Mexico or Petra in Jordan.
The second ritual complex is oriented in another direction, no longer
toward the earth but rather toward the horizon or even the sky. In some
sense, it cuts off the vision to the earth, which is now covered by temples
and galleries. Within these galleries, figurines, mummies, or other vener-
able things might have been displayed. Like the stones of the galleries,
they represent transportable elements. The whole cult in the later phase
appears to be one of portable things. Even the ancestors are transportable
in the form of mummies and the Sun in the form of statues. Of course,
mythology is transferable and can be symbolized in the form of the various
groups of four double-jambed niches or little “windows,” which appear on
several parts of the rock. These appear to have symbolized the Inka origin
myth of the four Ayar brothers and sisters.
The ritual system underlying these architectural elements does not
seem to be concerned with, or respectful of, the static earth elements. On
the contrary, it appears as a system of power, of subordination, and even
of disdain. Some of the symbols of the previous ritual system can be imag-
ined not just as ceasing to function, but as having disappeared. Several
of the basins, for example, had been filled with sand so as not to disturb
the new religious system in its symmetry and presentation to the subju-
gated people. This ritual perspective, cut off from the earth and directed
to the horizon and to the peaks of the mountains, was reinforced, trans-
forming the former platform steps into pillars that could serve as visual
lines (ceques) that delineate the landscape to the horizon and even the sky,
as a way of establishing possession. This is the view of a conquest society.
It transforms the religious symbols and places of the conquered society
into transportable state-owned goods applicable and redistributable to
dissimilar conquered societies throughout the Andes.
My interpretation of this conjunction of elements is not that of an
integration, as has been commonly suggested by those relying on written
sources, but that of a clear superposition, with demonstrable public effects
like the introduction of new transportable symbols to replace the existing
chthonic elements. In the same way, the increasingly studied artificial

240 Albert Meyers

 altars (ushnus), sometimes built around or on top of a natural rock
outcrop, can be seen not only as symbols of power and conquest but also
as the incorporation of military and ritual elements into the complex of
state religion. They can be observed especially on the periphery of the
entire empire from Rumicucho to the north of Quito (Almeida 1984)
through La Fortaleza, some 60 km east of Samaipata (Meyers and Ulbert
1998a), through El Shincal in Argentina (Raffino et al. 1998), and south
of modern Santiago de Chile (Stehberg 1995).
Several consequences result from these considerations. Assuming
that the two systems are manifestations of the same Inkaic culture, we
calculate that a considerable amount of time would be necessary for the
change from the one to the other. The history of religion tells us that in
archaic societies, no religious system changes from one day to another.
Myths, like the dream of a founder of a religion who claims to have
received his orders from god (“sunturwasi legend”), can be interpreted
as phenomena of etiology, that is to say, an explanation “ex post facto.”
Generally, they serve to legitimate a new cult created by the new state
elite or its leader.
A search for archaeological explanations of these changes will hardly
find local or closely related comparative elements, even in the eastern
Bolivian Andes. One has to extend the area of research past Lake Titicaca
to the Cusco Valley, as did other students of the carved rock of Samaipata,
such as Nordenskiöld and Uhle (1910) early in the twentieth century.
There is no doubt that the biggest concentration of elements of both reli-
gious systems exists in Cusco and its neighboring regions (Hemming and
Ranney 1990). Considering this as the core region for the development of
the Inka style, one should also assume a more or less “organic transition”
from one system to the other. Again, here we are concerned with meth-
odological problems and those of the presentation and interpretation of
concrete and comparable data from Cusco.
Beginning with the most important transportable material, ceramics,
we have to deal with the old and well-known “Killke problem,” which has
been treated in a rather dogmatic way since its first definition more than
sixty years ago (Rowe 1944). This problem can only be resolved by more
concrete archaeological work. It has been stated repeatedly that Killke
ceramic production persisted through the Inka period (e.g., Lunt 1988:
494, 498) and there seems to be evidence of various contemporaneous
“Killke-related” styles in the same region (Bauer 1992b; Rivera Dorado
1977). However, the problem in considering these new archaeological
data is that the authors’ interpretations are guided too much by historical
preconception. Sometimes, a separate archaeological argument is prom-
ised and in the final discussion questionable historic data again are taken
into consideration. However, one of the results in the field of chronology

Toward a Reconceptualization of the Late Horizon and the Inka Period 241
 is the modification of the “master-sequence” expanding the
Inka to a period starting from 1400 AD (Bauer 1992a).
We also have various hints of the superposition of
construction phases and probably construction styles, as has
been argued even for Machu Picchu (Kendall 1988: 474).5 A
very serious problem, however, is represented by the mass of
carved rocks and their association with standing architecture.
It is difficult to establish a method to identify sequences or
even modifications in these cases. But if we depart from the
stark vision of a single, short phase of construction and distri-
bution of Inka elements in the Cusco region, we become aware
of the sequential processes that can be presumed for sites like
Sacsahuaman, Ollantaytambo, Vilcashuaman, and, of course,
Machu Picchu.6 For the Andean flanks east of Cusco, we have
documented several examples of boulders enclosed by walls,
which are comparable to Samaipata (Fejos 1944).

HYPOTHESIS 2: MANIFESTATIONS OF TWO CULTURE COMPLEXES:

INKA AND PRE-INKA
A second and alternative explanation would ascribe the
rock carvings to two cultures: 1) the Inka and 2) pre-Inkaic
cultures, whose origin could be the local Andean culture or
even that of the eastern lowlands. Evidence of cult stones
from the latter regions exists, especially in or near the rivers
FIG. 21 Torso of a (Riester 1981). The presence of lowland ceramics at Samaipata (figs. 12,
zoomorphic ceramic 21) demonstrates that the site might have been attractive to people from
vessel found in a looter’s
pit, Sector 4, Samaipata the “Oriente” or eastern lowlands. Among the authors who postulate a
pre-Inkaic origin for Samaipata (Rivera Sundt 1979: 108), however, aside
from the mention of the mythic importance of animals such as the jaguar
and the boa constrictor (e.g., Pucher 1945), there have been no concrete
elements to confirm such views. Nevertheless, the region of Moxos will
have to be taken into account in future studies of this kind. The eastern
Bolivian valley region was also a considerable source of the late ceramics
recovered at Samaipata, but these have been too little studied to make
serious comparisons.
So again we return to the nuclear region of the development of the
Inka style. The stepped stones of Rodadero Hill in Cusco, the so-called
Throne of the Inka from Intincala, Copacabana, Lake Titicaca (Paternosto
1996: 98, 99), and the stepped structures of the carved rock of Samaipata
have a common stylistic and probably a common religious background.
However, there seems to exist a closer relationship to the stepped stone
carvings of the Pumapunku complex in Tiwanaku than to what has been
specified as Inka architecture until now. Also, similar structures in Cusco,

242 Albert Meyers

Sacsahuaman, Ollantaytambo, and even the stepped stone of Machu
Picchu and other elements in that area show this same pattern. To draw
the conclusion that we have to concern ourselves with the Tiwanaku
complex would not only be premature, but also inadequate. In the more
or less 3000 m2 excavated in Samaipata, we did not find a single ceramic
sherd of the classic Tiwanaku style.
In any case, our understanding of the Tiwanaku culture complex in
terms of the architecture as well as the ceramics suffers from the same
general problems as in the Inka case. For example, we still do
not have a convincing subdivision of the Tiwanaku style,
which is supposed to have lasted more than a millennium
in the altiplano, nor a definition of its patterns of expansion.
What we can assume is that in the Cusco to Titicaca region
there was a “creative climate” to forge a mixture of Wari
and Tiwanaku styles that resulted in a new stylistic
concept. It is my opinion that future studies of
Inka origin that concentrate on the stylistic lines
from these cultural complexes to the Imperial
Inka style, via possible mediator styles, such as
Churajón (Szykulski n.d.) and Mollo (fig. 22), will
be very promising.

HYPOTHESIS 3: THREE OR MORE PHASES OF ROCK CARVINGS

A third explanation derives from the evidence of at least three
phases of origin and use of the sacred complex of Samaipata. Given the FIG. 22 Mollo-style ceramic

fact that the veneration of rocks and working of stones is an old phenom- vessel combining late
Andean and eastern
enon in South America, beginning with Chavín, including San Agustín elements (Museo de Metales
in Colombia, and ending with various cultural manifestations even in the Preciosos, Colección Fritz
lowland areas, one could conclude that the huge boulder of Samaipata Buck, La Paz)

could have played a role as a sacred hill (loma santa) since remote times
(Riester 1976). However, at this moment we lack specific evidence for
early rock carving, although we have found Formative-period ceramics,
probably of lowland origin, in the excavations at Samaipata.
The second phase would be represented by the steps, seats, and similar
carvings. Its most important characteristic would be the conversion of
the rock into a giant “archisculpture.” Whether we call it early Inka or pre-
Inka is a question of definition of what is Inka in the core area. The third
phase would include the niched elements, especially the niched temples
and galleries, the libation canals, and probably also the cavity similar to
those of Cochasquí. The Inka use of the rock seems to correspond to the
two construction phases defined for the standing stone architecture in the
excavations. Finally, we could add a fourth phase, which consists of the
reutilization of the Inka houses after their destruction by the Chiriguanos

Toward a Reconceptualization of the Late Horizon and the Inka Period 243
and/or the Spaniards, and also the offerings of ceramics and seeds found in
the niched galleries that date from the time of the Spanish Viceroyalty.

Toward a Reconceptualization of the Late Horizon and Inka Periods

The problems with the so-called “master-sequence” established for the
Ica Valley by John Howland Rowe, when applied in the broader context
of Peruvian archaeology, have already been alluded to in this chapter. But,
given the lack of a convincing alternative to this chronological framework,
it seems reasonable to modify the existing model rather than to reject it. In
the south central Andes, especially in northern Chile and in Bolivia, there
has been a tendency to adopt this framework, at least when referring to
the Late Intermediate period and the Late Horizon.
The most important criterion for its application in these regions seems
to be the existence of a Middle Horizon, here represented by the Tiwa-
naku style. As Tiwanaku is believed to have lasted longer than Wari, the
period between the Middle and Late Horizon would be reduced consider-
ably. The situation becomes more complicated if we consider two other
Horizon styles mentioned in the literature, the Horizonte Tricolor del Sur
(Ibarra and Querejazu 1986) and the so-called Black-on-Red Horizon (e.g.,
Pärsinnen and Siiriäinen 1997: 265). The first of these styles developed
earlier than Tiwanaku and points to an independent formative cultural
development in the southeastern Bolivian and the northwestern Argen-
tinian Andes. The second style is a post-Tiwanaku phenomenon, which
belongs to the complex out of which the Imperial Inka style must have
originated. However, it must be clear that the application of the “master-
sequence” in Bolivia only makes sense in terms of a mere periodization
scheme that subsumes the different stages and periods of development in
that geographical area. The start of the Late Horizon (LH) is understood
not as the beginning of the military expansion, but as the spread of Inka
cultural material throughout the later empire.
The problem of defining the Late Horizon in the core area is that we
are still ignoring the origin of the Inka style—in cultural, temporal, and
geographical terms. Until now, as has been stated, almost all scholars have
searched for the origin of Inka culture in the Cusco region. Even though
we have two Inka origin myths, archaeologists as well as ethnohistorians
take into consideration only the “Paqaritambo version.” No archaeologist
until now has seriously examined the “Titicaca version” and considered
possible Inka precursor styles in that area. Stylistically, in the same way
Killke or Killke-related styles are called Early Inka, one could postulate
that the late circum-Titicaca styles belonged to an early Inka ceramic
complex. The “advantage” of Killke, of course, is its geographical closeness
to the later Inka capital. But no clear transition from Killke to Cusco Inka
and its expansion from Cusco to the Lake Titicaca area has been proven.

244 Albert Meyers

So there is still room for arguments for the development of Inka style in
the Lake Titicaca area and its expansion to the Cusco Valley. Similarly, one
could look for other foci in the “Inka core area.”
My conclusions based on these considerations imply a two-phase
vision of the development of Inka culture:
1) As discussed, I refer to the last two Pre-Hispanic cultural mani-
festations as Inka I and Inka II. Inka I is a cultural complex distributed
throughout the core area and characterized by common cultural patterns
inherited from the preceding Wari/Tiwanaku culture horizon. While
there may not have been political unification, the cultural expressions
demonstrate great similarity as exemplified by the stepped rock carvings,
pottery with similar shapes and decorative patterns, and probably also a
common domestic architecture. Inka II is a highly standardized cultural
pattern that, up to now, is evidenced mainly in the public sphere, particu-
larly in architecture, and in the other fancy art styles subsumed under the
denomination of Imperial Inka.
While the appearance of Inka I is rather rural and noncentralized,
devoting attention to the protection of the agricultural fields against
bellicose neighbors, the Inka II complex is that of an organization with a
highly developed public culture and technology and perhaps a stronger
degree of political unity. Judging by the lack of Imperial Inka fortresses
in the core area, one can assume that the process of state formation had
been finished.
2) This leads us to the temporal question. I calculated a time span of
200 years for the Inka I period beginning in 1100 AD, the estimated date
of the collapse of the center of Tiwanaku. As the Wari influence in Cusco
is thought to have ceased about 200 years earlier, a middle date of 1000
AD for the beginning of the Late Intermediate period for the whole area
could be acceptable (table 1). In the absence of reliable historical data for
the military expansion of the empire during the Late Horizon, I correlate
its beginning with the expansion of Inka Imperial style through military
expansion, trade, or other exchanges. The problem lies in the definition of
the area from which the expansion begins. In our case, where we assume a
very large core area, we have to estimate a considerable time for the distri-
bution of the style from one or multiple manufacturing centers and for the
consolidation of the Horizon style in the entire area. This process would

TABLE 1 Chronology of the Late Horizon (LH), the Late Intermediate period (LIP) and
the Inka periods
1550 AD–1600 AD: Inka IIc LH 3 (colonial Inka)
1400 AD–1550 AD: Inka IIb LH 2 (Cusco Inka, outside the core area)
1300 AD–1400 AD: Inka IIa LH 1 (Cusco Inka, in the core area)
1000 AD–1300 AD: Inka I LIP (Formative Inka, in the core area)

Toward a Reconceptualization of the Late Horizon and the Inka Period 245
represent the first phase of Late Horizon (LH 1) for which I calculate
100 years. Recent radiocarbon dates from sites at the periphery of the
core area, that is to say, northern Chile, northwest Argentina, and Bolivia
(altiplano and eastern slopes) (Adamska and Michczynski 1996) suggest
that by about 1400 AD the Imperial Inka style had already spread beyond
the frontiers of the core area. We accept this date for the beginning of LH
2, which does not end with the conquest by the Spanish but, depending
on the remoteness of the area, probably some decades later. In some areas,
as is well known and documented, the Imperial Inka style survived for
a considerable time after the Spanish Conquest, integrating some Inka
forms and decorative elements with certain techniques of the conquest
culture. Consequently, it seems suitable to define a third phase (LH 3),
which would last from approximately 1550 to 1600 AD.
In summary, we have a 300-year span for the Imperial Inka style,
which, in comparison with the earlier elaborate art styles prior to the
Inkas, is not very long. Of course, the styles’ further development was
brutally interrupted by the Conquest, but there is reason to believe that
there was considerable time for the “maturing” of such a style. It was these
and similar considerations that led me to reject the “short version” of
Inka chronology more than thirty years ago (Meyers 1975). The recently
presented C 14 measurements shed new light on the question, but have
to be considered from a methodological perspective. It is my hope that
future excavations in the Cusco and Titicaca areas, as well as on the
periphery, will deliver stratigraphic and other evidence to provide a more
differentiated picture of the development of the Inka style and its various
manifestations through time and space.

Conclusions
In this essay, I have considered the methodological problems of handling
the data with which Inka archaeology is concerned. It was my intention,
perhaps a bit exaggerated in some places, to discuss and object to some
preconceptions and scholarly traditions that have advanced our knowl-
edge in some fields of study, but have been an obstacle to creativity and
the development of alternative interpretations in others. In the case study
of Cochasquí, I presented a different vision of the Inka domination in that
northern periphery, one that was more intensive than generally presumed.
In Samaipata, I identified two cultural-religious complexes, whose inter-
pretation throws new light on the Inka culture in the core area. Many
scholars will disagree with my denomination of the earlier phase as Inka I;
this chronological framework is presented nonetheless in order to stimu-
late discussion and future studies. Its applicability and possible modifica-
tion must be tested empirically. I am aware that the last word in this matter
has not yet been spoken.

246 Albert Meyers

Notes
1 This essay is an attempt to combine some results of recent
research projects with observations on the state of Inka archaeology
dating back as far as the early seventies. Not all of the arguments are
accompanied by citations from the literature, especially when there
is agreement among most of the scholars. I am very grateful to Kerstin
Nowack for the correction of the English text of an earlier version and
for her detailed comments, although we disagreed on many of the
points under discussion. I also thank the students of a seminar on Inka
culture, held in 1996/97 at Bonn University, for their comments and
collaboration. Finally, I would like to thank Joanne Pillsbury for her
encouragement, and especially Grace Morsberger for her patience and
great ability in editing the final version. Rolando Marulanda and Iken
Paap kindly assisted in preparing the map of the carved rock (fig. 14) for
publication. Of course, all shortcomings fall under my responsibility.
Only minor changes to the original text and updates to the bibliography
have been made.
2 The same has been demonstrated for Cayash, Central Peru
(Krzanowski and Tunia 1986). This report is one of the best discussions
of Inka influence on local ceramics written.
3 As a precondition for my doctoral dissertation on the Inka style
in Ecuador in the seventies (Meyers 1976 [1998a]), I had to reconsider
the existing classifications of Inka ceramics, mainly from Bingham
(1915, 1930), Pardo (1938, 1939), and Rowe (1944). As I published the
reasons for my new classification many years ago in Spanish (Meyers
1975), I will simply state here that I relied strictly on material coming
from so-called “closed contexts,” applying form and quantity as my
method of analysis. The fourteen main shapes (leitformen) were orga-
nized into seven form classes, based mainly on the Sacsahuaman mate-
rial (Valcárcel 1934–1935; Valencia 1970), the grave goods from Machu
Picchu (Bingham 1915, 1930; Eaton 1916), and, to a minor extent, from
Ollantaytambo (Llanos 1936). A similar classification of the decoration
and rim profiles of materials from secure Cusco Inka contexts, excluding
all possibilities of an earlier or later intrusion, is still a desideratum.
4 The “Proyecto de Investigaciones Arqueológicas en Samaipata”
(PIAS) was carried out with the permission of and in coordination
with the Instituto Nacional de Arqueología (INAR), La Paz. I am very
grateful for the support of its directors, Oswaldo Rivera and Juan Albar-
racín-Jordán. I also thank Omar Claure, Willy Pantoja, Javier Gonzales,
and María de los Angeles Muñoz, who acted as codirectors during the
field campaigns. The project was supported by the Deutsche Forschungs-
gemeinschaft (DFG), Bonn and by minor grants. Thanks also to VARIG-
Airlines for their assistance in resolving transportation problems.

Toward a Reconceptualization of the Late Horizon and the Inka Period 247
5 Also here we have the intention of dating via historical deduction
(e.g., when Machu Picchu is attributed to the epoch of Pachakuti).
While I am not completely opposed to the method of identifying archi-
tectural structures as belonging to certain Inka lords at the time of the
Spanish Conquest, this method can be a supplement to the main task of
identifying architectural styles and differences.
6 I am convinced that Machu Picchu has a long occupation history,
starting from pre-Inka times, comparable to the situation we found at
Samaipata.

248 Albert Meyers

References cited
ADAMSKA, ANNA, AND RICHARD MICHCZYNSKI
1996 Towards Radiocarbon Chronology of the Inka State. Andes:
Boletín de la Misión Arqueológica Andina 1: 35–58.
ALMEIDA REYES, EDUARDO, AND HOLGUER JARA CHÁVEZ
1984 Investigaciones Arqueológicas en el Pucará de Rumicucho.
In El Pucará de Rumicucho (E. Almeida and H. Jara, eds.): 9–27.
Miscelánea Antropológica Ecuatoriana. Serie Monográfica 1.
Museos del Banco Central de Ecuador, Quito.
ARELLANO LÓPEZ, JORGE
1985 Mollo: Investigaciones arqueológicas. Impr. Nacional, La Paz,
Bolivia.
BAUER, BRIAN
1992a The Development of the Inca State. University of Texas Press,
Austin.
1992b Avances en arqueológia andina. Centro de Estudios Region-
ales Andinos “Bartolomé de las Casas,” Cusco.
BINGHAM, HIRAM
1915 Types of Machu Picchu Pottery. American Anthropologist 17
(2): 257–271.
1930 Machu Picchu, a citadel of the Incas: Report of the explorations
and excavations made in 1911, 1912 and 1915 under the auspices
of Yale University and the National Geographic Society. Pub. for
the National Geographic Society, Yale University Press, New
Haven and London.
BRAY, TAMARA
1992 Archaeological Survey in the Northern Highlands of Ecuador:
Inca Imperialism and the Pais Caranqui. World Archaeology 24
(2): 218–233.
n.d. The Effects of Inca Imperialism on the Northern Frontier.
Ph.D. dissertation, Department of Anthropology, State Univer-
sity of New York at Binghamton, New York, 1991.
D’ORBIGNY, ALCIDE
1835–1847 Voyage dans l’Amérique méridionale . . . 6 vols. Pitois-
Levrault, Paris.
EATON, GEORGE
1916 The Collection of Osteological Material from Machu Picchu.
Memoirs of the Connecticut Academy of Arts and Sciences 5.
Tuttle, Morehouse, and Taylor, New Haven, Conn.
FEJOS, PÁL
1944 Archaeological Explorations in the Cordillera of Vilcabamba,
Southeastern Peru. Viking Fund Publications in Anthropology
3. New York, N.Y.

Toward a Reconceptualization of the Late Horizon and the Inka Period 249
FRESCO GONZÁLEZ, ANTONIO
1984 La arqueología de Ingapirca (Ecuador): Costumbres funerarias,
cerámica y otros materiales. Comisión del Castillo de Ingapirca,
Consejo de Gobierno del Museo Arqueológico del Banco
Central del Ecuador, Cuenca, Ecuador.
HEMMING, JOHN, AND EDWARD RANNEY
1990 Monuments of the Incas. University of New Mexico Press,
Albuquerque.
HYSLOP, JOHN
1993 Factors Influencing the Transmission and Distribution of
Inka Cultural Materials throughout Tawantinsuyu. In Latin
American Horizons (Don Stephen Rice, ed.): 337–356.
Dumbarton Oaks Research Library and Collection, Wash-
ington D.C.
IBARRA GRASSO, DICK EDGAR, AND ROY QUEREJAZU LEWIS
1986 30,000 años de prehistoria en Bolivia. Editorial Los Amigos
del Libro, La Paz, Bolivia.
JULIEN, CATHERINE
1997 Colonial Perspectives on the Chiriguana. In Resistencia y
adaptación nativas en las tierras bajas latinoamericanas (Maria S.
Cipolletti, ed.): 17–76. Ediciones Abya-Yala, Quito, Ecuador.
KENDALL, ANN
1988 Inka Planning North of Cuzco between Anta and Machu
Picchu and along the Urubamba Valley. In Recent Studies in Pre-
Columbian Archaeology (Nicholas J. Saunders and Olivier de
Montmollin, eds.): 457–488. BAR International Series 421 (iii).
BAR, Oxford, England.
KRZANOWSKI, ANDRZEJ, AND KRZYSZTOF TUNIA
1986 Cerámica de la región Cayash. In Cayash prehispánico:
Primera parte del informe sobre las investigaciones arqueológicas
de la Expedición Cientifíca Polaca a los Andes, Proyecto Huaura-
Checras (Perú, 1978) (Andrzej Krzanowski, ed.): 84–186.
Zaklad Narodowy imienia Ossolínskich, Wrocław.
LLANOS, LUIS A.
1936 Trabajos arqueológicos en el Departamento de Cuzco bajo
la dirección del Dr. Luis E. Valcárcel. Informe sobre Ollantay-
tambo. Revista del Museo Nacional 5 (2): 123–156.
LUNT, SARAH
1988 The Manufacture of the Inca Aryballus. In Recent Studies in
Pre-Columbian Archaeology (Nicholas J. Saunders and Olivier
de Montmollin, eds.): 489–511. BAR International Series 421
(iii). BAR, Oxford, England.

250 Albert Meyers

MEANS, PHILIP A.
1931 Ancient Civilizations of the Andes. C. Scribner’s Sons, New York.
MEYERS, ALBERT
1975 Algunos problemas en la clasificación del estilo incaico.
Pumapunku 8: 7–25.
1976 Die Inka in Ekuador: Untersuchungen anhand ihrer mate-
riellen Hinterlassenschaft. Bonner Amerikanistische Studien 6.
Bonner Amerikanistische Studien, Bonn.
1989 Análisis de la cerámica tosca. In Excavaciones en Cochasquí,
Ecuador, 1964–1965 (Udo Oberem and Wolfgang Wurster,
eds.): 180–197. Verlag Phillip von Zabern, Mainz am Rhein.
1993 Trabajos arqueológicos en Samaipata, Depto. De Santa Cruz,
Bolivia. Primera Temporada 1992. Sociedad de Investigación
del Arte Rupestre de Bolivia (SIARB), Boletín 7: 48–58. La Paz.
1998a Los Incas en el Ecuador: Análisis de los restos materiales.
Colección Pendoneros 6–7. Ediciones Del Banco Central del
Ecuador: Abya-Yala. Quito.
1998b Las campañas arqueológicas en Samaipata, 1994–1996.
Segundo Informe de Trabajo. Sociedad de Investigación del Arte
Rupestre de Bolivia (SIARB), Boletín 12: 59–86. La Paz.
n.d. La crónica de Alcaya y otros datos históricos sobre el Fuerte
de Samaipata. Ms. on file. Seminar für Völkerkunde, University
of Bonn, 1997.
MEYERS, ALBERT, AND CORNELIUS ULBERT
1998a Archaeological Explorations in Eastern Bolivia: The
Samaipata Project. In Past and Present in Andean Prehistory
and Early History (Sven Ahlgren, Adriana Muñoz, Susana
Sjödin, and Per Stenborg, eds.): 19–31. Etnologiska Studier 42.
Etnografiska Museet i Göteborg, Göteborg, Sweden.
1998b Inka Archaeology in Eastern Bolivia: Some Aspects of the
Samaipata Project. Tawantinsuyu 3: 79–85.
MORRIS, CRAIG
1995 Symbols to Power: Styles and Media in the Inka State. In Style,
Society and Person: Archaeological and Ethnological Perspectives
(Christopher Carr and Jill E. Neitzel, eds.): 419–433. Plenum
Press, New York.
NORDENSKIÖLD, ERLAND
1915 Incallacta: Eine befestigte und von Inca Tupac Yupanqui
angelegte Stadt. Ymer 2: 169–185.
1917 The Guarani Invasion of the Inca Empire in the Sixteenth
Century: An Historical Indian Migration. Geographical Review
9: 103–121.

Toward a Reconceptualization of the Late Horizon and the Inka Period 251
 1924 Forschungen und Abenteuer in Südamerika. Strecker und
Schröder, Stuttgart.
OBEREM, UDO, AND WOLFGANG WURSTER (EDS.)
1989 Excavaciones en Cochasquí, Ecuador, 1964–1965. Materialien
zur allgemeinen und vergleichenden Archäologie 42. KAVA.
Verlag Phillip von Zabern, Mainz am Rhein.
PARDO, LUIS A.
1938 Hacia una nueva clasificación de las cerámica cuzqueña
del antiguo imperio de los Incas, Peru. Revista del Instituto
Arqueológico del Cuzco 3 (4–5): 1–22.
1939 Arte peruano: Clasificación de la cerámica cuzqueña (época
incaica). Revista de la Seccíon Arqueológica de la Universidad
Nacional del Cuzco 4 (6–7): 3–27.
PÄRSINNEN, MARTTI
1992 Tawantinsuyu: The Inca State and Its Political Organization.
Studia Historica 43. The Finnish Historical Society (SHS),
Helsinki.
PÄRSINNEN, MARTTI, AND ARI SIIRIÄINEN
1997 Inka-Style Ceramics and Their Chronological Relationship
to the Inka Expansion in the Southern Lake Titicaca Area
(Bolivia). Latin American Antiquity 8 (3): 255–271.
PATERNOSTO, CÉSAR
1996 The Stone and the Thread: Andean Roots of Abstract Art.
University of Texas Press, Austin.
PONCE SANGINÉS, CARLOS
1957 La derámica Mollo. Biblioteca Paceña, La Paz.
PUCHER, LEO
1945 Ensayo sobre el arte pre-histórico de Samaypata. Talleres
tipográficos salesianos, Sucre, Bolivia.
RAFFINO, RODOLFO, DIEGO GOBBO, ROLANDO VÁZQUEZ,
AYLEN CAPPARELLI, VICTORIA G. MONTES, RUBÉN ITTURRIZA,
CECILIA DESCHAMPS, AND MARCELO MANNASERO
1998 El ushnu de El Shincal de Quimivil. Tawantinsuyu 3: 22–39.
RIESTER, JÜRGEN
1976 En busca de la loma santa. Editorial Los Amigos del Libro,
La Paz, Bolivia.
1981 Arqueología y arte pupestre en el oriente boliviano. Editorial
Los Amigos del Libro, Cochabamba and La Paz, Bolivia.
RIVERA DORADO, MIGUEL
1977 La cerámica inca de Chinchero, Peru. Indiana 4: 139–170.

252 Albert Meyers

RIVERA SUNDT, OSWALDO
1979 El complejo arqueológico de Samaipata. In El Fuerte
Preincaico de Samaipata (H. Boero R. and O. Rivera S., eds.):
41–144. Editorial Los Amigos del Libro, La Paz, Bolivia.
ROWE, JOHN HOWLAND
1944 An Introduction to the Archaeology of Cuzco. Peabody
Museum, Cambridge, Mass.
1945 Absolute Chronology in the Andean Area. American Antiq-
uity 10 (3): 265–284.
1996 Inca. In Andean Art at Dumbarton Oaks (Elizabeth Hill
Boone, ed.): 301–320. Dumbarton Oaks Research Library and
Collection, Washington, D.C.
SAIGNES, THIERRY
1985 Los Andes Orientales: Historia de un olvido. Instituto Francés
de Estudios Andinos, Lima, Peru.
SAUNDERS, NICHOLAS J., AND OLIVIER DE MONTMOLLIN
1988 Recent Studies in Pre-Columbian Archaeology. BAR Interna-
tional Series 421 (iii). BAR, Oxford, England.
SINOPOLI, CARLA
1994 The Archaeology of Empires. Annual Review of Anthropology
23: 159–180.
SMITH, MICHAEL E.
1992 Rhythms of Change in Postclassic Central Mexico: Archae-
ology, Ethnohistory, and the Braudellian Model. In Archaeology,
Annales, and Ethnohistory (A. Bernard Knapp, ed.): 51–74.
Cambridge University Press, Cambridge and New York.
STEHBERG L., RUBÉN
1995 Instalaciones incaicas en el norte y centro semiárido de Chile.
Colección de antropología 2. Dirección de Bibliotecas,
Archivos Museos, Centro de Investigaciones Diego Barros
Arana, Santiago, Chile.
SZYKUSLKI, JOZEF
n.d. La cerámica de Churajon. Ms. on file, Seminar für
Völkerkunde, University of Bonn.
TRIMBORN, HERMANN
1967 Der Skulptierte Berg von Samaipata: Archäologische Studien
in den Kordilleren Boliviens III. Baessler Archiv, Beiträge zur
Völkerkunde, NF 5: 130–169.
UHLE, MAX
1910 Zur Deutung der Intihuatana. In Verhandlungen des XVI.
Internationalen Amerikanistenkongresses: 371–388. A. Hartleben,
Vienna and Leipzig.

Toward a Reconceptualization of the Late Horizon and the Inka Period 253
 1912 Los orígenes de los Incas. In Actas del XVII Congreso Interna-
cional de americanistas: 302–352. Impr. de Coni hermanos,
Buenos Aires.
1933 Die Ruinen von Cochasquí. Ibero-Amerikanisches Archiv 7 (2):
127–134.
VALCÁRCEL, LUIS E.
1934 Los trabajos arqueológicos del Cusco: Sajsawaman redescubi-
erto II. Revista del Museo Nacional 3: 3–36, 211–233.
1935 Los trabajos arqueológicos en el departamento del Cusco.
Sajsawaman resescubierto III–IV. Revista del Museo Nacional 4:
1–24, 161–203.
VALENCIA, ALFREDO
1970 Las tumbas de Saqsaywaman. Revista Saqsaywaman 1:
173–177.
WEDIN, AKE
1963 La cronológia de la historia incaica: Estudio crítico. Insula,
Madrid.
WURSTER, WOLFGANG
1989 Ruinas existentes. In Excavaciones en Cochasquí, Ecuador,
1964–1965 (Udo Oberem and Wolfgang Wurster, eds.): 11–103.
Verlag Phillip von Zabern, Mainz am Rhein.

254 Albert Meyers