Content uploaded by Erik J. Marsh
Author content
All content in this area was uploaded by Erik J. Marsh on Mar 31, 2015
Content may be subject to copyright.
169
Ñawpa Pacha, Journal of Andean Archaeology, Volume 32, Number 2, pp. 169–188. Copyright © 2012 Institute of Andean Studies. All rights reserved.
T F T
E K’
Erik J. Marsh
Tiwanaku is among the most prominent sites in the Andes. Despite nearly a century of research, it remains unclear when
the site was founded, currently thought to be around 300 B.C. Excavations in 2008 in the Kk’araña sector present pat-
terns suggested by previous research: the earliest material culture is from the first part of the Late Formative period. Tiwa-
naku was likely founded during the first or second century A.D. is refinement to the site’s history carries implications
for clarifying the ceramic chronology and the relatively rapid process of state emergence.
Tiwanaku se encuentra entre los sitios más preeminentes en los Andes, pero después de casi un siglo de investigación, per-
manece poco clara la fecha de su fundación, sugerida alrededor de 300 a.C. Ésta se basa en la presencia de fechados radio-
carbónicos tempranos, los cuales no están asociados con material cultural, y por tanto no se corresponden con ocupaciones
humanas del sitio. Excavaciones recientes en el sector de Kk’araña confirman los patrones de las investigaciones previas,
en donde el material cultural más temprano en Tiwanaku se ubica en el primer o segundo siglo d.C., durante el Período
Formativo Tardío. En Kk’araña, como en otros sectores, este material se encuentra en estratos inmediatamente por encima
de sedimentos estériles. Toda esa evidencia temprana de contextos domésticos, se localiza alrededor del Templete semi-sub-
terráneo, el núcleo probable de la fundación del sitio. Sugiero entonces que la fundación de Tiwanaku se produjo después
del 1 d.C. Ello se basa en un refinamiento en la cronología disponible para el sitio y en tres fechados nuevos, por primera
vez en 50 años, para el Período Formativo Tardío. Esto conlleva implicancias importantes para entender el proceso acel-
erado que pudo ocurrir en la conformación de este estado.
Erik J. Marsh, Department of Anthropology, University of California, Santa Barbara, CA 93106 (emarsh@umail.ucsb.edu)
This article is copyright protected and may not be used for commerical purposes without the written
permission of the publisher. No forwarding, web posting, or further distribution of this PDF is allowed.
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
170
Tiwanaku is among the most prominent sites in
the Andes, and was the capital of an archaic state
from around A.D. 500–1000 (Figure 1). Despite near-
ly a century of research, it remains unclear when the
site was founded. is basic point of chronology has
important implications for the dramatic social changes
that accompanied state emergence. Refining the chro-
nology of Tiwanaku’s rise can help clarify the nature of
this Andean state, which was different from others in
many respects (Goldstein 2007: 154–158). Sparse data
and radiocarbon dates with unclear contexts suggested
that the site was founded during the Middle Forma-
tive, but the lack of associated material culture has left
doubts that the site is so old (Bandy 2001: 201).
is paper reviews Formative evidence from differ-
ent sectors at Tiwanaku, which suggests that the site was
founded in the early part of the Late Formative (A.D.
1–500). It also presents the chronology of the Kk’araña
sector based on a 2008 excavation, carried out as part
of my dissertation research (Marsh 2008, 2012b). Stra-
tigraphy, ceramics, and radiocarbon dates confirm that
this sector, once thought to be the oldest at Tiwanaku
(Janusek 2004: 100), was first occupied during the Late
Formative. ese are the first Late Formative radiocar-
bon dates at Tiwanaku since the 1960s (Ponce Sanginés
1964: Table 2, 1970: Table 5). Kk’araña’s residents were
among Tiwanaku’s earliest occupants, who founded the
site in the first or second century A.D.
e Late Formative period was previously called
Early Tiahuanaco or Tiwanaku I, II, and III (Figure 2;
Bennett 1934: 448–453; Janusek 2003: 37–40, 52–54;
Ponce Sanginés 1993: 50–56). e diagnostic ceramics
of this period are Kalasasaya zonally incised, Kalasasaya
red–rimmed, and Qeya vessels (Janusek 2003: 37–54).
Figure 1. Perspective map of the southern Titicaca Basin, as seen from the southeast, indicating selected Late Formative ceremonial
centers (Marsh 2012a: Figure 1). Based on a composite map by Arik Ohnstad, using Landsat and SRTM topographic data.
Marsh: e founding of Tiwanaku
171
e current ceramic chronology suggests the Late For-
mative began around 200 B.C., but the absolute dates
of this period are based on many unreliable radiocarbon
dates. A re–evaluation suggests that Janusek’s ceramic
sequence begins during the first century A.D. (Marsh,
Marsh 2012a: 213–214). Current research is refining
the starting and ending dates of the Late Formative, the
timing of the shift from the phases Late Formative 1 to
Late Formative 2 (hereafter LF1 and LF2, respectively),
and variability in the use of decorated and undecorated
ceramics (see review in Marsh 2012b: 98–114).
Formative Occupations at Tiwanaku
ere are reports of scarce Middle Formative undeco-
rated sherds on the surface at Tiwanaku. However, ex-
cavation indicates that the earliest occupation was dur-
ing the Late Formative (Figure 3). is
section first addresses surface finds, and
then proceeds with discussions of specific
sectors in the order they were excavated.
Surface Evidence
Most considerations of the surface ceram-
ics at Tiwanaku have focused on estimat-
ing the maximum population (e.g., Par-
sons 1968; see Lémuz Aguirre 2005). e
first systematic surface collections did not
report Formative diagnostics (e.g., Giesso
2011: 152–155; Sutherland 1991). Por-
tugal Ortiz (1992: Figure 15) published
Figure 2. Late Formative ceramic chronology. Images modified
from Janusek (2003: Figure 3.20), photo by Wolfgang Schüler,
and Ponce Sanginés (1970: Figure 31), not to scale.
Figure 3. Tiwanaku’s ceremonial core, showing
excavations with LF1 evidence in blue. Portugal’s
La Karaña and Bennett’s Pit VIII include Late
Formative material, some of which may date to
LF1, but this is unconfirmed.
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
172
a schematic map that identified a few concentrations
of Middle Formative sherds on the surface, but at this
time the ceramic chronology did not clearly distin-
guish between undecorated Middle and Late Forma-
tive sherds. Pastes with coarse fiber were once consid-
ered diagnostic of the Middle Formative, though it is
now clear that fiber pastes were used during the Late
Formative as well (see Bandy 2001: 46–56; Bermann
1994: 52–53; Browman 1998: 301; Roddick 2009:
220–221, Figure 7.9).
e first detailed survey of the entire site, using
a significantly updated ceramic chronology, concludes
the following:
e Middle Formative presence at the site of
Tiwanaku is scarce, and restricted to isolated
finds and small concentrations, found in the
sectors east of the Akapana, Kk’araña, and the
area of qochas in Mollo Kontu (Lémuz Agu-
irre 2004a: 12, translation by author, empha-
sis added).
In stark contrast to the isolated find of Mid-
dle Formative sherds, the survey reports two large
Late Formative surface scatters, covering some 81 ha
(Lémuz Aguirre 2004a: 12; see Marsh 2012b: 452).
While the surface data seem to leave open the possibil-
ity for a Middle Formative occupation at Tiwanaku,
excavation data reveals a consistent pattern: Late For-
mative occupations and ceramics overlaying sterile soil.
Excavations by Bennett, Kidder,
and Cordero Miranda
e first well-documented, systematic excavations at
Tiwanaku suggested that settlement began in the Late
Formative. In 1932, Bennett (1934) excavated ten pits,
two of which included a few Late Formative decorat-
ed sherds mixed with domestic refuse overlying ster-
ile soil. In 1955, Kidder excavated two adjacent pits,
with the purpose of refining and dating the ceramic
sequence (Janusek 2003: 33; Ponce Sanginés 1995:
173; Ralph 1959: 54–55). e excavation was never
described in detail, but Late Formative ceramics were
reported in association with date P–123, cal A.D. 205
(80–333).1 e lowest level of Bennett’s pit was 50 cm
deeper, so perhaps dated to around the first century
B.C. (Rowe 1963: 8). While these data did not provide
a definitive picture, they roughly outlined the pattern
that has since emerged.
A few years later, Gregorio Cordero Miranda
directed excavations in the Kalasasaya temple (Ponce
Sanginés 1995: 226–228). Late Formative occupa-
tions were revealed in the deepest levels of a number
of units (Figure 4; Couture 2002: 95–101; Janusek
2003: 37–38; Ponce Sanginés 1993: 50–54). Sterile
soil was reached in stratum 8, which was “prior to hu-
man settlement” (Ponce Sanginés 1993: 53, translation
by author). Strata 5 and 7 both included Late Forma-
tive Kalasasaya ceramics and occupational refuse, and
radiocarbon dates with medians in the first three cen-
turies A.D. e earlier dates from this excavation are
from the “sterile” stratum 6, or the mixed layers of con-
struction fill in stratum 4 (Ponce Sanginés 1993: 53,
translation by author), and do not seem to correspond
to human occupation (see below; Marsh 2012a: 209–
211). Excavations of 73 units in the Kalasasaya covered
1825 m2, part of a project whose goals included inves-
tigating Formative period occupations (Ponce Sangi-
nés 1993: 51). Despite this goal, decades of extensive
excavations throughout Tiwanaku did not reveal any
Middle Formative cultural material.
In these excavations by Bennett, Kidder, and
Cordero Miranda, sterile soil was located immediately
Figure 4. West profile of pit E-17 in the Kalasasaya, redrawn
from Ponce Sanginés (1993: Figure 13.2, 1995: Figure 141).
Marsh: e founding of Tiwanaku
173
below occupational refuse that included Late Forma-
tive decorated ceramics. e radiocarbon dates directly
associated with this material suggest an initial occupa-
tion during the last century B.C. or first century A.D.
Excavations by Arellano López
In 1984, six units were excavated within the Puma
Punku, Kalasasaya, north of the Kalasasaya, and “west
and south of the Akapana (B–1, B–2)” (Arellano
López 1991: 269). Based on these excavations, a new
ceramic sequence was proposed (Arellano López 1991:
Table 3). Generally it is of “dubious utility,” as there are
a number of inconsistencies with both prior and later
research (Mathews 1992: 100).
Of interest here is the unexpected identification
of Chiripa, or Middle Formative, ceramics in the same
strata as diagnostic LF2 sherds. Arellano López (1991:
273, Table 3) argues that Chiripa ceramics “survived
at least until what has been called [LF2]” (Arellano
López 1991: 273, Table 3), echoing a previously pro-
posed temporal overlap between Chiripa and Tiwan-
aku cultures (Ponce Sanginés 1970: 55–61). However,
it seems more likely that these were in fact sherds from
undecorated Late Formative vessels.
Pit B–1, West of the Akapana. Excavation in unit
B–1 identified Tiwanaku period sherds in levels 1 and
2 (Figure 5). Level 2 also included diagnostic sherds of
LF2. Levels 3–6 included apparently Chiripa sherds,
sampled from inside a bell-shaped pit with ash and
bone.
Level 2 included Qeya decorated sherds, diag-
nostic of LF2 (Figure 2; Arellano López 1991: Figure
42c–d). In the same level, there were sherds from jar
necks with punctuations and incisions, with mica-
ceous sandy paste (Arellano López 1991: Figure 42a–
b). is paste was very common in Late Formative jars,
and neck sherds with punctuations have been found
in contemporary contexts at various other sites (e.g.,
Bandy 2001: Figure 7.2a–h; Bermann 1990: Figures
82a and 98; Burkholder 1997: 181, Figure 8.6; Ja-
nusek 2003: Figure 3.9a, Ponce Sanginés 1993: Fig-
ure 4.2; Roddick 2009: Figures 5.2f and 7.17; Rydén
1947: Figures 6a and 66h; Steadman et al. 2005: Fig-
ure 6.3a–c).
e lower levels (3–6) included sherds from ves-
sels with unusual sizes and forms according to Arellano
López (1991: 273). ese vessels seem to have orifices
that range from about 14 to 32 cm (Figure 6). Necked
vessels from the Late Formative and Tiwanaku periods
usually have diameters between 12 and 26 cm, though
exceptionally large vessels are also known, some with
orifices larger than 30 cm (Couture 2002: Figures 4.11
and 5.13; Janusek 2003: 57; Marsh 2012b: 273, Table
5.19; Roddick 2009: 336, 354; Steadman 2007: 89,
103). e vessels from levels 3–6 have forms that are
very similar to those of undecorated vessels from the
Late Formative and Tiwanaku periods (Figure 7). Ad-
ditionally, level 3 included a bowl with straight walls,
a shape known from the Late Formative, and in lev-
el 5 there was a goblet–shaped vessel, perhaps similar
to a proto–kero, a rare but diagnostic LF2 form (e.g.,
Bermann 1990: Figures 78–81; Janusek 2003: Figure
3.25; Marsh 2012b: 254, Figure 5.9e; Roddick 2009:
372, Figure 9.21). In summary, the apparently Chiripa
sherds seem to be from Late Formative vessels, and are
associated with diagnostic LF2 decorated sherds.
e profile clarifies that these sherds are strati-
graphically associated with level 2, probably depos-
Figure 5. Profile of pit B-1, west of the Akapana, redrawn from
Arellano López (1991: Figure 35).
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
174
ited during LF2 (Figure 5). e “ceramic collection
levels” are within a bell–shaped feature filled with ash
and bone (Arellano López 1991: 271). Such pits are
ubiquitous at Tiwanaku, and commonly filled with
undecorated ceramics (Janusek 2009: 171–172; Rod-
dick 2009: 201). e relationship between the ashy pit
and its stratigraphic origin implies that the undecorat-
ed vessels date to LF2 or perhaps the early Tiwanaku
period. Neither the style nor the stratigraphic positions
of the ceramics indicate occupation or material prior to
the Late Formative.
Arellano López (1991: 271, 277) concludes that
the earliest phase at Tiwanaku was defined by decorat-
ed ceramics such as Kalasasaya and Qeya vessels (Fig-
ure 2). e preponderant undecorated sherds, called
Chiripa, seem to be from Late Formative jars and ollas
(see Mathews 1992: 100). is set of ceramics current-
ly defines the Late Formative (Janusek 2003: 41–52).
If so much can be read from the chronological chart,
Arellano López (1991: Table 3) surmises that this set of
ceramics was used no earlier than A.D. 1, presaging the
general conclusion presented here.
Excavations by Portugal Ortiz in La Karaña
(Western Sector)
In 1988, Portugal Ortiz (1992: 15, 25) located a Late
Formative domestic structure in the western sector of
La Karaña. is excavation was located 112 m north
of the Kalasasaya’s northwest corner, and appears to be
unrelated to later excavations further east, where no
Formative materials were reported (K. Davis, personal
communication; Escalante Moscoso 1994: 255–259,
2003: 326). e domestic structure had stone founda-
tions similar to those found below the Kalasasaya and
at other contemporary sites (Marsh 2012b: 116–129).
e material included domestic refuse, lithics, and Late
Formative decorated ceramics. It is unclear if the ex-
cavation continued below these remains, but accord-
ing to the brief report, this seems to be the earliest
occupation.
Figure 6. Vessel shapes from pit B-1, redrawn from Arellano
López (1991: Figure 41).
Marsh: e founding of Tiwanaku
175
Excavations by Proyecto Wila Jawira
In the 1980s and 1990s, members of the Proyecto Wila
Jawira, directed by Alan Kolata and Oswaldo Rivera,
conducted extensive excavations throughout Tiwanaku
(Kolata 2003). ey revealed scattered Late Formative
occupations in Mollo Kontu and east of the Akapana,
and identified one sector with possible Middle Forma-
tive evidence, Kk’araña (Janusek 2004: 100).
Middle Formative ceramics have been reported
from the surface of the Mollo Kontu sector, especially
near an area with qochas, shallow basins dug out for
collecting water (Janusek 2004: 100; Lémuz Aguirre
2004a: 12, 2004b: 10). A surface survey of Mollo Kon-
tu identified 43 surface concentrations, 21 of which
included material from the Late Formative. Only one
concentration included potentially Middle Formative
material (Lémuz Aguirre 2004b: 10).
Excavations have revealed material from no earlier
than the Late Formative. e Mollo Kontu mound was
built during the Tiwanaku period, when a number of
dedicatory offerings were placed (Couture 2003). Ponce
Sanginés (1961: 23) directed excavations of 25 m2, re-
vealing some 30 tombs. In 1990 and 1991, over 400
m2 were excavated, which identified early Tiwanaku pe-
riod refuse located over sterile soil (Couture 2003: 210,
Figure 8.21). South of the mound, two large areas were
excavated from 2001 to 2008. Diagnostic ceramics and
radiocarbon dates confirmed that the earliest occupa-
tion was not until the late sixth century (Couture et al.
2010). Other excavations did reveal “thin midden lens-
es” that may date to the LF2; this is the earliest possible
occupation (Janusek 2004: 104).
Surface survey has also reported Middle Forma-
tive sherds east of the Akapana. However, excavations
east of the Akapana have reported sterile soil below
Late Formative occupations, such as in Bennett’s pit
V, one of the few places where LF1 material has been
found (Figure 3). To the southeast, in Akapana East 1,
Janusek reports an initial occupation during LF2, in-
cluding a small structure foundation and LF2 ceramics
(Janusek 1994: Figures 6.2–6.4, 106–108, 2004: 103–
104). In sectors further east, such as Akapana East 1
and Ch’iji Jawira, occupation probably began in the
early Tiwanaku period.
Finally, the 1990 excavations at Kk’araña identi-
fied it as the sector with the best evidence of Formative
occupation at Tiwanaku, as early as the Middle Forma-
tive (Janusek 2004: 100). In 2008, excavations I di-
rected at Kk’araña established that this sector was first
occupied no earlier than the second century A.D.
Kk’araña: Integrating Data
from the 1990 and 2008 Excavations
is section reviews data on the chronology of occupa-
tion of Kk’araña, based on field notes from the 1990
excavation and additional excavations in 2008 (Figure
8; for additional detail, see Marsh 2008, Marsh 2012b:
Chapter 7). Here, I limit my discussion to the chronol-
ogy; elsewhere I focus on domestic practices and rela-
tionships with residents of Tiwanaku and other sites
(Marsh 2012b: Chapter 8). Janusek (2004: 100–103)
tentatively proposed three occupations for Kk’araña,
based on Helsley–Marchbanks’ (1990) excavation re-
port and field forms (Figure 9), but here I suggest that
there was only one, during LF1.
Figure 7. Cooking and storage vessel forms from the Late Formative and Tiwanaku periods. Modified from Janusek (2003: Figure 3.27).
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
176
Kk’araña’s Single LF1 Occupation
Analysis of the ceramics excavated in 1990 suggested
“abundant” Middle Formative sherds, but they more
likely date to LF1 (Janusek 2004: 101). Janusek’s 1993
analysis identified ceramic attributes that were com-
mon during the Middle Formative: pastes with fi-
ber and mica inclusions, thick unoxidized cores, red
slips, and exterior burnishing. ese attributes were
also present in Late Formative assemblages, but with
subtle differences, and usually in lower frequencies
(Arellano López 1977: 82–83; Bermann 1994: 66,
269; Browman 1980: 110–115; Lémuz Aguirre 2001:
186–191; Mathews 1992: 71, 104; Roddick 2009:
Figure 8. Location of units excavated in Kk’araña in 1990, shown in dark gray, and those excavated in 2008, numbered and shaded in
light grey. Modified from Marsh (2012: Figure 7.3).
Figure 9. Feature 34 from
the 1990 excavations,
designated here as structure
5, facing south. Photo
courtesy of Alan Kolata,
from the Proyecto Wila
Jawira archive, University of
Chicago.
Marsh: e founding of Tiwanaku
177
220–221; Steadman 1995: 299–305). ese attributes
were most common in sherds from large jars, a pat-
tern also seen in other Late Formative assemblages. In
each excavation context with apparently Middle For-
mative sherds, Late Formative sherds were also identi-
fied. In some cases, these apparently mixed contexts
unexpectedly appeared above unmixed Late Formative
contexts. Janusek agrees that the sherds originally iden-
tified as Middle Formative more likely date to LF1 (Ja-
nusek, personal communication, 2011). is temporal
overlap in ceramic attributes highlights the difficulty
in confidently identifying an occupation’s phase with
small samples of undecorated sherds.
e 1990 excavations did not reach sterile soil,
so at the time it was reasonable to suppose that sherds
with early attributes might be associated with an ear-
lier, unexcavated occupation (Janusek 2004: 100).
e 2008 excavations identified culturally sterile soil
in a number of units, which was consistently located
immediately below the LF1 occupation (Figure 10;
Marsh 2012b: 371, 396, 413; Ulloa Vidaurre 2008:
38; Viviani Burgos 2008: 43). e sterile soil was prob-
ably a paleo-lake bed, a highly uniform deposit of very
fine silts and clays and a complete absence of artifacts.
is contrasts with potentially similar construction
fills, which often include mixtures or pockets of dif-
ferent sands and clays and sparse artifacts, especially
small Late Formative undecorated sherds. Kk’araña’s
residents built adobe structures with stone foundations
over sterile soil during LF1 (Figures 11 and 12).
LF2 Use Surface or LF1 Adobe Wall Melt?
During the 1990 excavations, a compact clay layer
was identified as a possible LF2 use surface, but in fact
seems to have been adobe wall fall from the LF1 adobe
structures (Figure 10; see Bermann 1994: Figure 3.6).
is layer most likely formed from collapsed adobe
walls, leaving “an irregular, compact surface,” similar to
Figure 10. Proposed LF1 (left) and LF2 (right) occupations of Kk’araña, modified from Janusek (2004: Figures 4.1c and 4.2a).
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
178
Figure 11. e 2008 excavation, facing
southwest, showing structure 3 in the
center of the photo. Within structure
3, a 1 m2 excavation revealed culturally
sterile deposits. Photo by D. Ulloa
Vidaurre.
Figure 12. Architectural features at Kk’araña. Based on hand drawn and digital maps by A. Helsley-Marchbanks, J. Janusek (2004b:
Figure 4.1c), R. Fontenla Alvarez, E. Marsh, D. Ulloa Vidaurre and L. Viviani (see chapters in Marsh 2008).
Marsh: e founding of Tiwanaku
179
the adobe wall melt identified in Akapana East 1 (Ja-
nusek 1994: 108).
e proposed LF2 use surface extends over about
32 m2, coinciding with the extent of stone foundations
(Figure 10). It is apparently associated with hearths
and refuse pits, but these features originate above the
clay stratum, suggesting they post-date it. e stratum
is 25–35 cm above the level of the stone foundations,
a plausible depth for collapsed and rain-melted ado-
be walls (Bermann 1994: 54). Profiles indicate wall
stumps and a low mound of structure 2, similar to oth-
er collapsed adobe structures (Figures 13 and 14, Table
1; Bermann 1994: 54; Goodman–Elgar 2008:3067;
Figure 13. Collapsed mud structure from West Africa, showing mound shape and surviving wall stumps. Modified from McIntosh
(1977: Figure 4).
Table 1. Key to profile in Figure 14.
APlowzone 5.yr 4/2
BAshy with red fill 5.yr 5/3, 7.5 yr 5/3
CReddish clayey fill 5.yr 5/3
DAdobe melt on floor 7.5 yr 4.5/2
EAdobe wall/floor preparation 7.5 yr
FReddish clayey fill 7.5 yr 5/3
GAshy reddish fill 7.5 yr 5/3
HBrown clayey fill 7.5 yr 5/2
IRed clay – adobe melt? 2.5 yr 5/4
JBrown clay fill 7.5 yr 4/2
KBrown clay fill 7.5 yr 4/3
SAshy light brown fill 5.yr 5/3, 7.5 yr 5/3
Figure 14. Profile (top) and plan
(bottom) of unit N8566 E4930,
re-excavated as unit 3 in 2008 (see
Figure 8). Profile drawn by A. Helsley-
Marchbanks, digitized by E. Marsh; see
key in Table 1. Plan view of structure 2
(bottom) drawn by D. Ulloa Vidaurre,
and digitized by R. Fontenla Alvarez.
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
180
McIntosh 1977: 187–193). e possibility that Tiwa-
naku was once a large urban center was first suggested
by the site’s numerous low mounds, correctly inter-
preted as collapsed adobe structures (Bandelier 1910:
225; Parsons 1968: 244).
Refuse Accumulation
Over the collapsed adobe walls, secondary refuse
began to accumulate in LF2 and continued dur-
ing the Tiwanaku period (Figure 15; Janusek 2004:
143–144). e refuse level was identified in the 1990
and 2008 excavations. Material includes temporally
mixed diagnostics: Late Formative ceramic buttons,
sherds from Tiwanaku period serving vessels, and
sherds from orange jars with fiber pastes, which were
used during the Late Formative and Tiwanaku peri-
ods. e 1990 excavations identified a few LF2 di-
agnostic Qeya sherds from this mixed fill level (see
Janusek 2003: 48, Figures 3.19 and 3.23). Tiwan-
aku period sherds are found throughout the midden,
though less common in deeper parts of the level. Nei-
ther Qeya nor Tiwanaku ceramics were found within
or below the wall fall, indicating that the occupation
of the structures and was limited to LF1.
Absolute Dating
ree samples of carbonized organic material were ra-
diocarbon-dated to LF1 (Table 2). e earliest two
(KK2 and KK3) are from the lowest levels of mid-
den that accumulated against the north and south
side of the principal compound wall, respectively
(Figure 15). e latest (KK1) was taken from the use
surface of a gravel patio south of structure 2 (Figure
12). Treated independently, the dates are statistically
indistinct. Toward a more precise picture of the start-
ing and ending dates of the occupation, I grouped
them as single phase in a Bayesian model. e phase
likely began around A.D. 200 (120–310), and ended
around A.D. 350 (240–410).ese dates and occupa-
tions are associated with typical Late Formative ce-
ramic assemblages, dominated by undecorated sherds
from cooking vessels and storage jars. e few deco-
rated sherds are diagnostic of LF1. ey include Kala-
sasaya red–rimmed small jars and bowls, as well as an
elaborate LF1 vessel in the Kalasasaya zonally incised
style (Figures 2 and 16). is vessel is from a midden
north of the compound wall, from the same strata as
the earliest date (KK2).
Figure 15. Western profile
of structure 1, showing ash
dumps in light grey. e
wall fall and occupation
surface are shown in dark
grey. Immediately south of
the structure foundation,
carbon sample KK2 dates the
occupation to LF1. Modified
from Marsh (2012b: Figure
7.11).
Marsh: e founding of Tiwanaku
181
Absolute Dates and Regional
Comparisons
ere are few reliable Late Formative radiocarbon
dates at Tiwanaku, though recent dates from Kk’araña
and large sets of dates from other sites have contribut-
ed to a refined regional chronology of the Late Forma-
tive and the poorly understood centuries immediately
prior (Figure 2).
Late Formative Chronology
Ponce Sanginés maintained for decades that Tiwan-
aku’s initial occupation was around 1580 B.C., based
on one very early date (1970: Table 5, 1993: 65–66;
see Janusek 2004: 65). However, this and other ear-
ly dates seem to be outliers, or not directly associated
with the site’s human occupation. e first phase, Ti-
wanaku I, was dated to 237 B.C., the arithmetic aver-
age of the uncalibrated means of the oldest radiocarbon
dates (Ponce Sanginés 1993: 65), a spurious approach
to estimating absolute age. is date seemed to gen-
erally agree with stylistic comparisons to Tiwanaku’s
earliest ceramics and monoliths, which suggested that
it was founded no earlier than the first few centuries
B.C.; the earliest described material was in fact associ-
ated with much later dates. Given the lack of any oth-
er data, the consensus for many years was that Tiwan-
aku was founded around 300 B.C. (Bermann 1990:
86; 1994: 66; Browman 1980: 114; Chávez and Mohr
Chávez 1975: 66; Janusek 2003: 37, 47; Lémuz Agu-
irre 2005: 5; Mathews 1992: 66–67, 1995: 85; Rowe
1963: 8; Stanish 2003: 117).
Currently, there are only two radiocarbon dates
directly associated with the ceramics used by Tiwan-
Table 2. Calibrated and modeled dates from Kk’araña. Positive dates are AD, negative date is BC.
Modeled
Boundary
Excavation
Locus Lab Code
Radiocarbon
Age ± Error
Calibrated Modeled
Median 1 2 Median 1 2 A index
End occupation 350 240–410 160–770
1.5.4 AA89414 (KK 1) 1742 49 300 230–380 130–410 290 230–340 160–400 108
1.12.8.B AA89412 (KK 3) 1753 50 290 220–380 130–400 280 240–430 160–390 111
1.19.12 AA89413 (KK 2) 1798 49 220 130–320 80–380 260 210–340 130–360 98
Start occupation 200 120–310 –230–370
Figure 16. Kalasasaya zonally incised
polychrome vessel, painted red (7.5 yr
5/6), black (7.5 yr 2.5/1), brown (7.5 yr
5/4) and orange (5 yr 6/6).
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
182
aku’s earliest occupants, KK2 from Kk’araña, cal A.D.
220 (130–320), and P–532 from the Kalasasaya, origi-
nally reported as 1653 ± 61 B.P. (Stuckenrath 1963:
95). A revised radiocarbon age for this date was in-
cluded in a 1962 letter to Ponce Sanginés from Rob-
ert Stuckenrath, director of the Pennsylvania labora-
tory that processed the sample (Steadman 1995: 147).
e letter indicates two counts for the sample, 1956
± 76 and 1847 ± 75 B.P. (Ponce Sanginés 1970: Table
5). Statistically combining and calibrating these two
counts produces a date of A.D. 110 (20–210), which
much more closely follows stylistic and stratigraphic
expectations (Marsh 2012a: 211).
ese two dates are from occupation levels that
overlie sterile soil. ey include domestic refuse and
both styles of LF1 decorated ceramics (Figure 2). e
date from the Kalasasaya is associated with a cache of
24 complete vessels in pit E–17, which includes 11
rare Kalasasaya zonally incised vessels (Figure 4; Ponce
Sanginés 1993: Table 1). Hence, these two radiocar-
bon dates are the most secure for dating the founding
of Tiwanaku and the use of the zonally incised style,
which seems to have been in circulation for less than
two centuries.
e rest of the Late Formative dates from Tiwan-
aku are from Kidder’s pits and the Kalasasaya, but have
unclear material associations. A re-assessment of these
dates, using Bayesian models, suggests that occupa-
tion in Bennett’s pits, Kidder’s pits, and the Kalasasaya
probably started around A.D. 110 (50–170, 1 sigma)
(Marsh 2012a: 213).
A similar chronology is apparent at other sites,
such as Lukurmata, Khonkho Wankane, and three
sites on the Taraco Peninsula (Figure 1). Large sets of
radiocarbon dates indicate that Late Formative ceram-
ics were first used around A.D. 1 at Lukurmata, and at
other sites, not until after A.D. 100. At most sites, peo-
ple stopped using these ceramics around the first half
of the fifth century A.D. (Bermann 1994: Appendix
III; Bruno 2008: Appendix 1; Janusek 2003: Table 3.3,
2011: Table 1; Marsh 2012b: Tables 5.7 and 7.2; Rod-
dick 2009: 170–176, Table 5.2; Smith 2009: Figure
3.24, Table 3.1). ere seems to have considerable in-
ter-site temporal variation in the use of ceramic styles.
For example, at Lukurmata and Kirawi, Late Forma-
tive ceramics may have been in use as late as the sev-
enth century A.D. (Janusek 2003: 52). While there are
general temporal tendencies, shifts in the use of differ-
ent styles were not coordinated or consistent (Janusek
2003: 88; Roddick 2009: 177).
is adjustment to the Late Formative chronol-
ogy has important implications for the regional ceram-
ic sequence, chronologies of survey data based on this
sequence (e.g., Bandy 2001), as well as the dating of
Yaya–mama stone sculptures. e ceramics and mono-
liths are chronologically sensitive material once thought
to have been much older, but in many cases, probably
date to the Late Formative, at Tiwanaku and other sites
(see Marsh, in press). An additional implication of this
chronological refinement is a conspicuous lack of data
during prior centuries (Figure 2).
e Middle to Late Formative Transition
ere is little information available on the final centu-
ries B.C. in the southern Titicaca Basin. e latest use
of Middle Formative ceramics was around 250 B.C.,
while the earliest use of Late Formative ceramics was
after A.D. 1. e lapse between these dates identifies
a poorly known period with no temporally diagnostic
artifacts or reliable radiocarbon dates.
e Middle Formative was traditionally dated
from 800 to 200 or 100 B.C., based on ceramic se-
quences and radiocarbon dates from Chiripa (Brow-
man 1980: 111; Janusek 2003: 37; Mohr Chávez
1988: 18). A re-assessment of the latest radiocarbon
dates and their contexts suggests that the Middle For-
mative occupation at Chiripa ended around 250 B.C.
(Bandy 2001: 115–118). Almost all data that define
the Middle Formative are from Chiripa itself, so the
regional chronology remains vague.
Recent excavations on the Taraco Peninsula spe-
cifically targeted contexts from the centuries following
250 B.C. ere are data from this period at Kala Uyu-
ni, but they are from mixed, secondary contexts (Bru-
no 2008: 390–391; Roddick 2009: 125–126). At Chi-
ripa, there are constructions and occupations that may
be contemporaneous (Bandy 1999: 13–14, 2001: 174;
Bennett 1948: 91; Browman 1980: 111, 118; Hastorf
Marsh: e founding of Tiwanaku
183
1992: 8–9; Mohr Chávez 1988: 20–21; Steadman and
Anthony 2008: 72–73, 80).
Beyond the Taraco Peninsula, a few other sites
have data potentially relevant to these centuries. Two
sites have radiocarbon dates, Palermo and Ch’isi (Stan-
ish et al. 1997: 52, 73–74, Figure 33). Ch’isi includes
a sunken court, carved stones, Chiripa and Kalasasaya
style ceramics, and five radiocarbon dates, whose over-
lapping 1 sigma ranges span cal 360 B.C to cal A.D.
120 (Mohr Chávez 1997: 4). Other sites without radio-
carbon dates include stratigraphic sequences that cover
this transition, such as Qeya Kuntu, Quesani, and Iro-
hito (Janusek and Kolata 2003: 136–140; Lémuz Agu-
irre and Paz Soria 2001: 105–107; Pérez Arias 2007:
230–238). Future research at these sites will be able to
clarify these poorly documented centuries.
At principal Late Formative centers such as Tiwa-
naku, Lukurmata, and Khonkho Wankane, there is no
clear evidence for occupation prior to the Late Forma-
tive. e large temporal gap between the Middle and
Late Formative periods suggests more critically evalu-
ating Janusek’s (2008: 91) suggestion that Late Forma-
tive centers “all cover relatively small Middle Formative
habitation sites that were at the peripheries of Middle
Formative political communities.” I consider it more
likely that these centers were newly founded around
the first century A.D., and there may have been little
direct historic link to Middle Formative communities.
e founding and rise of Late Formative centers may
have been part of rapid, emergent processes during a
period of accelerated historical change and population
movement. Such a period will be challenging to docu-
ment archaeologically.
Conclusion: e Founding
of Tiwanaku
Tiwanaku was founded during LF1, probably in the
first or second century A.D. e initial settlement was
focused around the Sunken Temple (also known as the
Semi–subterranean Temple), an LF1 edifice with an
LF1 monolith (Figure 3; Callisaya Medina 2009: 101;
Chávez and Mohr–Chávez 1975: 66; Janusek 2004:
109, 2006: 478–480; Marsh 2012b: 94; Ponce Sangi-
nés 1964: 61, 63; Smith 2009: 282). It remains unclear
if this settlement was contiguous, like the surface ce-
ramics, or if residential occupations were spatially sepa-
rated, such as Kk’araña and the residences below the
Kalasasaya (Figure 3). Other sectors were not occupied
until LF2, such Akapana East 1, Mollo Kontu, and the
Putuni (Couture 2002: 110–117, Figure 4.10; Portu-
gal Ortiz 1992: 32). Some LF2 data also come from
Ch’iji Jawira, the Puma Punku, and areas around the
Akapana (Janusek 2004: 133; Marsh 2012: 463–465).
e majority of sectors from were first occupied after
A.D. 600, marking the period of the densest popula-
tion, during which most immigrants arrived (Janusek
2004: 149).
e scarcity of Middle Formative ceramics on the
surface does leave open alternate possibilities, which
cannot be adequately tested with current data. For ex-
ample, Tiwanaku may have been first settled prior to
the Late Formative, but material from hamlets or tem-
porary occupations might not have not survived lat-
er disturbances (Bandy 2001: 201). Early occupations
may have been located away from the ceremonial core,
in areas that have not yet been excavated, such as near
the qochas southeast of Mollo Kontu. ey may have
been in sectors subsequently covered by deep layers of
fill, and remain undiscovered. ese potential scenari-
os may date to the sparsely documented centuries be-
tween the Middle and Late Formative periods.
Refining the cultural history of this preeminent
site carries implications for our understanding of the
rise of the state. Rather than a singular, dominant cen-
ter that grew gradually during two thousand years
(Ponce Sanginés 1981), Tiwanaku was probably simi-
lar to other contemporary centers founded during the
Late Formative, forming part of a network of centers
akin to Renfrew’s early state module” (Renfrew 1986:
2; see Bandy 2001: 201–204; Janusek 2008: 95–96;
Stanish 2003: 283–284). e emergence of the state
was a much more rapid process than previously imag-
ined, suggesting an accelerated process of punctuated
change (Marsh 2012: Chapter 9). e social changes
underway in the Late Formative led to an enduring
hegemony centered at Tiwanaku, a nexus of interac-
tions between people from distant places throughout
the Andes.
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
184
Acknowledgments
Excavations at Kk’araña in 2008 were organized as
Proyecto Kala Uta, funded by a Fulbright–Hays grant
(P022A070024). I thank my co–director, Delfor Ulloa
Vidaurre, and the other excavators, Ruth Fontenla Al-
varez and Luis Viviani Burgos. Organizing the logistics
of an excavation in Tiwanaku required the patient col-
laboration of Alejandra Gasco, Marcial Medina Huan-
ca and Don José Osco. Project equipment and labora-
tory space was generously offered by Maria Bruno and
Alexei Vranich. Field forms, reports, and photo in Fig-
ure 7 are from the Proyecto Wila Jawira archive at the
University of Chicago and were provided by Christine
Hastorf and John Janusek. Janusek introduced me to
Kk’araña, greased the logistical wheels, and has pro-
vided a constant stream of helpful feedback. Excava-
tions permits were facilitated by Javier Escalante, César
Callisaya, and the mayor and archaeological commis-
sion of the Municipality of Tiwanaku. ree anony-
mous reviewers provided valuable comments that sig-
nificantly improved the manuscript. Any errors of fact
and interpretation are of course my own.
Note
1. I calibrated dates with IntCal09 (Reimer et al. 2009)
and used OxCal 4.1 for the Bayesian models (Bronk
Ramsey 2008, 2009). Dates are rounded to the nea-
rest 10 years. In the text, I report the statistical me-
dian followed by the 1 sigma range in parenthesis; 2
sigma ranges can be found in Table 2 and the cited
literature.
References Cited
Arellano López, Jorge
1977 Determinación del antiplástico en algunas cerámi-
cas precolombinas de Bolivia y Perú. In Arque-
ología en Bolivia y Perú, Vol. 2, pp. 75–101. Casa
Municipal de la Cultura “Franz Tamayo”, La Paz.
1991 e new cultural contexts of Tiahuanaco. In Huari
administrative structure: prehistoric monumental ar-
chitecture and state government, edited by William
H. Isbell and Gordon F. McEwan, pp. 259–280.
Dumbarton Oaks, Washington, D.C.
Bandelier, Adolph Francis
1910 e islands of Titicaca and Koati. Hispanic Society
of America, New York.
Bandy, Matthew S.
1999 Excavaciones en el montículo de Chiripa. In
Proyecto arqueológico Taraco: 1998 excavaciones en
Chiripa, Bolivia, edited by Christine A. Hastorf,
pp. 13–17. Report submitted to the Unidad Na-
cional de Arqueología, La Paz. Electronic docu-
ment, http://escholarship.org/uc/item/0gt184mg
2001 Population and history in the ancient Titicaca
Basin. Unpublished Ph.D. dissertation, Depart-
ment of Anthropology, University of California,
Berkeley.
Bennett, Wendell C.
1934 Excavations at Tiahuanaco. Anthropological Papers
of the American Museum of Natural History 34:
359–494.
1948 A revised sequence for the South Titicaca Basin.
Memoirs of the Society for American Archaeology 4:
90–92.
Bermann, Marc
1990 Household and empire at Lukurmata, Bolivia.
Ph.D. dissertation, Department of Anthropolo-
gy, University of Michigan, Ann Arbor. University
Microfilms, Ann Arbor.
1994 Lukurmata: household archaeology in prehispanic
Bolivia. Princeton University Press, Princeton.
Bronk Ramsey, Christopher
2008 Deposition models for chronological records.
Quaternary Science Reviews 27: 42–60.
2009 Bayesian analysis of radiocarbon dates. Radiocar-
bon 51: 337–360.
Browman, David L.
1980 Tiwanaku expansion and altiplano economic pat-
terns. Estudios Arqueológicos 5: 107–120.
1998 Lithic provenience analysis and emerging material
complexity at Formative Period Chiripa, Bolivia.
Andean Past 5: 301–324.
Bruno, Maria C.
2008 Waranq waranqa: ethnobotanical perspectives on
agricultural intensification in the Lake Titicaca
Basin (Taraco Peninsula, Bolivia). Unpublished
Ph.D. dissertation, Department of Anthropology,
Washington University, St. Louis.
Burkholder, JoEllen
1997 Tiwanaku and the anatomy of time: a new ceramic
chronology from the Iwawe Site, Department of
La Paz, Bolivia. Unpublished Ph.D. dissertation,
Department of Anthropology, State University of
New York, Binghamton.
Callisaya Medina, Luis M.
2009 Estudio de las percepciones y experiencias en el
templete semisubterráneo de Tiwanaku. Anales de
la Reunión Anual de Etnología 23: 99–110.
Marsh: e founding of Tiwanaku
185
Chávez, Sergio J., and Karen L. Mohr Chávez
1975 A carved stela from Taraco, Puno, Peru, and the
definition of an early style of stone sculpture from
the altiplano of Peru and Bolivia. Ñawpa Pacha
13: 45–83.
Couture, Nicole C.
2002 e construction of power: monumental space
and elite residence at Tiwanaku, Bolivia. Unpub-
lished Ph.D. dissertation, Department of Anthro-
pology, University of Chicago.
2003 Ritual, monumentalism, and residence at Mollo
Kontu, Tiwanaku. In Urban and rural archaeology,
edited by Alan L. Kolata, pp. 202–225. Tiwanaku
and its hinterland: archaeology and paleoecology
of an Andean civilization, Vol. 2, Smithsonian,
Washington, D.C.
Couture, Nicole C., Deborah E. Blom, and Maria C. Bru-
no (editors)
2010 Proyecto arqueológico Jach’a Marka: informe de
investigaciones realizadas en 2008. Report submit-
ted to the Unidad Nacional de Arqueología, La
Paz.
Escalante Moscoso, Javier F.
1994 Arquitectura prehispánica en los Andes bolivianos.
Producciones CIMA, La Paz.
2003 Residential architecture in La Karaña, Tiwanaku.
In Urban and Rural Archaeology, edited by Alan
L. Kolata, pp. 316–326. Tiwanaku and its hinter-
land: archaeology and paleoecology of an Ande-
an civilization, Vol. 2, Smithsonian, Washington,
D.C.
Giesso, Martín
2011 La Producción de Instrumental Lítico en Tiwanaku:
El Impacto del Surgimiento y Expansión del Estado
en las Unidades Somésticas Locales/Stone Tool Pro-
duction in the Tiwanaku Heartland: e Impact
of State Emergence and Expansion on Local House-
holds. BAR International Series 2244. Archaeo-
press, Oxford.
Goldstein, Paul S.
2007 Settlement patterns and altiplano colonization:
new models and evidence from the Tiwanaku di-
aspora. In Sociedades Precolombinas Surandinas:
Temporalidad, Interacción y Dinámica Cultural del
NOA en el ámbito de los Andes Centro–Sur, edit-
ed by Verónica I. Williams, Beatriz N. Ventura,
Adriana B. M. Callegari and Hugo D. Yacobaccio,
Artes Gráficas Buschi, Buenos Aires.
Goodman–Elgar, Melissa
2008 e devolution of mudbrick: ethnoarchaeolo-
gy of abandoned earthen dwellings in the Boliv-
ian Andes. Journal of Archaeological Science 35:
3057–3071.
Hastorf, Christine A. (editor)
1992 Preliminary report on the 1992 excavations at Chi-
ripa, Bolivia by the Taraco Archaeological Project
(Proyecto Arqueológico Taraco–TAP). Archaeo-
logical Research Facility Field Reports, Universi-
ty of California, Berkeley. Electronic document,
http://escholarship.org/uc/item/28z3w6cv
Helsley–Marchbanks, Anne H.
1990 Kk’araña preliminary report #2: search for the re-
ported andesite wall (and associated excavations).
Report submitted to ProyectoWila Jawira. Depart-
ment of Anthropology, University of Chicago.
Janusek, John W.
1994 State and local power in a prehispanic Andean pol-
ity: changing patterns of urban residence in Tiwa-
naku and Lukurmata, Bolivia. Unpublished Ph.D.
Dissertation, Department of Anthropology, Uni-
versity of Chicago.
2003 Vessels, time, and society: toward a chronology of
ceramic style in the Tiwanaku heartland. In Urban
and Rural Archaeology, edited by Alan L. Kolata,
pp. 30–92. Tiwanaku and its hinterland: archae-
ology and paleoecology of an Andean civilization,
Vol. 2, Smithsonian, Washington, D.C.
2004 Identity and Power in the Ancient Andes: Tiwanaku
Cities through Time. Routledge, New York.
2006 e changing ‘nature’ of Tiwanaku religion and
the rise of an Andean state. World Archaeology 38:
469–492.
2008 Ancient Tiwanaku. Cambridge University Press,
Cambridge.
2009 Residence and ritual in Tiwanaku: hierarchy, spe-
cialization, ethnicity, and ceremony. In Domestic
Life in Prehispanic Capitals: A Study of Specializa-
tion, Hierarchy, and Ethnicity, edited by Linda R.
Manzanilla and Claude Chapdelaine, pp. 159–
179. Museum of Anthropology, University of
Michigan, Ann Arbor.
2011 Contextualizando el sitio de Khonkho Wankane:
objetivos, antecedentes, y resultados preliminares
del Proyecto Jach’a Machaca, 2001–2006. Nuevos
Aportes 5:3–30.
Janusek, John W., and Alan L. Kolata
2003 Prehispanic rural history in the Katari Valley. In
Urban and Rural Archaeology, edited by Alan L.
Kolata, pp. 129–172. Tiwanaku and its hinter-
land: archaeology and paleoecology of an Ande-
an civilization, Vol. 2, Smithsonian, Washington,
D.C.
Kolata, Alan L. (editor)
2003 Urban and Rural Archaeology. Tiwanaku and its
hinterland: archaeology and paleoecology of an
Andean civilization, Vol. 2. Smithsonian, Wash-
ington, D.C.
Ñawpa Pacha: Journal of Andean Archaeology Volume 32, Number 2
186
Lémuz Aguirre, Carlos
2001 Patrones de asentamiento arqueológico en la
Península de Santiago de Huata, Bolivia. Un-
published Licenciatura thesis, Carrera de Arque-
ología, Universidad Mayor de San Andrés, La Paz,
Bolivia. Digital version
2004a Normalización de datos de asentamiento en la
cuenca sur del Lago Titicaca. Report submitted to
the Unidad Nacional de Arqueología, La Paz.
2004b Prospección arqueológica. In Diversidad Urbana
en Tiwanaku: Arqueología funeraria y Residencial en
Mollo Kontu. Proyecto Arqueológico Jach’a Marka,
temporada 2001, edited by Deborah E. Blom and
Nicole Couture, pp. 6–11. Report submitted to
the Unidad Nacional de Arqueología, La Paz.
2005 Nuevas consideraciones sobre el tamaño y la po-
blación del asentamiento cívico, ceremonial y resi-
dencial de Tiwanaku. Paper presented at the 14th
Annual Meeting of Ethnology, La Paz.
Lémuz Aguirre, Carlos, and José Luis Paz Soria
2001 Nuevas consideraciones acerca del Periodo For-
mativo en Kallamarka. Textos Antropológicos 13:
93–110.
McIntosh, Roderick J.
1977 e excavation of mud structures: an experiment
from West Africa. World Archaeology 9: 185–199.
Marsh, Erik J.
2012a A bayesian re–assessment of the earliest radiocar-
bon dates from Tiwanaku, Bolivia. Radiocarbon.
54: 203–218.
2012b e emergence of Tiwanaku: domestic practices
and regional traditions at Khonkho Wankane and
Kk’araña. Unpublished Ph.D. dissertation, De-
partment of Anthropology, University of Califor-
nia, Santa Barbara. Electronic document, http://
db.tt/RxlF5ZzW
Marsh, Erik J. (editor)
2008 Proyecto arqueológico Kala Uta: informe de las
excavaciones en el sitio Kk’araña, Tiwanaku. Re-
port submitted to the Unidad Nacional de Arque-
ología, La Paz. Electronic document, http://db.tt/
ce2S9pPF
Mathews, James E.
1992 Prehispanic settlement and agriculture in the mid-
dle Tiwanaku Valley, Bolivia. Unpublished Ph.D.
dissertation, Department of Anthropology, Uni-
versity of Chicago.
1995 A re–evaluation of the Formative Period in the
southern Titicaca Basin. Journal of the Steward An-
thropological Society 23: 83–110.
Mohr Chávez, Karen L.
1988 e significance of Chiripa in Lake Titicaca Basin
developments. Expedition 30: 17–26.
1997 e temple site of Ch’isi on the Copacabana Pen-
insula: a view of local differences and regional simi-
larities within the Yaya–Mama religious tradition.
Paper presented at the 62nd Annual Meeting of
the Society for American Archaeology, Nashville,
Tennessee.
Parsons, Jeffery R.
1968 An estimate of size and population for Middle Ho-
rizon Tiahuanaco, Bolivia. American Antiquity 33:
243–245.
Pérez Arias, Adolfo
2007 Excavación e análisis de la cerámica de Irohito. In
Khonkho e Iruhito: Tercer Informe Preliminar del
Proyecto Jach’a Machaca, edited by John W. Janusek
and Victor W. Plaza Martínez, pp. 217–248. Report
submitted to the Unidad Nacional de Arqueología,
La Paz. Electronic document, http://www.khonk-
howankane.org/janusek_plaza_martinez_2007.pdf
Ponce Sanginés, Carlos
1961 Informe de labores (octubre 1957 – febrero 1960).
Publication No. 1. Centro de Investigaciones Ar-
queológicas en Tiwanaku, La Paz.
1964 Descripción sumaria del Templete semisubterráneo de
Tiwanaku. 1st ed. Publication No. 2. Centro de
Investigaciones Arqueológicas en Tiwanaku, La
Paz.
1970 Las culturas de Wankarani y Chiripa y su relación con
Tiwanaku. Publication No. 25. Academia Nacio-
nal de Ciencias de Bolivia, La Paz.
1981 Tiwanaku: espacio, tiempo y cultura. Editorial Los
Amigos del Libro, La Paz.
1993 La cerámica de la época I (aldeana de Tiwanaku).
Pumapunku (Nueva Época) 4: 48–89.
1995 Tiwanaku: 200 Años de Investigaciones Arqueológi-
cas. Producciones CIMA, La Paz.
Portugal Ortiz, Max
1992 Trabajos arqueológicos de Tiwanaku (I. Parte).
Textos Antropológicos 4: 9–51.
Ralph, Elizabeth K.
1959 University of Pennsylvania Radiocarbon Dates III.
American Journal of Science Radiocarbon Supple-
ment 1: 45–58.
Reimer, Paula J., Mike G. L. Baillie, Edouard Bard, Alex
Bayliss, J. Warren Beck, Paul G. Blackwell, Christopher
Bronk Ramsey, Caitlin E. Buck, George S. Burr, R. Law-
rence Edwards, Michael Friedrich, Pieter M. Grootes,
omas P. Guilderson, Irka Hajdas, T. J. Heaton, Alan G.
Hogg, Konrad A. Hughen, Klaus F. Kaiser, Bernd Kromer,
F. Gerry McCormac, Sturt W. Manning, Ron W. Reimer,
D. A. Richards, John R. Southon, Sahra Talamo, Chris
S. M. Turney, Johannes van der Plicht, and Constanze E.
Weyhenmeyer
2009 IntCal09 and Marine09 Radiocarbon Age Calibra-
tion Curves, 0–50,000 Years cal BP. Radiocarbon
51: 1111–1150.
Marsh: e founding of Tiwanaku
187
Renfrew, Colin
1986 Introduction: peer polity interaction and socio–
political change. In Peer Polity Interaction and So-
cio–political Change, edited by Colin Renfrew and
John F. Cherry, pp. 1–26. Cambridge University
Press, Cambridge.
Roddick, Andrew P.
2009 Communities of pottery production and con-
sumption on the Taraco Peninsula, Bolivia, 200
BC–300 AD. Unpublished Ph.D. D, Depart-
ment of Anthropology, University of California,
Berkeley.
Rowe, John H.
1963 Urban settlements in ancient Peru. Ñawpa Pacha
1: 1–27.
Rydén, Stig
1947 Archaeological Researches in the Highlands of Bo-
livia. Elanders Boktryckeri Aktiebolag, Göteborg,
Sweden.
Smith, Scott C.
2009 Venerable geographies: spatial dynamics, religion,
and political economy in the prehistoric Lake Titi-
caca Basin, Bolivia. Unpublished Ph.D. Disserta-
tion, Department of Anthropology, University of
California, Riverside.
Stanish, Charles
2003 Ancient Titicaca: e Evolution of Complex Societies
in Southern Peru and Northern Bolivia. University
of California Press, Berkeley.
Stanish, Charles, Edmundo de la Vega M., Lee Steadman,
Cecilia Chávez Justo, Kirk L. Frye, Luperio Onofre Ma-
mani, Matthew T. Seddon, and Percy Calisaya Chuquimia
1997 Archaeological survey in the Juli–Desaguadero Re-
gion of Lake Titicaca Basin, southern Peru. Fieldiana
Anthropology, New Series, No. 29. Field Museum
of Natural History, Chicago.
Steadman, Lee
1995 Excavations at Camata: an early ceramic chronol-
ogy for the western Titicaca Basin, Peru. Unpub-
lished Ph.D. D, Department of Anthropology,
University of California, Berkeley.
2007 Ceramic analysis. In Kala Uyuni: An Early Political
Center in the Southern Lake Titicaca Basin: 2003
Excavations of the Taraco Archaeological Project, ed-
ited by Matthew S. Bandy and Christine A. Has-
torf, pp. 67–112. Contributions of the Archaeo-
logical Research Facility, Vol. 64. University of
California, Berkeley.
Steadman, Lee, and Nicole Anthony
2008 Cerámica. In Proyecto Arqueológico Taraco: 2006
Excavaciones en Chiripa, Bolivia, edited by Chris-
tine A. Hastorf, pp. 63–83. Report submitted
to the Unidad Nacional de Arqueología, La Paz.
Electronic document, http://escholarship.org/uc/
item/1w3136jv
Steadman, Lee, Andrew Roddick, and José Luis Capriles
2005 Análisis cerámico. In Proyecto Arqueológico Tara-
co: Informe de las Excavaciónes de la Temporada del
2004 en los Sitios de Kumi Kipa, Sonaji y Chiripa,
edited by Christine Hastorf and Matthew S. Ban-
dy, pp. 65–75. Report submitted to the Unidad
Nacional de Arqueología, La Paz. Electronic docu-
ment, http://escholarship.org/uc/item/4zq0f18f
Stuckenrath, Robert
1963 University of Pennsylvania radiocarbon dates VI.
Radiocarbon 5: 82–103.
Sutherland, Cheryl
1991 Methodological, stylistic and functional ceramic
analysis: the surface collection at Akapana East,
Tiwanaku. Unpublished Master’s esis, Depart-
ment of Anthropology, University of Chicago.
Ulloa Vidaurre, Delfor
2008 Informe de las excavaciones realizadas en el área 1.
In Proyecto Arqueológico Kala Uta: Informe de las
Excavaciones en el Sitio Kk’araña, Tiwanaku, edited
by Erik J. Marsh, pp. 40–49. Report submitted to
the Unidad Nacional de Arqueología, La Paz.
Viviani Burgos, Luis M.
2008 Excavaciones en el área 1: unidades 1, 7, 14, 17,
19, 21. In Proyecto Arqueológico Kala Uta: Informe
de las Excavaciones en el Sitio Kk’araña, Tiwanaku,
edited by Erik J. Marsh, pp. 40–49. Report sub-
mitted to the Unidad Nacional de Arqueología, La
Paz.
