Content uploaded by Dennis Ogburn
Author content
All content in this area was uploaded by Dennis Ogburn on Sep 25, 2017
Content may be subject to copyright.
RADIOCARBON, Vol 54, Nr 2, 2012, p 219–237 © 2012 by the Arizona Board of Regents on behalf of the University of Arizona
219
RECONCEIVING THE CHRONOLOGY OF INCA IMPERIAL EXPANSION
Dennis E Ogburn
University of North Carolina at Charlotte, Department of Anthropology, 9201 University City Blvd., Charlotte, North Caro-
lina 28223, USA. Email: dogburn@uncc.edu.
ABSTRACT. The basic chronology of Inca imperial expansion in Andean South America derives from historical accounts
from the Spanish Colonial era, but several issues with this traditional chronology have arisen in recent decades. Advances in
radiocarbon dating and calibration now give us some ability to refine or rebuild the chronology, and guidelines for obtaining
the most useful dates are discussed. Dates recently obtained from the site of Chamical in the southern highlands of Ecuador
are evaluated according to those guidelines, and they suggest Inca expansion to the north began 1 to 2 decades earlier than
allowed in the traditional chronology. The chronology of Inca expansion presented in the Spanish chronicles is called into
question by these and other dates, and by a reconsideration of the nature of sources of Inca history utilized by Spanish writers.
Evidence suggests a primary Inca form of recording provincial conquests resulted in lists that were ordered geographically.
However, those records were interpreted by colonial writers as being chronologically ordered, which led to written histories
of Inca expansion that are not consistent with the actual historical sequence of events. As a result, the preferred approach to
building a chronology of Inca expansion should be based on 14C dates, with historical sources used to supplement rather than
structure the timeline.
INTRODUCTION
The Inca Empire of Andean South America was rather short-lived, as it expanded rapidly out from
the capital of Cuzco and endured for about a century until the Spanish conquest, which began with
the arrival of Francisco Pizarro in AD 1532. Beyond this general timeframe, precise dating of the
sequence of Inca imperial expansion has been a significant challenge both because of the absence of
written records produced before the Spanish conquest and because of the low resolution of applica-
ble dating techniques. But for scholars interested in studying the short-term processes of Inca impe-
rial expansion and maintenance, a refined chronology is of interest because it is central to under-
standing the duration of the imperial occupation of specific provinces and how Inca strategies in
those areas may have been tied to conditions or events in the capital or in other provinces as the
empire developed.
Given the unlikelihood of discovering a lost document equivalent to a Rosetta Stone for Inca history,
we must focus on nonhistorical means for improving the Inca chronology in terms of establishing or
adjusting the timing of incorporation of different regions into the empire. At present, radiocarbon
dating is the only method that has the potential for providing data of sufficient precision and accu-
racy. The primary questions for archaeologists are whether 1) we maintain the framework based on
historical data and employ 14C dates to refine it, or 2) we give priority to 14C dates and either treat
historical information as supplemental, to be interpreted in light of those dates, or create a chronol-
ogy independent of the historical record. Here, I address those questions, first discussing how we
can obtain data suitable for chronology building even though the accuracy and precision of 14C dat-
ing present a challenge for the short-lived Inca Empire. Toward that end, I suggest criteria for
obtaining and calibrating the most suitable dates for understanding Inca expansion in the provinces.
I then present a new set of dates, derived from the Inca site of Chamical in the northern quarter of
the empire, which adhere to those criteria. The dates suggest that the Incas expanded into this region
1 or 2 decades earlier than the historical chronology indicates. In light of the dates from Chamical
and the series of reported dates that indicate early expansion in the southern part of the Inca realm,
I argue we need to reconceive the chronology of Inca expansion based on 14C dates supplemented
with a reframed view of historical data. In particular, I find the historical outlines of Inca expansion
to be problematic because the Incas gave priority to geographical rather than chronological order in
recording lists of provincial conquests pertaining to the major expansion of the empire. This resulted
in data that were mismatched with the linear narratives of Inca history that writers sought to create
220 D E Ogburn
during the Spanish Colonial era, i.e. this created fundamental problems with the accounts that have
been our principal sources for understanding the chronology of Inca expansion.
HISTORICAL FRAMEWORK OF THE TRADITIONAL CHRONOLOGY
For decades, scholars have relied on the chronology of Inca expansion proposed by John H Rowe
(initially in Rowe 1944, then more fully developed in Rowe 1945). Before those publications, esti-
mates for the creation and duration of the Inca Empire were based on differing interpretations of the
myriad inconsistent dates found in the chronicles produced during the Spanish Colonial era. Work-
ing with the same set of sources used by previous scholars, Rowe argued that the dates provided in
the AD 1586 chronicle of Miguel Cabello Balboa (1945) were the most reasonable and believable
of the various dates given for events in Inca history. Given the dearth of other absolute dating meth-
ods that could be employed in the Andes at the time, Rowe proposed a chronology of the Inca Empire
based on Cabello Balboa’s dates for the reigns and conquests of the last several emperors, with the
assumption that dates supplied for earlier rulers were not credible. This chronology (referred to here
as the “traditional chronology”) was based on what Rowe termed the “standard version” (Rowe
1945:272) of Inca history, which he thought most closely reflected the “official history” as sanc-
tioned by the Inca rulers. While acknowledging that the official history was likely to have been
manipulated and censored because of politics, Rowe (1945:268) suggested that the order of con-
quests as passed down in the accounts was likely to be reasonably preserved. In sum, this traditional
chronology was significant not only for proposing plausible absolute dates for Inca expansion but for
also making a convincing argument for the rapid expansion and short duration of the empire.
The heart of the traditional chronology (Table 1, Figure 1) focuses on 4 phases of Inca imperial
expansion and dominance, beginning with the creation of the empire by the Inca leader Pachacuti
(Inca Yupanqui) after he assumed rule from his father in AD 1438, and ending with the death of his
grandson, Huayna Capac in AD 1525. The final years of the empire were marked by a struggle for
succession and did not result in any significant territorial conquests. The end of an independent Inca
Empire is marked by Pizarro’s capture of the eventual winner of that contest, Atahuallpa, in
AD 1532.
Much of the discussion of Inca chronology has centered on the Cuzco region (e.g. Bauer 1992) and
the reigns of these later rulers, but the archaeology of imperial expansion outside the Cuzco region
requires framing chronology specifically in relation to the dates of incorporation of different prov-
inces or sections of the empire. Many studies of Inca expansion do not directly tackle chronological
questions, often relying on accepted standardized dates for different regions, such as the AD 1476
date for the beginning of the Late Horizon of Peru that was proposed by Rowe (1962). Those that
give more specific dates for the Inca arrival typically estimate the date by analyzing the historical
Table 1 The traditional chronology based on Cabello Balboa (1945).
Year (AD) Events
1438 Pachacuti deposes his father Viracocha, embarks on campaigns of imperial conquest.
1463 Pachacuti remains in Cuzco while sons including Topa Inca lead the army on cam-
paigns of expansion.
1471 Topa Inca succeeds Pachacuti, embarks on further campaigns of expansion.
1493 Topa Inca dies, succeeded by Huayna Capac, who achieved only limited conquests
during his reign.
1525 Huayna Capac dies, succeeded by Huascar, whose rule was contested by Atahuallpa.
1532 Pizarro captures Atahuallpa in Cajamarca.
Reconceiving the Chronology of Inca Imperial Expansion 221
accounts to determine which Inca ruler is given credit for conquering the region or incorporating it
through peaceful submission, then relating that to the traditional chronology (e.g. Menzel 1959;
Lippi and Gudiño 2010). The great majority of Inca territory was said to have been incorporated by
Topa Inca (Figure 1), both during the rule of his father, Pachacuti, and after the latter’s death. So the
dates associated with Topa Inca’s efforts are particularly important to studying Inca expansion; those
include not only the years of his reign but also the year he began to undertake conquests during his
father’s rule.
RADIOCARBON DATES AND THE CHRONOLOGY OF INCA EXPANSION
Since the 1940s, researchers have widely employed the traditional chronology (Bauer 1992:38),
though in recent times they often note skepticism about the precision of the dates, particularly
regarding the earliest phases of expansion. Archaeologists have seldom relied to any significant
extent on 14C dates for understanding Inca imperial chronology on the larger scale; nonetheless,
numerous dates have been obtained from excavated Inca contexts. Typically, these have not varied
Figure 1 Extent of the Inca Empire along western South America, indicating areas con-
quered according to the traditional chronology (after Rowe 1945), location of Chamical
within the southern highlands of Ecuador, and the average position of the ITCZ along
the eastern Andes during the austral summer (according to Mayle et al. 2000).
222 D E Ogburn
appreciably from the traditional chronology, and a statistical analysis of combined dates (Adamska
and MichczyÒski 1996) seemed to support the historical dates in general terms.
However, several arguments have been made in favor of refining or remaking the imperial Inca
chronology. Bauer (1992:39) argued that because we have no evidence that the Incas or Andean
groups recorded the passage of years, we have no way of knowing if Cabello Balboa obtained his
dates from an indigenous source or if he invented dates he felt most plausible. While his analysis of
available 14C dates from the Cuzco region supported the model of rapid imperial expansion within
a few generations, Bauer (1992:46) suggested AD 1400 as a better approximate date for the begin-
ning of the Inca period in the Cuzco region. This is several decades earlier than the traditional date
of AD 1438 for the creation of the empire, but it is important to note that Bauer’s Inca period spe-
cifically marked the appearance of Inca imperial style pottery in the region rather than the expansion
of Inca political domination outside the Cuzco heartland. Also, beginning in the 1990s, a series of
published 14C dates from Inca contexts in the southern provinces of the empire in Argentina and
Chile have led some to argue that Inca expansion in the south may have occurred several decades
earlier than believed (e.g. Williams and D’Altroy 1998; Schiappacasse 1999; D’Altroy et al. 2007),
opening up the question of how much faith we should put in Cabello Balboa’s dates for both the
beginning of Inca imperial expansion and the incorporation of the southern provinces. A different
argument for producing a chronology independent of historical data was made by Zuidema (1982)
based on the theoretical position that Inca history as recorded in the Spanish Colonial era is prima-
rily mythological, reflecting Inca ideological concepts of social structure, religion, time, etc., rather
than presenting a linear Western-style historical narrative.
Given these arguments, although there is well-founded skepticism regarding the extent to which 14C
dating can refine or remake the chronology of Inca expansion (e.g. McEwan 2008), it is worth fur-
ther consideration in light of ongoing improvements in methods. Such attempts at improving the
chronology could follow 2 paths: 1) Refining the traditional chronology by establishing what are
considered better dates for the various waves of imperial expansion as carried out during the reigns
of Pachacuti, Topa Inca, and Huayna Capac. Adjustment of dates for the earlier campaigns of expan-
sion could have implications for the timing of subsequent conquests in other regions, the order of
which are understood in general outline from historical sources. 2) Establishing a chronology based
primarily on 14C dates, obtained from Inca contexts in many different provinces of the empire. This
could either be fully independent of the historical record, or employ historical data in a secondary
role. This approach would demonstrate the progression of expansion rather than assuming the verac-
ity of the historical sequence of conquests. However, it presents the formidable challenge of obtain-
ing a large number of 14C dates from clearly identified Inca imperial contexts from excavation
projects throughout the Inca realm. I will return to these 2 approaches in the discussion below, argu-
ing that there are compelling reasons to follow the second course.
Regardless of the approach favored, certain criteria need to be met to productively improve the chro-
nology of Inca expansion. Obviously, we need to aim for 14C dates of the highest possible accuracy
and precision. Of course, the statistical nature of 14C dating will not give us the level of precision to
be able to establish the actual order of incorporation of different provinces that may have been con-
quered within the same decade. But we should be able to discern differences of 30 to 40 yr, and
under the best circumstances we may be able to discern differences in the range of 20 yr or so. To
this end, we should follow the established recommendations for producing the highest quality dates,
e.g. 13C corrections should be measured rather than estimated, old-wood problems should be
avoided by dating short-lived plant materials such as seeds and annuals, and revising the timing of
the Inca conquest in any given province based on multiple rather than single 14C dates.
Reconceiving the Chronology of Inca Imperial Expansion 223
Another obvious requirement is sampling from contexts likely to represent the earliest Inca presence
in a region. This can be challenging to determine because any individual site or building could have
been constructed at any time during the Inca occupation of the area, and the lowest excavated Inca
strata will not necessarily date the Inca arrival. Moreover, we do not have a good sense of how long
after conquest the Incas began constructing imperial administrative sites, way stations, etc. within a
province—it could have been immediate, or after a delay of a decade or more. The best candidates
for marking the earliest Inca presence are sites related to the advance of the Inca army, such as forts
and other military outposts constructed and utilized during the conquest of a region. These could be
relatively rare, however, as the Inca army conquered many provinces quickly and was probably not
stationary long enough to require the construction of such facilities in many regions. Furthermore,
not all Inca forts were associated with expansion. Many were constructed to defend the borders of
the empire, house garrisons, or maintain control over strategic locales over the long term, and would
not necessarily mark the arrival of the Incas in an area.
Regardless of the site type, it is vital that samples for dating come from securely established contexts
related to the Inca imperial presence. Ideally, the site sampled would lack a pre-Inca component to
avoid the possibility of encountering a mixture of earlier and Inca-era material. Preference should be
for contexts within Inca imperial sites rather than local sites containing Inca-related material: local
sites could produce problematic dates if Inca-style ceramics initially arrived in a region through
exchange prior to conquest.
The final major requirement is determining which calibration curve to apply to any specific set of
14C dates. Dealing with the offset between calibration curves for the Northern and Southern hemi-
spheres may be our biggest challenge because even though the Inca Empire was primarily located
within the Southern Hemisphere, it also encompassed land within the Northern Hemisphere, span-
ning from about 34S to 1N. As a result of the offset, during the timeframe of interest (~AD 1400–
1532) dates calibrated with the Northern Hemisphere curve (IntCal09; Reimer et al. 2009) can be
between 15–20 yr earlier than those calibrated with the Southern Hemisphere curve (SHCal04;
McCormac et al. 2004). Twenty years is a significant difference for a relatively short-lived phenom-
enon like the Inca Empire, and inconsistent application of the different calibration curves could lead
to chronological confusion. Clearly, the geographical spread of the Inca realm means that a single
calibration curve is not going to be appropriate for all parts of the empire, but no consensus has
emerged on how to deal with this situation.
The issue of choosing a curve is complex because the air masses in the 2 hemispheres have different
concentrations of 14C. This difference exists because the greater extent of ocean in the Southern
Hemisphere leads to a greater exchange of CO2 between ocean and atmosphere compared to the
Northern Hemisphere. This produces an offset between calibrated ages, which varies over time
rather than staying constant (McCormac et al. 2002). Moreover, the division between the hemi-
spheric air masses is not the Equator, but the Intertropical Convergence Zone (ITCZ), where
tradewinds from north and south converge and mix the air masses. The ITCZ is not fixed, but moves
seasonally as influenced by the position of the sun and the topography of the landforms it passes
over. This complicates choosing the appropriate calibration for 14C dates because in some regions
the seasonal movement of the ITCZ brings CO2 from the Northern Hemisphere for part of the year
and from the Southern Hemisphere during the other part (McCormac et al. 2004:1088).
This is of direct concern because the seasonal shift of the ITCZ impacts a good portion of the Andes.
During the austral winter, the ITCZ lies primarily to the north, several degrees above the Equator.
But during austral summer, it exhibits a meridional shift, bending southeast down along the eastern
side of the Andes and turning east around Bolivia before heading back north (Figure 1). This brings
224 D E Ogburn
the Northern Hemisphere atmosphere close to the Andes during a good part of the year. The key
concern for calibrating 14C dates is the predominant source of air during the growing season, as this
affects the uptake of CO2 (and thus 14C) in plants (McCormac et al. 2004:1088). As Finucane et al.
(2007) argue for the Ayacucho region of Peru, due to the southward shift of the ITCZ in the primary
growing season, the austral summer, much of the CO2 absorbed by plants should represent the
Northern rather than the Southern Hemisphere. During that time of year, easterly winds prevail
along the Andes down to ~20S (Garreaud 2009), so the weather systems and the moisture they
bring derive from the east, across the ITCZ. South of Bolivia, however, the influx of Northern Hemi-
sphere air is much less significant.
Long-term migration of the austral summer position of the ITCZ is another potential issue (McCor-
mac et al. 2004). For the pre-Columbian Andes, the most recent major change in the ITCZ was a
southward shift that occurred around 3000 yr ago (Mayle et al. 2000). But on a finer scale, it appears
the ITCZ may have shifted to its southernmost position during the Little Ice Age, from ~AD 1420
to 1560/1640 (Sachs et al. 2009) and subsequently retreating to its current position. This would have
encompassed most, if not all, of the duration of Inca imperial expansion, but not all of Inca territory.
In general terms, because of this seasonal movement of the ITCZ, the IntCal curve may be the most
relevant for highland sites from Ecuador down to northern Bolivia. In highland areas in the mid-lat-
itudes from central Bolivia to the south, the SHCal curve may be most appropriate. The coastal
regions are subject to different conditions. Coastal desert zones are characterized by winds from the
west and south; thus, they are dominated by well-mixed Southern Hemisphere air. The Southern
Hemisphere calibration curve should therefore be applied to coastal Chile and Peru. In contrast, the
coastal Ecuadorean climate is far wetter and likely subject to the influx of Northern Hemisphere air
during austral summer, so the Northern Hemisphere calibration may be more suitable. Given the
apparent southward shift of the ITCZ during the Little Ice Age (Sachs et al. 2009), the Northern
Hemisphere curve may be most appropriate for the far southern coast of Ecuador, despite its currently
having an arid climate similar to the northern coast of Peru.
As a suggested approach to calibration in the different zones, I must point out that this is both tenta-
tive and general and that there are various complicating issues that cannot be easily resolved. For
one, this is only a basic characterization of the complex climatic dynamics of western South Amer-
ica, and does not deal with the significant regional variability that typifies many areas. For many
regions, including the Peruvian and Ecuadorean highlands, the air is going to represent some level
of mixture between the Northern and Southern Hemisphere atmospheres, but we currently have no
way of gauging the level of mixture. Plus, much of Inca territory lies within transitional areas, such
as the Peruvian western slopes near the coast, which fall between climate regimes heavily influ-
enced by the seasonal movement of the ITCZ and those dominated by Southern Hemisphere air
masses. Choice of calibration curve in those cases can thus be problematic, and we will need to con-
sider strategies for mixing the calibration curves, which can be done with software such as OxCal
(Bronk Ramsey 2009).
Another issue to consider is the nature of the materials dated and whether they were exposed to
Northern Hemisphere air while they were growing. In the highlands of Peru and Ecuador, food crops
grown during the austral summer certainly were, but some irrigated crops and wild plants growing
along water courses could also have been growing during other parts of the year. Also, the Incas
moved massive quantities of foods, clothing, and other materials between different climatic zones,
which means sometimes we will not know if the same calibration curve can even be applied to all
excavated remains within a single Inca site.
Reconceiving the Chronology of Inca Imperial Expansion 225
Because we cannot easily quantify all of these potential effects, the best approach may be for archae-
ologists to publish their dates using the latest versions of both calibration curves. This could help
limit any tendency for researchers to pick from the different curves to find the best fit for their data
or preconceptions. Then, for matters of interpretation and refining chronology, we may employ the
different curves as recommended above, but researchers should describe to the best of their knowl-
edge the reasons why one curve would be more suitable than the other, depending on the individual
sample and context.
Finally, the shapes of the calibration curves during the period of interest present an issue. For 14C
ages ~525–410 BP, the IntCal09 curve is quite favorable: it is relatively straight, producing narrowly
focused calibrated dates unaffected by multiple intercepts. Moving outside of that window, cali-
brated dates have poorer resolution because dips and spikes in the curve produce multiple intercepts,
with an especially pronounced effect on those dates that have larger errors. In effect, that curve is
best suited for dating objects and associated events between about AD 1415 and 1460. Because of
the hemispheric offset, the SHCal04 curve is most optimal for 14C ages between ~575–450 BP,
applicable to objects or events dating between about AD 1410 and 1470.
The lower limit for good resolution in calibrated dates could be problematic for improving the chro-
nology of Inca expansion if any conquests outside the Cuzco region actually occurred prior to
~AD 1410; fortunately, there is little to suggest any significant expansion at such an early date. The
upper limit restricts the potential for refining the dates for provinces incorporated much later than
~1460–1470 and also makes it difficult to distinguish events of the early Spanish Colonial era from
late Inca imperial times. This impacts dating of expansion during what is thought to be the latter part
of Topa Inca’s reign, and the entire reign of Huayna Capac seems to fall in this calibration “black
hole.” However, if Cabello Balboa’s dates are not grossly inaccurate and most Inca conquests out-
side the Cuzco region occurred during the reigns of Pachacuti and Topa Inca, as most historical
accounts suggest, then much of Inca imperial expansion would have occurred within the optimal
calibration window of AD 1410/15 to AD 1460/70. These limits mean that we can create a chronol-
ogy of Inca expansion that is based mostly, but not entirely, on 14C dates.
CHAMICAL: THE INCA INCURSION IN THE SOUTHERN HIGHLANDS OF ECUADOR
For exploring the potential to improve the chronology of Inca expansion, I present a set of 7 14C dates
from Chamical, a site that appears associated with the arrival of the Incas in the southern highlands
of Ecuador. Most historical accounts, including chronicles whose authors gathered most of their evi-
dence from Cuzco sources (e.g. Betanzos 1987; Sarmiento de Gamboa 2007) and others who had
first-hand experience in Ecuador (Cabello Balboa 1945; Cieza de León 1985), agree that the south-
ern highlands of Ecuador were conquered by Topa Inca during the reign of Pachacuti. According to
the traditional chronology, this would have been sometime between AD 1463 and 1471. This area
was populated by a subgroup of Cañaris, who apparently resisted incorporation into the Inca Empire.
This resistance could have led to the construction of temporary outposts by the Inca army.
Chamical is a small site situated on a point overlooking the confluence of the Paquishapa and León
rivers at 1225 m, 14 km west of the town of Oña. The compound was constructed of fieldstone walls
with mud mortar, including a main rectangular compound measuring 56 × 36 m and 2 separate
smaller rectangular structures. Inca imperial-style ceramics were found distributed on the surface
outside the compound walls, which included sherds with polychrome decoration (Figure 2) as well
as neck fragments and rope nubbins from aribalo-style vessels. Test excavations conducted within
the main compound in 2008 uncovered shallow deposits with few remains, indicating a limited
occupation with no evidence of pre-Inca activity.
226 D E Ogburn
There were no traces of Inca roads leading to the site, making it unlikely that Chamical served as an
administrative site. The limited excavation finds included a sling stone and several bone arrow
points, suggesting a military focus for Chamical, as does the walled enclosure plan with a single
entrance. The limited occupation indicates the site may have been constructed as a short-term out-
post, possibly to control or monitor the strategic location at the meeting of 2 major rivers. Thus, if it
were constructed during the Inca military advance rather than as a later, internal control point,
Chamical could help to establish the date of imperial expansion into the southern sierra of Ecuador.
Seven samples collected during the test excavations (Table 2) were dated by accelerator mass spec-
trometry (AMS) at the University of Georgia Center for Applied Isotope Studies (lab code UGa), and
the dates were calibrated using OxCal v 4.1 (Bronk Ramsey 2009; Table 3). The choice of calibra-
tion curve makes a difference of roughly 20 yr in these dates, with the Northern Hemisphere calibra-
tion producing dates earlier than the Southern Hemisphere curve. Because Chamical lies within the
climatic zone of the highlands that is greatly influenced by the ITCZ, the Northern Hemisphere cal-
ibration curve IntCal09 (Reimer et al. 2009) is seen as the most suitable for interpretation of the
dates. Two of those calibrated dates (UGa-3459 and UGa-8801) are consistent with the timeframe
for the Inca conquest of southern Ecuador according to the traditional chronology (i.e. the AD 1463–
1471 date range derived from Cabello Balboa). One date (UGa-3460) was associated with rubble
that blocked the entrance into the compound and appears to represent activity during the early Colo-
nial era. The second most recent date (UGa-8802) falls mostly within late Inca imperial times, sug-
gesting limited use of the site may have continued for several decades.
Of most relevance are the 3 earliest dates, which cluster together and likely represent the construc-
tion phase of the site. These included a seed incorporated within the mortar of the enclosure wall
(UGa-3458), a fragment from the outermost layer of a wooden post underneath a prepared floor in
Figure 2 Selection of polychrome Inca ceramics from Chamical; surface col-
lection.
Reconceiving the Chronology of Inca Imperial Expansion 227
one room (UGa-3457), and a piece of charcoal from an ash layer in the lowest stratum of the site
(UGa-8803). To more precisely gauge the arrival of the Incas in the region from this set of dates, I
used the Bayesian analysis (Bronk Ramsey 2009) functions of OxCal (v 4.1) to determine the prob-
able start boundary of the phase of Inca occupation at Chamical. The only assumptions included in
the Bayesian modeling were that the main body of 6 dates represented a single occupation phase and
that the most recent date (UGa-3460) postdated the Inca occupation. With the IntCal09 calibration
(Table 3, Figure 3), the start of the phase falls within the ranges of AD 1432–1451 (1) and
AD 1412–1457 (2); this places the Inca incursion 1 to 2 decades earlier than the mid-AD 1460s
timeframe of the traditional chronology.
As noted above, the choice of calibration curve makes a distinct difference in reading of Inca chro-
nology. In this case, the Southern Hemisphere curve produces broader start boundary ranges of
AD 1446–1477 (1) and AD 1416–1495 (2) (Table 3, Figure 4); the 1 timeframe is consistent
Table 2 Context of 14C samples collected from Chamical.
Lab # Sample # Material Context
UGa-3457 Cham-08-01 Wood Fragment from the outermost layer of a post under-
neath a prepared clay floor, Unit 2-B
UGa-3458 Cham-08-02 Seed Plant fragments extracted from mortar of the NE enclo-
sure wall
UGa-3459 Cham-08-03 Charcoal Burnt cane or reed (roofing material?), from above pre-
pared clay floor, Unit 2-A
UGa-3460 Cham-08-04 Charcoal Rubble filling primary entrance into the compound, N
corner of site, Unit 4-A
UGa-8801 Cham-11-01 Wood Fragment of post below prepared floor, Unit 2-C
UGa-8802 Cham-11-02 Charcoal Burnt cane or bamboo, above floor, Unit 2-A
UGa-8803 Cham-11-03 Charcoal Piece from an ash layer in the lowest stratum of the site,
Unit A
Table 3 14C dates from Chamical, Ecuador, calibrated with OxCal v 4.1; dates calibrated individu-
ally with no prior assumptions, start boundary of phase is derived from Bayesian modeling of the
dates as a group (see Figures 3 and 4).
Lab #
(UGa-)
13C
(‰)
14C age
(BP) cal AD 1
(IntCal09) cal AD 2
(IntCal09) cal AD 1
(SHCal04) cal AD 2
(SHCal04)
3457 –26.1 450 ± 30 1427–1454 (68.2%) 1414–1480 (95.4%) 1445–1491 (68.2%) 1434–1506 (82.6%)
1588–1618 (12.6%)
3458 –11.4 450 ± 25 1431–1451 (68.2%) 1419–1468 (95.4%) 1446–1485 (68.2%) 1439–1502 (87.9%)
1595–1613 (7.5%)
3459 –26.4 410 ± 25 1443–1480 (68.2%) 1435–1515 (87.7%)
1600–1618 (7.7%) 1460–1501 (49.6%)
1595–1612 (18.6%) 1451–1513 (56.6%)
1545–1624 (38.8%)
3460 –25.2 300 ± 25 1522–1574 (50.0%)
1627–1646 (18.2%) 1495–1602 (69.9%)
1616–1651 (25.5%) 1516–1540 (16.5%)
1625–1660 (51.7%) 1507–1586 (35.2%)
1618–1669 (59.0%)
1787–1793 (1.2%)
8801 –25.5 410 ± 25 1443–1480 (68.2%) 1435–1515 (87.7%)
1600–1618 (7.7%) 1460–1501 (49.6%)
1595–1612 (18.6%) 1451–1513 (56.6%)
1545–1624 (38.8%)
8802 –25.6 380 ± 25 1452–1498 (48.9%)
1505–1512 (5.1%)
1601–1616 (14.2%)
1446–1523 (67.0%)
1572–1630 (28.4%) 1490–1513 (15.6%)
1547–1623 (52.6%) 1464–1628 (95.4%)
8803 –23.8 440 ± 25 1433–1456 (68.2%) 1422–1481 (95.4%) 1450–1491 (68.2%) 1443–1505 (81.9%)
1590–1616 (13.5%)
Start boundary of phase 1432–1451 (68.2%) 1412–1457 (95.4%) 1446–1477 (68.2%) 1416–1495 (95.4%)
228 D E Ogburn
with the traditional chronology, while the 2 timeframe is too broad to be of much use in addressing
this question. But given the climatic situation of the southern sierra of Ecuador, it is hard to argue
that the CO2 taken in by plants during the growing season was purely derived from the Southern
Hemisphere. Furthermore, given the evidence that the ITCZ had shifted further southward during
the Little Ice Age (Sachs et al. 2009), it seems most likely that plants in the area received much of
their CO2 from Northern Hemisphere air. Thus, exclusive use of the Southern Hemisphere calibra-
tion curve cannot be considered appropriate in this case. If we allow that the air in the region during
the growing season was more of a mixture from both hemispheres, we could estimate a date that
roughly splits the difference between the 2 calibrated ranges, giving us a date range centering on the
AD 1450s. Whether we rely on that estimate or entirely on the Northern Hemisphere calibration, the
ramifications of the dates from Chamical are that the Incas seem to have expanded into southern
Ecuador a decade or more earlier than presumed in the traditional chronology.
Figure 3 Bayesian analysis of 14C dates for the phase of Inca occupation of Chamical, Ecuador, calibrated
with OxCal 4.1 using the IntCal09 curve. Dotted line indicates AD 1463, the date after which the region
should have been incorporated into the Inca Empire according to the traditional chronology.
Reconceiving the Chronology of Inca Imperial Expansion 229
DISCUSSION: RECONCEIVING THE CHRONOLOGY OF INCA EXPANSION
Because the dates from Chamical indicate an earlier Inca expansion into southern Ecuador, they
point to a need to revise the traditional chronology, but not necessarily to dismiss it entirely. A stron-
ger case for modifying the chronology can be made when also considering the body of early 14C
dates from Inca contexts in Argentina and elsewhere in the southern area of the empire (Williams
and D’Altroy 1998; Schiappacasse 1999; D’Altroy et al. 2007), which suggest Inca expansion in
that area 3 to 6 decades earlier than the timeframe of the traditional chronology. A difference of a
few decades is not major in many archaeological contexts. But in terms of Inca expansion and the
implementation of imperial strategies of control, that timespan can be quite significant. In particular,
it could allow for much greater integration of a region and its people into the Inca state. It also allows
for another generation to grow up under imperial control, and provides more time for imperial forms
of organization to become embedded in local culture. In essence, uncertainty in the length of Inca
occupation arising from conflicts in different sources of chronological information can be problem-
atic for the study of Inca imperialism. Thus, it is imperative to develop the most accurate chronology
possible with available methods.
Figure 4 Bayesian analysis of 14C dates for the phase of Inca occupation of Chamical, Ecuador, cali-
brated with OxCal v 4.1 using the SHCal04 curve. Dotted line indicates AD 1463, the date after which
the region should have been incorporated into the Inca Empire according to the traditional chronology.
230 D E Ogburn
Given that we cannot rely solely on historical sources for chronological data, we must decide on the
appropriate approach for integrating 14C dates with historical information to improve the chronol-
ogy of Inca expansion. Possible approaches to that problem vary depending on how much credence
is given to details of recorded Inca history, which is a challenging question because the historicity
of the chronicles and other sources treating the Inca Empire has been subject to much debate (e.g.
Zuidema 1964; Julien 2000). On one hand, Inca historical narratives were clearly subject to a variety
of alterations because of the passage of time before being written down, translation and filtering
through differing languages and cultures, manipulation by competing Inca factions, and alteration to
fit differing agendas of Spanish writers, among other things. On the other hand, recorded stories
derived from oral Inca histories can at times preserve events in accurate detail (Ogburn 2004). Thus,
the historical record can neither be accepted uncritically nor dismissed outright. In the end, we are
faced with the choice between the 2 primary approaches to refining the chronology of Inca expan-
sion stated in the introduction: 1) maintaining the framework based on historical data and employing
14C dates to refine it; or 2) giving priority to 14C dates and either treating historical information as
supplemental, to be interpreted in light of those dates, or creating a chronology entirely independent
of the historical record. As discussed below, an argument can be made for the first option, but it is
problematic and there are instead compelling reasons to abandon the traditional chronology in favor
of one based primarily on 14C dates.
Refining the Traditional Chronology with Radiocarbon Dates
Refining the traditional chronology would entail maintaining the basic outline of imperial expan-
sion, i.e. adhering to the sequence of territories incorporated by the different emperors as outlined by
Cabello Balboa and other chroniclers (represented in basic form in Figure 1). Modifying the chro-
nology would focus on adjusting Cabello Balboa’s dates for the primary stages of Inca expansion
based on 14C dates from the provinces in combination with analysis of data from other historical
sources. For example, taking into account the dates from Chamical, the entire wave of expansion led
by Topa Inca and his brothers should have begun approximately 10 to 20 yr earlier than the AD 1463
date provided by Cabello Balboa. Pushing this stage of expansion back in time suggests the need to
reassess the AD 1438 date for the commencement of imperial conquests led by Pachacuti outside of
the Cuzco region. Whether that should be pushed back 1 or 2 decades or even more is hard to esti-
mate without new data from the field.
A need for this type of adjustment to the traditional chronology is not surprising. Acceptance of
Cabello Balboa’s dates as the best approximation was a prudent approach at a time when there was
no method for deriving more accurate dates, as Rowe (1945:265) noted. But we have the long-stand-
ing issue of not knowing how Cabello Balboa obtained his dates. We do not know the identity of
possible indigenous sources (Rowe 1945:277) or whether Cabello Balboa simply arrived at his own
best estimates (Bauer 1992:39). While those dates were certainly the most “plausible” ones avail-
able from the historical record, plausibility alone does not ensure high accuracy. As Rowe (1946:
274) and others (e.g. Bauer 1992:36–7) have pointed out, while the Incas noted the passing of the
seasons from year to year, they did not appear to keep count of the passage of years. However, more
recently Rowe (2008; a posthumous translation of a lecture from 1996) argued that the Incas were
in fact interested in recording elapsed time, but used rounded numbers in giving these time intervals.
This resulted in “approximations, not precise measures” (Rowe 2008:159) of years in the written
accounts. Either way, the chronological data given by Inca informants would have been impacted.
For example, ages at death of the different emperors, which were significant chronological markers
used by chroniclers like Cabello Balboa, were typically rounded off to the nearest decade. It is also
clear that not even the Inca nobility of Cuzco knew the specific ages of emperors, as revealed in the
Reconceiving the Chronology of Inca Imperial Expansion 231
reports collected by Viceroy Toledo in the late 16th century (Levillier 1940; Covey 2006). Further-
more, there are a number of issues regarding which rulers were included in Inca king lists, which
were constructed subject to the politics of Inca and Spanish interests (Covey 2006); these could also
impact chronology building. Therefore, there is little reason to suspect that Cabello Balboa had
access to data that were more accurate than those available to other writers.
In refining the traditional chronology, it may not be difficult to find historical data to argue for shifts
in Cabello Balboa’s dates on the order of 1 or perhaps 2 decades. For example, to accommodate the
14C dates from Chamical, a case can be built to support the conquest of southern Ecuador earlier than
indicated by Cabello Balboa. This can be done by drawing on various accounts regarding the
Emperor Huayna Capac’s place of birth, age, and date of death, as follows: It was widely reported
that Huayna Capac was born in the Inca center of Tomebamba in the southern highlands of Ecuador
(e.g. Cabello Balboa 1945; Betanzos 1987; Sarmiento de Gamboa 2007). This implies that the area
must have been incorporated into the empire sometime prior to his birth, allowing for the passage of
a few years for the founding and construction of the Inca city. Huayna Capac died around AD 1525–
1528; while not very precise, this date is probably our most accurate because it was very late in the
history of the empire and there are multiple reports of his death that were derived from informants
with first-hand knowledge (e.g. Cieza de León 1985; Betanzos 1987). It appears he was somewhere
around 70 to 80 yr old, according to various Spanish chroniclers (e.g. Betanzos 1987; Sarmiento de
Gamboa 2007) and direct testimony of Inca nobility in the Cuzco area (Levillier 1940). An age of
70 indicates that he would have been born by AD 1455, indicating that Tomebamba must have been
conquered somewhat earlier. Interestingly, Rowe (2008) made a similar analysis of Huayna Capac’s
life according to data provided by Betanzos (1987). He arrived at dates that suggest the ruler was
born in AD 1457 or 1458, and that Topa Inca succeeded as ruler at the same time. In both cases, the
estimates of Huayna Capac’s birth run counter to the post-AD 1463 date from Cabello Balboa, but
are consistent with the AD 1440–1455 range indicated by the 14C dates from Chamical.
Similar arguments can be made for revising the traditional chronology to accommodate new 14C
dates that do not fit with Cabello Balboa’s dates. In the end, however, this approach is very problem-
atic because it hinges on arguing that specific pieces of information are more accurate than others.
Without corroboration from other lines of evidence, this would amount to researchers choosing the
historical facts they deem most accurate in the attempt to refine the traditional Inca chronology or
build a new one. As shown by Covey (2006), there is wide inconsistency in the chronological infor-
mation embedded in the historical sources, which could easily result in a profusion of conflicting
chronologies. In a sense, one of the benefits of following Rowe’s suggested dates for more than half
a century is that we have mostly avoided such a problem. Because it is desirable to continue avoid-
ing that problem, it would be preferable to follow a different course.
Abandoning the Traditional Chronology and Giving Priority to Radiocarbon Dates
The need for a different approach to refining the chronology of Inca expansion is made even stron-
ger by the body of early 14C dates from Inca contexts in Argentina and elsewhere in the southern
area of the empire (e.g. Williams and D’Altroy 1998; Schiappacasse 1999; D’Altroy et al. 2007).
Fully assessing the implications of all of those dates will require evaluating them individually in
terms of the criteria outlined above. However, even a basic review of the dates as calibrated with the
SHCal04 curve strongly suggests Inca incorporation of at least some southern provinces sometime
between AD 1410 and 1440. This is a significant deviation from the post-AD 1471 timeframe of the
traditional chronology and points to a sequence of provincial expansion out of line with that
described in most accounts.
232 D E Ogburn
The conflict between those 14C dates and the traditional chronology cannot be easily explained by
drawing from other historical information, and indicates a clear need to reconsider both the nature
of Inca history (D’Altroy et al. 2007) and the extent to which we can reliably draw from the recorded
accounts for our archaeological reconstructions of Inca imperial expansion. This divergence from
the historical record does not necessarily mean we need to dismiss the potential for the written
accounts to provide any historical information pertinent to chronology. But what is specifically
called into question is the accuracy of the general outline of imperial expansion given by Cabello
Balboa and other chroniclers, including the sequence of provincial conquests and how those con-
quests were tied to the reigns and activities of different emperors (as distilled in Figure 1). That
sequence of expansion is critical because it underlies our understanding of the chronology of Inca
imperial expansion, whether or not we choose to tie it directly to the dates given by Cabello Balboa.
In general, that chronological sequence of provincial conquests has appeared well grounded because
the same basic outline of conquests is repeated in what are deemed the most reliable accounts of Inca
history, such as those of Betanzos (1987) and Sarmiento de Gamboa (2007). Both of those authors
collected information directly from members of the Inca nobility, including many who were alive
before the arrival of the Spaniards. In theory, the general agreement in the sequence of Inca expan-
sion in those and other Spanish chronicles should reflect an accurate picture of the relative order of
provincial conquests as related by native sources. However, there may be a critical problem with
interpreting and using much of those data on the sequence of conquests. In essence, there appears to
have been a mismatch between a primary Inca form of record keeping and Spanish forms of histor-
ical narrative, which led to a standard history of Inca expansion that is actually more disordered (or
imprecise) than inaccurate. Specifically, the order of conquests of Topa Inca and his brothers as
recorded by Cabello Balboa and others is unlikely to reflect a strictly chronological sequence of Inca
expansion. Understanding the nature of this mismatch is crucial because those particular campaigns
of conquest accounted for the vast majority of Inca expansion, and our knowledge of the sequence
of those campaigns form the basis of how we relate Inca expansion in different regions to the dates
given by Cabello Balboa.
Because this problem is critical to the utility of historical data for understanding the chronology of
Inca expansion, it merits some discussion in detail. Central to the problem is how the accounts of
Inca history were produced from Inca sources, and Julien’s (2000) insights into that process are par-
ticularly relevant. As she demonstrated, the Incas preserved their historical information in multiple
genres, which were focused on specific subjects such as royal genealogy and life histories of rulers.
Colonial writers had to integrate information from multiple Inca genres to create linear narrative his-
tories according to Spanish canons. The main contents of specific Inca historical narratives were not
extensively detailed accounts, as they were restricted by the genres in which they were created; spe-
cific stories or events related in those accounts tended to be brief. For example, life histories of rul-
ers listed major events in their lives such as important conquests, but they did not list all of the
emperor’s military campaigns. To form their histories, it appears Spanish writers took as their basis
the chronological framework provided by the genealogical genre, and then interwove data from
other genres to flesh things out. This approach often resulted in problems of sequencing of events,
which are evident when details of different written histories are compared.
The key to understanding the recorded chronological sequence of Inca expansion is the existence of
an Inca historical form focused on recording military campaigns, another of the Inca historical
genres identified by Julien (2000:19). This historical genre essentially consisted of lists of the prov-
inces and peoples conquered by specific royal Incas, and it appears they were a major source for
Spanish chroniclers. We are fortunate to have a document that derives directly from that genre, the
Reconceiving the Chronology of Inca Imperial Expansion 233
“Probanza de los Incas nietos de conquistadores,” published by Rowe (1985). This document was
one of the most important Inca-related texts discovered in the 20th century (Julien 2008), yet it is not
widely known among English-speaking Andean archaeologists because it has only been available
through Rowe’s Spanish-language article. Nonetheless, it merits wider recognition because it pro-
vides vital insight into Inca historical practices, politics, and imperial expansion, as well as Inca-
Spanish relations during the Colonial era.
The “Incas nietos” document included the testimony given to Spanish colonial authorities in
AD 1569 by Inca nobility regarding the provinces conquered by the founders of their particular
royal kin group: the 3 brothers Topa Inca, Amaro Topa, and Topa Yupanqui. As Rowe (1985:194)
noted, this presented the first case where we have direct, voluntary declarations from Inca infor-
mants about their own history, remarkable because they were not subject to the literary elaboration
of Spanish authors. The features and content of the list of conquests suggest that it had been pre-
served on a quipu (the Inca method for recording information via knotted strings), which was then
read to the Spanish authorities (Rowe 1985:197). Notably, Rowe found a significant correspondence
between the list of conquests given by the “Incas nietos” and those listed in the chronicles of Cabello
Balboa, Pedro Sarmiento de Gamboa, and Fray Martín de Murúa. This indicated that Colonial writ-
ers were drawing on sources of information very similar to the “Incas nietos” record to create their
linear narratives of Inca imperial expansion.
Most vital for understanding Inca chronology is how the information in this list was organized.
Notably, there is no explicit or implicit indication that the document was meant to record a strictly
chronological order of conquests. Instead, the “Incas nietos” listed the conquests of their ancestors
in geographical order according to the 4 quarters of the empire, beginning with Chinchaysuyu, then
proceeding through Antisuyu, Collasuyu, and Condesuyu. This is the same canonical (Julien 2000:
132) or hierarchical (Pärssinen 1992:75) order the Incas used for listing the 4 imperial quarters in
many other contexts, e.g. it replicates the sequence of quarters in the list of ceque shrines in Cuzco
(Rowe 1985:197). This standardized format for ordering these data has significant ramifications.
Most importantly, because the conquests were grouped in this formalized geographical order and the
list lacked chronological references, we have no way of knowing if the conquests in one quarter
were conducted before or after those in another quarter.
The structuring of the list of conquests within each quarter is less clear. Pärssinen (1992) argues that
the internal order was primarily geographical in this and similar accounts that were presumably
based on the same type of quipu records. The sequence of provinces listed by the “Incas nietos” is
roughly in order as one travels from Cuzco, which could reflect a standardized geographical list of
provinces. However, such a sequence could also logically result from the systematic conquest of
contiguous territories as the Inca army moved away from Cuzco. Thus, the sequence within each
quarter listed by the “Incas nietos” could be roughly chronological. More significant is the likelihood
that the list for each quarter combined conquests resulting from multiple military campaigns. The list
was structured as if the conquests of each quarter comprised a single military campaign (Julien 2000:
132), and the “Incas nietos” even noted that the Incas returned to Cuzco at the end of the lists for 2
of the quarters. However, a historical sequence of just 4 major military campaigns by Topa Inca and
his brothers, one to each quarter, is highly unlikely. As described in various Colonial accounts, the
Incas of this generation conducted multiple campaigns in some of these provinces, which were led
by differing combinations of Inca nobles. For example, Sarmiento de Gamboa’s (2007) list of the
provinces conquered by Topa Inca in Chinchaysuyu parallels those listed by the “Incas nietos,” but
he described 2 distinctly separate military campaigns: one up to the land of the Cañaris (chapter 44),
and a second from Cañari territory north to Quito and then along the coast of Ecuador (chapter 46).
234 D E Ogburn
It is this compilation of multiple campaigns within each quarter that creates another significant chro-
nological problem: even if the military campaigns and resulting conquests were basically ordered in
time within each quarter, we have no way of comparing the sequence of campaigns between the dif-
ferent quarters. For example, there could have been a first campaign in Collasuyu, then one in Chin-
chaysuyu, followed by another in Collasuyu that continued into Antisuyu. The Inca quipu-based
records, as structured in this genre, would have listed the conquests in Chinchaysuyu first, followed
by Antisuyu, ending with the combined conquests of the 2 campaigns in Collasuyu. Drawing from
such a record under the assumption that it was chronological would clearly lead to the creation of a
historical sequence of conquests that was far from accurate.
As noted, the correspondence between the list of conquests given by the “Incas nietos” and the
chronicles of Sarmiento, Cabello Balboa, and Murúa (Rowe 1985) indicate that the Colonial writers
drew from sources organized in this fashion. A similar ordering of provinces conquered by Topa
Inca and his brothers is also provided in Betanzos (1987), and Cieza de León (1985) includes similar
information but lists the quarters in reverse order. These accounts of Inca conquests vary in their
details, suggesting they were not necessarily derived from the same quipu-based account of the
“Incas nietos,” and some may have been drawing from secondary versions of such accounts. None-
theless, the consistent grouping of conquests by imperial quarter indicates they were all drawing on
sources that were geographically structured, and integrating these data with information derived
from different genres of Inca history.
Because this standard form of listing conquests was primarily geographical rather than chronologi-
cal, Spanish chroniclers would have faced the challenge of fitting these campaigns within their own
chronological narrative form. They may have been unaware of the geographical structuring of the
data, and assumed a chronological order to the conquests. In that case, the sequence of conquests
simply needed to be pinned to certain events in Topa Inca’s life history that were derived from other
historical genres. The problem of this fusing of Inca historical genres is illustrated by the variability
in how Pachacuti’s death is placed in relation to campaigns of warfare carried out by Topa Inca and
his brothers. For example, Betanzos (1987) and Sarmiento de Gamboa (2007) describe Topa Inca
conquering and suppressing rebellions in Antisuyu and Collasuyu after the death of Pachacuti,
whereas Cabello Balboa (1945) places Pachacuti’s death during Topa Inca’s campaigns in those 2
provinces. Notably, the account of the “Incas nietos” made no reference to whether the conquests
listed occurred during or after the lifetime of Pachacuti, giving no guidance for meshing data from
that list with those from genealogical or other genres.
This challenge of meshing data from different Inca historical genres can be seen in similar issues of
sequencing elsewhere in the Colonial accounts (Julien 2000:231). In fact, the chronicler Martín de
Murúa (2001) directly acknowledged the existence of conflicts between different native sources. In
his history of Peru, he listed the Chinchaysuyu conquests of Topa Inca altogether in a short passage,
presented as a single campaign during the reign of Pachacuti. His list closely paralleled that of the
“Incas nietos” (Rowe 1985), with a few added details regarding some conquests, and as in the native
account, Murúa’s ends with noting the defeat of Hahua Chumpi and Nina Chumpi. In a later chapter,
he describes these places as being conquered on a separate trip by Topa Inca, but after the death of
his father and conquests in the other 3 quarters of the empire. Murúa directly attributed the conflict
in chronology to what was said by different sets of native informants and stated “entrambas opin-
iones se pueden tener, pues no va mucho en que haya sido en un tiempo o en otro” (“one could hold
either opinion, as it did not much matter what happened in one time or the other”) (Murúa 2001:83).
Reconceiving the Chronology of Inca Imperial Expansion 235
In sum, the major implication is that imperial incorporation of the great majority of provinces—
those conquered by Pachacuti’s sons—did not happen in the general order presented by Cabello Bal-
boa and other chroniclers. Instead, accounts from the Colonial period most likely combined a num-
ber of separate campaigns of expansion within each quarter to present them as occurring in a linear
fashion, when there had actually been multiple campaigns that alternated between different quarters.
The writers may have preserved the internal chronological order of places conquered in each cam-
paign, but not the relative order of campaigns between the different quarters of the empire. Thus, it
is quite conceivable that Topa Inca engaged in conquests in Argentina and Chile quite early in his
career, and that they were recorded by the Spanish as occurring much later because those campaigns
were memorialized by the Incas according to a geographical structure that put conquests in Colla-
suyu toward the end of the list.
This means that our chronology of Inca expansion can no longer be based on the historical accounts,
and we should not be judging 14C dates against the traditional chronology. Instead, we must base our
chronology on 14C dates, and evaluate the historical record in relation to them. One option would be
to eschew the historical record and create a chronology of Inca expansion based solely on 14C assays
of samples collected from a vast range of sites throughout the Andes. However, creating a satisfac-
tory chronology with this approach may be impractical given the time and money required, plus we
would not be able to deal with any expansion that occurred after ~AD 1470 because of the nature of
the calibration curves. The more practical approach would be to obtain as many 14C dates as feasi-
ble, and try to fill in the blanks in the map by critically analyzing the historical record to ascertain
which provinces were conquered together in specific campaigns, similar to the analysis of Pärssinen
(1992). It may be difficult to dissect the accounts of conquests of each quarter to reveal separate
campaigns, but they may in part be identified by the various combinations of leaders said to be
involved in differing series of conquests.
CONCLUSIONS
The dates given to us by Miguel Cabello Balboa, which served as the basis for the traditional chro-
nology of Inca expansion, have not stood the test of time. The traditional chronology promoted by
Rowe (1944, 1945) has been instrumental in emphasizing the rapid growth of the Inca state out from
the heartland of Cuzco while giving us a conception of the timing of stages of expansion in the
absence of applicable absolute dating methods. But refinements in 14C dating and calibration meth-
ods, a growing database of dates from diverse parts of the former Inca realm, and a reconsideration
of the sources of information used to create accounts of Inca expansion in the Spanish Colonial
period together point to a need to take a different approach to Inca imperial chronology.
14C dates will not provide the level of precision we might desire within the short span of the Inca
Empire. But they can still provide a level of accuracy that is not possible within the framework of
the historical accounts of Inca provincial expansion, which are based to a significant extent on geo-
graphically rather than chronologically structured data. Under the best circumstances, we may be
able obtain dates of sufficient precision and accuracy to date the Inca incorporation of different
provinces within a range of about 2 decades. While not ideal, this may be sufficient in some cases
to allow ordering of different campaigns of expansion that were spread out in time. Conversely, it
may prevent us from establishing the relative order of campaigns that happened much closer in time
than is suggested in the historical accounts.
The criteria presented here for providing the best 14C dates are suggestions that are certainly open to
discussion. Archaeologists working on Inca imperial expansion will need to decide what level of
precision and accuracy are acceptable or necessary for 14C dates and how specifically to calibrate
236 D E Ogburn
the dates and interpret them. Thus, rebuilding the chronology will not be easy. As one step in that
direction, the dates from Chamical are interpreted here as indicating Inca expansion into southern
Ecuador sometime in the range of AD 1440–1455. Also, we could add that they expanded into parts
of Argentina and Chile sometime prior to AD 1440. If the approach suggested here is followed and
more dates are produced and integrated into this incipient framework, it will be fascinating to see
how a new chronology of Inca imperial expansion unfolds over time.
ACKNOWLEDGMENTS
Test excavations at Chamical were supported by a Faculty Research Grant from the University of
North Carolina at Charlotte. 14C dates from that project were funded through the same grant as well
as the College of Liberal Arts and Sciences and the Department of Anthropology at UNC Charlotte.
My thoughts on Inca chronology also benefited from discussion with a number of colleagues, espe-
cially Gary Urton, and comments from Donna Lanclos have helped me to refine the manuscript. I
also thank the anonymous reviewers for their insightful comments and suggestions, which likewise
helped improve the final product.
REFERENCES
Adamska A, MichczyÒski A. 1996. Towards radiocarbon
chronology of the Inca State. Andes - Boletín de la
Misión Arqueológica Andina 1:35–60.
Bauer B. 1992. The Development of the Inca State. Aus-
tin: University of Texas Press. 185 p.
Betanzos J. 1987. Suma y narración de los Incas [1551].
Martin Rubio M, transcriber. Madrid: Ediciones Atlas.
317 p. In Spanish.
Bronk Ramsey C. 2009. Bayesian analysis of radiocar-
bon dates. Radiocarbon 51(1):337–60.
Cabello Balboa M. 1945. Obras, Volume 1 [1586].
Quito: Editorial Ecuatoriana. 451 p. In Spanish.
Cieza de León P. 1985. La crónica del Perú segunda
parte (El señorío de los Incas) [1553]. Lima: Pontifi-
cia Universidad Católica del Perú Fondo Editorial.
238 p. In Spanish.
Covey RA. 2006. Chronology, succession, and sover-
eignty: the politics of Inka historiography and its mod-
ern interpretation. Comparative Studies in Society and
History 48(1):169–99.
D’Altroy T, Williams V, Lorandi AM. 2007. The Inkas in
the southlands. In: Burger R, Morris C, Matos R, edi-
tors. Variations in the Expression of Inka Power.
Washington, DC: Dumbarton Oaks. p 85–133.
Finucane BC, Valdez JE, Pérez Calderon I, Vivanco
Pomacanchari C, Valdez LM, O’Connell T. 2007. The
end of empire: new radiocarbon dates from the Ayacu-
cho Valley, Peru, and their implications for the col-
lapse of the Wari State. Radiocarbon 49(2):579–92.
Garreaud R. 2009. The Andes climate and weather. Ad-
vances in Geosciences 7:1–9.
Julien C. 2000. Reading Inca History. Iowa City: Univer-
sity of Iowa Press. 350 p.
Julien C. 2008. Incas nietos de conquistadores (1569). In
Pillsbury J., editor. Guide to Documentary Sources for
Andean Studies 1530–1900. Volume II. Norman: Uni-
versity of Oklahoma Press. p 311–2.
Levillier R. 1940. Don Francisco de Toledo, supremo or-
ganizador del Perú; su vida, su obra (1515–1582).
Tomo II. Buenos Aires: Biblioteca del Congreso Ar-
gentino. 646 p. In Spanish.
Lippi R, Gudiño A. 2010. Inkas and Yumbos at Palmito-
pamba in northwestern Ecuador. In: Malpass M, Alco-
nini S, editors. Distant Provinces in the Inka Empire.
Iowa City: University of Iowa Press. p 260–78.
Mayle F, Burbridge R, Killeen T. 2000. Millennial-scale
dynamics of southern Amazonian rain forests. Science
290(5500):2291–4.
McCormac F, Reimer P, Hogg A, Higham T, Baillie M,
Palmer J, Stuiver M. 2002. Calibration of the radiocar-
bon time scale for the Southern Hemisphere: AD
1850–950. Radiocarbon 44(3):641–51.
McCormac F, Hogg A, Blackwell P, Buck C, Higham T,
Reimer P. 2004. SHCal04 Southern Hemisphere cali-
bration, 0–11.0 cal kyr BP. Radiocarbon 46(3):1087–
92.
McEwan G. 2008. The Incas: New Perspectives. Santa
Barbara: ABC-CLIO. 269 p.
Menzel D. 1959. The Inca occupation of the south coast
of Peru. Southwest Journal of Anthropology 15(2):
125–42.
Murúa M. 2001. Historia general del Perú [1590].
Madrid: Dastin. 571 p. In Spanish.
Ogburn D. 2004. Evidence for long-distance transport of
andesite building blocks in the Inca Empire from
Cuzco, Peru to Saraguro, Ecuador. Latin American
Antiquity 15(4):419–39.
Pärssinen M. 1992. Tawantinsuyu, the Inca State and Its
Political Organization. Helsinki: SHS. 462 p.
Reimer PJ, Baillie MGL, Bard E, Bayliss A, Beck JW,
Blackwell PG, Bronk Ramsey C, Buck CE, Burr GS,
Edwards RL, Friedrich M, Grootes PM, Guilderson
TP, Hajdas I, Heaton TJ, Hogg AG, Hughen KA, Kai-
ser KF, Kromer B, McCormac FG, Manning SW, Re-
Reconceiving the Chronology of Inca Imperial Expansion 237
imer RW, Richards DA, Southon JR, Talamo S, Tur-
ney CSM, van der Plicht J, Weyhenmeyer CE. 2009.
IntCal09 and Marine09 radiocarbon age calibration
curves, 0–50,000 years cal BP. Radiocarbon 51(4):
1111–50.
Rowe J. 1944. An Introduction to the Archaeology of
Cuzco. Cambridge: Papers of the Peabody Museum of
American Archaeology and Ethnology. 69 p.
Rowe J. 1945. Absolute chronology in the Andean area.
American Antiquity 10(3):265–84.
Rowe J. 1946. Inca culture at the time of the Spanish con-
quest. In: Steward J, editor. Handbook of South Amer-
ican Indians. Volume 2. Washington: Smithsonian In-
stitution. p 183–330.
Rowe J. 1962. Stages and periods in archaeological inter-
pretation. Southwest Journal of Anthropology 18(1):
40–54.
Rowe J. 1985. Probanza de los Incas nietos de conquis-
tadores. Histórica IX(2):193–245. In Spanish.
Rowe J. 2008. A question of time: Juan de Betanzos and
the narrative of the Incas. Julien C, translator. Ñawpa
Pacha: Journal of Andean Archaeology 29:155–62.
Sachs J, Sachse D, Smittenberg R, Zhang Z, Battisti D,
Golubic S. 2009. Southward movement of the Pacific
intertropical convergence zone AD 1400–1850. Na-
ture Geoscience 2:519–25.
Sarmiento de Gamboa P. 2007. The History of the Incas
[1572]. Bauer B, Smith V, translators. Austin: Univer-
sity of Texas Press. 266 p.
Schiappacasse V. 1999. Cronología del Estado Inca.
Estudios Atacameños 18:133–40. In Spanish.
Williams V, D’Altroy T. 1998. El sur del Tawantinsuyu:
un dominio selectivamente intensivo. Tawantinsuyu 5:
170–8. In Spanish.
Zuidema RT. 1964. The Ceque System of Cuzco: The So-
cial Organization of the Capital of the Inca. Leiden:
EJ Brill. 256 p.
Zuidema RT. 1982. Myth and history in ancient Peru. In:
Rossi I, editor. The Logic of Culture: Advances in
Structural Theory and Methods. South Hadley: JF
Bergin. p 150–75.
