Violence after imperial collapse: a study of cranial trauma among Late Intermediate Period

burials from the former Huari capital, Ayacucho, Peru

Author(s): Tiffiny A. Tung
Reviewed work(s):
Source: Ñawpa Pacha: Journal of Andean Archaeology, No. 29 (2008), pp. 101-117
Published by: Left Coast Press, Inc.
Stable URL: http://www.jstor.org/stable/27977841 .
Accessed: 04/11/2012 02:22

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

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

Left Coast Press, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Ñawpa Pacha:
Journal of Andean Archaeology.

http://www.jstor.org
 Violence after imperial collapse:
a study of cranial trauma among late intermediate period
BURIALS FROM THE FORMERHUARI CAPITAL,AyACUCHO, PERU

TiffinyA.Tung

This study documents thefrequency and patterning of cranial fractures to evaluate the role
of violence afterHuari
imperial collapse. These Late Intermediate Period buriah were interred at the sector at Huari, the
Monqachayoq
theHuari 31 adults exhibit healed cranial Perimortem
former capital of empire. Twenty-two of fractures (71%).
cranial fractures were observed on 42% of adults (n-31) and 30% of children (n-10). Men, women, and children
all sufferedfrom lethal attacks, suggesting that theymay have been victims of raids and killing sprees.Although the
skeletal sample is not wholly representative of all LIP communities in theformer imperial heartland, thedata indicate
that thepost-Huari period was a violent timefor numerous individuah, or sex.
regardless of their age

Este estudiopresenta h frecuencia y patr?n de trauma craneal para evaluar el

papel de violencia despu?s del coUpso del
imperio Huari. Los entierros corresponden al Intermedio Tard?o y fueron enterrados en el sectorde en
Monqachayoq
la capital antigua del imperioHuari. Veintid?s de los 31 adultos (71%) demuestran
Huari, fracturas craneales
saneadas (antemortem). Fracturas perimortem del cr?neofueron observadas en 42 de adultos (n-31) y 30% de los
j?venes (n=10). Ambos sexos tienen porcentajes similares.Hombres, mujeres, y ni?os sufrieron ataques mortales, y
sugiere que ellosfueron v?ctimas de asaltos y eventosde matanzas indiscriminadas. Aunque la muestra de esqueletos no
es
representativa de todas las comunidades del Intermedio Tard?o, losdatos osteol?gicos indican que elperiodo despu?s
del reinado de Huari fue un tiempo violento para muchas personas de varias categor?as de edad y sexo.

Periods followingimperialcollapsemay be char afterwards), or at the least, had to reorganize sub

acterized by social instability and increased ten stantially socio-political networks, trade relationships,
sions that are often manifested as higher levels of vio and agricultural and craftproduction systems (Alcock
lent conflict between newly autonomous groups. 2001; Parsons and Hastings 1988; Parsons et al.
These periods of upheaval may be caused by the 2000; Schreiber 1992).
power void left by the dissolution of an empire, as Researchers who have studied this and other "in
various polities jockey for control over trade routes, termediate periods" in theAndes often note the po
resources, labor, the "souls" of the populace, etc. The litical fragmentation that characterized these periods,
a case in and contrast that to "horizon" eras when states and
prehispanic Andes may provide point, where
after the collapse of two states?the Huari and the empires integrated large segments of land and people
Tiahuanaco?at the start of the 11th century AD, (Parsonsand Hastings 1988;Willey 1991). In par
local and regional polities experienced social and ticular, the Late Intermediate Period (LIP), ca. A.D.

political (either immediately or centuries 1000-1450, has been described as a time of immense
upheaval

TiffinyTung. DepartmentofAnthropology,
VanderbiltUniversity,
VU Station #356050,Nashville,TN 37235, t.tung@vanderbilt.edu.
 ?awpa Pacha 29

were achieved, aswell as the outcomes,

instability, especially in the highlands. In support of bywhich these
this, researchers note the disappearance of fine poly were affected a
by variety of factors, including local
chrome ceramics and elaborate textiles, the dissolu histories, unique ecology, individual leaders, and the
tion of widespread trade networks, a decrease or ab varied social and political footprints left by the pre
sence ofmonumental works an increase vious Middle Horizon states. One commonality that
projects, and
site locations and fortifications, likely a
in defensible seems to emerge in these
post-imperial settings, how
result of increased conflict or the threat thereof ever, is the prevalence, or threat, of violence in these
(Arkush2005; Conlee 2003; Parsons et al. 2000; apparently unstable times.
Schreiber1992; Stanish 1992;Wernke 2003). The goal of this study, therefore, is to look at
An excellent archaeological case study that docu levels of violence in the period following the col
ments the increased threat of violent conflict
during lapse of Huari in the former heartland of the Huari
theLIP isArkushs (2005) studyof theColla polity a
empire. This contributes much-needed perspective
in the highland Lake Titicaca Basin. Notably, how from the imperial core. Many studies have focused
ever, she does not directly attribute this instability to on or
regions far from the centers, where imperial
the disintegration of theTiahuanaco state, noting that state on local
collapse still had major repercussions
most fortifications were not constructed until two lifeways and social and political networks (Arkush
centuries after Tiahuanacos collapse. In the lower 2005; Conlee 2006; Torres-Rouff and Costa
This is remarkable because these
lying lands of Peru's south coast, Conlee's (2003, Junqueira 2006).
analysisof localNasca politiesafter
2006) insightful groups presumably could have relied on some form
the collapse of Huari reveals significant changes, such of local organization to to the
adjust major changes
as more with state
dispersed local elites and increased production associated collapse (Arkush, personal
and exchange. There also may have been increased 2007). This was done to some degree,
comunication
threats of violence as evidenced by stockpiles of sling but not markedly so as shown by archaeological evi
stones on a an LIP sector at the site of La dence demonstrating adverse conditions for the lo
ridge along
Tiza personal communication
(Conlee, 2007). cal populations. In the imperial center, the contrast
to these to
Contributing archaeological studies of from imperial post-imperial times may have been
ancient lifeways after imperial collapse in theAndes, even more extreme, as institutions, trade networks,

Torres-Rouff and Costa (2006) docu

Junquiera of provisioning the urban center, to
and methods
mented cranial trauma frequencies among San Pedro name but a few, declined and in some cases
collapsed.
de Atacama populations as a more direct way tomea The ramifications of these events would have been
sure instances of violence before, during, and after profound, perhaps manifest in frequent incidences
Tiahuanaco rule. Their thorough study found that of violence, both lethal and non-lethal.
the highest rate of head injuries occurred during the To address this issue, I document the frequency
LIP, when Tiahuanaco s power had waned and amajor and patterning of cranial trauma among an LIP
was underway. The skeletal evidence for in population that was interred in theMonqachayoq
drought
sector at the
creased tensions coincided with archaeological data capital site of Huari (Figure 1). Previous
that showed elevated population concentrations and studies have shown that documenting cranial frac
more fortified structures (Torres-Rouff and Costa tures is a reliable proxy for
estimating the pervasive
Junqueira2006: 66). ness of violence in past
populations (Lovell 1997;
These "responses"
to
periods of political frag
Martin and Frayer 1997;Walker 2001). Although
mentation states once unified
in areas where some head stem from accidental falls,
regions injuries may
share important attributes: these various polities had the spatial patterning of head wounds and the de
to reformulate and new identities, modes distribution of the skull trauma can be
renegotiate mographic
of governance, systems of production, and socio-po used to suggest whether or not injurieswere susrained
litical networks, while also rebuilding infrastructure in violent encounters.
and other aspects of material culture. The processes

102
 Tung: Violence after imper?alcolfopse

80eW75?W 70?W

80eW75?W 70eW

Figure 1.Map of Peru showing location ofHuari and otherHuari sites.

103
 ?awpa Pacha 29

The Site of Huari Monqachayoq

The capital of theHuari

state was located at the site
Monqachayoq is located in the architectural core of
ofHuari, in theDepartment ofAyacucho in the cen the site and includes a D-shaped structure and sev

tral, highland Andes of Peru. It is large site,mea

a eral underground, stone-lined galleries. The Quechua
some 250 hectares in the architectural core term used to denote the sector loosely trans
suring by locals
(Schreiber 2001), and approximately 15 square ki lates to "el lugar de las monjas' or "the place of the
lometers when habitation and trash disposal areas are nuns." Based on the sex
profile and
trauma analysis
included (Isbellet al. 1991) (Figure2). The site in of the human remains fromMonqachayoq, the name
cludes ceremonial, habitation, craft production, and couldn t be more of a misnomer, at least in terms of

mortuary areas, the last of which tended to be for how itwas utilized in the post-Huari era.
status individuals, such as those recovered from The Monqachayoq sectorwas excavated by Fran
high
the Cheqo Wasi sector (Benavides C. 1991). Large cisco Solano, and students from the Universidad
burial areas with Huari era lower elites or common Nacional de San Crist?bal deHuamanga (UNSCH)
ers have yet to be uncovered at Huari, so we know in Ayacucho in 1977-1978. Their excavations fo

nothing of the demographic patterns, health status, cused on the D-shaped structure and four of the
or 12 m
frequency of violence, body modification prac underground galleries. Each gallery measured
tices among the general population that lived at the in length and averaged 2 m in height, and theirwidth
times. In contrast, skeletal samples m on the m at the
capital inHuari averaged 1.65 ground and 1.25
from LIP contexts at the site of Huari have been re a
trapezoidal shape in profile
ceiling, giving them
covered, enabling inquiries into the role of violence (Solano andGuerrero 1981) (Figure3). The human
in a post-imperial context. remains derive primarily fromGallery 3, though some

Figure 2. Aerial photo ofHuari. Courtesy of Servicio Aerofotogr?ficoNacional de Peru.

104
 Tung: Violence after imperialcollapse

were recovered from

Gallery 4 (Solano and Guerrero populations used those spaces to shelter camelids
1981: 44, 186). In both galleries, the human remains (Solano and Guerrero 1981: 186).
were
"totally disturbed from destructive agents such Until the present study, itwas unclear to what
as looters, students from the
university, and harsh time period the human remains corresponded. Solano
weather" (Solano and Guerrero 1981: 186). The cra and Guerrero (1981: 215) had suggestedthat they
nia were stored together without reference to their dated to theMiddle Horizon based on Huari ceram
so it is unknown which ics found in the galleries, which they interpreted as
original gallery provenience,
came fromGallery 3 andwhich from
Gallery4.Only possible grave goods. They further suggested that the
a few
postcranial remains fromMonqachayoq have disturbed human remains from the gallery "repre
been located thus far, though hundreds were certainly sented collective burials that probably corresponded
excavated (Solano and Guerrero 1981: 195). For this to a certain social group, such as those connected to
reason, only cranial trauma is reported here. administrative tasks or artisanal activities" (1981:216).
to Solano and Guerreros ceramic To establish more securely the date of the buri
According
analysis(1981: 79-159, 230-238), thepotteryfrom als, four radiocarbon dates (using AMS)
were ob
theMonqachayoq sector includes both tained from the skeletal remains themselves, and they
Huarpa (pre
Huari) and Huari styles; among theHuari styles, they fall within the range cal A.D. 1020-1430, calibrated
classify theHuamanga ceramic type as "marking the at 2
sigma (Table 1). This places the human remains
finalphases of the
Wari empire" (1981: 231). They squarely within the LIP. The Middle Horizon ceram
also note that the camelid remains from Galleries 1 ics found in the galleries by Solano and Guerrero
and 2 likely date to post-Huari use when
subsequent may have been present prior to the disposal of the
bodies, coterminous with the burials, or mixed in
later as a result of looting. At this point, it is unclear
how the Huari ceramics and LIP human remains
came to be
together in Galleries 3 and 4. The sug
gestionbySolano andGuerrero (1981) thattheHuari
ceramics were grave goods is difficult to verify given
the disturbed nature of the deposits. Regardless, the
later dates obtained directly from the human bones
suggest that these were used to
impressive galleries
inter the dead after the collapse of theHuari empire.

Methods

were observed for both antemortem

Complete crania
and perimortem trauma; the frequencies for each are

reported separately to distinguish which injurieswere

non-lethal (antemortem) and which may have con
tributed to the cause of death (perimortem). All cra
nia were observed with a lOx handheld magnifying
lens. Bone fractures were identified as antemortem
as the
using evidence for bone healing primary crite
rion. They were identified based on areas
depression
on the surface of the skull or as uneven nasal bones,
sometimes with remnants of fracture lines, that indi
Figure 3. An underground gallery atMonqachayoq, Huari. cate a broken nose.
Although
ante- and postmortem

105
 ?awpa Pacha 29

Table 1. Radiocarbon (AMS) dates fromMonqachayoq skeletal remains.

Lab Code Bone Code Bone ?13C 14CAge 1 sigma range 2 sigma range

(Years b.p.) (a.D.) (a.D.)

Beta-229247 MC-Cran44 Scalp frag. -12.6 940+-40 1030-1160 1020-1200

Beta-229245 MC-Rad-AY Radius frag. -11.5 780+-40 1220-1270 1200-1280

Beta-229246 MC-Cran43 Cran frag. -13.3 600+-40 1300-1400 1290-1420

Beta-229244 MC-Rib-II-2-5 Rib frag. -9.7 560+-40 1320-1420 1300-1430

breaks can sometimes be difficult to distinguish, radi mortem head wounds are as on
reported occurring
at the one of five areas of the skull: anterior,
ating fracture lines, homogenous color margins posterior, right
of breaks, and the "freshness" of the fracture (i.e., and left sides, and superior.
adherent bone fragments or "hinging" at the fracture
were used to trauma
margins) identify peri-mortem
Results
(Berryman and Jones Haun 1996; Ortner 2003).
The locations of these various wounds can be
Age-at-Death and Sex Profiles
as these
quite informative, provide insight into the
kinds of violent actions thatmay have led to the in on counts of non-over
Based complete crania and
a on the front of the head were at least 47 in
jury. For example, wound lapping cranial fragments, there
may be the result of face-to-face conflict, while a frac dividuals in the sample. Two were infants and 10were
ture to the back of the head may indicate the person children under 14 years of age. The other 35 indivi
was
injured while fleeing from
an assailant (Walker duals consisted of an older adolescent, 11 young adults,
1997;Webb 1995), or ducking froman oncoming eight middle-aged adults, 13 old adults, and two
blow. To reconstruct the behavioral aspect of the vio adults thatcould not be specifically
aged (Figure4).
lent interaction, the locations of ante- and peri Of the 29 adults whose sex could be estimated,

30%

25%

20%

15%\

10%

5%]

Infant Child Late Teen Young Adult Mid-Adult Old Adult Adult
3-14 yrs 15-19yrs 20-34 yrs 35?49 yrs 50+yrs 20+ yrs

Figure 4. Age-at-death profile of theMonqachayoq crania.

106
 Tung: Violenceafterimperialcollapse

18 were male (62%) and 11 were female (38%). Al tern cranial trauma, and rate is
although the female
therewere more males than females, the dif the difference is not statistically
though higher, significant
ference was not statistically significant relative to an (Fishers exact test, p=0.330; n=29). Two adult cra

equal distribution (Fisher's

exact test,
p=0.251). nia were of unknown sex, and one exhibited a healed
head wound. No child crania displayed healed head
Antemortem Cranial Trauma wounds (n=10).
Seven of the 22 affected adults exhibited more
Among the 31 adults with complete crania, 22 ex
than one healed head wound, indicating that they
hibitedat leastone healedheadwound (71%) (Table
either received several blows to the skull in one vio
2 and Figures5 and 6). Nine out of 11 females(82%)
lent incident, or were in separate violent interactions.
and 12 out of 18 males (67%) exhibited antemor
One of these adults?an old adult female?showed
as many as five antemortem fractures. Four of the
Table 2. List of injured adults and the counts of ante
and cranial fractures individual.
nine affectedfemales(44%) and threeof the 12 af
perimortem per
fected males (25%) exhibited more than one healed
Cranium Sex No. No. head wound, but this sex-based difference is not sta
Code Wounds Perimortem
sexes were
Wounds tistically significant, suggesting that both
to receive more than one head
equally likely injury
72 F 1 0 (Fisher's exact test, p=0.319; n=21).
60 F 5 0 All together, therewas a total of 41 healed cra
42 F? 1 2 nial wounds among the 22 injured persons (Table
43 F? 3 0 3). Roughly half of the healed wounds were on the
44 F? 1 1 facial and frontal bones?21 of the 41 wounds were
located there?suggesting that face-to-face violent
63 F? 4 0
interactions were common. The second most fre
76 F? 4 0
area was the posterior, where 10
F? 1 2 quently injured
wounds (24.5%) were 10
observed. The other
C F? 1 0
wounds were distributed among the sides and supe
F? 0 2
rior portion of the head (Figure 7).
15 M 1 1
Examining sex-based differences in location of
22 M 3 1 healed head wounds shows that females have dispro
23 M 3 0 more fractures on the anterior and a less
portionately
26 M 1 1 random distribution of head wounds overall; 62%
30 M 1 0 of female wounds are on the facial and frontal, and
are present on
39 M 1 0 the other 38% only three other cra
40 M 0 2 nial areas (posterior and right and left sides). In con
trast, only 42% of male wounds are on the anterior,
41 M 4 1
and the other 58% of the wounds are distributed
69 M 1 0
over all other cranial locations
75 M 1 0 (posterior, right and
left sides, and superior). The more restricted distri
34 M? 1 3
bution of female wounds may suggest that women
37 M? 0 1
were
injured in similar social contexts, many ofwhich
46 M? 1 0
involved hits to the facial area while thewomen were
64 M? 1 1
facing their attackers.
Q M? 0 3
s 1 0
Total 41 21

107
 ?awpa Pacha 29

5. Healed trauma on adult crania.

Figure Monqachayoq
C.46: Healed blunt force trauma on C.76:
right parietal.
Healed blunt force trauma on left parietal boss. C.60:
Healed nasal fracture and an tern o rtem tooth loss re
likely
lated to facial trauma, not dental disease C.26:
Partially
healed blunt force trauma on rightparietal. C.15: Partially
healed blunt force trauma on rightparietal boss.

108
 Tung: Violenceafter imperialcollapse

Perimortem Cranial Trauma

In addition to sublethal head wounds, several indi

viduals suffered perimortem cranial fractures, many
of which may have been the mechanism of death

(Figure8).Thirteenof the31 adults (42%) displayed

at least one perimortem head fracture. Specifically,
nine out of 18males (50%) and fourout of 11 fe
males (36%) were affected, but these sex-based dif
ferences were not statistically significant (Fisher s ex
act test, p=0.372; n=29). Neither of the two unsexed
adults showed perimortem cranial fractures.
Six of the 13 affectedadults receivedmultiple
nine males
potentially fatal head wounds. Three of the
(33%) and threeof the four females (75%) with
perimortem cranial fracturesexhibited multiple

perimortem wounds, but that sex-based difference

was not statistically s exact test, p=
significant (Fisher
0.118; n=13). In one case, an adult male received
twohitson therightsideofhis skullthatappearedas
separate, round depression fractures, likely resulting
from a sling stone or an attacker wielding a mace or
a rock. In another an adult female was struck
example,
at least twice: once on the frontal bone and another
timeon theoccipitalbone (Figure9). If theseblows
occurred one after the other, thewounds on the front
and back suggestthatshehad timeto changebodily
one aspect of her skull while
position, protecting
was in control
exposing the other (assuming that she
of her own It is to
bodily movement). impossible
determine in this case which of the two injuries was
sustained first, as there are no radiating fracture lines
that intersect to determine which precedes the other.
There was a total of 21 perimortem skull frac
tures among those six affected adults (Table 4). Ex
most
on
Figure 6. Healed trauma Monqachayoq adult crania. cluding sex-based differences for the moment,
C.64: Healed trauma on nasal and leftorbit. C.43 and fatal wounds were located on the posterior of the
C.63: Healed nasal fracture. skull (7/21=33%), followedby theanteriorand right

Table 3. Number of antemortem head wounds per affectedadult.

1 2 3 4 5 Total
Head Wound Head Wounds Head Wounds Head Wounds Head Wounds

No. of adults 15 331 22 0

No. ofwounds 15 0 9 4112

5
in sample

109
 ?awpa Pacha 29

14

Anterior (n=21) Posterior (n=10) Right Lateral (n=3) LeftLateral (n=5) Superior (n=2)
Location offHealed Head Wounds (N=41)
I Female ?Male!

on each portion of the cranium among males and females.

Figure 7. Counts of healed head wounds

Figure 8. C.42: Large perimortem fracture that dislodged portions of the right frontal,
sphenoid, and temporal bones. Those displaced cranial fragmentswere not recovered.The
radiating fracture line, commonly visible in cases of perimortem blunt force trauma, runs
along the posterior of the parietals.

110
 Tung: Violence after imper?alcollapse

on leftfrontalbone.
Figure 9. Perimortem fractureson Cranium B. Top: rectangularperimortem depression fracture
Bottom left:ectocranial view of perimortem depression fracture in center of occipital bone. Bottom right:endocranial
view showing that the blow to the occipital was forcefulenough to dislodge part of the internal surfaceof the cranium.

Table 4. Number of perimortem head wounds per affectedadult.

3 12 Total
Head Wound Head Wounds Head Wounds

No. of adults 7 213 4

No. ofwounds 7 8 21 6
in sample

111
 ?awpa Pacha 29

side where both areas had six perimortem wounds assailant or when
ducking their head
to avoid an
each. There were two perimortem fractures on the
oncoming blow. It is also possible that these victims
leftside (Figure 10). were forced to bow their heads before receiving the
More detailed analysis of the spatial distribu fatal strike.
tion of perimortem head wounds shows that there Among men, the locations of lethal head inju
are sex-based differences. the seven peri ries differ from non-lethal ones. Lethal blows tend
Among
mortem fractures on female skulls, 43% are on the to be on the
posterior, while non-lethal ones are on
anterior, and the others are evenly distributed on the the anterior. This likely reflects the social circum

posteriorand rightside of thehead (Figure 10), a stances inwhich men found themselves inmortally
pattern generally mirroring that of healed head situations versus less serious violent en
dangerous
wounds on women. Whether the blow was lethal or counters. As far as head
injuries
are concerned, men
not, women were more
commonly hit while facing
seem to have fared better (i.e., recover from their
their assailant. were
injury) if they facing their attacker.
Men, in contrast, exhibit relatively more peri Nine of the 13 adults with frac
perimortem
mortem wounds on the tures also exhibit healed head wounds,
posterior: five of the 14 indicating that
(36%) perimortem skull fractures were located there, had at least two separate violent
they experienced
followedby fouron therightside (29%). The others incidents in their lifetime: one nonfatal violent en
were on the anterior and left side of the skull counter earlier in life, and one around the time of
(Figure
10). That lethal head wounds were more common death. Of those nine adults with both injury types,
on the threewere females and sixwere males. However, be
posterior than the anterior suggests that these
fatal injuries were sustained when fleeing from an cause there aremore males than females in the
sample,

I3
O
O

Anterior (n=6) Posterior (n=7) Right Lateral (n=6) Left Lateral (n-2) Superior
Location of Perimortem Head Wounds (N=21)

Figure 10. Counts of perimortem head wounds on each portion of the cranium among males and females.

112
 Tung: Violence after imperialco?apse

this does not suggest thatmen were more likely than skulls, as evidenced by radiating fracture lines and
women to be involved in multiple violent events. portions of the occipital.
missing
Rather, their frequencies are similar: 3/11 fe
quite
males (27%) and 6/18 males (33%) exhibitboth
antemortem and perimortem cranial fractures. Discussion and Conclusion
Children also suffered perimortem cranial frac
tures; three out of 10 children were affected, and all Re-use ofHuari Spaces in the
were between the ages of 10 to 13 years old. All three Late Intermediate Period
had receivedblows to the leftside of theirskull. In
two of those cases, it portions of the left The radiocarbon dates obtained from the human
dislodged
two
temporal and sphenoid bones (Figure 11). Those bones indicate that post-Huari populations interred
children also received blows to the posterior of their the dead in elaborate subterranean galleries thatwere

""
\Iff*'J& S?/1Eff5

on children.Top photos: Blunt force trauma to the leftside and posterior aspect
Figure 11. Perimortem cranial fractures
of cranium (C.51). Photo detail shows radiating fracture lines and adherent bone fragments,evidence of perimortem
fracture.Bottom photos: Blunt force trauma to the leftside of child s cranium (C.50). Photo detail shows "hinged"
bone, evidence of perimortem fracture.

113
 ?awpa Pacha 29

constructed during the time ofHuari rule. This may
suggest that LIP populations continued to view the
imperial capital, and these underground spaces in
as an important locale ideal as a final rest
particular,
ing place for specific individuals. Furthermore, sub
radiocarbon dates are an accurate reflection of long
term use of the
gallery, then it is possible that the
bodies represent victims from several different kill
ing sprees.Whether these skeletal remains represent
heroic warriors, injured victims, or both remains

sequent use of these elaborate spaces thatwere once unclear; the disturbed and commingled bones pre
associated with a powerful polity may have been a clude determining if the deceased were formally bur

savvy means by which newly autonomous groups ied or simply tossed into the gallery.
could their quest for authority. Indeed, Despite the ambiguity of themanner inwhich
legitimize
of empires'(Alcock2001) show
studiesof the 'afterlife they
were
finally deposited, the skeletal
trauma data
that descendent groups often borrow or build upon tell a story of vicious attacks during killing sprees,
the social and symbolic capital (Bourdieu 1977, where blows to the head were hard-hitting and re

1986) of the preceding authority. And while destruc petitive?and likely intended to kill. In many in
tion of imperial symbols or structures may also be stances, the blunt force trauma to the head was so
used to convey the rise of a new authority, that does forceful, entire sections of the cranium were cracked
not appear to be the case in the re-use of theMon and pieces of bone were dislodged. Moreover, about
as there is no evidence of its inten half of those with perimortem trauma exhibited more
qachayoq sector,
tional destruction, only evidence for its ongoing use. than one re
impact scar, indicating that they had
stones thrown at
peatedly been hit by high velocity
Warrior orDispose of theVictim?
Bury the using
a honda
(sling).

The demographic profile shows that infants and chil Indiscriminate Violence in the Post-Huari Era
dren constitute only 25% of the burial sample?a
low frequency for prehistoric groups (Hoppa and In addition to the numerous fatal head on
injuries
Vaupel 2002; Paine and Boldsen
2002)?so it is men, women, and children, an extraordinary num
is representa ber of adults?71 %?had also received a severe blow
highly unlikely that this burial sample
tive of a once-living community. Rather, this skeletal to the head earlier in life, as discussed above. This is
a "vil a statistically
series represents a mortuary population?not significantly higher frequency than that
lage" population?where
mourners may have interred among adults from theHuari era site of Conchopata,
select persons from one or several different settle located 10 km south of Huari, where only 23% of
ments. Given the high rate of healed cranial trauma, adults (n=44) exhibited healed cranial trauma (Tung

perhaps warriors or defenders of nearby habitations n.d., 2007) (Fishers exact, p<0.001; n=75). Focus
were in this space. on percentages alone, the data suggest an increase
preferentially deposited ing
It is also possible given the high rates of lethal in non-lethal violent to post
conflict from Huari
cranial trauma among men, women, and children Huari times in the imperial heartland, but when the
that this burial sample represents massacre victims. are considered, the non-normal
population profiles
In studiesof otherpopulationswith high levelsof population distribution among the Monqachayoq

perimortem trauma,
massacres are often
suggested group renders this conclusion tentative. Until a skel
as when several etal sample that is apparently more representative of
possible explanations, particularly
traumatized bodies are found buried together (Ferllini LIP populations from the heartland is excavated and

1999; Hinton 2004; Semelin 2003; Ta ala, et al. analyzed, extrapolations based
on data in this study

2006; Willey and Emerson 1993). Thus, the corpses remain to be confirmed.

may have been deposited in the underground galler Although I do not conclude that this high
ies by those who committed the acts of violence, or trauma
frequency is representative of all post-Huari
a in the former imperial heartland, I do
by the survivors, perhaps in the aftermath of single populations
era was
massacre or several different lethal raids. If the four suggest that violence in this post-imperial

114
 Tung: Violence afterimperialcolUpse

somewhat indiscriminate.Children, men, and References Cited

women were in violent encounters. During
targeted
Susan E.
this prolonged absence of a unifying power, violence Alcock,
2001 The afterlifeof empires. In Empires, edited by
rippled through various segments of society, affect Susan E. Alcock, Terence N. Kathleen
D'Altroy,
ing the young, themiddle-aged, and the old, regard
D. Morrison, and Caria M. Sinopoli, pp. 369
this case study demonstrates that
less of sex. Thus, 373. Cambridge University Press, Cambridge.
post-imperial times were seriously detrimental to
Arkush, Elizabeth .
many people who were interred at the former Huari
2005 Cotta fortified sites:warfare and regionalpower in
capital. theUte prehispanic Titicaca Basin, Peru. Unpub
numerous
As previous studies have demon lisheddoctoral dissertation,Department ofAnthro
strated, periods after imperial collapse can be char at Los Angeles.
pology,University of California
acterized by instability,which likely stems from the
Benavides C, Mario
demise of social and political structures that long
state and
1991 Cheqo Wasi, Huari. InHuari administrativestruc
organized and shaped internal and external ture: monumental architecture and state
prehistoric
net
community relations and social and political government,edited byWilliam H. Isbell and Gor
works. If one accepts that outbreaks of vio
repeated don F. McEwan, pp. 55-70. Dumbarton Oaks,
lence are evidence for unstable, even chaotic times, Washington.
then it is appears that this characterizes portions of
Berryman,Hugh E., and Susan JonesHaun
Late Intermediate Period society in the Ayacucho
1996 Applying forensic techniques to interpretcranial
Basin afterHuari imperial collapse. Thus, the results fracture patterns in an
archaeological specimen.
of this research coincide with other studies on post InternationalJournal ofOsteoarchaeology6: 2-9.
imperial collapse in theAndes that have demonstrated
Bourdieu, Pierre
an increased threat of violence and/or its actual mani
1977 Outline of a theoryofpractice. Translated by R.
festation. The cranial trauma data from those interred
Nice. Cambridge University Press, Cambridge.
at the sector atHuari
Monqachayoq strongly suggest 1986 The forms of capital. InHandbook of theoryand
that the collapse of the Huari empire led to volatile
for thesociologyofeducation,edited by John
research
times, at the least, for a subset of post-Huari popula Greenwood Press, New
Richardson, pp. 241-258.
tions living in the former imperial heartland. York.

Conlee, Christina A.
Acknowledgments 2003 Local elites and the reformationof Late Interme
diate Period sociopolitical and economic organi
I thank Liz Arkush and Christina Conlee for insight zation in Nasca, Peru. Latin American
Antiquity
ful feedback on an earlier draft of this manuscript, 14(1): 47_66.
and I thank the anonymous reviewers for their help 2006 as transformation: so
Regeneration post-collapse
ful comments. I am also grateful to Katharina
ciety in Nasca, Peru. In the regen
After collapse:
Schreiber, the Nawpa Pacha editor, for providing eration of complex societies,edited by Glenn M.
and detailed comments on my article. I Schwartz and John J.Nichols, pp. 99-113. Uni
thoughtful
of Arizona Press, Tucson.
thank SteveWernke for editing my Spanish abstract. versity

to Jos?Ochatoma
Finally, I express immense gratitude
for granting me permission to study this important
skeletal collection. This research was supported by a
Vanderbilt Discovery Grant and aWork Group Grant
from theVanderbilt Center for theAmericas.
Ferllini, Rox?na
1999 The role of forensicanthropology inhuman rights
issues. InForensic osteologicalanalysis:a book ofcase
studies,edited by Scott I. Fairgrieve,pp. 287-302.
Charles C Thomas, Springfield.

Hinton, Alex
2004 The poetics of genocidal practice: violence under
theKhmer Rouge. In Violence, edited byNeil L.

115
 ?awpa Pacha 29

Whitehead, pp. 157-184. School of American pires, edited by Susan E. Alcock, Terence N.
Research Press, Santa Fe. Kathleen D. Morrison, and Carla M.
D'Altroy,
/
Sinopoli, pp. 70-92. Cambridge University Press,
Hopp?, Robert D., and JamesW. Vaupel
Cambridge.
2002 Paleodemography: age distributionfrom skeletal
samples.Carnbridge University Press,Cambridge. Semelin, Jacques
2003 Analysis of a mass crime: ethnic cleansing in the
Isbell,William H., Christine Brewster-Wray,and Lynda E. formerYugoslavia, 1991-1999. In The spectorof
Spickard mass murder in historical ed
genocide: perspective,
1991 Architecture and spatial organization at Huari.
ited by Robert Gellately and Ben Kiernan, pp.
InHuari administrative structure: monu
prehistoric 353-370. Cambridge University Press, Cam
mental architecture and state edited
government,
bridge.
byWilliam H. Isbell and Gordon E McEwan,
pp. 19-54. Dumbarton Oaks, Solano Ramos, Francisco, and Venturo Guerrero
Washington. Anaya
1981 Estudio arqueol?gico en el sector de Monqa
L?vell, Nancy C.
chayoq-Wari. Unpublished bachelor's thesis,De
1997 Trauma analysis in paleopathology. Yearbook of
partment of Archaeology, Universidad Nacional
Physical AnthropologyAO: 139-170. de San Crist?bal de Huamanga, Ayacucho.

Martin, Debra L., and David W. Frayer Stanish, Charles

1997 Troubled times: violence and warfare in thepast.
1992 Ancient Andean political economy.University of
Gordon and Breach, Australia.
Texas Press, Austin.

Ortner, Donald J. Ta ala, Sabrina ,Gregory E. Berg, and Kathryn Haden

2003 Identification ofpathological conditions in human 2006 Blunt force cranial trauma in theCambodian kill
skeletal remains. Academic Press, Amsterdam.
ing fields.Journal ofForensic Science 51(5): 996
1001.
Paine, Richard R., and JesperL. Boldsen
2002 Linking age-at-death distributions and ancient Torres-Rouff, Christina, and Maria Antonietta Costa
a case In Paleode
population dynamics: study.
Junqueira
mography: age distributionsfrom skeletal samples, 2006 Interpersonal violence in prehistoric San Pedro
edited by Robert D. Hoppa and JamesW. Vaupel,
de Atacama, Chile: behavioral implications of en
pp. 169-180. Cambridge University Press,Cam vironmental stress.American Journal ofPhysical
bridge. Anthropology 130: 60-70.
R., and Charles M. Hastings
Parsons, Jeffrey
Tung, Tiff?nyA.
1988 The Late Intermediate Period. In Peruvian pre n.d. The bioarchaeology ofWart imperialism. Book
an overview and Inca
history: of pre-Inca society, Press of Florida, Tallahassee.
manuscript. University
edited by Richard W. Keatinge, pp. 190-229.
2007 Trauma and violence in theWari empire of the
Cambridge University Press, Cambridge.
Peruvian Andes: warfare, raids, and ritual
fights.
Parsons, Jeffrey R., Charles
M. Hastings, and Ramiro Matos American Journal ofPhysical Anthropology 133:
Mendieta 941-956.
2000 Prehispanic settlement Mantaro
patterns in theupper
and Tarma drainages,Junin,Peru. Memoirs of the Walker, Philip L.
1997 Wife beating, boxing, and broken noses: skeletal
Museum of Anthropology 34, University of
evidence for the cultural patterning of violence.
Michigan, Ann Arbor.
In Troubled times:violenceand warfare in thepast.
Schreiber,Katharina J. Vol. 3, edited by Debra L. Martin and David W.
1992 Middle Horizon Peru. Anthro
Wari imperialismin Frayer, pp. 145-179. Gordon and Breach Pub
pological Papers 87, Museum of Anthropology, lishers, Australia.

University ofMichigan, Ann Arbor. 2001 A bioarchaeological perspective on the history of

2001 The Wari empire ofMiddle Horizon Peru: the violence. Annual Review ofAnthropology30: 573
an
epistemological challenge of documenting 596.
without evidence. In Em
empire documentary

116
 Tung: Violence afterimperialcolUpse

Webb, Stephen Willey, Gordon R.

1995 Palaeopathology of aboriginal Australians: health 1991 Horizontal integrationand regional diversity: an
and disease across a continent. Cam in the rise of civilizations.
hunter-gatherer alternating process

bridge University Press,New York. American Antiquity 56(2) : 197-215.

Wernke, Steven A.
Willey, P., and Thomas E. Emerson
2003 An archaeo-historyofAndean communityand knd 1993 The osteology and archaeology of theCrow Creek
scape: theUte prehispanic and early colonial Coica massacre. Plains
Anthropologist38(145): 227-269.
valley,Peru. Unpublished doctoral dissertation,
Department ofAnthropology, University ofWis
consin at Madison.

117