Female Kiva Societies in the Taos Valley:
Mortuary and Bioarchaeological Evidence
Presented in the Symposium
Bodies of Evidence: Integrating Mortuary and Osteological Analyses
Organizers: Ryan Harrod and Kyle Waller
Discussant: Gordon Rakita
78th Annual Society for American Archaeology Meeting, Honolulu, Hawaii.
April 2013
Catrina Banks Whitley
Research Associate
Office of Archaeological Studies
Museum of New Mexico
1
Abstract:
The incorporation of a wider range of bioarchaeological data with mortuary practices can
provide a clearer method for archaeologists to perceive social structure. In isolation, each can
provide an interpretation of an individual’s position within their community, but results from
each study may be contradictory. For example, evidence of egalitarianism in mortuary ritual may
not be a complete reflection of the complexity inherent within a group. Likewise, poor health
may not always indicate lower social position or lack of prestige. Only by merging the two lines
of evidence can a more complete understanding of social structure and how an individual’s social
position can affect their health and life expectancy be revealed. This paper presents a concrete
example of the need to combine mortuary and bioarchaeological data to garner a more complete
interpretation of health, risk, and social structure by using data from Pot Creek Pueblo (AD
1260-1320) in the Taos Valley. Inclusion of osteological datasets, beyond simple sex and age
divisions, with the mortuary datasets and architecture revealed a structured division of labor
among the females, resulting in one labor group potentially having ritual prestige. However, the
prestige and status of these females did not buffer them from poorer health.
Introduction
“Context is everything. . . .context is key in interpretation”
Goldstein (2006:377)
In many bioarchaeological studies, the individual has been taken out of the context of the
archaeological site. Studies focus on topics such as health, disease, indicators of diet and
entheseal changes that elucidate important information about the lives of those in the past, but a
more robust interpretation of how the community was structured and social position can be
ascertained when the bioarchaeological interpretations are incorporated into site and mortuary
contexts. Results from these studies in isolation can be contradictory. Analysis of mortuary
practices without bioarchaeological data may not reflect the complexity inherent within a group;
particularly if the mortuary data indicate egalitarianism. Poor health may not always indicate
lower social position or lack of prestige, particularly in a society not based on power through
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
2
economic wealth. Prestige can come from knowledge, spiritual power or ritual duties, or special
access to tasks that may not economically benefit the individual. Thus, economic egalitarianism
does not always indicate social egalitarianism. Only by merging the two lines of evidence,
complex bioarchaeological analyses and complex understanding of site data and/or mortuary
behavior, can a more complete understanding of social structure and how an individual’s social
position can affect their health and life expectancy be revealed. To simplify, we need to place the
individual into the space and environment in which they lived - including interpretations of ritual
practices. This paper presents a concrete example of the need to combine mortuary and
bioarchaeological data to garner a more complete interpretation of health, risk, and social
structure by using data from Pot Creek Pueblo (AD 1260-1320) in the Taos Valley (Whitley
2009).
Materials
Pot Creek Pueblo was occupied approximately AD 1260-1320 (Figure 1). Construction of the
aggregated pueblo rooms began in the late 1260s with roomblocks 2, 3, and 6, and these
roomblocks were expanded from 1280 to 1300. A building surge in the 1310s created plaza
spaces surrounding an unfinished Great Kiva, dating to AD 1318, and the addition of roomblocks
1 and 4 (Crown 1991). Two mealing rooms are associated with roomblocks 2 and 3, and
Ramadas are only associated with roomblock 4. The northern portion of the site has not been
dated and only one room was excavated. It is blocked in grey in this last slide because we have
no construction dates. The site was excavated over a fifty-year period by Southern Methodist
University and burials were encountered between 1957 and 1984.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
3
Skeletal remains available for study included 47 individuals- 18 males, 21 females, and 8
children 0-5 years of age. All adult skeletal remains available for analysis were able to be sexed,
although not all remains were intact for hypoplasia or entheseal changes. Life expectancy was
24.47 years at birth (Whitley 2009). Average stature for males was 156.80 +/- 7.72 cm and
females was 150.92 +/- 5.19 cm (Whitley 2009).
All remains, excluding two buried on the second floor of roomblock two and a child excavated
from a D-shaped kiva in the north roomblock, were found with one or more stones, ash, and trash
covering the body and were in an extramural context (Whitley 2009). It is possible the
exclusionary burials are associated with the abandonment of the site. The child in the D-shaped
kiva is interpreted by Fowles (2004) to be part of a ritual deposit.
Methods
The bioarchaeological analysis discussed in this paper include entheseal changes,
musculoskeletal stress markers- which can indicate the muscles regularly subjected to stress,
presence and location of osteoarthritis, age-at-death, and dental hypoplasia. Since the average
age-at-death for the population is low, with life expectance at 24 year and few individuals
surviving past 50 - 77% of adults dying before an average age of 40- entheseal changes and
osteoarthritic changes are more likely to correlate with occupation than aging stressors.
Entheseal changes, or musculoskeletal stress markers (MSM), can indicate habitual use of a
muscle that is regularly subjected to minor stress. Patterns in muscle use, or the lack thereof, can
identify groups of individuals that habitually move their bodies in similar manners, particularly
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
4
in groups that require heavy labor for subsistence (Capasso et al. 1999; Hawkey and Merbs
1995). i Initial changes were calculated looking for occupational differences between males and
females. No statistically significant differences were present between the sexes, suggesting
similar levels of occupational stress. However, I chose to also look within each sex to identify
whether hierarchical or occupational differences existed.
Additionally, elevated levels of osteoarthritis may indicate increased mechanical demand in areas
of the body subjected to higher mechanical loading(Larsen 1997), particularly related to habitual
corn grinding, carry heavy loads, heavy lifting, or habitual bending or stooping. Osteoarthritis
was analyzed to investigate whether incidence was similar in males and females as well as within
each sex.ii
The presence of dental hypoplasia is linked with dietary deficiencies, fevers, infectious disease,
or other childhood diseases (Hillson 2008). Dental hypoplasia were scored and the ages at which
they occurred were assessed (Reid and Dean 2006; Hillson 2008). All observable teeth were
scored.
Burials were mapped to identify potential group clusters. Maps were created based on age, sex,
and by female groups. Clustering of graves may indicate bounded disposal areas for certain
roomblocks with several inividuals buried before roomblock construction expanded over their
graves.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
5
Results
Entheseal Changes
Analysis indicated a correlation between sex and MSM changes (r=0.75, p=0.02). No
correlations existed between humeral size, MSM changes, and age in the males, but were
statistically significant in females. Simple comparisons of male and female MSMs obscured a
significant difference within the females and how they were using their muscles (Figure 2).
These females were placed into two Groups, one interpreted as corn grinders (Group 1) and the
other as non-corn grinders (Group 2).
•
Statistically significant differences were present at the deltoid, pectoralis
major, and teres major sites between the Group 1 and Group 2 females.
•
Females in group 2 exhibited no hypertrophy at these MSM site markers,
indicating they did not habitually use these muscles during biomechanical
stress .
Osteoarthritis
A similar pattern was found in the osteoarthritis analysis. Osteoarthritic changes indicates no
correlation between age and presence of OA in females (Weiss and Jurmain 2007). Arthritic
changes were analyzed to ascertain if certain osteoarthritic changes correlated with Groups 1 and
2.
•
Group 1 females had higher rates of costovertebral osteoarthritis (p=0.041) than
Group 2 females (Figure 3).
•
Costovertebral osteoarthritis, or arthritis at the junction of the rib and
vertebrae, is caused by elevation of the rib cage.1
•
Group 2 females had greater incidences of thoracic and lumbar osteoarthritis.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
6
Dental Hypoplasia
Dental hypoplasia were calculated for all males and females. More females than males had
dental hypoplasia and only two Group 1 females had dental hypoplasia (Table 1). However, the
presence of dental hypoplasia correlated with a younger age-at-death; 9 of the 11 dying before
age 30 - five of those were under 20. Additionally, individuals with dental hypoplasia had
multiple hypoplasia, some exhibiting one to two per tooth. Age assessments indicate the
hypoplasia correlate with weaning ages, 1-3 years of age.
Age-at-Death
Females from Group 1 died an average of 5 years younger than those in Group 2 (Figure 4).
Though the sample size is small, there is a distinct difference in the age-at-death patterns for the
two groups. Discrepancies in health and longevity may simply be the result of their expending
greater amounts of energy performing more strenuous tasks than the group two females. Habitual
strenuous activities can have significant affects on health and longevity in individuals because
increased labor results in greater physiological stress on the body, resulting in higher morbidity
and mortality of those under greater physiological stress. Corn grinding for that many hours
would burn more calories than gathering or sitting to perform tasks. If each group of females
had access to similar foods and amounts of food, Group 1 corn grinders would need more
calories to stay healthy. Therefore, the females from Group 1 suffered the consequences of an
earlier death. In other words, they could have been an elite group of females in terms of power,
but they did not secure the benefit typically associated with privileged groups, who usually have
access to better or greater amounts food.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
7
Mortuary Practices
With this information, each individual was placed within the site to look for patterns in burial
location (Figure 5). Clusters in close proximity to each roomblock indicate a strong link between
habitation sites and burial placement and suggests bounded disposal areas. This assessment is
based upon the random pattern of individuals by age and sex buried near each roomblock.
Spatial patterning of Group 1 and 2 females indicates females from each group are also randomly
distributed at each roomblock. Mortuary context was consistent throughout the site as all
individual were buried in the exact same manner - buried on sterile ground, covered with ash and
"trash", pottery fragments etc., covered with cobbles or large stones, and no associated funerary
objects - suggesting egalitarianism among the population. If this was the only line of evidence
used, it would mask social distinctions within the population. Even the addition of age-at-death
or sex with spatial patterning, it appears that there are group areas for burials for each
roomblock, but still social egalitarianism. This data does not elucidate Fowles (2004) argument
that Pot Creek was a theocrocacy - that prestige was held through knowledge of ritual and
spiritual duties.
Discussion/Conclusions
Ethnographic evaluation of habitual activities by pueblo females generally includes grinding
corn for consumption and ritual use, food preparation, cooking, later stages of pottery
production, gathering wild foods, and making baskets, while men were responsible for farming,
occasional hunting, and religious ceremonies (Crown 2000; Hegmon et al. 2000; Harrington
1989; Martin 2000). Differences in muscle use between the female task groups indicates two
distinct occupational groups, one responsible for grinding corn, while the other performed less
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
8
strenuous upper body tasks but require bending or sitting, such as plant gathering, pottery
production, or cooking (Whitley 2009). Corn grinding is of most interest because it is a
strenuous task that, as Martin (2000) indicates, required that females spend up to eight hours per
day grinding corn to make enough meal for consumption. In order to grind corn, one must
utilize the deltoid, pectoralis major, and teres major muscles - the arm muscles- and these
changes were not present in Group 2 females. This slide demonstrates that an individual
grinding corn at a metate and mealing bin will kneel in front of a bin, feet braced, and raised up
on the knees. During grinding, mechanical stress is placed on the upper limbs as the grinder
flexes and extends the arms below and above head to grind the corn. During this process, the
ribs are elevated in order to keep the arms in proper position, particularly at the point of the
greatest extension of the arms. The mechanical loading on the thoracic vertebrae and
costovertebral joint and muscles in the arms necessary to grind corn for such extended periods of
time indicate Group 2 females did not participate in corn grinding activities.
Yet, simply recognizing the existence of the two groups does not provide information regarding
the role these females had within society or the social organization of the site; only that they had
different tasks. This could be interpreted as a division of labor for efficiency in feeding a larger
group of people. However, "social integrative facilities of a more monumental nature tend to be
constructed and used almost exclusively for ritual activities" (Adler and Wilshusen 1990).
Hegmon, et. al. (2000) argues the presence of mealing rooms indicates the present of task
groups, or “grinding work groups” and that these task groups, if recognized with architecturally
defined gendered work spaces, may be an important source of power. Specialized mealing
rooms with multiple grinding bins built into the architecture were located in Roomblocks 2 and
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
9
3. The presence of these specialized mealing rooms, which is not open to public view, can be
interpreted as female kivas, since difference in status and prestige can be created by excluding
certain individuals from spaces, resources, or information (Fowles 2004; Hegmon et al. (2000).
Differences in ritual participation in the Southwest are key to differentiation of social groups,
and task groups associated with restricted architectural spaces can be an important source of
power (Hegmon et al. 2000).
Random spatial patterning of Group 1 and 2 female burials may hold a key to the nature of
power and prestige held by the females. Coordination and cooperation with work groups are
important aspects of food processing in many societies as well and may be associated with
improved status of women. Hegmon (2000) argues that the presence of task groups and the
construction of facilities for women’s work, such as formal mealing rooms, suggest the
importance of corn grinding and the power of women over this separate realm. The spatial
patterning of females from Groups 1 and 2 throughout the site suggests that no single roomblock
was responsible for grinding corn for the entire village (Figure 6). Instead, females from each
roomblock were responsible for this task. There are no indications that control over certain
productive resources was concentrated in some households over others because of the dispersal
of females in the corn grinding society. Such distinctions in female labor and the presence of
mealing rooms also suggests that the knowledge associated with corn grinding, and access to
preparing corn for meals and ceremonies, was limited to specific females within the society. In
such circumstances, the division of labor may signal individuals with “high social honor”.
The presence of specialized mealing rooms may also link the corn grinding females with
specialized ritual knowledge forming a “corn grinding society”, indicating a more prominent role
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
10
for female ritualism and gender specific roles in ritual. As the community aggregated to the
pueblo lifestyle, the move from family to community control may have been an attempt to
organize the community at a larger scale (Whitley 2009). Since women’s roles often center on
the preparation of food for use and consumption in rituals, and as the community increasingly
relied upon corn as a staple food source, the ritualization of corn grinding may have served as a
community based integrative activity. Thus, certain females from each kin group or roomblock
were given access to the specialized knowledge of grinding corn, had access to these specialized
spaces, and would have been responsible for grinding corn not only for mundane, everyday use
but also for rituals. As the kin groups aggregated into a single location, they may have needed a
mechanism for the increasing scalar stress of living within a larger population by dividing female
labor (Whitley 2009). Formal mealing rooms may represent a change from generalized, familial
ritual control to ritual control by external groups that cross-cut familial line, forming ritually
bounded spaces for females or “female kivas”. As a result, gender became an element defining
participation, but did not result in an increased in economic standing or greater access to foods.
Instead, the "corn grinding society" actually had a lower life expectancy even though they held
specialized ritual power and social prestige. Through altering the scale of analysis, analyzing
bioarchaeological differences within each sex, and integrating the data with mortuary practices,
obscured patterns of female labor and female access to power and prestige was identified;
creating a more robust interpretation on the female role within their society.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
11
References
Adler M.A., and R.H. Wilshusen
1990 Large-Scale Integrative Facilities in Tribal Societies: Cross-Cultural and Southwestern
US World Archaeology, Vol. 22, No. 2, Monuments and the Monumental (Oct., 1990),
pp. 133-146.
Capasso, L. et al.
1999 Atlas of Occupational Markers on Human Remains, First Edition. Journal of
Paleontology - Monographic Publication Edigrafital S.p.A., Teramo, Italy.
Crown, P.
1991
Evaluating the Construction Sequence and Population of Pot Creek Pueblo, Northern
New Mexico. Am. Antiquity 56(2):291-314.
2000
Women and Men in the Prehispanic Southwest: Labor, Power and Prestige. SAR
Advanced Seminar Series. SAR Press, Santa Fe, NM.
Fowles, S.
2004
The Making of Made People: The Prehistoric Evolution of Hierocracy Among the
Northern Tiwa of New Mexico. University of Michigan, Unpublished Ph.D. dissertation.
Goldstein, Lynne
2006 Mortuary Analysis and Bioarchaeology. In Bioarchaeology: The Contextual Analysis of
Human Remains, edited by Buikstra, J. E. and Beck, L. A., pp. 375-388. Academic
Press, Burlington, MA.
Harrington, J.
1989 Indian Tales from Picuris Pueblo, edited by Marta Weigle. Ancient City Press, Santa Fe,
NM.
Hawkey, D. and Merbs, C.
1995
Activity-induced Musculoskeletal Stress Markers (MSM) and Subsistence Strategy
Changes Among Ancient Hudson Bay Eskimos. IJO 5:324-338.
Hegmon, M. et al.
2000
Women, Men, and the Organization of Space. In Women and Men in the Prehispanic
Southwest: Labor, Power and Prestige, Edited by Patricia L. Crown, 43-90. SAR
Advanced Seminar Series. SAR Press, Santa Fe, NM.
Hillson, S.
2008 Dental Pathology. In Biological Anthropology of the Human Skeleton 2nd ed., M.A.
Katzenberg and S.R Saunders, pp. 301-340. Wiley-Liss, Hoboken, NJ.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
12
Larsen, C.
1997 Bioarchaeology: Interpreting Behavior from the Human Skeleton. Cambridge
University Press, Cambridge.
Martin, D.
2000 Bodies and Lives: Biological Indicators of Health Differentials and Division of Labor by
Sex.. In Women and Men in the Prehispanic Southwest: Labor, Power and Prestige,
edited by Patricia L. Crown, 267-300. SAR Advanced Seminar Series. SAR Press,
Santa Fe, NM.
Molnar, P.
2006 Tracing Prehistoric Activities: Musculoskeletal Stress Marker Analysis of a Stone-Age
Population on the Island of Gotland in the Baltic Sea. AJPA 129:12-23.
Reid, D.J. and M.C. Dean
2006 Variation in Modern Human Enamel Formation Times. Journal of Human Evolution
50:329-346.
Weiss, E.
2003 Understanding Muscle Markers: Aggregation and Construct Validity. AJPA 121:230–
240.
2007
Muscle Markers Revisited: Activity Pattern Reconstruction With Controls in a Central
California Amerind Population. AJPA 133:931–940.
Weiss E. and Jurmain, R.
2007
Osteoarthritis Revisited: A Contemporary Review of Aetiology. IJO 17: 437–450.
Whitley, CB.
2009
Body Language: An Integrative Approach to the Bioarchaeology and Mortuary
Practices of the Taos Valley. Unpublished Ph.D. Dissertation, Southern Methodist
University.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
13
Figure 1. Map of Pot Creek Pueblo
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
14
Figure 2. Entheseal Changes
Common Extensors (o)
Group 2
Group 1
Muscle or ligament attachment
Pronator Quadratus (o)
Supinator (o)
Extensor Carpi Radialis Longus (o)
Costoclavicular Ligament
Linea Aspera
Gleuteus Maximus (i)
Biceps Brachii (i)
Brachialis (i)
Teres Major (i)
Flexor Policis Longus (o)
Deltoid (i)
Pectoralis Major (i)
0
0.5
MSM mean score
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
1
1.5
2
2.5
15
Figure 3. Osteoarthritis
80.0%
70.0%
Group 1
Group 2
Percent of Individuals Affected
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Cervical
Thoracic
Lumbar
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
Ribs
Hip/Knee Shoulder/Elbow
16
Figure 4. Female Age-at-Death Frequencies
7
6
Number of Individuals
5
4
G1
3
G2
2
1
0
15-20
20-29
30-39
Age-at-Death Groups
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
40+
17
Figure 5. Pot Creek Burials by sex and cluster.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
18
Figure 6. Burial Locations of Group 1 and 2 Females
M=Mealing Room, 1=Corn grinding females, 2=Non-corn grinding females
.
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
19
Table 1. Hypoplasia
NA
No Hypoplasia
Hypoplasia
Male
7
10
3
Female
8
8
8
Total
15
18
11
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR
20
End Notes
i
ii
In order to ascertain the presence of task groups:
•
Analysis of MSM sites were recorded in accordance with the scales devised by
Hawkey and Merbs (1995)7.
•
Two-sample Wilcoxon rank sum (Mann-Whitney) tests and Wilcoxon matchedpairs signed-rank tests were utilized to analyze the MSM data; following Molnar
(2006).
•
Two sample t-test were utilized to check for variances in the observed frequencies
of specific MSMs between and within males and females, including lower and
upper limb scores.
•
Spearmans rank correlations were calculated to control for age (defined into three
groups 15-24, 25-34, and 35+), humeral size and sex; following Weiss(2003;
2007).
•
Some MSMs were excluded from analysis due to small sample size.
•
Right and left MSM scores revealed no bilateral asymmetry.
In order to ascertain osteoarthritic changes:
•
Analysis was based in the presence or absence
•
Chi-square tests were used to control for age and test for differences within and
between males and females.
•
Vertebral changes refer to the changes present on the anterior vertebral bodies
•
Rib changes refer to costovertebral osteoarthritis.1
© Catrina Banks Whitley, 2013
DO NOT CITE WITHOUT WRITTEN PERMISSION OF AUTHOR