ArticlePDF Available

Political Agency of Royal Women



Comparative analysis of women rulers and main wives of kings in eight premodern states around the globe reveals similar patterns of political agency, or the opportunity and ability to take political action. Queen rulers, regents, and main wives substituting for their husbands in their absence made policy, but they had somewhat less political agency than male rulers. Main wives’ political agency took the forms of influencing policy and people’s behavior (sometimes through their role as patron to others), interceding between their kin and their husbands, advocating for one party or the other, spying, and conspiring. Therefore, women’s political agency ought to be part of any political study. This study builds on the anthropological/archaeological study of agency by drawing attention to royal women’s political agency and showing how the analysis of structural rules and the roles of kings, queen rulers, and main wives illuminates the societal structure in which agency is embedded. By analyzing premodern societies this way, we learn that there is remarkable similarity of agency behaviors among royal women in the eight sample societies, even though the societies emerged independently of one another.
1 23
Journal of Archaeological Research
ISSN 1059-0161
J Archaeol Res
DOI 10.1007/s10814-019-09131-y
The Political Agency of Royal Women: A
Comparative Analysis of Eight Premodern
States According to Societal Rules and Roles
Paula L.W.Sabloff
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
Science+Business Media, LLC, part of
Springer Nature. This e-offprint is for personal
use only and shall not be self-archived in
electronic repositories. If you wish to self-
archive your article, please use the accepted
manuscript version for posting on your own
website. You may further deposit the accepted
manuscript version in any repository,
provided it is only made publicly available 12
months after official publication or later and
provided acknowledgement is given to the
original source of publication and a link is
inserted to the published article on Springer's
website. The link must be accompanied by
the following text: "The final publication is
available at”.
Journal of Archaeological Research
1 3
The Political Agency ofRoyal Women: AComparative
Analysis ofEight Premodern States According toSocietal
Rules andRoles
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Comparative analysis of women rulers and main wives of kings in eight premodern
states around the globe reveals similar patterns of political agency, or the opportu-
nity and ability to take political action. Queen rulers, regents, and main wives sub-
stituting for their husbands in their absence made policy, but they had somewhat less
political agency than male rulers. Main wives’ political agency took the forms of
influencing policy and people’s behavior (sometimes through their role as patron to
others), interceding between their kin and their husbands, advocating for one party
or the other, spying, and conspiring. Therefore, women’s political agency ought to
be part of any political study. This study builds on the anthropological/archaeo-
logical study of agency by drawing attention to royal women’s political agency and
showing how the analysis of structural rules and the roles of kings, queen rulers, and
main wives illuminates the societal structure in which agency is embedded. By ana-
lyzing premodern societies this way, we learn that there is remarkable similarity of
agency behaviors among royal women in the eight sample societies, even though the
societies emerged independently of one another.
Keywords Agency· Political agency· Premodern state· Royal women· Women’s
agency· Patron–client relations
Based on the study of eight premodern states spread over the globe, I propose a new
perspective on their queen rulers and main wives (the small coterie of elite women
selected to beget legitimate heirs). The behaviors of both sets of women reveal
remarkable political agency. Women’s political agency is infrequently mentioned
* Paula L. W. Sabloff
1 Santa Fe Institute, 1399 Hyde Park Road, SantaFe87501, NM, USA
Author's personal copy
Journal of Archaeological Research
1 3
and often ignored in scholars’ analysis of premodern state politics. But royal women
were active players in determining succession, governing the polity, building
inter- and intrapolity alliances, and expanding or defending territory. By studying
royal women’s behavior through the lens of agency theory, I found that queen rul-
ers had almost the same political agency as kings. Main wives exhibited consid-
erable agency as well, for they influenced policy (e.g., decisions of war or when
to plant crops), influenced behavior often by controlling relationships of obligation
(patron–client relations), interceded between or advocated for parties (especially
royal kin and affines), and even spied on or conspired against their spouses.
The eight premodern states represent three different kinds of regions (Fig. 1).
The first is a region with independent states or city-states, including the Mari King-
dom during Old Babylonian times (2000–1600 BC) and Protohistoric Hawai’i (AD
1570–1590 to 1778). The second regional type is an empire, represented by Old
Kingdom Egypt (2686–2181 BC), Late Shang China (1250–1046 BC), the Aztec
Empire (AD 1440–1520), and the Inca Empire (AD 1460–1532). The premodern
states in regions that contain both independent states and empires are the Late Clas-
sic Maya (AD 600–800) and the Postclassic Zapotec (AD 1050–1500).
This comparative analysis of royal women’s political agency contributes to
agency theory as used in the archaeological literature. Agency theory refers to peo-
ple’s actual behaviors when they have the capability and opportunity to act and
emphasizes the embeddedness of agents’ behavior in their social structure. I define
social structure as the rules that govern behavior, such as rules of succession and
descent. Agency theory also depicts action as behavior in relation to others. This
study investigates royal women’s actions in relation to the main political actors of
the time, namely, kings and emperors. Agency theory concerns the issue of intent,
or motivation: are people motivated by self-interest alone? Royal women used their
agency to aid their kin and children and sometimes their husbands or themselves.
Fig. 1 Map of premodern states
Author's personal copy
1 3
Journal of Archaeological Research
Agency theory embeds people’s actions in their sociocultural context. Although
each premodern state had unique features, through my analysis I found amazing
similarity in royal women’s roles, behaviors, and agency. Thus agency should not be
treated as idiosyncratic.
Archaeologists largely derive their definition of agency from Bourdieu (1977,
1985, 1990), Giddens (1979, 1984), and Ortner (1984) (see Ahearn 2001, pp.
117–119; Dobres and Robb 2000, p. 5; Dornan 2002; Gardner 2007; Joyce and
Lopiparo 2005, pp. 366–367; Robb 2010). They analyze agency as the behaviors
that are both constrained and enabled by the social structure in which people live.
Such behaviors are performed either by historical individuals or categories of peo-
ple—statuses such as king, queen, or viceroy. Prior to 2000, adapters of agency the-
ory focused on the behavior of individuals (Moore 2000, p. 260). The more recent
trend has been to focus on the interplay of behavior and structure, including people’s
action in relation to others. Robb (2010, pp. 501–502) calls this “relational agency.”
Scholars also have broadened the definition of agent to include social groups act-
ing as a unit and such non-human agents as mountains or abstract gods (animistic
agency) (see also Harrison-Buck and Hendon 2018; Sillar 2007, pp. 154–155).
Within this general framework, archaeologists’ interest varies. Their first concern
is to describe the constraints on agents. These frequently include power constraints,
as described by Foucault. What are the structural conditions (such as political organ-
ization or overpopulation and drought) and more immediate stimuli (such as warfare
or factionalism) that influence agents (Ahearn 2001, pp. 116, 120; Banning 2010;
Blanton etal. 1996)? Some archaeologists isolate the intentions or motivations of
agents (Hubert 2016, p. 2; Joyce and Lopiparo 2005, p. 368; Joyce and Winter 1996,
pp. 33–34). A subset of this pursuit is to learn for whom people take action—them-
selves or some collective group. This is often phrased as acting “on behalf of” or
“agency for.” Archaeologists also are concerned with the consequences of agency,
especially structural change (Blanton etal. 1996, p. 2; Brumfiel 2000; Moore 2000,
p. 260; Pauketat 2001). Flannery (1999) devises ten strategies for structural transfor-
mation from chiefdom to premodern state based on the life histories of three men.
Feminist archaeologists have focused on women’s economic or religious agency,
that is, gendered production (weaving, pottery), reproduction, and ritual behavior
(Ardren 2008; Brumfiel 1991; Claassen 1997, pp. 83–85; Conkey and Gero 1997,
pp. 415, 419; Joyce 1993; Marcus 1998), although some have studied political
agency as well (Brumfiel 1992; Gillespie and Joyce 1997). My goal is to extend
their work by formalizing and instantiating agency theory in royal women’s political
agency. To that end, I organize the data in four sections, or groupings of questions—
action, intent, structure, and constancy—that are inspired by Dornan’s (2002) analy-
sis of agency.
Action. What did royal women actually do? What do these behaviors tell us about
their political agency?
Intent. For whom did the royal women take action? Were they motivated by self-
interest or the needs of their kin group or their affines? Joyce and Winter (1996, p.
46) see the agency of Monte Albán elites as the goal-seeking, conscious strategies
of individuals. Bell (1992) and Banning (2010) take the same approach. But Ahearn
(2001, p. 112) posits that agency also may be “supraindividual—the property,
Author's personal copy
Journal of Archaeological Research
1 3
perhaps, of families, faculties, or labor unions.” Robb (2010, p. 502), summarizing
Gillespie (2001) and Fowler (2004), writes that agency is sometimes reserved for
a lineage or some other group rather than for individuals. When looking at royal
women, we see that they mostly acted on behalf of their kin, not their husbands or
Structure. How did society (here, royal women’s ruler-kin and -affines) struc-
ture—constrain and enable—women’s political agency? For example, did their right
to control property, especially arable land, affect their ability to act? How can we
isolate the structure that affects behavior? I suggest that studying rules and roles—
societal rules and the roles expected of rulers (male and female) and royal women—
can help focus our analysis of agency.
Constancy. Why is the political agency of royal women so similar across the
eight premodern states? Agency theorists such as Barrett (2000) and Johnson (2000)
argue that agency is too rooted in idiosyncratic societal context to allow for cross-
societal comparison. Yet I found remarkable similarities among the royal women of
this study, even though they could not possibly have learned such behavior from one
another because of the spatial and temporal distances between the premodern states
in the sample. Perhaps we should turn the question around and see the behaviors of
royal women as providing independent proof that humans developed similar politi-
cal structures independently all over the world (e.g., Johnson and Earle 2000; Trig-
ger 2003; Turchin etal. 2015, 2018).
I pose these questions to make two points. First, we must include women in our
analysis of premodern states’ politics, including warfare and alliance building, if
we are to understand the process by which decisions are made and actions taken.
Gero (2000) raises this issue when she argues against “flattening” human actions or
reducing actions (including discourse) to those performed by men, usually powerful
men. If one component of agency is what people do, then influencing, mediating,
and interfering (spying) are just as valid agency activities as warring, subjugating,
and controlling. Although Collier showed that Zinacantan (Mexico) women were
politically active back in 1974 in her chapter in the ground-breaking book, Woman,
Culture and Society (Rosaldo and Lamphere 1974), few have taken up her call to
arms, so to speak. In a delightful chapter on women’s political roles in a School for
Advanced Research edited volume, Bowser and Patton (2010) describe Conambo
(Ecuador) women building consensus and alliances, resolving conflicts between fac-
tions, directing others in some activity, and speaking persuasively. Conambo women
are able to do so because they are the major food producers and have the ability to
tie together disparate kin groups. But such analyses are few and far between (e.g.,
Ardren 2008).
My second point is that it is difficult if not impossible to separate self-interest
from altruistic motivation in this and perhaps any database. Main wives usually
acted on behalf of others—their powerful kin or affines or their children. Were the
women’s actions undertaken in free will or coercion? We may never know. However,
their motivation do not diminish the fact that the women exhibited political agency.
Influencing policy, patronizing clients, advocating for others or interceding on their
behalf, and spying are as legitimate kinds of political agency as are making policy,
warring, alliance building, and controlling access to resources.
Author's personal copy
1 3
Journal of Archaeological Research
I used a comparative methodology to find patterns of royal women’s behavior
that cut across premodern state societies. The behavior patterns emerged from an
intensive five-year project to isolate all possible statuses and the roles associated
with them (Linton 1936). With the help of research assistants and citizen scien-
tists, I created databases of 14 premodern states. After having them checked by
an expert in each society (see Sabloff 2018), I was able to double the data entries
as I discovered more sources and gained a better idea of what to look for. From
these databases, I found sufficient information on royal women in the eight socie-
ties listed in the Introduction. This inductive approach revealed trends in royal
women’s agency that extended beyond the particular societies in which they lived.
The societies selected for this project are the same ones that I included in a
previous article on the strategies that kings and emperors used to reduce the risk
of losing at war (Sabloff 2018; see also Sabloff and Cragg 2015). In both cases,
I used data from the time when premodern states had some form of writing that
could complement the archaeological record. Without the writings, stories of
individuals could not be collected and therefore royal women’s agency could not
be extracted. Use of this time period did not allow us to note structural change,
that is, the period of state formation or collapse. Therefore, I cannot address
one of the major issues of agency in archaeology, namely, the possible effect of
agency on societal change. It should be noted that the data for all the New World
cases (including Protohistoric Hawai’i) come not only from the archaeological
material and accounts of conquered people but also from the European invaders.
The latter, needless to say, brought their own bias to their interpretation of the
societies they conquered. For example, Sahagún (1961a, b) documented Aztec
life extensively but never mentioned women when he wrote about how Aztec rul-
ers took pleasure. Therefore scholars have judiciously interpreted these accounts,
and I have followed suit.
I compared the premodern states according to royal women’s roles and politi-
cal agency, social structural rules for societal elites, and political roles expected
of royal women’s counterparts, namely, kings and emperors (summarized in
Tables1, 2, 3). To conduct the comparison, I grouped the societies according to
the political complexity of the regions in which they existed: regions of independ-
ent states, empires, and regions that held both independent states and empires.
While I hypothesized that types of political complexity would determine or at
least correlate with differences in royal women’s political agency, this was not
the case. Instead, patterns of agency crisscrossed the regions, the chronology, and
relative complexity of society.
To learn whether or not there were causal relationships or co-occurrence
between royal women’s political agency and their societies’ structures or role
expectations, I spent over a week doing simple regression analyses and then
Excel’s CHITEST (chi-square) function. Both exercises yielded little evidence
of either. I also generated Johnson’s Hierarchical Clustering and non-metric
MDS (multidimensional scaling) in UCINet (Borgatti etal. 2002). I did not learn
Author's personal copy
Journal of Archaeological Research
1 3
anything new from these exercises and so eliminated them from the analysis, too.
From all of these trials (and tribulations!), I have to concur with my mathematical
and physicist colleagues at the Santa Fe Institute that eight cases are insufficient
for statistical analysis.
Statistics may not work in this study, but the epigraphic and archaeological data
strongly correlate women’s behavior with the societies in which they live. And logic as
well as life experience confirm that behavior is influenced by multiple societal factors.
The following descriptions illustrate not only the prevalence of royal women’s political
agency but also the connection between, or co-occurrence of, societal traits and such
Table 1 Royal women’s political agency by state
Presence of a trait (marked as 1) is based on literature and logic, or speculation. Absence of a trait
(marked as 0) means that literature or logic implies that the trait did not exist or no information on the
trait could be found.
Makes policy as ruler/co-ruler: Woman makes policy as ruler or co-ruler in her own right.
Makes policy as regent: Woman makes policy as regent for her underage son.
Influences policy: Woman provides information and advice regarding war, planting crops, etc.
Influences behavior: Woman influences the behavior of her spouse as well as others (beneath her in rank).
Intercedes/advocates: Woman intercedes for others—often her kin—with the ruler. She may advocate for
others to the ruler or advocate the ruler’s position to others, e.g., her husband or kin.
Patronizes: Woman has unequal exchange relationship with underlings, who are her clients.
Spies: Woman passes information to her husband or her kin.
Conspires: Woman conspires against her husband in order to have him replaced on the throne.
Empires Combination No. of states
Mari Hawai’i Egypt China Inca Aztec Maya Zapotec
Makes policy as ruler/
0 1 0 0 1 0 1 0 3
Makes policy as regent 0 1 1 0 0 1 1 0 4
Influences policy 1 1 1 1 1 1 1 1 8
Influences behavior 1 1 1 1 1 1 1 1 8
Intercedes/advocates 1 1 1 1 1 1 1 1 8
Patronizes 1 1 1 1 1 1 1 1 8
Spies 1 0 0 1 0 1 1 1 5
Conspires 0 0 1 0 0 0 0 0 1
No. of traits 5 6 6 5 5 6 7 5
Author's personal copy
1 3
Journal of Archaeological Research
Table 2 Structural rules and practices by premodern state
Presence of a trait (marked as 1) is based on literature and logic, or speculation. Absence of a trait
(marked as 0) means that literature or logic implies that the trait did not exist or no information on the
trait could be found
Polygyny: Rulers marry more than one wife
Only men can rule: 1 = only men can rule unless women are acting as regents. The presence of women
rulers and co-rulers is marked 0
Primogeniture: Ideal succession is oldest son of the ruler
Bilateral descent: Legitimacy to rule may come from the father or mother or both. In cases marked 0, the
mother’s lineage usually has some import but not as strong as bilateral descent
Close-kin marriage: Preferred marriage is brother–sister, brother–half-sister, or uncle–niece
Gender separation: Men and women are segregated across the society. Does not refer to royal women
being housed in their own quarters
Ruler divine in life: Rulers were considered divine when ruling rather than post-mortem. 1 means rulers
are the sons of a god; does not include rulers who impersonate/speak as if gods during rituals but not
Ruler divine after death: Ruler may or not may not be considered divine in life, but definitely deified as
a sacred ancestor after death. If this and the row above are 0, then the ruler is never considered divine
(except, maybe during sacred ceremonies)
Ruler monopolizes access to main god: Ruler monopolized access to the prime deity; by acting as the
sole mediator between the primary god and the people, ruler controls messages from the gods
Ruler protects the people: Ruler is responsible for the well-being of the people
Ruler protects territory: Ruler is responsible for expanding or defending the polity’s territory
Advisory council: An advisory or governing council helps the ruler make policy and may select the ruler
from among several candidates
Empires Combination No. of states
Mari Hawai’i Egypt China Inca Aztec Maya Zapotec
Structural rules for elites
Polygyny 1 1 1 1 1 1 1 1 8
Only men can rule 1 0 1 1 0 1 0 1 5
Primogeniture 0 1 1 0 0 0 0 1 3
Bilateral descent 0 1 0 0 0 1 1 1 4
Close-kin marriage 0 1 1 0 1 1 0 0 4
Gender separation 0 1 0 0 1 0 0 0 2
Roles expected of rulers
Ruler divine—in life 0 1 1 0 1 0 0 0 3
Ruler divine—after death 0 1 1 1 1 0 1 0 5
Ruler monopolizes access
to main god
1 1 1 1 1 1 1 0 7
Ruler protects the people 1 1 1 1 1 1 1 1 8
Ruler protects territory 1 1 1 1 1 1 1 1 8
Advisory council 1 1 1 0 1 1 1 1 7
No. of traits 6 11 10 6 9 8 7 6
Author's personal copy
Journal of Archaeological Research
1 3
Regions withIndependent, Competing States (in Chronological
Regions where premodern states competed with others for resources and domination
Table 3 Roles of queen rulers and main wives
a Presence of a trait (marked as 1) is based on literature and logic, or speculation. Absence of a trait
(marked as 0) means that literature or logic implies that the trait did not exist or no information on the
trait could be found.
Rules: They ruled the polity. 1 means they ruled in their own right or as regents.
Co-rules: They shared rulership status and duties with their husbands, either temporarily (e.g., when the
ruler was away) or permanently (e.g., when women ruled the female population).
Considered divine: Considered divine during their lifetime, not just post-mortem.
Mediates between gods and the people: They could mediate between the gods and people—of any social
Leads war: They led troops in war or developed military strategy to gain/defend territory.
Manufactures: They actually made fine goods, especially textiles.
Controls estates: They controlled and/or owned their own estates. They had the right to determine what
to do with the produce from their estates.
Alliance pawn: They were pawns in their royal kinsmen’s alliance games.
Begets royal children: They were responsible for begetting royal children—possible heirs to the ruler.
Administers: They administered or supervised estates, the household (palace), or the production of fine
Gains wealth: They could gain wealth from their dowries, gifts, their own estates, and sometimes tribute.
Receives gifts/tribute: They gained wealth from tribute/gifts.
Empires Combination No. of states
Mari Hawai’i Egypt China Inca Aztec Maya Zapotec
Rules 0 1 1 0 0 1 1 0 4
Co-rules 1 1 0 0 1 0 1 1 5
Considered divine 0 1 1 1 1 0 0 0 4
Mediates between
gods and the people
1 1 1 1 1 0 1 0 6
Leads war 0 0 0 1 0 0 0 0 1
Manufactures 0 0 0 0 0 1 1 1 3
Controls estates 1 1 1 1 1 1 0 1 7
Alliance pawn 1 1 1 1 1 1 1 1 8
Begets royal children 1 1 1 1 1 1 1 1 8
Administers 1 1 1 1 1 1 1 1 8
Gains wealth 1 1 1 1 1 1 1 1 8
Receives gifts/tribute 1 1 1 1 1 1 1 1 8
No. of traits 8 10 9 9 9 8 9 8
Author's personal copy
1 3
Journal of Archaeological Research
are represented by two of the societies. The Mari Kingdom in Old Babylonian times
was one of several kingdoms in the Tigris–Euphrates area until it was conquered by
its southern neighbor, Babylonia. The other is the Hawaiian archipelago just before
European contact. About that time, four polities in the Hawaiian archipelago were
frequently shifting alliances and warring with each other to dominate the region. But
domination (empire) did not occur until after European contact. I expected that the
royal women in these two regions of competing states would have in common more
kinds of agency than the women in the other societies, but I did not find this to be
so. I surmise that other patterns of behavior, such as descent rules, obfuscate a clean
pattern here.
Mari Kingdom (Old Babylonian Period, 2000–1600 BC)
In the time of Old Babylonia, Amorite kingdoms flourished along the Euphra-
tes River. Each kingdom was composed of several nomadic tribes, but factions
and allegiances in the region shifted so frequently that Sasson (2015, p. 344), the
major American scholar of Mari, writes that men played “musical thrones,” and rul-
ers made tribal chiefs, top administrators, diviners, and even their own wives swear
oaths of fealty (Sasson 2015, pp. 29−30, 345; Ziegler 1999b, pp. 59, 97, 209).
One of the Amorite kingdoms was Mari, which ruled four provinces and was cen-
tered in a city of about 40,000 people 400 km north of Babylon (or one hour south
of present-day Aleppo). When ruled by the Lim Dynasty (ca. 1820–1761 BC), it was
a trade center that linked the Euphrates to the Levant. The last Lim king, Zimri-Lim,
was an ally of King Hammurabi of Babylon until the latter conquered the polity and
destroyed the urban center. Although Zimri-Lim then disappeared from history, his
archives of over 17,000 clay tablets immortalized him and his predecessors (Sas-
son 2015, p. 4). Most tablets were official reports, but several hundred were letters
between Zimri-Lim, his principal wife, Shibtu (Shiptu), and his married daughters.
These letters describe their lives, activities, and agency.
King Zimri-Lim had two main wives, 12 known secondary wives, and numer-
ous concubines. (Mari rulers practiced polygyny but did not marry close kin, i.e.,
a sibling, half-sibling, child, or first cousin.) His first principal wife, Dam-hurasi,
had been the daughter of a king and the queen consort of the man Zimri-Lim
overthrew to regain the throne. She bore his heir-apparent, who had to be male
(Stol 2016, p. 491). Shibtu, the daughter of the king of Yamhad (near Aleppo),
replaced Dam-hurasi as principal wife (Ziegler 1999b, pp. 54–55). Although she
did not produce the heir-apparent, she bore him three sons and many daughters.
Her sons would have helped Zimri-Lim secure the succession for his dynasty
had not King Hammurabi destroyed the kingdom, while her daughters gave their
father the means to secure outside alliances and his vassals’ fealty (Postgate 1994,
p. 48; Sasson 1973). Furthermore, Shibtu’s father’s superior rank to Zimri-Lim
and Dam-hurasi meant that she out-ranked them both. Therefore, she along with
her entourage had special privileges. She got to live in her own residence out-
side the palace walls for part of each month and in the palace women’s quar-
ters the other days (Lafont 2001, pp. 135–136; Ziegler 1999a, pp. 55–56, n. 7, b,
Author's personal copy
Journal of Archaeological Research
1 3
pp.16–18, 42). The other women lived in the women’s quarters all the time, but
scholars believe they were free to come and go as they wished (Lafont 2001, p.
136; Ziegler 1999b, p. 7).
Zimri-Lim’s mother (Addu-duri) out-ranked all other palace women. The queen
mother might have lived in the women’s quarters for a short time, but she soon
obtained her own quarters outside the palace, too. She had her own wealth from gifts
(such as oil and prestige objects) and from her ownership of estates and livestock.
She administered her own properties, employing many servants (Ziegler 1999b, p.
44, 50−51).
Zimri-Lim had at least 23 daughters (Sasson 2015, p. 110). He married many
of them to minor kings or his administrators. Other daughters stayed home or were
made priestesses of high rank (Ziegler 1999b, p. 46). Like their mothers, Zimri-
Lim’s married daughters out-ranked their husbands, most of whom were his vas-
sals—rulers of smaller urban centers that he had incorporated into Mari territory
through conquest (Sasson 2015, p. 110). The letters between father and daughters
make clear that these princesses did not choose their husbands; their father assigned
them mates for his own purposes (Sabloff 2018; Sasson 2015, p. 344).
Mari princesses’ superior rank to their vassal husbands meant that they usually
became main or principal wives (Ziegler 1999b, p. 42). Their rank (really, their
fathers’ rank) influenced succession, for their sons had a greater probability of
becoming the next king than the sons of other wives or concubines. Mari royalty
did not practice primogeniture. Although Mari royalty traced succession through the
father (except insituations of usurpation), the mother’s rank (stature) was important
in determining the next ruler (Sasson 2015, pp. 22–27). Thus Zimri-Lim’s daughters
were a significant gift. In fact, one of the cuneiform tablets is a request from a vassal
for Zimri-Lim to grant him a daughter in marriage (Sasson 1973, p. 62).
Despite their wealth and high status in their husbands’ courts, Zimri-Lim’s
daughters’ movements were circumscribed. They were assigned to the women’s
quarters in their husbands’ homes just as they had been in their natal home. Since
their ruler husbands apportioned food and supplies to the women’s quarters (as did
Zimri-Lim), their lives were circumscribed. One daughter complained that her hus-
band kept her from firewood and food. Another wrote that her husband threatened
to kill her (Sasson 1973, pp. 68–69, 2015, pp. 116–117). The princesses were not
allowed to leave their husbands without Zimri-Lim’s permission. At least two peti-
tioned their father to be brought home. One sought refuge when her husband refused
to make her his main wife and abused her. She wrote,
… [The main wife] has forced me to sit in a corner, and caused me, like an
idiot, to hold my cheek in my hand. His eating and drinking is continuous
before this woman, [while I could not] open my mouth. He has strengthened
the guard over me, invoking my lord’s [Zimri-Lim’s] authority. What about
me? Should I witness (such) unhappiness here? (Sasson 1973, p. 65).
Zimri-Lim’s wives and daughters were expected to stay loyal to their fathers
and continue worshipping their natal gods after marriage (Sasson 1973, p. 77; Stol
2016, p. 492). They were dependent on their fathers’ continuing generosity, for he
sent them to their weddings with gifts and servants. After marriage, the daughters
Author's personal copy
1 3
Journal of Archaeological Research
continued to request that their father send them extra servants (Sasson 1973, p. 63,
2015, p. 68).
Like Zimri-Lim, his main wives and daughters were not considered divine in life
or in death. The authority of Lim Dynasty rulers derived partially from their posi-
tion as “chief servant” of the gods and because the main god, Itur-Mer, supposedly
favored them (Sasson 2015, pp. 22–26, 35, 132, 268; Ziegler 1999b, p. 39). Royals
also communed with the gods and prophesied from the “knowledge” they learned
while in that communication (Sasson 1973, p. 60, 1995, p. 911).
Mari main wives, whether the principal wife of the king or one of his married
daughters, perhaps exhibited a full spectrum of political agency. It is possible that
they made policy. Queen Shibtu was second-in-command, right behind her husband
(Sasson 1973, p. 69; Westenholz 1990, p. 518; Ziegler 1999b, p. 43). Shibtu fre-
quently represented her husband, making official visits in his name and traveling
throughout the kingdom for him. When Zimri-Lim was away from the palace—per-
haps touring his kingdom or waging war, he entrusted the affairs of the kingdom
to her, although the governing council was consulted for major decisions (Maisels
2010, p. 118). She collected tribute and taxes. She received foreign dignitaries,
supervised local officials and provincial governors, and ran the palace (Lafont 2001,
p. 130; Sasson 2015, pp. 73–75, 136, 232; Stol 2016, p. 504; Ziegler 1999b, pp. 30,
43). While it may appear that the women ruled in their husbands’ absence, in effect,
the women played at autonomy, for they checked everything with their husbands
through correspondence (Ziegler 1999b, p. 43). Therefore, it is debatable whether or
not they made policy.
Main or principal wives certainly influenced policy. Their visions and dreams of
political events influenced their father’s and husbands’ military decisions (Sasson
1973, p. 60, 1995, p. 911). In that way, they acted as a kind of counsel to the king.
The royal women influenced their husbands’ behavior in other ways as well. They
had frequent access to the king and his entourage, for they frequently dined with
him, his major officials, and visiting dignitaries (Sasson 2015, pp. 306–307). While
I found no examples of women actually changing the men’s behavior in literature, it
would seem logical that the women at least modified it on these and other occasions.
The women also influenced people below them in rank. Shibtu administered the
palace and its economy (Ziegler 1999b, p. 30). A sister of Zimri-Lim bragged that
her husband “entrusted his large palace to my control, thus giving me full satisfac-
tion. Under my authority are now 200 women, including singers, weavers, and care-
takers” (Sasson 2015, p. 69). Shibtu and other royal wives also supervised textile
workshops (Lafont 2001, pp. 130–131, 137–138; Stol 2016, p. 510; Ziegler 1999b,
pp. 30, 55 n. 347). Zimri-Lim’s mother was authorized to administer reserves of
precious metals and wool. And she oversaw the construction of a stable for her son.
When she died, Shibtu took over her responsibilities (Stol 2016, p. 494; Ziegler
1999b, p. 51).
Main and principal wives were patrons in their own right. Their wealth gave them
the means to take on this role. Shibtu owned several estates, as did Dam-hurasi and
other queens. Shibtu received tribute and continuous gifts from her father, among
others. This wealth gave her the wherewithal to patronize such men as “a merchant,
two goldsmiths, towns in the countryside” (Stol 2016, p. 493). Zimri-Lim’s mother
Author's personal copy
Journal of Archaeological Research
1 3
even acted as patron to her son, for she loaned him great quantities of grain (Sasson
2015, pp. 67–68; Ziegler 1999b, pp. 50–51 n. 318, 54, 56–57, 96–97).
Zimri-Lim’s daughters obtained wealth through their dowry (Sasson 2015, pp.
107–108). This suggests that they owned livestock. As their mother, aunts, and
grandmother owned estates, I infer that they did also. Once they became the main
wife, they received “gifts of the city” (Sasson 1973, p. 65). Wealth and the abil-
ity to make people indebted to them must have resulted in the princesses becoming
patrons. Their clients would then become obligated to supply things they wanted,
such as information (see below).
Royal women sometimes acted as go-betweens for kin and affines. Zimri-Lim’s
daughters advocated on behalf of their husbands to their father. Princess Ibbatum
defended her husband’s loyalty to her father even though he was scheming behind
her father’s back. The married princesses also reminded or even chastised their
father for not supporting their husbands’ war efforts (Sasson 1973, pp. 62–63,
72–76, 2015, p. 117).
These women also spied for their father, who sent a female scribe as part of their
wedding party so that she could “write home” continuously. The information they
gleaned from partaking in meals and rituals with their husbands and courts was dic-
tated to their scribes and sent to their father by courier. They warned him of possible
threats, and they passed along messages and warnings from other officials (Sasson
2015, p. 110). Sasson (1973, pp. 69–70) writes, “While a queen in Ilansura, Sima-
tum [Zimri-Lim’s daughter] still considered herself a Mari princess. . . . She retains
interest in her father’s struggles and tries to keep him abreast of political and mili-
tary events. . . .”
In sum, the lives of main wives in the Mari Kingdom were circumscribed. They
had no say in whom they married; they lived in the women’s quarters; and their
husbands controlled their allotment of firewood, cooking oil, food, etc. When they
wanted something extra, like a hairdresser, they appealed to their father. Yet they
covered most forms of political agency, acting mostly for their fathers and some-
times for their husbands. Perhaps they made policy—at least minor decisions—
when they met with foreign diplomats and provincial administrators. They clearly
influenced policy and the behavior of anyone from their ruling husbands to others
beneath them through their ability to prophesy and distribute their personal wealth
among followers. They also advocated for their husbands and spied for their father.
Their letters home are full of obeisance to and praise of their father. Yet they wrote
Protohistoric Hawai’i (ca. AD 1570–1590 to1778)
Beginning with the consolidation of the districts of Hawai’i Island under the lead-
ership of ‘Umi-a-Liloa (ca. 1570–1590), states emerged, dissipated, and reformed
in the archipelago. By the time of European contact (1778), four major states had
formed on the islands of Hawai’i, O’ahu, Maui, and Kaua’i. They built temporary
alliances and warred with one another for territorial supremacy over a population
Author's personal copy
1 3
Journal of Archaeological Research
estimated to be between 300,000 and 396,000 (Hommon 2013, pp. 225–226; Kirch
2010, p. 33, table2.1).
Archaeological information, historical records, and ethnohistoric accounts made
by indigenous people and Europeans soon after contact purport that Protohis-
toric Hawaiian society was highly stratified. Ali’i nui, or rulers of states and their
royal families, held the highest rank. They came from the elite (ali’i) stratum, as
did chiefs, warriors, and administrators. Within this stratum were about 11 grada-
tions of rank (Hommon 2013, pp. 247–251). Rank determined how much power and
privilege people had, and so elite men and women strove to raise their (and their
children’s) rank. Men could rise through valiant deeds (military prowess) and hyper-
gamy—marrying someone of higher rank. Women could rise only through hyper-
gamy. In short, the ideal marriage for men and women alike was hypergamy (Kirch
2010, pp. 204–205; Linnekin 1990, pp. 102–108).
Gender separation was another major characteristic of the society. Men and
women ate separately and worshipped separate deities. Division of labor fell along
gendered lines. Men fished, farmed, and cooked the meals, while women made
mats, bark cloth, personal ornaments, and other goods used for gifts, tribute, and
offerings to the deities. There is some information that women sometimes worked
in the fields, too, but it was not the norm. Women were forbidden access to men’s
spaces, where wealth and power accumulation occurred, but like the men, women
could own land and the goods they produced (Linnekin 1990, pp. 5, 13–15, 55, 107).
Elite women were more restricted than commoner women. Elites slept in gen-
der-specific dwellings while commoners did not. Elite women were forbidden from
having liaisons before they had produced at least one heir for their husbands and
generally remained married longer than commoners (Linnekin 1990, pp. 107, 123).
But Kirch (2010, p. 105) recounts that a “high-ranking chiefess” (of ali’i, not ali’i
nui rank) had liaisons with several Hawaiian Island rulers. The added restrictions on
ali’i women were to help men determine the paternity of offspring, which factored
into the islands’ ranking system.
Most rulers were men, but women sometimes attained this position, especially
when they ranked higher than their husbands (Hommon 2013, p. 133). Kalani-mau-
nia ruled several O’ahu districts. She improved agricultural and aquacultural produc-
tivity. Her kingdom broke apart after her death because she left her polity to all three
sons and a daughter (Hommon 2013, pp. 263–264; Kirch 2010, pp. 112–113). Kam-
akahelei ruled Kaua’i alone for 10 years. After that, she co-ruled with Ka’eokulani
(Linnekin 1990, pp. 22–23). Other women co-ruled with a husband or brother. (Such
diarchy also existed between brothers.) The co-ruling women, who often had supe-
rior rank to their male co-rulers, were usually symbolic heads, while their partners
were active in war and governance (Valeri 2014, p. 183). Kaikilaninui and Keakea-
lani-wahine on Hawai’i Island and Kalaimanuia on O’ahu were symbolic queens
(Hommon 2013, pp. 263–65, 270). Male or female, Hawaiian rulers owned and con-
trolled the land. When new leaders took power, they had the authority to reallot the
land, coastal waters, and administrative offices to their subordinates, or loyal follow-
ers. They received and redistributed tribute as well. Rulers had the right and duty
to predict the future or start and stop wars, although a council advised them (Kirch
2010, pp. 73–74; Valeri 2014, pp. 44–46, 179–180).
Author's personal copy
Journal of Archaeological Research
1 3
Like their male counterparts, ruling women were considered divine, the voice of
the gods. They interceded with the gods for “the well-being of all of the people” and
interpreted the gods’ wishes. Ruling women participated in parts of men’s ceremo-
nies and presided over their own as well. They had their own deities separate from
the men’s (Linnekin 1990, pp. 25–26; Valeri 2014, pp. 140–153).
Ruling women were immune from punishment if they broke the kapu rules. They
had the same access to wealth as male elites, for they had the same access to per-
sonal property as the men. They exchanged and distributed gifts to influence behav-
ior just as the men did. They apportioned land rights, patronized their followers, and
“made their own deals with traders and kept separate accounts” (Linnekin 1990, pp.
55–56). They had the power of life and death over their subjects, like male rulers.
And they supervised the making of feather capes, just as their male counterparts
did. Symbolically, their worn garments were considered kapu (here, forbidden) to
anyone else, just as the garments of the male rulers were (Linnekin 1990, pp. 13–14,
35, 55).
Despite the rosy picture painted above, female rulers’ ritual–political participa-
tion—and therefore their power—was limited. They were not allowed to engage in
battle or take part in political discussions with male rulers/chiefs or lead wars. And
their participation in war-temple rituals (including human sacrifice) was limited.
Even though rulers, they did not worship the gods of the state, i.e., the war gods.
Therefore, their power was somewhat limited (Hommon 2013, p. 132–133; Linnekin
1990, pp. 25–30; Valeri 2014, p. 183).
The women who became main wives to polygynous rulers always held the high-
est rank among wives. If the third wife a man married had higher rank than the first
two, she was considered the main wife. Rank was determined by bilateral descent,
that is, one’s genealogy traced through the mother and father. Main wives were usu-
ally the kin of ali’i nui or ali’i from other islands. By the 17th and 18th centuries,
the ideal main wife was the ruler’s sister, half-sister, or niece (Hommon 2013, pp.
19–20; Kirch 2012, p. 236). When married siblings produced an heir, their offspring
had the highest possible bloodline and were considered divine. As the ideal succes-
sion pattern was primogeniture, the high rank of the heir-apparent was invaluable
to his attaining the throne. Despite the rule of primogeniture, the incumbent often
named a different successor, or kin fought each other for the throne. And the high
priest could deny a contender’s legitimacy (Kirch 2010, p. 37, 2012, p. 220; Valeri
1985, p. 136, 2014, p. 50).
The functions of the main wife were to produce heirs of equal or higher status
than their parents and cement alliances among rulers/chiefs. As rank depended on
the lineage of both parents, women were considered “essential to the dynastic strate-
gies of male chiefs. ‘To seek a lord’ was to seek a chief’s high-ranking daughter. As
the sisters of other male chiefs, they were also critical as the means of forging alli-
ances” (Linnekin 1990, p. 153).
Main wives maintained some power and therefore agency. At least some became
rulers and others became regents. As such, royal women must have made policy,
although their preclusion from the main temple and from battle probably limited
their ability to act (Kirch 2010, pp. 65, 105). Queens, main wives, princesses, and
commoners tried to change policy when they attempted to overturn the rules of kapu
Author's personal copy
1 3
Journal of Archaeological Research
or at least break their husbands’ constraints on their sexual behavior (Linnekin 1990,
pp. 24, 35, 153).
While main wives (as opposed to female rulers) were not decision makers, their
roles as seers or prophets must have given them the authority to influence their hus-
bands’ decisions to go to war or support a family member (Kirch 2010, p. 100).
Main wives influenced some behavior, for they had a say in who married their
daughters, and they generally influenced their children’s behavior. They had their
own retainers and loyal followers and maintained and distributed their own posses-
sions. They controlled land and the wealth derived from it. Therefore, they had the
means and opportunity to become patrons to traders, workmen, and other elites just
as ruling women did (Linnekin 1990, pp. 5, 25–26, 59–61, 153).
Could main wives have acted as intermediaries or spies? While the literature
does not specify, I assume that women had limited opportunity to mediate or advo-
cate between husband and kin because of the strict separation of men and women
(Linnekin 1990, p. 25). However, they might have passed information about other
women to their spouse. The probability that royal women acted as spies is rather
low as well. Rulers controlled information flow via messengers and spies (Hommon
2013, p. 132). Whether or not these messengers and spies communicated with the
married daughters of rulers is not clear.
In sum, perhaps we should consider Protohistoric Hawaiian royal women as “sep-
arate but (almost) equal” to the men. Women lived in their own spaces, worshipped
their own deities, and controlled their own wealth. Royal women had authority over
their own households and other women. But because women were blocked from
total access to male rulers and the state religion (i.e., the war temple), they were not
equal to men. Their inferiority was symbolized by the limited kinds of food they
were allowed to eat and their circumscribed political agency.
Old Kingdom Egypt, Late Shang China, Aztec, and Inca were all empires, ruling
vast territories that encompassed conquered polities. As in the other comparisons,
all four limit rulership to men except when women acted as regents. There is no pat-
tern of female political agency that distinguishes this group from the other two.
Old Kingdom Egypt (2686–2160 BC, orDynasties 3 Through 8)
Old Kingdom Egypt—the time of major pyramid construction—is the earliest
period in which centuries of excavation and decipherment of monuments and text
give us sufficient information to reconstruct at least some of the political agency
of main wives. At that time, the empire stretched north to south from the Delta to
Aswan and west to east from Libya to southwest Asia. Wenke (2009, p. 70) esti-
mates that the empire encompassed somewhere between 1 and 2 million people. Old
Kingdom rulers traded with Aegean communities, raided as far north as Syria, and
Author's personal copy
Journal of Archaeological Research
1 3
fought wars to enrich the kingdom with slaves, herds, and goods (Frankfort 1978, p.
55; Smith 1962, pp. 39–43; Wengrow 2009, pp. 147–157).
Rulers decided policy (with the help of the vizier and other advisors) and con-
trolled critical resources such as trade goods and land distribution, although this
power eroded over time (David 1998, p. 42; Moreno Garcia 1997; Smith 1962, p.
53). As divine beings, they were responsible for maintaining the orderliness of the
cosmos, the rains and the Nile’s flooding cycle, and therefore the well-being of the
populace. They were called di ankh, “the giver of life,” to their people (Baines and
Yoffee 1998, p. 206; Frankfort 1978, pp. 54, 59).
Rulers’ main wives were supposed to be their close kin (Smith 1962, pp. 24,
32; Tyldesley 2006, pp. 36–63). In the Fourth Dynasty, Meritites (Meryetyotes I),
daughter of King Snefru (Seneferu), married her half-brother, King Cheops (Khufu).
The Sixth Dynasty King Pepy II married two of his sisters (Dodson and Hilton 2004,
pp. 52–53, 73). However, the lack of information about the parentage of many main
wives suggests that not all had been royal. Roth (2009, pp. 7–8) reports that at least
some were foreigners (see also Smith 1962, pp. 49–50).
Although the rule of succession was primogeniture through the patriline, rulers
also were supposed to be the sons of one of the king’s main wives. Main wives
attained the status of official (principal) wife when they gave birth, especially to
sons who could become the next ruler. The king’s mother held the second high-
est position at court (following the ruler) (Roth 2009, pp. 2–5; Smith 1962, pp. 10,
24, 49, 54). No matter the parentage of main wives, the wedding ceremony made
them the “earthly embodiment of [their] divine counterparts” (Roth 2009, p. 6). In
other words, their direct connection to deities determined their high rank. Like the
rulers, queens wore headdresses with symbols of the gods (goddesses) to associate
themselves with the divine. Some were buried in pyramids or tombs next to their
husbands and were deified after death. Their divine status matched that of the rulers,
who embodied the god Horus in life and became divine after death (Kemp 1992, p.
82; Richards 2010, p. 56; Smith 1962, pp. 27–28, 32–33, 49–50).
While women did not rule in their own right in this period, they sometimes acted
as regents for their sons. In the Fifth Dynasty, Khentkaues I acted as regent (Dodson
and Hilton 2004, pp. 64–65). And in the Sixth Dynasty, King Teti’s wife, Iput, may
have acted as regent for her son (Smith 1962, pp. 48–49).
Main and official wives (when not regents) had the opportunity to exhibit some
political agency. Even though Roth (2009, p. 7) writes that “there is little evidence
that queens held political influence,” their ability to represent the wishes of the gods
to their husbands gave them the capability to influence policy. They also influenced
peoples’ behavior. They ran their own households and controlled their own agricul-
tural estates. Therefore, they had the wealth and means to build their own relation-
ships of obligation, or patron–client relationships (Dodson and Hilton 2004, p. 78;
Kemp 1992, p. 82; Smith 1962, p. 36). They also transferred the offerings they made
to the deities in their own temples or tombs to the tombs of others nearby. Such
donations gave them superiority over others and therefore some control over their
The access of main and official wives to their husbands and importance in the
kings’ religious ceremonies meant that they had the opportunity to intercede to the
Author's personal copy
1 3
Journal of Archaeological Research
king on behalf of their kin (Roth 2009, p. 7; Trigger 2003, pp. 163, 185). As par-
ticipants in their husbands’ court, main wives had the opportunity and authority to
conspire—often against their husbands. They helped their sons become ruler (as pri-
mogeniture was not always followed) and selected their daughters’ husbands (Roth
2009, p. 6; Smith 1962, pp. 23, 32).
In sum, when principal wives acted as regents for their sons, they displayed
political agency by making policy. Main wives’ agency was circumscribed by their
exclusion from state political events (Roth 2009, p. 7). However, their divine author-
ity, their wealth, and their access to the ruler and his court gave them the oppor-
tunity and capability to practice several types of political agency. They influenced
(rather than made) policy by interpreting the wishes of the gods. They controlled the
behavior of their households and influenced the behavior of courtiers through their
patronage. They acted as go-betweens for their kin, thus influencing the decisions of
the king. And they conspired to replace their husbands either with their sons, their
kin, or their favorites.
Late Shang China (ca. 1250–1045 BC)
Modern excavations and analyses of oracle bones, bronze vessel inscriptions, and
documents from later dynasties provide a picture of Late Shang society. The empire
spread from north of the Yiluo (Yellow) River to beyond the central plain. However,
interspersed among pockets of Late Shang control lived autonomous non-Shang
peoples—agricultural villages and steppe nomads. While most non-Shang within
and beyond the empire’s borders were enemies, some were allies (Feng 2013, p.
111; Keightley 1983, p. 532; Shelach-Lavi 2015, p. 221).
Keightley (1999a, p. 270), a leading American sinologist, writes that Late Shang
political organization was based on “a confederation of patrilineal descent groups”
that fissioned and fusioned over time. Lineage groups (zu) were ranked according to
their distance to the sitting king. They served the king as administrators and warriors
and received rewards according to their position and deeds.
Rulers practiced polygyny but did not marry close kin, although they sometimes
married their patrilineal cousins (Linduff 2003, p. 71). King Wu Ding (Wu Ting),
the 11th king of the dynasty, had three main wives and 58–63 consorts (Wang 2004,
p. 111). Whichever main wife produced a son became the principal wife. Fu Hao
(Lady Hao) was named first wife when she bore a son. Upon the boy’s death, she
lost that position but remained a main wife. Fu Jing replaced her as principal wife,
for she bore several sons. Rulers had to be born into the ruling clan, but they could
be the oldest son or the brother of the deceased king. Despite this emphasis on the
patriline, the lineage of princes’ mothers and their position at court also influenced
their right to rule (Feng 2013, pp. 103–106; Nelson 2015, p. 94; Wang 2004, p.
Main wives were usually the daughters of foreign leaders and held high rank
in their husbands’ courts. However, they did not become rulers or even regents.
Their value came from three sources. Their marriage cemented alliances between
their (often non-Shang) kin and their husbands (Keightley 1979, p. 32 §37). Their
Author's personal copy
Journal of Archaeological Research
1 3
begetting of legitimate heirs enabled their husbands’ patrilineal clan to maintain
control of the state. And they assisted their husbands in ruling. They performed
various administrative tasks such as announcing royal decrees, delivering formal
messages, administering the towns assigned to them, and supervising agricultural
production. Fu Hao was “second to the king as administrator. . . .” (Childs-John-
son 2007, p. 23; see also Barnes 1999, p. 134; Li and Zhu 1989).
Wu Ding’s three main wives took an active part in hunting and military affairs.
Fu Hao recruited an army of over 13,000 troops and even led military campaigns
against the Qiang (Fiskesjö 2001, p. 132; Keightley 2012, p. 175; Linduff 2003,
pp. 70–71; Nelson 2015, pp. 95–96).
Main wives assisted their husbands and sons in sacrificing to the gods and
ancestors. They prepared oracle bones (used in divination) for cracking, thereby
“consecrating” them. Even though rulers monopolized communication with the
main god Di, main wives had the authority to appeal directly to their husbands’
ancestors (Campbell etal. 2011, p. 1287; Keightley 1983, p. 550, 1999a, pp. 262,
275; Nelson 2015, pp. 95, 97). Upon death, queen consorts could become divine
ancestors in their husbands’ ancestral clan (i.e., not their own clan), just as the
rulers did (Keightley 1999a, p. 256). All three of Wu Ding’s main wives became
royal ancestors (Nelson 2015, p. 95).
The position of main wives at court gave them the opportunity to engage in
several types of political agency—for their kin and themselves. They did not
make policy decisions, but they did influence policy. As diviners or interpreters
of oracle bones in their own right, they helped predict the future of wars, har-
vests, etc. (Linduff 2003, p. 70). Fu Hao and Fu Jing might have had shamanic
roles as some grave goods (mirrors, musical instruments) suggest (Nelson 2015,
p. 97).
The position and wealth of main wives gave them the means to influence the ruler
and courtiers. The women gained wealth from significant tribute goods and gifts
(Childs-Johnson 2007, p. 22; Wang 2004, p. 104). They administered or controlled
land (Barnes 1999, p. 134; Keightley 2012, pp. 144–145, 152; Li and Zhu 1989).
Since they were responsible for harvests, they must have had access to wealth from
agricultural production. Fu Hao’s tomb included carved jade jewelry, bone hairpins,
carved bronze ceremonial pots, and thousands of cowrie shells as well as 16 human
and four dog sacrifices (Nelson 2015, p. 96).
By giving the king tribute (partly in the form of plastrons for making oracle
bones) and participating in royal hunts, they could build their own networks in the
kings’ court (Childs-Johnson 2007, pp. 22–23; Fiskesjö 2001, p. 132). By giving
feasts and presents to noblemen, they could establish patronage ties. By maintaining
their own “entourage of servitors,” they developed loyalty relationships for them-
selves and their own lineage (Keightley 1999b, p. 33).
Main wives also acted as intermediaries, especially between the ruler and the
women’s kin group (Trigger 2003, p. 169). Fu Hao held special audiences with the
Many Royal Daughters and blind seers (Childs-Johnson 2007, p. 23). We have no
information on main wives spying on their husbands for their kin or vice versa. Fu
Hao’s life story shows that at least some main wives were extremely loyal to their
Author's personal copy
1 3
Journal of Archaeological Research
In sum, the Late Shang patrilineal structure made the women ineligible to rule,
even as regents. Therefore, they did not make policy. Still, they had access to power.
They participated in family and public rituals; they joined the court on royal hunts;
and they acted as intermediaries between the king and his subjects or foreign emis-
saries. They controlled whole settlements and agricultural estates, which gave them
access to wealth. Because of their position, divination powers, and wealth, they had
the means to influence policy. And their access to a wide network of individuals
gave them the opportunity to influence individuals, including the ruler. Aside from
acting as patrons to their clients, they also acted as intermediaries, often between the
ruler and their own kin.
Aztec Empire (ca. AD 1440–1520)
Scholars have recreated the history of the Aztec Empire from excavations, decipher-
ment of texts, and the accounts of Spanish friars and Indian nobility written soon
after conquest (1519–1521). These sources explain that the Aztec Empire arose
when Tenochtitlan managed to dominate the Triple Alliance of Tenochtitlan, Tex-
coco, and Tlacopan. Aztec rulers conquered the surrounding area, and by the early
16th century, the empire encompassed 13 to 25.2 million people, with about 1.5 mil-
lion living in the Basin of Mexico (Hassig 2016, p. 62; Townsend 2009, p. 12).
Aztec rulers were considered “the embodiment of the deities” (Berdan 2014, p.
152) or the voice of Tezcatlipoca, the most powerful god, but they were not con-
sidered divine in their lifetime. It would seem that the main trait that enabled rulers
to stay in power was their ability to maintain a complex social network of follow-
ers whom they kept loyal by constantly rewarding with position and material goods
(Berdan 1982, p. 100; Clendinnen, 1991, pp. 80–81; Smith 2003, pp. 149–150).
None of the imperial rulers, selected by the council of nobles, were female. Three
women (including Moctezuma I’s daughter Atotoztli) inherited the throne but prob-
ably passed governance to their husbands or sons. The daughter of Motecuhzoma II,
Tecuichpo (later, Isabel), also inherited the right to rule but never did (Berdan 2014,
p. 144; Gillespie 1989, pp. 18–20, 101, 106–109). Still, women served as regents
(Diel 2005, pp. 83, 90–92, 2007, p. 265).
Aztec rulers were polygynous, having many wives and concubines. Main, or
principal, wives were selected for two potential benefits. First, Aztec society did
not follow the rule of primogeniture. Instead, upper noblemen (tecuhtli) fought for
rulership until Moctezuma I instituted the rule that their peers and men of stand-
ing (priests, brave warriors, etc.) elect their ruler from among those of royal blood.
Their royalty could come from a collateral relative (e.g., an uncle) or direct descend-
ant of a former king (son, grandson, etc.). He also had to be a warrior (Diel 2005, p.
101; Hassig 2016, pp. 46–47). Because elites traced their social rank through either
parent, the son whose mother had the highest rank among women was supposed to
inherit the throne as long as he demonstrated a virtuous nature and success in battle
(Berdan 2014, p. 146, 262; Hassig 2016, pp. 40–41; Pennock 2011, pp. 100–101;
Sahagún 1961b, pp. 61–64). “[P]ower, in the sense of legitimacy of rule, came with
or through these females even though only males appear in the pictorial king lists”
Author's personal copy
Journal of Archaeological Research
1 3
(Gillespie 1989, pp. 19–20). Royal women—main wives—were so valued that one
was even depicted in a codex on the same plane as her husband, facing him and ges-
turing toward him while he speaks to her (Diel 2007, p. 261).
Toward the end of the empire, rulers of Tenochtitlan and other kingdoms chose
main wives from their own kin or from descendants of a previous king, including
their agnates (Diel 2007, pp. 260, 270; Schroeder 1992). The last king, Cuitla-
huac, married one of Montezuma II’s daughters, i.e., his niece (Hassig 2016, p. 56).
Agnatic or sibling marriages gave rulers several advantages. Marriage to a woman
in their own descent group assured them and their sons the greatest legitimacy to be
named emperor (Berdan 2014, p. 146; Diel 2007, p. 259). Emperors were neither
inferior (by blood) nor obligated to some affine. And close bloodlines “diminished
the threat of split loyalty inherent in the taking of an outside bride” (Diel 2007). “[T]
he matriline was the hidden determinant of succession among the Aztecs” (Hassig
2016, p. 37).
The value of main wives also came from their ability to link rulers (their hus-
bands or fathers) to others, thus increasing the odds that alliances and patron–client
relations would be long term. Texcoco and Tlatelolco rulers used this method to ally
themselves against the more powerful Tenochtitlan ruler. Others used royal women
to declare their autonomy from the Tenochtitlan ruler. While some abused their
wives, Nezahualpilli, ruler of Texcoco (1472–1515) had his main wife executed,
ostensibly for adultery (Diel 2007, pp. 261–270).
The heightened value of main wives afforded them little political agency. Those
acting as regents may have made policy, but there is little evidence to suggest they
made or influenced policy directly. They “typically” had no position in court (Has-
sig 2016, p. 37). They had authority within the women’s compound on the palace
grounds, for they administered large, complex households and supervised the pal-
ace women’s textile and craft production (Berdan 2014, p. 180; Van Zantwijk 1994,
p. 103). Sahagún (1961a, p. 46) described the expectations of royal women: “The
noblewoman [la señora principal] [is] a woman ruler, governor, leader—a provider,
an administrator. . . . She is heeded, obeyed; she creates order; she establishes rules.”
He added, “A good noblewoman [is] . . . careful of her estate. She governs, leads,
provides for one, arranges well, administers peacefully” (Sahagún 1961a, p. 45). But
he gave no examples of women actually ruling or governing. What did they rule?
Although royal women owned estates, their husbands or sons often administered
them (Hamann 1997, p. 167; Schroeder 1992).
Main wives could modify the behavior of individuals. As in other cases, royal
women were expected to honor their own lineages, even after marrying an emperor.
By providing their kin access to the king, they indirectly influenced his behavior
and perhaps his policies. “Royal women played an important part not only in politi-
cally strengthening their husbands in the internal politics of Tenochtitlan, but also in
integrating the Aztec Empire” (Hassig 2016, p. 101). Because women could inherit
land from either parent and because they wore finery in the codices, we assume that
they had some means to accumulate wealth. In fact, Aztec society allowed women
to inherit from either parent and even their husband (Berdan 2014, pp.193−194).
Therefore, they had the means to act as patrons to some—at the least, noble women
and women in the rulers’ harem.
Author's personal copy
1 3
Journal of Archaeological Research
Royal women acted as intermediaries, especially between the ruler and their own
kin. They provided their kin cohort (father, brothers, uncles, and descendants) spe-
cial access to the ruler, interceded on their kinsmen’s behalf, and gave them infor-
mation as well. As singers and story-tellers in the palace, they had access to the
court as well as the king (Berdan 2014, p. 180; Diel 2007, p. 260). They used this
access to act for their children, their kin, and themselves.
Once in their marital home, royal women spied on their husbands for their home-
land (Diel 2007, p. 260). One daughter of a minor king told her father of her hus-
band’s war plans. “Whenever they talked war, she sent word to the ruler Tecocoat-
zin. And for this reason, when the battle had begun, she was very definitely on the
watch” and actually “set fire to the temple of Toltitlan” (Diel 2007, p. 269); Diel (p.
270) adds, women “were seen as potential threats in need of neutralization.
In summary, Aztec women had some power and respect, at least within the court
and their own households. Furthermore, because they could inherit and pass on
property, they had the means to build wealth and therefore act as patrons to clients
in their own right. Perhaps the most interesting aspect of their political agency is the
evidence that royal women acted as spies.
Inca Empire (ca. AD 1460–1532)
Knowledge of the imperial Inca period comes from intensive excavation and schol-
ars’ analysis of Spanish chroniclers’ reports. These inform us that the Inca settled
in the Cuzco Valley ca. AD 1100. About 1438, King Pachacuti initiated the impe-
rial period, which ended with the Spanish conquest ca. 1531. By then, the empire
encompassed almost 8 million people, extending about 2,500 miles north to south
and 300 miles east to west (Bauer 1998, p. 1; Cobo 1990, p. 39; D’Altroy 2002 p.
Inca rulers (Sapa Inca) controlled resources, including water, although they were
advised by their kin, or panaqa, and the high priest. They also received advice from
diviners who helped them communicate with the gods (D’Altroy 2002, p. 99). They
controlled the Inca nobility in Cuzco and the provinces by distributing gifts as well
as meting punishments (Rostworowski de Diez Canseco and Murra 1960, p. 418).
They distributed provisions to the nobility from their own storehouses (D’Altroy
2002, p. 91). They assigned to their underlings wives and titles as well as rights to
land and water. In this way, Sapa Inca built an elaborate patronage system (Kolata
2013, pp. 211–212, 217, 230; Rostworowski de Diez Canseco 1999, pp. 42–44).
Sapa Inca were polygynous. Emperor Wayna Qhapaq had about 6,000 “consorts”
(D’Altroy 2002, p. 103). Rulers married to form alliances, and so their wives and
concubines came from many ethnic groups. However, their main wives (qoya) were
usually their own kin—usually sisters or first cousins. The progeny of these incestu-
ous marriages would therefore have the purest blood that a son of the sun god could
produce (Cieza de Leon 1883, p. 20; Cobo 1893, p. 187; Isbell 1997, p. 63; Jenkins
2001, p. 178).
Potential rulers fought for succession, for there was no rule of primogeniture. If
the main wife produced no “fit” sons, the “royal elders” chose an heir that the main
Author's personal copy
Journal of Archaeological Research
1 3
wife then had to adopt (Kolata 2013, pp. 207–208; Rowe 1946, pp. 209, 257–258).
This was because rulers “identified closely with their mother’s [kin group]”
(D’Altroy 2002, pp. 103–104). It appears that Inca royalty practiced bilateral, or
“parallel,” descent (Silverblatt 2015, p. 301).
Inca rulers claimed descent from the sun god and were worshipped as divinities
in their lifetime. In death, they were gods who could mediate “between heaven and
earth” (Kolata 2013, p. 225). Their mummies were feted, fed, and consulted. Despite
divine status, rulers had to frequently demonstrate their warrior abilities (Kolata
2013, pp. 203–204; Rostworowski de Diez Canseco 1999, pp. 81–82; Rowe 1946,
p. 257; Silverblatt 1987, p. 45). Main wives who married the Sapa Inca ascended
the throne on the day of their husbands’ ascension to the throne. As daughters of
the moon goddess, they were divine and were mummified upon death, just as their
spouses were (Isbell 1997, pp. 45, 56; Silverblatt 1987, pp. 57, 59).
Inca society was divided not only by class but also by gender. Because royal
women represented the moon goddess and diverse female deities, they comple-
mented royal men who represented the sun god and other male deities. Fertility (of
the land and people) depended on the interaction of both sun and moon, or male and
female. Therefore, men and women were originally interdependent. But as Inca rul-
ers became more and more focused on conquest, women’s place diminished (Silver-
blatt 2015, pp. 303–304). Still, Inca queens ruled the parallel universe of women’s
lives. Their duty—second only to begetting potential heirs—was to assure the well-
being of the empire’s women. They mediated between deities and women in cer-
emonies. They supervised the education and activities of young noblewomen who
lived in Cuzco and surrounding areas (Silverblatt 1987, p. 63). One month a year
the entire empire celebrated the queens and female deities, who symbolized and pro-
moted agricultural and human fertility, in the Qoya rami festival. The rest of the
year, main wives led rituals to honor female deities (Silverblatt 1987, pp. 55–65).
Queens’ religious, political, and economic position was manifest in their political
agency. They probably made some policy, for they ruled in their husbands’ absence
(D’Altroy 2002, pp. 106, 109; Silverblatt 1987, pp. 59–60). “If the Inca’s privy coun-
cil, composed of delegates from the empire’s four provincial subdivisions, could not
reach an accord, the matter was turned over to the queen” (Silverblatt 1987, p. 59).
Queens influenced policy, acting as advisors/counselors to the ruler.
Royal women undoubtedly influenced the behavior of others. They helped the
Sapa Inca select marriage partners for all the nobility. They influenced the marriages
of their own daughters, advocating for particular matches or blocking a daughter’s
marriage to the king (D’Altroy 2002, pp. 103, 106). In supervising underlings—from
agricultural field hands to priestesses and the concubines the ruler assigned them,
they developed “bonds of allegiance and obligation” among women, often women
from several ranks in Inca society (Silverblatt 1987, p. 63). They could bind men
as well as women to their bidding, for they had the wherewithal to act as patrons.
While they brought a dowry to their marriage, they also received gifts (tribute) and
owned estates (D’Altroy 2002, pp. 103−106, 130; Rostworowski de Diez Canseco
1999, p. 43). Like all Inca women, qoya could pass on their property to their kin.
Thus main wives had the means to patronize men, who gave them gifts and received
patronage—sometimes in the form of feasts in their own palaces.
Author's personal copy
1 3
Journal of Archaeological Research
Main wives acted as intermediaries between their powerful husbands and
their own kin, often representing the interests of their kin group to the Sapa Inca
(D’Altroy 2002, p. 103). It does not seem logical that they acted as spies, for their
kin and their affines were one and the same. However, factions within the royal fam-
ily probably used the information that qoya learned.
In short, imperial women could not become the Sapa Inca or lead military expe-
ditions. But they controlled the world of women and were responsible for all wom-
en’s well-being. They also ruled in the men’s absence. They had the power to control
or influence people’s behavior—from advising their husbands to selecting marriage
partners and acting as patrons. They acted as intermediaries and advocates between
their husbands and kin. Whether or not they spied is not known.
Regions Containing aCombination ofIndependent States
The Late Classic Maya and Postclassic Zapotec are an interesting comparison
because the sociopolitical patterns of the former arose from a less complex state for-
mation while the latter illustrate what happens after the breakdown of a large state
(Monte Albán). Although both regions are in Mesoamerica, the time gap between
them suggests that the Postclassic Zapotec did not borrow ideas or organization
from the Late Classic Maya.
Late Classic Maya Lowlands (AD 600–800)
Encompassing over 10 million people in 250,000 km2, the Late Classic Maya were
organized into a network of kingdoms (Blanton et al. 1996; Chase et al. 2002, p.
253). Some were autonomous city-states while others—Tikal, Calakmul, Copan,
and Caracol— dominated vassal states at various times (Marcus 1976, pp. 44–50,
191; Martin and Grube 2008, p. 104; McAnany 2013b, p. 144; Munson and Macri
2009, pp. 429–432).
Archaeological and epigraphic research informs us that rulers in the different
types of states were military leaders who competed for trade routes, prisoners, and
natural resources, including land and water (Marcus 2006, pp. 216–217; McAnany
2013a, p. 239; Miller 1993, p. 408). Late Classic rulers wielded significant power,
although there was always a governing council that advised them (Sharer and Trax-
ler 2006, pp. 89, 697), and there is good evidence that factionalism mitigated autoc-
racy (McAnany 2013b, pp. 144–154; Pohl and Pohl 1994).
Stelae glyphs show that women as well as men became rulers (Chase etal. 2002,
pp. 256–257). There are enough examples of royal women ruling as ajaw (ruler)
or kaloomte ajaw (supreme warrior ruler, or leader who ruled other ajaw) as well
as their own polity (Guenter 2014, p. 414) to warrant the idea that they were not
necessarily exceptions to the rule of male succession but acceptable occupiers of
the throne (Ardren 2002, p. 1; Sharer and Traxler 2006, p. 89). Women attained
rulership through various means. Josserand (2002, p. 148) suggests that it was more
Author's personal copy
Journal of Archaeological Research
1 3
important to have royal blood than to be male, and so some inherited their position
from a male relative. Lady K’awiil Ajaw ruled Cobá after her father or uncle. She
ruled for 40 years even though she was married (Guenter 2014, p. 410). Princesses
whose kin married them to rulers of inferior rank sometimes ruled or co-ruled their
adopted home (Guenter and Freidel 2009, pp. 75-76; Martin and Grube 2008, pp.
38–39, 74–77).
There is as yet no information about women usurping a throne—leading a fac-
tional fight in order to become ruler—as some men did. (Neither McAnany nor Frei-
del could think of such a case [personal communications 2017].) Some women had
the right to rule but ceded their right to their husbands. Others ruled but later ceded
authority to their sons when the boys were old enough to take charge. Still others
ruled as regents for their heirs (Freidel and Schele 1997; Guenter and Freidel 2009,
p. 76; Marcus 2001, pp. 324–325, 2006, p. 216, fig.11.1; Martin and Grube 2008,
p. 76).
No matter what the basis for their power, women rulers had the same author-
ity and responsibilities as men. They controlled trade routes and resources, col-
lected taxes and tribute, and were responsible for the people’s welfare. Lady K’awiil
Ajaw of Cobá was “sovereign of a group of very successful warriors and statesmen
who either conquered Yaxuná or consolidated this prize of their fathers . . . . When
she died . . . her kingdom was one of the most powerful the Maya world had ever
known” (Guenter 2014, p. 418).
Joyce (2008, p. 99) provides further evidence that women rulers were equal to
men, for they were depicted on stelae holding the same symbols of power as men.
One was a bar with serpent heads on either end. Another was posing standing on the
backs of captives. Lady K’ab’al Xook of Yaxchilan, who held the title of kaloomte
ajaw, is pictured on lintels performing rituals and dressed as a warrior emerging
from a serpent’s mouth (Guenter and Freidel 2009, p. 75). Several Late Classic ste-
lae at Cobá represent women kaloomte ajaw treading on captives.
From this evidence, we may safely conclude that women rulers had all the politi-
cal agency of men rulers with the exception of selecting multiple mates and perhaps
usurping thrones for themselves. Men also participated to a greater extent in “politi-
cal events, warfare, and dancing” than women (Robin 2003, p. 326). Still, Chase
etal. (2008, pp. 137–138; Chase and Chase 2017, p. 220) point to the many tombs
and stelae featuring women and suggest that the modern emphasis on male-domi-
nated society in the Late Classic may be an outcome of Mayanists’ interpretation of
monuments rather than a true reflection of Mayan society.
Main wives of paramount (kaloomte ajaw) or autonomous (ajaw) rulers were
depicted on monuments and tombs, just as women rulers were (Chase etal. 2008,
pp. 137–138; Martin and Grube 2008, p. 145). Such portraits illustrate their impor-
tance in the minds of the political elite. As Late Classic Maya rulers practiced polyg-
yny but not close-kin marriage, their main wives usually came from other places
(Houston and Inomata 2009, p. 150; Marcus 2006, p. 216). It is not yet clear whom
paramount rulers married. There is evidence that a royal Palenque woman married
into the Copan dynasty (Pohl and Pohl 1994), but there is little evidence for the
origins of the main wives at Tikal or Calakmul, for example. There are many cases
of the female kin of powerful rulers marrying subordinate rulers and founding new
Author's personal copy
1 3
Journal of Archaeological Research
dynasties. The Dos Pilas ruler sent his daughter Lady Six Sky to Naranjo to estab-
lish “a new dynastic line” (Munson and Macri 2009, p. 429). And a ruler of Holmul
was the grandson of the Naranjo ruler, Ajnumsaaj Chan K’inich (Estrada-Belli and
Tokovinine 2016, p. 162).
The primary responsibility of main wives was to produce potential royal heirs,
for Late Classic Maya did not practice primogeniture. Descent was traced bilater-
ally—through the father or mother or both. The mother’s royal status comes from
her father, not her mother. The higher the lineage of the main wife, the higher the
status of the children (Bell 2002, pp. 92–93; Chase etal. 2002, pp. 257–267; Fox
and Justeson 1986, pp. 22–25; Marcus 1974). Marcus (2001, pp. 324–325) notes
that King Pacal of Palenque “uses the name of a mythical goddess to refer to his
mother, thereby creating for himself a kind of divine right to rule, as Hatshepsut did
when she described herself as the child of Amun.” Of course, not all rulers were the
children of the ruler and his main wife. A secondary wife produced a royal heir in
Yaxchilan (Freidel and Schele 1997, p. 77). As at Piedras Negras, sometimes a son-
in-law or nephew inherited the throne (Fox and Justeson 1986, pp. 15–25). Usurpers
existed in the Late Classic Maya Lowland just as they had in Old Kingdom Egypt,
Protohistoric Hawai’i, and other societies. Sometimes usurpers married a female
heir to improve their legitimacy (Josserand 2002, p.139).
Joyce (1996) proposes that women were in gender complementarity to men. That
is, each gender had its own roles, and each was necessary for the other to perform
their roles properly. Palenque Classic texts often include the names of rulers’ moth-
ers as well as fathers, and other Classic monuments depict royal women—some-
times to their husbands’ right and at a slightly lower elevation (Joyce 1996, pp. 169,
185). Her argument for gender complementarity may be based on genealogical com-
plementarity, i.e., bilateral descent rules.
The stature of royal women came from their ability to mediate between deities
and humans as well as their lineage and child bearing. The wives of Yaxchilan rulers
acted not only as priestesses but also as conjurers (Houston and Inomata 2009, p.
148). Main wives participated in public rituals and ceremonies (Freidel and Schele
1997, p. 77; Sharer and Traxler 2006, p. 436). The wives and mothers of Yaxchilan
kings “drew blood and conducted visionary rites” (Martin and Grube 2008, p. 129).
And one Yaxchilan main wife dedicated a temple (Freidel and Schele 1997, p. 77).
In all these roles, main wives were the agent (the power) or had access to the agent.
But just as rulers were not divine in life, so main wives were not divine either.
Female rulers and main wives acting as rulers, regents, or co-rulers were active
players in politics, government, and the royal economy. These women had practi-
cally the same agency to make policy as rulers. At Dos Pilas, the mother of a young
ruler (and her brother) even “fought wars in his name against Tikal and Tikal allies”
(Freidel and Schele 1997, p. 76).
As priestesses, conjurers, and seers, main wives must have influenced policy in
the same way that Mari Kingdom royal women did. That is, they added their prog-
nostications derived from communion with the gods to the kings’ own divinations
(see Houston and Stuart 1996, p. 299; Pohl and Pohl 1994; Sharer and Traxler 2006,
pp. 696–697, 722). By creating wealth for their husbands through their spinning
and weaving, they perhaps had another opportunity to influence the men’s policy
Author's personal copy
Journal of Archaeological Research
1 3
decisions (Bell 2002, pp. 97–99; Houston and Inomata 2009, p. 148; Robin 2003, p.
By influencing policy, main wives modified the decision-making behavior of
their husbands. They also prepared their sons for the throne and manipulated the
clients they had cultivated through distribution of their personal wealth. Such wealth
came from their dowries and gifts from the nobility. Scholars are not yet certain
whether or not royal women owned land. Harrison-Buck (2017, pp. 114–115) com-
pares the Caluco of El Salvador to the Classic Maya Lowlands, explaining that a
Caluco cacique controlled 29 cacao orchards, but 26 of them were reserved for the
tribute he owed his wife’s family and for his daughters’ bride price. Several elite
women even owned cacao orchards (Fowler 2006, pp. 318–319). It is hard to judge
whether or not we can extrapolate from the colonial period in El Salvador to the
Late Classic Maya Lowlands.
As pawns in marriage, main wives were naturally in a position to act as interme-
diaries between their husbands and kin. We might think of Late Classic main wives
as the glue that held together the factionalized, constantly warring Maya. These
daughters of kings traveled near and far to help their royal kinsmen consolidate alli-
ances (Houston and Inomata 2009, p. 150; Pohl and Pohl 1994). In short, princesses
were used to turn conquests into long-term patron–client relations (Sabloff 2018).
There is no record of main wives transmitting messages or acting as intermediar-
ies between their husbands and kindred, but it seems impossible that they did not do
so. Estrada-Belli and Tokovinine (2016, pp. 164–165) imply that women did so. It
seems inconceivable that women whose ruler-kin sent them to found a new dynasty
or cement an alliance did not report back to their kin, i.e., spy. But there is not yet
evidence of them actually passing information from one party to the next.
In conclusion, main wives’ agency came from their position, the support of their
own lineage, and their wealth. They made and influenced policy, influenced the
behavior of everyone from their spouses to the royal court and members of their
entourage. While we have no examples of them spying, they did act as intermediar-
ies and advocates. Late Classic Maya royal women seem to have had greater agency
and more opportunities to rule than women in the other societies in this comparative
analysis. Perhaps it is inappropriate to attribute causality between the volatile sta-
tus of Late Classic Maya kingdoms and the impression that royal women had more
opportunity to act than royal women in the other societies presented here.
Postclassic Zapotec (ca. AD 950–1530)
Postclassic Zapotec peoples mostly lived in settlements that ranged from autono-
mous or small regional states (i.e., one city-state would rule several others) to vassal
towns of those city-states (Feinman and Nicholas 2016, pp. 59–60). Some lived in
wards (barrios) of Mixtec towns that dotted parts of the valley (Blomster 2008, pp.
22–23; Marcus and Flannery 1983, p. 217; Whitecotton 1977, pp. 138–139). Blan-
ton etal. (1993, p. 99) estimate the total valley population at 160,000.
As the bulk of information on Postclassic Zapotec royalty comes from epigraphic
and archaeological records of Zaachila (Teotzapotlan), I focus on this kingdom.
Author's personal copy
1 3
Journal of Archaeological Research
Zaachila, which had long been an important community in the valley, was most
powerful early in the Postclassic period. Its rulers governed with advice from
a council (Marcus 2006, p. 215). At least seven other towns (including Mitla and
Macuilxochitl) paid tribute to the Zaachila rulers (Spores 1965, p. 966; Whitecotton
1992, p. 61). But a Mixtec ruler conquered Zaachila about 1490, and its Zapotec
rulers fled to the Isthmus of Tehuantepec. There King Cocijoeza established control
of the town of Tehuantepec, which Zapotec rulers governed until Spanish conquest
(Marcus 1983a, pp. 302–303; Oudijk 1998, pp. 16, 19, 26–27; Oudijk and Jansen
2000, pp. 288–300). Marcus and Flannery (1996, p. 13) estimate the entire num-
ber of Zapotec speakers—including those in the Oaxaca valleys and the Isthmus of
Tehuantepec—as close to 500,000.
It is difficult to assess the position of royal women, for little has been written
about them. There is no example of a woman actually ruling a city-state, although
there are examples from the Mixtec (Hamann 1997, p. 171; Joyce 2010, p. 47). Per-
haps Postclassic Zapotec royal women had less stature than Mixtec women, for the
former ground their own maize while the latter did not. Although Mixtec women in
the codices and ethnohistoric references “actively engaged in the political sphere
as warriors, queens, and oracles” (Hamann 1997, p. 171), there is no evidence that
Postclassic Zapotec royal women did the same. Yet the Zapotec pictographic lien-
zos (“maps or diagrams of a community and its communal lands” [Boone 1996, p.
193]) and the Nuttall and Bodley codices show men and their main wives ascending
together to rulership. Ruling houses traced their descent back to a “founding couple”
(Joyce 2010, p. 44). The implication is that a couple succeeded to the throne, but
there is no documentation that they actually shared governance or power (Hamann
1997, pp. 171–172; Pohl 2003a, p. 177).
Postclassic Zapotec rulers were polygynous but not incestuous—although some
Mixtec did marry their close kin (Whitecotton 1977, pp. 144–145). The Span-
ish chronicler Troncoso (1579–1581, pp. 198, 218) wrote that Zapotec rulers took
15–20 wives, far fewer than imperial Inca rulers. They had a main wife, usually the
woman with the highest status (most royal blood) before marriage. The women they
married were from the same noble class as they, but they came from different poli-
ties. Postclassic Zapotec rulers married strategically, using their unions to consoli-
date or extend their domains. This did not preclude them from warring against each
other (Blomster 2008, pp. 24–25; Hamann 1997, pp. 168–172; Oudijk 1998, pp.
15–16, 2008, p. 106; Spores and Flannery 1983, p. 340).
Women who married Zapotec kings were prized for several reasons. Those who
were of higher birth than their spouses could raise their husbands’ and children’s
stature (Blomster 2008, p. 25; Oudijk 1998, p. 21). Postclassic Zapotec rulers traced
their ancestry through their mother and their father, which had implications for royal
succession. While the ideal was for the eldest son of the rulers’ main wife to suc-
ceed his father, primogeniture was not always followed (Paddock 1983, p. 313; Pohl
2003b, p. 247; Whitecotton 1977, pp. 153–155). The dynasty founder, Zaachila I,
was succeeded by his grandson, probably because the founder’s own son died before
he did. The penultimate ruler was succeeded by his uncle (his father’s half-brother)
because the former had no sons (Oudijk 2008, p. 101). So the rank of a ruler’s
mother figured in their right to rule.
Author's personal copy
Journal of Archaeological Research
1 3
Main wives cemented alliances, although these dissipated upon the death
of either spouse (Joyce 2010, pp. 45–46). The ruler of the Zapotec town Cui-
lapan married Lady 6 Monkey, from a Mixtec town, thus binding the two rul-
ers together. Another Mixtec princess, Lady 4 Rabbit, married the scion of the
Zaachila founder (Oudijk 1998, pp. 19–21). A few years before Spanish conquest,
Zapotec and Mixtec rulers successfully banded together to prevent an Aztec take-
over of the trade network. The alliance quickly fell apart when the Postclassic
Zapotec ruler Cocijoeza (1487–1529) married the daughter of the Aztec ruler
Ahuitzotl. As a result, Cocijoeza started paying tribute to Ahuitzotl, and Ahu-
itzotl gained a foothold in the valley (Marcus 1983a, pp. 301–303; Marcus and
Flannery 1996, p. 11; Spores 1965, p. 966; Whitecotton 1992, pp. 62, 64).
Royal women (main wives and princesses) were probably not considered
divine, although two princesses in the Codex Nuttall are depicted wearing the
costume of goddesses (Hamann 1997, p. 161). Perhaps their dress symbolizes
their affiliation with or impersonation of the goddesses, but it does not neces-
sarily signify their divinity. Zaachila rulers were considered descendants of the
gods. During their lifetime, they were believed to be conduits to the gods, as were
priests and other leaders (Joyce 2010, p. 48). That is, unlike Late Shang China
rulers, they did not monopolize access to the gods. After death, they became dei-
fied ancestors (Marcus 1983b, p. 348). Whether or not royal women shared the
rulers’ transformation into sacred ancestors is not clear. Still, they were valued as
producers or supervisors of royal textile manufacture, for textiles were used for
exchange and alliance building (Pohl 2003a, p. 177). Lady 4 Rabbit is credited
with stimulating textile production, probably by bringing to court Mixtec weavers
(Hamann 1997, pp. 167–170).
Did Postclassic Zapotec main wives have agency? Clearly, their ruling kin used
them to cement alliances or even reinforce patron–client relations (Whitecotton
1977, p. 145). But there is information that at least one princess refused to marry
and was not forced to (Marcus 1983a, p. 303). Because women did not rule in their
own right, they did not make policy. However, Zapotec main wives and princes
“aided” the king in governance, which implies that they influenced policy (Marcus
2006, p. 215). At the very least, they influenced rulers’ behavior. Hamann (1997,
p. 171) suggests that royal women used their textiles to advance their “own politi-
cal goals, creating the literal fabric of alliances.” But there is no mention of royal
women prophesying for the kingdom, even though they divined for their own homes
as did commoner women (Marcus 1998, p. 12). Rather, high priests had the respon-
sibility of divining to advise the ruler on political decisions (Marcus 1983a, p. 303,
1983b, p. 350; Whitecotton 1977, p. 147).
Main wives also must have acted as patrons (thereby influencing behavior), for
they had the means to do so. They owned property, which sometimes meant whole
towns, for Lady 5 Flower “inherited Tlaxiaco and married Lord 5 Rain . . . . Their
daughter, 8 Deer ‘Quetzal-Cobweb’ then inherited Tlaxiaco and married Lord
10 Alligator… of Achiutla” (Oudijk 1998, p. 23). Royal women also might have
gained wealth and power from their production of textiles. Royal women did serve
as intermediaries or advocates for their kin, husbands, or both. Perhaps they spied
for their kin or even their spouses. The Aztec wife of King Cocijoeza remained loyal
Author's personal copy
1 3
Journal of Archaeological Research
to her husband even when her father planned to assassinate him (Marcus 1983a, p.
302–303; Paddock 1983, p. 313).
In sum, although there is little information about Zaachila or Zapotec royal
women, we can surmise that they influenced policy and the behavior of everyone
from their spouses and kin to those they patronized. In addition, they exercised some
power by acting as intermediaries and choosing the side (husband or kin) they would
Royal women—main wives, regents, and rulers—exercised considerable political
agency in the eight societies presented. Of course, this statement depends on one’s
definition of political agency. I adapt the analyses of agency proposed by Ahearn
(2001), Dornan (2002), and Robb (2010), defining agency as the ability—the oppor-
tunity and capability—to act, whether that action is for oneself, one’s kin or affines,
or anyone else. One’s ability to take action takes place in a social structure and in
relation to other statuses or people. Political agency is action taken concerning the
politics and governance of a polity, a premodern state in this study.
Using the four categories of questions from the Introduction as a framework, I
summarize the trends present in the eight societies. My original goal was to use sta-
tistical (regression analysis) and network analysis to point out correlation or causal-
ity between societal rules or expected behaviors (roles) and royal women’s political
agency. However, of the eight types of political agency, half were found in all the
societies and one-quarter were found in four or more. As a consequence, the statisti-
cal and network analysis tests I tried turned up little of significance. Therefore, the
causal or correlational language that I use is based not on statistics but on chronol-
ogy as found in the literature or on logical suppositions on my part.
The agency of queen rulers and main wives took eight forms (Table1). The former,
be they women who inherited rule, co-ruled with their husbands, or acted as regents,
had the authority to make policy decisions. Such policies included when to plant and
harvest, when to start or stop wars, when and from whom to collect taxes or demand
tribute. The Late Classic Maya and Protohistoric Hawaiian women who ruled in
their own right had these powers. Women who co-ruled in Protohistoric Hawai’i and
the Inca Empire were limited to making policy about the female half of the popula-
tion. They did not determine when to go to war, for example. Main wives—actu-
ally widows of rulers—acting as regents for their sons had the same authority as
male rulers to make policy. Such regents were found in half the states: Protohistoric
Hawai’i, Old Kingdom Egypt, Aztec, and Late Classic Maya.
In all eight societies, main wives enjoyed four types of agency. They influenced
policy by representing the deities’ wishes and by linking their husbands with their
kin. The opportunity to influence policy came from their unique access to the ruler
Author's personal copy
Journal of Archaeological Research
1 3
and his courtiers in rituals, the court, and the bedroom. Main wives affected their
husbands’ behavior when they interceded between their spouses and others. Some-
times their intercession took the form of advocacy for either party. At other times,
they passed information from one party to the next, usually information about their
husbands’ plans to their fathers. There are actual accounts of Mari Kingdom main
wives (Zimri-Lim’s married daughters) spying for their father, and inferential infor-
mation from Late Shang China and the Postclassic Zapotec kingdom of Zaachila
suggests that they did so as well. Anecdotes about Aztec and Late Classic Maya
married princesses suggest that they, too, passed information about their husbands
to their kin. It would be surprising if royal women in the other case studies did not
engage in this activity. Some royal women conspired against the ruler. Scholars of
Old Kingdom Egypt report that queens sometimes conspired to overthrow their hus-
bands’ rule, thus helping their child or a usurper take over. While such behavior
probably occurred in other societies, there is as yet no firm data to corroborate such
Main wives controlled the behavior of those beneath them in rank (main wives
usually ranked second, just below the ruler), especially when they acted as adminis-
trators. Such opportunities arose when main wives supervised the palace (or at least
the women’s quarters of the palace), craft workshops in the palace, and their own
estates. (An exception to the latter is Aztec women, who owned estates, but their
male kin or overseers administered them.) This pattern was found in the Mari King-
dom, Aztec Empire, Late Classic Maya, Postclassic Zapotec, and probably Proto-
historic Hawai’i. Late Shang China main wives in the early years of the dynasty
directed armies, but they seem to have been unique.
Main wives altered underlings’ behavior when they turned wealth into gifts,
thereby forming patron–client relations. The women’s wealth came from the dow-
ries they brought to their marriage as well as the wealth they accumulated in their
new homes. Women in five states received gifts and tribute: the Mari Kingdom, Late
Shang China, Aztec Empire, Late Classic Maya, and Postclassic Zapotec. I infer that
Protohistoric Hawaiian and Inca Empire women also received gifts, if not tribute.
Royal women also gained wealth from the estates they controlled in every society
but the Late Classic Maya. While the latter may have controlled estates, too, evi-
dence is not yet definitive.
Like their male counterparts, women rulers and co-rulers acted on behalf of their
kin. But rulers of both genders were responsible for protecting or increasing ter-
ritorial boundaries. Furthermore, they were supposed to protect their subjects. To
that end, they sought to prevent their subjects’ capture, consulted the gods, and pro-
tected the harvests or livestock. Main wives in all eight states also took action on
behalf of their royal kin—their fathers, brothers, or uncles. They sought rulership for
their sons and good marriages for their daughters. Some main wives acted for their
husbands, pleading their case to the women’s kin, as did Zimri-Lim’s daughters.
Author's personal copy
1 3
Journal of Archaeological Research
Whether or not such actions were in their own best interest cannot be discerned from
The eight states exhibited greater variability in structural rules related to politics
(Table2) than royal women’s political agency (Table 1). Only one out of six struc-
tural rules was shared by all eight states, whereas four out of eight political agency
traits were shared by the states. The only rule that all the states followed was polyg-
yny. Among their many wives, rulers designated main wives—one to three women
of the highest birth whose major responsibility was to produce legitimate royal heirs
and marriage pawns who could be used to cement alliances. The need for legiti-
mate heirs gave main wives high status and special access to their husbands, prob-
ably correlating with their ability to influence the men’s policies and behavior. In
the societies where women ruled in their own right, there is as yet no information on
whether or not they practiced polyandry.
The second structural rule was that rulers should be men. Five of the states
adhered to this rule. The exceptions were the Late Classic Maya and Protohis-
toric Hawai’i, which accepted women as rulers in their own right. These women
had almost all the agency of kings, with a few exceptions. While the Inca and Pro-
tohistoric Hawaiians followed the rule that rulers should be men, their practice of
extreme separation of the genders throughout the society meant that rulers’ main
wives ruled over the women. Co-rulers’ agency was limited to the women’s sphere;
they did not wage war or address the war gods. Main wives sometimes substituted
for their husbands, ruling in their stead. In Old Kingdom Egypt and the Aztec
Empire, main wives acted as regents until their sons were ready to take charge; in
the Mari Kingdom, main wives ruled in their husbands’ absence. These women were
expected to perform all the roles of kings. They collected taxes and tribute, met with
foreign ambassadors, performed ceremonies and other actions to protect the people
from risk, etc. Their power and agency were limited by their husbands. Therefore, I
marked the societies allowing female regents and women ruling in their husbands’
absence as adhering to the structural rule that men should rule.
Some of the major characteristics of premodern states concern the rules for
selecting the next ruler. One was primogeniture, the rule that the eldest son should
inherit the throne. Only three states followed this rule: Protohistoric Hawaii, Old
Kingdom Egypt, and Postclassic Zapotec. The other states used a combination of
royal birth and heroic character as criteria for rulership. A brother, uncle, or any
son of a main wife could succeed a deceased king. These rules of succession were
broken in several states, for usurpers also gained the throne. When the right to rule
depended on being the son of a ruler and one of his main wives, these women had
to be of high rank. Their rank, in turn, boosted their opportunity to take political
Four states (Protohistoric Hawai’i, Aztec, Late Classic Maya, and Postclassic
Zapotec) used bilateral descent to limit the pool of candidates for ruler, whereas oth-
ers used the rule of patriliny. Although the other four societies (Late Shang China,
Author's personal copy
Journal of Archaeological Research
1 3
Old Kingdom Egypt, Inca Empire, and the Mari Kingdom) used patrilineal descent
as the first criterion for the right to rule, the higher the mothers’ lineage, the greater
the chance that their children would become ruler. States whose rulers practiced
some form of close-kin marriage (with their sister, half-sister, niece, or some other
close kin) might follow the patrilineal (Old Kingdom Egypt and Inca) or bilateral
(Aztec and Protohistoric Hawai’i) rule of succession. Rules of descent and succes-
sion determined main wives’ position at court, their opportunity and authority to
influence both policy and behavior, and their ability to gain wealth, which, in turn,
gave them the means to patronize men and in lower stations.
Another possible influence on the agency of main wives was rulers’ divinity,
which varied by society. Old Kingdom Egypt, Protohistoric Hawai’i, and the Inca
Empire considered rulers to be the sons of gods in life, or at least once they attained
the throne. Late Classic Maya believed that the gods entered rulers’ bodies during
certain rituals; otherwise living rulers were not divine. After death, Late Shang
China and the Late Classic Maya believed that rulers became divine ancestors. Post-
classic Zapotec and Aztec considered their rulers to be the descendants of gods or
super-humans (founders of the dynasty) but not gods per se. And the Mari King-
dom did not regard rulers as divine but simply as mortals who had the ear of the
main god. Royal women followed the same pattern as their husbands. In the socie-
ties where living rulers were divine, their main wives also were divine; in the socie-
ties where rulers were descendants of gods, so were their main wives, etc. Royal
women’s position in relation to the gods raised their stature, thus adding to their
political agency.
Rulers’ expected behaviors also varied by state. Only two out of six roles were
found among the eight states (Table2). Kings in all but Postclassic Zapotec (where
there is insufficient information to judge the kings’ uniqueness) monopolized access
to the primary deity, acting as sole interpreter of the main god. In that way, rulers
controlled the message, invoking the deity to legitimize political decisions. Royal
women consulted other deities. As appeasers and oracles of the gods, main wives
had the power to influence the policy decisions and behaviors of rulers as well as
others (e.g., courtiers or other women in the palace).
Rulers were responsible for protecting the people at large. They were expected
to reduce the risk that the populace might face destruction. To that end, rulers were
responsible for defending people’s economic well-being (good harvest) and pro-
moting their health. In two cases (Protohistoric Hawai’i and Inca Empire), rulers
assigned their main wives the responsibility for assuring the well-being of the states’
female population. Rulers also were responsible for the people’s safety. Rulers in all
but Late Shang China were expected to be warriors who would lead their troops in
battle, at least when they were young. When rulers defended the polity, their people
were not made captives and their arable land (their livelihood) was not destroyed.
When the ruler conquered new territories, the people gained access to additional
land and resources (including slaves). Women’s agency clearly came into play, for
they spied for their fathers while participating in their husbands’ courts and thus
aided their father’s war plans. A few main wives in Late Shang China led troops in
battle, and there is some indication that Late Classic Maya women rulers also were
involved in warfare.
Author's personal copy
1 3
Journal of Archaeological Research
Closely connected to the protector role is the duty of rulers to sponsor and per-
form public rituals. By engaging in public spectacles, they appeased or pleased the
gods who, in turn, would determine the fate of the people. By sponsoring feasts for
the populace—redistributing goods/food to all, rulers made themselves patrons to
entire populations. Royal women attended and assisted their husbands and fathers in
these public rituals and thus were privy to political information that they could use
when spying, mediating, or advocating.
The 12 roles expected of queen rulers and main wives also delimited their politi-
cal agency (Table3). Whereas four sample states (Protohistoric Hawai’i, Old King-
dom Egypt, Aztec, and Late Classic Maya) had some women rulers or regents, four
did not. Royal women in five states co-ruled, sharing rule of the entire population
with their husbands or ruling just the female population. Of the main wives, three
states (Protohistoric Hawai’i, Old Kingdom Egypt, and Inca) considered them divine
in their lifetime. However, six states believed that main wives could communicate
directly with the gods. These states were the three that considered women divine as
well as the Mari Kingdom, Late Shang China, and the Late Classic Maya. In only
one state, Late Shang China, did women engage in war. There is sufficient informa-
tion on three states (Aztec, Late Classic Maya, and Postclassic Zapotec) that have
royal women actually manufacturing luxury or prestige goods. It is highly likely that
royal women in other states also made such goods, but there is no information on
that as yet. Main wives in seven states controlled estates, either owning or supervis-
ing them but definitely controlling the distribution of the estates’ production. Infor-
mation on the Late Classic Maya is not known at this time.
The rest of main wives’ roles are found in all eight states: being pawns in their
kins’ alliance schemes during their marriage, begetting royal children—possible
heirs to the throne, administering or supervising such people as the occupants of
the women’s quarters or their own estates, increasing their wealth through gifts and
tribute but especially controlling arable land. All of these roles gave main wives the
capability and opportunity to exert political agency on others. For example, their
ability to communicate with the deities gave them the ability to influence policy;
their accumulation of wealth gave them the capability to form patron–client relation-
ships and influence behavior; and their presence in the court and access to the king
gave them the opportunity to influence policy and behavior, sometimes even spying.
Although I found similar patterns in royal women’s agency as well as the contexts
in which they acted, I do not belittle the variability found among the different states
in the sample. They ranged in population size from 40,000 (the Mari Kingdom,
which was an autonomous kingdom among several during the Old Babylonian era)
to 13–25 million (the Aztec Empire). Political organization also varied, for some
state-level regions encompassed several autonomous kingdoms while others were
dominated by paramount rulers who controlled vassal kingdoms (Aztec Empire,
Old Kingdom Egypt, Protohistoric Hawai’i, and Inca Empire). The Late Classic
Maya Lowland region contained both. And as discussed above, the position of royal
Author's personal copy
Journal of Archaeological Research
1 3
women differed, for some states considered them divine in life or after death (Late
Shang China, Old Kingdom Egypt, Protohistoric Hawai’i, Inca Empire, and Late
Classic Maya); some (Protohistoric Hawai’i, and the Inca Empire) separated them
from men; and still others allowed them to rule either as regents (Aztec Empire, Old
Kingdom Egypt) or in their own right (Protohistoric Hawai’i, Late Classic Maya).
What do we learn about royal women in premodern states, and what do we learn
about agency from this research? First and foremost, the stories of royal women
show that they exhibited considerable political agency in at least five ways. Queen
rulers made policy. Main wives sometimes made policy, but mostly they influenced
policy, influenced the behavior of those above and below them in rank, acted as go-
betweens (usually for their kinsmen or spouses), and passed information to their
kinsmen, their affines, or both. Royal women of either status acted as patrons to vari-
ous clients.
When women were rulers in their own right or when they were regents, they per-
formed almost all the same roles and exhibited almost all the same agency as male
rulers. However, women rulers of the Late Classic Maya did not practice polyg-
amy, usurp thrones for themselves, or directly participate in interpolity fighting the
way kings did. Women rulers in Protohistoric Hawai’i did not have equal access to
the war temple as did men rulers, and so their ability to make war was somewhat
The roles and agency of main wives sometimes overlapped with those of their
ruler husbands. When they acted as ruler in the absence of their husbands, they,
too, made policy. They met with visiting dignitaries, supervised tax collection, etc.
However, the roles and agency of main wives were not the same as for male rul-
ers. The tablets that Queen Shibtu of the Mari Kingdom sent to her husband while
he was fighting battles or touring his kingdom show that she checked her decisions
with him. In Protohistoric Hawai’i and the Inca Empire, main (principal) wives
ruled over the women while their husbands ruled the polity. Such queens lacked the
authority to make interpolity policy decisions or decisions affecting the entire popu-
lation. And royal wives who used their great wealth to patronize clients did not do
so on the same scale as their husbands. R. Joyce (2000, pp. 88–89), the author of
gender complementarity, would agree that it does not necessarily mean equality of
position or power.
Some of the roles and agency of main wives differed from those of their hus-
bands. Whereas kings made policy, main wives influenced policy. In the Mari King-
dom, Late Classic Maya, and Late Shang China, main wives who acted as spokes-
persons for the gods had the capability to influence policy. The women’s access to
rulers and their court gave them the opportunity to influence policy. As go-betweens
or advocates for their husbands and kin, they influenced the behavior of both parties.
They also were the conduits of information between the two, sometimes acting as
spies. Even though main wives lacked the power and resources of their husbands,
their actions were efficacious. Therefore, analysts of royal behavior—from warfare
Author's personal copy
1 3
Journal of Archaeological Research
to network analysis and alliance building—should not dismiss the women’s part in
royal history or societal change. Women archaeologists have argued this point for
decades (Gero 1990; Marcus 2001; Pyburn 2008), but little progress has been made
in the general literature. I understand that archaeologists will have to combine his-
toric and epigraphic data with archaeological research to find information on royal
women, but the extra effort would benefit our understanding of agency, not to men-
tion the political behavior of premodern state actors.
Main wives’ position as pawns in their liminal stage of life, i.e., the marriage
ceremony that marks their transition from princess to main wife, does not seem to
limit their agency as adults. As they were used by rulers jockeying for position in the
regional ranking game, we cannot understand how politics and “international rela-
tions” actually operated in premodern states if we do not include royal women in our
Whether women were rulers in their own right or main wives, they shared cer-
tain agency behaviors across the eight societies. Although we cannot claim that their
behaviors were universal among premodern states because of the limited sample, the
patterns described here might be worth studying elsewhere. They suggest that royal
women’s political agency may be discerned in the archaeological, epigraphic, and
historical records just as their economic agency has been. Furthermore, they sug-
gest that once societies became states, royal women in most of the sample shared
roles and agency. I found this to be true no matter whether the polities were small
states operating in a region of similar small states (Mari Kingdom, Protohistoric
Hawai’i), empires (Late Shang China, Old Kingdom Egypt, Aztec, or Inca), or states
of varying complexity in a single region (Late Classic Maya and Postclassic Zapo-
tec). In sum, we cannot yet fully account for the fact that so many societies scat-
tered across the globe developed the same structural and behavior patterns, but we
must acknowledge the remarkable similarity among these eight premodern states.
The kinds of agency described here are so prevalent (Table1) that we cannot discern
which variables are independent and which are dependent.
As for the concept of agency, I propose that this comparison of royal women’s
political agency helps clarify certain trends in the application of agency theory to
archaeology. The data on main wives advance the notion that people do not act
solely from the motivation of self-interest. We accept that people may act not in
“rational” (self-interested) ways but in ways that support others—their children, hus-
bands, kin group, or community (Dobres and Robb 2000, p. 4; Shennan 2007, pp.
19−21). The Aztec princess who burned her husband’s town so that her father could
more easily conquer it is a clear example of acting for others (Diel 2007, p. 269).
Perhaps this analysis of motivation allows anthropologists to account for altruistic
behavior, which seems to operate within a community more than beyond that social
space (Bowles and Gintis 2011). The royal women discussed here certainly acted for
others. Whether or not they were coerced or self-motivated will probably never be
known from the archaeological record.
Archaeologists follow Giddens and Bourdieu in describing agency as embed-
ded in a particular social structure, that is, “‘the rules and resources’ which actors
draw upon” to take action (Gardner 2007, p. 7). A good way to begin the analysis
of agential context, then, is to outline the societal rules that circumscribe people’s
Author's personal copy
Journal of Archaeological Research
1 3
actions. For main wives in premodern states, these rules include descent and suc-
cession to the throne. Structural context also is encoded in the roles expected of
the people with whom the agents interact, for agency is a dyadic, or intersubjec-
tive, act; it is action in relation to another person, object, or situation (Gardner 2007,
pp. 4–5, Robb 2010, p. 499). Finally, structural context includes the agents’ roles,
for expected behaviors circumscribe agents’ habitual behavior. These three contexts
provide agents with opportunity and at least the baseline for their capability to take
It is in the distinction between role and agency that we can truly understand
agency. Yet as a sociocultural anthropologist, I have frequently seen a confusion
between status, rank, role, and agency in archaeological writing. According to Lin-
ton (1936), the originator of the concept in anthropology, status refers to the position
one holds in society. Positions may be ranked, and individual people may occupy
more than one status at a time, or at least in their lifetime. Roles are the expected
behaviors associated with each status. Linton emphasizes that roles are combina-
tions of the right to behave in such a manner and the expectation that one will ful-
fill one’s duties. A ruler may be expected to fight wars or perform in public rituals.
Agency, then, refers to the actual behaviors that individuals do. To act, people have
to have the opportunity and capability to do so. Main wives had to have access to the
king and his court to spy for their fathers. They also had to have the ability to com-
municate with their fathers.
The study of women rulers and main wives highlights one more distinction in
literature, and that is between power and agency (Dornan 2002; Gardner 2007,
pp. 5−8; Robb 2010). “Power is the transformative capacity of an agent to achieve
an outcome which can either reproduce or change system and structure (Giddens
1979:88–94)” (A. Joyce 2000, p. 73). Power has been analyzed as “power of,
“power over,” and “power to.” Clearly, a certain amount of power (of and over) is
necessary to have “power to.” But is not “power to” another way of describing capa-
bility and opportunity to act—the major components of the definition of agency?
What is so interesting about the royal women is that they were able to build polit-
ical agency (“power to” and “power over”) even though they had been treated as
pawns in interregional marriages. Information on the eight societies suggests that
some roles expected of kings were not allowed women rulers/regents or co-rulers
of the female population. In Late Classic Maya, the Inca Empire, and Protohistoric
Hawai’i, these roles included engaging in polygamy, usurping the throne for them-
selves, and participation in war ceremonies. Main wives’ “power to” (the ability to
make or participate in policy decisions) was certainly limited, relegating them to
influencing rather than making policy in most situations. But main wives in most
of the societies had the power to influence behavior, i.e., influence policy decisions,
patronize clients, and advocate for or pass information to one party or another.
Acknowledgments This project was initiated under the John Templeton Foundation grant to the Santa Fe
Institute (“The Principles of Complexity: Revealing the Hidden Sources of Order among the Prodigies of
Nature and Culture,” Grant No. 15705). With this support, a new database on eight premodern states was
developed under the guidance of leading archaeologists Jeremy Sabloff, Henry Wright, Timothy Kohler,
and Charles Stanish. Researchers Robert Weiner, Kong Fai Cheong, and Jonah Nonomaque, as well as
citizen scientists Jeffrey Cohen, George J. Haddad, Jack M. Jackson, and Shelley Waxman helped develop
Author's personal copy
1 3
Journal of Archaeological Research
the original database. This database was then checked by Laurel Bestock, Gary Feinman, Michael Galaty,
Abigail Holeman, Peter Peregrine, Patrick Kirch, Gideon Shelach-Lavi, Adam D. Smith, Michael E.
Smith, Charles Stanish, Stephen Tinney, and John Ware. Following this process, I tripled literature search
to find answers to questions about royal women’s roles and behaviors in their societies. Therefore, the
errors are my own. When constructing the tables and attempting to place the data in statistical format, I
consulted several colleagues at the Santa Fe Institute: Aaron Clauset, Mimi Kohl, Cris Moore, and Van
Savage. It was Michael Lachmann who finally convinced me that statistical analysis of this particular
database was fruitless and Laura Fortunato who recommended the format used in the tables presented
here. My special thanks go to Jeremy Sabloff. I could not have written this without his support, his love,
and his library.
References Cited
Ahearn, L. (2001). Language and agency. Annual Review of Anthropology 30: 109–137.
Ardren, T. (2002). Women and gender in the ancient Maya world. In Ardren, T. (ed.), Ancient Maya
Women, Altamira, Walnut Creek, CA, pp. 1–11.
Ardren, T. (2008). Studies of gender in the Prehispanic Americas. Journal of Archaeological Research
16: 1–35.
Baines, J., and Yoffee, N. (1998). Order, legitimacy, and wealth in ancient Egypt and Mesopotamia. In
Feinman, G., and Marcus, J. (eds.), Archaic States, School of American Research Press, Santa Fe,
NM, pp. 199–260.
Banning, E. (2010). Houses, households, and changing society in the Late Neolithic and Chalcolithic of
the Southern Levant. Paléorient 36: 49–87.
Barnes, G. (1999). The Rise of Civilization in East Asia: The Archaeology of China, Korea and Japan,
Thames and Hudson, London.
Barrett, J. (2000). A thesis on agency. In Dobres, M., and Robb, J. (eds.), Agency in Archaeology, Rout-
ledge, London, pp. 61–68.
Bauer, B. S. (1998). The Sacred Landscape of the Inca, University of Texas Press, Austin.
Bell, E. (2002). Engendering a dynasty: A royal woman in the Margarita Tomb. In Ardren, T. (ed.),
Ancient Maya Women, Altamira, Walnut Creek, CA, pp. 89–104.
Bell, J. (1992). On capturing agency in theories about prehistory. In Gardin, J., and Peebles, C. (eds.),
Representations in Archaeology, Indiana University Press, Bloomington, pp. 30–55.
Berdan, F. (1982). The Aztecs of Central Mexico: An Imperial Society, Holt, Rinehart and Winston, New
Yor k.
Berdan, F. (2014). Aztec Archaeology and Ethnohistory, Cambridge University Press, Cambridge.
Blanton, R., Feinman, G., Kowalewski, S., and Peregrine, P. (1996). A dual–processual theory for the
evolution of Mesoamerican civilization. Current Anthropology 37: 1–14.
Blanton, R., Kowalewski, S., Feinman, G., and Finsten, L. (1993). Ancient Mesoamerica: A Comparison
of Change in Three Regions, revised edition, Cambridge University Press, Cambridge.
Blomster, J. (2008). Changing cloud formations: The sociopolitics of Oaxaca in Late Classic/Postclassic
Mesoamerica. In Blomster, J. (ed.), After Monte Alban: Transformation and Negotiation in Oax-
aca, Mexico, University Press of Colorado, Boulder, pp. 3–46.
Boone, E. (1996). Manuscript painting in service of imperial ideology. In Berdan, F., and Smith, M.
(eds.), Aztec Imperial Strategies, Dumbarton Oaks, Washington, DC, pp. 181–206.
Borgatti, S., Everett, M., and Freeman, L. (2002). Ucinet 6 for Windows: Software for Social Network
Analysis, Analytic Technologies, Harvard University, Cambridge, MA.
Bourdieu, P. (1977). Outline of a Theory of Practice, Nice, R. (tr.), Cambridge University Press,
Bourdieu, P. (1985). The social space and genesis of groups. Theory and Society 14: 723–744.
Bourdieu, P. (1990). The Logic of Practice, Stanford University Press, Stanford.
Bowles, S., and Gintis, H. (2011). A Cooperative Species: Human Reciprocity and Its Evolution, Prince-
ton University Press, Princeton, NJ.
Bowser, B., and Patton, J. (2010). Women’s leadership: Political alliance, economic resources, and repro-
ductive success in the Ecuadorian Amazon. In Vaughn, K., Eerkens, J., and Kantner, J. (eds.), The
Author's personal copy
Journal of Archaeological Research
1 3
Evolution of Leadership: Transition in Decision Making from Small-scale to Middle-range Socie-
ties, School for Advanced Research Press, Santa Fe, NM, pp. 51–71.
Brumfiel, E. (1991). Weaving and cooking: Women’s production in Aztec Mexico. In Gero, J., and Con-
key, M. (eds.), Engendering Archaeology: Women and Prehistory, Blackwell, Oxford, pp. 224–251.
Brumfiel, E. (1992). Distinguished lecture in archaeology: Breaking and entering the ecosystem—gender,
class, and faction steal the show. American Anthropologist 94: 551–567.
Brumfiel, E. (2000). On the archaeology of choice. In Dobres, M., and Robb, J. (eds.), Agency in Archae-
ology, Routledge, London, pp. 249–255.
Campbell, R., Li, Z., He, Y., and Jing, Y. (2011). Consumption, exchange and production at the great set-
tlement Shang: Bone-working at Tiesanlu, Anyang. Antiquity 85: 1279–1297.
Chase, A., Chase, D., and Haviland, D. (2002). Maya social organization from a “big site” perspective:
Classic period Caracol, Belize and Tikal, Guatemala. In Blos, V., Cobos, R., and Robertson, M.
(eds.), La organización social entre los mayas prehispánicos, coloniales y modernos, Instituto
Nacional de Antropología e Historia, Mexico, pp. 253–276.
Chase, A., Chase, D., Zorn, E., and Teeter, W. (2008). Textiles and the Maya archaeological record.
Ancient Mesoamerica 19: 127–142.
Chase, D., and A. Chase (2017). Caracol, Belize, and changing perceptions of ancient Maya Society.
Journal of Archaeological Research 25: 185–249.
Childs-Johnson, E. (2007). Fu Zi, the Shang woman warrior. In Hong, L., and Stefanowska, A. (eds.),
Biographical Dictionary of Chinese Women: Antiquity through Sui 1600 BCE618 CE, M. E.
Sharpe, Armonk, pp. 19–25.
Cieza de Leon, P. (1883). The Second Part of the Chronicles of Peru, 1653, Markham, C. B. (tr.), Hakluyt
Society, London.
Claassen, C. (1997). Changing venue: Women’s lives in prehistoric North America. In Claassen, C., and
Joyce, R. (eds.), Women in Prehistory: North America and Mesoamerica, University of Pennsylva-
nia Press, Philadelphia, pp. 65–87.
Clendinnen, I. (1991). Aztecs, Cambridge University Press, Cambridge.
Cobo, B. (1893). History of the New World, Imp de Rasco, Sevilla.
Cobo, B. (1990). Inca Religion and Customs, Hamilton, R. (tr.), University of Texas Press, Austin.
Collier, J. (1974). Women in politics. In Rosaldo, M., and Lamphere, L. (eds.), Woman, Culture, and
Society, Stanford University Press, Stanford, CA, pp. 89–96.
Conkey, M., and Gero, J. (1997). Programme to practice: Gender and feminism in archaeology. Annual
Review of Anthropology 26: 411–437.
D’Altroy, T. (2002). The Incas, Blackwell, Oxford.
David, R. (1998). The Ancient Egyptians: Beliefs and Practices, Sussex Academic Press, Brighton.
Diel, L. (2005). Women and political power: The inclusion and exclusion of noblewomen in Aztec picto-
rial histories. Res: Anthropology and Aesthetics 47: 82–106.
Diel, L. (2007). Till death do us part: Unconventional marriages as Aztec political strategy. Ancient Mes-
oamerica 18: 259–272.
Dobres, M., and Robb, J. (2000). Agency in archaeology: Paradigm or platitude? In Dobres, M., and
Robb, J. (eds.), Agency in Archaeology, Routledge, London, pp. 3–17.
Dodson, A., and Hilton, D. (2004). The Complete Royal Families of Ancient Egypt, Thames and Hudson,
Dornan, J. (2002). Agency and archaeology: Past, present, and future directions. Journal of Archaeologi-
cal Method and Theory 9: 303–329.
Estrada-Belli, F., and Tokovinine, A. (2016). A king’s apotheosis: Iconography, text, and politics from a
Classic Maya temple at Holmul. Latin American Antiquity 27: 149–168.
Feinman, G., and Nicholas, L. (2016). After Monte Albán in the Central Valleys of Oaxaca: A reassess-
ment. In Faulseit, R. (ed.), Archaeological Perspectives on Resilience, Revitalization, and Transfor-
mation in Complex Societies, Occasional Paper No. 42, Center for Archaeological Investigations,
Southern Illinois University, Carbondale, pp. 43–69.
Feng, L. (2013). Early China: A Social and Cultural History, Cambridge University Press, Cambridge.
Fiskesjö, M. (2001). Rising from blood-stained fields: Royal hunting and state formation in Shang China.
Bulletin of the Museum of Far Eastern Antiquities 73: 48–191.
Flannery, K. V. (1999). Process and agency in early state formation. Cambridge Archaeological Journal
9: 3–21.
Fowler, C. (2004). The Archaeology of Personhood. Routledge, London.
Author's personal copy
1 3
Journal of Archaeological Research
Fowler, W. (2006). Cacao production, tribute, and wealth in sixteenth-century Izalcos, El Salvador. In
McNeil, C. (ed.), Chocolate in Mesoamerica: A Cultural History of Cacao, University Press of
Florida, Gainesville, pp. 307–321.
Fox, J., and Justeson, J. (1986). Classic Maya dynastic alliance and succession. In Spores, R. (ed.), Hand-
book of Middle American Indians: Ethnohistory, Supplement to Vol. 4, University of Texas Press,
Austin, pp. 7–33.
Frankfort, H. (1978). Kingship and the Gods: A Study of Ancient Near Eastern Religion as the Integra-
tion of Society and Nature, Oriental Institute, Chicago.
Freidel, D., and Schele, L. (1997). Maya royal women: A lesson in Precolumbian history. In Brettell, C.,
and Sargent, C. (eds.), Gender in Cross-Cultural Perspective, Pearson/Prentice Hall, Upper Saddle
River, NJ, pp. 74–78.
Gardner, A. (2007). Introduction: Social agency, power, and being human. In Gardner, A. (ed.), Agency
Uncovered: Archaeological Perspectives on Social Agency, Power, and Being Human, Left Coast
Press, Walnut Creek, CA, pp. 1–15.
Gero, J. (1990). Archeology: Women in prehistory. American Anthropologist 94: 1033.
Gero, J. (2000). Troubled travels in agency and feminism. In Dobres, M., and Robb, J. (eds.), Agency in
Archaeology, Routledge, London, pp. 34–39.
Giddens, A. (1979). Central Problems in Social Theory, University of California Press, Berkeley.
Giddens, A. (1984). The Constitution of Society, University of California Press, Berkeley.
Gillespie, S. (1989). The Aztec Kings: The Construction of Rulership in Mexica History, University of
Arizona Press, Tucson.
Gillespie, S. (2001). Personhood, agency, and mortuary ritual: A case study from the ancient Maya. Jour-
nal of Anthropological Archaeology 20: 73–112.
Gillespie, S., and Joyce, R. (1997). Gendered goods: The symbolism of Maya hierarchical exchange rela-
tions. In Claassen, C., and Joyce, R. (eds.), Women in Prehistory: North America and Mesoamer-
ica, University of Pennsylvania Press, Philadelphia, pp. 189–207.
Guenter, S. (2014). The Queen of Cobá: A reanalysis of the Macanxoc stelae. In Stanton, T. (ed.), The
Archaeology of Yucatan: New Directions and Data, Archaeopress Archaeology, Bilingual edition,
Oxford, pp. 395–421.
Guenter, S., and Freidel, D. (2009). Warriors and rulers: Royal women of the Classic Maya. In Bret-
tell, C., and Sargent, C. (eds.), Gender in Cross-Cultural Perspective, Pearson/Prentice Hall, Upper
Saddle River, NJ, pp.74–80.
Hamann, B. (1997). Weaving and the iconography of prestige: The royal gender symbolism of Lord 5
Flower’s/Lady 4 Rabbit’s family. In Claassen, C., and Joyce, R. (eds.), Women in Prehistory: North
America and Mesoamerica, University of Pennsylvania Press, Philadelphia, pp. 153–172.
Harrison-Buck, E. (2017). The coin of her realm: Cacao as gendered goods among the pre-hispanic
and colonial Maya. In Matthews, J., and Guderjan, T. (eds.), The Value of Things: Commodities
in the Maya Region from Prehistoric to Contemporary, University of Arizona Press, Tucson, pp.
Harrison-Buck, E., and Hendon, J. (2018). An introduction to relational personhood and other-than-
human agency in archaeology. In Harrison-Buck, E. and Hendon, J. (eds.), Relational Identities
and Other-Than-Human Agency in Archaeology, University Press of Colorado, Louisville, pp.
Hassig, R. (2016). Polygamy and the Rise and Demise of the Aztec Empire, University of New Mexico
Press, Albuquerque.
Hommon, R. (2013). The Ancient Hawaiian State: Origins of a Political Society, Oxford University
Press, Oxford.
Houston, S., and Inomata, T. (2009). The Classic Maya, Cambridge University Press, Cambridge.
Houston, S., and Stuart, D. (1996). Of gods, glyphs, and kings: Divinity and rulership among the Classic
Maya. Antiquity 70: 289–312.
Hubert, E. (2016). Figuring identity in everyday life. Journal of Anthropological Archaeology 44: 1–13.
Isbell, W. (1997). Mummies and Mortuary Monuments, University of Texas Press, Austin.
Jenkins, D. (2001). The Inka conical clan. Journal of Anthropological Research 57: 167–195.
Johnson, A., and Earle, T. (2000). The Evolution of Human Societies: From Foraging Group to Agrarian
State, 2nd ed., Stanford University Press, Stanford, CA.
Johnson, M. (2000). Self-made men and the staging of agency. In Dobres, M., and Robb, J. (eds.), Agency
in Archaeology, Routledge, London, pp. 213–231.
Author's personal copy
Journal of Archaeological Research
1 3
Josserand, J. (2002). Women in Classic Maya hieroglyphic texts. In Ardren, T. (ed.), Ancient Maya
Women, Altamira, Walnut Creek, CA, pp. 114–151.
Joyce, A. (2000). The founding of Monte Alban: Sacred propositions and social practices. In Dobres, M.,
and Robb, J. (eds.), Agency in Archaeology, Routledge, London, pp.71–91.
Joyce, A. (2010). Mixtecs, Zapotecs, and Chatinos: Ancient Peoples of Southern Mexico, Wiley-Black-
well, Chichester.
Joyce, A., and Winter, M. (1996). Ideology, power, and urban society in prehispanic Oaxaca. Current
Anthropology 37: 33–47.
Joyce, R. (1993). Women’s work: Images of production and reproduction in prehispanic southern Central
America. Current Anthropology 34: 255–274.
Joyce, R. (1996). The construction of gender in Classic Maya monuments. In Wright, R. (ed.), Gender
and Archaeology, University of Pennsylvania Press, Philadelphia, pp. 167–195.
Joyce, R. (2000). Gender and Power in Prehispanic Mesoamerica, University of Texas Press, Austin.
Joyce, R. (2008). Ancient Bodies, Ancient Lives: Sex, Gender, and Archaeology, Thames and Hudson,
Joyce, R., and Lopiparo, J. (2005). Postscript: Doing agency in archaeology. Journal of Archaeological
Method and Theory 12: 365–374.
Keightley, D. (1979). The Shang state as seen in the oracle-bone inscriptions. Early China 5: 25–34.
Keightley, D. (1983). The Late Shang state: When, where, and what? In Keightley, D. (ed.), The Origins
of Chinese Civilization, University of California Press, Berkeley.
Keightley, D. (1999a). The Shang: China’s first historical dynasty. In Loewe,M., and Shaughnessy, E.
(eds.), The Cambridge History of Ancient China: From the Origins of Civilization to 221 BC, Cam-
bridge University Press, Cambridge, pp. 232–291.
Keightley, D. (1999b). At the beginning: The status of women in Neolithic and Shang China. NAN NU 1:
Keightley, D. (2012). Working for His Majesty: Research Notes on Labor Mobilization in Late Shang
China (ca. 1200–1045 BC), as Seen in the Oracle-Bone Inscriptions, with Particular Attention to
Handicraft Industries, Agriculture, Warfare, Hunting, Construction, and the Shang’s Legacies,
Institute of East Asian Studies, University of California, Berkeley.
Kemp, B. (1992). Old Kingdom, Middle Kingdom and Second Intermediate Period, c. 2686–1552 BC. In
Trigger, B., Kemp, B., O’Connor, D., and Lloyd, A. (eds.), Ancient Egypt: A Social History, Cam-
bridge University Press, Cambridge, pp. 71–182.
Kirch, P. (2010). How Chiefs Became Kings: Divine Kingship and the Rise of Archaic States in Ancient
Hawai’i, University of California Press, Berkeley.
Kirch, P. (2012). A Shark Going Inland is My Chief: The Island Civilization of Ancient Hawai’i, Univer-
sity of California Press, Berkeley.
Kolata, A. (2013). Ancient Inca, Cambridge University Press, Cambridge.
Lafont, B. (2001). The women of the palace at Mari. In Bottéro, J. (ed.), Everyday Life in Ancient Meso-
potamia, Johns Hopkins University Press, Baltimore, pp. 127–140.
Li, M., and Zhu, Z. (1989). Zu Yi qian xing yu bu ci Jing fang (On ‘[King] Zu Yi moved to Geng’ and the
tribe of Jing in the oracle bones), Chen, J. (tr.). Journal of Zhengzhou University (Philosophy and
Social Sciences Edition) 6: 13–19.
Linduff, K. (2003). Many wives, one queen in Shang China. In Nelson, S. (ed.), Ancient Queens: Archae-
ological Explorations, Altamira Press, Walnut Creek, CA, pp. 59–75.
Linnekin, J. (1990). Sacred Queens and Women of Consequence: Rank, Gender, and Colonialism in the
Hawaiian Islands, University of Michigan Press, Ann Arbor.
Linton, R. (1936). The Study of Man: An Introduction, Appleton-Century-Crofts, New York.
Maisels, C. (2010). The Archaeology of Politics and Power: Where, When and Why the First States
Formed, Oxbow Books, Oxford, UK.
Marcus, J. (1974). The iconography of power among the Classic Maya.World Archaeology6: 83–94.
Marcus, J. (1976). Emblem and State in the Classic Maya Lowlands: An Epigraphic Approach to Territo-
rial Organization, Dumbarton Oaks, Washington, DC.
Marcus, J. (1983a). The reconstructed chronology of the later Zapotec rulers, AD 1415–1563. In Flan-
nery, K., and Marcus, J. (eds.), The Cloud People:Divergent Evolution of the Zapotec and Mixtec
Civilizations, Academic Press, New York, pp. 301–308.
Marcus, J. (1983b). Zapotec religion. In Flannery, K., and Marcus, J. (eds.), The Cloud People: Divergent
Evolution of the Zapotec and Mixtec Civilizations, Academic Press, New York, pp. 345–351.
Author's personal copy
1 3
Journal of Archaeological Research
Marcus, J. (1998). Women’s Ritual in Formative Oaxaca: Figurine-making, Divination, Death and the
Ancestors, Memoirs 33, Museum of Anthropology, University of Michigan, Ann Arbor.
Marcus, J. (2001). Breaking the glass ceiling: The strategies of royal women in ancient states. In Klein, C.
(ed.), Gender in Pre-Hispanic America, Dumbarton Oaks, Washington, DC, pp. 305–340.
Marcus, J. (2006). Identifying elites and their strategies. In Elson, C., and Covey, R. (eds.), Intermediate
Elites in Precolumbian States and Empires, University of Arizona Press, Tucson, pp. 212–246.
Marcus, J., and Flannery, K. (1983). The Postclassic balkanization of Oaxaca. In Flannery, K., and Mar-
cus, J. (eds.), The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations, Aca-
demic Press, New York, pp. 217–226.
Marcus, J., and Flannery, K. (1996). Zapotec Civilization: How Urban Society Evolved in Mexico’s Oax-
aca Valley, Thames and Hudson, London.
Martin, S., and Grube, N. (2008). Chronicle of the Maya Kings and Queens: Deciphering the Dynasties
of the Ancient Maya, 2nd ed., Thames and Hudson, London.
McAnany, P. (2013a). Artisans, Ikatz, and statecraft: Provisioning Classic Maya royal courts. In Hirth,
K., and Pillsbury, J. (eds.), Merchants, Markets, and Exchange in the Pre-Columbian World, Dum-
barton Oaks, Washington, DC, pp. 229–253.
McAnany, P. (2013b). Living with the Ancestors: Kinship and Kingship in Ancient Maya Society, rev. ed.,
Cambridge University Press, Cambridge.
Miller, M. (1993). On the eve of the collapse: Maya art of the eighth century. In Sabloff, J., and Hender-
son, J. (eds.), Lowland Maya Civilization in the Eighth Century AD, Dumbarton Oaks, Washing-
ton, DC, pp. 355–411.
Moore, H. (2000). Ethics and ontology: Why agents and agency matter. In Dobres, M., and Robb, J.
(eds.), Agency in Archaeology, Routledge, London, pp. 259–263.
Moreno García, J. (1997). Estates (Old Kingdom): Abstract. In Frood, E., and Willeke, W. (eds.), UCLA
Encyclopedia of Egyptology (UEE), http://digit al2.libra /zz001
ndr97 .
Munson, J., and Macri, M. (2009). Sociopolitical network interactions: A case study of the Classic Maya.
Journal of Anthropological Archaeology 28: 424–438.
Nelson, S. (2015). Shamans, Queens, and Figurines, Left Coast Press, Walnut Creek, CA.
Ortner, S. (1984). Theory in anthropology since the sixties. Society for Comparative Study of Society and
History 26: 126–166.
Oudijk, M. (1998). The genealogy of Zaachila: Four weddings and a dynastic struggle. In Jansen, M.,
Krofges, P., and Oudijk, M. (eds.), In the Shadow of Monte Albán: Politics and Historiography in
Postclassic Oaxaca, Mexico, Research School CNWS, School of Asian, African, and Amerindian
Studies, Leiden, pp. 13–35.
Oudijk, M. (2008). The Postclassic period in the Valley of Oaxaca: The archaeological and ethnohistori-
cal records. In Blomster, J. (ed.), After Monte Alban: Transformation and Negotiation in Oaxaca,
Mexico, University Press of Colorado, Boulder, pp. 95–118.
Oudijk, M., and Jansen, M. (2000). Changing history in the Lienzos de Guevea and Santo Domingo
Petapa. Ethnohistory 47: 281–331.
Paddock, J. (1983). Comments on the Lienzos of Huilotepec and Guevea. In Flannery, K., and Marcus, J.
(eds.), The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations, Academic
Press, New York, pp. 308–313.
Pauketat, T. (2001). Practice and history in archaeology. Anthropological Theory 1: 73–98.
Pennock, C. (2011). Bonds of Blood: Gender, Lifecycle and Sacrifice in Aztec Culture, Palgrave Macmil-
lan, London.
Pohl, J. (2003a). Ritual ideology and commerce in the southern Mexican Highlands. In Smith, M., and
Berdan, F. (eds.), The Postclassic Mesoamerican World, University of Utah Press, Salt Lake City,
pp. 172–177.
Pohl, J. (2003b). Royal marriage and confederacy building among the Eastern Nahuas, Mixtecs, and
Zapotecs. In Smith, M., and Berdan, F. (eds.), The Postclassic Mesoamerican World, University of
Utah Press, Salt Lake City, pp. 243–248.
Pohl, M., and Pohl, J. (1994). Cycles of conflict: Political factionalism in the Maya Lowlands. In Brum-
fiel, E., and Fox, J. (eds.), Factional Competition and Political Development in the New World,
Cambridge University Press, Cambridge, pp. 138–157.
Postgate, J. (1994). Early Mesopotamia: Society and Economy at the Dawn of History, Routledge,
Author's personal copy
Journal of Archaeological Research
1 3
Pyburn, K. A. (2008). Shaken, not stirred: The revolution in archaeology. In Brumfiel, E., and Robin, C.
(eds.), Gender, Households, and Society: Unraveling the Threads of the Past and Present, Archae-
ological Papers No. 18, American Anthropological Association, Arlington, VA, pp. 115−124.
Richards, J. (2010). Kingship and legitimation. In Wendrich, W. (ed.), Egyptian Archaeology, Blackwell,
Malden, MD, pp. 55–84.
Robb, J. (2010). Beyond agency. World Archaeology 42: 493–520.
Robin, C. (2003). New directions in Classic Maya household archaeology. Journal of Archaeological
Research 11: 307–356.
Rosaldo, M., and Lamphere, L. (eds.) (1974). Woman, Culture, and Society, Stanford University Press,
Stanford, CA.
Rostworowski de Diez Canseco, M. (1999). History of the Inca Realm, Iceland, H. (tr.), Cambridge Uni-
versity Press, Cambridge.
Rostworowski de Diez Canseco, M., and Murra, J. (1960). Succession, coöption to kingship, and royal
incest among the Inca. Southwestern Journal of Anthropology 16: 417–427.
Roth, S. (2009). Queen. In Frood, E., and Willeke, W. (eds.), UCLA Encyclopedia of Egyptology. UCLA,
Los Angeles, http://digit al2.libra /zz001 nf7cg .
Rowe, J. (1946). Inca culture at the time of the Spanish Conquest. Bureau of American Ethnology Bul-
letin 21: 183–330.
Sabloff, P. (2018). How pre-modern state rulers used marriage to reduce the risk of losing at war: A com-
parison of eight states. Journal of Archaeological Method and Theory 25: 426–452.
Sabloff, P., and Cragg, S. (2015). Status and role in early states: A comparative analysis. Santa Fe Insti-
tute Working Paper 2015-06-018.
Sahagún, F. B. (1961a). Florentine Codex: General History of the Things of New Spain, Dibble, C., and
Anderson, A. (trs.), Part XI, Book 10, Monographs of the School of American Research and the
University of Utah, Santa Fe, NM.
Sahagún, F. B. (1961b). Florentine Codex: General History of the Things of New Spain, Dibble, C., and
Anderson, A. (trs.), Part IX, Book 8, Monographs of the School of American Research and the
University of Utah, Santa Fe, NM.
Sasson, J. (1973). Biographic notices on some royal ladies from Mari. Journal of Cuneiform Studies 25:
Sasson, J. M. (1995). King Hammurabi of Babylon. In Sasson, J. M. (ed.), Civilizations of the Ancient
Near East, Vol. 2, Charles Scribner’s Sons, New York, pp. 901–915.
Sasson, J. M. (2015). From the Mari Archives: An Anthology of Old Babylonian Letters, Eisenbrauns,
Winona Lake, IN.
Schroeder, S. (1992). The noblewomen of Chalco. Estudios de Cultura Nahuatl 22: 45–86.
Sharer, R., and Traxler, L. (2006). The Ancient Maya, 6th ed., Stanford University Press, Stanford, CA.
Shelach-Lavi, G. (2015). The Archaeology of Early China: From Prehistory to the Han Dynasty, Cam-
bridge University Press, Cambridge.
Shennan, S. (2007). An evolutionary perspective on agency in archaeology. In Gardner, A. (ed.), Agency
Uncovered: Archaeological Perspectives on Social Agency, Power, and Being Human, Left Coast
Press, Walnut Creek, CA, pp. 19−31.
Sillar, B. (2007). Acts of God and active material culture: Agency and commitment in the Andes. In
Gardner, A. (ed.), Agency Uncovered: Archaeological Perspectives on Social Agency, Power, and
Being Human, Left Coast Press, Walnut Creek, CA, pp. 153–189.
Silverblatt, I. (1987). Moon, Sun, and Witches: Gender Ideologies and Class in Inca and Colonial Peru,
Princeton University Press, Princeton, NJ.
Silverblatt, I. (2015). Women. In Urton, G., and von Hagen, A. (eds.), Encyclopedia of the Incas, Row-
man and Littlefield, Lanham, MD, pp. 301–304.
Smith, M. (2003). The Aztecs, 2nd ed., Blackwell, Malden, MA.
Smith, W. (1962). The Old Kingdom in Egypt, Vol. 1, Cambridge University Press, Cambridge.
Spores, R. (1965). The Zapotec and Mixtec at Spanish contact. In Willey, G. (ed.), Handbook of Middle
American Indians: Archaeology of Southern Mesoamerica, Part 2, University of Texas Press, Aus-
tin, pp. 962–987.
Spores, R., and Flannery, K. (1983). Sixteenth-century kinship and social organization. In Flannery, K.,
and Marcus, J. (eds.), The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civiliza-
tions, Academic Press, New York, pp. 339–342.
Stol, M. (2016). Women in the Ancient Near East, Richardson, H., and Richardson, M. (trans.), Walter de
Gruyter, Boston.
Author's personal copy
1 3
Journal of Archaeological Research
Townsend, R. (2009). The Aztecs, 3rd ed., Thames and Hudson, London.
Trigger, B. (2003). Understanding Early Civilizations: A Comparative Study, Cambridge University
Press, Cambridge.
Troncoso, Francisco del Paso y (1579–1581). Papeles de Nueva España, segunda serie, geografia y esta-
distica, Vol. 4: Relaciones geograficas de la diocesis de Oaxaca, Manuscritos de la Real Academia
de la Historia de Madrid y del Archivo de Indias en Sevilla.
Turchin, P., Brennan, R., Currie, T., Feeney, K., Francois, P., Hoyer, D., Manning, J., etal. (2015). Seshat:
The global history databank. Cliodynamics: The Journal of Quantitative History and Cultural Evo-
lution 6: 77–107.
Turchin, P., Currie, T. E., Whitehouse, H., François, P., Feeney, K., Mullins, D., Hoyer, D., etal. (2018).
Quantitative historical analysis uncovers a single dimension of complexity that structures global
variation in human social organization. Proceedings of the National Academy of Sciences 115:
Tyldesley, J. (2006). Chronicle of the Queens of Egypt, Thames and Hudson, London.
Valeri, V. (1985). Kingship and Sacrifice: Ritual and Society in Ancient Hawaii, University of Chicago
Press, Chicago.
Valeri, V. (2014). Rituals and annals: Between anthropology and history. HAU: Classics of Ethnographic
Theory Series, Vol. 2, Westwater, L. (tr.), revised by Elliot, A. https ://haubo ls-and-
annal s/.
Van Zantwijk, R. (1994). Factional divisions within the Aztec (Colhua) royal family. In Brumfiel, E.,
and Fox, J. (eds.), Factional Competition and Political Development in the New World, Cambridge
University Press, Cambridge, pp. 103–110.
Wang, Y. (2004). Rank and power among court ladies at Anyang. In Linduff, K., and Sun, Y. (eds.), Gen-
der and Chinese Archaeology, Rowman Altamira, Lanham, MA, pp. 95–114.
Wengrow, D. (2009). The voyages of Europa: Ritual and trade in the eastern Mediterranean circa 2300–
1850 BC. In Parkinson, W., and Galaty, M. (eds.), Archaic State Interaction: The Eastern Mediter-
ranean in the Bronze Age, School for Advanced Research Press, Santa Fe, NM, pp. 141–160.
Wenke, R. (2009). The Ancient Egyptian State: The Origins of Egyptian Culture (c. 80002000 BC),
Cambridge University Press, Cambridge.
Westenholz, J. (1990). Review: Towards a new conceptualization of the female role in Mesopotamian
society. Journal of the American Oriental Society 110: 510–521.
Whitecotton, J. (1977). The Zapotecs: Princes, Priests and Peasants, University of Oklahoma Press,
Whitecotton, J. (1992). Culture and exchange in Postclassic Oaxaca: A world-system perspective. In
Schortman, E., and Urban, P. (eds.), Resources, Power, and Interregional Interaction, Plenum,
New York, pp. 51–74.
Ziegler, N. (1999a). A questionable daughter-in-law. Journal of Cuneiform Studies 51: 55–59.
Ziegler, N. (1999b). La population féminine des palais d’après les archives royales de Mari: Le harem de
Zimrî-Lîm, Memoires de N.A.B.U. (Nouvelles Assyriologiques Brèves et Utilitaires), No. 1 (Mari),
Société pour l’étude du Proche-Orient Ancien, Antony.
Bibliography of Recent Literature
Ayala Falcon, M. (2002). Lady K’awil, Goddess O, and Maya warfare. In Ardren, T. (ed.), Ancient Maya
Women, Altamira, Walnut Creek, CA, pp. 105–113.
Blanton, R. (1998). Beyond centralization: Steps toward a theory of egalitarian behavior in archaic states.
In Feinman, G., and Marcus, J. (eds.), Archaic States, School of American Research Press, Santa Fe,
MN, pp. 135–172.
Bottéro, J. (2001). Women’s rights. In Bottéro, J. (ed.), Everyday Life in Ancient Mesopotamia, Nevill, A.
(tr.), Johns Hopkins University Press, Baltimore, pp. 112–126.
Campbell, R. (2009). Toward a networks and boundaries approach to early complex polities: The Late
Shang case. Current Anthropology 50: 821–848.
Campbell, R. (2015). Animal, human, god: Pathways of Shang animality and divinity. In Arbuckle, B.,
and McCarty, S. (eds.), Animals and Inequality in the Ancient World, University Press of Colorado,
Boulder, pp. 251–273.
Author's personal copy
Journal of Archaeological Research
1 3
Chase, D., and Chase, A. (2003). Texts and contexts in Maya warfare: A brief consideration of epigraphy
and archaeology at Caracol, Belize. In Brown, M., and Stanton, T. (eds.), Ancient Mesoamerican
Warfare, Altamira, Walnut Creek, CA, pp. 171–188.
Conkey, M., and Gero, J. (1991). Tensions, pluralities, and engendering archaeology: An introduction
to women and prehistory. In Conkey, M., and Gero, J. (eds.), Engendering Archaeology: Women in
Prehistory, Blackwell, Oxford, pp. 3–30.
Connell, S., and Silverstein, J. (2006). From Laos to Mesoamerica: Battlegrounds between superpowers.
In Arkush, E., and Allen, M. (eds.), The Archaeology of Warfare: Prehistories of Raiding and Con-
quest, University Press of Florida, Gainesville, pp. 394–433.
Fash, W. (2005). Toward a social history of the Copan Valley. In Andrews, E., and Fash, W. (eds.),
Copan: The History of an Ancient Maya Kingdom, School for Advanced Research Press, Santa Fe,
MN, pp. 73–101.
Feng, L. (2008). Bureaucracy and the State in Early China: Governing the Western Zhou, Cambridge
University Press, Cambridge.
Fleisher, J., and Wynne-Jones, S. (2010). Authorisation and the process of power: The view from African
archaeology. Journal of World Prehistory 23: 177–193.
Foias, A. (2002). At the crossroads: The economic basis of political power in the Petexbatún region. In
Masson, M., and Freidel, D. (eds.), Ancient Maya Political Economies, Altamira, Walnut Creek,
CA, pp. 223–248.
Freidel, D. (2007). War and statecraft in the northern Maya Lowlands: Yaxuna and Chichen Itza. In Kow-
alski, J., and Kristan-Graham, C. (eds.), Twin Tollans: Chichen Itza, Tula, and the Epiclassic to
Early Postclassic Mesoamerican World, Dumbarton Oaks, Washington, DC, pp. 345–375.
Freidel, D., and Guenter, S. (2003). Bearers of war and creation. Archaeology Online Features, http://
archi ve.archa eolog e/featu res/siteq 2/.
Gardner, A. (2007). Agency Uncovered: Archaeological Perspectives on Social Agency, Power, and Being
Human, Left Coast Press, Walnut Creek, CA.
Garrido, F., and Salazar, D. (2017). Imperial expansion and local agency: A case study of labor organiza-
tion under Inca rule. American Anthropologist 119: 1–14.
Gero, J. (1994). Excavation bias and the woman at home ideology. In Nelson, M., Nelson, S., and Wylie,
A. (eds.), Equity Issues for Women in Archaeology, Archaeological Papers No. 5, American Anthro-
pological Association, Arlington, VA, pp. 37–42.
Graeber, D., and Sahlins, M. (2018). On Kings, Hau Books, Chicago.
Handy, E. (1999). Government and society. In Kamehameha Schools (ed.), Ancient Hawaiian Civiliza-
tion: A Series of Lectures Delivered at the Kamehameha Schools, Mutual Publishing, Honolulu, pp.
Harrison-Buck, E. (ed.) (2012). Power and Identity in Archeological Theory and Practice: Case Studies
from Ancient Mesoamerica, University of Utah Press, Salt Lake City.
Harrison-Buck, E. and Hendon, J. (eds) (2018). Relational Identities and Other-Than-Human Agency in
Archaeology, University Press of Colorado, Louisville.
Hays-Gilpin, K., and Whitley, D. (eds.) (1998). Reader in Gender Archaeology, Routledge Press, New
Yor k.
Hodder, I. (2000). Agency and individuals in long-term processes. In Dobres, M., and Robb, J. (eds.),
Agency in Archaeology, Routledge, London, pp. 21–33.
Hodge, M. (1996). Political organization of the central provinces. In Berdan, F., Blanton, R., Boone, E.,
Hodge, M., Smith, M., and Umberger, E. (eds.), Aztec Imperial Strategies, Dumbarton Oaks, Wash-
ington, DC, pp. 17–45.
Hutson, S. (2017). Unanticipated Politics: Ancient Chunchucmil and Exclusive Maya Collectivities, Uni-
versity Press of Colorado, Boulder.
Jones, P. (2003). Embracing Inana: Legitimation and mediation in the ancient Mesopotamian sacred mar-
riage hymn Iddin-Dagan A. Journal of the American Oriental Society 123: 291–302.
Martin, S. (2003). In line of the founder: A view of dynastic politics at Tikal. In Sabloff, J. (ed.), Tikal:
Dynasties, Foreigners, and Affairs of State: Advancing Maya Archaeology, School for Advanced
Research Press, Santa Fe, NM, pp. 3–45.
Mayerová, H. (2015). The queens of Lagash in the Early Dynastic period, especially the last three royal
couples. In Dittmann, R., and Selz, G. (eds.), It’s a Long Way to a Historiography of the Early
Dynastic Period(s), Ugarit-Verlag, Munster, pp. 259–265.
Martin, S. (2008). Wives and daughters on the Dallas Altar. Mesoweb: www.mesow les/marti
n/Wives &Daugh ters.pdf.
Author's personal copy
1 3
Journal of Archaeological Research
McCafferty, S., and McCafferty, G. (1994). Engendering Tomb 7 at Monte Alban: Respinning an old
yarn. Current Anthropology 35: 143–166.
Meskell, L., and Joyce, R. (2003). Embodied Lives; Figuring Ancient Maya and Egyptian Experience,
Routledge, London.
Nichols, D., Berdan, F., and Smith, M. (eds.) (2016). Rethinking the Aztec Economy, University of Ari-
zona Press, Tucson.
Sabloff, P., and Cragg, S. (2018). Status, role and behavior in premodern states: A comparative analysis.
In Sabloff, J., and Sabloff, P. (eds.), The Emergence of Premodern States: New Perspectives on the
Development of Complex Societies, Santa Fe Institute Press, Santa Fe, NM, pp. 53–102.
Sasson, J. M. (1998). The King and I: A Mari king in changing perceptions. Journal of the American
Oriental Society 118: 453–470.
Smith, M., and Berdan, F. (1996). Introduction to Part II. In Berdan, F., Blanton, R., Boone, E., Hodge,
M., Smith, M., and Umberger, E. (eds.), Aztec Imperial Strategies, Dumbarton Oaks, Washington,
DC, pp. 109–113.
Steadman, S., and Ross, J. (2016). Agency and Identity in the Ancient Near East: New Paths Forward,
Routledge, London.
Sweely, T. L. (ed.) (1999). Manifesting Power:Genderand the Interpretation of Power inArchaeology,
Routledge, New York.
Wengrow, D. (2006). The Archaeology of Early Egypt: Social Transformations in North-east Africa,
10,000 to 2650 BC, Cambridge University Press, Cambridge.
Wilkie, L., and Howlett Hayes, K. (2006). Engendered and feminist archaeologies of the recent and docu-
mented pasts. Journal of Archaeological Research 14: 243–264.
Wright, R. (2016). Cognitive codes and collective action at Mari and the Indus. In Fargher, L., and Espi-
noza, V. (eds.), Alternative Pathways to Complexity, University Press of Colorado, Boulder, pp.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.
Author's personal copy