3

Immanence in the
Andes (1000–1700 ce)
Divine Kingship, Stranger-Kingship, and Diarchy
Peter Gose

T
his chapter discusses the Andes as a case of “pure” immanence
that, unlike the other regions this volume covers, did not develop
an indigenous transcendentalism.1 It will explore Andean political-
religious life as it existed during the Late Intermediate period (1000–1400 ce),
a period of sustained conflict among hierarchically organized regional polities
that ultimately generated the Inca conquest state (1400–1532 ce), which in turn
gave way to the Habsburg period of indirect Spanish colonial rule (1532–1700
ce) that relied on the regional indigenous elites whom the Incas had previously
subordinated. During this span of seven centuries, the ayllu, a territorialized
descent group under the direction of a sacralized ruler with a privileged relation
to the founding ancestor, was the basic unit of Andean society, out of which the
Inca Empire arose and into which it collapsed. These scalable divine kingships
clearly aspired to mediate basic life processes and had nearly all of the imma-
nentist features that Strathern identifies.2 Rather than present them as a mere
instantiation of a type, however, the goal here will be to explore their internal
dynamics, which establish (among other things) the immanent worldly founda-
tions from which transcendentalism can emerge.
Transcendentalism has a polemical relation to immanence3 and often distorts it.
Most of what we know about Andean ayllus comes from colonial “extirpation of
idolatry” documents that describe them from inquisitorial Catholicism’s hostile
transcendentalist perspective and so require careful symptomatic reading when
these documents are used as sources.4 Transcendentalism similarly taints our
contemporary descriptive language: to establish the concept of divine kingship,
Frazer had to speak of priestly kings and propose an original nondifferentiation
54 l Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce)

of temporal and spiritual domains.5 What defined the king as such were not
secular duties but his identification with the realm and its crops, his mediation
of collective access to life, the supreme good in these regimes. Inevitably, this
mediation had a strong ritual (and to that extent, “priestly”) component but not
one separable from the polity’s personification. Indeed, Hocart argued that the
state’s administrative function was a slow outgrowth of a more primordial ritual
responsibility, often embodied in a king who “reigns, but does not govern,” leav-
ing most practical decision making to decentralized civic or tribal processes.6
Even secularized administrative states, Hocart held, never entirely overcome
their origins in ritual. Hocart challenges transcendentalism by showing how its
distinctions between sacred and secular emerge incrementally out of undivided
immanence, not a sudden revolutionary sundering. Here, I continue that argu-
ment by outlining an immanent but nonteleological progression from divine
kingship through stranger-kingship, to diarchy and finally transcendentalism,
each as a potential point of departure for the next, subject to the active realiza-
tion or suppression of immanent possibilities in their actual development. The
goal is not only to reaffirm Hocart’s immanentist demystification of transcen-
dentalism but also to pose immanence dynamically and articulate it through
relevant anthropological concepts.
Divine kingship, with its fundamental emphasis on securing life through
undifferentiated political-religious means, lies at the core of immanence as an
in-worldly order.7 Basically a ritual association for the common good, its relation
to centralized political power is variable, including many cases where the latter is
entirely absent and others (like the Incas) where it is demonstrably derived from
ritual. Among divine kingship’s ubiquitous and defining features are the priestly
nature of the king, his identification with the realm, and his responsibility for the
success of the crops and promotion of life within it: a burden for which he might
pay with his life in times of dearth or when his own vitality diminishes.8 This
tension between the divine king as deity on the one hand and scapegoat or sacri-
ficial offering on the other was undoubtedly conditioned by the development of
institutional state power around him but ultimately intrinsic to his role as repre-
sentative of and mediator for the realm. While we can treat divine kingship as a
type or even a structure, it is more basically a strategy for securing the conditions
for life, an answer to a question. That the answer involves elevating a person into
a being above and beyond the human condition to mediate with extrahuman
forces suggests a view of life as exogenous to human society. Thus, the divine king
may drift immanently into the stranger-king.
First identified by Dumézil, the stranger-king was later elaborated conceptu-
ally by Sahlins to emphasize the disjuncture between king and people or state
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 55

and civil society.9 This disjuncture is initially one of origins, such that the peo-
ple are indigenous and the king foreign, thus enabling him to mediate with
the outside for influxes of life and fertility as in divine kingship. However, the
contrast carries over into social morality because the king breaks with impunity
prohibitions against murder, incest, and so on, that the people are expected to
uphold. Here we see the emerging dark side of divine kingship, where the king is
no longer the people’s scapegoat but a usurping transgressive power that stands
above them, often (but not always) as the state. The people’s challenge therefore
becomes how to incorporate and domesticate this alien power without extin-
guishing its productivity. Marriage to indigenous women is the most frequent
strategy but also common is the representation of the foreigner as orphan, whose
aggressive demands for care and preternatural growth destine him to rule. Both
variants significantly feminize indigenous people, but also render them senior
to the intrusive stranger-king in emergent kinship relations. Out of such a pro-
cess (that both conserve and integrate foreign-indigenous distinctions), diarchy
emerges immanently from stranger-kingship.
Diarchy is a worldwide pattern that distinguishes an active, junior, and intru-
sive sovereign often identified with war and thunder from a passive, senior,
and indigenous sovereign often identified with agriculture and the sun. These
correspond, respectively, to the heroic and cosmological modes of immanent
sacralized kingship articulated by our editors. Both are sacred and reign simulta-
neously, but the nature of their differentiation predisposes the junior partner to
usurp whatever executive power may have developed, and the senior partner to
retreat into priestly activities, such that the two can easily be mistaken for king
and priest. Diarchy always stops short of any transcendental distinction between
the temporal and the spiritual, however, because it retains a complementarity
between diarchic functions that transcendentalism necessarily severs. That split
begins with the priestly critique of kings that secularizes and morally denounces
power but culminates in a more thoroughgoing rejection of the world itself. This
radicalization derives practically from literacy and commodity exchange, whose
“real abstraction”10 (of meaning from time, space, and matter, of exchange-value
from use-value) generates transcendental sensibilities and betrays their social
provenance. Thus, full-blown transcendentalism registers priests’ and merchants’
will to autonomy and refusal to submit to the king, who subsequently struggles to
recuperate their innovations hegemonically. Such developments presuppose but
exceed diarchy so that a real continuity from it into transcendentalism can occur,
beyond the endless category mistakes that arise from describing immanent real-
ities with transcendental concepts. Diarchy is therefore key to an immanentist
critique of transcendentalism. It is the prior basis from which transcendentalism
56 l Imm an en ce i n t h e An des ( 10 0 0 –170 0 ce)

can emerge from immanence and into which it may then backslide. It does not
abolish the difference between them but qualifies and situates their contrast,
helping us identify what is actually at stake in their differentiation.11
In expounding the Andean case below, this chapter will chart a specific ver-
sion of this immanent drift from divine—through stranger-kingship to diar-
chy. However, it will also identify the factors that limited diarchy in the Andes
and with it, any internal impulse toward a transcendentalist order. The point of
this conceptual discussion, then, is not to provide an abstract or transcendental
model but rather to highlight broader dynamics that emergent real interests may
promote or block. Identifying these points of contention should greatly aid com-
parative discussion. For expository purposes, I break the three-part immanent
progression just outlined into two pairs, divine kingship/stranger-kingship and
stranger-kingship/diarchy, each as a section below.

DIVINE KINGSHIP/STRANGER-KINGSHIP IN THE ANDES

Ayllus were localized orders founded around exemplary ancestors who invari-
ably came from afar. The most comprehensive accounts describe these ancestors
arising in cohorts from either the Pacific Ocean or Lake Titicaca, from whence
they journeyed through a series of paqarinas or dawning points before arriving
at the localities they were to settle. Typically, this sequence of dawning points
proceeded through regionally important lakes, then to mountains, and finally to
localized caves, springs, rock outcroppings, or trees. This transition from aquatic
to terrestrial is significant, as we will see below. These journeys were subterranean
for ancestors arising from the Pacific and aerial for those who arose from Lake
Titicaca. Dawning points occurred where those journeys touched the earth’s
surface. There, ancestral cohorts rested before dispersing in ever-smaller group-
ings toward their divergent destinations, until they arrived singly, in male-female
pairs, or father-son groupings, either from the bowels of the earth or by light-
ning strike. Upon arrival, they built settlements and agricultural infrastructure
or conquered other groups already in the area, thus establishing a patrimony for
the localized descent groups they established. After completing their coloniz-
ing missions, accounts often mention their spontaneous petrification, an act that
marked their metahumanity, not ordinary mortality.12
In addition, founding ancestors often left behind mortal bodies that were
mummified and that presided over mortuary caves or towers in which the
mummies of their ayllu descendants accumulated. These founding ancestors
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 57

were the reference points for calculating rank and the transmission of political
office within the ayllu so that a succession of elder-rulers (kurakas) ostensibly
descended from them. But this descent was established less on any genealogical
grounds than by presiding over the mummification and mortuary rites of one’s
predecessor13 and continuing to offer sacrifice in an active cult. Because descent
was fundamentally performative, it gave great political significance to the fulfill-
ment of key priestly duties, which no kuraka could neglect or delegate beyond
a certain point,14 and which frequently occasioned significant intrigue. Ances-
tor worship and filial piety did not provide a missing moral legitimation (as in
transcendental regimes) for the outcome of sordid political power struggles but
instead expressed them directly because their entire point was to tap into the
founders’ animating power. Thus, Andean divine kingship might distinguish
military prowess from ritual observance, but it did not fundamentally oppose
them because both aspired to the same immanentist ends. A successful ruler had
to shift fluidly between these modalities and prevent their delegation from hard-
ening into permanent occupational specializations from which he was removed.
Andean statecraft thus required a proliferation of surrogates for the ruler, most
commonly statues, which he could dispatch to perform various duties in far-
flung places while continuing to animate them all in oracular mode as extensions
of himself, very much on an ancestral model.15
Ancestors remained secluded from their descendants throughout most of the
year, stored in mortuary facilities associated with the final dawning points on
their journeys. While close to their living descendants on the territories they col-
onized, these mortuary facilities also pointed to the ancestors’ remote origins and
the foreignness of their power, which extended to that of the living kuraka, who
remained somewhat secluded from ayllu commoners by prohibitions that could
include touching, direct conversation, eye contact, and even public visibility.
More powerful kurakas also avoided contact with the earth by sitting on a throne
and being carried in a litter,16 as if such contact would establish new dawning
points in the manner of the founding ancestors on their journeys of colonization.
Here we see that, although the power of ancestors and the rulers descended from
them was decidedly immanentist, it nonetheless operated at some remove from
ordinary mortals and the life-worlds they inhabited. There was a certain tension
between the immanence of this power and its derivation from afar, which dis-
tinguished sources or foci of animating power from the world they animated.
Andean divine kingship was not a decentered animism but a hierarchical circu-
lation of life that metapersons enacted. To arise as such, these agents had first to
be consolidated against their background life-world, and only then could they
return to it as an animating force. Andean immanence was thus imminent and
58 l Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce)

not fully present. Rather, it consisted of a pulse involving withdrawal and return
in alternation.
This pulse appeared most obviously in annual cycle rituals. Their number and
timing varied from one place to another, but minimally and universally included
festivities around planting (poqoymita) and harvest (karwaymita), during which
the ayllu brought its living and mummified members together for a multiday
occasion, called the vecochina, in a ceremonial plaza in front of its mortuary cave
or tower. Prior to these celebrations, the living had to abstain from sex, corn
beer, chile, and salt for five days; confess all transgressions to ayllu ministers;
and undergo purificatory bathing. These prohibitions involved no elaborated
priestly morality or transcendentalist notions of purity and pollution. Rather,
they addressed ayllu unity and set an appropriate equilibrium of wetness and
dryness in living bodies prior to receiving the ancestors’ animating power across
the osmotic frontier between the living and the dead. The celebrations proper
began by recounting the founding ancestors’ journeys, then moved to many days
of festive drinking, sacrifices, oracular communication, and even dancing with
the mummified dead. Perhaps the most intimate care the living offered their
mummified dead was to change their clothes and repair any damage to their
flesh with copper appendages. Once these festivities ended, the living returned
the mummified dead to their mortuary caves or towers until the next such cele-
bration.17 Thus, intense festive contact between the ayllu’s living and dead con-
trasted with their quotidian separation. It generated an animating charge that
drove agrarian production and sexual reproduction within the ayllu. As with the
founding ancestors’ and living rulers’ journeys, however, these points of contact
with the earth and its life had to be limited and strategic in order to be manage-
able. The immanence of ancestral power did not mean it was equally present in
all things and at all times. It necessarily withdrew and advanced in phases and
through ritual interventions.
These seasonal festivals also served as mortuary observances for the ayllu’s
recently dead, integrating them into the ancestral community. Like established
ancestors, they received food and libations during these festivities to meet their
ongoing corporeal needs, including a journey to the afterlife. Living relatives
danced through the night around the ceremonial plaza and entrance to the mor-
tuary caves with the bodies of the recently dead on their backs in an observance
called paqarikuq or paqarikuspa, meaning “vigil,” but also paqarina, meaning
“ancestral origin point.”18 This is exactly what mortuary caves and towers were,
but one that marked the beginning of their journey to the abode of the dead
or upaymarka, which was located in their ancestors’ remote places of origin.19
Thus ayllu dead passed through the same sequence of paqarinas as their founding
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 59

ancestors, but in reverse, starting from the localities their ancestors colonized
and ending at their point of departure.20 The vecochina appears to have marked
the arrival of the recently dead in the upaymarka and submerged their individual
biographies in the ayllu’s collective seasonal rhythms,21 essentially an alternation
between wet and dry.
Death divided the person into separate components that were spatially dis-
tributed. On the one hand, there was the mummified body dried and hardened
ideally to the point of imperishability22 and retained in the locality. On the other,
there was a more fluid aspect lost to the body and its locality to remote aquatic
ancestral origin points. The Cajatambo idolatry corpus designates these entities
as kamaqnin and upani, respectively.23 A kamaqnin derived from the kamaq or
ancestral animator. In the first instance, it was that ancestor’s statue or mum-
mified body but also designated the transfer of a soul-like entity to the hearts
of its decendants via an internal replication.24 Thus, descendants were “infused”
(kamasqa) by the ancestor. The Quechua verb kamachiy expressed this process.
On strictly analytical grounds, it meant “to cause to exist” but in actual usage
“to command.”25 Both meanings converged in an immanentist ontological order
that inseparably linked animation to political rule. In death, the kamaqnin’s
power persisted as a localizing or fixing influence manifest in the accumulation
of mummified bodies in the mortuary cave over which it presided. Extirpation
of idolatry documents frequently mention the mummified dead’s thirst,26 and
all dealings with them required offerings of corn beer. Thus, their preservation
implied a loss of water and vitality embodied in another aspect of the person that
did not remain under the kamaqnin’s command.
The upani, which meant “shadow” or “deaf-mute being,”27 was a much
weaker shade than the kamaqnin but nonetheless eluded its localizing con-
trol on death. In contrast to the kamaqnin’s fixity, it journeyed to the abode
of the dead (upaymarka) generally located in distant, large bodies of water.28
Thus, the upani was affiliated to the moist part of the person lost at death.
This moisture was not merely lost to the dead person but also to the locality
of his or her ayllu and to that extent was a geographical issue. The upani’s jour-
ney to the upaymarka enacted in reverse the founding ancestors’ journey from
their remote points of origin in the Pacific or Lake Titicaca, which were not
only the maximal paqarinas of the Andean cosmos but also the ultimate col-
lecting points for the watery aspect of the dead.29 Thus, ancestral colonization
and the upani’s postmortem return to their ancestors’ origin points defined a
cycle based on the differentiation of dry, local paqarinas from distant wet ones,
which drove the redistribution of life between them. Petrified or mummified
ancestral bodies lost water to the distant centers from which they originated
60 l Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce)

but also commanded its return in a cycle where animation and political control
were endlessly lost and regained.30
To summarize, a kamaqnin contrasted with an upani in several ways. While
the former was an exemplary and powerful ancestor who remained in the locality
it colonized, an upani was a weak and insignificant shadow that drifted away. The
kamaqnin was a repository and replicator of group life, whereas the upani was a
remnant of an individual life lost to the group and its locality. A kamaqnin stayed
dry and hard aboveground, whereas an upani journeyed underground through
the earth’s aquatic interior. Finally, a kamaqnin never really died but continued
to speak through oracular media and was deemed mute (upa) or done (atisqa) if
it could not, signaling the end of its ancestral career.31 By contrast, the deceased
upani was categorically mute.
These distinctions (which at death became separations) were primarily mate-
rial in nature and to that extent were immanent, not transcendental. Death
withdrew and recycled water from desiccated bodies, allowing its redistribution
across the landscape and enabling the Andean circulatory cosmos.32 Geograph-
ical separation between the dry and wet poles of ancestral paqarina itineraries
ensured that vital circulation was not tightly confined to humans in any rein-
carnation cycle but instead suffused the land itself and was shared with other
beings that lived on it, including plants and animals. This circulation’s driving
force, however, was the founding ancestors whose original journeys established
the pathways by which life moved and whose dry, imperishable bodies drew lost
vitality back across the landscape to animate the local life-worlds over which they
presided. Their remote origins and local fixity defined these animating circuits.
Thus, the very dryness of founding ancestral mummies established not just a par-
titive contrast with aquatic life lost to the periphery but also a means of summon-
ing it across what I call the osmotic frontier in a bidirectional flow. In times of
drought, for example, people paraded desiccated mummies across the landscape
to petition rainfall.33 An alternative title for ancestral mummies was mallki, a
term that also designated “seed,” “young plant ready for transplanting,” “sapling,”
and “fruit tree.”34 These entities all awaited the arrival of water to renew life and,
to varying degrees, could draw it in. Mummification was thus comparable to
the conservation of seeds by drying, and interment was comparable to plant-
ing.35 Aquatic life and the forms it vitalized were sufficiently interdependent that
the very dryness of seedlike mummies allowed them to affect distant sources of
water, attracting them back toward the localities that were also continuously los-
ing animating liquids. As with other such systems, Andean immanence was not
fully monistic or a flat, unmediated presence but more an anticipated arrival of
vitality circulating across spatial frontiers.36
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 61

STRANGER-KINGSHIP AND DIARCHY

Thus far, we have considered Andean divine kingship/stranger-kingship as a

relatively unified matter, fundamentally concerned with the loss of life to and
attraction of life from distant points of origin. We will return to reaffirm this
view, but let us now turn to the duality of the local and the distant implicit in
stranger-kingship, which was also basic to moiety distinction within Andean ayl-
lus during our time period. Throughout the Andes, moiety divisions between
an “upper” (hanan) and a “lower” (hurin) group were ubiquitous and typically
defined the former as intrusive conquerors who originated from Lake Titicaca
and the latter as conquered indigenous people who originated from the Pacific.37
These distinctions tended to be recursive within Andean sociopolitical organi-
zation such that upper and lower divisions could be found anywhere from the
minimal level of a hamlet up to the maximal level of the polity itself, in classic
segmentary fashion. At each level, the upper group outranked and encompassed
the lower group such that its leader represented the unity of the two groups
to the next ascending level, which at the top of the polity’s internal hierarchy
became the outside.38 Thus, Andean moiety distinctions were an outgrowth of
stranger-kingship, in which mediation with the external became the dominant
principle of internal organization.
The most detailed accounts of how these dualities operated are in the Cajat-
ambo idolatry corpus and the Huarochirí manuscript, both from the central
Peruvian highlands.39 Here, the conquering “upper” group was called llakwash,
worshiped lightening, and practiced pastoralism, whereas the indigenous
“lower” group was called wari, worshiped the sun, and practiced agriculture.40
The Huarochirí account vividly describes the arrival of these intrusive conquer-
ors; the battle of their ancestor Pariaqaqa to displace the indigenous ancestor
Wallallo Karwincho as the animator of the land; the intruders’ control over
water as decisive in subjugating the native agriculturalists; and the relations of
intermarriage, reciprocal ancestor worship, and economic interdependence that
followed.41 Of note is that, while the intrusive pastoralists were militarily dom-
inant, they were represented as poor and needy, whereas the indigenous agri-
culturalists were depicted as economically prosperous. The result was that the
latter had to intermarry and share their maize-growing land with the former:
interdependence, not ethnic cleansing. Indigenous ancestors who made the area’s
agricultural infrastructure still received veneration for their ongoing stewardship
of the crops, but the intrusive pastoralists’ lightening deity was recognized as the
distributor of rainfall and thus held the preeminent position in local relations of
ritual complementarity. Once again, the stranger-king pattern emerges, not only
62 l Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce)

in the pastoralists foreign origins but also in their mediation with the outside for
the inward flow of life, now specifically as water.42
These substantive associations of upper and lower moieties correspond quite
remarkably to those Hocart outlined for junior and senior diarchs, where the
former occupy the active seat of power and are associated with thunder, and
the latter are more passive and associated with law, regularity, and the sun.43
They also conform to the distinction between heroic and cosmological modes
of immanentist sovereignty from this volume’s introduction. Diarchy integrates
those distinctions and renders them complementary and processual, providing
a certain basis from which the secular-sacred dichotomy of authority associ-
ated with transcendentalism may spring. In what follows, therefore, I will try to
unpack the subtleties, dynamics, and contradictions of Andean diarchy, show-
ing how they remain within an immanent framework despite creating openings
for transcendence.
Hierarchical complementarity was the normative relation between upper and
lower moieties but the dominant position within that hierarchy was contestable
and never entirely stable. Moieties retained enough autonomy and competitive
symmetry that their complementary integration was not the only dynamic in
play. Ritual battles at key points in the agricultural cycle were one occasion where
they could pit forces against each other, sometimes with the “upper” group’s tri-
umph as a predetermined outcome but always hinting at the possibility of a real
competition. Should the latter break out and the “lower” group prevail, the result
would be a pachakuti, or cataclysmic overturning of the dominance relations
between upper and lower divisions and the establishment of a new animating
order. At least two Inca crises of rule (Pachakuti Inca’s defeat of the Chanka and
the Waskar-Atawallpa war of succession) featured an aspirant to power recruit-
ing hitherto dormant “lower” forces to overcome a rival and establish a new
dispensation.44 Thus, previously vanquished powers could return to defeat their
conquerors in an open-ended succession of animating regimes. This possibility
mitigated against their transformation into the fully domesticated complement
of the upper group’s political power, that is, into something like disempowered
spiritual authority.
Sometimes upper and lower moieties appear to have pursued separate alliances
to outside forces as part of their internal rivalry. For example, Medinaceli argues
that during the Spanish “conquest,” Paullu and Manqho Inca effectively served as
diarchs, pursuing policies of collaboration with and resistance to the Spaniards,
respectively.45 Although she does not map these diarchs onto lower and upper
divisions, such associations would be entirely in keeping with their respective
political stances. A related case was that of the Wanka, a powerful ethnic polity
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 63

that allied with the Spaniards against the Incas during the “conquest,” presum-
ably with the objective of winning back a greater degree of autonomy. So firm was
this stance that the Wankas became the most academically recognized Andean
ally of the Spaniards in defeating the Incas, although several other Andean pol-
ities were equally committed to that cause. It is a surprise, therefore, to find that
on Wanka territory, military preparations linked to the Manqho Inca resistance
were detected in 1564 and 1565 during the Taki Onqoy anti-Spanish movement,46
despite the fact that the Wankas continued to be ruled by the collaborationist
Apoalaya clan. Another faction of the same polity apparently pursued an opposed
political strategy. The advantage of such contradictory stances for a group is that
they could always be sure of having one faction come out on the winning side of
any larger conflict. By the same token, however, they would always be conflicted
over internal supremacy and external loyalties. Because Andean diarchy and moi-
ety divisions were already locked into that competitive dynamic by definition, it
follows that their internal rivalry would occasionally spill over into contradictory
diplomatic mediations with outside powers.
Of particular interest in this regard are details from “Inca history” that
bear on the relation between Upper and Lower Cuzco in the imperial capital.
Following the broader Andean pattern, the Inca ruling elite claimed moiety
affiliation with Upper Cuzco47 and left Lower Cuzco to the descendants of the
local indigenous people they conquered and those of mixed unions. Along the
same lines, the sun and its priesthood were located in Lower Cuzco, even
though these priests appear to have been recruited primarily or exclusively
from the Inca elite in Upper Cuzco.48 Thus, an opposition within the Inca elite
between the ruler (sapa Inca) and the priesthood, particularly the high priest
or oracular medium of the Sun (willaq umu), superimposed itself on the basic
moiety division between conquerors and indigenous people. What interac-
tions and effects resulted?
One tendency was toward complementarity, in which the roles of ruler and
priest became differentiated but interlocking halves of an encompassing whole
on the model of Andean moieties. For example, Molina underlines the senior-
ity relations involved by describing the Inca sovereign as the “Sun’s son” (indi
churi) and the high priest as the “Sun’s servant” (indivianan).49 By acting in
place of the Sun, the high priest simultaneously claimed seniority and ceded
the active seat of power to the sovereign. Indeed, the willaq umu’s principal
duty was to act as kingmaker: the final, oracular voice of the Sun in ratifying
succession to high office.50 This role often extended into offering tutelage for
young sovereigns without sufficient experience to govern properly.51 Using
such observations, some commentators argue not only that the Incas had a
64 l Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce)

developed diarchy but that it began to approximate a sacred/secular split typical

of transcendentalism.52
A countertendency was toward overlap or nondifferentiation between the
supreme Inca and the high priest of the Sun. For example, the Inca sovereign
performed important priestly functions in many imperial rituals, during which
he left his governing duties.53 In this scenario, the high priest was a stand-in who
acted on behalf of the Inca, who could reclaim at will the priestly powers he
delegated. Thus, when Huayna Capac took office, his first reported deed was to
remove the high priest and act as such himself, inspecting all the shrines, oracles,
and their estates.54 Other sovereigns similarly intruded on the role of the willaq
umu.55 Even when relations with the willaq umu were cooperative, the Inca sov-
ereign would still join or replace him in his priestly duties, particularly during
oracular consultations of the Sun. Conversely, the willaq umu could assume the
duties of the Inca sovereign, as happened during the rebellion of Manqho Inca,
when he played an important military role.56 On these and other grounds, one
can challenge the idea that the Inca sovereign and willaq umu had a fixed, hier-
archical, and complementary division of labor along junior and senior lines.57
While they certainly joined most Andean peoples in distinguishing upstart
militarist modes of power from their established priestly counterparts through
moiety system and related forms of ritual practice, cosmology, and narrative,58
the Incas failed to consolidate these distinctions into stable offices with specific
and limited duties. Arguably this made them a more authentic diarchy than pol-
ities that did. Rather, the original impulse of divine kingship toward the unified
deployment of military and priestly power prevailed. In short, no entrenched dis-
tinction between secular political power and transcendental religious authority
arose under the Incas: the sovereign not only outranked the high priest in gen-
eral but also specifically within the priestly domain. Dumont’s analysis of South
Asia59 cannot be extended to the Andean case because complementarity never
replaced delegation there as the basis of the king-priest relation. The comple-
mentarity that did exist in Andean moiety relations thus failed to recast the Inca
king-priest relation when the two were superimposed.
The competitive, symmetric aspect of Andean moiety relations was probably
the main countervailing force behind the formation of a clear, complementary,
and hierarchical division of labor between Inca rulers. Molina states that the
high priest was the ruler’s “second person”60 in much the same way that the rep-
resentative of any lower moiety would be in relation to his counterpart in an
upper moiety. As such, his relation to the Inca ruler featured the same mixture
of subordination and rivalry that all “lower” authorities had with their “upper”
counterparts. According to Cieza, the willaq umu’s influence competed with that
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 65

of the Inca.61 Indeed, Valera treats the willaq umu as formerly having been more
powerful than the Inca sovereign, a situation that was only reversed during the
reign of Topa Inca.62 Using additional sources, Zuidema argues that this rever-
sal enacted the basic categorical assumptions of Andean moiety systems, which
posit the defeat by young or intrusive forces of a previously dominant power that
becomes priestly, quietist, and associated with indigenous people,63 in short, the
basic stranger-king scenario. While this normative reading is undoubtedly cor-
rect as far as it goes, it nonetheless presupposes the possibility of struggle and
contention, a more symmetric moment in which moieties’ hierarchical coding
hangs in the balance and is vulnerable to inversion.
During the time period under discussion here, such instability was the rule,
not the exception. The Late Intermediate Period out of which Inca imperialism
arose was an epoch of petty conquests in which neighboring groups imposed
themselves upon each other, a reality registered in the invader/indigenous cod-
ing of local moiety systems. While the Inca expansion nominally imposed an
overarching order on this panorama, the death of each Inca sovereign nonethe-
less opened the door to provincial secessionist rebellions and wars of succes-
sion among the Inca elite. The Spanish invasion, at least in the short run, only
exacerbated this chronic political instability and expanded the possibilities for
maneuver. These considerations should caution against an overly “structural”
reading of the asymmetric duality of Andean moieties and their hierarchical
logic, which attempted to contain political realities that were far more fluid.
This logic was aspirational, performative and never precluded the possibility of
challenge and reversal from below. If successful, challengers could themselves
be assimilated into the same framework as the new dominant (“upper”) entity
in an endless succession of ex post facto revisions. Thus, some have preferred to
treat Andean moieties diachronically as a developmental progression whereby
political regimes arise and run their course into senility rather than as a static
synchronic order.64
All of this is consistent with divine kingship’s fundamental premise that sov-
ereignty is subject to violent, reinvigorating renewal. Despite their very real flir-
tations with diarchy as hierarchical complementarity, Andean polities constantly
returned to this disruptive point of departure. Complementarity developed in
provincial moiety systems to the point of distinguishing junior intrusive mili-
tarists from senior indigenous priests who remained interdependent within a
single ancestral framework in which the former dominated. Conquests estab-
lished new ancestral patrimonies for priestly regulation, which in turn posited a
larger cosmos available for imperial control. The Inca imperial project embodied
exactly this dynamic: instead of producing two divergent social logics, it strove
66 l Imm an en ce i n t h e An des ( 10 0 0 –170 0 ce)

to perfect divine kingship’s basic identification of political sovereignty with the

totality and fertility of the land. In realizing this unitary agenda of political and
cosmological control, the highest administrative levels of their massive conquest
state largely suspended the recursive dualisms of Andean moiety organization to
reveal a maximalist divine kingship. Securing a single regime of animating power,
whatever its internal hierarchy and multiplicity, was a broadly shared Andean
goal that the Incas sought to realize. In so doing, they showed that diarchic ten-
dencies were perfectly compatible with political-cosmological immanence and
implied no necessary transition toward transcendentalism.

The Andean material covered here thwarts any teleological expectation that the
continuum of divine kingship into stranger-kingship and diarchy discussed at
the outset should necessarily lead to transcendentalism. Although Andean diar-
chy produced something formally comparable to the king-priest distinction of
transcendental regimes, the dynamics of that relation turn out to be fundamen-
tally different. Instead of generating two supposedly separate realms defined by
opposed values, the worldly versus the transcendental, the Andean case used this
distinction to outline divine kingship’s progression from vigorous beginnings to
senile demise. Each of these states amounted to a different point of view with
divergent interests and concerns that might complement or compete with each
other but never escaped divine kingship’s defining preoccupation with the cir-
culation of life and thus in-worldly immanence. Similarly, a realm of the beyond
existed in the ancient Andes, but it was a vast pool of aquatic life at the world’s
edge, not a spiritual domain entirely outside it.65 As long as in-worldly power
addressed that periphery and attempted to mediate with or control it, the result
was an immanentist regime. The first conclusion is therefore that the distinction
between immanence and transcendence survives even the mapping of a possible
transformational continuum between them. Morphological opportunities for
transcendentalism existed but were not realized.
To account for this road not taken is to show how Andean divine kingship
imposed its logic on the entire social field and prevented the emergence of sub-
versive practices and actors within it. As a form of hegemony, all divine kinghips
directly addresses the primary class relations of an agrarian society by mediating
peasant livelihood concerns through sovereign power so that rulers and ruled
can find a common cause. The Incas elaborated on this template by linking war to
agriculture when the sovereign initiated the agricultural year by breaking ground
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 67

to the accompaniment of martial songs66 and undertook imperialist expansion

to assert agrarian control over water lost to an aquatic periphery. Priests played
an important role within this hegemonic process, sometimes as the sovereign’s
delegates and others as a distinct occupational group, but their activities never
became detached from it to claim a moral monopoly or initiate an alterna-
tive transcendentalist social logic. To do so, they would have to draw on social
groups and practices that could be detached from divine kinghip’s immanentist
logic. Comparative work on the Axial Age suggests that merchants were a likely
social source of this development because the “real abstraction” of commodity
exchange favored transcendental thinking.67 Therefore, it is significant that the
Incas actively suppressed commodity exchange and attempted to subsume it into
their tributary system,68 which they presented as part of a cosmic circulation of
life in classic immanentist fashion. That Andean people outside the Inca milieu
practiced commodity exchange suggests this Inca policy was deliberate, strate-
gic, and about control. A final impediment to transcendentalism in the Andes
was the minimal development of displacement in knotted cord (khipu) literacies,
whose primary function was the ritualized public accounting of ayllu tributary
obligations.69 While some still-undecoded narrative functions also existed, the
khipus’ deployment in contextually disembedded discourse, another key tran-
scendentalist precondition, was limited relative to other literary technologies.
In a remarkably thorough immanentist hegemony, then, Andean divine kingship
contained most existing social practices but suppressed or incorporated those
that might have mounted a transcendentalist challenge.
Perhaps trying to explain why transcendentalism did not emerge endoge-
nously in the Andes is a fool’s errand that only perpetuates its false claims to
universality and death grip on our descriptive language. By showing that Andean
immanentism had its own dynamics that correspond to concepts of divine king-
ship, stranger-kingship, and diarchy, this chapter has argued, however, that it
was not the crude idolatry, empiricism, or metaphysics of presence that transcen-
dentalism disparagingly perceives it to be. Andean immanentism accepted the
partiality of human experience, the limits of local political horizons, trafficked
in many of the same existential mysteries as transcendentalism, and promoted
cosmological elaboration to at least the same degree. In so doing, it produced
some formal outlines that resembled those of transcendentalism but retained
an alternative social grounding and logic. By highlighting the contingency and
hegemonic struggle between the two modes of religiosity, this chapter has argued
that transcendentalism has an immanentist basis, but one that failed to material-
ize endogenously in the Andes.
68 l Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce)

Notes

1. Of course, Spanish colonialism did bring an ambiguously transcendental version of

Christianity to the area in 1532, one that was still too immanent for the protestant
critique. Spanish Catholicism transferred the charge of idolatry that it experienced
itself in Europe to its Andean subjects on the colonial periphery, as is evident in
nearly all the primary sources on Andean politico-religious practices. Out of this
dynamic, a wonderfully inventive Andean Christianity nonetheless emerged, one
that is frequently dismissed as such because of its resolutely immanent character. See,
among many other studies, Thomas Abercrombie, Pathways of Memory and Power:
Ethnography and History among an Andean People (Madison: University of Wiscon-
sin Press, 1998); Peter Gose, Deathly Waters and Hungry Mountains: Agrarian Rit-
ual and Class Formation in an Andean Town (Toronto: University of Toronto Press,
1994); Peter Gose, Invaders as Ancestors: On the Intercultural Making and Unmaking
of Spanish Colonialism in the Andes (Toronto: University of Toronto Press, 2008);
Olivia Harris, “The Dead and the Devils among the Bolivian Laymi,” in Death and
the Regeneration of Life, eds. M. Bloch and J. Parry (Cambridge: Cambridge Univer-
sity Press, 1982), 45–73; Tristan Platt, “The Andean Soldiers of Christ: Confrater-
nity Organization, the Mass of the Sun and Regenerative Warfare in Rural Potosí
(18th–20th Centuries),” Journal de la Société des Américanistes 73 (1987): 139–92;
Michael Sallnow, Pilgrims of the Andes (Washington, DC: Smithsonian Institution
Press, 1989).
2. Alan Strathern, Unearthly Powers: Religious and Political Change in World History
(Cambridge: Cambridge University Press, 2019), 27–46.
3. Srathern, Unearthly Powers, 61–63, 68, 242.
4. Pierre Duviols, La lutte contre les religions autochtones dans le Pérou coloniale (L’ex-
tirpation de l’idolatrie entre 1532 et 1660) (Lima: Institut Français d’Etudes Andines,
1971); Gose, Invaders as Ancestors.
5. James G. Frazer, The Golden Bough: A Study in Magic and Religion, 3rd ed., 12 vols.
(London: MacMillan, 1911–1915), 1: 44–51, 2: 139.
6. Arthur M. Hocart, Kings and Councillors (Chicago: University of Chicago Press,
1970), 30–40, 128–55, 135.
7. Frazer, The Golden Bough, 1: 420; Hocart, Kings and Councillors, 81.
8. Frazer, The Golden Bough, 1: chapter 1.
9. Marshall Sahlins, Islands of History (Chicago: University of Chicago Press, 1985),
chapter 3; David Graeber and Marshall Sahlins, On Kings (Chicago: HAU, 2017),
chapter 1.
10. Alfred Sohn-Rethel, Intellectual and Manual Labour: A Critique of Epistemology
(London: MacMillan, 1978).
11. Hocart, Kings and Councillors, chapter 12.
12. Pierre Duviols, “Un symbolisme Andin du double: La lithomorphose de l’ancêtre,”
Actes du XLIIe Congrès International des Américanistes IV (1978): 359–64.
13. Juan de Betanzos, Suma y naración de los Incas (Madrid: Ediciones Atlas, 1551 [1987]),
207–209, 285–6.
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 69

14. José Luis Martínez Cereceda, Autoridades en los Andes, los atributos del señor (Lima:
Pontificia Universidad Católica del Perú, 1995), 39–43.
15. Peter Gose, “Oracles, Mummies, and Political Representation in the Inka State,”
Ethnohistory 43, no. 1 (1996): 1–33.
16. Martínez Cereceda, Autoridades, 91–101.
17. See Pierre Duviols, Cultura Andina y represión: Procesos y visitas de idolatrías y
hechicerías, siglo XVII (Cuzco: Centro de Estudios Rurales Andinos «Bartolomé de
las Casas», 1986), 86–87, 93, 179–80, 188, 195, 281, for the best primary sources on the
vecochina. For commentary and analysis, see Mary Eileen Doyle, The Ancestor Cult
and Burial Ritual in Seventeenth and Eighteenth Century Central Peru (Ann Arbor:
University of Michigan Dissertation Services, 1988); Kenneth Mills, An Evil Lost
to View? An Investigation of Post-Evangelization Andean Religion in Mid-Colonial
Peru (Liverpool: Institute of Latin American Studies, 1994); and Gose, Invaders as
Ancestors.
18. Duviols, Cultura Andina, 8–10, 13, 16, 19, 25–28, 64–65, 72, 77–79, 81.
19. See Hernando de Santillán,“ Relación del origen, descendencia, politica y gobierno de
los Incas” in Historia de los Incas y relación de su gobierno, Colección de Libros y doc-
umentos referentes a la historia del Perú, vol. 9, ed. H. Urteaga (Lima: Imprenta
y Libreria Sanmarti, 1553 [1927]), 33; Pedro de Cieza de León, La crónica del Perú,
Parts 1 and 2, in Obras completas, vol. 1 (Madrid: Consejo Superior de Investigaciones
Cientificas, Instituto Gonzalo Fernández de Oviedo 1553 [1984]), 122; Felipe Gua-
man Poma de Ayala, Nueva corónica y buen gobierno (Paris: Université de Paris, 1615
[1936]), 70; José de Arriaga, “Extirpación de la idolatría en el Perú,” in Biblioteca de
Autores Españoles, vol. 209 (Madrid: Ediciones Atlas, 1621 [1968]), 202, 216; and
Duviols, Cultura Andina, 150, 171, 200, 227, 268–9.
20. Doyle, The Ancestor Cult, 241; Peter Gose, “Segmentary State Formation and the
Ritual Control of Water under the Incas,” Comparative Studies in Society and History
35, no. 3 (1993): 480–514.
21. Doyle, The Ancestor Cult, 232–3, 240–3, 257.
22. Frank Salomon, “ ‘The Beautiful Grandparents’: Andean Ancestor Shrines and
Mortuary Ritual as Seen through Colonial Records,” in Tombs for the Living:
Andean Mortuary Practices, ed. Tom D. Dillehay (Washington DC: Dumbarton
Oaks Research Library and Collection, 1995), 328.
23. Pierre Duviols, “Camaquen, Upani: Un Concept Animiste des Anciens Péruviens,”
in Estudios Americanistas, vol. 1, eds. R. Hartmann and U. Oberam (Bonn: Collecta-
nea Instituti Anthropos, vol. 20, 1978), 132–44.
24. Duviols, Cultura Andina, 67, 143–44.
25. Gerald Taylor, “Camay, Camac et Camasca dans le Manuscrit Quechua de Huaro-
chirí,” Journal de la Société des Américanistes LXIII (1974–6): 236.
26. Duviols, Cultura Andina, 198, 217, 230.
27. Duviols, “Camaquen, Upani,” 143–44; Gerald Taylor, “Supay,” Amerindia 5 (1980):
51–55.
28. Duviols, “Camaquen, Upani,” 136; Taylor, “Supay,” 58.
29. Duviols, Cultura Andina, 150, 171, 200, 227, 268–69.
70 l Imm an en ce i n t h e An des ( 10 0 0 –170 0 ce)

30. Jorge Polo de Ondegardo, De los errores y supersticiones de los indios, sacadas del trat-
ado y averiguación que hizo el Licenciado Polo, in Informaciones acerca de la Religión
y Gobierno de los Incas, vol. 1 (Lima: Imprenta y Libreria Sanmarti, 1554 [1916]), 10.
Peter Gose, "Segmentary State Formation and the Ritual Control of Water Under
the Incas," Comparative Studies in Society and History 35, no. 3 (1993): 480–514.
31. Duviols, Cultura Andina, 180; Cristóbal de Albornoz, “Instrucción para descubrir
todas las guacas del Piru y sus camayos y haziendas,” Journal de la Société des Améri-
canistes LVI, no. 1 (1967): 18, 37.
32. Peter Gose, “The Andean Circulatory Cosmos,” in The Andean World, eds. L.
Seligmann and K. Fine-Dare (London: Routledge, 2019), 105–27.
33. Polo de Ondegardo, De los errores errores y supersticiones, 10.
34. Domingo de Santo Tomás, Lexicon o vocabulario de la lengua general del Peru (Lima:
Universidad Nacional Mayor de San Marcos, 1560 [1951]), 314; Diego González
Holguín, Vocabulario de la lengua general de todo el Perú llamado lengua Qqichua o
del Inca (Lima: Imprenta Santa María, 1608 [1952]), 224.
35. Frank Salomon and Jorge Urioste, The Huarochirí Manuscript (Austin: University of
Texas Press, 1991), 43; Luis Valcárcel, “La Religión Incaica,” in Historia del Perú, vol. 3
(Lima: Mejía Baca, 1980), 81.
36. Strathern, Unearthly Powers, 31–3.
37. Pierre Duviols, “Huari y Llacuaz: Agricultores y Pastores. Un Dualismo Prehis-
pánico de Oposición y Complementariedad,” Revista del museo nacional 39, 153–91.
38. Juan de Matienzo, Gobierno del Perú (Lima: Institut Français d’Etudes Andines, 1567
[1967]), 20–21.
39. Duviols, Cultura andina; Salomon and Urioste, The Huarochirí Manuscript.
40. Duviols, “Huari y Llacuaz.”
41. Salomon and Urioste, The Huarochirí Manuscript.
42. Gose, “Segmentary State Formation.”
43. Hocart, Kings and Councillors, 161–6. An alternative but equally applicable defini-
tion of diarchy that distinguishes the young, vigorous, militarist, and foreign from
the old, luxurious, priestly, and indigenous can be found in Marshal Sahlins, Islands
of History (Chicago: University of Chicago Press, 1985), 90.
44. Gose, Invaders as Ancestors, 62–63.
45. Ximena Medinaceli, “Paullu y Manco ¿Una diarquía Inca en tiempos de Conquista?,”
Boletín del Instituto Francés de Estudios Andinos 36, no. 2 (2007), 241–58.
46. Gose, Invaders as Ancestors, 84.
47. José de Acosta, Historia natural y moral de las Indias (Madrid: Historia 16, 1590
[1987]), 420.
48. Betanzos, Suma y Naración, 291; Pedro Pizarro, Relación del descubrimiento y con-
quista de los reinos del Perú (Lima: Pontificia Universidad Católica del Perú, 1571
[1978]), 91.
49. Molina, Relación, 74, 76.
50. Sarmiento, Historia, 174, 205, 250–1, 265–6, 268.
51. Betanzos, Suma y Naración, 291–2; Sarmiento, Historia, 253.
52. R. Tom Zuidema, The Ceque System of Cuzco (Leiden: Brill, 1964), 110, 126–8, 245–6;
Pierre Duviols, “La dinastía de los Incas: ¿Monarquia o diarquia? Argumentos
 Im m an en ce i n t h e An des ( 10 0 0 –170 0 ce) m 71

heurísticos a favor de una tesis estructuralista,” Journal de la Société des Américan-

istes LXVI (1979): 67–83; Pierre Duviols “Algunas reflexiones acerca de las tesis de
la estructura dual del poder incaico,” Histórica IV, no. 2 (1980): 183–96; Franklin
Pease, Los ultimos Incas del Cuzco (Lima: P. L. Villanueva, 1981), chapter 1; María
Rostworowski de Diez Canseco, Estructuras andinas del poder (Lima: I.E.P, 1983),
chapter 5.
53. Betanzos, Suma y naración, 51–52, 149–50, 181–2; Cieza, La crónica del Perú, 2,
chapter 7.
54. Betanzos, Suma y naración, 182–3; Pedro Sarmiento de Gamboa, Historia de los Incas
(Buenos Aires: Biblioteca Emecé, 1572 [1942]), 238.
55. Sarmiento, Historia, 191–2.
56. Betanzos, Suma y naración, 292–4, 299; Sarmiento, Historia, 253.
57. Peter Gose, “The Past Is a Lower Moiety: Diarchy, History, and Divine Kingship in
the Inka Empire,” History and Anthropology 9, no. 4 (1996): 383–414.
58. Isabel Yaya, The Two Faces of Inca History: Dualism in the Narratives and Cosmology
of Ancient Cuzco (Leiden: Brill, 2012), chapter 2.
59. Louis Dumont, Homo Hierarchicus (Chicago: University of Chicago Press, 1970).
60. Cristóbal de Molina (el almagrista), Relación de muchas cosas acaescidas en el Perú
in Biblioteca de Autores Españoles, vol. 209 (Madrid: Ediciones Atlas, 1553 [1968]),
75–76.
61. Cieza, La Crónica del Perú, 2, chapter 30.
62. Blas Valera, Relación de las costumbres antiguas de los naturales del Pirú, in Biblioteca
de Autores Españoles, vol. 209 (Madrid: Ediciones Atlas, 1590 [1968]), 161.
63. Zuidema, The Ceque System, 111–3, 156, 161.
64. Gose, “The Past,” 402–3; Yaya, Two Faces, 88–9.
65. Gose, “Segmentary State Formation.”
66. Molina, Relación, 82–3.
67. Richard Seaford, Money and the Early Greek Mind: Homer, Philosophy, Tragedy
(Cambridge: Cambridge University Press, 2004); David Graeber, Debt: The First
5000 Years (Brooklyn: Melville House, 2011).
68. For a good recent summary of debate on this subject, see Kenneth Hirth and Joanne
Pillsbury, “Redistribution and Markets in Andean South America,” Current Anthro-
pology 54, no. 5 (2013): 642–47.
69. Frank Salomon, The Cord Keepers: Khipus and Cultural Life in a Peruvian Village
(Durham, NC: Duke University Press, 2004).