Continuity and Change in the Cult of the Dead: Shaping Kin Formation in the Peruvian Andes (AD 1000–1600)

Aim and setting

This article reports excavations of funerary contexts at Marcajirca, a large mortuary complex that was also a ceremonial and village settlement in the Puccha Basin of the Ancash highlands, part of the Conchucos region of north-central Peru. The site was occupied from about AD 1000 to 1600, spanning the Late Intermediate Period (LIP), the Late Horizon under the Inca, and the early Colonial period. The study asks what its tombs reveal about the people buried there, how burial changed over time, and how the evidence relates to the ayllu, the kin-based community that organized highland society at Spanish contact.

Regional background

Occupation in the basin runs from about 1200 BC to AD 1600. Chullpas, the aboveground tombs central to the study, first appear with the Recuay culture around AD 350, alongside underground tombs and cists, while funerary caves are absent in that period. In the LIP, large Recuay towns gave way to many small hilltop villages of two to four hectares, yet tombs did not move with them. Survey recorded settlements with tombs of several forms, settlements with no burials, and isolated cemeteries of mainly caves. Ethnohistorical sources place two groups in the area, the Huari north of the Puccha River and the Pincu to the south, each associated with a sacred mountain, Llamoq and Wiñaj respectively.

The site and the excavations

Marcajirca stretches along a ridge on Llamoq's southern slope. Its core covers about four hectares and has three sectors: a southern residential sector of 43 houses, a central funerary sector, and a northern ceremonial sector containing ushnu platforms, an amphitheater, a large circular building called the Torreon, and a standing stone, or huanca. After surveys in 1999–2000, excavations ran from 2005 to 2014, alongside conservation and bioarchaeology programs. Of 61 excavation operations, 25 targeted funerary contexts: 17 at chullpas, seven at caves, and one at a rock-shelter tomb. The site holds 37 chullpas and 22 caves. Roof timbers and stucco straw were dated to establish construction, and human teeth to establish duration of use. A detailed inventory of the remains was made on site, and the bones were reburied out of respect for the community.

Chullpas, caves, and other tombs

The chullpas average about 3 by 2.5 meters, are mostly single-story, and have small elevated doorways of roughly 40 by 50 cm, suggesting bodies were placed as compact bundles. Roofs are largely conical and built by corbelling. Masonry falls into two traditions: walls finished with chinking stones, a technique inherited from Recuay builders, and walls covered with painted stucco. Chinking-stone chullpas tend to be earlier. Doorways face several directions, but twelve face Llamoq, as does the largest chullpa, built in the Inca period. Excavations in front of chullpas found hearths at eight of seventeen, with burnt camelid bones, matching chroniclers' accounts of food being burned for the dead. Courtyard walls postdate the chullpas, which the author reads as a response to a growing need for private ritual space, with ceremonies later led by direct descendants in smaller numbers.

The caves, concentrated within the core, use natural outcrops and boulders, with walls added at entrances. Some also had courtyards, but caves and chullpas never shared one, implying that different groups oversaw their rituals. The single rock-shelter tomb lies beneath rock art and resembles tombs in isolated valley cemeteries. Two underground burials are unusual in the highlands. They hold 12 and 48 articulated individuals of all ages, with no offerings, and they look like mass graves, possibly from an epidemic between Inca rule and Spanish arrival.

Human remains

Bones inside the tombs were commingled and disarticulated, and the minimum number of individuals across excavated tombs totals 620. Chullpa 3, with a floor of about 5.5 square meters, held 66 individuals, indicating repeated burials over generations: older remains were rewrapped and moved aside to make room for new bundles. The author documents 33 modified crania, including annular erect, annular oblique, and bilobular forms, plus unintentional flattening from cradling. Cave 19 contained crania of every type, and dates from three individuals overlap, so modification differed within one group. The author therefore argues that skull shaping here did not mark ethnic identity, challenging a common assumption.

Objects

Grave goods include stone clubs, gourd vessels, spindle whorls, pierced marine shells including Spondylus, copper discs, a miniature Inca-style wooden kero cup, a llama-bone flute, bone snuffers suggesting hallucinogen use, and needles. Caves and chullpas held similar goods, indicating no marked status differences among the buried.

Chronology

Nineteen radiocarbon dates place the main occupation between about AD 1000 and 1470. Chullpas appear slightly earlier than caves, consistent with the chullpa tradition inheriting from Recuay, and Cave 19 was used for roughly four centuries. Inca influence appears in the largest chullpa, rectangular buildings, and Cuzco-style ceramics, yet did not fundamentally change funerary practice.

Discussion: ayllus and ancestors

The author defines the cult of the dead as the actions the living perform for the dead, through tomb construction and periodic ritual, to keep them present in daily life. Strontium isotopes from the remains form three clusters tied to different parts of the valley, carbon isotopes show little maize consumption, tomb masonry reveals at least three builder groups, and the rock-shelter tomb suggests another. Together these point to at least three ayllus using Marcajirca as a regional burial place, possibly including people from Pincu territory. Hearths in front of chullpas are read as evidence of cults to ancestors known as mallquis, and the huanca in the amphitheater, built after the chullpas, suggests growing ceremonial activity centered on a founding ancestor. Llamoq is interpreted as the mythical ancestor of the community.

Conclusions

Marcajirca's variety reflects several groups burying their dead together. The key change from the preceding period is the move from burial outside residential areas to tombs embedded in houses and courtyards, signaling the cult of the dead as a strategy of social cohesion. Individuals within chullpas were probably biologically related, as inherited conditions in the bones suggest. Whether chullpas marked territory remains unresolved. The author suggests ayllus kept houses at Marcajirca for rituals and as a base for pilgrimage to Llamoq.