Rust Family Foundation: Archaeology Grants Program

The Tarvagatai Valley Project

Principle Investigator: Jargalan Burentogtokh, Anthropology Department, Yale University

Importance of the Project

The Tarvagatai Valley Project was initiated in 2010 as a collaboration between the Mongolian Institute of Archaeology and Department of Anthropology, Yale
University. The main objective of the project is to understand how and why social complexity was formed in mobile pastoral societies through archaeological field research and various analyses. The field research consist of full-pedestrian survey and excavations of specific sites. A special attention was given to identify and study habitation sites of early herders as we still know very little about their day-to-day lifeways. This is especially true for the studies of the Late Bronze Age (c. 1450-700 BC) when major economic and social transformations took place in the Mongolian steppe. Towards the end of the second millennium BC, dependence on domestic animals increased significantly as well as the construction of large stone monuments known as khirigsuurs. 

Fig.1: The Teshig research area in the Tarvagatai Valley.

This research provides evidence to evaluate various models for function of these khirigsuurs.  The current dominant model argues that these monuments were used by elite rulers to legitimize their status. The investigator, however, argues that Mongolian society at this time was not stable enough to produce elite rulers and instead interprets khirigsuurs as monuments that were built by herding communities to stabilize the social landscape where low population density along with increased movement and interaction was becoming the norm. In order to manage pastoral resources among the dispersed herders, these monuments were also used as landmarks that represented a community’s presence and cohesiveness to outsiders.

Previous Research

The Mongolian khirigsuur monuments began attracting the attention of archaeologists in the early 1990s following many decades in which they were 
considered relatively uninteresting in comparison to later period burials with rich assemblages. Mongolian and international researchers initially worked on establishing khirigsuur chronology, functions, and spatial arrangements (Tseveendorj et al. 2003). From this early work a series of interpretations were proposed to explain how and why steppe people began construction of such large-scale monuments. In order to test competing hypotheses for the emergence of khirigsuur constructions, the Teshig Valley Project was initiated in 2010 as a collaboration between Yale University and the Mongolian Institute of Archaeology.  Prior to 2016 this project surveyed a total of 43 km2, documented 108 khirigsuurs and 30 habitation sites, and opened test excavations of 2 habitation sites (Galdan et al. 2011, 2012, 2013; Jargalan et al. 2014, 2015).

One of the largest khirigsuurs in the valley (found in previous survey) with most of its satellite features on the eastern side.

Project Funded in 2016 (RFF-2016-13)


The main goal of the project was to gather spatial data of khirigsuur monuments in order to analyze their spatial distribution and their relationship to the pastoral landscape.  The excavation of khirigsuur satellite features would follow two main objectives. First, to collect samples for radiocarbon dati
ng from a single monument complex in order to determine if it was built over long period or was a result of a single construction campaign.  Second, the collect samples of faunal material (horse bones) for isotope and digital scanning analysis.  Both of these analyses aim to provide more information about the nature of pastoral activities during the Late Bronze Age, including long-distance interaction. The isotope analysis of horse skeletons should reveal if any of the horses were brought from another region with a different environment and climate. Digital scanning of horse skulls could reveal if any of them were bridled. This may indicate to what extent the Late Bronze Age population was engaged in horseback riding; the introduction of horseback riding would have opened a completely new world of possibility to Mongolian society.


One of the specific objectives for the 2016 fieldwork was to conduct pedestrian survey in the last section of the survey area and record the monuments. Habitation sites were also identified by using augers during survey (Gardner and Burentogtokh, forthcoming). In a place with minimum ground visibility, this methodology has proven to be quite successful (almost 70% of our habitation sites were identified solely from auger probes).

Results of 2016 Research

Survey:  The Tarvagatai Valley is a relatively independent valley system that is surrounded with mountain ridges to the north and the south. The pedestrian survey in 2011-2013 recorded over a hundred khirigsuur monuments and 102 habitation sites from different time periods, indicating an active presence of early pastoral communities in the valley. As supporters of the elite model argue, khirigsuurs were usually built in clusters and most daily activities would have taken place in close proximity to them. These supra-local communities were in competition with other valley communities over pastoral resources and therefore the monumental landscape was made of monument clusters and ‘buffer-zones’ in between them. In other words, members of the upper social strata were focused on establishing their social status on the local level through monument construction and ceremonial rituals.

The monument distribution in Tarvagatai Valley tells a different story. Rather than organized in clusters, khirigsuurs are equally distributed across the Tarvagatai Valley and are constructed at the most visible places: in the middle of the wide, bottom valleys or at the mouth of the side valleys. Larger monument complexes were built at the crosspoint of several major side valleys, representing the labor of the occupants of these valleys. Relatively smaller khirigsuurs were often built at the upper sections of the side valleys, indicating the presence of a relatively small part of the community, perhaps several families.

The pedestrian survey and sub-surface tests near the large khirigsuur complexes did not find any evidence of consistent activity around these sites. In order to test the distribution hypothesis at intra-valley level, reconnaissance was conducted between the Tarvagatai and Teshig Valleys. Teshig Valley is another relatively independent river valley system with a large concentration of khirigsuurs.  In 2016 the reconnaissance covered a total of 16 km2 and recorded 32 khirigsuurs in the space between these two major valleys. These monuments were built in similar locations that are highly visible for passersby. Each monument was photographed, measured, and included in the general monument database.

Excavation: A total of nine satellite features from three khirigsuur complexes were excavated, all of which yielded remains of sacrificed horses. Samples for radiocarbon dating have been prepared for shipping to the US but dates have not yet been obtained.

Isotope and digital scanning analysis:  Most of the horse skulls were poorly preserved. Only three skulls qualified for digital scanning analysis. Presently, samples for both analyses are being prepared at the Mongolian Institute of Archaeology by a professional zoologist and will be shipped to the US and Germany as soon as export permissions are issued.

Fig.3: Skull of a sacrificed horse (arrow) placed in one of the satellites excavated in 2016. This is one of the relatively well-preserved skulls and will be used for  analysis.

The research from this project enables us to better understand how monumental traditions emerged and developed in mobile pastoral societies and clarify what were the preceding social and economic foundations of it.


Tseveendorj, D., Bayar, D., Tserendagva, Ya., Ochirkhuyag, Ts. (2003). Mongolyn Arkheologi [Archaeology of Mongolia]. Academy of Sciences, Ulaanbaatar.

Galdan, G. and Jargalan, B. (2011-13). Teshig hundiin arheologiin sudalgaani tailan [Report on archaeological research in Teshig valley]. Report submitted to the Institute of Archaeology, Mongolian Academy of Sciences, Ulaanbaatar.

Gardner, W. and Burentogtokh, J. Forthcoming. iSurvey: Systematic survey and electronic data collection on the Mongolian steppe. Journal of Field Archaeology.


Recent Foundation grants: general        Archaeology Grants Program w/map  

Recent Foundation Grants in archaeology

Copyright  ©  2018 Rust Family Foundation