“Serfs” on the Roof of the World: The Importance of Terminology in Discussions of Politically Sensitive History

By Contributing Editor Kristin Buhrow

Most commonly associated with Medieval Europe, the term “serf” is commonly used to describe a certain type of peasant class with a particular set of living conditions.  From the European Middle Ages, academic discourse in English has imposed the term “serf” onto populations of landless peasants in nineteenth century Russia (Melton, 1987), early twentieth-century East Africa (Huntingford, 1931), and beyond.  This essay will explore the application of the term “serf” to peasants of a particular social and political context: the mi ser class within Pre-Modern Tibet, the political and social effects of the modern use of this term, and the importance of recognizing the politicized use of terminology in academic and political discussions more generally.


Farmland outside of Meldro Gungkar, Tibet.  August, 2017. Photography by Kristin Buhrow.


While each of the cultural contexts mentioned above exhibits a different sort of peasant class affected by unique conditions, resources, and norms, the designation of these as “serf” classes is dependent upon the existence of an interminable relationship with a particular member of the higher class—a “lord”. According to Melvyn Goldstein’s multiple treaties on the application of the term “serf,” a class’ designation as such indicates a “hereditary superordinate-subordinate relationship in which the subordinate possesses a legal identity independent of the superordinate” (Goldstein, 1971 p.522), the subordinate’s lack of legal right to terminate the relationship (Goldstein, 1971 p. 522), and “a degree of judicial control” by the superordinate over the subordinate (Goldstein 1986 p. 82).  Whatever the other conditions experienced by a peasant class, these criteria serve to determine which social classes may be designated a “serf” class (as opposed to a slave or more general peasant class) regardless of culture.

Based on the principles above, some academics and political personages have recognized Pre-Modern Tibet (1642-1959) a culture practicing serfdom.  More complex than a simple array of lord and serf relationships, however, the taxonomy of Pre-Modern Tibetan social classes involves several branches.  Firstly, Pre-Modern Tibet might be divided into two social classes: the monastic population and the laity. Focusing on the laity, this category could be divided into two clear classes:  the landed aristocracy, sger pa, and the peasants, or mi ser[1] (Goldstein, 1971 p.522-3).  Despite both sger pa and mi ser contributing children to the monastic population, those who remained part of the laity were obligated to fulfill unequal social roles.  Far outnumbering the sger pa, who constituted only about 200-350 aristocratic families, the mi ser were a more diverse population in terms of wealth, lifestyle, and location on the Tibetan plateau.

While responsive to a particular sger pa family, the condition of the Tibetan mi ser diverges with that of other global peasantries classified as serfs with regard to their ability to challenge their social superiors, accumulate resources, and even change locations.  Unlike their feudal European counterparts, Tibetan mi ser were not only able, but publicly encouraged to appeal to other members of the aristocracy and even the central court system to challenge or prosecute nefarious or unjust members of the aristocracy (Goldstein, 1971 p. 523; Bischoff, 2013 p. 12). This access to the court system and the influence of other aristocrats allowed mi ser a more powerful role in their social relationships than “serfs” in other cultural contexts.  In addition to providing peasants with prosecution rights, premodern Tibetan lords were less powerful than their medieval counterparts because they could not unilaterally alter the amount of taxes required or the amount of land serfs held (Goldstein, 1971 p.522), lords often provided minimal oversight so long as corves were met (Goldstein, 1971 p. 526). Perhaps related to a standard of minimal oversight, Tibetan mi ser did not necessarily live in destitute conditions.  The status of mi ser then, were not analogized to serfs because of high levels of poverty or total incapability to challenge the will of their lords; instead, mi ser have been analogized to serfs due to a lifelong bond to the land of one aristocratic family, extending to the lives of their descendants[2]  for multiple generations.

Within the social category of mi ser is an allowance for an alternative lifestyle which further complicates the discussion of mi ser as land-bound serf: the geographically mobile mi bogs. Translated to “human lease,” mi ser who obtain the status of mi bogs were leased to work on other estates.  This status was usually taken on voluntarily by the mi ser with approval from their sger pa landowner to allow for more freedom as to geographic location as well as the ability to travel.  Some Tibetologists claim that the existence of a status like mi bogs indicates that mi ser were free to travel, not bound to the land of a sger pa family, and therefore not able to be categorized as serfs at all (Alice Travers going as far as to assert that mi bogs were “a masterless sect of people—people without land, without master, and without a servant” [2013 p.151]).  However, while the availability of the mi bogs path does differentiate the experiences of Tibetan mi ser from other groups of “serfs,” it is important to note that even while away, mi ser still had to respond to their original sger pa family’s commands, pay an annual fee to the sger pa, travel where the sger pa commanded, and return home upon his order (Goldstein, 1986 p. 94). As such, even with the allowance of a certain number of mi ser to attain the status of mi bogs and live away from the immediate lands of the sger pa, the social relationship between mi ser and sger pa remained not only extant, but influential.  Even in cases of geographic removal, mi ser were still responsive to the owners of the lands they were obligated to work.  In this context, the term “serf” may render a fairly accurate picture of certain aspects of the lives of mi ser in Pre-Modern Tibet.

Despite its application to the Tibetan peasants of more than sixty years ago, the modern description of mi ser as a “serf” class has functioned in support of a narrative condoning the communist party’s occupation of and Sinicization efforts within Tibet.  Since entry of the communists in1959, Chinese political publications have emphasized the parallels between the mi ser sger pa relationship and the serf – lord relationship. This emphasis includes making direct reference to the mi ser population as “serfs” in English language publications. Taking one popular example, the word “serf” occurs some thirty-five times in the 2001 edition of the English language Chinese government public relations leaflet “100 Questions on Tibet” (Barnett, 2008 p. 81).  While it is true that some aspects of the Pre-Modern Tibetan peasant experience allow the mi ser to be categorized alongside serfs according to loose, culturally relative definitions, it is also likely that associating Pre-Modern Tibetan peasants with a historical system known for oppression of laborers was a conscious choice by the Chinese Communist Party to illicit support for the seizure and Sinicization of Tibet.

This support from admirers of the Chinese Communist Party and Communists worldwide was sought by appealing to a Marxist perspective of linear societal development.  In the Marxist view, a feudal society is considered less “evolved” than an industrial society; this perspective allows communists to deduce that any cultural operating on a system similar to feudalism is inferior to the more developed China.  While a necessary stepping stone to reach the higher evolutions of social structure, the stark differences between the landed and laboring classes in feudalism render the system, in the Marxist eye, “inseparable from extreme abuse” (Barnett, 2008 p.82). This narrative of Pre-Modern Tibet as a society characterized by violently abusive inequality is propagated to justify its seizure of Tibet and excuse some of the human rights abuses which take place there today.  As noted by Travers in her 2013 article, this Marxist narrative has been reinforced by the Chinese Communist Party and Western observers with communist backgrounds since 1959 (p. 143).  Understanding this politicization of historical social hierarchies, while the term “serf” is technically accurate, its use legitimizes a construction of Tibetan history which marginalizes traditional Tibetan norms and allows the Communist Party to take the role of a savior society.

In the quest for honest and fruitful discourse, it is important that all discussants be aware, not only of the historical realities of the past, but also the political context of the present.  A responsible contributor must therefore be able to earnestly describe the likely difficult conditions for life among Pre-Modern Tibetan mi ser and must acknowledge that mi ser did live within a system which could be easily abused by their sger pa counterparts.  Simultaneously, an informed discussant can not act as if openly associating the conditions of the mi ser with medieval serfdom does not have a history of politicization.  Under conditions such as these, academic and political discussions must remain open, with the choice to analogize the mi ser with serfdom available for those who have no qualms with the modern connotations.  Likewise, the choice to refrain from making the analogy or to argue against the appropriateness of the analogy should be regarded as a form of academic activism by those who wish to knowingly partake.

[1]Other variations on these Tibetan terms include sku drag for aristocracy and smad rigs for peasants (Travers, 2013 p.141).

[2] Descendants of peasants were ascribed through parallel descent, with sons going to their father’s lord, and daughters go to mother’s lord (Goldstein, 1971 p. 522).


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s