Posted by: Christian | May 6, 2007

Afghanistan and the Qawm: An Important Yet Unknown Concept

May 7, 2007.

You must know this word: qawm (also transliterated as qaum or qowm). Apparently anthropologists are hoarding this word and not wanting to share it with others. But I’ll set it loose on the interweb so that the google may find it. But first…

If you believe what you read in the media and what you see and hear on television then the fault lines and loyalties in Afghanistan are obviously apparent: ethnic groups and Islamic sects. Maybe if the media is feeling smart they will discuss Pashtun tribal loyalties as well. However, I’ll deal here primarily with attacking the idea of ethnic loyalties being a strong determining factor for mobilization and social organization on the ground in Afghanistan.

Olivier Roy argues against assuming that all members of an ethnic group defined by its spoken language actually share a coherent identity with a “will to express themselves politically.” Many others agree that loyalties are strongest within local communities, not at a national or ethnic level. Roy concedes that ethnic identities are important but argues that “primordial” local identities take precedence (Roy 2002: 4). These local identities are usually called qawm:

A qawm is the term used to describe any segment of society bound by solidarity ties, whether it be an extended family, clan, occupational group or village. Qawm is based on kinship and patron-client relationships; before being an ethnic of tribal group, it is a solidarity group, which protects its members from the encroachments of the state and other qawm, but it is also the scene of internal competition between contenders for local supremacy (Pierre Centlivres, Olivier Roy, and Whitney Azoy quoted in Roy 1989: 71).

Nazif Shahrani describes the importance of qawm in the mobilization process [and note the different linguistic terms used to describe a similar concept]:

Ethnicity and kinship, which are expressed linguistically through the same terms, qawm (people, tribe, group), wulus (nation, tribe, relatives), and tyfah (clan, tribe, group), represent the same or similar ideological frameworks in Afghanistan. Together with Islam, they provide the most fundamental bases for individuals and collective identities and loyalties, and they are the most persistent and pervasive potential bases for the organization of social formations, for the mobilization of social action, and for the regulation of social interaction among individuals and between social groups […]” (Shahrani 2002: p. 717).

Shahrani notes that local loyalties and responsibilities between leader and follower are not static but rather change according to such circumstances as “shifting boundaries” and factional struggles (Shahrani 1998: pp. 219-20. See also Dorronsoro 2005: p. 111).

Richard Tapper also cites the flexibility of the qawm:

According to context and situation, qawm may involve a varying number of individuals, close kinsmen, a village, an ethnic group, a religious sect or a linguistic group. It is therefore a highly ambiguous and flexible concept allowing for strategic manipulations of identity (Tapper 1988: p. 27).

Changes in economic conditions can also cause realignments in qawm structure. The change in Afghanistan towards a modern market economy has lessened the importance of genealogical relations and increased the significance of patron-client economic relations, encouraging new qawms to appear based on patronage networks (Rasuly-Paleczek: 2001: p. 152; Roy 1995: p. 108). Roy, Shahrani and Tapper all give definitions for qawm that demonstrate not only the importance of the qawm, but also its flexible nature. This clearly has significance when considering the possibility of strategic manipulation of identity, as well as the shifting of individual and group loyalties from one identity to another.

During the Soviet-Afghan conflict qawm affiliations became more relevant than in the preceding decades. For example, it became common for a qawm leader to make a deal with a government militia and bring all his followers (Sinno 2002: p. 169). This happened more among Uzbeks since the “centralized” and “undemocratic” nature of the Uzbek qawms ensured that members would willingly join and not dissent from their leader’s decision (Sinno: p. 187; Giustozzi 2002: p. 128). Similar recruitment tactics among Tajiks and Pashtuns failed since they would not follow their leaders into the government forces as readily as Uzbeks (Sinno 2002: p. 190). Although at the beginning of the war local commanders used their local qawm to mobilize, at a later point their power bases expanded geographically and gave political significance to larger identities (Dorronsoro 2005: pp. 20, 211. Dorronsoro gives the example of the Tajik Panjshiris). This process of consolidation encouraged many qawms to identify themselves as part of a larger macro-ethnic identity in order to associate with a stronger group and thus acquire political representation in national politics (Roy 1990: p. 224).

Afghans usually will identify themselves by their qawm. Roy states that when they identify themselves by the language they speak they do so “without any ethnic connotation” (Roy 1995: p. 24; Roy 2002: p. 4). Gilles Dorronsoro argues that macro-ethnic identity (Uzbek, Pashtun, etc.) is too encompassing to be used as a mobilizing framework, meaning that appeals by ethnic entrepreneurs to mobilize to protect one’s own ethnic group are likely to be ineffective (Dorronsoro 2005: pp. 108-9, 258). Robert Canfield claims that people are aware of their broader macro-ethnic identity but it is the kin networks and patron-client networks that are more important to the people and that form cleavages within and across the ethnic group identities (Canfield 1986: p. 76). Yet despite the limitations to ethno-political mobilization in Afghanistan, an Uzbek-dominated faction emerged as a result of the mobilization process as qawms and individuals attached themselves to larger, more powerful units. Roy and Dorronsoro concede this point 15 years apart (Roy 1990: p. 224; Dorronsoro 2005: pp. 20, 211. The emergence of a broad Uzbek identity is discussed by Gabriel Rasuly-Paleczek. Rasuly-Paleczek 2001: 151-2, 161, 174, 176). But that was during civil war and now what of the Uzbeks? They seem to be not operating as a coherent group with a single strategy. Same as the Pashtuns, Tajiks, Hazaras, etc…

Conclusion: Afghanistan is not Europe in terms of ethnicity. The phrase “all politics are local” can be applied to Afghanistan with much accuracy. While the importance and awareness of ethnicity increased during the 1980s and 1990s, ethnicity never became the sole driving focus of people’s loyalties and social organization. These “Afghanistanis” are just so pragmatic.

PS: Yes. I occasionally try to pawn-off parts of my academic research in the form of blog entries.


Azoy, Whitney. 2003. Buzkashi: Game and Power in Afghanistan. Waveland Press.

Canfield, Robert L. 1986. ‘Ethnic, Regional, and Sectarian Alignments in Afghanistan’ in The State, Religion, and Ethnic Politics: Afghanistan, Iran, and Pakistan. Banuazizi and Weiner (Eds.) Syracuse: Syracuse University Press.

Dorronsoro, Gilles. 2005. Revolution Unending: Afghanistan, 1979 to the Present. New York: Columbia University Press.

Giustozzi, Antonio. 2002. ‘Afghanistan: The Problems of Creating a New Afghan army- and the critical dangers of failure!’, International Industrial Information (April 2002).

Rasuly-Paleczek, Gabriele. 2001. ‘The Struggle for the Afghan State: Centralization, Nationalism and their Discontents’, in Schendel and Zurcher (editors) Identity Politics in Central Asia and the Muslim World. London: I.B. Tauris Publishers.

——- 1998. ‘Ethnic Identity versus Nationalism: The Uzbeks of North-Eastern Afghanistan and the Afghan State’, in Atabaki and O’Kane (editors), Post-Soviet Central Asia. London: Tauris Academic Studies.

Roy, Olivier. 2002. ‘Afghanistan: Internal Politics and Socio-Economic Dynamics and Groupings’, WRITENET Paper, No. 14.

—— 1995. Afghanistan: From Holy War to Civil War. Princeton, NJ: Darwin Press.

—— 1990. Islam and Resistance in Afghanistan. Cambridge: Cambridge University Press.

—— 1989. ‘Afghanistan: Back to Tribalism or on to Lebanon?’ Third World Quarterly, Vol. 10, No. 4. October 1989, pp. 70-82.

Shahrani, M. Nazif. 2002. ‘War, Factionalism, and the State in Afghanistan’, American Anthropologist, Vol. 104, No. 3. September 2002, pp. 715-722.

——- 1998. ‘The Future of the State and the Structure of Community Governance in Afghanistan’, in Maley (editor) Fundamentalism Reborn? Afghanistan and the Taliban. New York: New York University Press.

Sinno, Abdulkader H. 2002. Organizing to Win in Afghanistan and Beyond: How Organizational Structure Affects the Outcome of Strategic Interaction in Politicized Group Conflicts. Ph.D. Dissertation in Political Science, September 2002, UCLA.

Tapper, Richard. 1988. ‘Ethnicity, Order, and Meaning in the Anthropology of Iran and Afghanistan’, in J.-P. Digard (Ed) Le Fait Ethnique en Iran et en Afghanistan. Paris: Editions du CNRS.

Note: If you read this far, from top to bottom, then you have won a free cookie. Congratulations. Redeemable wherever free cookies are being distributed by someone else.


  1. […] 7th, 2007 by Ronin Afghanistanica has an interesting discussion of the term “qawma.” It’s a loose term that may imply physical similarity […]

  2. […] discusses “qawm,” a term for bounded social groups of various types, and what this means for identity in Afghanistan. Share […]

  3. […] is the word “qawm” which comes from Afghanistan. There are ethnic and tribal and local rivalries in Afghanistan. Often […]

  4. […] You can read the full discussion on the qawm at Afghanistanica. […]

  5. […] Generally I’m talking about those at a much lower level on the village or extended village or qawm level (and therefore usually more accountable and sensitive to local […]

  6. […] seems essential to comprehending the intricate relationships of the myriad peoples of Afghanistan. Oliver Roy defines a qowm as the term used to describe any segment of society bound by solidarity ties, […]


%d bloggers like this: