Focused Series »

Indo-European Origins
Northern California
The Caucasus
Imaginary Geography
Home » Borders, Central Asia, Linguistic Geography, Myth of the Nation-State, South Asia

The Complex and Contentious Issue of Afghan Identity

Submitted by on November 19, 2011 – 7:04 pm 26 Comments |  
Nigel Allan's Map of Babur's Use of the Term "Afghanistan"“Afghanistan” is an oddly constructed place name. It is usually said to be a Persian word meaning “land of the Pashtuns.” The widely used suffix “stan” is Persian for “place of” or “land of,” cognate with the English “stead” (as in “homestead”) and ultimately with “stand.” “Afghan” is usually considered synonymous with “Pashtun.” From the Pashtun perspective, “Afghanistan” is an exonym, a geographical term of foreign origin. For Pashto-speakers to call their country “Afghanistan” would be a bit like Germans calling their country not Deutschland but Germany, or the Japanese calling theirs Japan instead of Nihon or Nippon. But Pashtun-speakers in Afghanistan, unlike German speakers in Germany, do not form the majority linguistic community. In Persian (or Dari), the country’s most widely used language, “Afghanistan” is native term, but one that refers to a different people and to some extant a different place. (In Pashto, the official name of the state is Da Afġānistān Islāmī Jomhoriyat, translated into English as “the Islamic Republic of Afghanistan”; in Persian it is Jomhūrī-ye Eslāmī-ye Afġānistān.)

The identification of “Afghan” with “Pashtun,” however, turns out to be a knotty issue. Geographer Nigel J. R. Allan of the University of Nevada at Reno has examined this subject in detail. The earliest use of “Afghanistan” that he has found is in the Baburnama of the early 1500s, the autobiographical story of Babur’s creation of the Mughul Dynasty of northern India. In this account, “Afghanistan” denotes a limited area south of Peshawar in what is now northwestern Pakistan. Subsequently, “Afghan” came to denote a handful of Pakhtun/Pashtun tribes living in and around the Vale of Peshawar. The designation was gradually generalized to cover all Pashtun people, but the idea lingers that the “real Afghans” are still those of Peshawar, arguably the most important Pashtun city (although the multi-ethnic metropolis of Karachi now has the largest Pashtun population). With the creation of the modern state of Afghanistan, “Afghan” was extended to cover all residents of the country, regardless of their language or ethnicity. Allan associates this usage with 19th century British imperial agents. As a place name, “Afghanistan” is thus both exonymic and geographically displaced, having originally denoted an area outside of the borders of the modern country of that name.

To be sure, Allan’s narrative would be challenged by a few Afghan nationalists, who see earlier versions of “Afghan” in historical place names such as “Abgan.” A few have gone so far as claim a geo-historical essence for the Afghans and their country. In 1969, Abdul Hai Habibi argued that, “The word Afghan … represents an indivisible unit under all historical, economic and social conditions in the heart of Asia … with a historical background of one thousand and seven hundred years.” Yet, as usually the case, such an insistently nationalistic interpretation twists the past in accordance with modern-day dreams of national unity. The result leans more toward wishful thinking than scholarly analysis.

Allan’s finds similar complexities with other ethnic designations used in Afghanistan. Most texts and maps divide the country into roughly a dozen ethnicities. As the Wikipedia puts it, “The ethnic groups of the country are as follow: Pashtun, Tajik, Hazara, Uzbek, Aimak, Turkmen, Baloch, Pashai, Nuristani, Arab, Brahui, Pamiri and some others.” The term “some others” indicates uncertainty, which is indeed warranted. Following the Ethnologue and Erwin Orywal, Allan discerns forty-five languages and fifty-five ethnic groups in Afghanistan. Several conventional groups turn out to be composite units composed of distinct peoples speaking separate languages; “Nuristani,” for example, encompasses five languages. “Tajik” is an especially a fraught category. It too is foreign, derived from the Turco-Mongol term for “non-Turk.” Although it has long been used by Persian speakers for self-designation, “Tajik” retained pejorative connotations until recent decades. The institutionalization of the term has been linked to the Soviet manipulation of ethnic categories in Central Asia. As Allan shows, most of the current ethnic designations of Afghanistan were imposed by outsiders on the basis of limited knowledge. The resulting scheme ignored the indigenous concept of manteqa, used to divide most of Afghanistan into ethno-geographical units. As the Wikipedia puts it, “In Afghanistan, the Tajiks … refer to themselves by the region, province, city, town, or village they are from; such as Badakhshi, Baghlani, Mazari, Panjsheri, Kabuli, Herati, Kohistani.”

            As a term of self-designation, “Pashtun” stands on stronger grounds than “Tajik. “Pashtun identity tends to be pronounced, and is often a source of considerable pride. But the mere fact that this ethnic group spans the border challenges the national formations of both Afghanistan and Pakistan. Pakistan also signals a claim to Pashtun territory by way of the word “Afghan.” The “A” in “Pakistan”—an acronymic country-name—refers to “Afghania,” just as the “P” refers to Punjab, the “K” to Kashmir, and the “S” to Sindh.

Map showing different definitions of the term "Pashtunistan"The Pashtun people themselves have on occasions hoped to break out of this geopolitical bind, proclaiming their own separate nationality and agitating for the creation of a new state. The proposed boundaries of an independent Pashtunistan, however, vary significantly, as can be seen in the maps posted to the left. In one version, Pashtunistan would be limited to the Pashto-speaking areas of both countries; in another it would add Pakistani Balochistan and a few other areas; in another it would encompass all of Afghanistan as well as the western half of Pakistan; and in yet another it would include only the western half of Pakistan. Although the idea of Pashtunistan is often considered dead, some maintain that the fear of its revival pushes policy in both Pakistan and Afghanistan. As stated recently in the anti-militarist World War Four Report, “Kabul and Islamabad both feel the need to appease Pashtun tribal leaders, fearing the specter of an independent ‘Pashtunistan’—which would take a critical chunk of both states’ territory, and widen the war yet further…”

As a final note, it has long seemed odd to me that Afghanistan is usually portrayed in the U.S. media as a Pashtun-dominated country, even though most of its residents belong to other ethnic groups, and even though Persian is more widespread and more prestigious than Pashto. Nigel Allan links this habit to U.S. diplomatic and military maneuvering in the region. In the 1950s and 1960s, almost all American developmental projects in Afghanistan focused on Pashtun areas. In the 1980s, the U.S. military embraced a “southern strategy” based on alliances with Pashtun militias aimed at expelling Soviet forces from the country. This Pashtun focus, Allan argues, stemmed in part from the British imperial designation of the Pashtun people as one of South Asia’s “martial races.” Whether the U.S. reliance on such Pashtun leaders as Hamid Karzai has been an effective strategy is a different matter altogether.

For more information, see Nigel J. R. Allan, “Defining Place and People in Afghanistan.” Post-Soviet Geography and Economics, 2001, 42(8), 545-560. (Note: the journal is now called Eurasian Geography and Economics.)

Previous Post
Next Post

Subscribe For Updates

It would be a pleasure to have you back on GeoCurrents in the future. You can sign up for email updates or follow our RSS Feed, Facebook, or Twitter for notifications of each new post:

Commenting Guidelines: GeoCurrents is a forum for the respectful exchange of ideas, and loaded political commentary can detract from that. We ask that you as a reader keep this in mind when sharing your thoughts in the comments below.

  • Robbie III

    Dr. Lewis,

    Interesting post, I think the Afghan/Pashtun question needs to be resolved in U.S. foreign policy circles. If we assert that they are equivalent then there needs to be a more realistic objective on the ultimate “look” of the Afghan state, meaning its authority is going to be problematic outside the eastern borderlands.

    Personally, I think Afghanistan’s way forward is toward a concept of “Afghan” separated from “Pashtun,” much like “American” has been divorced from the “original” “Anglo” colonists. An Afghan political identity centered around a multi-ethnic Afghan would, hopefully, be a bit more stable within the current boundaries.  Moreover, it may serve as an effective example for the rest of the region.

    This may be a bit optimistic.

    • Anonymous

      Afghanistan is a weak land locked country that borders China, Pakistan, Iran, Turkmenistan, Tajikistan and Uzbekistan.  The ethnic and linguistic groups living within those borders have done so for millennia.  

      The people of the USA (Americans) live in the worlds only superstate, with borders to 2 vassal states (Canada & Mexico) and access to two oceans. Most Americans are descended from migrants of the last 300 years (probably less) from all over Europe, much of Asia, Latin America and black African slaves.

      I don’t think we should project New World models onto ancient lands, and in this case to do so, would be an Orientalist erasure of the agency of the people of Afghanistan.

      • Robbie III

        You’re contradicting yourself. “Don’t project New World models” (you mean Old World Greek/Roman models of democracy?) in fear of erasing “the agency of the people of Afghanistan”?

        Further, you’re aware that the “borders” of Pakistan were set in the late 1940s? That the other Central Asian “stans” can thank the Soviet Union (I’d want to blame Mr. Stalin but this might be only applicable for the Caucasus) and that China annexed Tibet in the 1960s? I don’t necessarily disagree that that these groups have not been in this general area for “millennia”, but your containing borders certainly have not.My original point isn’t so much democracy but a re-envisioning of what it means to be “Afghan,” i.e. not purely Pashtun but also Tajik, Uzbek, Hazara, Turkmen, within the borders (such as they are). This is probably a more realistic goal than a remapping of the borders.

        • Interesting debate. I avoid making any policy recommendations, as that it not the point of GeoCurrents. I will say that there have been major efforts to “re-imagine Afghan national identity” as ethnically inclusive, encompassing all peoples of the country.  Such efforts have been partially successful, but only partially. I also agree that a “remapping of the borders” is highly unlikely. But existing borders are problematic, both here and in much of the rest of the world. As a result, a certain amount of tension and conflict seems inevitable. 

        • Anonymous

          Robbie, I didn’t mean geo-political models, I meant cultural models.  I was using New World to refer to countries that are largely inhabited by recent migrants or their descendants. 

          A better term for New World might have been Settler States and for Old World perhaps Established States – but I’ve never seen them used, because I just invented them – maybe the academy has some other terms – anyone?

          I’ve lived & worked in Established States (England, Slovenia & Japan) and Settler States (USA, Singapore and Australia).. A feature I’ve observed in Settler States is their ability to create harmonious multi-racial/ethnic/cultural/religious societies, which is a feature that most (all ?) of the Established States lack.

          So the point I was making is that you can’t project US, Singaporean or Australian multi ethnic/linguistic/culture models on to Afghanistan, England or Japan – its just don’t work.

          I also think the geo-political circumstances of Afghanistan are so divergent from the USA as to make comparison or projection foolhardy.  If you’d suggested Switzerland as a model, it might have been better. Imagining Afghanistan could become the Switzerland of Central Asia is a long stretch; but they have a heck of a lot more in common, than Afghanistan has with the USA.

          The Pak border with India was settled in 1947, but the border with Afghanistan is essentially the Durand Line “drawn” in 1893, through the middle of the Pashtun tribal lands.

          The other ‘stans were already “annexed” to the Russian Empire in 1918, and at that time most (all) of the Caucasus were within the Russian Empire, as were the Baltic states. The Baltic states regained their independence in 1918 as a result of the Bolshevik Revolution, then they lost it again as a result invasion by Hitler, then they were invaded by the Red Army. And because of Yalta they ended up in Stalin’s USSR, then came 1989.  Some of the Caucasian states regained their freedom in 1989, others didn’t.  The Soviet Union didn’t add much new territory to the Russian Empire – although most people think it did.

          There are precedents for redrawing borders – the treaties at the end of WW1 redrew the map of much of Europe, including the creation of Czechoslovakia and Yugoslavia; the former dissolved and the latter blew up in the 1990s.   Russia redrew the borders of Georgia a few years ago.

          As for democracy – I think the people of Afghanistan would probably settle for a one party Marxist Capitalist state like Singapore – but they’re not Confucians and there’s very few Lee Kwan Yew’s in Afghanistan 🙂

  • I seriously doubt the Balochis can see themselves as part of the Pashtun ethnicity. The Balochis are fighting their own wars against Pakistan and Iran for freedom but they are doing so in Balochi ethnic guerrillas. Nothing to do with Afghan/Pashtuns.

    • Anonymous


    • I think you are not familiar with Baloch history that’s why you stated that Baloch can be seen as part of Pashtun ethnicity, You should read the Baloch history they are not Pashtun however they got historical and political relations because we are neighbours.If you can think that Baloch are part of Pashtun ethnicity than I am sure  tomorrow you will predict that Slavic peoples are part of Arab ethnicity ! 

      As a Baloch I request you people please first do research ,find out the facts than pass comments on Baloch and Balochistan 

      • Where do I say that the Baloch are part of the Pashtun ethnic group?  Nowhere. There is nothing in the post to indicate that the Balochs have anything to do with Pashtun ethnicity. Rather, the post merely shows that some Pashtun nationalists claim all of Pakistani Balochistan (but not Iranian “Baluchestan”) in their maps of their imagined state of Pashtunistan. Please examine the maps, and you will see that this is indeed the case. I can understand why this upsets you, but I do not understand why you would find fault with GeoCurrents merely for mentioning this fact. Your disagreement should be directed at those Pashtun partisans who dream of annexing Balochistan, not with the blog that informs you of their desires.   

        The issue of Balochi national/ethnic identity was explored at length in earlier GeoCurrents posts. Please refer to the GeoCurrents MasterMap to obtain access to these posts. Alternatively, one can search the blog archives. 

        There are, however, many connections between the two groups. But as Frederik Barth showed years ago, it is vastly easier for a Pashtun person to become Baloch than it is for a Baloch person to become Pashtun.  

        • Martin, I believe that Jamal Nasir Baloch responded to Maju’s comment “I seriously doubt the Balochis can see themselves as part of the Pashtun ethnicity” (and has misread it to say that Balochis do indeed see themselves as part of the Pashtun ethnicity), not to your post.

          • Thank you, and I must apologize to Jamil Nasir Bloch in that case. I obviously need to pay more attention to the discussion format!

          • This is just the sort of thing that we linguists catch more easily… I should have pointed out that misunderstanding when I first noticed it… sorry.

          • Sorry Sir, I was just Replying to Maju’s comment and I read your all article about Balochistan these are great piece of information and I hope you will write more on Balochistan ! 

  • Anonymous

    Then there’s the Brahui who are Dravidians

    They’re geographically co-mingled with the Baluchi – I think.  Anyone know how they came to be in the border region of Af/Pak/Iran – presumably they migrated there from SE India, but why, when, etc

    • The current linguistic consensus about Brahui is not that this group migrated from southern India, but rather that it didn’t migrate to southern India, like other Dravidian-speaking groups did, being displaced by advancing Indo-European speakers. A similar story is said of other (small) pockets of Dravidian speakers currently in the mountainous eastern regions of India (Gondi, Kurukh, etc.)…

      • Thanks, Asya, for the clarification. It does seem, however, to be a more contentious issue.  The absence of old Persian loan-words in Brahui has been taken by some scholars as evidence that the group originated in central India and moved north circa 1000 CE.  (see Overall, however, the Dravidian language family may have originated in “Greater Iran” rather than in South Asia. Elamite, spoken in ancient times in SW Iran (Khuzistan/Ahwaz), may have been a Dravidian language

        • You are probably right in that the Brahui moved from central India, but the Dravidian group as a whole moved to (mostly) southern India from much further north (perhaps even “Greater Iran”, depending on how it is defined). There is also some intriguing evidence of Uralic-Dravidian contact, in the form of both cognates and grammatical similarities (I discuss the latter in chapter 3 of my forthcoming book). While at least some grammatical similarities can be explained away independently, cognates (if substantiated) suggest that Dravidian languages were once spoken much further north.

        • Anonymous

          Fascinating – as an amateur I was too “scared” to suggest that the Dravidians might have migrated north to south.  Maybe the reason they’ve not borrowed (m)any old Persian loan-words, is because first arrivers borrow less words from late arrivers than visa versa – why would I say that, just a hunch 😉

          I recently finished reading Lesley Adkins biography of the Rawlinson who transcribed the Bisitun and other cuneiform texts. I think Elamite is one of the 3 languages at Bisitun.  Rawlinson was an East India Company officer, worth reading if only to get a feel of the scope of their work. 

    • Brahvi is a language of Baloch nation. Baloch nation speak two languages 1) Balochi and 2) Bravhi or Brahui. Balochistan was an Independent state till 27 March 1948 and Kingdom was ruled by Khans(or Kings) most of kings’ first language was Bravhi, but the official language of the state was Balochi and persian. When British came to Balochistan,they used their famous tactic “Divide and Rule” British writer were first among all who wrote that Brahvi are not Baloch, after the complete liberation the same tactics utilized by Pakistani state. I should remind all of you that First man who took arms for Baloch people ,the man who started the War of Liberation of Balochistan in 1948 was Agha Abdul Karim Baloch, and his first language was Bravhi. 

      • Interesting comment — many thanks.  But Balochistan was not an “independent state” until 1948. Under the British Raj there were several “Princely states” in Baochistan. They were autonomous, but not independent, as they remained under British authority. By definition, an “independent state” has full sovereignty, which they lacked. 

  • Chris in Binghamton

    An interesting parallel when it comes to names for countries and geographical regions is Macedon(ia). I remember reading quite a bit in the 1990’s and early 2000’s about how Greece objected to the use of the term Macedonia because the region included large portions of northern Greece. [I don’t remember if Macedon(ia) actually had expansionist aims directed against Grecce.] For some time, that new state had to go by FYROM, the Former Yugoslavian Republic of Macedon(ia). There may be other examples out there for a state’s name that doesn’t quite match the geographical homonymn.

    Apologies for the use of Macedon(ia), but I’ve seen both used in serious literature concerning this state and geogrpahical region and am unsure which is correct.

    • Good point. I have spent a bit too much time recently looking at nationalist YouTube videos propounding “Greater Nations” in this part of the world: Greater Turkey, Greater Greece, Greater Bulgaria, Greater Serbia, and so.  In almost all of these maps, Macedonia disappears. 

      Ancient Macedonia included most of the modern country of Macedonia as well as much of what is now northern Greece, largely the area that the Greeks call Μακεδονία or Makedonía . The ancient Macedonians spoke a Greek dialect, whereas Macedonians of the modern country of Macedonia speak a Slavic dialect very similar to Bulgarian.  

    • Anonymous

      Chris, I don’t think the problem’s gone away – Greece had no real choice with the change of name when “FYROM” joined the UN.  But if/when the Macedonia joins EU (assuming its still there) the issue will come up again, and Greece has a veto in the EU.

      Martin, about a 25% of Macedonians are Albanians,

      The Slavs didn’t arrive in the region until the 4th Century AD or later, so it’s a bit of a cheek to call yourself Macedonian in the hope that some of Alexander’s glory will accrete to your country.  An better alternative name would be Paeonia 🙂

  • Omar katawazay

    There are even more Pashtuns in Afghanistan than showing in the map above, this map by Afghanistan side is totally wrong

  • Pingback: The New York Times’ List of Potential New Countries, and Others As Well | GeoCurrents | Humbert.Cat()

  • shahid h khatttak

    In Pakistani Balochistan those who speak Balochi as their first language/mother tongue are no more than 22% of the population. An early British gazetteer(1885-1890) on Balochistan considered Balochi language to be native to 18% of the people of Balochistan. In 25 out of 33 districts today Baloch are either non-existent, speak languages other than Balochi or are tiny minorities. This totally ties up with the 20% of the population of the Paki province of Balochistan considered baluchi speaking.

  • Pingback: The New York Times’ List of Potential New Countries, and Others As Well - Languages Of The World | Languages Of The World()

  • Nakash Chit

    Excellent post. Without teachers like you international diplomacy will continue to be (mis) informed by ignorance. One detail that might interest you: Though the name Pakistan was decomposed into an acronym of the West Pakistan states, it seems most unlikely that this was the origin of the name. “Pak” is a Persian word meaning pure or holy, used in Urdu and Hindustani generally, so Pakistan translates as the pure land, which is clearly the intended meaning in the Pakistani national anthem “pak sar zameen shad bad”. I am afraid that what you may have thought a delightful oddity is rather more prosaic.