Finding Data in the Soviet-Afghan Borderscapes

The semester, and your humble narrator’s introduction to Harvard, continues to speed by too quickly. After bidding farewell to visitors to Cambridge at the beginning of the week, I gave my first seminar talk at Harvard – kindly hosted and catered for by the Russian & East European History kruzhok (circle). Presenting some re-worked material from one of the more polished chapters of Developing Powers gave me the chance to think more about issues I had missed until now: the role of “free” places like Afghanistan or Chernobyl in the mentality of the last Soviet generation; that same generation’s perception of what really mattered in terms of world politics (Afghanistan was seen as “hot,” Poland in spite of Solidarity not so much); and the ways in which Soviet and Afghan cartographic practice imposed certain developmental visions on the landscape of Afghanistan. After the Q&A (and smuggling some pizza back home), your humble narrator finds himself back in the rhythm of scrambling between writing & editing, ordering obscure maps of Central Asia from Harvard’s wonderful map librarians, and cleaning up the data that I do have to add the cartographic and digital spine that I desire for the book project.

Which raises all kinds of reflections. This week, for the Digital History seminar, we’re asked the following: “Describe the data you plan to use for your project. In what ways (if any) does it differ – in breadth, depth, format, etc. – from the “data” you have used to write papers in the past? How are you dealing with problems of information density, sample size, etc.?” Had you asked me this question as recently as a year ago, I’d have given a pretty conventional answer: Russian and American archives, special collections of Afghan material like the fantastic Arthur Paul Collection at the University of Nebraska-Omaha and the University of Arizona’s digitized collection of government newspapers, and the usual litany of secondary materials. The considerable amounts of VLKSM (Komsomol) reports from the ground in Afghanistan I viewed primarily useful for their prose, not the tables and charts that they themselves contained. But in part as a result of taking the Digital History seminar, I’ve been challenged to think about more creative ways to “scrape” and make use of the considerable amounts of data – not just prose – that exists in the sources I do have. Add to that new materials like the Afghan Death Lists, and my own growing awareness of sources on the refugee crises in Iran and Pakistan that followed the invasion, and the question becomes not what to use, but rather how much, and why.

Omaha and its special collections loves historians back

In any event, to answer the question more directly, as I go ahead with the project for the seminar, I’m concretely working with several sources that have to do with state-led violence and forced migration in and from Afghanistan in the 1970s and 1980s. The overall aim is to provide a robust quantitative picture of what I see as one of the major ironies in the story I’m telling: Afghan leaders successfully enlisted foreigners to build the developmental hardware of Pashtunistan (the idea of a state that would be a national homeland for much of the region’s tens of millions of Pashtuns) throughout the 1960s and 1970s, but the combination of an ultra-nationalist PDPA Pashtun Communist government from 1978-1979 and, even more importantly, the Soviet invasion, destroyed the possibility of such a Pashtun commonwealth, at least on the terms originally envisioned by leaders like Mohammad Daoud Khan and Zahir Shah. As the UCLA historian Nile Green observes, this crucial period of the late Seventies and early Eighties is one of several mid-century moments in which Afghans and foreigners move clumsily (and violently) from one vision of state-building in Central Asia to another: one in which both Afghans and foreigners play a role in state construction and state destruction. With the dream of Pashtunistan destroyed via mass emigration (prompting Pakistani authorities to fret over a Pashtunization of Pakistan on terms not their own), Soviet nation-builders embark on a doomed quest to build a Soviet-style territorial state in Afghanistan – doomed because of the lack of an economic base and the fact that many parts of the country were depopulated by up to a half of their 1979 demographics.

In order to tell this story, I’m trying to link together several sources, each of which tells a slightly different sub-story of its own. There’s the Death Lists themselves, which are a story of state-led killing rather than forced migration. They, however, are useful for contrast in that they allow us to get a sense of what regions (provinces, and even subdistricts), professions, and religious groups of Afghanistan Nur Muhammad Taraki and Hafizullah Amin targeted, and when. Having geo-coded a good chunk of the data already, there’s some findings there that already surprise me. I am struck by how urban a phenomenon the killings were: if I find enough good maps of 1970s-era Kabul, the data about which neighborhoods of Kabul people were plucked from (to be shot and killed at Pul-i Charki prison) is there to produce an urban micro-history. Whereas the Soviet invasion led to depopulation in primarily those provinces along the border with Pakistan – Nangarhar, Kunar, Paktia, Paktika – the terror conducted by the PDPA appears to be more local and concentrated along the Soviet- and American-constructed ring road. Provinces like Ghazni (to the south of Kabul) and Baghlan (a cotton-growing center north of the Salang Tunnel) were particularly hard hit. Now that I’ve finished coding professions for victims, too – classified according to International Labor Organization standards and the data supplied in the original Death Lists – I can begin to get a sense of both the “professional topography” of the country right before the war, and of what groups were particularly targeted by the PDPA (Army officers and students took huge hits).

Neatline: my vehicle for telling the story of depopulation and state destruction in Afghanistan?

The second major source base – one that I’m pleased to report is getting more robust with every day – are reports from the refugee camps in Pakistan and Iran which tell us more about the provinces from which refugees were ejected, or fled from, as they went into Afghanistan’s non-Soviet neighboring countries. These sources – now from 1985, 1989, and 1990 – allow us to get a better sense of how the Soviet invasion emptied out the country, how it terrorized the country, in ways that differed from domestic terror. Some of these sources even include data from small (n=2,000) surveys on refugees’ professions, showing us that the refugee crisis was one driven more by farmers and the uneducated (whereas the Death Lists tells us that Taraki and Amin’s terror was one against officers and the intelligentsia). As I accumulate more data from refugee publications, I can even begin to get some sense – one has to remember how damned difficult statistical collection was – of the rates of depopulation of those border provinces (most of which were emptied out by as quickly as 1982; word of the Soviets appears to have spread quickly). In the next couple of weeks, I need to start to think more seriously about how I all want it to look; Professor O’Neill reports that Harvard is moving quickly on making it easier for graduate students and post-docs to build implementations of NeatLine, a chronological visualization tool that works with Omeka. Building something like a digital and chronological atlas of “Pashtunisan” that walks the reader from the spring of 1979 (the beginning of the Death Lists’ data) to 1990 (when the refugee figures end), and shows us how three million people performed mass migrations during that period – one of the largest forced migrations in recent history – in a way that had major ramifications for Pashtun, Pakistani, and Soviet visions of political order in Central Asia.

How does it all differ from sources I’ve used in the past? Obviously, many of these sources are data in the sense that there’s relatively little prose to accompany them. As a historian – even one trained in Economic and Social History with wonderful teachers at Oxford – I’m still inclined to think of text as the raw material from whence we get started. And in many of the drafts of chapters that I’ve worked on, I’ve tended to view documents like the VLKSM reports as just that – text, products of a specifically Soviet discursive system, and so on. The “really” interesting moments for me were those tidbits that come out of the woodwork when you pore over the bureaucratic prose: Soviet-Afghan marriages, advisers glossing over the execution of 5,000 mujahideen near lazurite mines in Badakhshan, and the ways in which these borderscapes and episodes of mass killing, respectively, became covered up by the “government of paper” that was Komsomol and, arguably, the Communist Party of the Soviet Union more broadly.

A post-1991 World Food Programme map of Badakhshan, a remote battleground for humanitarian aid and development and complicated part of the Soviet-Afghan-Pakistani borderscape in the 1980s

But working with these more data-driven sources has had a perceptible effect on how I work with the same sources. Take the case of Badakhshan – a rugged mountainous region of northeastern Afghanistan that borders with Tajikistani Badakhshan, Pakistan’s northern areas, and even the extreme southwest of the People’s Republic of China. During the Soviet invasion, the high-altitude trails and passes of the region were traversed not only by mujahideen (many of whom were, as the aforementioned report makes clear, engaged in trading lapis lazuli for weapons) but also Western doctors and NGO workers who provided humanitarian aid to the locals. With the Soviet Union, Afghanistan, Pakistan, China, and the nascent institutional vehicles of global health all in one place, writing about Badakhshan during this period is like being a mako shark at a seafood buffet. Even more so after my attempted “digital turn.” Many of the VLKSM reports from Fayzabad (the provincial capital of Badakhshan) tell us about  not only Soviet- and Afghan-led attempts to recruit youth into the PDPA during the 1980s, but also the recruitment figures for various mujahideen groups in the mountainous province, down to the level of the subdistrict. I’ve yet to do anything concrete here, but I wonder: if I can find out more information about these subdistricts (geography; accessibility of roads to Fayzabad vs. passes to the Panjshir, Nuristan, and eventually Pakistani Chitral; the extent to which they were depopulated by the invasion), might it not be possible to better explain and visualize (especially important for such a rugged, inaccessible place) the battle between the PDPA and mujahideen groups for Afghan youth in these area? I’m not sure yet. But the process of becoming more fluent with data – and viewing it as equally useful as bureaucratic prose – lends a new vision and sense of adventurism when I see tables like those in the archives.

It’s time to get to work.


Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s