This conversation between researcher Madiha Tahir and system-design engineer Majd Al-shihabi is the continuation of one started in Beirut after PalFest 2019. They discuss the way data and digital instruments are used by the state and by activists in Pashtun tribal areas administered by Pakistan and Palestine.
MADIHA TAHIR: What is the relationship between data, knowledge, and political resistance? Part of my impetus for this question comes from my brief travels through occupied Palestine, and contrasting that with the sites, spaces, and communities I engage with that are related to the Tribal Areas along the Pakistan-Afghanistan border. Called the Federally Administered Tribal Areas or FATA, the region has been bombed by American drones since 2004 and subject to multiple ground and aerial assaults by the Pakistani military as part of the so-called “war on terror,” even though this is Pakistani territory. The Tribal Areas were governed through indirect rule by the British colonial administration, specifically though the draconian Frontier Crimes Regulations (FCR). At independence, the Pakistani state incorporated the FCR into its governance structure, and while it initially applied to various territories within Pakistan, it was repealed in other parts of the country while it continued on in this border zone.
Under the FCR, ethnic Pashtuns from FATA didn’t get the right to vote till 1996, and once they did, they were in the bizarre position of voting for assembly members who had no capability to make laws or changes to the region. Instead, they were governed through a federally appointed Political Agent (PA) in concert with a range of maliks (loosely meaning tribal elders), who were given funds by the government to police their own tribes. The FCR allowed the PA to administer ruthless forms of collective punishment to entire tribes on mere suspicion of wrongdoing. They razed homes, arrested entire groups of people, levied heavy fines, and so on. So we are speaking here of people who are ostensibly Pakistani citizens but who are subject to checkpoints, bombardment, arrests, military blockades, curfews and other methods found in occupied zones.
Both Palestine and the Tribal Areas are instances of domination in their separate ways and there are different registers, qualities, and quantities of knowledge production and data about each of them. Occupied Palestine is probably one of the more examined conflicts by scholars as well as human rights organizations. And, in the context of the erasure that Israel wants to enact upon Palestinians, it seems to me that there are two streams of data-driven projects around occupied Palestine. One set consists of archival and mapping projects that recuperate and make visible the long, indigenous history of Palestinians in Palestine. Here, I am thinking of your work with Palestine Open Maps which has been collecting and extracting data from historical public domain maps, and also of Visualizing Palestine. The other are projects that document the occupation itself, such as B’Tselem’s maps of Israeli checkpoints, settlements and other architectures of occupation.
By contrast, that kind of rich documentation by independent parties doesn’t really exist for Pakistan’s Federally Administered Tribal Areas (FATA). Unlike Palestine, I don’t think most people have ever even heard of FATA though the U.S. has been drone bombing this region along the Pakistan-Afghanistan border since 2004. Since 9/11, the Pakistani security forces have blockaded various parts of the Tribal Areas. Internet and cell phone service is patchy to non-existent in various regions there, and sometimes forcibly shut down by the state. This isn’t simply about lack of infrastructure but a built, digital enclosure that is intended to interrupt, delay, and curtail communicative practices.
So, traveling through Palestine and thinking about the Tribal Areas, I felt somewhat unsettled. First, I was astounded by the architectural scale of the Israeli occupation. I have read about its materiality through Helga Tawil-Souri’s, Nadia Abu El-Haj’s, and Eyal Weizman’s work, among others. I knew of it intellectually already, but it is another thing to witness first-hand the chillingly methodical strangulation of Palestinian neighborhoods and sites. But, second, perhaps more cynically, it brought home to me that there is a fair amount of information about what is happening in occupied Palestine, and yet the occupation and the theft continue in broad daylight, as it were. What hope then is there for the Tribal Areas? To what end do we do our mapping projects, our statistical projects charting theft and murder and forcible disappearances and whatever else?
Where the Tribal Areas are concerned, it’s been interesting to see the rise of a non-violent protest movement there over the last year, and what that has done to how information circulates. To explain, I’ll just say a bit more about the political context of FATA: unlike Palestine, which is explicitly occupied and drone-bombed by Israeli forces, there is a split situation in FATA. It’s drone-bombed by the U.S. and militarily managed on the ground by the Pakistani forces. What the ethnic Pashtuns who live there have been calling for largely is not independence but assimilation into the Pakistani state — though that may change given the ruthless way the Pakistani military has been attempting to handle the movement.
Last year finally, after the non-violent Pashtun Tahaffuz Movement (Pashtun Protection Movement or PTM) became a national protest movement, the government quickly repealed the FCR and legally assimilated FATA under the constitution. But, it’s still uncertain what that practically means and things on the ground continue the same as ever.
PTM has really shifted the resonance of media artifacts. Before PTM, photos of people who had been killed or disappeared in the Tribal Areas did get out and circulate, but their circulation didn’t really create any disturbance or interruption in the general flow of national Pakistani publics. In other words, if the assumption among journalists, scholars, and activists is — and I think a lot of forms of scholarship and activism implicitly assume this — that the right set of data or the right visual product will produce an affective response, maybe revulsion or pity or whatever, that will compel people to say, “Enough is enough,” that hasn’t really happened. It was after PTM that I think people started to see these images differently, that is, to be impacted by them. The movement shifted the relationship between an artifact (an image, a video, a document) from the Tribal Areas and its viewer or reader. Suddenly, these stories and these images started to circulate beyond the smaller networks directly connected to FATA. Suddenly, I hear broader groups of people talking about what happened at that checkpoint in Swat (not in FATA, but a Pashtun region) or about how Rao Anwar, a policeman who killed a young Pashtun man, Naqeebullah Mehsud, tried to label him a “terrorist’ — all of a sudden people are talking about all of this.
So, I guess what I am saying is that there is a tendency to assume that data or information or media artifacts beget movements or a compulsion to do something, but perhaps it is in some ways the other way around: movements give force and meaning to data, information, and media. And, perhaps recognizing that has an impact on how we should or could think about how we put such projects together.
The last thought I’ll mention here is about how evidence is constructed. Testimony, for instance, became evidence of the horrors of the Holocaust. But, today, evidence largely seems to mean “data,” that is the construction of calculable, statistical information and the mobilization of large-scale data-sets and their attendant scientific expertise. It is a process that can supplement witness testimony, but it seems to me that it can equally displace first-hand accounts by the already marginalized.
So for instance, earlier, we had a discussion about mapping projects. We looked at some GIS and hand-drawn maps and talked about what different kinds of maps foreclose and reveal, and the various kinds of epistemologies at work depending on the map. Hand-drawn maps can reveal details about the experience of distance and time in spaces of war and occupation, but they can be impossible to commensurate with GIS maps and sometimes, even with each other. In other words, they tend away from “ontological stability” — Miriam Posner’s term for how data is constituted through the stabilization of relationships.
Another way to put it this issue is to ask whether data-driven projects that derive their legitimacy through large-scale, calculable data-sets play into the dismissal of the Palestinian witness or the Pashtun survivor. It reminds me of that well-known quote by Toni Morrison:
“The very serious function of racism […] is distraction. It keeps you from doing your work. It keeps you explaining, over and over again, your reason for being. Somebody says you have no language and so you spend 20 years proving that you do. Somebody says your head isn’t shaped properly so you have scientists working on the fact that it is. Somebody says that you have no art so you dredge that up. Somebody says that you have no kingdoms and so you dredge that up. None of that is necessary” (emphasis mine).
I sometimes wonder if that’s what we are doing, though I admit it is a cynical way of looking at things. They say we don’t exist, so we run around gathering data to prove our histories. They say we are lawless, so we run around trying to prove that we were good law-abiding “civilians.” It keeps you from doing your work. What is our work? What could it be?
MAJD AL-SHIHABI: Does the ability to produce knowledge about a territory, in a way, an assertion to the right to that territory? And I use the term “knowledge production” carefully, to contrast it with collecting data, or turning it into information.
Take for example the Pakistan’s Federally Administered Tribal Areas (FATA). OpenStreetMap (OSM) is the world’s foremost repository of open geographic data. This data includes road network, building footprints, shops, checkpoints, and any data that can be verified by the “ground truth” rule of OSM: if you can see it in real life, you can place it on the map. The data recorded in OSM about the entirety of Pakistan is about 46 MB of data, while the city of Berlin alone has about 54 MB of data recorded on it. But there is also a lot more data in other, less accessible repositories, such as the map that you shared with me. Overlaying it on the OSM map reveals numerous urban clusters, visible from satellite and aerial imagery, which are not recorded on the map itself. Why is this data not recorded in public maps? From what I understand, under the FCR, whole populations could be forcibly transferred, rendering some of those urban clusters empty. If one day the entire population of, say, Zaukai, was moved from its location to another location many kilometers away, then with the ground truth principle of OSM, the latitude and longitude of the village itself would need to be moved. That would require the OSM editors to make a political decision, justifying the movement of an entire population: do we call this forced population displacement? Do we call it ethnic cleansing?
Another layer of data is that of satellite and aerial imagery. It’s very easy to go to Google or Bing maps and see reasonably high resolution satellite images of the FATA. Similar images, albeit undoubtedly at higher resolution, are available to drone pilots carrying out “targeted killings” there.
However, having this data does not suffice to understand the territories of the FATA. The U.S. Army is able to turn this data into information, that is, turning latitudes and longitudes into maps. However, in order for this information to be useful, it needs to be contextualized within intimate knowledge about the territory. The tribes who have historically inhabited those lands and continue to do so have that knowledge, whereas the U.S. Army does not. This is why they protect themselves by carrying out their warfare via drones, by pilots commuting to the comfort of their bases in Texas and New Mexico, instead of boots on the ground.
This is true outside of FATA as well: since its creation as defined in the Oslo Accords (1993), the Palestinian Authority started producing high quality data using Geographic Information Systems (GIS) and publishing it on their portal, Geomolg. It is a way of asserting control over a territory: they map everything, in order to protect whatever has remained of Palestine from the creeping encroachment of Israeli settler colonialism. This results in a comical representation of the lands designated by the accords as the Palestinian territories, showing aerial images of the West Bank and Gaza as decontextualized lands floating on a white space (no pun intended!) — and a more tragic representation, where you see the division of Jerusalem in a very stark way.
The issue of aerial imagery is very interesting: in the U.S., the Kyl-Bingaman Amendment (KBA) was passed in 1997, restricting the sale of satellite imagery of Palestine (including Israel) to a 2 meter resolution, contrasted with the 50-70 centimeter resolution available for most of the world. The amendment is justified because publishing high resolution imagery “could unintentionally have a deleterious impact on the national security of the state of Israel,” in the words of U.S. Senator Kyl.
Here we have two territories: Palestine/Israel with detailed GIS data, but poor aerial imagery, and FATA, with poor data but finer satellite imagery. But I want to focus less on the data itself, and think more about what data counts as legitimate, and what data does not.
We have different repositories of knowledge, and their perceived legitimacy depends on their uses. Oral history is a great example: oral testimonies, especially those about traumatic events, often contain contradictions, if they are viewed literally. However, in many oral traditions, the details of stories mutate over time, but the core truth that the stories are telling stays the same. Those stories even preserve events such as tsunamis over geological time. In the July-August 2018 issue of The Funambulist, Elise Misao Hunchuck described ancient markers dotting the landscape of eastern Japan, warning people not to build there because of the danger of tsunamis. The 2011 tsunami proved those markers right.
Earlier this year (2019), there was a conflict in the Turkish Parliament. In the perpetual denial of the Turkish state of the Armenian genocide, Erdoğan questioned the legitimacy of an Armenian Member of the Turkish Parliament, Garo Paylan, saying that the any evidence of the genocide would be in the state archives. But of course, those archives are censored. Paylan responded that he doesn’t need state archives to prove that the Armenian genocide took place because, as he said: “My grandmother is my archive.”
So when knowledge is produced, the questions are: who is controlling its flow, and for what purposes? Many indigenous communities have struggled to keep their sacred knowledge from being exploited by capitalist forces. Aboriginal tribes have sought to hide their sacred places from European cartographers. An example from Māori culture was very interesting: an Māori organization called Te Hiku Media has compiled a corpus Māori language, both written and oral. This data can be used in many commercial products, for teaching the language, translation, and more. The people who produced the data also produced a license called the Kaitiakitanga License, based on the Māori concept of Kaitiaki. The license acknowledges that the corpus is not owned by the organization, but was given by the community. Thus, the organization’s role is not to collect fees for its usage, but it is to be the guardian of a collective resource, with authority to grant or deny access to it, in accordance with the interest of the community.
I want us to dwell a bit more on the last idea in your text. The quote by Toni Morrison that you used describes the effect of racism very well: knowledge production is happening by all people, all the time, at different scales. The problem becomes: who is the audience of the knowledge that we are producing? Why is it being produced? What are the dominant epistemic frameworks for producing the knowledge? How is globalization affecting our answers to the previous questions? At what scales?
MADIHA TAHIR: Reading you email, especially your closing questions, I am reminded of philosopher Jacques Rancière as, perhaps, a useful theorist to think with on these questions. Rancière thinks of the police in opposition to politics. He argues that the police is a partitioning of the sensible (“Ten Theses on Politics,” 2001), that is, a way of constituting the social and shaping our perceptual fields. This partition distributes the world. It establishes what is visible and what is not, what can be heard, and what is made inaudible.
I thought about Rancière in reference to your discussion about the restriction on the resolution of satellite imagery where Palestine is concerned. I think the problem goes beyond restriction to a question of the shaping of our senses, and sense, of the world. As you may know, even degraded, publicly available 50 centimeter pixel resolution essentially erases the human subject since it produces a conflation between pixel and subject. This retains the privacy of individuals captured by satellite images, but it also invisibilizes certain kinds of human flows and political violence (cf. Eyal Weizman “Violence at the Threshold of Detectability” 2015). Interestingly, it’s not these concerns but market demands that have pushed that resolution down to approximately 25 centimeter now since there are international companies that can compete. And, for sensitive issues, even though the American government can legally demand shutter-control — asking a company to shut down its satellite — it has largely chosen to use its market power and buy up exclusive rights to certain kinds of images.
Other countries are trying to legislate the terms of mapping and aerial photography. Pakistan proposed the Land Survey and Mapping Bill in 2012 that essentially puts the military’s central command in-charge of all mapping. The Act was passed in 2014. It’s vague, broad, and criminalizes other types of mapping by independent groups or organizations unless they are registered with the government’s mapping organization. India proposed its own Geospatial Information Regulation Bill in 2016. Again, it’s extremely broad. Anything, even a geo-tagged selfie could be subject to penalties if done without a license. It also criminalizes the depiction of international boundaries that do not conform to Indian governmental perceptions. This is obviously a key issue where India’s occupation of Kashmir is concerned.
So, if I think of all these practices and laws in terms of a distribution of the sensible, then — going to your question about who produces knowledge and for whom — I really like projects where the police line is undone. For instance, there is an app that tracks White Collar Crime Risk Zones. The app works like an early warning system predicting where white-collar crimes might occur and displaying them with a heat map. It’s brilliant because it reveals the racism embedded within predictive policing as it is used by several police forces in U.S. cities. The latter takes working class neighborhoods and non-white populations as its target and ignores crimes of greater magnitude and impact carried out by the predominantly white and wealthy.
Another example was the app designed by Josh Begley (also guest of PalFest 2019): Metadata+. It alerted the user when a drone bombardment had occurred. So, in a small way, it interrupted the flow of daily life in the U.S. This was deemed so troublesome that Apple rejected Josh’s app 12 times! The company stated that the app violated its clause against “excessively crude or objectionable content.” I find that language really interesting — “excessively crude or objectionable” — because obviously what’s objectionable and crude for Apple is not the fact of incinerating people to death, but that an app should record and disperse this fact to its users. In other words, because the app interrupted what Apple thought the distribution of the sensible should be, the company rejected it. And, it did so not on an overtly political basis, but in affective/sensible terms (“crude,” “objectionable”).
These projects help me think more broadly about what digital projects could do that goes beyond aggregation.
Another way of putting that last point is that I am concerned about what we reveal about ourselves in these projects. As you wrote, information needs contextualization — this is key. I find it interesting that anthropology, my own discipline, doesn’t register for journalists at all. Journos aren’t running to anthropologists for ‘expert’ quotes. And yet, the U.S. military has been interested in a certain kind of actionable anthropological insight and contextualization. This is what the military’s Human Terrain System (HTS) was all about. Teams of social scientists were deployed alongside “violence workers,” — I’m borrowing the term from David Correia and Tyler Wall — in Afghanistan and Iraq in a bid to understand “culture” and thereby minimize fatalities and capture “hearts and minds.” David Kilcullen’s brief article on the Tribal Areas begins with a quote by an anthropologist admonishing Kilcullen to know the “terrain and the tribes,” and learn how to play “their game.” In fact, there’s a long and shady history of anthropology’s complicity with American power that the anthropologist David Price has extensively documented. It is in the shadow of that history that the discipline developed “thick description,” the notion that multi-layered, interpretively oriented descriptions can yield hidden meanings that help contextualize behaviors and norms in any given community. Anthropologist John L. Jackson criticizes thick description as a kind of anthropological hubristic ambition towards “full social knowing.”
The dark irony of thick description is that its most famous practitioner Clifford Geertz failed to mention that he developed his ethnographic fieldwork at the Center for International Studies funded by the CIA and among a milieu invested in U.S. imperial interests. His exemplary thick description of the Balinese cockfights similarly omits the tense national situation that engulfed Indonesia, one that would result in the massacre of 500,000 to 2 million Indonesians backed by the CIA (cf. David Price’s Cold War Anthropology, 2016, and Nancy Scheper-Hughes’ “The Primacy of the Ethical: Propositions for a Militant Anthropology,” 1995).
The question then is perhaps less of thickness than that of strategic elision: do we elide on behalf of power or in solidarity with those resisting it and surviving its conditions? Race and tech scholar Ruha Benjamin explicitly discusses these issues and uses Jackson’s elaboration of “thin description” in her work, Race After Technology (2019), to think about strategic knowledge production. Speaking about the inexorable expansion of exploitative data regimes, she writes, “If the New Jim Code seeks to penetrate all areas of life, extracting data, producing hierarchies, and predicting futures, thin description exercises a much needed discretion, pushing back against the all-knowing, extractive, monopolizing practices of coded inequity.” This discretion is a political relation, a way to respect, protect, and defend racialized and surveilled communities.
I’m thinking here also of Édouard Glissant’s demand that “we clamor for the right to opacity for everyone” (Poetics of Relation, 1990). This is a different kind of political and ethical relation, one that doesn’t engage the politics of recognition. I think this partly means we have to become different kinds of subjects and reconfigure ourselves. So much social capital is now tied to constantly presenting, performing, and make our identities and communities available and knowable as a means towards some minimal kind of recognition. Against this, Simone Browne’s (also guest of PalFest 2019) analytics of “dark sousveillance” (Dark Matters, 2015) offer ways to think about and challenge surveillance. Dark sousveillance encompasses the tactics and strategies used to counter pervasive surveillance during Black enslavement, tactics like anti-surveillance pranking and counterveillance. But, drawing on this work would mean rethinking our historical frames. It would mean rethinking the terms of our politics.
So, this is some of what I am both inspired and haunted by as I try to think through the possibilities of digital projects. I wonder whether you have come across such concerns in your own work? Are their ways in which strategic invisibility is a practice or a tactic within your projects?
Also: Vive la révolution! Isn’t the digital also inflected in these protests?
MAJD AL-SHIHABI: Between my last letter to you and your response, a revolution has erupted in Lebanon, the country I’m calling home for now. I will talk about the role of technology and data has played in the protests, but first, I want to acknowledge how politically mature this protest movement has been. Many western media outlets tried to call the revolution a “WhatsApp Revolution,” because the immediate trigger for it was a proposed tax on VoIP calls. This is similar to the flattening that we saw in 2011, where the same media outlets called the revolutions “Facebook revolutions” or “social media revolutions.” The 2019 protestors were very quick to shut down that rhetoric, and made it clear that what was happening was a revolt against the cumulative effect of decades of the political class plundering the public resources of the country. Yes, technology has catalyzed the protests, and WhatsApp, Telegram, and Signal groups have sprung up to disseminate information about actions while much of Lebanon’s media has enacted a complete blackout against the protests. However, the organizing has been happening through human to human connection, and people would have taken to the streets regardless of which path the information took to reach them.
While we’re speaking of mobile phone applications, I want to point out another app that was approved, and then removed from the Apple App Store. HKmap, an application that shows live, crowd sourced updates on locations of police, protests, and usage of tear gas. The app was withdrawn from the App Store because it “allows protestors to escape arrest by the police.” Meanwhile, there’s another app, called Waze — which is incidentally Israeli! — that also allows people to avoid police, this time, it is the traffic police. Waze is still on the app store, and it was bought by Google for almost a billion dollars.
So depending on the level of political engagement of your activity, Apple and Google would either hand you over to the police, or help you evade capture.
This raises the stakes about the discussion surrounding the role of the tech giants in controlling our lives. They are not just mediating our social lives, they are stepping up and becoming mediators of our political expression, with a profound effect on the political systems that govern us.
While the revolution was breaking political taboos in Beirut, I was in London, at the heart of the empire, attending the Mozilla Festival. One of the central themes of MozFest is the discussion around decentralization. Can we create robust alternatives to Facebook, Twitter, Google, and the likes, that operate on the community level? Governed by each community’s rules, terms of service, and codes of conduct? This is strongly connected to the question of the production of spatial data as well: the Survey of Pakistan is centralizing the authority of map making, and imposes heavy restrictions and fines on any entity that produces maps outside of that framework. The result is a hegemonic view of what is worth mapping, and what is not, obliterating the diversity of etymologies and ontologies embedded in maps.
I want to bring up one more example that complicates our narrative about data sovereignty and self-determination. A group of young residents of the Bourj al-Shamali Palestinian refugee camp in the South of Lebanon were taught by Greening Bourj Al Shamali on how to create an aerial map from images they capture using a helium balloon. They spent a lot of time going around the neighborhood, to rooftops, and flying the balloon photographing each section of the camp, then assembling the individual photos into one giant mosaic photo. Once the map was done, the map was presented to the popular committee of the camp, which then decided to not publish the complete map at all, citing security reasons. However, I would argue that most adversarial entities that would want a high resolution aerial photo of the camp can get one easily. The party that has the most to lose by such a decision is not an adversarial entity, but the residents of the camp themselves, who cannot see a map made that reflects them. The point here is that the popular committee is represents the camp’s residents officially, but is not democratically elected, and positions are assigned by different Palestinian factions. So here we have a case of nominal data sovereignty, that in practice has exactly the opposite effect of its intention.
One final thought: one of my favorite moments from the revolution: one of the things that makes this revolution unique is that the protests are diffused all over Lebanon. In Beirut, one of the main sites of protests is the downtown area. Until October, most people would refer to it as “Solidere”, the name of the company that was tasked with the reconstruction of the downtown area after the civil war, and instead of restoring it, they turned it into a rich people’s Disneyland. Someone put up a giant banner which read “It’s called Balad, not Solidere!”, Balad being the common name for a city center in Levantine Arabic. The people of Beirut are re-discovering the parts of their city that were previously off-limits, learning how to use public space again, and to do that, they need to reclaim the names of those spaces, and the maps that they use to navigate it.
MADIHA TAHIR: As we go to press, major, multi-ethnic student protests have erupted in several cities across Pakistan! The protesters have several demands, among them, an end to mass surveillance and the securitization of university campuses where the state repeatedly intervenes to shut down political activism, unions, and any gathering it construes as a threat.
I love this protest poster by a young Pakistani-based artist Saniya Kamal because it foregrounds the gendered surveillance of women’s bodies that is part of securitization practices. A woman who dresses differently — say, she wears a leather jacket (an incident that happened recently) — or who acts boldly is viewed as a threat not only by the state but also by various sectors of society, liberal and conservative. To me, Kamal’s poster neatly stages the intersection of gender, patriarchy, and securitization.
I don’t know where the protests will be by the time this conversation is published, but two major protest movements during the span of one conversation isn’t bad! In that, there is hope. ■