Monthly Archives: August 2021

Afghanistan and the Tools of War

The haunting GIF in the header to this post tracks the rapid return of the Taliban to power as a drawdown of the Forever War. It echoes a sense of inevitable loss–a dramatic ceding of territory, echoing the “loss” of Korea, China, or Vietnam–an un-imagined conclusion to the War on Terror. The terrifying denouement of a collapse of provinces across this virtual Afghanistan seems to suggest a logic deflating bravura of the Forever Wars, in which arms and military materiel were funneled at unprecedented rate to Afghanistan–at a rate that would only be later superseded by the rush of arms into Ukraine. This was hardly, the GIF suggests, the conclusion Americans would have expected from Donald Trump’s promise to “ending the era of endless wars,” but was the end of an era of pretenses to American empire, that sent hundreds of billions of military spending to Afghanistan, inflating the budget for the Department of Defense in unsustainable fashion, and, intentionally suggests an ominous terms a haunting pivot to an unknown future without imperial plans. This is a future where the return of military forces from Afghanistan will upset a global military playing field, where war will no longer be fought in terms of a map of Afghanistan or a level field.

But if the glass can be called half-empty or half-full, its apparently overpowering logic of loss also obscures, by flattening to a few months the long history of post-9/11 period, how wars waged since 2001 has left the United States without any control over the ground game. For by failing to find allies in the ground we’ve been pummeling , unsuccessfully seeking to construct alliances on the ground, the arrival of arms and military technologies have re-written the situation of Afghanistan, or the conflict there in which we were long immersed, in ways few Americans have any memory, and surely won’t be aided in the dramatic GIF that suggests the collapse of the house of cards on which we created a power vacuum filled with only intensified high-powered arms, in what was virtually a powder keg of massive American forces across the Middle East, in an extended military apparatus designed to keep a geo-political map afloat that had no endgame or even game.

It is hard to come to terms with the 9/11 wars without tracking the flow of military technology and tools overseas. Over 9,000 Americans have died, or the hundreds of thousands who returned from the wars, injured in body or psyche, the roughly 6,200 U.S. military personnel, contractors, humanitarian workers and journalists killed in Afghanistan since the U.S. government invaded are left off the map, but the legacy may be greatest for the huge amounts of military materiel shipped into the Middle East–arms that helped in some way to “modernize” the current Taliban, who may have received training from Pakistan intel–as well as the huge losses of population and infrastructure in Afghanistan, where about 71,000 Pakistani and Afghan civilians are estimated to have been killed–a staggeringly disproportionate number in crossfire, bombing raids, drone attacks, suicide bombings in Kabul and other bases, IED’s and night-time raids by NATO or American troops.

The GiF that purports to document the effects of American withdrawal renders the battlefield of Afghanistan as the rapid falling of provinces as if they were a gameboard, or a mock battlefield, creating a sense of causation due to American withdrawl by the proverbial falling of a set of dominoes. But the limited long-term strategy of these wars is handily elided in what seems the result of an immediate retreat of military presence. The retreat was, however, only the last act of a tragedy on a massive scale, the result of funneling arms rather than promoting national infrastructure in a nation that has limited infrastructure–and which even American forces were compelled to cast and indeed to consider as a tribal society that had no social structures that could be trusted or built upon. The increased lack of trust that dominated relations on the ground were more revealed by the map–as well as the lack of effort to foster a functioning government. Donald Trump may have escalated the arms trade into the Middle East to levels far beyond his predecessor, but the frustration of his successor Joe Biden was perhaps more clear-eyed than is given credit, if intentionally so: “We provided our Afghan partners with all the tools — let me emphasize: all the tools.”

But were tools of war ever enough? Biden’s remarks revealed a combination of deep dissatisfaction at returning to government after four years, and finding the same boondoggle on the table from the Bush years, and apparent exasperation. If he was trying to justify his rapid withdrawal of U.S. troops from Afghanistan as a pivot in prioritizing strategy he had long seen as of limited benefit or without exit strategy, it betrays a deep sense of what might have been different in Afghanistan, or how the map of civil government could have been different–if the arms sent to Afghanistan in military aid was not seen as a sufficient basis to forge a civil society. The vague circumlocution “all the tools” may well come back to haunt both Biden and the world. For in the course of training and equipping a military force of 300,000 provided the basis for delivering much military support, America created spiraling costs of a global arms industry, even if the range of arms offered was not as well-suited to Afghani terrain or as protective as equipment offered NATO troops. (Oryxblog notes the poor protection these vehicles offer against feared improvised explosive devices (IEDs) compared to the MRAPs available to NATO forces in Afghanistan, and offered to police departments across the United States, but not offered to Afghan special forces.)

While the messy exit from Afghanistan appeared an uncoordinated relinquishment of control, the reliance on firepower and bombing raids as the sole veneer of stability in earlier maps of the region is revealed by the map, far more than the crumbling of a once united front of control. The GIF dramatically collapses the past four years as they unravelled over the months from May to April 13 to August 16, 2021; if it is only one of the several theaters of war, it seems to offer a compelling, if distorting story of a fall of provincial provinces in the state that the United States and the failure of rebuilding an infrastructure to which NATO committed from 2008, a loss that seems to ratchet up one’s sense of a lost opportunity. The failure of being able to control Bagram Airfield thirty miles north of Kabul–its control ceded to an Afghan army able to provide cover for fleeing Americans–was a final tragic episode in sustained lack of commitment in the ground game over more than two decades of ignoring the level of local trust that might have better created the nation’s infrastructure.

Indeed, the fraught planning of the withdrawal from Afghanistan, too easily blamed on a failure of “listening to those on the ground” who grasped the critical strategically critical nature of operations of drawing down the war rests is imbued with a sense of loss the mock up maps released by outfits as Long War Journal communicated to the viewers that reveal incomplete tactical awareness of a long-term ground game, but cunningly erased the costs of a war that inflicted such sustained damage on the country–and introduced escalating levels of violence and anti-government opposition–that little trust or loyalty remained after intense military efforts over all those years.

The costs of the pursuing of war and of bombarding much of the nation are never referenced in the maps of the advance of Taliban forces across the nation that suggest a strategic meltdown of ground-game. The “loss” of territory in the flip-book like sets of images recorded a real-time reaction to the transmission of power from American military camps, a transfer of power that was so poorly coordinated to not even allow the departing United States troops to secure Bagram Airfield, miles outside of Kabul, and the Hamid Karzai Airport to coordinate departures.

The narrative of Taliban advance is however mapped as an optic of loss. But the loss is almost hidden from visibility in the very same maps. The failure to compel Afghanistan to present Osama bin Laden and Taliban officers or training camps created the false sense of security of a show of power. It was based on and predicated the false concept of a submission of Afghanistan as best achieved by bloody bombing campaigns, drone strikes, and military incursions. For the loss of what we imagine territory held by our troops seems almost to cleanse the bloodiness of that past history. The advance of the Taliban into areas that were allegedly once in “government control”–or are labeled as such–reveal the spread of an ominous wash of deep crimson across the country as the tragic end of the War on Terror, something of a blood bath in the making, a spurt of pink and deep crimson red–as if the bloodshed was not cast by an American show of power.

Yet it erases the effects of a sustained numbers of deaths, violence and loss of blood, and the deaths of civilians that might have been prevented, already destabilized what was left of the civil government. The absences of governmental structures or webs of local allegiance allowed the superficial sense of stability that the provinces had retained, as American air power left them , and as stockpiling of arms and munitions in many former American bases provided the materiel for Taliban forces to advance even more quickly across space than they had ever expected. The insufficient supervision of arms that arrived at American bases suggested a landscape long permeated by naivite about the agency of Afghan people, and the utter the absence of training of local forces, that anticipated local governmental failure across the Forever Wars.

The readiness to point blame at a new President for not listening to the on-the-ground sources is concealed in the maps that suggest an abandonment of areas “under government control” as a betrayal–rather than a culmination of the long-term costs of a failure of effective governance of a land that long lacked centralized governance of the sort that is signified–but not demonstrated–by a map. The very national borders of what was shown to be a “nation” created a sense of false security, belied by the appearance of relatively few areas of insurgent activity across the terrain since 2018, and with little sense of the infrastructure destroyed by sustained bombing campaigns.

Afghanistan: Background and US Policy In Brief
Afghan “District Stability” and Sites of Insurgent Activity (2018)
SIGAR, January 20, 2019, Quarterly Report to the United States Congress

But the arrival of bloodshed to Afghanistan was something that the United States, of course, brought there on a scale no one had ever before imagined, flooding the nation with arms of a level of modernity as if they would defeat the society we had once called ‘tribal’ and incapable of tactical maneuvering or high-tech weaponry. As the United States assures we are As the area under “Government Control” contracts to an isolated the limited area, leaving us asking how the United States mapped it so badly. As the Government four Presidents promoted military ties contracts to a dot, but the dream of such an independent state now apparently eclipsed and recast into what may now seem more of an inter-regnum between two rulers–Hamid Karzai and Ashraf Ghani–in a Taliban regime. Rather than being cast as a restoration of power, the map illustrated to Americans the fall of an American dream, and an eclipse of the idea of nation-building as a primarily military prospect, that the US Army took over from NATO.

The hope to recreate firm borders of Afghanistan at untold expense fell like a house of cards. The Taliban’s strategic operations for controlling the very roads on which they once attacked American and NATO forces had destroyed the structures long before the troops retreated, as they had paralyzed the country’s movement and flexibility of its soldiers or national infrastructure. The fiction that was long nourished of an Afghan state that America had been able to try to fortify by the importing armaments–the “tools of war”–over more than twenty years. While the map is a visualization that derives from the work of the Foundation for the Defense of Democracies, and poses as a vision charting the erosion or loss of the coherence of a liberal state in the borders of Afghanistan, it both isolates the nation from its broader context in the Middle East and War on Terror–from the United States Central Command (CENTCOM) in Qatar, from the allies of Taliban in Pakistan and elsewhere, or the exit of many Afghan forces as refugees, or the seizure of weapons, humvees, and armored vehicles abandoned by the Afghan National Security Forces (ANSF) who left them behind as they fled north across the border or abandoned their posts. A map of the arrival of firearms and materiel–the procurement of Foreign Military Financing (FMF) and International Military Assistance (IMET) programs that American Presidents are authorized, and with Donald Trump escalated and Barack Obama had previously–would be as helpful, as it would track a vision of a significant increase of security assistance for geopolitical dominance.

Finally revealed: UK drone strikes in Afghanistan by province – Drone Wars  UK
UK Drone Strikes in Afghanistan
Tableaux Heat map of Drone Strikes in Afghanistan by Amderican Military under Presidents Bush, Obmaa, Trump tps://dronewars.github.io/narrative/Map of Drone Strikes in Afghanistan by American Military, Presidents Bush, Obama and Trump (2018)
DRONE WARS | Narrative
Total Drone Strikes in Afghanistan and Somalia by Presidents Bush, Obama, and Trump

The investment in drone escalation as a tactical relation to “space” redefined territorial dominance to replace one of community building, often confusing targets with the territory. Drone strikes not only served to “take out terrorist commanders”–but as if this did not destroy the stability of the fabric of a nation America was allegedly trying to rebuild since 2008–defined a view far from the ground. Over 13,000 drone strikes on Afghanistan alone–a minimum of 13,072 strikes killed in Afghanistan alone over 10,000–conducted by the United States Reconnaissance created a landscape being invaded by foreign powers. The dynamic of incessant drone strikes–conducted by a tool not owned by the U.S. military before the Forever Wars, and now showcased in targeted strikes is an invaluable prism to understand the mapping of the land that appears a hope for peace and end to the Forever Wars, as much as a lack of training, strategy, or American assistance. In ways that make drone strike fatalities pale, the recent estimate of 46,310 Afghan civilians–if below half of the estimated 95,000 dead Syrian civilian casualties of the War on Terror–suggests the way that the United States has benefited form the low presence of reporters on the ground.

The war in Afghanistan was located predominantly in the countryside, and across the many provinces that “fell” to a Taliban newly fortified by the windfall of armaments they accumulated as provincial cities, abandoned by the AFSN, fell. The logic that we had supplied the ANSF with sufficient arms to defend the territory reveals a confusion between the territory and the map–and the theater of combat and the situation on the ground. When Joe Biden marveled at how American-trained Afghan security forces Americans out-numbered Taliban fighters fourfold, and possessed better arms, the 298,000 armed ANSF were thinly spread and at low morale; if trained and armed by Americans, perhaps amounting to but 96,000, they lacked decisive advantage against Taliban force of 60-80,000 whose leaders effectively exploited internal weaknesses off the battlefield.

The real map–or the inside story of the progress of the Taliban across the nation–lay the perhaps not control over districts’ capitols, but the many well-stocked bases, airfields, and army depots long cultivated by American troops. The long-running bases across the country–sites with often mythic and storied names, like Kandahar and Bagram airfield, where tens of thousands of United States soldiers had been stationed from 2001–had posed a site of immense military materiel that the . The Bagram Airfield was a site for drones, of course, but also for storing cutting edge Blackhawk helicopters that the United States committed to Afghan forces, even if they were not well-trained in using or maintaining them, munitions, and firearms, even if the larger American aircraft and drones were withdrawn. As American forces withdrew, the rifles, ammunition, and tactical vehicles–as well as cars–were left at bases that the Taliban had long attacked–as Bagram—and had their eyes and were particularly keen. American commanders, as if intending to disrupt the withdrawal’s smoothness, disrupted the smooth transition by not even telling Afghans before they arrived at the Kabul airport–allowing the looting of laptops from Bagram, as a sort of bonanza, by local residents, before the arrival of Taliban forces.

Over three million items were abandoned by the U.S. Army in Bagram, from food to small weapons, ammunition, and vehicles–presuming that the “tribal” Taliban did not know how to use them–before they down-powered the entire base. Did the generals doubt that the Taliban could ever operate them, or just trust they were secure with Afghan forces? The weapons were poorly monitored. As ammunition for weapons not being left for the AFSN was destroyed, the abandonment of materiel, planes, helicopters and ground vehicles followed departure from ten other bases before Biden took office, often over NATO objections–that bestowed a huge symbolic victory of sorts to the Taliban of having driven foreigners from the land as they long promised, if not one of military materiel as wall. If American military argued “They can look at them, they can walk around — but they can’t fly them. They can’t operate them,” the ludic inversion of Taliban displaying armaments of Americans was profound theater of deep symbolic capital.

Taliban forces celebrate the withdrawal of US forces in Kandahar.
Taliban Forces Celebrate American Withdrawal from Kanadahar

If the hundreds of bases that Americans sent soldiers had long declined to dozens, the withdrawal of American forces without clear coordination with Afghans left a vast reserve of symbolic military material ready for the taking. How much was left at the bases closed in Helmand province, Laghman province, or Kunduz, as well as the bases in Nangahar, Balkh, Faryab and Zabul? Did these sites, and the reduction of American presence in Jalalabad Air Field, Kandahar Air Field, and Bagram not provide targets on which the Taliban long had eyes? The seizure of Kandahar provided an occasion for a triumphal procession of sorts, showcasing armored vehicles, as Blackhawk helicopters flying the Taliban flag flew in the skies overhead. In a poor country, the large prizes of American bases stood out like centers of wealth inequality, stocked with energy drinks, full meals, medical care and other amenities, and stockades were impossible to fully empty as the American bases closed from 2020.

Sites Supported by United States Military in Afghanistan, 2006/Globalsecurity.org

Few gave credence to Taliban boasts 1,533 ANSF joined the Taliban by May, or that June saw another 1,300 surrender, but the numbers of deserters only grew, expanding “contested” areas where Government forces lost ground without a fight. All of this crucial information is absent from the map, but we still believe, despite all we might have learned from Tolstoy, that generals and strategists determine the state of play on a battlefield, without knowing how the war was waged, or that the war was never seen as geopolitical–as it was waged–but across borders and rooted much more locally on the ground, as Taliban entered sites of former bases, and amassed arms caches in a drive of increasing momentum to Kabul–one of the only areas that wasn’t bombed so intensively, hoping it would be a reprieve from the violent bombed out landscapes on the ground.

For a war that was long pursued remotely, the image of territorial “loss” obscured the failure of engineering a transition to democracy. We have already begun debating the extent to which an executive decision-making shouldered full responsibility for the folding of the government of Afghanistan that followed the withdrawal of United States soldiers. –and air cover. We like to imagine that an American President has continued to steer global dialogue about the Afghanistan War, the remainder and reduced proxy of the War on Terror. Perhaps it is that we have a hard time to imagine a sense of an ending, and loose the ability to imagine one, and have lost any sense of a conclusion to the War on Terror that was long cast as a “just war,” against evil, and in terms of a dichotomy between good and bad, as if to disguise its protracted disaster. If we could never “see” the results of a an end to the War on Terror, Orwellianly, we were told it was not endless–Americans must have patience, said President George W. Bush as he promised us he had, to pursue a simple, conclusive, and final end to terrorism, assuring us the war would not, appearances to the contrary, grow open-ended, with a “mission creep” even greater than the Vietnam War. Barack Obama, after he presided over the military surge, hoped to “turn the page” on it in 2016. But any “exit” receded, and may not even be able to be dated 2021–as we imagine–but more protracted and indefinite than resolute–as Barack Obama, who presided over the military “surge”–hoped to “turn the page” and wind down by 2016. The logic of the war grew, as if deriving from Bush’s refusal to negotiate as was requested after the eight day of the bombing campaign, or move Osama bin Laden to a third country, but employ military might to force destruction of the camps of the Taliban, and delivery of all Taliban, fixating on the Taliban escalated the war far as an American struggle, far beyond attention to the situation on the ground.

The nightmarish reversion of Afghan territories was seen as the culmination of the withdrawal of American troops at large levels, almost achieved by President Obama in 2016, after the heights of the first “Surge” in 20011, but which was delayed by President Trump. The war that refused to end or conclude was never seen as a protracted struggle–or presented as one–but it was, and perhaps because of this never had any end in sight. “This is not another Vietnam” was announced by the father of that President, President George H.W. Bush in 1990. Americans changed the organizational structure and leadership of Afghan troops with each U.S. President, making it hard to conclude or manage, shifting how Afghans were trained, that must have encouraged a sense of clientelism and corruption of which the Afghan government became increasingly accused–and perhaps introducing a lingering suspicion of corruption and clientelism, more than bringing anything like a modern fighting army or New Model Army. There was never a sense of refusing to leave for fear that the failure that the maps depicted of the collapse of all districts of the new “Afghanistan” depended on continued American investment and support to endure.

Although the rapid reversion of districts to Taliban is far more likely to remain perceived by Republicans as a fiasco in leadership, the poor state of the country and ineffectiveness to work with the increased military materiel it was provided as if the army members did not have to be motivated and organized. The impossibility of mapping the geopolitical interests America felt onto the Security Forces–Lt. General William Caldwell IV reflected Defense Dept. opinion in the military when he assured the world Afghanistan National Security Forces were effective and trained, in fact “probably the best-trained, the best-equipped and the best-led of any forces we’ve developed yet inside of Afghanistan,” by June 2011, after a decade of military training, and only able to get better, even if American Generals were clear they would tolerate a degree of chaos, and didn’t want Afghans to be defining priorities, but only to instill a “particular kind of stability“: by 2016, National Security officials openly worried about the lack of any metrics–levels of violence, control over territory, or Taliban attacks that presented or projected confidence. The distrust, missed assessment and mutual mis-communications between American Generals who promoted and mistrusted Afghan troops whose efficiency they promoted created a disconnect between Americans as they downplayed the military ability of the Taliban, regarded as lacking sufficient air capacity or military prowess to command the nation or pose a threat to the Afghan Security Forces who folded before the Taliban’s military and threats of reprisals.

Is it possible to trace a transfer of military technologies and armaments in the twenty years since the crashing of airplanes into the Twin Towers by jihadist militants and the appropriation of sophisticated arms, night-goggles and humvees of members of the same Taliban who now occupy Baghdad? At the same time as American purchasers of handguns and firearms grew, the transfers of weapons and military firearms to the Afghan areas–UAE; Saudi Arabia; and especially Qatar–in a massive transfer of military technology that paralleled the emergence of the very groups cast as primitive rebels who had commandeered aircrafts to strike the Twin Towers into an efficient user of enhanced military tools and technologies, rather than the primitives who occupied the outer peripheries, but were both trained and prepared to occupy a nation’s center in disarmingly modern ways. Although the image of the plans flying into the Twin Towers presented an image of modernity versus premodernity, a lens through which the protracted war was pursued, as we cast the Taliban as “tribal,” and drove the Taliban into the opium production business, selling “modern” weapons and military tools into Afghanistan, the dichotomy of modern and primitive failed to present anything like a proper lens to pursue the war, although it was one American military had adopted on cue from an American President who had promised a “crusade” in no uncertain terms.

9/11

Perhaps the story of the War on Terror, in both its Afghanistan chapter and in other ways, demands to be written, when it is, as a massive transformation from the perspective of a shift of military engagement on the ground, and the military experience of the soldier, or what John Keegan called “the face of battle,” rather than the grand narratives of a conflict of civilizations in which it was framed. If the experience and strategic outlook Keegan emphasized might well be expanded, following increased awareness, to the long-term psychological and physical costs to those who were fighting, the erosion and fraying of the sense of nation and national motivation for combat must be included in the history as well, but the shift in war experience of the soldier must have shifted far more dramatically for how the “sharp end of war” appeared for the generation of the Taliban who matured in a terrain where American weapons had increasingly arrived in abundance to become part of the landscape of the state, and might be understood in terms of the shifting eras of military engagement from being attacked by bombers, targeted by drones–none of which were owned by the U.S. Army before the war, a telling index of engagement that reflects the way the war was in fact pursued at its sharp face. While in America disdain candidate Obama showed for how his opponent thought the military operated by measuring might by its navy or air force–“we have these things called aircraft carriers . . .,” suggesting one might use cavalry or bayonets as metrics in the Presidential debates in condescending tones–the shifting theater of military engagement of the Taliban, from placement of IED devices to the mastery of roadways and local influence–greater than the American soldiers on the ground.

From IED placement to suicide bombers, to rifles, kalashnikov, helicopters, and humvees, Taliban developed a new mastery of terrain, control of road networks for shipping materiel, to a n increasingly sophisticated tactical and performative use of arms and modern fighting tools that altered its experience and skill at the “sharp face of war” that we ignore, or attribute to outside assistance from Pakistani military, preferring to see the Taliban as primitive fighters without access to the technology America possesses and our provision of military “aid” as destined for “Security Forces” alone, rather than for a theater of war.

1. The current appeal of the clear mapping of the “fall” of Afghan districts to Taliban omits any senses of the line of battle. This is perhaps convenient for the military observers, who digest the war as it is pursued by American interests alone, even the NATO presence was increasingly defined in terms of the development of Afghan forces and democracy, although the “military alliance” shared by America and its Afghan ally is most often understood only in American terms. In mapping the “fall” of districts as if they were of purely strategic outposts in a geopolitical game, the map not only ignores the face of battle, but emblematizes the mis-mapping of American geopolitical interests onto Afghan interests. Despite the continued perhaps overzealous promotion of the skills of Afghan Security and the continued presence of American and NATO military failed to transition to Afghan Security Forces, even if we have continued to equip them with robust “tools of war,” without having trained them fully to fight our wars or to imagine their territorial mastery as anything like a strategic advantage for themselves.

Although the first elected President of Afghanistan, Hamid Karzai, was a friendly figure for Americans, trained in international relations and fond of Islamic philosophy, the promise invested in him as a “transitional figure” uniting “all Afghans” was better received by the British Queen and American President, Americans have been more concerned to map Afghan strategy as if it aligned with American interests, and a global war on terror, which Afghan Security Forces were deputized to adopt. We had long mapped the Taliban Resistance or “neo-Taliban” after the Taliban had been crushed as confined in the mountians, rather than in terms of its engagement with the “sharp face” of battle and its toll on both soldiers and the civilians who lived it. We saw the Taliban as an “insurgency” confined to the mountains as if these were the margins of the nation, and located them in Tribal grounds that were opposed to the vision of a central state–or as the inhabitants of a “Triangle of Terror” they had created.

File:Neotaliban insurgency 2002-2006 en.png
“Neo-Taliban Insurgency, 2002-6”
“Triangle of Terror”

In the images of Afghanistan’s “fall,” the “face of battle” is conveniently absent. In the visualizations of “district control” that were produced in the maps of the Foundation for the Defense of Democracy and reproduced across Western media, serving lambasted President Biden for some sort of dereliction of duty in concluding a forty-year old poorly thought out war? Democracy becomes something that the United States defends in these maps–or deputized Afghans to learn to defend–but the American President is suddenly seen as asleep at the wheel and not vigilant, the reverse of the image of a powerful Commander-in-Chief we desire, or the necessary and needed military “genius” who can strategically protect the national interests these visualizations reveal to have been tragically imperiled. And so we watch the “fall” of districts that had never gained independent unity, as if they failed to protect themselves from a theocratic opposition. We pretended that the failure was not the entry of increased materiel to the nation, but the global dismay at the levels of arms that are left in Afghanistan–more than are possessed by some NATO countries, and an unknown remainder of the $83 billion of materiel shipped to that nation–and the failure of Afghans to learn to use them against the Taliban, as if they were the exponents shaped by a Triangle of Terror, not affected by the shifting face of battle and “sharp edge” of war.

Increasingly, the promotion of the image of success in containing the Taliban that the U.S. Government promoted was doubted in the press, and seen as not an accurate reflection of the dominant role that the Taliban already had gained and controlled in Afghanistan, but which United States military assessments had rather dishonestly diminished, a scneario in which the maps of the Foundation for the Defense of Democracy provided a needed reality check as the true crowd-sourced story of the limited amount of control that the Afghan Government controlled. The extent to which the misleading military map by which the US government was seen as exaggerating and misleading the public on Afghanistan was US government is exaggerating and misleading the public on Afghanistan reflected the more bracing judgements of the right-wing Long War Journal, which valued its ability to present a clear-eyed view of America’s strategic interests in an unvarnished or not sugar-coated geopolitical assessment that America needed in the Trump era, when the confidence in our own government declined.

We did not ever map the “sharp edge” of war, preferring to view the nation from above, either against a “Triangle of Terror” we sought to bomb and domesticize, or parsed into tribal affiliations that became the preferred means of translating Afghanistan to an American audience, which almost acknowledge the failed imperial fantasy to project Afghanistan as a nation with clear sovereign borders, or to define an objective for Afghan independence that is not backward-looking, and rooted in the cartographic attempts of Great Britain in the nineteenth century, translated into the crucial “buffer” function that might contain Pakistan, and stabilize Central Asia in a geopolitical struggle defined by the War on Terror, and not the situation on the ground, or how Americans altered that situation by their increasing military presence and profile. As the Taliban slowly gained ground over the years, and in which the logic of waging war as a protracted struggle had ceased to be worth the $6.4 trillion American taxpayers have invested in post-9/11 wars through FY2020, in Iraq, Syria, Afghanistan, and Pakistan–and the escalating future costs that the war would mean. As we have lost sight of the logic of continuing the “forever wars” into the Biden Presidency, and the vision of a “just war” has become clouded and polluted in the Trump yeas, we have lost site of any ability to imagine the ground plan for the resolution of the continuation of a War on Terror or imagine at what scale such a conclusion might ever occur.

To be sure, the advance of Taliban was not how we wanted to imagine it as a restoration of “normalcy” or a status quo, and a rejection of a theocratic government for a secular liberal ideal. But perhaps the image of Afghanistan as a liberal state was indeed a failed project, and it only existed in maps that had outlived their usefulness or reflection of the area on the ground. The “fall” of Afghanistan reflects the inability to contain the Taliban from the nation, and the weird blindness that America–and the American military and perhaps military intelligence–have to the effects of war on Afghanistan on the ground, wanting to believe in a clear chain of command, recognizable in other militaries, in the AFSN. The GIF seems to raise as many questions as it resolves of the fall of Afghanistan’s provinces to imagine what that ending looks like. As much as the number of districts that speedily negotiated a resolution of hostilities with the Taliban, the fall of Afghanistan and painful and deadly withdrawal from Kabul has been cast as the final cataclysmic episode of the War on Terror, as if President Joseph R. Biden–and Donald Trump before him–had already decided on a military withdrawal from the region was both long planned, and was indeed a means of cutting losses and leaving a region to re-dimension or re-scale the War on Terror that had been fought.

The mapping of the collapse of Afghan districts to the Taliban, cast as sudden and without any sense of occurrence, seem to justify the continuation of that war, but track the erosion of a territorial war, long morphed into a struggle whose aims are unclear. Maps that suggest a “country” of Afghanistan as land that was lost help us imagine that the authority of US forces might have trumped geography. And so we are retrospectively questioning the reporting of intelligence on the ground, trying to read the records of intelligence, or debate the false confidence projected by U.S. military through the final years of the campaign, as if this were an American decision, and a reflection of American global authority, as a microcosm of the image of the United States in the world theater, and seem to present the reassuring picture of a scenario of global politics in which wars are still fought on the ground, and which the loss of the War on Terror was not a failure of the American military, but the ceding of land by Afghans themselves who lacked ability or conviction to fight the war against theocracy that was largely scripted by American Presidents and military–who were unwilling to share their sense of their mission in Afghanistan with Afghan leaders, certain, as last as 2016, that Afghan “priorities are different from ours”–perhaps making it impossible for Afghans to take charge, as leadership of the nation was less of a gridded battlefield that became the dominant graphic that filtered, processed and mediated the withdrawal of American forces across the mainstream media.

In viewing a nation as a battlefield, we are not looking at the right map, or perhaps not looking at the right maps at all–or at the role that the arrival of military weapons played in the rendering “Afghanistan” all the more difficult to map. Perhaps the exportation of arms to the Middle East and to Afghanistan in the years since the nation’s invasion provides a better legend, and indeed a necessary legend, to map how control slipped out of the increasingly corrupt Islamic Republic of Afghanistan, established in 2004 after the United States as it assumed control of most of the country, which has been ceded–and destroyed–by the advance of the Taliban. The drawdown of troops in the country from the heights of the first surge under President Obama of 10,000 men and women has in fact been declining for years, but we have not noticed, or even looked closely at it. Yet the compelling nature of visualizations of “control” over individual districts by 2020 seemed a sudden loss of the nation, a progression of a fall of provinces culminating in the Taliban taking control over almost all of Afghanistan’s provinces, and entering Kabul, perhaps as Afghanistan seems a fitting theater or field for the master-trope of America’s imperial decline. Indeed, the attention in media maps to the delusion at an apparent absence of groundplan for American extrication or withdrawal.

These graphic visualizations are hardly accurate maps, but conveniently omit all information about the “sharp end” of battle, falling back on the geostrategic place of “control” over provinces–is this by the flags flying in their capitals? what is control in a war-torn area?–that can be understood as an element of a “Global War on Terror,” rather than the ways that the war was fought. As uncomfortable as such images might be, we prefer the “objective” GPS image “mapping” control, not pausing to ask what they miss or distort, or process the war in an episode on the War on Terror, or a lost field of battle for Afghan independence which it has long ceased to be.

The time-lapse visualization in the header to this post, of Afghan provinces shifting from “Government Control” or “Contested” to “Taliban Control” offers an image of dramatic impact, as if it were real-time, compelling as a tragic narrative, but erases the deep roots of the “lightning drive” of Taliban forces, fueled in large part both by absence of administrative unity and a massive uncoordinated influx and abandonment of arms–both left to Afghan Security forces or in caches. So strong was the flow of arms to Afghanistan and Qatar from the United States that the Biden administration only suspended arms contractors from delivering pending arms sales. Caches of arms left abandoned by Afghan Security Forces and, presumably, American military who had left them to be used by Government forces, not only destabilized the landscape of local government, but amplified a landscape by men with guns long fed by the over $40 billion contracts for firearms and ammunition flowing to the Middle East since 9/11. But if Biden assessed the Afghan Security Forces as being “as well-equipped as any army in the world” in contrast to the Taliban–and greatly outnumbering Taliban fighters–the long-term distrust of Afghan priorities and concerns left them with little sense of a common grounds for defense. As Americans were making similar assurances, Afghans were already fleeing in July to Tajikistan, where over a thousand Security Forces had already fled.

The arrival of the Taliban did not embody the victory of a theocratic to a secular regime that Americans have cast the War on Terror. The arrival of the Taliban as an armed infantry group, with its own modern military power, is an unwritten history, but was fueled by the arrival of an increased number of weapon that arrived in the region, and the transmission of military technologies across borders in ways that American governments could not perhaps imagine. Whether they were not exposed to the arrival of high tech arms of US manufacture in previous years or not, the idea that the arms that allowed Taliban members to arrive with speed in Kabul and negotiate a ready capitulation of districts, perhaps with Pakistani assistance, the seizure of of an unaccounted number of weapons caches turbocharged the advance to Kabul, in ways that not registered adequately in daunting images of the shift in districts to Taliban control. Such visualizations map a checkerboard of district that seem to track the government “control” of districts that image the erosion of a secular vision of Afghanistan. The division of Afghan lands into “districts” is almost a shorthand for the localism of Afghan politics, an admission of the difficulty of knitting together a secular state from into a centralized state, was never resolved by occupying forces or the Islamic Republic of Afghanistan. More than confirm the alienation of ethnic groups from the vision of an allegedly secular government, inter-ethnic divisions have dramatically grown in the place of a coherent strategy for forging a multi-ethnic state, emblematized by an unknown CIA analysts’ map of circa 2017, that continued to map a nation bound by the red line of Afghanistan’s historical border–the “Durand” line, negotiated in the last decade of the nineteenth century–a conceit bisecting a region of Pashtun dominance and mountainous terrain that poses questions of Afghanistan’s ‘borders’ as much as it answers them. Was the retention of this imperial cartographic imaginary not suited for the sense that Afghanistan, as Samuel Moyn argued, offered a chance for the “last gaps if imperial nostalgia” in the post-Trump years, that was, improbably, able to play across the political spectrum?

This image has an empty alt attribute; its file name is cia-map-durand_line_border_between_afghanistan_and_pakistan-1.jpg
CIA Analyst’s Map of Afghanistan, Pashtun dominance in Blue “Tribal Belt” (CIA, c. 2007)

Is it possible that the among of weapons funneled into Qatar, United Arab Emirates, and Saudi Arabia that have disguised the cost of the War on Terror to some degree have created a huge concentration of arms in Afghanistan.

Stockholm International Peace Research Institute (SIPRI)

If a rationale for the increased ability of Taliban members both to manipulate negotiations may lie in their attention to negotiations at Doha, their use military weapons may lie in the increased arrival of arms in the region. The escalation of imports and sales of arms to Afghanistan–many not registered or under the radar–escalated in the course of the Afghanistan War, and reflect a growing geopolitical significance that the nation was given to the United States, rather frighteningly similar to Vietnam, if the withdrawal from Afghanistan has been most focussed on as the greatest similarity between these two long wars, both fought at considerable hemispheric remove, only conceivable as they were logistically mapped by GPS. In both cases, wars were pursued across a complex and often oversimplified logistic chain, pursuing an elusive vision of global dominance or geopolitical strategy, whose obstacle appeared a lack of geopolitical “vision”: but was the presumption of a possibility of “global military dominance” that mismapped both military projects from a purely American point of view. The flattening of the effects of waging war only seems to have increased, paradoxically, as the geopolitical significance of Afghanistan overwhelmed the well-being of its residents, blotting it out, as the country modernized by force as it became a focus of the arms trade.

2. The investment of American taxpayers’ monies in the region was astounding, and hardly democratic, so much as a tantamount to a massive dereliction of national vision amidst the faulty reprioritization of mission creep that may be attributed as much to the military-industrial complex as to leadership or governance. Over half of all American foreign military financing arrived in Afghanistan directly by 2008, but aid had long flowed to Mujahideen and other insurgents through Pakistan, yet in later years billions of substantial materiel flowed via Qatar, location of the $1 billion CENTCOM headquarters where Americans coordinated all air operations in Afghanistan–a small nation that became the tenth largest importer of arms in the world, after South Korea, Iraq, United Arab Emirates, from 2015-19, largely from the United States, with contributions from France and Germany, jumping by 631% from 2010-14–becoming the eighth-largest market share in arms imports for 2016-2020 behind South Korea.

The absence of attention to the situation in the ground is nowhere more apparent than in the GIF that is the header to this post, which reveals the “fall” of Afghan districts to the Taliban from April, 2021. We map the hasty conclusion of the long war in GIF’s of districts, as in the header of this post, the flattening of a country that has been divided for over forty years, a form provided by the Long War Blog. The division of inhabitants of the land, or the effects of previous combat on the nation’s infrastructure and sense of security, is hardly rendered in the shape-files that flip from one hue to the other, suggesting a “lightning” advance of a militarized Taliban, evoking a sudden loss of a territorial advantage for which Americans long fought, and for which Aghans are to blame. Yet as much as the linked maps of “district control” suggest a traumatic collapse of the Islamic Republic of Afghanistan, the ally of the past five American Presidents, the maps collapse or elide the deep disturbances the war and importation of arms has brought to the territory that lies beneath the map, or oversimplified visualization of regional control.

Financial Times via Global Investigative Journalism, “The Taliban’s March”
source: Foundation for the Defense of Democracy’s “Long War Journal” by Mike Roggio

The quandary of designating Afghan regions by questions of “control” presumed a sense of stability and allegiance more akin to an idealized military map than to the situation on the ground. The checkerboard image of areas of “government” and areas of “Taliban” control became thinly veiled covers for a Global War on Terror in which the United States defined itself on the side of the good, that was current in a variety of maps long after the First Surge. In the context of the broad drawdown of American troops after the First Surge, as US troops level fell below 10,000 and Afghan Security Forces were celebrated for their effectiveness, the Taliban made steady gains on the ground. But the maps that suggested “stability” in government-held areas created a cocoon from which to affirm stability of a regime that never had broad institutional support as if the dangers it faced were from an “insurgency” 2002-6, and promoted an image of government control within the outlines of a national map, arriving from outside of a nation that still had retained its integrity and clear bounds as if they were able to be preserved.

“Neo Taliban Insurgency, 2002-6”

Even as Taliban presence was more clearly established than we liked to map, the image of the Taliban as outsiders in Tribal lands created a sense of justifying a “civilizing mission” that was understood as more pacific than military, underpinned by a myth or conceit that the disciplined bodies of American warriors would beat the undisciplined bodies of the Taliban. This myth was confusing the goals of the military occupation, but creating an increasingly real edge for Afghans who experienced much more fully “the sharp edge of war” both forged increased bonds between the members of the military and the fighters and the landscape among the generations of Taliban fighters, and their logic of responding to a military strategy American generals mismapped on a geostrategic checkerboard–the very checkerboard that Foundation for the Defense of Democracies encouraged us to understand the success, progress, or challenges of combat, and indeed control their fears and responses to technologies of combat imported to the region by the United States.

Fall of Districts by July 1, 2021, documented by Fazl Raman Muzary, from local media and on the ground reports

The deep concern of a lack of “strategic vision” was not the best way to understand military engagement of Taliban forces, or to cast the compact shift of district loyalty after the American withdrawal.

But these terms provided the terms to condemn and bewail the broad geopolitical military failure read into the maps of Taliban advance in August, 2021, apparently confirming that the AFSN had built up as our surrogate was unable to “face” the Taliban militia we continue to cast as “rebels” or “insurgents.” But the negotiated settlement allowed te rapid fall of a number of districts, as while it required the Taliban cease hostilities with NATO and American troops who had negotiated the settlement, the terms allowed Taliban forces to concentrate on negotiating settlements with local regions, exploiting divisions and existing corruption of Ghani’s Afghan government, boosted by the concessions to release 5,000 prisoners in the past, and the opening of jails in districts whose centers they captured or negotiated a solution.

Donald Trump may have escalated the arms trade into the Middle East to levels far beyond his predecessor, but the frustration of his successor has perhaps provided a far more clear-eyed assessment, perhaps more than he is given credit. “We provided our Afghan partners with all the tools — let me emphasize: all the tools,” U.S. President Joseph R. Biden sternly told the nation, in a combination of evident dissatisfaction and apparent exasperation, in justifying his rapid withdrawal of U.S. troops from Afghanistan. The vague circumlocution “all the tools” may well come back to haunt both Biden and the world. For in the course of training and equipping a military force of 300,000 provided the basis for delivering much military support, America created spiraling costs of a global arms industry, even if the range of arms offered was not as well-suited to Afghani terrain or as protective as equipment offered NATO troops. (Oryxblog notes the poor protection these vehicles offer against feared improvised explosive devices (IEDs) compared to the MRAPs available to NATO forces in Afghanistan, and offered to police departments across the United States, but not offered to Afghan special forces.)

It is hard to tally or come to terms with the human cost of post-9/11 wars. Over 9,000 Americans have died, or the hundreds of thousands who returned from the wars, injured in body or psyche, the roughly 6,200 U.S. military personnel, contractors, humanitarian workers and journalists killed in Afghanistan since the U.S. government invaded are left off the map, but the legacy may be greatest for the huge amounts of military materiel shipped into the Middle East–arms that helped in some way to “modernize” the current Taliban, who may have received training from Pakistan intel–as well as the huge losses of population and infrastructure in Afghanistan, where about 71,000 Pakistani and Afghan civilians are estimated to have been killed–a staggeringly disproportionate number in crossfire, bombing raids, drone attacks, suicide bombings in Kabul and other bases, IED’s and night-time raids by NATO or American troops.

Continue reading

Leave a comment

Filed under Afghanistan, Afghanistan War, Kabul, military maps, War on Terror

The New Arid Regions of the United States

The southwest and states east of the Sierras magnify the effects of global warming in the intensity of their aridity. But global warming reveals a new relation of regions to overheating, and reveals the depths of inflexibility to accommodate water scarcity, as well as the tragedy of its effects. As aridity of the soil and reduction of groundwater reaches unprecedented scales, our passivity is accentuated as we are suspended before maps that try to visualize unprecedented aridity magnified by global warming and its magnifying effects.

For the cascading effects of warming on the land and environment might be mapped in ways that cannot essentialize the greater “aridity” of the region, but the effects of increased aridity of soil, air moisture, and dry air on a region that we have remade into a region of food supplies, agriculture, and livestock, but, beyond, on hydropower. While the Colorado mountains long provided an effective basin to gather rainwater for western states that have been funneled to state reservoirs for agricultural irrigation, the man-made irrigation networks were drying up as the snowpack determinedly fell, and warmer temperatures evaporated what snowpack fell.

The logic of this longstanding pattern of appropriation of water from across the Colorado Basin was in a sense begun with the Hoover Dam, but was, writ large, organized by very process of appropriating water rights to redistribute water that had been enshrined in California from the turn of the century, circa 1914 and the policies of filling reservoirs to redistribute water rights. While we have considered appropriative water rights a distinct feature of how water is redistributed unique to the Golden state, the appropriation of water rights by reshuffling of water in state’s now precarious supplies how diverts over 99 million acre feet of surface water diverted along the Sacramento and San Joaquin Rivers to farmland, created a powerhouse of national agriculture. Much of the 75 million acre feet that flow from reservoirs across the state evaporates before it arrives at crops, however–far more that actually reaches the farms or cities.

–and the growing heat of the Great Plains have likewise diminished surface flow of the Colorado Basin already reduced by diminished rainfall. The increasingly warmer atmosphere of recent years has created a new “Arid Region” of the United States, of even greater aridity than when it was first mapped by John Wesley Powell in 1890, and the renewed aridity of the region not only challenges the calculus of water distribution according by appropriative rights that is structured by the Interstate Compact, but the very logic of redistributing water.

The past two decades have seen the departure of seven trillion gallons from Lake Mead, the largest reservoir on the Colorado River, holding rainflow from the Upper Basin before it crests the Hoover Dam. The drop will trigger hydrologic stresses across western states, as ever ever-increasing amounts of water are sucked up into drying out air and atmosphere, requiring more abundant irrigation of croplands and grazing grounds. This new and expanded “Arid Region” suggests a return of the repressed, returning at even greater scale and aridity to haunt the nation by a lack of groundwater again.

Evapotranspiration Rates in Colorado River Basin, Landsat 13

that is not to say it ever left. But the solutions of diversion have been undermined by the cascading effects of climate change and increasing temperatures, across an expanse of irrigated lands where water from the Upper Colorado, as from the western slopes of the Sierra Nevada, are funneled to cities, farms, and irrigation projects, and used to generate electricity. Even as Californians and westerners face the threat of further fires more destructive than any in recorded history–potentially enough to energize an implausible recall effort in the state of California–we face the problems of managing not only historic drought, mandated energy shortages, reduced water supplies. The climate crisis appears to have provoked a deep crisis in leadership, but one without easy means of resolution.

The most improbable political candidates–global warming skeptics after Donald Trump’s heart–have argued drought, wildfire, and electrical storms reveal Gavin Newsom’s lack of leadership, even as they stridently object to aggressive climate legislation aimed at emissions reduction as restraining the free market business,– preferring a free market approach for all climes that would be the laissez-faire redistribution of water to the highest bidder, monetizing a scarce resource to consolidate financial profits and gain in response to diminished water supplies.

As more water is being released from Upper Basin reservoirs to make up for the shortfall in Lake Powell, but the shortage of water in Lake Mead–the largest reservoir in the United States–to less than 40% capacity by 2022 will mean reducing water for lower basin states like Arizona of 812,000 acre-feet, Nevada–by 21,000 acre feet, and Mexico, by 80,000, that have led to the call for new “water markets” to be created across the western states. Indeed, even western states no longer carry the brunt of increased use of freshwater for irrigation–

High Country News

–the demand for conserving water in agriculture is increasingly incumbent on western states, so much so that the shift to less water-intensive crops–like California’s almonds–at a time when many crops require more irrigation, and a shift toward fewer acres of pasturage for livestock–good luck–have become a necessity. Increasing the efficiency of irrigation systems is necessary–ending customs of flooding fields, increasing drip irrigation, center-pivot irrigation, or micro-irrigation, in a New Deal for agriculture, even regulating irrigation systems before water markets price rural communities out of their accustomed access to freshwater. The increased trend toward shifting the distribution of water by “water markets” from lower- to higher-value use is dangerous for farmers, and indeed all rural areas, but also for the western ecology, as it would be the most difficult to preserve water in rural communities or farming areas less able to pay for pricing of water for higher use-value, although they currently consume over 70% of the water in the Colorado basin, or encourage sustainability in regions that are increasingly facing realities of sustained drought, if not megadrought of unprecedented intensity.

United Stated Drought Monitor for Western States, October 2021

Yet the systems of allocating water from the Colorado River by a system of dams, diversions, and canals have led to broad calls to end further projects of water diversion, as the diversion of water to western states may be drying up itself by up leads to calls for new policies of allocating water, not based on the highest bidder, as the river we have made increasingly mobile across boundaries will be divided or redivided between agriculture, urban use, indigenous Americans, and land trusts, as we are in need of redefining the working basis for conserving the redistribution of water rights beyond capture and diversion, and outside of existing water markets and appropriative water rights within states. While the Bureau of Land Reclamation has run the reservoirs, dams, canals, and hydroelectric plants and contracting with individual districts, a broad reconception of practices of regulating water markets, allocations of water, and costs of large-scale water diversion, as demand for water outstrips supply.

Yet as increased farmers are withdrawing water from the ground, or from rivers, from California’s Central Valley to the Lower Colorado River Basin, in Arizona, New Mexico, and Nevada, the need to reduce the eighty percent of water dedicated to agriculture across the west will demand new practices of conservation, beyond what John Wesley Powell mapped, in 1880, when he advocated new practices of land use, as climate change increasingly destabilizes the Basin, including the thirty sovereign Native American tribes along the river basin. The need to manage demand and riverflow that will begin with the start of the “Tier One Shortage” from 2022, will introduce new rules on water-use and supply that stand to reduce the amount of water flowing to Arizona by a third. Water diversion from the Colorado River has transformed the land west of the hundredth meridian by re-engineering its flow to make the “desert bloom.” Yet the recent dramatic reduction of rainfall, river flow, and increased aridity of the lands, leave us contemplating the viability of relying on water diversion.

John Wesley Powell, “Arid Region of the United States, showing Drainage Zones” (1880)

The new arid region is reflected in weather maps, but will be a region of radically reduced piped water and a new landscape of hydrologic diversion. If the “Arid Region” was mapped in earth-tones of clear distinction as a cautionary way by explorer and geologist John Wesley Powell, to alert the government to the distinct climate of lands west of the hundredth meridian, the recent area is both based on more detailed and specific remote sensing records, often from satellite observation, but suggests a far more complex area to manage.

For the western states are linked, both by projects of water diversion, and by hydropower, to a region where rainfall and snowpack has declined, and far less water enters into the river-flow of the rivers whose diversion allowed the expansion of agriculture and livestock across the western states. Due to global warming, the earlier “arid region” expanded, returning bigger and better than ever since it was described as extending west from the hundredth meridian by John Wesley Powell, in one of the foundational maps of climate aridity. In today’s parched California, dangerously low levels of rainfall across the central valley seem to belong to the Arid Region. But we have hardly come to terms with its new expanse or migrating edges. The “lands of the ‘Arid Region'” that Powell had hand-colored with earth-tones to communicate the dramatically falling rainfall west of the hundredth meridian long ago mapped a biting response to the eagerness of homesteaders to Go West, cautioning about constraints on water-rights that division by states–rather than drainage districts–would bring. If current rainfall maps of USDA or EPA seem to engage in dialogue with Powell’s old polemical cry, the limited traction of mapping policy against increased pressures of climate change place most maps in a sort of Scylla and Charybdis, located not in the Straits of Messina, but the scissors of decreased rainfall, rising temperatures, and lack of groundwater retention.

1. The problems of managing water rights and ensuring flow are now far greater than what Powell’s creative palette of before the fact overlays even imaged was able to depict, and was a puzzle beyond the interlocking pieces of drainage districts that he–as if akin to the first puzzle-boards composed of hand-painted maps, as this forty-nine piece puzzle map of ca. 1849, painted by Kelly & Levin, of the similar region, that curiously compressed the western United States. Was Powell’s map indicative of the difficulty of solving the puzzle of allocating water resources across arid western states?

Puzzle “Map of California, Mexico, Texas and the United States,” ca. 1849, Kelly & Levin. Boston MA

While the puzzle pieces rarely echoed the shapes of individual states, undoubtedly because o the difficulty of their cutting the contours of states, the puzzling of how the rivers of the west would align with states in this roughly contemporary 1880 Milton Bradley map-puzzle, an “Outline Map of the United States,” posed by including light blue rivers across a map with little sense of varied topography.

ca. 1860, M. H. Traubel, Lith., Philadelphia PA/American Antiquarian Society

In contrast to the resolution of assembling individual pieces of a map of fixed bounds, the expanded arid region mapped by remote sensing spans a farther territory and expanse, and raises deep questions of access to water or even soil moisture in a region that developed as an agricultural breadbasket and locus of husbandry of livestock.

The growing puzzle broached by how the water supply of the west will be reassigned is rarely faced or addressed, although it is ruminated upon as the sub-text–or super-text?–of terrifying maps of rising aridity and low rainfall across the western states, that magnify a new “arid region” with less clear suggestion of an outcome of land management but pause before the cyclically compounded effects of rising heat, low soil moisture, limited run-off, and the specter of drastic irrigation cuts.

Current remotely sensed maps use far less clearly set boundaries or edges of water-shortages, but pose similarly pressing puzzles of how to resolve the appropriative logic of water rights, as drought intensity reduces the water that once flowed from the “upper basin” of the Colorado, feeding the river and redistributed water, and even more surface water is lost to evaporation.

Snow drought is worsening the American West's water woes | The Economist

The puzzle of hydrological access to land-water has become so curtailed across western states, that increased pumping of groundwater risking depleting aquifers by draining vital aquifers, irreparably damaging rivers and riverine waters. The New Arid Region, afflicted by far more aridity and low soil moisture than at any time, parallel to increased global suffering of warming and increased heat, the persistence of private water “rights” to agrarian expanse stand increasingly on a collision course with global warming throughout the new arid West in ways we have yet to address, even as we recognize that we are facing a climate emergency of the sort without precedent in modern memory.

2. No single visualization can, perhaps, adequately come to terms with the unprecedented aridity of the recent years. For no visualization can fully capture the cascading and magnified effects of declining water and soil health, and their effects on ecosystems, as much as on livestock or irrigated crops: the distance from reduced irrigation and new climate specters demands an intensified map. But the terrifying nature of the intense aridity of western states in part lies in how we have seem to forgot the semi-arid nature of the region. The deeper effects of a drying out atmosphere were evident in the huge deficit in water vapor in the past decade during the “fire season” from August to September, dramatically unlike how fire fighters navigated the same terrain in previous decades, when many fire containment strategies were developed and many active firefighters had trained. The map is one that should raise immediate fears of the loss of a landscape of future irrigation, and the need for tightening agricultural belts and shifting our conceptions of food supply and water budgets–as well as the same landscape’s increased combustability and inability to manage or control by an old playbook.

Decreased Water Vapor Present in the Air in Past Decade from Two to Three Decades Previous

The previous month has brought an even more pronounced record of drought across the Upper Basin of the Colorado on which so much hydropower relies, as do other schemes of water diversion.

US Drought Monitor for Colorado River Basin, September 23 2021/Brad Rippey, USDA

The revelation of a new intensity of exceptional drought in many pockets of the Upper Basin of the Colorado River presses the bounds of how we imagine dryness, aridity, and their consequences, even as we rely on older methods of fire-fighting, and fire-prevention, and outdated models of water diversion and energy resources.

The historical denial of what John Wesley Powell had already called the “Arid Region” west of the hundredth meridian, has become a snare for ecological disaster translating into a process of the drying out of long-irrigated zones, with consequences that the nation has not been able to comprehend–and demand a New Deal of their own to replace the diversion of water and generation of energy in the Hoover Dam. Or have we forgotten the intensity of a differential of climate, soil moisture, and increased aridity that Powell long ago mapped in order to illustrate the new regime of government its unique atmospheric conditions it would require, using his uniquely designed palette to hint at the best way to organize the region of water scarcity according to the units that its drainage districts–rather than the state lines surveyed by latitude and longitude?

John Wesley Powell, “Arid Region of the United States, Showing Drainage Districts” (1890)

Powell had explored the canyons, rivers, and plains, as he addressed the Senate Select Committee on the Reclamation of Arid Lands in 1890, he crafted an eloquent seven-color map of rich earth-tones to impress readers with the sensitivity of the region’s texture and urge restraint for expanding the westward flow of homesteaders with hopes to make the desert bloom. Indeed, by circumscribing areas for which sufficient water in this “Arid Region” would be able to providently allow future settlement, Powell neatly divided areas for settlement in a region by hydrographic basins collecting sufficient rainfall for farming. Whereas rainfall maps of previous years mapped a blank spot of water scarcity, Powell hoped to direct attention by a devising a map of the region’s subdivisions that called attention to its soil quality and decreased moisture, focussing on its distinctly variegated terrain in ways foreign to Senators in Washington. Powell hoped to convince who were removed from the region to acknowledge the commanding constraints created by these drainage districts for all future agricultural development and settlement–an unpopular position that ran against the notion of allocating free land in an age of expansive homesteading. If the image of a “drainage district” was foreign to existing state lines, Powell’s image of an “arid region” long haunted the geography of the American West–and contributed in no small part to the subsequent reengineering of the waters of the Colorado River.

In light of the dramatically increased aridity now endemic to the western states, Powell’s map gains terrifying relevance as western states enter severe drought, placing the breaks on once-expanding developments across western states. Powell’s map articulated a historical vision of the limited infrastructure of water in the American west. While the technologies of irrigation that allowed such a massive project of damming and canalization only later developed, did his map inspire the need for a project of such scale as a better model of land management? The intensified aridity that afflicts the western states responds not only to low levels of rainfall. We continue to hope groundwater depletion that afflicts the lower basin won’t extend to the Upper Basin of the Colorado River that has captured water on which so many farmers rely–and thirty-five million north of the border and three million living in Mexico depend, across its Lower Basin. The escalating megadrought has created pressures across the overpopulated west that the water-sharing model Powell proposed for drainage districts cannot resolve, but the distinct forms of water management he advocated have been forgotten, as the declining water level on the Colorado River seems a time bomb as its waters have fallen so far below capacity that while the waters that drain from the Upper Colorado into Lake Mead, the largest reservoir in the western states, are only 37% full, and Lake Powell stands at 34% capacity. As less and less water enters the river system of a drying-out west, the future of the river on which so many rely for irrigation and energy is all but uncertain.

The water-level of Lake Mead, the largest reservoir in the US and a critical source of water for millions across the Southwest, has fallen 140 feet since 2000, a third of capacity.  Can we come to terms with the increased aridity across the west that the drying out of the Colorado River may bring?   The western states are haunted by the return of the "Arid Region" John Wesley Powell once mapped.
Lake Mead, May 2021

Demand for water in the upper basin and older technologies have meant far less water reaches the lower basin, but what does has been redistributed across western states–absolutely none reaches the ocean at the river’s old delta. Supplies of surface water and groundwater barely provide for the border region, as the overdraft of the basin’s aquifers have made trans-border water management a crisis often overlooked in favor of water management north of the border. As unprecedented soil aridity currently seems to run off the rails, after three summers of no rainfall have depleted soil moisture, may remind us how we have missed the lesson of Powell’s map of instilling new set attitudes toward the land, as the volume of riverflow consistently dropped as it crosses the Mexico border since the filling of the Glen Canyon Dam.

Does selective amnesia underlie how we map the drying out of the west? Most data vis of rising temperatures and low rainfall across the western states is already magnifying and escalating the effects of unprecedented heat over twenty years in a deeply melancholic vein, daunted by the scale of dryness across such an interstate expanse, and passive before an absence of atmospheric moisture that seems a modern casualty of global over-heating. If we were already “living in the future” in California’s frequent and increasingly extreme fire regimes, the multi-hued data visualizations electrify the landscape–and not with power or hydro-energy, but by the all-too familiar color ramp of the extremes of climate change we have been trying hard not to normalize. These images chart a landscape that has gotten away from us, outside seasonality changes, making the American West a cautionary case study for global climate change inspires melancholy.

The additive logic and graphic syntax of maps, long before the separate map-“layers” that accommodate information from GPS, provided a basis to define the fungibility of water and the emergence of “rights” to water across the Arid Region, enabling the idea of governing the transference of water and water “rights” across the region, that separated water from the landscape and environment. The flow of water had long been understood and reconstrued in the west by a logic of irrigation needs–and the “rights” to unpolluted water for livestock raising, pasturage, and agricultural needs of land owners–that was removed from conserving groundwater needs. The increased nature of the fungibility of water as able to be transacted across basins, state lines, and counties reflects the legal fiction of considering water as a “good” tied to the needs of property owners, that, long before global warming, had already sanctioned the removing water from the ground.

If we use metaphors rooted in temporality that try to come to scale with the new era of global warming that cut down and perhaps minimize the era of water scarcity. in which we are entering–“heat waves,” for example, that broke records in states from Washington to Idaho in June and July, breaking or matching records of hot temperatures, the levels of aridity that have allowed the ground to grow arid and degrade have not only led to a spate of western wildfires, but have changed the levels of soil moisture over the long term in ways we have difficulty to map in the scale of our weather maps, or even the maps of the U.S. Drought Monitor, as the cascading influence of such unprecedentedly dry conditions–where stresses on river water create extraction of groundwater that stresses aquifers and groundwater supplies–can be scarcely imagined, or confined to the conventions and color ramps of weather maps.

We have struggled for decades to process the cascading effects of waves of unprecedented heat that over time have produced a drying out of soil and reservoirs over the past twenty years, resulting in an expanded and far more destructive fire season and parched lands whose effects we cannot fully come to terms or comprehend, as we have not seen or experienced the extent of dryness of subsoil, soil, and low rainfall which the US Drought Monitor seems to have mapped, as drought expanded not across the entire Pacific Northwest, from Oregon to Idaho, or 86% of Idaho–by the land’s combustibility, impossible to read without premonitions of lost forests–including old growth forests–melancholic fears more than tinged by an acute sense of a lack of agency.

The sense of struggle with an absence of agency–at the same time as an almost moral urgency–reflects the difficult to process such absence of water as a landscape we have inherited from the rapidly accelerating dynamics of climate change. The history of the increased aridity is all the more poignant as a source of melancholy not only because exceptional drought was the standard before President Trump, and a national emergency before his Presidency. We have failed to register this national emergency with the same immediacy, even as the theater of the border was magnified in disproportionate ways in public discourse on migration. The sense of melancholy is compounded as the map seems haunted, if only tacitly, and perhaps without acknowledgment, by the fact that the head of the USGS in 1890 admonishingly illustrated virtually the same basins now suffering severe and moderate drought as distinguished by semi-aridity–if the current levels are nothing like those faced over a century ago, when the transition of public to private lands. We have recently mapped the substantial threat of increased aridity to the Great Plains–less than a tenth of whose croplands are irrigated–where farmers depend entirely on rainfall to grow soybeans, sunflowers, cotton, and winter wheat, the fear of greater “dry spells” as anthropogenic emissions drive decreasing rainfall and groundwater reserves–a term that tries to convince us they are not permanent–led red flags to be drawn in broad brushstrokes in those states, where extreme and exceptional ‘drought’ .

But climate change has created a new concept of “water stress”–stresses best be pictured not by the isotherms of weather maps, but the watersheds and drainage districts that were the basis of Powell’s revolutionary map, and matching the very region of the Arid Zone where the soil scientist Powell turned viewers’ attention to the crucial index of ground and soil moisture, the true determinant of the future of agrarian settlement and the future of food. The regions determined of greatest future stress were the very basins that Powell mapped, and suggest the relevance of his map, as well as his caution of the difficulties of governance in an area of severe water stress-stress being understood and indexed as a relation between supply and demand, as well as rainfall, in national watersheds.

3. The “Arid Region” of the Untied States had been austerely and admonishingly described by John Wesley Powell as a geologist to caution against the administration of its future settlement with a level of clarity that reveals his Methodist upbringing. It is hard to know how clearly we can ever parse aridity, in an age when rising temperatures have unremittingly drained soil of water. As if informed by a deep respect for the map as a clarity of record, possessing the power to reorient readers to the world by preaching a new relation to the land, Powell had placed a premium on cartographic form as a tool to re-envision local governance–and prepared his striking eight-color map of the limited rainwater that arrives west of the hundredth meridian, the eastern border of what he baptized as the Arid Region, an almost zonal construction akin to a torrid zone.

The imposing title of this reclassification of the interior of the United States revealed Powell’s own keen sense of the map as a visual record of the territory, whose transparency as a record of the quality of the land would be a basis for all discussion of settlement. Powell parlayed his own deep study of the geography of the Colorado Basin to query the value of parsing the administration of water rights by state lines in 1890, convinced of the need to oversee later apportionment outside the jurisdiction of the arbitrary boundaries of western states, but joined them to his sense of duty of preparing a legible map of striking colors to convey the constraints and difficulties for its future settlement– not only by the scarcity of the threads of rivers curled against its topography, but the few watersheds.

Powell trusted the map might mark the opening of the “Great American Desert” in order to alert the US Congress that the dry lands west of the hundredth meridian was a divide. Even if the meridian no longer marking as clear a divide of reduced rainfall, as we confront the growth with unprecedented degree of global warming of a parched west–both in terms of reduced rainfall and declining soil quality–it may serve as a model for the map we need for the future governance and administration of already contested water rights. Powell’s place in the long story of soil quality reflects how neatly the American west as a microcosm of global warming is rooted in the conversion of public lands to private ownership, into which warming has thrown such a significant wrench.

Arid Region of the United States, Showing Drainage Basins (1890)

For the Arid Region’s aridity has since been unremittingly magnified, producing a region more arid than we have ever experienced and struggle to find an adequate color ramp adequate. But we would do well to try to map the forgetfulness of that arid region, even as we confront the quandary of the stubborn continuity of sustained dryness of a megadrought enduring multiple years, compounding the aridity of the soil, and multiplying fire dangers–and the conditions of combustibility of the region–far beyond what the west has ever known or Powell imagined possible. If aridity of soils and poor land quality has spiraled out of control due to “global warming,” raising questions about the future of farms and livestock, the absence of groundwater and surface water alike, global warming demand we shift from national lenses of water shortage to beyond American territory,–but also to discuss the warping nature of national lenses on the remaking of the sediment of the west–and Colorado Basin.

The difficulties of parsing river-flow by “states” as helpful political aggregations for future settlement was rebutted by the map, which sought to direct attention to the aridity of the ground’s soils to orient its administration in a region where water was destined to remain front and center on settlers’ and residents minds for the foreseeable future. The subsequent attempt to jerry-rig the question of scarcity of water by entitlements that rely on re-apportioning unused water escaped the constraints Powell located in the basic common denominator of groundwater.

As much as the region needs to be mapped outside a national context–despite the national nature of climate tracking–the hope of revealing imbalances of the drought indeed exist across borders, and impact water-sharing agreements, much as the smoke from recent northwest fires has traveled across the Pacific northwest. National territory is as meaningless an analytic category for global warming, or water scarcity, which, this blogpost argues, exists in a global contest of migration, as the migration or transborder transit of fires’ smoke.

4. The conditions of aridity that Powell described in the Colorado Basin and its neighbors offer an oddly productive image of the dryness of the ground, in an era before irrigation, that may be useful for how we can come to terms with the fear of a suspension of irrigation across western states. But it is as if the very definition of aridity was forgotten, as infrastructures of irrigation have re-mapped the region that John Wesley Powell in 1890 mapped as an area of difficult agrarian settlement, as farmlands of agrarian fertility and wealth. Powell proposed to view the “arid region” of the United States east of the Rockies with a clarity approaching scripture in a powerful eight-color map to instructively show how limited water constrained settlement of the region after surveying the Colorado Basin.

Powell probably imagined his map in somewhat revisionary as much as rebarbative, reorienting attention to the dry nature of the soil of the semi-arid region of the Colorado Basin by parsing it in areas by which the availability of water constrainted the settlement of the “open” government lands of the west, obscuring that they were seized from indigenous, to correct the mythic geography propounded by official state-sponsored geologists. Unlike Powell, most state geologists had boosterishly endorsed a site for future pasturage, to be enriched by unknown artesian springs, and ripe for settlement by homesteaders, and Powell’s map posed a more tempered image of resettlement that would obey the laws of the availability of water in the Colorado plateaux and other regions he knew so well, cautioning against the encouragement of settlement and sale to prospective farmers in ways that have improbably made the map something of an icon of conservationist thought. Against promise of prospective bucolic lands of pasture, the dry colors chasten viewers by communicating scarcity of water of drainage basins.

The arid region that Powell correctively propounded was long inscribed in the psycho-geography of the United States to be forgotten, but the arrival of irrigation infrastructure allowing irrigation of western states continues to inform, even in our own era of global warming, the return of the boosterist sloganwhere water flows, food grows,” that is still raised in Northern California’s San Joaquin Valley, to protest “cuts to farmwater” in the recent order of an “emergency curtailment” across rivers of the Sacramento-San Joaquin Delta watershed — essentially the entire Central Valley. The recourse to an engineering “miracle” of making water flow uphill and redistributing more water from reservoirs contest calls for conservation–and only demand the further construction of dams, reservoirs, and water storage for better irrigation. The very promises that the flow of the Colorado River would irrigate lands, that made good on the promises made to homesteaders by describing the region to settlers as a New Canaan, where the growth of future streamflow and even rainfall that had never been documented, would make it suitable for the expansion of animal pasturing and farming, suggests a mythic geography of timeless bounty has replaced its actual conditions.

Friant-Kern Canal Flowing past Kern Dam/Septmeber 2020, Eric Paul Zamora, Fresno Bee

The mythic geography led to a rewriting of America’s irrigation infrastructure that in itself may be one of those pieces of infrastructure just no longer adaptable to extreme climate change. And as we face the scale of the national emergency of water shortages about to be triggered by falling reservoir levels, the crisis of using and recycling water, and the inefficiency of desalination plants of riverwater and groundwater, on which the world currently relies–and were predicted by the US Bureau of Reclamation back in 2003 to provide a “sustainable” solution to the dwindling water provided by the Colorado River, which had allowed the unexpected expansion of the settlement of western states. While desalination plants currently generate worldwide over 3.5 billion gallons daily, with 50 million gallons produced daily in Carlsbad, CA alone, desalination plants in one hundred and twenty counties, only half using sea-water, its energy expense justified as Colorado River decreased, promoted as a “sustainable and drought-proof water supply in Southern California” in an era of climate change, as if to calm our concerns at the dramatically decreased groundwater of western states.

Reclamation scientists assured the nation in 2016 of future recharge in the Upper Colorado Basin would offset temperature increases in their modeling scenarios through 2099, projecting basin-wide precipitation, the fears of the persistence of a mega-drought of extreme aridity with little recharge that may last decades has left fifty-sevens million living in drought conditions across the west according to the U.S. Drought Monitor, that has brought a new era of mega-fires. The thin blue line of the Colorado River is but a crack or thread coursing through a combustible landscape in this recent map of the expansion of unprecedented extreme drought in western states from National Geographic:

For all the disturbing and disquieting elegant if terrifying spread of deep red isotherms in Riley D. Champine’s map, the consequences of such exceptionally below-average levels of precipitation and aridity are difficult to comprehend as cumulative and deep in our nation’s history, as well as the effect of man-made climate change.

The utter saturation of this data vis of growing dryness of a region where rain far below previous norms fell forces the viewer to process an undue range of measures of aridity that they must struggle to process-if the deep orange and reds approaching emergency warning to suggest that surely a climate emergency is at hand. The absence of text in the visualization invites viewers to acknowledge they stand an eery remove of familiarity with an irrevocably landscape, posing unspoken if also unanswered questions about hydrological infrastructure in the Colorado basin, and greater west, that all but erases the geopolitical formation of this landscape–interruption of a rich color ramp at the southwestern border compartmentalize the large-scale decline in precipitation apart from national categories; but the danger lies in its focus on the economically developed north, more than the global south, as if it lacked adequate resources to prudently respond to groundwater shortages, but as an emergency for the developed world.

The focus of the climate emergency is on a large scale, daunting the possibility of individual response, but focussing on prudence at a local level, even if its scale is not defined, questions whether state politics can even resolve the intensity of the dilemma of declining rainfall levels below a thirty-year norm, a deviation on so broad a scale to be impossible to process save in local terms, but that omits the way the basin has been engineered as a site where groundwater now all but fails to accumulate, increasing the basin’s deep aridity more than the color ramp reveals.

The trust that Powell placed in his maps stand in sharp contrast to the “purple” coloration of regions of extreme heat introduced across western states to suggest so many “red-flag” warnings of excessive heat. In a year already tied with 2017 for receiving “excessive heat” warnings from the National Weather Service, already in early summer at a rate that is increasingly alarming, purple designates the need for caution when leaving air-conditioned environments, and suggests the booming of electric cooling across the west: the metric of a prediction of temperatures reaching 105°F for a two-hour stretch has paralleled the debate in Washington on infrastructure spending that suggest a similar disconnect that Powell confronted when he tried to describe the need for constraints on planning settlement west of the hundredth meridian in 1890.

Four Excessive Heat Warnings issued from late May 2021 have introduced yet a new color to prominence in National Weather Service maps, the new deep purple was introduced in weather maps in 1997 as a venture of the NWS into health alerts; rarely used in other weather maps, which in recent years have shifted from urban areas to large stretches of the nation, shifting from a use of red to designate high temperatures to purple to designate risk of triple-digit temperatures, especially in man-made surfaces like asphalt (able to rise to 170°-180° Fahrenheit–territory of third-degree burns–or cars which can rise thirty degrees above air temperature.

Heat Advisories, July 11, 2021/National Weather Service

During the decade before 2003, the water-level of Lake Mead had begun to decline precipitously, inaugurating a historical decline that led it to fall to but 35% of its storage abilities. While the decline was not more precipitous than the two earlier declines in its water-levels in the reservoir from the mid-1950’s and mid-1960’s, the current decline in storage capacity of what is the largest reservoir of water in the United States has raised the unthinkable and unimaginable arrival of water cutbacks, as Arizona’s share of the Colorado River’s waters will be reduced by 7%, and Mexico–where the Colorado runs–will lose 5% of its share, in a scenario never foreseen in the dam’s history, but that reflects the increased aridity of the watershed from which the Colorado River draws. The decline to 1,075 feet in the reservoir’s depth that triggered the Tier 1 reductions in flow may only be a harbinger of the arrival of future Tier 2 reductions, should Lake Mead drop to 1,065 feet, as is expected in 2023, and raises the fear of a Tear 3 reduction, should the lake level fall below 1,025 feet, reducing the water allocated to western cities. In ways that the infrastructure of irrigating the Arid District of the United States could never have foreseen, the arrival of the driest period that the basin has ever experienced in 1200 years has brought longer periods of drier weather without rainfall that have reduced the riverwater that fills the reservoir.

The declining level of Lake Mead plunged below average lake elevation of 1173 feet, by 2003, in ways that should have sent alarms across the west, were we not consumed by a war against terror. The Bush administration’s attacks on global warming grew, questioning the science of global warming and the dangers of increasing aridity. But the disconnect between the expectation for irrigation by the farming industry and farming states was dismissed, with global warming and climate change, as temporary shifts that wouldn’t alter the landscape of irrigation or river flow.

Robert Simmon, based on data provided by the U.S. Bureau of Reclamation

Continue reading

Leave a comment

Filed under border studies, climate change, data visualization, Global Warming, mapping global warming