During the spring and summer months in the more arid parts of Central Asia’s Ferghana Valley, the distribution of irrigation water has a distinctive rhythm. Domestic plots are allocated water for a few hours every two weeks, on a rotating cycle of up- and downstream homes. The shout from a neighbor that “the water has arrived!” (suu keldi) is met with frenetic activity. During the few hours that water has been released along one’s section of roadside ditch, it needs to be actively channeled through the garden’s corners to ensure that it doesn’t simply drain off downstream.

The natural work of gravity is assisted by a complex assortment of improvised plugs and sluices made of roofing tiles and plastic sheets. Gullies are flattened with a hoe to “chase the water” (suu aiduu), a term that speaks to water’s unruly materiality and the urgency of distribution. Get distracted at the height of the planting season and your tomato plants are liable to wither, followed by your potatoes and, eventually, your apricot trees. Fail to control the flow during your allocated slot and you may find parts of your garden flooded and others parched.

Networks of sluices, gullies, and improvised side channels provide an insight into the dynamics of neighborhood irrigation in a semiarid environment and an insight into the demands of mahalla sociality around a zero-sum resource. We can, as Morgan Liu (2012, 132) does for urban Kyrgyzstan, track the branching networks of canal irrigation to read relations of cooperation and what he calls “collective coping” (as well, perhaps, as the limits of both). But sluices also focus our attention on blockages and flows between riparian states in ways that are pertinent to thinking about sovereignty both volumetrically and barometrically.

A sluice is a distinctive technology of allocation. In the dynamics of upstream/downstream water management, it is a technology that influences intercommunal relations. Too little flow and gardens remain parched. Too much and roads and fields can get washed away. Diplomatic relations in Central Asia often escalate and subside in line with watery projects: a dam that might improve electricity output to one state can deprive its neighbor of water; an initiative to divert water for the benefit of cotton cultivation might turn downstream villages into dust bowls. Commenting on Kyrgyzstan’s geopolitical predicament in relation to its oil-rich neighbors, my acquaintances would often respond by referencing not just Kyrgyzstan’s natural abundance of water but also its ability, as the upstream partner, to determine geopolitical flow. El bashchy bolguna, suu bashchy bol—“Better be the head of the water than the head of the people.”

Sluices work through the creation of pressure differentials. In the Isfara basin, which begins high in the Alay Mountains, transecting Tajikistan and Kyrgyzstan multiple times before ending in Uzbekistan, an aging system of vertical sluices directs water from the Isfara river through an elaborate system of concrete canals according to a protocol of seasonal use dating from 1980. Near the village of Khodji-A’lo where Kyrgyzstan’s Batken district and Tajikistan’s Isfara district meet in a complex section of chessboard-style border, the sluice system or “water-distribution knot” (vodoraspredelitel’nyi uzel) is the de facto boundary, even though the land on which this metal-fenced Soviet infrastructure was constructed remains disputed between the two states.

Open the sluice-gate on the right, and the enormous force of the Isfara river is set gushing off into late-Soviet concrete channels toward the Tortgul reservoir, an artificial lake built in 1971, on which most of the downstream Batken district depends for its irrigation. Close it, and the water flows on to Tajikistan and Uzbekistan toward cotton fields and parched postwar mining villages. Standing on the sluice’s bridgehead at the start of the Tortgul canal, the sheer force of raging water hitting metal is a reminder of just how much watery power is at stake in seemingly arcane determinations of flow, pressure, population size, and irrigation need. Perhaps it is no surprise that the Vodkhoz, the state agency responsible for water allocation, is shrouded in rumors of deliberate mismanagement, or that the Soviet interrepublican protocol, conceived at a time when the population of the valley was about half what it is today, is rarely observed to the letter. Perhaps it is no surprise, too, that when interstate relations became fraught in January 2014 over the construction of a disputed bypass road, it was precisely this canal-head sluice construction that was the epicenter of military crossfire.

In his study of vertical geopolitics, Stuart Elden (2013) draws attention to the need to think about the vertical as a critical dimension of security, both in its subterranean and atmospheric dimensions. The stakes of much contemporary politics hang precisely on the question of who controls skies, tunnels, sea, and ice in all of their intransigent, three-dimensional materiality. Along the Isfara valley, where international borders cross and recross a glacial, fast-flowing river in a largely arid mountain zone, it is only by engaging the international border architectonically that we can gauge the stakes of contention over prospective infrastructures that would cut, transect, or divert fragile transboundary flows. A sluice, holding back water one moment, releasing it in a torrent of violent flow the next, is part of this three-dimensional geopolitics.

But a sluice’s power is as much barometric as volumetric. It acts; its modality is pressure. Its infrastructural capacity derives not just from the fact that it extends below the surface of a river or above the height of the flow, but rather from its ability to block, channel, and flood, and in so doing to create a volume with force. This has implications for “securing the volume” as an object of anthropological concern. Volumes, like lines on the sand, are produced, not found. The politically urgent questions, in the Isfara basin as elsewhere, are who gets to close the sluice gates and to what ends.