Pokémon GO and three kinds of capture

Nick Seaver
Jul 25, 2016 · 6 min read

These are some reflections on Pokémon GO in relation to anthropological theorizing about traps, written hastily after a recent panel on “anthropological traps” at the European Association of Social Anthropologists meeting. Thanks to Alberto Corsín Jiménez and Rane Willerslev for organizing, to audience members for their probing questions, and to my fellow presenters for their stimulating work: Anthony Pickles, Christina Leeson, Patrick Laviolette, Stefan Helmreich, Martin Holbraad, Heather Swanson, Robert Wishart, and Hermione Spriggs.


The first kind is obvious: you gotta catch them all. Find Pokémon in the environment, throw pokéballs at them, and collect them. Really, you collect them in two ways: first materially, to serve as combatants under your control — objects through which you exert your agency against other players in the game; second informatically, as entries in the pokédex, a ledger of Pokémon you have successfully captured. These twinned modes of capture should be familiar to anthropologists, who know well the “butterfly collecting” tendencies of certain periods in the discipline’s history, when a primary goal was amassing information about the peoples of the world into the ethnographic record (a sort of collaborative, Euro-American pokédex). This informatic capture went hand in hand with material capture, as understanding was used for purposes of enclosure — enabling colonial administrators to destroy or encircle those peoples collected.

Like other traps, pokéballs contain a model of both hunter and prey — the ball is physically suited to human throwing (or, as the case may be, flicking across a screen with a finger) — but the heterogeneity of the entities they capture (from tiny Weedles to enormous Snorlaxes) raises a question: What happens inside? Pokémon not only vary in terms of their size, but also in strength. They can escape or resist capture, and some of the items available in the game are intended to combat these capacities. The pokéball opens, a bright white light shines out, and, if the random number generators line up in your favor, after a brief wiggle of the ball, the creature is caught. “Gotcha! Pidgey was caught.” This suggests, I think, that the relationship between player and Pokémon is primarily informatic — you catch an entry in the pokédex, in which respect all the creatures are the same in outline. Capture is domestication, transformation from a wild, environmental encounter (“A wild Pidgey appeared!”) to knowledge — a row in a ledger, a glowing white light zapping between domains.


As with traps more generally, capture goes two ways: you, the player, catcher of Pokémon, will find yourself caught up (if the trap works) in the game, your attention captured, your desires anticipated, the game tuned to keep you signing in. Visit a cluster of pokéstops on a weekend, and you will see a crowd of people peering into their phones, which are likely plugged into external batteries (the game not only captures your attention, but also your electricity). That the central mechanics of the game are entangled with random number generators (the likelihood an attempted catch will succeed, the qualities a Pokémon will have, etc.) makes evident the family resemblance to the slot machines Natasha Dow Schüll describes in Addiction by Design: if you want to capture humans, physically and mentally, carefully tuning random number generators (game designers call this “balancing”) is a good way to do it.

Ironically, the regular failure of the game’s servers, which mean that much of the time, one cannot even load the game or pursue one’s goals in it, might heighten the effect of such traps, serving as the kind of variable reward that behaviorists learned could drive pigeons insane. Players themselves serve as trappers of each other as they battle for control of Pokémon gyms, the regular transfers of power serving as a kind of engine in which players may be trapper or trapped. The gyms are thus another kind of object to be caught in the game.


The game’s loading screen warns of a third kind of capture: “Remember to be alert at all times. Stay aware of your surroundings.” This is useful advice if you want to avoid falling into a thatch-covered pit (or just down the stairs), but it takes on extra meaning given some news reports that have come out in the wake of the game’s sudden, astonishing popularity. Players have reportedly been robbed, either caught while their attention was captured by the game, or “lured” to pokéstops by villains who then proceeded to mug them. In-game “lures” are items that are used to draw out Pokémon at particular locations, marked on the map by a cascade of pink confetti. In physical space, these lures serve to draw people to catch the Pokémon, who are evidently sometimes caught themselves by people who have reconfigured the game as a different kind of trap.

These three modes of capture all suggest that capture occurs at the “material and epistemic interfaces between worlds,” as Alberto Corsín Jiménez has argued—at the informatic boundary of the pokédex, in the map of the game, or at the edge of the “magic circle” of gameplay. You, in trying to catch Pokémon, your attention caught by the game, may yourself be caught out. This chaining effect seems typical as well: often the entities we try to catch are themselves in the process of trying to catch something (an animal pursuing a lure, for instance, or in Robert Wishart’s Gwich’in ethnography, a lynx trying to catch a rabbit that it thinks might have been caught in a hunter’s rabbit trap, but which is in fact a lure in a trap designed to catch the lynx). In the manifold adjacencies of worlds, entities capture each other, like in the classic behavioral trap: the raccoon whose hand is caught in a jar because it is trying to catch hold of something within the jar, caught until it lets go — until it breaks the chain of capture, which otherwise will end with the trapper taking hold of it.

The raccoon makes clear the blurriness between mental and physical capture, not only at the extreme of addiction (in which mental captivation, via medicalization, comes to be understood as physical captivation and thus a legitimate object of interest), but rather in the ambivalent middle, where mental and physical tendencies meet in environments, and in the material and epistemic webbing of infrastructure. The persistence of the captured leading to further ensnarement is a common theme in trapping (salmon pushing further into a pound-net; elk running farther into a snare; and even the chimpanzee, whose tragic persistence is not in moving forward, but in staying still and investigating, resulting in death by falling weighted spear). Like the raccoon, the trapper is caught up in broader structures out of his control — tendencies in game design, for instance, which are contoured by culturally particular ideas about accumulation, collection, and the kinds of value that will be valued by those whose valuations have come to matter.

Rane Willerslev reports that Siberian Yukaghir hunters consider three kinds of trap: The first are the prosaic, ordinary kind (a heavy rock propped up on a stick, for instance). These require work and always risk failure. The second kind are a style of magical trap to which animals are mysteriously drawn, flinging their pelts at it. These cannot help but succeed. The third are wild, shamanic traps, so outlandish in their design that only a shaman could make them work, and which, if an animal escapes, results in their acquiring human properties, such that you might find sables smoking cigarettes in trees or bears making dough for bread by the river. To its detractors, Pokémon GO seems something like the second kind of trap: an otherwise unremarkable construction to which its prey is inexplicably and irresistibly drawn. But, being sober-minded, we might do better to suggest that it in fact an example of the first, a construction that was always at risk of failing to capture its desired prey (as the game’s predecessor, Ingress, did). I do not know what a shamanic game would look like—one that was so enticing that those who escaped came out transformed in the likeness of those who tried to trap them, a glitchy collapse through the material semiotic interface between worlds.

Nick Seaver

Written by

automatic anthropologist, sound/media/technology studier, resonator