Sensory Co-laboring: Mine Detection Dogs and Handlers in Humanitarian Demining in Colombia

Abstract

Demining has not been an exclusively human affair. Mine detection dogs have been indispensable in the work of detection and in the slow but essential effort to regain trust in mine-suspected landscapes. Famously renowned for their extraordinary sensory perception, physical strength, and mental traits, they are part of human-nonhuman units training and working together to perceive explosives’ odors. This article considers the role of these units, known in Colombia as binomios caninos, in the strenuous task of mine clearance. Drawing on eighteen months of ethnographic engagement with global and local humanitarian demining efforts in Colombia, it examines detection choreographies and daily interactions, proposing to think of their joint work in terms of sensory co-laboring. Bringing anthropological work on collaboration between worlds, sensory labor, and animal work into dialogue, this composite term foregrounds detection as labor and as a result of human-nonhuman cooperation. It also highlights the asymmetrical field in which these collaborators converge and the divergent desires, affects, and attachments that mobilize their participation in demining. Mine detection is conceptualized as a sensory task through which dogs and humans intra-act, both together and apart. Recognizing this partial connection allows us to rethink how humans and other creatures are ontologically reconstituted and how overlapping histories of warfare and humanitarianism, legacies of animal behavioral practice, and instrumental-affective interactions shape these reconstitutions.

A Detection Choreography

The canine team trains in the afternoon, when the burning midday sun has given way. This team of international dog handlers and mine detection dogs, originally from Bosnia and Herzegovina, Ethiopia, and Norway, was recently assigned to Colombia to investigate areas suspected of being contaminated with improvised land mines. Deployed mainly by rebel groups during the sixty-year armed conflict, these makeshift devices have killed or maimed more than 12,380 people.¹ From 2008 to 2014, Colombia was the second-highest mine-affected place in the world. The canine team is part of Norwegian People’s Aid (NPA), a well-known demining organization that handles breeding, training, and circulation of mine dogs and the preparation of dog handlers for humanitarian demining efforts worldwide. In Colombia, NPA coordinated the technical and political aspects of the Pilot Project of Humanitarian Demining, a demilitarization initiative that occurred during the 2012 to 2016 peace talks between the national government and the Revolutionary Armed Forces of Colombia (FARC-EP).

Although the canine team trains together, exercises are conducted in separate human-dog pairs known internationally as “mine detection dog units.” In Colombia, these units are called binomios caninos, a local idiom that has expanded from military to humanitarian and civilian circles.² With no English equivalent or presence in global standard demining vocabulary, the term refers to a duo (binomial) while emphasizing dogs, whose olfactory capacity is critical to the detection task. I usually observed the binomio of Ibrahim and Hamilton (fig. 1).³ Ibra, as he is known, a former Bosnian soldier in his forties, now serves as NPA canine team’s chief technical adviser and trainer for the Pilot Project. Hamilton is a nine-year-old Belgian Shepherd Malinois, a working dog breed famous for its extraordinary sensory perception, physical strength, and mental traits. Reared and trained in Bosnia and Herzegovina, he is the oldest and most experienced mine-sniffer in the canine team assigned to Colombia.

Their daily training is a forty-five-minute, highly choreographed, joyful encounter in a one-hundred-square-meter imaginary minefield. In El Orejón, the first demining site of the Pilot Project, the training field is in a steep green meadow in front of the village school; in Santa Helena, the second demining site, it is a grassy flat area next to the worn-out local soccer field, about five hundred meters from the base camp. Their dance starts with a simple bodily gesture: Ibra leans forward, extends his arm, and slowly opens his closed hand, his fingers subtly touching the grass. This is the command “Go and search!” While Ibra stands outside the “safety line,” Hamilton enters the “minefield,” searching with great concentration—an illustration of the unequal vulnerability framing their joint detection work. Hamilton’s attentiveness and thoroughness are clear in his corporeal language; his position, pace, and speed are signs of interest. He walks in a straight line, snout touching the ground and wagging tail pointing skyward. Ibra, engrossed in Hamilton’s movements, typically ignores the humans around him. As Hamilton moves down the lanes, Ibra offers praise and instruction in Norwegian: Rolig! means “Slow down!” Kom søk designates “Return and keep searching.” “Venstre!,” Ibra finally yells, ordering Hamilton to return. Ibra briefly touches him and sends him back, commanding, “Søk!”

The multispecies dance culminates in one last movement: the indication. In front of the targets (pieces of explosives placed before training), Hamilton wags his tail, sits down, and stays still, staring intently at the ground. Ibra walks over with a Kong—a red, bubbled plastic canine toy/training device—and quickly puts it in front of Hamilton. He slowly passes his hand through his fur, from head to tail, once, twice, three times. “This is to relax him,” Ibra explains. Hamilton remains concentrated on the toy. Ibra moves his hand to communicate that Hamilton can take his reward. Kong in mouth, Hamilton swings in what seems to me the epitome of happiness.⁴

This is a condensed version of numerous dog-human training sessions I observed during my time in the Pilot Project, the unprecedented demining initiative that framed both Ibra and Hamilton’s collaboration and my research on the nonexplosive capacity of land mines. Dubbed a “peace laboratory,” this project brought army and FARC-EP guerrilla soldiers together to clean up explosive remnants of war, as a testament to their will for peace and reconciliation. In the demining effort, dogs were indispensable to detection and the slow but crucial effort to foster trust in mine-suspected landscapes. Together with their human companions, they helped identify sixty-six explosive artifacts and clear more than 40,363 square meters of war-contaminated land.

Drawing on eighteen months of ethnographic fieldwork in Colombia, this article examines the detection choreographies and daily interactions of mine-sniffing dogs and their handlers, what Haraway may call “dances of relating.”⁵ It considers the tangled political, material, and affective relationships that constitute their mutual conditioning and make this multispecies collaboration possible. I trace the infrastructures, economies, and encounters that allow dogs and humans to become sensory professionals and collaborators in the risky business of demining, foregrounding the mutual yet deeply unequal character of their joint becoming. This becoming is grounded in an anthropocentric framework that broadly defines the terms and purpose of the parties’ participation. I call this joint detection work sensory co-laboring, a composite term that brings recent anthropological work on collaboration between worlds, sensory labor, and animal work into dialogue. It emphasizes detection as labor and as a result of human-nonhuman cooperation, highlighting the asymmetrical field in which these collaborators converge and the divergent desires, affects, and attachments that mobilize them to participate in mine clearance.

In this analysis, mine-sniffing dogs inhabit ambivalent and even contradictory positions, since they are valued and make value as “organisms shaped for functional performance in human worlds.”⁶ In the words of my human interlocutors and in my observations, dogs emerge as a product of biotech breeding and training practices, a cost-efficient tool within the global humanitarian market, and a working companion species with whom one bonds. Although dogs are considered experts and partners “with equal and complementary roles” in humanitarian demining, they are also thought of as biological artifacts and tools that can be bred, trained, and relocated according to humanitarian ends and markets.⁷ Attentive to dogs’ complex subject/object positions, I am interested in emphasizing their sensory work; even if we do not know what work is for them or what interests them, their engagement in detection activities entails subjective involvement.⁸ Their subjectivity manifests in their capacity to learn and develop new dexterities, incite consideration, trust, and respect, and affect their human partners’ sensibilities.

I begin with an exploration of the global humanitarian canine demining industry and the short history of this technology in Colombia, offering insight into the canine team’s workday in the Pilot Project. I then examine puppy training to illustrate how dogs and humans cultivate dogs’ olfactory fascination and kinesthetic disposition toward the Kong, the training object at the core of mine detection. Looking at the development of handlers’ haptic and hermeneutic capacities, I show how touching well is critical to effective land mine detection and the circulation of this sensory expertise. I demonstrate ethnographically that dog-human collaboration does not presuppose warm attachments, sensory equivalences, or symmetry of power. I argue that sensory co-laboring has as much to do with convergence as it does with divergence; addressing their incommensurabilities allows us to better track the uneven distribution of obligations and vulnerabilities, as well as the distinct interests driving them to detection work.

Sensory Co-laboring: Cooperation and Divergence between Humans and Nonhumans

I borrow the term co-laboring from Marisol de la Cadena, who uses this term to describe her conceptual and empirical relationship with Mariano and Nazario Turpo, two Quechua speakers with whom she engaged in long-lasting conversations.⁹ Contoured by disciplinary practices and regional belongings, their dialogues were riddled with linguistic and conceptual translations, controlled equivocations, and ahistorical conditions and archives. Although their differences disrupted each other’s analytical semantics, they did not impede communication. Rather, they allowed them to explore translations that created “a shared space of sensations, practices, and words.”¹⁰ This space did not rely on “sharing unique and cleanly identical notions” or avoiding or capturing excess.¹¹ Citing Strathern, de la Cadena frames this relationship as a partial connection: “My world was included in the world that my friends inhabited and vice versa, but their world could not be reduced to mine, or mine to theirs.”¹² Instead of maintaining the separation brought about by this condition, de la Cadena and the Turpos worked together to understand each other and co-labor using tools from each of their worlds: “From this place of coincidence (which is also one of divergence), I meet the differences that made a connection between us possible.”¹³

In my analysis, co-laboring speaks of the sensory labor carried out by dogs and humans in training on war-contaminated fields. This labor is at once common, unequal, and divergent. Co-laboring emphasizes their efforts to understand each other and partially align their sensory worlds to jointly identify explosive devices, a task that neither can do alone. They must train constantly and be attuned to each other’s corporeal sensitivities and capacities—hence, this co-laboring is sensory. As scholars in anthropology and science and technology studies insist, sensations are not merely passive, biological responses; they are active, intricate practices embedded in social, cultural, and technological arrangements and knowledge-making regimes.¹⁴ Through these daily, repetitive choreographies of detection, dogs and humans learn to affect and be affected by certain materialities, bodies, and relationships. Here affect adopts two different, related meanings: it refers to the relational capacity of bodies to (im)mobilize, (re)orient, (de)attach, and (un)make,¹⁵ and it speaks to the emotional investments and feelings that emerge between subjects and communities and bind them together.¹⁶

As dogs and humans hone their sensory skills, they work to establish channels of embodied communication and create common grammars and vocabularies.¹⁷ While each has a concrete and lived milieu, what ethologists have described as Umwelt, they strive to create and rehearse shared meaningful signs. Consider bodily signs such as the indication, or verbal commands such as Kom søk and Rolig: given that mine detection unfolds in spaces of life or death, these signs must be consistent, univocal, clear, and effective. This obligation happens even if they “are training each other in acts of communication [they] barely understand.”¹⁸

Co-laboring also emphasizes how the convergence of dog and human sensory and semiotic worlds for a shared task is nonetheless asymmetrical and divergent. Indeed, mine detection is not an innocent, neutral practice. It is traversed by power relations that reinforce hierarchical bodily, cognitive, and ethical regimes; fantasies of human mastery conflicting with desires for more-than-human partnership; and conceptions of cost-efficacy and global economies of expertise. Thus, the experiences of dogs and humans are shaped by differential corporeal sensibilities and vulnerabilities, affective attachments, and spatial distributions. Called to fulfill specific functions, dogs are unequally exposed to the violence of land mines and other occupational hazards, such as heatstroke. These distributions of labor and risk are actively produced and naturalized through daily training and interactions.

Mine detection as sensory co-laboring involves not only the refinement and alignment of multisensory capabilities but also the re-production of values through them. As studies on sensory labor show, senses are more than means of apprehending material realities; they are also spaces for the transmission and constitution of aesthetic, epistemic, political, and economic valuations.¹⁹ I particularly consider how dog-human detection fashions and is fashioned by the global humanitarian industry and its values. Like many other sectors in the aid world, this industry increasingly embraces standardization, cost-effectiveness, and evidence-based decisions, justifying humanitarian interventions through economic rationales and measurable results.²⁰ Informed by logics of global capitalism, these values facilitate the asymmetrical mobility and deployment of humans and nonhumans across war-ravaged countries.

Thus, the notion of sensory co-laboring highlights that even when dogs and humans engage in a shared task, their motivations, the practice, its meaning, and its impacts may not be the same. As Stengers states, “Common does not mean having the same interest in common, only that diverging interests now need each other.”²¹ Trainers allude to this sensory and affective divergence when they say dogs are searching not for explosives but for the Kong, suggesting that dogs experience detection as play. NPA canine detection training takes the form of highly choreographed games (e.g., hide-and-seek, tug-of-war, catch-it-if-you-can), a practice recently adopted in military training circles in Colombia. Trainers believe that humans can interest dogs through playful, pleasurable exercises, enlisting their olfactory abilities in the humanitarian task of detection. The anthropocentric character of this reading aside—we do not know with certainty what interests dogs or what they find pleasant—I am interested in its suggestion that, even when their modes of attention and communication converge and are deliberately transformed, their worlds continue to be distinct and incommensurable. Neither dog nor human is fully comprehended by the other.

I have come to understand these games as playful intra-actions: joyful yet rehearsed encounters through which dogs and humans are made and remade together in difference and in relation to one another. I am inspired by Barad, who defines intra-action as an agential cut through which “the boundaries and properties of the causally related components of phenomena become ontologically determinate and that particular concepts become meaningful (that is, semantically, determinate).”²² Neither subject nor object is inherent; both are enactments. At stake in these games is what dogs and humans become. They do not precede, nor are they external to, these rehearsed joyous exercises; rather, they emerge in and through them: mine detection dog and handler. The former is a highly skilled animal able to move systematically through minefields and indicate the presence of explosive substances in the ground. The latter is a qualified human able to understand, guide, and maintain their partners’ mine-sniffing capacities and attend to their physical and mental interests. In this sense, the binomio canino cannot be understood as the sum of two unconnected and preexisting units, nor as a single one. It is “more than one, less than many.”²³ Or, in de la Cadena’s formulation, it is a “complex we.”²⁴ These playful intra-actions, however, also make other becomings impossible or unlikely. As dogs and humans become co-laborers and experts, they cannot be something else (e.g., a pet or a human owner).

By describing dog-human detection work as sensory co-laboring, I do not seek to extend sensory labor to animals since it can risk reproducing and sedimenting a logic in which wage relationships disproportionately shape bodies, relationships, and environments.²⁵ With sensory co-laboring, I am more interested in highlighting how the work of humanitarian mine detection co-constitutes dogs and humans in different, complementary, and fundamentally asymmetrical ways. Recognizing this partial connection might enrich understandings of human and more-than-human becomings. Sensory co-laboring allows us to rethink how humans and other creatures are ontologically reconstituted and how overlapping histories of warfare and humanitarianism, legacies of animal behavioral practice, and instrumental-affective interactions shape these reconstitutions.

The Global Humanitarian Canine Industry: From Bosnia Outward

Mine-sniffing dogs were introduced to demining during World War II. With varying levels of success, countries including the United States, the United Kingdom, France, and the Soviet Union attempted to implement canine detection programs and study the reliability of this biotechnology.²⁶ This detection work was an addition to the historical weaponization of dogs against Native people and African Americans; the multiple sanctioned roles they played in the nineteenth- and early twentieth-century wars as messengers, sentinels, guardians, and rescuers; and their more recent deployment against decolonization struggles and in the so-called War on Drugs and War on Terror.²⁷ As with other animals, dogs’ capacities have served as a strategic resource that consolidates colonial relations and powers.²⁸

The history of dogs in humanitarian demining is shorter. Although there are records of their deployment in Afghanistan in the 1980s, it was not until the 1997 Anti-personnel Mine Ban Convention (also known as the Ottawa Treaty) that dogs became a crucial civilian demining technology.²⁹ Humanitarian dog detection shares vocabularies, bodies, and organizational structures with its military equivalent, but its objectives, security conditions, and procedures are different. Notably, it is generally carried out in pacified areas and follows internationally standardized clearance procedures. These conditions weigh heavily on the material and affective interactions that can and should be established between mine-sniffing dogs and handlers.

Norwegian People’s Aid (NPA) has played a key role in this process through its breeding program and mine detection dog academy in Bosnia and Herzegovina, the Global Training Center (GTC). Since 2004, the GTC has aimed to professionalize and standardize canine detection in humanitarian settings. They conducted comparative studies to determine “the optimal demining dog breed” and developed dog and handler training methods. Based on what Weaver would call “fuzzy science,” a sense of science that “actively shape[s] experiences of differences such as gender, race, sexuality, species, and race,”³⁰ the GTC opted to primarily breed and train Belgian Shepherd Malinois (fig. 2). Compared with the German Shepherds and Golden Retrievers widely deployed in military and detection tasks, Malinois are smaller and lighter (which generally prevents them from triggering mines), and they have shorter coats and longer attention spans. They are also said to have the right temperament: high adaptability, extreme natural drives for hunting and herding, courage, a stable personality, and, as an NPA trainer commented, “good intraspecies relationships.”

Saúl, a Colombian Army canine expert I interviewed in late 2016, described this breed to me as a “formula uno [Formula One]” racing car: “Malinois are tireless.” They are faster, more reliable, and more hardworking than other dogs: “They ask for work, for more pista [track].” Although they tend to be nervous, they are a “great working machine when raised properly.” Saúl concluded: “Malinois have the best motor, better revolutions.” Certainly, this analogy illustrates how this breed has been instrumentalized, thought of as machines in service of so-called human problems—an argument that scholars have explored significantly.³¹ Here, however, I am more interested in how the analogy naturalizes “nature as working,” presenting and celebrating biological capacities as work.³² Indeed, Saúl’s analogy highlights Malinois’ supposedly inherent ability to sniff for long hours in demanding environmental conditions, making it only natural to train and deploy them to detect explosives. Furthermore, analogies and scholarly analyses conceiving of dogs as detection machines ignore how corporeal skills are the product of strenuous training processes and how sensory work depends on interspecies collaboration.

The standardization of the Malinois has allowed the GTC to supply qualified detector animals to different NPA offices and other mine-clearance operators. A significant percentage of dogs working in demining today come from this academy, making NPA the leader of the international humanitarian dog industry. Frequently, the transfer of mine-sniffing dogs to other war-affected countries involves the disintegration of an established dog-human partnership, depending on local and global humanitarian needs and handlers’ capacity and willingness to relocate and become expatriates. The global circulation of humans and nonhumans reveals a profound imbalance between national and international aid work, including their different conditions of locality and mobility.³³

A Dog’s Life: Binomios Caninos in Colombian Minefields

The majority of the canine team arrived in Colombia in May 2015 and settled in the national Army base in Tolemaida for a few weeks to complete their adjustment process and gain access to heavily regulated explosive substances for training. The team’s arrival marked a crucial moment in the evolution of demining in Colombia. Historically, mine detection dogs have been central to the army’s binomios caninos in the tactical fight against improvised land mines. Unlike humanitarian operations, military demining removes explosive devices to break defense lines or secure strategic positions against an enemy, so mine-contaminated territories are not thoroughly inspected or cleared. These dogs have also been deployed in military territorial control, counterinsurgency, and coca eradication.³⁴ Their extensive presence in military settings contrasts with their relative absence in the few humanitarian tasks carried out since the approval of Law 1421 authorizing civil demining in 2011. The Pilot Project introduced canine detection as a humanitarian technology in Colombia.

As the coordinator of the Pilot Project, NPA promoted mine detection dogs as a necessary and cost-effective response to the country’s unique war contamination problem. Globally, canine teams are part of the NPA’s comprehensive demining toolbox, an approach that combines manual, mechanical, and animal detection removal practices in an effort to professionalize and streamline mine clearance. Deployed during the second stage of the current mainstream demining methodology of Land Release, the Technical Survey, detection dogs can significantly reduce the extent of suspicious areas, particularly those with low mine density or nonmetallic land mines. In Colombia, land mines are made with plastic containers and chemical activation systems that easily evade metal detectors and are planted irregularly and temporarily, scattered in remote rural areas.³⁵ Detection dogs can also clear confirmed hazardous areas in a safe, cheap, and reliable manner. An NPA poster highlights this point, suggesting that dogs can find bombs five times faster than manual detectors and discriminate between metal and explosives. According to NPA’s calculation for the Pilot Project, a manual deminer inspects 360 square meters in forty-five days while a canine unit can do that in three and a half days, surveying 11,400 square meters in one month.

In the Pilot Project, the canine team was deployed only in the largest minefields, previously prepared by a minesweeper machine known colloquially in Colombia as la barreminas. Divided into binomios, dogs follow the GTC choreography, moving from left to right in ten-meter-long and one-meter-wide lanes. Unlike in daily drills, handlers in minefields wear a personal protection vest and visor. The canines do not have protective equipment. This is not the only significant difference between training and operations. The detection dance is extended in operations; it can last between four and six hours, divided into fifty minutes of work followed by a fifteen-minute break. The team usually has a dedicated shaded rest area near the minefield. Furthermore, the Kong, the affective device that catalyzes dog-human encounters, is absent during operations, which means that dogs and humans must be able to delay gratification when so-called serious work happens.

With few exceptions, the canine team was the first to arrive at the minefield around 6:00 a.m., primarily to avoid dense fog, heavy rain, or high temperatures that make explosive molecules even more elusive. Heat was a critical factor, since it reduces dogs’ stamina and increases their heatstroke risk. Trainers argue that dogs are more likely to suffer and die from heat than from stepping on an explosive device. An overly warm dog also has a less reliable nose. The search for explosive particles involves sniffing the lanes with their mouths closed, breathing only through their muzzles. Because dogs cool down by panting, their bodies heat up with each closed-mouth breath. Only upon completing the search and exiting the minefield can they cool off as usual, mouths open and tongues out. Indeed, heat is “a material signifier of work,” making practitioners “aware of work’s limits, its wastes, and its excesses.”³⁶

Dogs and humans would return to the base camp around noon. Their nonworking time looked utterly different. While the humans spent most afternoons in a semiopen space functioning as a common dining room, eating, playing cards, and listening to music (or secluding themselves in their respective tents, as Ibra used to do), the dogs would be confined to kennels, away from human contact. These circumstances were not unusual; their initial two-year training at the GCT also took place in a kennel environment. Unlike interspecies relationships in military operations where dogs and humans share everyday spaces, eat next to each other, sleep in the same tent, and have uninterrupted moments of affection, dogs and humans were kept in separate, clearly delineated spaces in the Pilot Project.³⁷ Dogs only left their kennels for sustained contact with their humans twice a day—while working in the minefields early in the morning and training at the end of the day. Dogs received food and water in their kennel, provided exclusively by their trainers.

For trainers, canine confinement had two functions. It allowed dogs to recover physically and mentally from arduous sessions—confinement was a form of caring for them. But it also solidified the association of detection activities with entertainment, joy, and human attention. This clear distribution of affective activities produced the kennel as a space of repose and care but also boredom and indifference, indexing the calibrated distance-intimacy and detachment-engagement required for sensorial co-laboring in humanitarian mine action. Monotony and excitement imply the distribution, intensification, and specialization of certain attachments and forms of labor to produce particular kinds of demining coworkers.³⁸ This work draws attention to matters of care, the engagements involved in the constitution of techno-scientific assemblages and human/nonhuman relationships.³⁹ As Puig de la Bellacasa states, “We must take care of things in order to remain responsible for their becomings.”⁴⁰ This responsibility implies recognizing the interdependent relationships that constitute our worlds and how our care practices reproduce particular social systems while discouraging and precluding others. The human-dog assemblages in mine detection involve asymmetric, instrumental, and profoundly anthropocentric regimes; they are also regimes of sensory distribution and collaboration.

The Kong and the Making of a “Nose for Demining”

In April 2016, I participated in a one-day tour of the newly inaugurated dog training center of the HALO Trust, a British organization that was the first certified civil demining operator in Colombia. Our guide, Oscar, was an experienced civilian dog trainer. Before coordinating HALO’s program, he had worked with the Colombian Army and police in their dog training centers, a personnel shift not uncommon in demining, demonstrating the links and continuities between the military and the humanitarian demining sector.

After outlining his organization’s plans, Oscar invited us to observe a session of “instinct empowering.” Two (male) trainers and a four-month-old puppy played tug-of-war and hide-and-seek with two Kongs in a bright room. While the head trainer and the dog playfully fought over one of the Kongs in a cleared half of the space, the second trainer hid the other Kong in the back of the room, littered with loose bricks, chairs, mannequins, plastic bottles, tubes, and pieces of wood on tires. Once the second trainer returned to the front, the head trainer snatched the first Kong from the puppy and prompted him to chase it. Orienting the dog’s body with the Kong so that he faced the back of the room, the trainer pretended to throw it. Fooled by the trick, the pup ventured out, sniffing his way in search of the Kong. The puppy moved nimbly through the rubble, snout pressed to the ground and tail toward the sky. After navigating obstacles, moving around corners, and narrowing the search area, he found the hidden Kong. The lead trainer praised him enthusiastically while the other remained at a distance, silent. Once trainer and dog were reunited, the trainer took the Kong back and enticed the doggie again. Catch it if you can!

As the game unfolded, Oscar commented that this is a daily routine for pups eight to sixteen weeks old. The time and level of difficulty of these encounters increase gradually over weeks, but it must be a joyous experience. “The trainer should not overwhelm the dog,” Oscar said; exercises must stop if the dog grows bored, frustrated, or tired. This is especially important, as these activities are key to developing dogs’ sensory attraction and attachment to the Kong. Saúl, the canine expert who compared the Malinois with a racing car, described this affective attachment in terms of desire and obsession, though he recognized that this fixation is neither natural nor inherent. Rather, it is cultivated during two years of training, mainly through playful sessions. Such recognition highlights that “desire is never an undifferentiated instinctual energy, but itself results from a highly developed, engineered setup rich in interactions.”⁴¹ The fixation is a collective fabrication that emerges in a carefully orchestrated convergence of bodies and encounters.

Technically, however, the Kong is conceived of as a reward, which is critical to operant conditioning, the animal learning process that currently informs most humanitarian mine detection training. Unlike military animal teaching methods focusing on human mastery, unconditional obedience, and negative reinforcement, this technique celebrates so-called desired behaviors while ignoring unwanted ones. Peter, a world-renowned expert on the topic and keynote speaker in the first official workshop on canine detection in Colombia in 2016, explained that this approach works by association. Typically, a “natural” or “voluntary” behavior is associated with a reward or a punishment. Punishment, he clarified, is not physical violence but, rather, “affective withdrawal,” which might involve disregard, absence of physical and vocal praise, and no playing time. Through such reinforcements, the likelihood of a particular behavior recurring in the future is increased. For mine detection dogs, the desired behaviors include playful and inquisitive behavior, good hunting drive, obedience, good search pattern, and correct indication of mines.

Vinciane Despret is critical of such behavioral animal conditioning because it tends to see “the animal as no more than an automaton for whom understanding is limited to simple associations.”⁴² In thinking about what moves animals to be interested and participate in human experiments, she warns us “not [to] confuse what the work makes possible with its motive. Or, at any rate, abandon the concept of instinct, but guard preciously what it makes us feel, what feels like a force in the face of which being must bend.”⁴³ I have thus come to see the Kong not merely as a reward but as a catalyst, an affective device to attract, shape, and assemble the sensory capabilities of dogs and humans, influence their dispositions and moods, and facilitate mutual obligations and attachments. The Kong allows them to encounter and affect each other, albeit differently. Mutual yet unequal, this influence is essential for them to become sensory co-laborers, experts and partners in the joint but divergent practice of mine detection.

The puppy training session illustrates the affective power of the Kong. Through prosaic games in carefully curated spaces, a simple toy becomes a lure that mobilizes the dog and redirects his olfactory attention—moved by the scent of the Kong, the dog increases his ability to ignore any other materials in his path. Oscar noted this corporeal orientation, kinesthetic disposition, and sensory sophistication, commenting that this was the first time this puppy had been exposed to a large group of people during a work session. Despite our presence, the dog continued to “work with determination and concentration.” Of course, we were not the only “distraction”—the room was purposefully filled with objects and sounds designed to challenge the pup’s temperament and divert his orientation, movement, and focus. I highlighted the word work in my field notes to remind me that play and work might be the same in mine detection training; games of hide-and-seek and catch-it-if-you-can are also working sessions. This overlap emphasizes how working relationships organize affects and facilitate their embodiment.

Through these encounters, the dog begins to become “a nose for demining.” Here I invoke Latour’s description of a nose as “someone able to discriminate more and more subtle differences and able to tell them apart from one another, even when they are masked by or mixed with others.”⁴⁴ Latour argues that the body is an interface that acquires a better articulation capacity as it becomes more sensitive to what constitutes the world and learns to let itself be affected by more elements. As the body’s awareness of difference increases, it develops capacity to propose richer and more complex odoriferous articulations of the world, actively participating in its composition. Becoming a nose is “thus a progressive enterprise that produces at once a sensory medium and a sensitive world.”⁴⁵

Detection choreographies are part of such sensory progressions. When trainers establish and cultivate the magnetism between the Kong and the dog, other associations and affective connections may be built on that magnetism—mainly, the one that must exist between the toy’s scent and the scent of explosive components. During the puppy program, Kongs of decreasing sizes are impregnated with chemicals frequently used to produce explosive devices (fig. 3) and then arranged on carousels, brick walls, or outdoor tracks to find during training. These “apparatuses of attunement” make elusive explosive substances perceptible by rendering them “available to be sensed.”⁴⁶ They also hide nonexplosive substances such as rocks, dirt, iron, and grass to make the search challenging and hone dogs’ discriminatory capacity. Through this systematic technical process, the toy gradually comes to stand for explosives. In Colombia, mine sniffers need to develop (fond) associations with ANFO (ammonium nitrate and fuel oil), R1 (ammonium nitrate, black aluminum, and fuel oil), R14 (potassium permanganate and aluminum in powder), pentolite, TNT, and nitroglycerin, components frequently used in improvised land mines.

However, the possibility of improving sensory abilities is not limited to the dog. As recent interdisciplinary research on dog-human training illustrates, this practice is not a one-sided affair.⁴⁷ Rather, as I illustrate in the next section, it is an exercise of interdependency and co-constitution that sustains and transforms everyone involved; the haptic and hermeneutic abilities of the handler are also being reconfigured in daily choreographies.

Learning to Touch Well

One misty morning in October 2015, I found myself sitting on a concrete floor pampering Drag, a five-year-old Malinois who belonged to the army’s demining battalion. “Stop petting him! He is not a pet; he is a hard-working dog!” Ibra barked, taking me by surprise. Drag’s handler did not mind my touch, and Drag seemed to enjoy it—as his belly in the air and his continuous movements suggested. “stop touching him!” Ibra repeated, louder and in a more hostile tone. I realized it was not a suggestion; it was a command. I stopped.

This was the first time I heard Ibra order people not to touch his dogs. It was not the only time I heard him make distinctions between his canines, whom he portrayed as tireless explosive-sniffers, and other dogs, whose roles as pets did not require, according to him, the same physical capacity and mental focus. Oliver, my then-two-year-old mutt dog, frequently served as an example: “Your dog sleeps, eats, and waits for you to pet him. My dogs detect.” Ibra allowed no interaction between his dogs and local dogs. Local dogs left to fend for themselves were more likely to get sick and spread disease. In discussing his dogs’ welfare, Ibra generally highlighted their economic value: a dog explicitly bred for mine detection but not yet trained—referred to as a green dog—costs approximately $400. The cost of a fully trained and certified hound like Hamilton can range anywhere from $25,000 to $35,000. Their price earned them the sarcastic title “the jewel in the crown” among local army dog handlers.

Initially, I associated Ibra’s commands with the military history of canine detection and the male-dominated and arguably deeply militarized sector of the demining industry.⁴⁸ When pressed with questions about the differences between companion animals and working animals, Ibra said: “A working dog is like a soldier: he works, plays, rests; he works, plays, rests: he is always under control.” His words illustrate the legacies of warfare and dominance training in dog-human mine detection, which requires the handler to become the pack’s alpha and remind his canine who is master. I assumed that these military relationships were marked by distance, detachment, and dominance, but I soon learned that the training and deployment of these animals on battlefields historically depended on practices of intimacy and bonds of affection.

Over time, I also understood that even the military’s affective idioms did not resonate with how Ibra and other handlers understood their relationships with their dogs. Indeed, despite NPA posters suggesting otherwise, they adamantly opposed discussing their dogs in terms of love or friendship. Nor did they recognize them as family. Ibra criticized the army’s binomios caninos, whose physical endurance and sensory abilities were threatened by the lack of discipline, rigorous training, and emotional detachment. NPA trainers prefer to speak of interest, joy, and trust. These affects bring what some trainers call “directed energy,” which is about having highly motivated dogs. Thus, they relate interest and joy to their dogs’ professionalism and expertise, offering descriptors such as hard work, reliability, and competence.

After a few months, I began to grasp what was at stake in not touching mine detection dogs. Rather than reading Ibra’s command as an imperative of physical distance and affective disinterest, I came to see it as an obligation to attend to the instances in which literal and metaphorical touch is performed—an obligation to learn to touch well. The phrase touching well does not refer to a legitimate and morally appropriate form of bodily relationship; it neither demands virtuous contact nor requires soft, tender, and nonviolent touch, even if it involves good treatment and affability. Rather, it is about bodies coming into contact on time and with care to produce specific effects and affects. To touch well is to touch with tact, “a form of sensorial politeness, understood as a political art of gauging distance and proximity.”⁴⁹ It entails developing a sensitivity to navigating spaces of engagement and withdrawal, regard, and inattention. The absence or regulation of touch is not a radical alterity to engagement; rather, it is part of the ambivalent spectrum of human/nonhuman relationships that must balance distance and intimacy to yield particular results, which requires attention to the other’s needs, expectations, and dispositions.⁵⁰ Hence, it implies considering the other that emerges during contact, being attuned to the encounter, and responding in relation.

Handlers are the only ones authorized to interact with mine detectors. Just as dogs practice their olfactory capacities and bodily language, humans work to hone tactile dispositions. Trainers must develop a specific sensitivity to know who can touch, when to touch, how to touch, and why to touch, as well as what counts as touch. Furthermore, trainers need to sharpen their capacity to read and listen to their canine partner’s body—the sometimes-hard-to-detect signals that dogs may send with their tails, ears, teeth, breath, and barks and growls. Refining their understanding of their partners, they can identify and modify their movements, rhythms, moods, and behavior to promote maximum team performance.

Like their canine companions, trainers hone haptic and interpretive skills in daily playful choreographies. In contrast to his “do not touch” command, Ibra frequently engaged in varied contact with Hamilton, timed precisely and tailored to specific interactions and purposes. Consider the detection dance I described in the introduction: whenever Hamilton returned to the “safe line,” Ibra offered a gentle, quick touch before the dog started another lane. When Hamilton detected the target, Ibra gave him a soft, slower touch to calm him and reward his correct indication. Calming the dog is vital; an overexcited mine-sniffer does not search rigorously. Another kind of touch emerged when they played tug-of-war with the Kong; the blunt, assertive, and energetic contact served to provoke enthusiasm and increase “motivation”—an apathetic canine does not search with exertion.

Touch is critical to the work of mine detection dogs and trainers. A touch offered at the wrong time, too often, or by the wrong person (me, for example) might encourage unwanted behaviors, produce a different olfactory capacity, and materialize a different dog—a pet rather than a working dog. An ill-timed or aimless contact could also cause the handler to lose his authority, making him something more than a collaborator—an owner or playmate. Given the bodily and affective obligations and realities that physical contact sets in motion, careless physical contact might disrupt the mutual conditioning of dogs and handlers, potentially jeopardizing their collaboration. As with other human-nonhuman work practices, mine detection entails the mobilization, instrumentalization, and circulation of affects.⁵¹ Learning to better read, align with, and respond to their companions, dogs and humans foster specific engagements and detachments. The control of contact and its correlated affective economy are central to their becoming sensory experts and partners and the performance of safe, reliable, and cost-efficient detection practices. They also facilitate the circulation of the sensory expertise of dogs and handlers to other war-polluted settings.

Conclusion

The detection choreographies and mundane interactions between explosive sniffing dogs and their handlers in Colombian minefields bring forth various elements that constitute what I call sensory co-laboring: the daily multispecies efforts to partially bring together incommensurable sensorial and semiotic worlds for the sake of a shared but profoundly unequal task.

Sensory co-laboring joins a growing body of work concerned with constitutive relationships between humans and nonhumans.⁵² Through different methodological approaches, these studies offer a passionate immersion into the multilayered biographies of animals, plants, fungi, and microbes to reveal their active participation in our human worlds and the need to move away from anthropocentric analytical frameworks. Some scholars have focused on the labor capacity of nonhuman entities and the value-production and accumulation power of interspecies encounters, drawing attention to relationships of cooperation, exploitation, and resistance.⁵³ While acknowledging the centrality of nonhumans in past and current scientific, political, and economic systems, other interdisciplinary environmental scholars question how acknowledging other species’ work sustains or transforms capitalism. They warn us about expanding what counts as work, as this can disrupt our understanding of the logics underlying formations of capital, labor, and nature.⁵⁴ These studies invite other ways to conceptualize human/nonhuman engagements outside wage relationships and accumulation.

In turn, anthropologists and feminist science and technology scholars advance an argument about perception as a form of labor, whether performed by humans or nonhumans.⁵⁵ Based on various ethnographic sites, they insist that sensing “requires work and produces value” and that sensory experiences are stimulated and disciplined through specific settings and apparatuses.⁵⁶ Hence, perception does not involve singular bodies but is distributed among people, techniques, technologies, and infrastructures. Additionally, scholars have explored the use of animals as sentinel devices in public health, environmental science, and postconflict scenarios for the detection of invisible threats, such as viruses, chemicals, radioactive materials, and explosives.⁵⁷ Although ambiguous and contestable, the signals provided by these biosensors reveal past, present, and recurring violence and can have political force—nonhuman sentinels afford ways to recognize invisible injuries as they become consequential allies in uncertain futures.⁵⁸

Sensory co-laboring builds on these diverse works of literature on multispecies relationships, nonhuman work, and sensory labor. Grounded in ethnographic engagement with more-than-human global and local humanitarian demining efforts, it captures the living and working conditions of Belgian Shepherd Malinois dogs—bred, trained, and deployed to detect explosive remnants across war-polluted countries—and their complex material and affective relationships with handlers, humanitarian professionals whose biographies often link them to warfare and militarism. I detailed the technical and everyday contact zones in which these multispecies detection units materialize, and, in a very literal sense, what they are capable of together. Through playful intra-actions—highly orchestrated games around the Kong, which is a toy, a training device, and an explosive (substitute)—dogs and humans become more sensitive to odors and bodily gestures and learn to attune themselves to their partners, building relationships of joy, interest, and even admiration. These sensory alignments and affective engagements are critical for mine detection dogs and handlers, the expert parts that make up the dog-human “unit,” the binomio canino.

As it foregrounds these conditioning practices, sensory co-laboring emphasizes the mutual yet uneven character of mine detection training. This labor disparity cannot be separated from the production of economic, political, and ethical values within the humanitarian industry, including in the distribution of biotechnologies. However, the sensory capacity of dogs and humans cannot be solely attributed to global value-producing dynamics. Doing so would overlook how dogs and humans engage in the joint task of sensing from different subject and power positions and distinctive sensory and ontological experiences. Mine detection is a common yet divergent practice: it may not be or mean the same thing for dogs and humans. Ultimately, sensory co-laboring highlights that humans and nonhumans can cooperate in a shared task even if and when it may emerge and be experienced as two different activities. This difference does not cancel the possibility of collaboration; detection binds dogs and humans together and brings them into relation through difference. Recognizing this partial connection enriches our understanding of human and more-than-human worlds, complicating the dichotomies often used to think about working relationships between dogs and humans, including work and pleasure, discipline and play, dominance and cooperation, and engagement and distance.

Acknowledgments

This work would not have been possible without the encouragement and intellectual generosity of numerous writing communities, escritoras superpoderosas y las divinas. I am particularly grateful to Lina Pinto, Diana Ojeda, Emma Crane, Nathalia Hernández, and Teresa Gil for their comments and suggestions on earlier versions of this document. I also thank the editors and anonymous reviewers for providing detailed and constructive feedback. Finally, I want to acknowledge my four-legged companions and friends for helping me think about what it means to work together but in difference: Hamilton, Rambo, Zarex, Oliver, Oso, Olaf, and Tinto.

Notes

References