Doing and undoing Caribou/Atiku: diffractive and divergent multiplicities and their cosmopolitical orientations*

More than one and less thanmany, has become a refrain to depict the notion of multiplicity. Borrowed from Mariyln Strathern, Annemarie Mol mobilized the refrain to succinctly capture the complex result of a series of operations that make a variety of practices hold together as a singular thing. In this article, I seek to explore some consequences of the proposition that multiplicity can be figured in, at least, two different ways: as diffraction, where the operations of singularization explored by Mol are more easily carried out, and as divergence, where singularization is not necessarily an option. The exploration is part of a larger project to rework the notion of cosmopolitics first proposed by Isabelle Stengers and later taken by Bruno Latour. Elsewhere I have argued that their conception of cosmopolitics as a project oriented towards the composition of a common world is predominantly informed by the figuration of multiplicity as diffraction, and thus it very much resembles a process of singularization writ large. In this context, foregrounding multiplicity as divergence opens a path to probe the limits of this conception of cosmopolitics, inquire into the different ways in which multiplicity holds together, and envision alternative forms of cosmopolitics. I organize my exploration around two entities caribou and atîku that, so to speak, occupy the same space at the same time in terms of bodily presence, albeit dominant common sense would have it that atîku and caribou are two words for the same entity.

In her now classic book The Body Multiple, Annemarie Mol (2002) used the refrain "more than one and less than many" to succinctly capture the result of operations of singularization that make a plurality of practices to hold together as a singular "thing" (as the disease "atherosclerosis", in her case). In this article, I explore some consequences of the proposition that multiplicity can be figured in, at least, two different ways: as diffraction, where the operations of singularization explored by Mol are more easily carried out, and as divergence, where singularization is not necessarily an option. The exploration is part of a larger project, which I started in a previous article (Blaser 2016), to rework the concept of "cosmopolitics" originally proposed by Isabelle Stengers in the mid 1990s and later further developed in dialogue with Bruno Latour (see Stengers 1997Stengers , 2005andLatour 2004, 2007).
In my previous article, I mobilized ethnographic materials from an ongoing conflict between the Canadian province of Newfoundland and Labrador (NL) and the Innu people in order to probe the scope of the concept of "cosmopolitics." My purpose in the present one is to move my exploration of cosmopolitics not necessarily forward but rather sideways, in a direction that I consider crucial but that in the context of the previous article's narrative line could not have been developed. Thus, in this introduction I provide a brief summary of the main argument of my earlier article.
The conflict that grounded my initial exploration on the scope of cosmopolitics hinges on the NL government's ban on hunting a declining caribou herd in Labrador and the Innu people's refusal to accept it. Succinctly, wildlife managers in the provincial government think that hunting in present conditions could mean the disappearance of the caribou. In contrast, and for reasons I will revisit below, Innu hunters and elders consider that being prevented from hunting according to "traditional" protocols almost assuredly would mean the disappearance of atiku (at first sight, the word Innu use to refer to what Euro-Canadians call caribou). A key argument of the article, which I will not fully develop here, was that the concept of cosmopolitics made it possible to grasp this kind of conflict without falling into what I called the trap of "reasonable politics," a trap deployed when these conflicts are taken to be between different perspectives on the same "thing." 1 I argued that cosmopolitics avoids this trap because it refers to the possible articulations between "different reals" rather than between different perspectives on the "same real." And yet, while cosmopolitics is useful in that regard, it also has some limitations that I attribute to the orientation that was given to it in its original formulation, an orientation towards the composition of the common world as the horizon to be pursued (even if never quite reached).
Why would an orientation to the common world constitute a limitation? Because it offers few alternatives for articulating "worlds" that are not amenable to become common. Quoting Latour, I pointed out that, in relation to the common world being constituted, those "un-common-izable" worlds would be rendered an externality about which "an explicit collective decision has been made not to take them into account; they are to be viewed as insignificant" (Latour 2004). This insignificance, of course, could justify their destruction. Thus, I argued, a conception of cosmopolitics oriented to the constitution of the common world can at best offer, as Stengers version does, a slowing down of this externalization and a recognition of the victims thereby generated. 2 I further surmised 1 Blaser: [R]easonable politics operates on the basis of turning differences into perspectives on the world. Differences made into perspectives are amenable to be ranked according to putative degrees of equivalence between perspectival representations of the world and the factual world itself. This ordering, in turn, makes it possible to deem some perspectives irrelevant, erroneous, or dangerous, and thus dismissible or, worse, destroyable. (2016,(549)(550) 2 Much of my tinkering with the notion of cosmopolitics has been done in collaboration with my colleague Marisol de la Cadena (see de la Cadena 2010de la Cadena , 2015 and, of course, we have both been greatly inspired by and in conversation with Isabelle Stengers, who has participated in two seminars that we organized (the first in 2009 the second in 2012). However, only slowly we fully realized that the kinds of situations for which Stengers coined the concept of cosmopolitics was not precisely what we had in mind when we started to think with it. As Stengers points in her contribution to a forthcoming volume De la Cadena and myself are co-editing (see Stengers 2018), she coined the concept of cosmopolitics that this orientation towards the common world was informed by the situations STS scholars had been more used to investigate until recently, that is, controversies through which matters of concern are progressively transformed into matters of fact (or, what is the same, through which a plurality of practices are made to hold together as a single "thing"). 3 In these cases, the kind of multiplicity at stake in a matter of concern that gathers an assembly is already visible and/or legible for the entire public assembled. This is not the case in situations like the ones involving the atiku/caribou conflict where, as I will soon show, another kind of multiplicity is at stake. I concluded my initial exploration in that article with the idea that, if we took the kind of multiplicity revealed in the atiku/caribou case as the starting point, another cosmopolitics, with a wider repertoire of responses to "un-common-izable" worlds, might be possible. My rationale was that if the original conception of cosmopolitics was inspired by situations where one kind of multiplicity (which here I will call diffractive) is made to hold together as a single thing, the possibility of a more expansive cosmopolitics might hinge upon an (ethnographically rich) examination of how another kind of multiplicity (which I will call divergent) is made to also hold together. In this article, I begin to make some headways in this examination.
Before I begin, I must state upfront that while the case study I work with implies these issues, the article is neither about insights to be gleaned from "Indigenous knowledges" nor about the relations that Indigenous and "Scientific" knowledges might sustain with each other. My immediate aims here are more modest. Drawing on the case study, I seek to sharpen for analytical purposes the contrast between diffractive and divergent multiplicities; and then, with empirical support from the case study, start discussing some ideas on how divergent multiplicity might "hold together," that is, how divergent practices might go on together without interrupting or canceling each other. 4 In doing this, I expand on my closing contention in the previously mentioned article, namely, that what Viveiros de Castro (2004) calls "equivocations"i.e. situations in which interlocutors appear to be speaking of the same thing when they are actually referring to different onesoffer a template to chart both potential avenues and roadblocks for such "holding on together." inspired by the European anti-GMO movement that brought together young urbanites, farmers, and biologists from the Continent along with African and Indian peasantsall with their own specific reasons to resist GMOs. The movement thus expressed an interest in common that was not the same interest. Yet, what was remarkable was precisely that all participants recognized the importance of the "issue" gathering them, "agreeing that each of them has a legitimate voice and is entitled to contribute to the issues that concerns them." But Stengers worried wether those gathered by the issue would be able to listen to those who would claim that the issue as such did not concern them, but would nevertheless be affected by the decisions taken. Here is where she saw the prefix "cosmo-" crucial to slowdown and disrupt political deliberation: "Cosmopolitics means that politics should proceed in the presence of those who will bear the consequences, who will be the victims of political decisions[.]" For us cosmopolitics was not a concept to think about the possible disruption of political deliberations but a tool to think about disputes (we can also call them gatherings) that concerned and included participants whose presence was not recognized by all who participated in the gatherings. Our paradigmatic ethnographic examples involved a mountain that is also an earth being (de la Cadena 2015) and forest animals that are also spirit masters of their world (Blaser 2009(Blaser , 2016. As Stengers (2018, 95) points out, in these circumstances cosmopolitics would not attach to situations in which the normal process of political deliberation is disrupted but rather to situations in which "ontological clashes would have to be anticipated anywhere as no issue can any longer be considered a matter of free deliberation [.]" 3 Latour coined the terms "matters of concern" and "matters of fact" to distinguish two opposing moments in the trajectory of a "thing," from being uncertain, under discussion, perhaps even nonexistent to being a definite entity, that holds on its own "out there." 4 Helen Verran's (2013Verran's ( , 2018 powerful refrain, "going on together doing difference" is an important beacon guiding this exploration.
In order to move on in this exploration, I will revisit only partially some of the ethnographic materials I used in my previous article and will expand on others that were not included there. 5 I begin by looking at caribou in a similar way as Mol did with atherosclerosis in her book. Using her work as a template and counterpoint, enables me to highlight the particularities of diffractive multiplicity in the caribou case. In the second section, I show how atiku is not only different from caribou, but different in a way that makes evident divergent multiplicity. And yet, this multiplicity nevertheless "holds together" in such a way that gives credence to the commonly held idea that atiku and caribou are just two words for the same "thing." This apparent contradiction will provide the springboard to discuss in the third section the important role that equivocations play in "holding on together in divergence." In the conclusions, I draw together the points raised by this exploration and reconnect them to the larger project of rethinking cosmopolitics in light of the limitations that an orientation to the common world seem to reveal.

The Caribou multiple 6
What is Caribou? Thus opens a landmark publication, Caribou and the North: A Shared Future (Hummel and Ray 2008) that sought to bring into the public eye the need for urgent actions to ensure the survival of caribou from Alaska to Labrador. The question posed is not an ontological one, it is clear that the authors know what caribou is: "a large-bodied ungulate … [that] belongs to the group Artiodactyla," and that "despite the fact that [they] actually encompass a remarkable variety of different forms of the same animal," it is one single species, Rangifer tarandus (Hummel and Ray 2008, 29-30). The challenge the authors want to stress with the question "what is caribou" is classifying properly this diversity into "types" using different criteria such as morphological differences (shape, size, antlers' form), behavioral differences (migratory and sedentary); and ecotypes where the caribou lives (tundra, boreal forest, mountain). Why is this classification important?
[A] good classification system … captures the full diversity of the wildlife species being classified, at a scale that is meaningful for both conservation and management purposes. For any given caribou herd or population, we need to know what behaviors, movements, foods, pressures and habitats might be unique. Only then can we … make sure caribou are maintained in healthy populations across the different landscapes where they are found. (Hummel and Ray 2008, 41) After providing readers with an overview of what characterizes each caribou type, and discussing the importance these have for people and particular ecologies, the book moves on to discuss the natural and anthropic pressures affecting the herds' health. All this discussion provides the premises for a logical conclusioni.e. specific steps to protect the herdsupon which the various perspectives of stakeholders would have to converge.
The challenges of making different perspectives on a given "natural resource" to converge for "co-management" has generated very intense debates, particularly when 5 More details about the empirical case and the research project from where the ethnographic materials come are available in the article (Blaser 2016) and an associated interview in the Journal of Cultural Anthropology website https:// culanth.org/fieldsights/1075-film-and-cosmopolitics-an-interview-with-mario-blaser. 6 I switch between plural and singular in reference to both, caribou and atiku, and caribou/atiku according to context to stress the "more than one" or the "less than many" aspects of multiplicity.
Indigenous "perspectives" are involved. I will not dwell on these debates (for an overview see Spaeder and Feit 2005) besides pointing out that the old relativist/universalist debate is not far from the surface in them and, as such, at bottom they share a common assumption, that the problem is epistemological: it is about perspectives on "the world" (singular) and the possible relations they may hold or not. 7 As readers of this journal will be aware, materialsemiotics versions of STS would engage the differences being alluded to by the word "perspectives" as involving the performance of different reals. Works in this "tradition" call for a sort of displacement of the gaze from the "thing" at stakethat, according to the standard realistic assumption supposedly preexist the perspective that some might have upon itto the practices that perform the "thing." Annemarie Mol's aforementioned book, The Body Multiple, is now a classic example of this approach.
In that book, Mol shows how, in a Dutch hospital, the practices of radiologists, clinicians and pathologist perform different versions of atherosclerosis and, then, how this multiplicity is rendered singular, even if temporarily for the purpose of intervention. This, she argues, is accomplished through coordination and distribution: operations by which different performances are either made to hold together as a single entity or are kept apart to avoid mutual interference. Succinctly, coordination works by adding performances as if they were multiple perspectives on a single object and by discarding dissonant ones. Distribution, in turn, works by keeping different performances apart so that inconsistencies between them do not turn into clashes where some sort of adjudication of "truth" has to occur to preserve the unity of a given object.
In our case, to follow a similar approach would imply to look at the different practices that in the book Caribou and the North, are presented as being the "perspectives" of stakeholders. Without constituting an exhaustive list, the participants gathered at the North American Caribou Workshop that has been meeting for the last 15 years provide an entry point to briefly explore what these different practices might entail. 8 Typically, these meetings are attended by caribou biologists, wildlife managers, and representatives from environmental NGOs (e.g. the World Wide Fund for Nature, the Wildlife Conservation Society of Canada, and so on), outfitters associations, resource extractive industries and, Indigenous groups.
Each of these "stakeholders" is involved in practices that render slightly different versions of caribou. For instance, biologists "do" caribou as a techno-scientific artifact, the animal species R. tarandus. This artifact is done through the mobilization of a variety of practices nested into each other in the forms of established theories (from genetics and etiology, for example); techniques (such as statistical sampling, and aerial tracking of movements and surveys among others); instruments (like satellite collars), and institutions (universities, governmental agencies and so on), to mention a few. Wildlife managers, in turn, enfold this techno-scientific artifact as part of a complex "equation" that must estimate how human activities impact on the herds and strike a (hypothetical) balance. The unstated points of reference for the balance are (supposedly) the minimum required for the sustainability of a herd and the maximum profit that any activity that affect them can obtain. As the authors of Caribou and the North point out "[r]esearch is needed to better understand thresholds of human harvest, as well as our industrial footprint with its corresponding deterioration in habitat quality. How much of this can be tolerated by caribou, and what are their minimum protection need" (Hummel and Ray 2008, 189, emphasis added). The quote also indicates why corporations involved in forestry, mining, and oil and gas extraction are prominent sponsors of the annual North American Caribou Workshop and how they do caribou. Articulated with environmental legislation, and nested within the wildlife managers' equations, this techno-scientific artifact commonly called "caribou" is another element in their own equations to plan operations and perform their calculus of profitability; for instance they might consider that mitigating the impacts of their activities on caribou is too onerous financially, or that disregarding these impacts is too onerous in terms of public relations. Outfitters do caribou in a similar but slightly different fashion, the resource being the herds themselves. Their calculus is more directly aligned with having continued access to it. Environmental NGOs, in turn, might do caribou as part of larger entities such a species at risk, biodiversity, symbol of the north, and so on. I will leave Indigenous practices aside as I will focus on them in the next section. For now, and using Mol's analysis of atherosclerosis as a template and counterpoint, I want to briefly explore how the different versions of caribou that I have schematically characterized here hold together.
To use Mol's terminology, the predominant operation that holds together the various versions of caribou is coordination through adding up. Interestingly, more often than not, caribou as techno-scientific artifact operates as the invariant term upon which the addition is performed. This is perfectly reflected in Caribou and the North where the discussion about "what is caribou" is solely based on the disciplinary practices of biology. This does not mean that there are no controversies and dissonances between different versions of caribou, but these tend to refer back to controversies between performances of caribou as a techno-scientific artifact. For example, wildlife managers, environmental NGOs, corporate representatives, outfitters and Indigenous groups might back their own position on how to manage herds by reference to contending methodologies used to establish "significant units" for management; or might find fault with specific techniques used to generate data; or might refer to the uncertainty and perfectibility inherent to scientific knowledge to refuse a course of actions that contradict their own versions. What rarely happens is that caribou as a techno-scientific artifact is simply discarded or kept apart as it may happen with other versions. In effect, caribou as a source of income for outfitters might have to be discarded if it runs against what biologists say. While extractive industries' version of caribou might be harder to simply discard, they can only persevere as long as they can claim grounding on the techno-scientific version. By distinguishing in their practices aspects that can be attributed to culture (for example, the "belief" in spirit owners of the animals) and those that can be attributed to "empirical observation" (for example, migration routes) Indigenous versions can be kept apart, in the first case, or added up, in the second. 9 In a similar way, environmentalists' positions grounded on scientific (supposedly detached) observations will be engaged, while positions perceived as "emotional" (i.e. "tree-hugging") would be kept apart.
The predominance of one version of caribou reflects some illuminating contrasts between Mol's and our case. First, in the Dutch hospital, the various versions of atherosclerosis are in principle relatively symmetrical to each other; there is no established hierarchy between them. Second, the practices of the radiologist, the pathologist and the clinician are about diagnosing a condition, and hence any one of them being asked "show me atherosclerosis" will show something different that relates to their practice (a graph, a calcified artery, claims of pain by a patient). Then, and given the first point raised before, how this multiplicity is ordered and sorted in such a way that holds together constitutes a question that can only be answered case by case for there are no established hierarchies between practices. In our case, asking anyone "show me caribou" would result in a finger pointing to a four-legged being with antlers. Here it seems self-evident that caribou exists out-there, holds together by itself and preexists the various practices of knowing. 10 This reinforces a series of assumptions about the basic stability, orderliness and "given-ness" of the world/ reality that is independent and prior to knowledge practices. As John Law (2004) has argued, these assumptions are commonsensical for moderns and call for a specific way of knowing which is epitomized in the "the scientific method." Thus, for a variety of reasons including but not limited to the perfect fit between a commonsensical notion of what the scientific method can achieve and dominant assumptions about the nature of reality the techno-scientific version of caribou has become dominant when it comes to specifying what this "thing" is and, thus, how humans have to manage it.
The role that commonsensical assumptions about reality play in establishing a hierarchy of versions of caribou raises a larger question about their role in enabling the kind of relatively "civil" process through which the various versions of atherosclerosis are made to hold together in Mol's case. The point becomes sharper if we attend to practices that, operating on different assumptions (surprisingly close to those of these material-semiotics versions of STS), are not as easily enrolled in processes of singularization. To this, we turn next.
Once an old man and his son were very expert in hunting. And it happened that the son dreamed that he cohabited with the caribou. … he said to his father "I will depart. And I will kill caribou enough for the whole winter. So do not wait for me … I am going to go with the caribou." Then he sang: "The caribou walked along well like me. Then I walked as he was walking. Then I took his path. And then I walked like the caribou, my trail looking like a caribou trail where I saw my tracks. And so indeed I will take care of the caribou. I indeed will divide the caribou. I will give them to the people … He who obeys the requirements is given caribou, and he who disobeys is not given caribou … For so now it is as I have said. I, indeed, am Caribou Man (Ati'k'wape'o)". (Speck 1977, 81) This is one of the many versions of an atanukan (narratives about the origin of the current status of "things") accounting for the special relation the Innu sustain with atiku, what Speck translates as "caribou." Other versions refer to a man who went to live with the atiku enticed by one of their females, married her and became the boss or spirit master of atiku, also known as Kanipinikassikueu. 11 As in many similar Indigenous narratives 10 I stress "seems self-evident" to indicate that this understanding of what is implied by pointing to the four-legged being is not universally valid, as it will become evident in the next section.

11
There is no standard spelling for this and other Innu words as different writers have tried to capture the phonetics that varied across various Innu groups. I generally follow the spelling used by my Innu partners but retain the spelling used in original quoted sources.
accounting for how things are now, the underlying assumption is that nothing begins ex nihilo. In effect, as several Indigenous scholars and ethnographers have pointed out, in spite of specific variations, in many philosophical traditions in the Americas, there is a widespread shared assumption that whatever exists in the present is the product of a transformation in the previously existing web of relations (see Deloria 1999;Cajete 2000;Ingold 2000;Cordova 2007;Simpson 2011). In other words, the world or reality is conceived as in a permanent state of becoming, the emergent effect of relations, and therefore "relatedness" or "relationality" is primordial. As this literature shows, the preeminence of relatedness or relationality grounds a specific epistemology: knowledge is mainly about attunement with the ongoing unfolding of life-giving relations (see Burkhart 2001Burkhart , 2016. This way of knowing is not about generating (mental) representations of an already existing world to then inform action, but about participating through appropriate actions in the worlding of a world that can sustain the web of life-giving relations of which the knower is part. Thus, hunting for the Innu, for example, is not an action pursued following a mental map of a "known world", but a way of participating in a particular kind of worlding.
One crucial point to keep in mind is that atiku (as other non-humans) have full personhood and will of their own, therefore hunting is not mainly about outsmarting "animals" but rather about enticing these fully volitional beings and their leader to be generous with their bodies (see Kaniuekutat and Henriksen 2009;Castro 2015). This is achieved through a series of practices that show respect and recognition of these generous acts. Among these practices are very detailed protocols to dispose of the bones of hunted atiku (from which new specimens will regenerate), the injunction to not waste any part of their bodies, and the requirement that meat has to be generously shared among people. Other prescriptions, like keeping atiku in one's thought through storytelling, singing and drumming and celebrating a ceremonial meal (mokoshan), are geared to receive the blessings that this relationship generates for the general wellbeing of nitassinan, the land (see Henriksen 1973;Armitage 1992;Andrew and Castro 2016).
Considering the points above, it is no surprise that for many Innu, a decrease in the frequency of atiku giving themselves to hunters is a symptom that Kanipinikassikueu is angry. Reporting on one such circumstance, Henriksen (cited in Armitage 1992) wrote, The search for reasons to explain Katipinimitautsh's anger amounts to a self-examination to find possible failings in [the Innu] spiritual and moral relationship with nature, among themselves, and between themselves and animals … It may be something which is very serious: someone may have breached the rules of sharing, or someone may have been careless when handling the marrow from the caribou long bones. For example, after a ritual of makushan, held in the spring of 1976, one man said "Yesterday we had makushan in my house. One man went outside the house with the marrow he was eating. This is very dangerous. Something bad is going to happen".
Most of the elders and hunters that my collaborators and I have engaged in conversations, workshops, and interviews, understand the ongoing decline of atiku in this key. 12 Since the 12 Since 2009 I have been involved in a collaborative research project with a group of Labrador Innu elders and hunters from a nonprofit called Tshikapisk. This group was concerned about the precipitous decline of atiku, and wanted to find ways to address it. Initially the collaboration was geared towards collaboratively designing interventions with the Innu Nation (the Labrador Innu regional government) that would address the problem as defined by Innu experts. The collaboration has continued until now and expanded. As I write these lines, our team of Innu experts and academic collaborators are 1990s, when it numbered about 800,000 individuals, the George River Herd (GRH) has been decreasing exponentially, with its numbers reaching 9,000 in the last published survey of 2016. 13 In the above-mentioned conversations, elders and hunters said that the disrespect of younger generations for atiku was rampant. They spoke of atiku remains being carried away by dogs, of the people selling meat, and of a general lack of interest about life on the land by younger people. But they were not angry; they were worried. For most of them the consequences of all this were obvious not only in the decline of the herds but also in the epidemic of addiction, suicide and diabetes that has plagued younger generations of Innu for the last twenty years (see Samson 2003). In a workshop we held in 2010, the elder Ponas Nuke expressed the problem thus: "Without atiku we are nothing. If we are not in the land, hunting, it will come the day Kanipinikassikueu will not know us, it will ask "who are you people?" … and if we do not have its blessing, things will get worse." In 2010, the Newfoundland and Labrador's provincial government commissioned a study to evaluate the status of the GRH in order to generate a management plan in consultation with the main local stakeholders. The measures proposed hinged mostly on the conclusion that hunting was "now significant and cumulative to natural mortality," in other words, caring for the caribou population would require restricting hunting (Newfoundland and Labrador Wildlife 2010). In the presentation, it was also made clear that, from the government's perspective, conservation took precedence over Aboriginal rights to hunt. Since then, non-Aboriginals began to mount strong pressure on the government to put a ban on all hunting of the species. 14 Finally, on 28 January 2013 the government, citing scientific evidence and recommendations, announced a hunting ban on caribou that continues until now. 15 The very next day, Prote Poker, the Innu Nation Grand Chief said that the Innu elders did not agree with the ban and the communities would proceed to hunt. 16 Since then, the hunting ban has been a matter of constant friction between the Innu and the Provincial government as the former consistently refuse to follow it. Although several instances of consultations and discussion have been put into place and the Inuit of Labrador (represented by the Nunatsiavut Government and the NunatuKavut Council) have accepted the argument that the ban is necessary to respond to the decline, the Innu "perspective" cannot be brought into compliance. What this brings to the foreground is that atiku is an altogether different "thing" rather than one of the practices/versions that constitute the caribou multiple. This, I contend, makes atiku recalcitrant to the usual operations of singularization.
The multiplicity at stake, in this case, is different from the multiplicity upon which Mol and other STS scholars have focused. For the latter, the refrain that multiplicity is "more than one but less than many" can be captured by the image below.
carrying on an applied research program, that also enrolls the Canadian Wildlife Service, with the goal of fostering an institutional context that can generate responses that are appropriate for caribou but also for atiku. As with Mol's praxiography of atherosclerosis, multiplicity can be seen in the image as the various traces of the contour of the caribou which do not quite coalesce into a single line. When seen through practices (or the lines), the "oness" of atherosclerosis (or the caribou) appears diffracted, multiplied. Yet as it happens with the image of a caribou, the assumption that there is a single thing called atherosclerosis also contributes to hold this multiplicity together. The multiplicity of caribou can be also thought along these lines. The technoscientific, wildlife management, environmentalist, corporations and outfitters' versions of caribou do not always quite align with each other, but they are all oriented by a similar assumption, that there is a "thing" out there, the animal caribou. Such an assumption, and the tacit acceptance that the most accurate perspective on the "thing" is the scientific, enables the operations that allow the different practices to hold together. And I would add that such an assumption also plays a role in enabling the relatively "civil" procedures radiologists, clinicians and pathologists use to stabilize and singularize the multiplicity of atherosclerosis when their practices are rendered as perspectives.
Image by author.
Taken on their own, the concrete practices that constitute atiku perhaps could be described in a similar way; that is, practices such as pursuing atiku, singing to it, dreaming with it, sharing meat, or performing rituals might not align all perfectly with each other. But there is an important caveat; in contrast to caribou, there is here an awareness that each of those practices are intrinsic to the being of atiku, hence the constant concern that the practices be carried out in the proper ways. Carelessness is very dangerous as it will not render the "state of affairs" being sought. Now, given that the various practices associated with caribou and atiku encounter each other in the flesh of a being (so to speak), as I pointed out before, the multiplicity at stake is different. It is better captured perhaps by a popular trick image like the one below.
Here we have a bird looking to the left and a rabbit looking to the right, more than one but less than two. There is a bird and a rabbit, and yet they are not two units; and while the traces overlap, there is not just one drawing. If we imagine that the bodies of the bird and the rabbit do not overlap as neatly as their heads, we can grasp the idea that there might be partial co-occurrence of the entities, but the difference is not canceled. In a similar fashion, the material-semiotic assemblages and practices, from which the more than one less than many atiku/caribou emerges, partially co-occur (most evidently in bodily presence) but they remain distinct. 17 In this case, multiplicity refers to mutually entangled but divergent worldings. 18 Although the specific multiplicity of atiku/caribou has often been a factor in the conflicts between the Innu and various agents of the nation-state and corporations, it had not before irrupted in such a way as it has now with the hunting ban. This is a situation that makes it more visible that at stake here are different worldings. In other words, before the conflict over the hunting ban, atiku/caribou held together well enough to give credence to the idea that each word represented cultural perspectives on the same "thing." In the next section, we will take a closer look into how "equivocations" have helped in this "holding together" and why they can no longer do the same.

Holding on together in divergence
In this section, and under the rubric of "equivocations," I will discuss a number of "crossings" or points of encounter between practices associated with caribou and/or atiku. For this, it bears repeating that an equivocation refers to a situation in which interlocutors appear to be speaking of the same thing when they are actually referring to different ones; in our case, assuming that caribou and atiku refer to the same thing would be an equivocation. In my analysis, I use the term equivocation not only in relation to words or concepts but also to practices that might seem to have as referent one and the same "thing" but actually have as referent atiku or caribou. This does not mean that the said practices have no impact beyond their intended referent. Rather, precisely because they affect both referents, the practices are equivocal. The question I pursue is, in which ways does the equivocation affect the quality of the crossing or encounter?
As pointed out above, caribou population studies are central to governmental decisionmaking regarding the management of the species. These studies are the aggregate result of theories and models about population cycles and ethology, aerial surveys, tracking of movement trough radio-collars, analysis of individual specimens' physical status (body fat, weight, parasites, etc.), and reports from hunters (including harvest data, location of herds, observation of body condition of specimens hunted and so on). Here, we have a first crossing or encounter of practices that we can follow in pursuit of understanding how equivocations help in holding together the divergent multiplicity of caribou/atiku.

17
Partial co-occurrence in bodily presence enables the equivocation inherent to the assumption that, in pointing to a fourlegged being when asked "show me a caribou or an atiku," Innu and Euro-Canadians are pointing to the same "thing."

18
The presence of diverging worldings under the (imposed) appearance of similarity has been a point consistently made by Indigenous analysts and scholars (see Goeman 2013;Watts 2013). The fact that such a point may appear as novel to a non-indigenous audience speaks to the asymmetries inherent to coloniality, as I will further discuss soon.
While hunting practices contribute "data" for population studies, some data gathering techniques contribute to the performance of hunting practices. In effect, until recently, the location of the GRH transmitted by radio-collars was posted with very little delay in a web site of the Quebec ministry of the environment. Innu hunters in Labrador used this information to travel to those areas where the hunt was more likely to be successful. Neither the data provided by the hunters nor the location the radio-collars provided are determinant to either set of practices (hunting atiku and researching caribou) but they reinforce each other even if unwittingly. Notice that in this case atiku and caribou (or some of the practices that constitute them) relate to each other in "productive" ways without a requirement that both entities be mutually equivalent. Somehow, the difference between entities remains invisible or irrelevant; they both "flow" without interrupting each other. But this is not always the case.
In a recent ethnography Damian Castro, one of my collaborators, reports an incident that shows how the contrasting assumptions about atiku and caribou might engender conflicts involving the collars.
One day, while I was in the Innu Nation office, a very experienced hunter who had been recently charged with illegal hunting came to the office where I was working and told me "they found a Red Wine [protected herd] collar close to lake Kamistastin; see, atiku wants to go there". Lake Kamistastin is located about 400 kilometers north of Sheshatshiu, very far from the Red Wine Herd range, and right in the migration area of George River herd. This information, as he and other Innu argue, shows that the Red Wine woodland herd and the George River migratory herd intermingle, therefore, there is no point in declaring the hunt illegal on the basis of the assumption upheld by government scientists that the herds are different: for the Innu there is only atiku. Furthermore, the words of this hunter obliquely indicate differences in how the collar information is used. The government uses it to obtain the information the scientists need to learn about caribou behavior, such as their whereabouts, while the Innu use this information to know what atiku wants. In other words, while the government administer the collars to satisfy their will to learn, the Innu use it to learn the will of atiku. Like human beings, atiku has will. (Castro 2015) The collars are nowadays a key component of caribou population studies and the calculations that inform regulations impacting the practices of hunters. But regulating hunting is only the tip of the iceberg of a larger wildlife management dispositive. Lets take a look at how atiku and caribou encounter each other in this larger dispositive of management.
In 1996, the Voisey's Bay Nickel Company Ltd. (VBNC) filed an application to proceed with a mine and mill project in northern Labrador. Subsequently, and following standing legislation, the company had to present an Environmental Impact Assessment Study (EIAS) that would then become part of a larger process of environmental impact assessment by an independent review panel. The Innu refused to participate in the EIAS as they staunchly rejected the project due to the effects that it could have on atiku and other non-humans (see Innes 2001). Nevertheless, the EIA went ahead, including an estimation of the impact the project would have on the George River Herd that hinged upon studies made on that population. Based on these estimates, the review panel made a series of recommendations to mitigate the impacts and that would allow the project to proceed. Thus, nested within them, the techno-scientific artifact "caribou" (at the center of caribou population studies) became a building block of the environmental impact assessment, the recommendations of the panel, and the planning of operations of the mine.
Although the Innu struggled and opposed the mining project, even physically disrupting operations on the ground during a two-weeks standoff with the Royal Canadian Mounted Police, they were aware that the deck was stacked against them: What choice do we have to protest or negotiate? We are trying to protect the land and the animals and on the other hand [are] the big companies. No matter where we stand, it's like giving away our land. And if we try to stop it, we won't go nowhere. Mining developments will go ahead anyway. (George Gregoire, Innu Nation Officer, cited in Innes 2001) Eventually, the Innu signed an Impact and Benefit Agreement (IBA) with the mine. The mine's website provides a concise and clear explanation of what is an IBA: Impacts and Benefits Agreements (IBA) are typical where a significant project is proposed for development on a First Nation's "traditional lands". Traditional lands are the First Nation's ancestral lands over which they have Aboriginal rights. Mining projects, as an example, have the potential to have social, cultural and environmental impacts on traditional lands and on local communities … IBAs are formal, written agreements between companies and First Nations that help to manage the predicted impacts associated with an industrial development and to secure economic benefits for neighbouring communities affected by that development … . 19 Although the terms of the IBA that the Innu reached with the mining company are confidential, some of the economic benefits derived from it have been used to sustain an outpost program where Innu families are supported in going by aircraft to live in camps in the bush for as much as three months (Castro 2015). These kinds of programs have become very important for those Innu who have been progressively driven to live in permanent settlements and depend largely on wage labor yet want to sustain their connection with Nitassinan and Kanipinikassikueu. Hence here we have another (more mediated) "crossing" or encounter of practices that at the same time disrupt and reinforce each other. In effect, the caribou population studies "nested" within the EIA and the review panel's recommendations paved the way for calculations that would make feasible a project that disrupts hunting practices in a certain place. The defense of these hunting practices (and more generally Nitassinan) led the Innu to initially reject the project, which in turn led the mine to the negotiation of an IBA that now contributes to sustain Innu hunting practices in the present circumstances.
Thus, while in a particular place/moment (the impact assessment) the nested practices of Caribou population studies, on the one hand, and hunting, on the other, interrupt each other, in another place/moment (the IBA) the interruption is worked around (although not eliminated) enabling the continuation of both (including of course the entities which they contribute to enact, i.e. atiku and caribou). We do not have here a process that produces "common ground" between diverging entities and the practices associated with them but rather a process that through "displacement" enables the sustenance of these divergent yet entangled entities; but only up to a point. In effect, the divergence is grounded on an asymmetrical equivocation, and thus the "holding on together" is not very robust.
What does it mean that the equivocation, in this case, is asymmetrical? For those mobilizing caribou, atiku is seldom anything more than a "word" in another language to reference the same "thing." The signing of the Impact and Benefits Agreement reinforced this 19 See http://www.vbnc.com/iba.asp. presumption of synonymity after a period of turbulence (while the Innu opposed the project). In effect, the IBA was acceptable from the mining and governmental stand point because, beside (now) attending to Innu "interests," it was ultimately based on the Environmental Impact Assessment within which was nested caribou as a techno-scientific artifact. In other words, after having had their "interests" (narrowly assumed as economic) addressed, the Innu had supposedly agreed on the soundness of caribou studies, or what is the same, had accepted the equivalence between caribou and atiku. For the Innu, in contrast, that caribou is not exactly atiku is made evident by the constant interruptions of the practices that bring the latter into being. Just remember the reference in Castro's quote above to the hunter charged with illegally hunting caribou from what biologists have established is a protected and distinct herd different from another one that was not then protected.
"Displacement" is different from Mol's "keeping apart" practicesto avoid that their crossing forces an adjudicationin that the adjudication rather than totally avoided is only deferred. In effect, while for those doing caribou the signing of the IBA may push the equivocation to the background (i.e. they end up thinking, "it seemed as we were talking about different things, but at last we have agreed that caribou and atiku are the same thing"), for the Innu it is evident that the equivocation will hold only so long; as Edward Piwas, an Innu elder, presciently stated: The smog from the milling plant will kill the plants and animals. And it will float into our community. We will not see the smogit will slowly kill the animals and us … . The wildlife officer will know when he can't find any animals. He will blame us for the lack of them but he will not think about the drilling. (cited in Innes 2001) In relation to my argument, I take the statement to indicate that the practices that constitute caribou (and through it, the EIAS and its suggested mitigation measures) would not sustain atiku and this would become evident in time, and when this happens those doing caribou will not question their own practices, such as industrial activities enabled by the caribou artifact, but those that do atiku, such as hunting. The passing of the ban in 2013 brought this scenario to fruition. But, in contrast to what happened with Voisey's Bay mine, now the interruption of atiku cannot be displaced as it occurs in one of the most crucial practices that constitutes it, hunting (as the giving and receiving of gifts from Kanipinikassikueu). Hence, the Innu refusal to accept the ban. The asymmetry in the equivocation of assuming that "caribou" and "atiku" refer to the same thing is foregrounded by an ironic call by Simeon Tshakapesh, chief of Natuashish, to make caribou and atiku symmetrical through the practices associated with each: If they're so sincere about the George River herd … I think they should stop the … mining and exploration. I wouldn't hunt if they say, "Okay, we want to save the caribou herd, we're going to shut all the exploration, mining companies, and other projectswe're going to shut it down for the next five years". 20 In other words, by making what he knows is an impossible demand on the colonial state Tshakapesh tries to foreground the extent to which the ban is an impossible demand for the Innu as well. Symmetry by the negative.
My final, very brief, example involves the attempt my Innu partners and I have been making since 2012 to stage an equivocation that would enable atiku and caribou to hold together symmetrically while avoiding the hunting ban. In my previous article (Blaser 2016), I described our first attempt to do this before the ban was passed; here I retell the argument we have been mobilizing. Succinctly, we have been proposing to wildlife managers that a limited hunt of atiku be allowed to the Innu communities under their strict "traditional" protocols. The pitch is that the amount of work required to follow the protocols and their enforcement by the communities themselves would have better results in terms of addressing the biologists concerns with population control than a ban that no Innu will ever abide by. In other words, a restricted hunt would cater to Innu concerns with keeping a relation with Kanipinikassikueu and to biologists' concerns with keeping tabs on how many and what kinds of caribou are being hunted (male, females, calves, healthy, unhealthy, and so on). A homonymic practice with two different referents. 21 In summary, from our brief overview of "crossings" in the caribou/atiku case it is evident that rather than singularization what "holds together" divergent multiplicity are various versions of equivocation, that is situations that by chance or design allow for divergent practices to be sustained through their mutual crossing or encounter. This may occur through equivocations that allow different practices to come together either without interference with each other, or even enabling each other as it happens with caribou collaring that enables more time-efficient atiku hunting. This kind of mutually enabling crossing might also be consciously staged as we have been trying to do since the ban was passed. Yet, that equivocations will do this work is never guaranteed, especially when they are premised on displacement. Here it seems that a dynamic is at play that might make their results perhaps to appear less stable than "matters of fact." In effect, in these cases, mutual interruptions between divergent practices seem only be postponed. But one may point out that "matters of fact" are only apparently more solid, as they can always turn into "matters of concern" again (Law 2004). Thus, the unstable character of the co-existence that displacement allows divergent yet entangled practices is not necessarily problematic. What it does flag out, however, is the need to be attentive to situations in which "displacement" enables the maintenance of asymmetrical equivocations to remain hidden. As our case seems to indicate, these situations might contribute to the intractability of encounters between divergent practices down the road. Overall, however, equivocations as a possible template to hold together divergent practices seem to offer a promising path, particularly if we consider that singularization as the other possible path might be unescapably connected to coloniality. To this, I turn in the conclusions.
Conclusions: coloniality and diffractive multiplicity I concluded my initial exploration of "another cosmopolitics" pointing out that my suggestion of pursuing equivocations as an alternative to singularization should not be understood as a matter of either/or but rather of both/and. However, the sharper 21 As I narrate in the article, in 2012 the idea was taken up by the wildlife officers (and has been consistently well received since then by on-the-ground agents). But soon after, under public pressure, the decision was made higher in the ministerial hierarchy to pass the ban (and the refusal at that level has remained constant as well).
contrast between diffractive and divergent multiplicities my case study has allowed me to stage here suggests that such an inclusive approach might need to be taken with caution.
Here I am coming back to the issue of asymmetrical equivocation and the point I raised before that the presence of "diverging worlds" under the appearances of sameness has been evident for Indigenous (and more generally colonized) peoples all along, but not so for many non-Indigenous ones. This is because, through the practices that constitute it, coloniality produces a "zero point" of observation from which the colonizers cannot but see their own world constantly being reinforced as the only one (Castro-Gómez 2005). In effect, read through a colonial matrix that renders them erroneous and/or unreal, colonized worldings are amenable to translation and/or pedagogical coercion that effectively disappears or invisibilizes them as alternatives, while simultaneously rendering colonial categories commonsensical (see Vázquez 2011;Santos 2014;Escobar 2016). When the Innu expressed that mining would go ahead regardless of whether they negotiated and accepted some form of compensation or not, they were expressing a "lesson" learned. And part of the lesson was that opposing the "knowledge" of the colonizerthat said that the mine could go ahead with no impact on caribouwould be met with coercive measures supposedly justified by common sense. As I pointed out, the displacement of adjudication that the IBA enabled contributed to reinforce (and naturalize) the "commonsensical" yet colonizing assumption that atiku is just the Innu word for caribou.
I would argue that this dynamic of invisibilization and erasure of divergent worldings through more or less overt coercion might not be unrelated to the very possibility of treating caribou (or for that matter, any other entity) as a diffractive multiplicity and of using operations of singularization to hold this multiplicity together. In effect, the looming threat of coercion (through the presence of the RCMP during the mine conflict and the threat of arrest under the current ban) seems to be a "normal" part of the procedures used to stabilize caribou in the face of the disturbances produced by atiku (and its associated practices). This situation certainly contrasts with the more civil procedures used in the Dutch hospital to singularize the multiplicity of atherosclerosis, and the same goes for the procedures through which various "stakeholders" may coordinate their "perspectives" on caribou that, as we saw, might be diffractively multiple but amenable to singularization via adding up upon the basic techno-scientific version of it.
This leads me to my concluding point about being cautious with my original call to an all-inclusive approach to various versions of cosmopolitics. The contrast between procedures used to hold together these different multiplicities raises the question of whether in order for diffraction to be the working ground for a version of cosmopolitics, divergence would not have had to have already been reigned in. In other words, might multiplicity as diffraction not be already a product of coloniality's way of dealing with multiplicity as divergence? And if so, what invisibilized divergent worldings might be there in situations where diffractive multiplicity is all that seems to be at stake?