Power in a minor key: Rethinking anthropological accounts of power alongside London’s community organisers

Anthropological accounts of power remain characterized by an enduring tension. Social scientific theories of power allow anthropologists to situate subjects and mediate between contending perspectives. However, in doing so, such theories inevitably also end up displacing the grounded perspective of interlocutors themselves. This tension sustains a contentious debate, which positions attention to power and attention to grounded perspectives in opposition. In this article I draw on ethnography conducted with the UK’s largest community organising body, Citizens UK, to trace an alternative approach to this tension. For Citizens UK organisers this tension becomes a way of driving change by enrolling diverse actors in collective projects and by displacing hegemonic understandings from within. Good theories, for Citizens UK organisers, are characterised by the practical ability to mediate between contending positions and, in doing so, transform them. To make sense of this mode of theorisation I take up queer theorist Jack Halberstam’s notion of ‘low theory’, geographer Cindi Katz’s notion of ‘minor theory’, and I draw on the linguistic anthropology notion of ‘register’. This allows me to unpack how organisers use theory to act, but also to trouble established anthropological understandings of what theory is and what it ought to do.

Power plays an odd role in the anthropological imaginary. The turn, since the mid-1980s, toward what Ortner (2016 calls 'dark anthropology' -that is a concern with 'questions of power, inequality, domination, and exploitation' -has made power into a key term within the discipline. Nonetheless, an unresolved tension persists between attempts to trace and analyse power across varying scales and contexts, and attempts to ground description and analysis in the standpoints of others. Each of these perspectives seems to foreclose, or at least significantly constrict, the other. This tension has been an animating force behind many of the theoretical pendulum swings of the past several decades -ranging from the deconstruction of structuralist and Geertzian culturalist paradigms by feminist, postcolonial and 'writing culture' critiques, to the more recent development of the ontological turn and the anthropology of ethics.
In this piece I do not attempt to resolve this tension -in fact, I follow the approach of those such as Marilyn Strathern (1988) and Donna Haraway (1991, 1994 who argue that this tension is, perhaps even ought to be, irresolvable. What I aim to do instead is to draw on these arguments for why and how we might leave this tension unresolved, but to trace a different modality of irresolution than that offered by Strathern or Haraway -one that holds the tension not between different theoretical perspectives, but between hegemony and transformation, and the forms of academic and everyday knowledge that support them. To do so, I draw on ethnography conducted with the UK's largest community organising body, Citizens UK (CUK). Between 2019 and 2020, I spent 12 months shadowing CUK community organisers, undertaking training as an organiser, getting involved in campaigns and spending time at the diverse organisations that make up CUK's membership. Most of my time was spent in London, but the ethnography here draws on experiences with local 'alliances' across five areas, also including Birmingham, Cardiff, and Tyne and Wear. In each locale, CUK organisers worked to construct alliances of institutions that could pursue shared political goals. In London, Albanian refugees worked alongside secondary school students to campaign for police reform; in Newcastle churches, mosques and university academics worked together to demand greater protection against gendered hate crime. My driving interest was in whether and how CUK were able to build successful political coalitions across the lines of meaningful difference -drawn in terms of faith, ideology, cultural background or material interest -that existed between the different institutions that they worked with.
Based on this work, I want to make a somewhat unusual move. For anthropologists interested in power, accounts of what power is and how it operates are almost always dominated by academic theory. Meanwhile, for those who believe a focus on power displaces our ability to understand the world in emic terms, local accounts of power are usually read as part of something else -such as cosmology, ethical hope and so on. Either way, emic accounts of power are subsumed within a broader framing. In contrast to these approaches, here I hope to describe how organisers conceive power, and to take this seriously as a sufficient theory of power in its own right.
This doesn't mean I do away with academic theory altogether, however. As we will see, the ways in which CUK organisers imagine power might initially appear to do more to reproduce dominant structures than to challenge them. To understand how this seemingly hegemonic language can open up the possibility of transformative change, I take up the linguistic anthropology notion of register, building on its conventional use. Register refers to a mode of expression and interpretation -a way of using language, that draws on, but also inflects and reinvents existing communicative possibilities. Familiar registers include ritual language, street slang, or 'newspeak'. Most frequently, registers are described in terms of particular expressive forms -lexicons, tones of voice, patterns of syntax and so on (Agha, 2008) -but they also incorporate meta-communicative conventions, which involve shared expectations around how meaning is encoded and decoded, what language is and does. As such, registers can provide a collectively accessible, collectively cultivated means for reflexive understanding and action (Keane, 2003) that sits within but is not wholly determined by more dominant frameworks of meaning and practice. Here, I argue that organisers' accounts of power serve as one such register -which sits within and often depends upon more dominant, and often ideologically constrained frameworks, but which is not exhausted by them, and which has a certain capacity to transform them from within. In turn, I draw on notions such as 'low' (Cox, 2015;Halberstam, 2011) and 'minor' (Katz, 1996) theory to highlight the ways in which organisers' theories of power gain their efficacy precisely by being 'squeezed through the pores' (Katz, 1996: 489) of everyday, even hegemonic worlds, rather than being situated within the 'high' registers and spaces of the academy.
Of course, bringing in these bits of theory inevitably displaces organisers' own accounts of power, to some degree -by implying that these accounts alone do not tell the full story. The main displacement I intend to effect, however, is on academic ideas as to the location and adequacy of theory. That is to say, I attempt to take my lead from organisers' own theory of power, while drawing on notions such as register or minor theory to help translate what this (emic) theory is doing, in relation to more established ideas of what (academic) theory is and does. In the conclusion, I draw on these gestures of translation to offer some reflections on the potentials and limits of theory to support social change.

Power inside-out
To make sense of power, Daniel Souleles (2021) argues persuasively, it is not sufficient simply to produce a 'thick description of a particular group's ambient social life in that group's own terms'. Rather, he argues, such accounts need to be supplemented by 'theories of motion' (in press) -which capture how particular subjects shape and are shaped by others, across a range of places and scales. Souleles uses the example of financial traders, who often approach trading as a playful, competitive, gambling-like pursuit. He argues that while it would be possible to trace the meanings and dynamics produced by traders in terms of jockeying for status, or as enacting and modelling a system of thought -that is, by producing a thick description of how gambling operates within the social field shared by traders -this would be misguided. Specifically, this would overlook how this gamblers' attitude intersects with a hegemonic discourse where financial markets and the traders within them are imagined and legitimated as being rational agents for the efficient allocation of capital -and how this intersection serves to impose forms of volatility and risk inherent to gambling onto the lives of countless ordinary citizens. Crucially, Souleles shows that drawing on theories of motion and going beyond inwardlooking analysis is not simply important for understanding traders' impacts on the lives of others, but also for understanding traders themselves as subjectswith their gambling practices taking on new meaning, as an elitist form of selffashioning, by dint of their structural position.
Souleles' call for theories of motion makes explicit an argument that first began to crystallise as anthropological interest in power increased in the 1980s. This interest, sustained to this day, has no single genealogy, but was heavily shaped by the intersection of several strands of scholarship, including feminist, Marxist and postcolonial critiques -both of social relations in the places anthropologists studied and of anthropology itself -practice theory, studies of globalisation and cultural transformation, Foucauldian theories of power and knowledge, as well as the reflexive turn within anthropology (Ortner, 1984(Ortner, , 2006(Ortner, , 2016. Across these different strands of work, power emerged as a key term of analysis. For instance, postcolonial scholars pointed to the ways in which forms of knowledge -including that of anthropology -were entangled with projects of colonial rule, and to the ways in which techniques of colonial rule continued to structure life in the postcolonial present (Asad, 1973;Mbembe, 2001;Stoler, 2002Stoler, , 2016. Feminist anthropologists called for scholars to trace how 'the reality of male privilege affects the lives of every woman, whether she is conscious of it or not' (Wolf, 1992: 133), or sought to centre the perspectives of women and other marginalised actors, which were often obscured by dominant voices or homogeneous understandings of culture (Abu-Lughod, 1991;Cowan, 1990;Stacey, 1988). Scholars of globalisation and nationalism interrogated the power that went into the social construction of seemingly bounded or homogeneous places and cultural groups (Gupta and Ferguson, 1997;Hannerz, 1986). All these approaches stressed the ongoing differences of power and positionality that shaped contemporary fieldwork encounters.
These entangled approaches, of course, did not lead anthropologists to adopt a single model of power, but rather positioned power as a key focal point of debate (for a distilled overview see Ortner, 2006). Nonetheless, I would argue, different anthropological approaches to power are united by a shared metaphysics, which takes power as a pervasive dimension of social life -where all relations are, at least in part, relations of power.
This view of power as inherent to the social makes plurality into a tricky problem. The more anthropologists acknowledged that the social world was composed of multiple positionalities, the more it fell inescapably to anthropologists themselves to mediate between these positions. Although anthropologists often sought to centre marginalised perspectives -sometimes even championing these as 'less partial and less distorted' (Harding, 1989: 83) -and to account for their own positionality, partiality and distortion nonetheless remained inescapable. Faced with irreconcilable perspectives and contending truth claims, anthropologists increasingly turned to social theory to mediate such tensions. Theory offered the promise of an external language for situating the perspectives of different actors (including anthropologists) in relation to each other -seemingly irreducible to any of the perspectives in play. It's in this light that Souleles' description of theories of power as 'theories of motion' becomes especially helpful, highlighting the ways in which theories of power are called on to do this mediating work -to join up heterogeneous positions, interests and worlds. Yet, if social theory is being called on to join up and mediate between diverse perspectives, then it also necessarily takes on an encompassing, even transcendent, quality -operating by providing a higher-level account than those which it purports to explain (Lutz, 1995).
This reliance on social theory to account for power, however, has created an enduring tension -one influentially unpacked by Marilyn Strathern (1987) in her reflections on the 'awkward relationship' between feminism and anthropology. Feminism, Strathern argues, builds on an assumption that relations of (gendered) otherness are relations of oppression. This assumption is necessary to sustain an epistemology that aims to uncover what oppression occludes. Anthropology, meanwhile, builds on an assumption that relations of otherness are dialogic. This assumption is necessary to sustain an epistemology, where ethnographers allow interlocutors to make transformative claims on their own ways of knowing, so as to open up new ways of understanding both oneself and others. Each perspective also envisions different processes of subject formation: where for feminism the goal is 'becoming conscious of oppression from the Other' in order to 'discover the self' that lies beyond oppression, in anthropology the goal is to break with one's past self, through the transformative process of trying to understand and represent the Other (Strathern, 1987: 289). These approaches sit in an 'awkward' relationship to one another because each risks undermining the basic epistemological claims and processes of subject formation which drive the other. In other words, theories of motion do not simply connect -in doing so, they also displace.
An acute sense of this epistemological impasse, between approaches which attempt to take power seriously and approaches which attempt to take the perspectives of others seriously, remains visible in recent theoretical turns. In an influential call for a new anthropology of ethics, Laidlaw (2013: 2) has argued that 'modern social theory suffers from a propensity to describe social life as if ethics were a superficial, marginal, illusory, or fraudulent dimension of it' -taking aim at theories of power as a primary source of this tendency. Similarly, in setting out what it would mean for anthropologists to take indigenous ontology seriously, Eduardo Viveiros de Castro (2015) takes a piece from David Graeber, which argues that Merina idols reveal the power of collective ideology, as a counterexample. He emphasises that sensitive though Graeber's analysis is, it relies on an ultimate understanding that 'Magical powers, as the Merina conceive them, cannot exist' (Viveiros de Castro, 2015: 13), since, in Graeber's analysis they are ultimately an instance of something else -of ideological power. In order to take difference seriously, Viveiros de Castro argues, analytical gestures such as Graeber's, which foreclose local understandings, must be declared an 'illegal move' 'in the disciplinary language-game' (Viveiros de Castro, 2015: 13). The stark, sweeping nature of these critiques reveals a corresponding sense of epistemological constraint, even incommensurability, which many anthropologists have come to associate with the encompassing nature of social-scientific theories of power.
Rather than attempting to circumvent or suspend questions of power, however, here I hope to follow and build on a different strategy for navigating this impasseone which Strathern (1988) sets out in The Gender of the Gift. Here, she traces three discussions: the anthropological debate around relativism versus universalism; feminist questioning over whether and how to interrogate women's oppression in different cultural contexts, such as Melanesia; and the contrast between gift and commodity exchange. Strathern argues that it is impossible simply to reduce any of these questions to the others and instead opts to hold them in tension with one another, enabling each conversation to inflect and displace the others. Thus, for instance, she argues that it is impossible to assess whether Melanesian women are exploited without taking into account both Melanesian notions of personhood and value (vis-a`-vis gift exchange) as well as the relative cultural location of feminist models of power and exploitation. Donna Haraway (1994) offers a helpful metaphor for understanding this strategy in similar reflections on the incommensurability between perspectives emerging from science studies, feminism and cultural studies. She uses the image of a game of cat's cradle -a game played by weaving a loop of string between the fingers of both hands into shifting, net-like patterns -as a strategy for productively committing to the 'tension of holding incompatible things together because both or all are necessary and true ' (1991: 149). Strathern (1988), however, reminds us that useful as this strategy is, it is fundamentally a strategy for producing academic knowledge. That is to say, the forms of knowledge being displaced and inflected are not commonplace cultural understandings, per se, but the academic understanding of Western culture, Melanesian personhood, the feminist project and so on. Thus, on the one hand, Strathern and Haraway redirect theories of motion, away from the domain of transcendent, all-encompassing abstraction, but at the same time, in doing so, they ultimately lead us firmly back to academic ground.
A different take on this solution is signalled by Webb Keane and developed by Aimee Meredith Cox. Keane (2003: 223) argues that anthropology has long been characterised by 'a tension between epistemologies of estrangement and of intimacy' -between abstract theory and irreducible local understandings and experiences. But, Keane notes, the worlds anthropologists study are already, themselves, characterised by this same tension, where the ability to reflect and act on the world unfolds through locally grounded 'metalanguages of agency' that abstract from but remain linked to complex, situated knowledge and experiences (Keane, 2003: 230). Different metalanguages -different theories -Keane stresses, allow actors to understand, join up and act on different parts of the world in different ways. In other words, both the substance -what theory purports to explain, and how -and the legibility of theory matters tremendously for what can be done with it. Taking up this insight concretely, in her ethnography of the citizenship practices of black girls living on Detroit's margins, Cox (2015) centres her interlocutors as the primary theorists of their own lives. The forms of care and citizenship they enact, Cox notes, have -at times -already been traced within feminist and anti-racist theory. Even so, she argues, it matters to make these points through the voices and lives of black girls themselves. Partly, Cox is concerned with the politics of representation -with how black girls are always spoken for by others. More significantly, however, Cox is interested in taking what black girls do seriously as a form of citizenship -as a connective exercise in its own right which creates new relations of entitlement and responsibility, that play out in fraught circumstances. What Cox's interlocutors hold in tension are different relations of care and constraint -using the possibilities in different locales, or different performances of self, to displace and inflect one another. Rendering this in the encompassing language of academic theory smooths over the cracks that they navigate, and their everyday efforts in holding different spaces, selves and relations together. Crucially, by staying close to her interlocutors' own experiences and dilemmas, Cox holds onto the possibility for theory to drive action, by ensuring that theory remains linked to local epistemologies of intimacy -the experiences and struggles of everyday life (see Middleton and Samanani, 2021 for a similar argument).
In the rest of this article, I follow Keane and Cox's lead, in approaching theory as an everyday practice and as a tool for transformation, capable of holding material constraints and creative practices, understanding and action, all in productive tension with one another. I provide an account of how organisers think and talk about power that attempts to take it seriously as a theory of power, and I interrogate both the potentials and limits of this mode of theory.
Organising 101: 'If you want change, you need power' 'Community organising' is sometimes a loose term and has been used to refer to a variety of activist and community-oriented work. For Citizens UK, however, community organising refers explicitly to a specific tradition of thought and practice emerging from the USA, associated most prominently with Saul Alinsky (considered the 'father of modern community organising') and the Industrial Areas Foundation, which Alinsky helped found. There is plenty of contestation surrounding this tradition, including over what legitimately falls within it, who the appropriate forebears are and so on but, importantly, this contestation departs from an initial consensus that a distinctive tradition does exist. At CUK this tradition was elaborated both explicitly -in trainings run by organisers for local groups, or aimed at training new organisers, in lesson-plans and notes produced by organisers, in talks given by CUK organisers or guest speakers, in circulated reading lists rivalling those of any Master's course in scope and complexity, and in the guidance and encouragement organisers provided to those they worked withas well as implicitly, as a logic guiding the practice of organising. Here, I focus on those elements of this tradition which serve to outline a distinctive theory of power.
CUK operates as a membership-based organisation. Local institutions would join as dues-paying members, forming part of local 'alliances'. Member institutions fell into three broad categories. Most common were faith-based institutions, including churches, mosques, synagogues and gurudwaras. This was followed by schools and universities, and then the occasional charity or political organisation, such as a youth club, neighbourhood association or union branch. In London, where the majority of my ethnography took place, these alliances typically spanned a borough -so, for instance, local alliances would be known as 'Brent Citizens', 'Waltham Forest Citizens' and so on -although in some cases alliances operated at smaller scales, and elsewhere, nationally, they covered wider geographies, such as multi-city regions.
These alliances were often seen, and often presented themselves first and foremost as collectives fighting for grassroots political change. Different alliances took up a range of issues, from lobbying local councils to build more social housing, to fighting for greater rights and support for undocumented migrants. Member institutions were expected to conduct 'listening' within their own ranks, before then coming together with other local institutions to democratically decide on collective priorities. CUK organisers, whose salaries came in large part from membership dues, positioned themselves as 'working for' alliances, supporting them both with local listening and deliberation, and then with efforts to fight for change.
Political change, however, was only one part of the work of organising. Organisers would often tell me that CUK had three intertwined goals: to develop the capacity of local 'leaders' to take part in political life; to strengthen the capacity of member organisations; and to win 'meaningful' political change. Organisers shared an understanding that all three goals were fundamental to organising, but would differ as to the relative importance of these. Several organisers had joined CUK as young, frustrated activists, disillusioned by what they saw as a contrast between the high-minded ideals found in activist circles and a persistent inability to produce actual change. These organisers were attracted by CUK's record of successful campaigning. For instance, the 'living wage movement', which CUK pioneered in the UK, was described as 'perhaps the most successful grassroots campaign of the last decade' (Toynbee, quoted in Bolton, 2017). These organisers often put particular emphasis on the 'urgency' of political changewhile identifying the development of individual leaders and local institutions as essential steps towards this. Others, however, insisted that all three goals were of equal importance, while yet others acknowledged this while feeling they themselves were more drawn towards, or adept in supporting, one goal in particular.
Taken together, the three core goals of organising expressed a distinctively civicrepublican vision of political life (see Feldman, 2013), where politics was not simply about the exercise of detached reason or the pursuit of individual interests, but about the interweaving of several elements, including: everyday experience; feelings of belonging; relationships of attachment and accountability; collective traditions such as those found in faith institutions or in the culture of a school; and particular forms of subjecthood. This intersection was understood as reimagining political commitment less as a matter of personal conviction and more as conviction which emerged through the embodied experience of collective life. In this regard, the goals of developing leaders and institutions were viewed by many organisers not only as instrumental to political change but, more broadly, as essential to reinvigorating and reorienting political life. Tellingly, while Citizens UK was outwardly best known as a campaigning organisation -with these questions around the nature of politics largely escaping public attention -their slogan was nonetheless 'reweaving the fabric of society'.
Nonetheless, organisers also recognised that for many 'leaders' -the term used for all those within member institutions who engaged with the organising processpolitical change was the most immediately compelling of its three goals. This provided a point of departure for thinking about power. 'If you want change, you need power', organisers often repeated. In turn, they simply defined power as 'the ability to act', often invoking its etymological origin from Latin, posse, 'be able' -or its cognates in romance languages poder, pouvoir. Organisers would unpack the nature of power by drawing a distinction between 'organised money' and 'organised people' as the two primary forms of power. The actors that were likely to be the targets of organising efforts -found within 'the state' and 'the market' -had plenty of organised money, but communities could counter this through organised people. Organisers then specified that communities were most effective at organising people through a specific modality of power -'relational power' -which was sometimes contrasted to the ability of the state to organise people through the coercive force of law.
To explain how community groups might build relational power, organisers brought in another concept -'self-interest'. Many people claimed to care about a range of issues in the abstract, organisers would argue, but self-interest was the root of committed, long-term action. There were various dimensions to self-interest, such as one's close relationships, formative experiences, material needs, and fears and anxieties, but however this was achieved, building relational power required connecting with the self-interests of others. This meant that issues needed to be accessible and relatable to a range of perspectives, and through different relationships.
To identify and connect with the self-interest of others, the best methodorganisers argued -was through one-to-one conversations. These conversations were meant to be public, in the sense that they ought to revolve around or incorporate a public issue or the question of involvement in public life, but also personal -sharing and probing for how people related to these issues in terms of how they saw themselves, and what mattered to them. To achieve this, organisers or leaders who initiated one-to-one conversations were often encouraged to model this dimension of personal openness for others -sharing personal pain, formative experiences or pressing struggles, and tying these in with collective issues, before soliciting a similar account from the other party. Partly, one-to-one relationships were meant to identify and draw out concrete, shared interests. Partly, however, they worked as forms of 'phatic labour' in Julia Elyachar's (2010) sense -as cultivating forms of sociality and attachment that became valuable in their own right, and which could serve as 'channels' for action.
It's worth pausing here to note that on the surface, many of the concepts used by organisers look very much like reflections of a hegemonic liberal-individualist world view, where self-possessed individuals identify and act rationally on their interests. This world view has been subject to plenty of anthropological critiqueboth for its false universalism and for its capacity to disguise and normalise parochial interests. Perhaps unsurprisingly, then, when organisers taught this model of political change, it was met with plenty of vocal resistance from a range of perspectives. People from religious backgrounds often resisted the anti-idealist stance that tied change to power, and dismissed abstract, principled commitments. University students and those with activist backgrounds were quick to contest the reduction of power down to 'organised money' and 'organised people'what about ideology? Media? The disciplinary apparatus of the state? Dissenters of various stripes frequently challenged the alleged centrality of self-interest in everyday life, and its framing as the foundation of political engagement. Other foundations, such as altruism, empathy and principled conviction were offered up as more persuasive and less socially corrosive alternatives. The emphasis put on one-to-one conversations was less controversial, but even this was sometimes singled out for privileging the building of relations between individuals, as opposed to the sorts of more collective understandings or sentiments that might be cultivatedfor instance -through church services or in a classroom.
Once every few months, organisers would run training sessions for local alliances, or for clusters of alliances together, teaching the rudiments of organising to several dozen leaders from across different institutions. Once a year, those leaders who had been more involved were invited to a six-day residential training to learn the tenets of organising in greater depth. Faced with the range of possible misgivings sketched above, it was very rare for a training session to proceed without an expression of vocal scepticism. Often, the first objection would open the floodgates. Strangely, in response to these, organisers would frequently respond in ways which affirmed but did not accept these critiques, before then pushing the task of answering back to the group ('That's a good point. What do others think?'). Meanwhile, although organisers themselves were well aware of this range of objections, and although in general organisers stressed the importance of learning pragmatically from experiences -following virtually everything they did with rounds of collective 'evaluation' -they made little effort to alter their teaching in response to these misgivings -whether over the short or long term. Finally, in a range of conversations following training sessions, I would often hear from those who had raised objections that these concerns were really quite profound and personally troubling. Nonetheless, very few of the 'leaders' whom I witnessed raising vocal objections in training ultimately abandoned their (voluntary) involvement with organising -while several gradually became vocal champions of organising and its principles themselves. In short, the theory of power elaborated by organisers generated plenty of objection, but simultaneously provided a collective register through which diverse leaders and communities continued to work, even despite misgivings. To understand these puzzling dynamics, we need to move on to . . .

Organising 102: Reweaving
Analysing the 'semiotics of the public/private distinction' Susan Gal (2002Gal ( , 2005 contrasts two ways in which the categories public/private can be understood. The first is as a clear-cut distinction, between two non-overlapping domains. The second is 'fractally' as never-complete distinctions, where each domain contains something of the other, recursively. To illustrate this latter understanding she gives the example of the American home, characterised as a quintessentially private space. Within the home, however, the living room is imagined as its public aspect -a public space within the private. And then again, within interactions in the living room, there are various conversational signs, from tone to posture, that will reinscribe a sense of privacy within public exchanges, such as when conversational partners turn their bodies in toward one another, or when voices drop in tone. Gal argues that the first understanding is hegemonic -it is the dominant way in which liberal societies imagine the categories of public and private -and it is also symbolic -in that these categories derive their meaning on the basis of a Saussurean logic of associations and oppositions. Meanwhile, the second, fractal, understanding is inscribed 'through indexical signs [that] are often hard to notice even for the social actors who use and impose them' (Gal, 2002: 79). Indexical signs operate like gestures -they 'are always relative: dependent for part of their referential meaning on the interactional context in which they are used ' (2002: 80). In other words, they trace relationships, and change their meaning when these underlying relationships change. For Gal, the public/private divide is indexical because the applicability of these categories depends on the framing: what may be private, such as a living room, or a turn inward towards others, in one frame, may be public in another. Importantly, then, these different understandings emerge from different registers -different modes of expression and interpretation -which can be applied to a common set of signs or situations.
Gal frames her analysis as a critique of the supposedly universal categories of liberalism. Here, the categories of public and private are (through the symbolic register) naturalised as existent and clear cut, when in fact they are (through the indexical register) constantly being reinscribed in inconsistent ways, that give the lie to any claim that they are natural or settled. Yet it is also possible to read this analysis as an illustration of liberalism's social dynamism -as an argument that while discrete, stable domains of public and private may not exist, liberal ideology is nonetheless able to flexibly evoke these concepts to channel understanding and action across a range of contexts (Anderson, 2016).
Here's the catch: community organisers themselves, especially among themselves or around more seasoned leaders, will sometimes explicitly describe organising itself as 'fractal'. To understand how this relates to their theory of powerand the vocabulary used to outline this -it helps to start with a simple example. Although all those involved with organising are referred to as 'leaders', organisers will also frequently say that 'to be a leader, you must have followers'. This statement serves a simple, practical purpose: it builds people power by pushing leaders to bring others along. But it also turns 'leader' into a recursive category. A leader may have followers who are themselves leaders to others, and so on. In fact, organisers will frequently refer to some leaders also as organisers. This links to a broader claim, that organisers do little more than practice what they preach -they too work to expand alliances, support processes of listening and action, and cultivate leadership, using the same tools and philosophies they teach to others. Meanwhile, as leaders deepen their involvement they are often expected to grow from campaigning on specific issues, to being invested in strengthening their own institution's capacity to pursue change, in general, to working to strengthen this same capacity across their alliance as a whole -effectively taking on the same tasks as the paid organisers. In other words, 'organiser' is another recursive category. In turn, who qualifies as an organiser, leader or follower often depends on who is positioned as such, in any given situation -giving these categories an indexical dimension. This recursiveness and reliance on indexicality, it turns out, applies to much of the conceptual apparatus of organising -including understandings of power.
This recursive notion of power becomes evident in how organisers pursue change. The notion of relational power extends not only to the community building power but also to their targets -those to whom demands for change are addressed. Organisers talk about seeking three things from opponents: 'relationship', 'recognition' and 'reaction'. Communities take action -whether disruptive protests or more formal negotiations -in order to provoke an understanding of their issues and interests, but also, crucially, to make opponents recognise that their interests are intertwined with those of others (even if, perhaps, this is partly the case because these others are barricading your building, or embarrassing you publicly). In turn, opponents should come to see that realising their own interests requires forging collaborative relationships. In effect, these goals bring the targets of power into the fold of relational power, meaning success is imagined not as a matter of acting against others, but as a matter of acting through or with others. This leads to the intentional pursuit of what others might term complicity. Two contrasting examples help illustrate this.
When I first started my fieldwork, I witnessed the conclusion of a campaign to get London City Airport, one of the largest local employers in the London borough of Newham, to pay the living wage -an independently calculated standard, meaningfully above the national minimum wage. The campaign was led by a local school, which identified low pay as a significant issue in their community. They singled out City Airport as a major local employer, but their initial requests to meet and discuss the issue went unanswered. Guided by Ryan, their organiser, the campaigners first worked to identify who, at the airport, held the power to respond to their demands and, second, what their targets' self-interest might be -effectively treating them as potential leaders. Recognising airport executives' concern with reputation, they hit on the idea of disrupting a community day run by the airport, by coming and having the school choir sing a song about low pay -composed by their music teacher with the aid of a parent, who was a modestly well-known rapper. They again wrote to airport executives to inform them of their plan, and this time received a quick reply, imploring them not to sing at the event -and instead inviting them for a meeting. Within a year, City Airport had become an accredited living wage employer. During this process, the school shifted their orientation towards the airport, offering them support in the accreditation process and celebrating their commitment. As a result, executives from City Airport themselves eventually became champions of the living wage movement -lobbying other businesses to become accredited themselves.
It's not hard to imagine how to formulate a scholarly critique of this approach to creating change. A large body of work exists on the workings of corporate social responsibility programmes, which trace the ways in which corporate concessions toward local demands almost always form a part of broader legitimating processes which ultimately serve to sustain and deepen, rather than rectify inequality (Dolan and Rajak, 2016). Interestingly, however, in conversations with the Living Wage Foundation team -who are themselves a spin-off organisation from CUK, created as the Living Wage movement grew in scale -I was told that even when factoring in reputational gains, benefits from better employee retention and so on, there wasn't always a 'business case' to be made for the living wage, and that many of the major employers who paid it did so with an awareness that it would reduce their profitability. Nonetheless, we may well object that even if benefits do not accrue to (all) individual employers, such gestures nonetheless maintain a hegemonic order where corporations are considered as legitimate arbiters of pay, working conditions, wellbeing and ultimately survival -however this power is usedand where alternative economic arrangements become more difficult to imagine. These charges are certainly harder to answer.
Not long after City Airport had acceded to the demand to pay the living wage, I accompanied Ryan to a meeting with August, an academic working at a university in central London. Several universities are members of CUK, and organisers work with both students and academics interested in organising. August was angry about the ways in which care workers were treated and valued in the UK. Her own research had highlighted some of the difficult, intangible skills that went into providing good care -but which resisted easy valuation and often escaped public notice -as well as the exploitative conditions under which many carers were employed. Her passion for the issue was evident, and she talked pointedly about how society as a whole needed to rethink the value of care and care workers. Ryan began to quiz her around the possibilities for taking action -who were the prominent employers? Were there care workers August knew who would make good leaders? What change did they want to see -would it be another living wage campaign, or would it be around another dimension of employment? August made her own suggestion -she wanted to start by organising an event, highlighting care worker voices. She could get some of those she had worked with to share personal stories capturing the difficulty, poor conditions and low rewards of the profession, alongside stories which highlighted the vital roles they played -publicly staging the absurdity of the hegemonic contradiction between the collective reliance on care work, and its systemic devaluation. With the right audience, she argued, something like this could play a role in shifting social perceptions. Ryan made a lateral attempt to object, asking if she had a specific employer she wanted to target, and August responded more pointedly that larger scale change was needed, that went beyond any particular employer. A few months later, the event went ahead and, after that, little else.
These two stories illustrate differing conceptions of power, which endeavour to identify and mediate different sorts of relations. The theory of power elaborated by CUK organisers may seem slippery, tautological, perhaps even self-contradictory: power is the ability to act, but to act, you need power; relational power is built by appealing to self-interest, but self-interest is constituted through relationships; to secure change you must forge a relationship with your opponents, but doing so involves allowing this relationship to reshape your interests, imperilling the original impetus for action -and so on. Yet must theory be coherent? In describing the citizenship practices of black girls in Detroit, Cox (2015) draws on the queer theorist Jack Halberstam's notion of 'low theory'. In defining the concept, Halberstam elaborates: Low theory is a model of thinking that I extract from Stuart Hall's famous notion that theory is not an end unto itself but 'a detour en route to something else' [. . .] we can think about low theory as a mode of accessibility, but we might also think about it as a kind of theoretical model that flies below the radar, that is assembled from eccentric texts and examples and that refuses to confirm the hierarchies of knowing that maintain the high in high theory. (2011: 16) Crucially, for Halberstam (and Cox) low theory can be elaborated through slippages and silences, forgetting and unknowing -it need not be coherent or strive for mastery, so long as it leads somewhere worthwhile. More specifically, I would suggest that the form of low theory elaborated by CUK organisers can be thought of as a form of 'minor theory'. Exploring the work of Franz Kafka, a Czech Jew living in Prague under the Austro-Hungarian empire, Deleuze and Guattari (1986) develop the notion of a 'minor literature' -which they also link to other authors, such as Samuel Beckett. Writing both within and against a literary tradition that resisted the expression of what he hoped to say, Kafka wrote in German in a way that seemed to subvert it from within -making displacement, alienation and tension features of a literature that would resist them. From this, geographer Cindi Katz (1996) develops a notion of 'minor theory' -understood as a form of theory that, in a particular context, sits within a more dominant one and transforms it from within. Katz's examples are feminism (as the minor) and Marxism (as the major), within her own academic circle, and, quoting Gayatri Spivak she talks about the need to conceive of a 'marxism squeezed through the pores of feminism' (Katz, 1996: 489). She adds that: 'Constituting the minor is not about naming but about consciously working in a vocabulary in which one is not at home -where one has become 'deterritorialised'but where one works that deterritorialisation to its limits, forcing it to express something different' (Katz, 1996: 489). In this regard, the linguistic anthropology notion of 'register' is very similar to that of minor theory. Register too, is a body of communicative elements and meta-communicative conventions, which sits within but is not wholly determined by dominant frameworks of meaning and practice (Agha, 2008). Yet while the concept of 'minor theory' emphasises the potential for displacement, the idea of register reminds us how this this potential is more often cultivated and sustained not through the heroic striving of individual authors or theorists but through collective effort and negotiation.
Building on these ideas, I would argue that for CUK organisers, what matters is not the coherence or encompassing scope of theory, but, first, the practical capacity of theory to lead somewhere worthwhile and, second, the capacity of theory to transform a given world from within. First, the recursive nature of the categories used by organisers gives them an unstable, ambiguous quality, but this same quality allows them to be continually re-specified in novel ways. Categories such as leaders and followers, relational power or self-interest are deployed across a range of contexts to index different things: the need for church parishioners to follow established leaders in one instance and to act as leaders in their own right in another; the need for community alliances to build relational power to challenge opponents in one moment, and the need to incorporate opponents into webs of relations in the next. At the same time, the categories through which organisers theorise power are far from arbitrary. Products of an established, gradually evolving tradition, these categories seem to be ones which invoke seemingly stable, hegemonic understandings but which retain a capacity to accommodate a range of different indexical referentssituated understandings, experiences, relationships and feelings -when deployed across diverse contexts. The result is a vocabulary that is at once familiar, capable of travelling widely (if never without some friction) and flexible, capable of taking on distinctive meaning and resonance in particular contexts. The pairing of familiarity and flexibility allows people occupying different situated positions, or driven by different beliefs to maintain a sense of shared understanding and purpose -providing a channel for acting collectively, across difference. Part of the reason why leaders stuck with organising even when they disagreed with its theoretical underpinnings was -as several leaders told me, explicitly -because it got things done, often through cultivating unexpected coalitions.
Second, however, this theory of power doesn't simply work to hold diverse possibilities together, but mutually transforms these different possibilities -including those associated with dominant orders -from within. As noted, the categories organisers draw on to set out their theory of power are often drawn from the language of a hegemonic world view, while the methods of organising often entail forms of complicity. And yet, by staging these categories and methods in recursive relation to one another, organisers allow them to inflect and displace one another. Thus, for instance, self-interest is simultaneously presented as composed of relationships (among other elements) and as existing prior to them -as the grounds for forming new relations. In turn, this means that the process of building relationships, through identifying and speaking to others' self-interest, is simultaneously also a process of displacing and reshaping self-interest. Here self-interest is imagined neither as a pre-given individualised possession, nor as wholly relational to an extent where individuals might dissolve into dividuals (Strathern, 1988) -but as situated in dynamic tension between these possibilities, capable of drawing individuals out some distance beyond themselves (for a similar account of self-interest, see du Gay, 2007). Rather than staging individual self-interest and relational-selfhood as distinct, each is 'squeezed through the pores' of the other. Another part of the reason why leaders stuck with organising despite initial misgivings, then, was that these misgivings, and their understandings of organising theory, were displaced and transformed over time.
Tracing this recursive, mutually inflecting dynamic across the whole range of concepts used by organisers in thinking about power, enables us to recognise how organisers attempt to theorise power in a 'vocabulary in which one is not at home'. This creates a tense situation where the possibilities and limits associated with how organisers conceive power remain inextricably bound together -but dynamically so, as they continue to push on and run up against the limits of dominant understandings in new ways. Here, the contrast to the implicit theory of power sketched by August in the second example above becomes especially illustrative. August imagines transformation at the transcendent level of society, writ large -implying that anything at a lesser scale remains uncomfortably complicit in upholding existing inequities or, at least, is insufficiently transformative. It's at this abstract, encompassing scale, that it becomes possible for her to imagine the possibility of resolving contradictions -such as that between the widespread reliance on care labour and its systemic devaluation, a problem that operates at the level of regimes of value (Appadurai, 1986). Here there is a link between the critical imaginary, the sorts of expansive links that it traces and the sorts of transformations which come to be seen as possible or worthwhile. The contrast to this approach is best illustrated by a final pair of concepts that organisers often used to talk about powerthe tension between 'the world as it is' and 'the world as it should be'. For organisers, social change was about continuously occupying the space between these two, and so about deliberately doubling back to get under the skin of familiar concepts and powerful actors -about holding the given and possible in recursive tension -rather than racing towards utopian horizons. This gave their claims a scope and efficacy that developed in step with their complicity and fragility -a project organisers described as 'reweaving' the social.

Conclusion: The pragmatics of power
Power has become a fundamental part of the anthropological vocabulary. Anthropology, however, teaches us to be wary of over-familiarity. When certain ideas come to be taken for granted, we risk losing sight of their historical and cultural contingency, and with this, the possibility of imagining the world otherwise, or of understanding those who already do so. I began this article by tracing an enduring dilemma within anthropology between the need for 'theories of motion' that are able to connect up and mediate between diverse standpoints, and ways in which such theories inescapably displace other perspectives -often including those of subjects themselves. In this article, I have attempted to affirm both the need for theories of motion and that these theories inevitably do have a displacing effect. Yet I have also attempted to resist taking this for granted as a dilemma, per se. Against the grain of ongoing debates between 'dark anthropology' and various perspectives which call for centring everyday understandings, hopes and ontologies -which continue to restage a disciplinary dilemma -I have attempted to follow the lead of CUK's community organisers and look at how the tension between epistemologies of intimacy and estrangement might instead be taken as a generative force for social change.
Even for seasoned scholars, theory can often make for tough going. The language of theory often operates in its own register, full of sui generis terminology and self-referential debates. In other words, academic theory too, carves out its own social location, its own relationship to dominant regimes of meaning. As anyone who has been disoriented or inspired by a theoretical text can attest, such theory inevitably effects its own forms of displacement upon established understandings. What CUK's organisers remind us is that displacement does little as an end in and of itself. Instead, what is needed is a critical interrogation of what is being displaced, and what spaces, in turn, are being opened up. As generations of scholars focusing on the voices and struggles of marginalised people have reminded us, we need theories of motion to make sense of the world and those within it. Perhaps even more importantly, however, if theory is to hold the sort of transformative promise academics so often attribute to it, we must first ask where particular theories of motion are likely to lead.