Which Markets, Whose Rationality? Markets as Polyvalent Political Devices

This article explicates and critiques an understanding of markets that is dominant in much contemporary political theory. Drawing on the insights of new materialist economic sociology, it argues that the divide between “the political” and “the market” that grounds many recent analyses cannot ultimately be sustained. Conceptualizing markets not as abstract, impersonal mechanisms but as polyvalent assemblages, the paper develops a view of markets as material devices subject to a wide variety of political inflections and deployments. This understanding is then used to clarify some of the disputes between market-friendly neo-republican theorists and their critics. The article argues that markets are best conceptualized as political institutions (rather than as alternatives to politics). It commends an approach to political theorizing that moves beyond “pro-” and “antimarket” positions, focusing instead on the material details of market configurations and their consequences for agency and social power.


INTRODUCTION
N eoliberalism has become an object of sustained critical attention in recent years.Articulated in many variants and from multiple theoretical vantage points, a consensus has been emerging-and gaining significant steam since the financial crisis of 2007-that intensifying inequality and the fateful unraveling of the social fabric attending it are due in no small part to the growing dominance of "the market" in all aspects of our lives.Critiques of contemporary neoliberalism-understood as relentless "marketization" (Brown 2015) or "frictionless market rule" (Peck 2010)-now abound in the literature.Such criticisms echo Karl Polanyi's famous critique of economic liberalism's aspiration to "[run] society as an adjunct to the market" (Polanyi 2001, 60).Markets, in this view, are abstract and impersonal economic mechanisms, imposing their coercive, homogenizing logics on qualitatively distinct entities and submerging them all in the icy waters of monetary equivalence.At the limit, this understanding points to the possibility of a dystopian future in which all domains of life are "marketized" and we are "only and everywhere" market subjects, no longer capable of revolutionary subjectivity nor even of thinking in a "distinctly political way" (Brown 2015, 31, 39).
As noted, these lines of criticism are indebted to the framework developed in Karl Polanyi's masterful account of the emergence of economic liberalism, The Great Transformation.This seminal work, which inaugurated a whole tradition of understanding the market as necessarily "embedded" in society, has been of salutary and lasting influence in alerting us to the ravages and dislocations attendant upon attempts to allow "free markets" in land, labor, and money to eclipse other social values and practices.By showing up the inhumane consequences of the version of economic liberalism imposed on English society at the dawn of the industrial revolution, Polanyi provided a forceful argument for the need to ensure that market activity is subordinated to social needs.His insight that society moves "to protect itself" when threatened with the brutal commodification of all aspects of life is of lasting significance for elucidating contemporary political dynamics (Polanyi 2001, 3). 1espite its immense influence and usefulness, however, I argue that Polanyi's approach has also bequeathed to us a problematic theoretical framing whose shortcomings hinder our efforts to grapple with contemporary problems of political economy.His analysis was undoubtedly salutary in highlighting the problematic features of economic liberalism's violent imposition of pitiless monetary logics on societies organized according to alternative moral economies.This line of criticism-of the devastating consequences of attempting to reduce all social relationships to commodified exchange-is of lasting relevance to contemporary concerns over rampant economic inequality and the cruel responsibilization attending the "great risk shift" of recent decades (see Hacker 2006).In developing this trenchant analysis, however, Polanyi also inadvertently reinforced some of economic liberalism's central claims regarding the fundamental incompatibility between the logic of the market and that of society, morality, and politics.In doing so, he helped to erect an opposition-"the market" versus "society"-that continues to shape debates over capitalism, liberalism, and justice.
This paper argues that the framework derived from the Polanyian tradition continues to structure our critical imaginations in ways that are severely limiting.To reiterate, I wholeheartedly affirm Polanyi's basic insight that the violent subordination of vulnerable populations to monetary logics, supported by the coercive arm of the state, was and is a gross violation of justice and a source of immense suffering, immiseration, and domination.The persuasiveness of this analysis, however, should not cause us to overlook the problematic features of the theoretical framework in which it is embedded.As a result of this framing, I argue, Polanyi's (correct) insight that society has needs that are independent of monetary considerations has been expanded into the more general conviction that markets as such inherently possess dehumanizing and/or antipolitical logics and should thus be opposed in the name of alternative logics that are more humane and democratic.In short, the market has itself become an object of political opposition, calling for critique and resistance from anyone concerned with defending freedom, equality, and humanity as choiceworthy political values.Although Polanyi's normative judgments were surely sound, the social-theoretic framework in which he articulated them leaves us with an inadequate grasp of the nature of market forms (and thus with an impoverished understanding of contemporary politics).
It is this framing that this paper seeks to explicate and contest.Although it is surely true that the forms of commodification and marketization that Polanyi so ably critiqued were unjust and deserving of political opposition, we should resist the generalization of this observation into the broader claim that emancipatory politics entails opposition to market logics as such.Indeed, the very notion of a "market logic," I shall argue, is confused to begin with.As will be demonstrated below, insightful work in economic sociology has questioned the coherence of the view that anything like a singular market logic can be identified.In particular, new materialist economic sociology has developed a compelling view of markets as sociotechnical assemblages subject to a wide range of applications and political inflections.In this view, markets are best seen as devices, "sociotechnical agencements,"2 that recompose and redistribute forms of agency and capacities for calculation and exchange.The disciplinary and classbiased imposition of particular market forms, such as those associated with the "stark utopia" of economic liberalism analyzed by Polanyi, are one-particularly worrisome and unjust-form in which markets can be politically constituted and deployed.However, objecting to this particular form of market imposition leaves open the question of what other forms of market organization might be available, what political projects they might play a role in advancing, and which political values they might aid in securing.
As I argue below, the framing derived from the actornetwork theory (ANT) research tradition questions the rigid binary between "market" and "political" logics.Drawing on the work of Michel Callon, Donald Mac-Kenzie, and Timothy Mitchell, I suggest that the notion of a putatively "free" market-unencumbered by politics and existing as a purified domain of technical calculation-is itself an ideological legacy of the disciplinary form of market liberalism critiqued by Polanyi. 3  To oppose the market in the name of politics (or society) is to capitulate to the fantasy of the possibility of a free and apolitical space of purely technical calculation and exchange (see Mitchell 2002, 3-9 and chap. 3).It is this very framing that the ANT tradition exposes as fallacious.The ambition of economic liberalism was, in fact, to impose a specific class-biased form of domination on a politically vulnerable segment of society and to mask this coercive political project with the veneer of apolitical, technical rationality and timeless economic laws. 4f, however, the very notion of a pure domain of technical calculation, unencumbered by politics, is itself an ideological obfuscation, then it is a mistake to allow the framing of market versus political logics to ground one's resistance to such disciplinary and class-biased projects.Recognizing that markets are infused with politics and have a variety of rationalities and logics potentially associated with them raises the question of the role that market arrangements-considered not as apolitical mechanisms but as sociotechnical agencements-might play in an emancipatory politics.Such a politics would remain opposed to the form of market imposition that the Polanyian critique exposed and criticized.However, it would maintain that alternative market logics are available and might provide useful resources for those who are intent on opposing the unjust sociopolitical dynamics associated with contemporary neoliberalism.Markets, seen as polyvalent political tools, prove to be compatible with a much wider range of political projects and commitments than contemporary political discourse allows.
The argument proceeds in three sections.First, I explicate Polanyi's framework through an engagement with the insightful work of Block and Somers (2014).I show that even their sophisticated interpretation of Polanyi does not altogether escape the theoretical confusions noted above.I then introduce an alternative theoretical framework derived from actor-network theory, which is capable of giving "the Polanyian intuitionthat economics is a political institution and a social project (Polanyi 2001)-the theoretical scheme it was lacking" (Cochoy, Giraudeau, and McFall 2010).Having presented this alternative paradigm, I then illustrate its usefulness by showing how it can help to clarify recent debates in political theory between neo-republican advocates of market mechanisms and one of their most trenchant Polanyian critics.

"THE MARKET" AGAINST "THE POLITICAL"
Much criticism of contemporary social and political life involves the claim that we are, either actually or aspirationally, dominated by the market.Wendy Brown has famously associated neoliberalism with the desire to marketize all domains of social life (Brown 2015).Jamie Peck has identified neoliberalism with an aspiration to frictionless market rule (Peck 2010).Wolfgang Streeck has described contemporary capitalism as embodying the dominance of market logics over the incompatible political logic of democracy (Streeck 2014a;2014b).Gerald Davis (2009) has characterized the contemporary political condition of Americans as that of being "managed by the market."In various forms and various guises, the dominance of markets in organizing social and political life is decried as a major source of inequality and exploitation in our contemporary experience.Much of this work draws, explicitly or implicitly, on Polanyi's seminal intervention, which identified the "stark utopia" of economic liberalism as involving the aspiration to "[run] society as an adjunct to the market" (Polanyi 2001, 3, 60).
As commentators have noted, Polanyi's influential account in The Great Transformation seems to contain an unresolved tension (see Block 2003;Block and Somers 2014).Over the course of the book, he frames his argument in divergent (and potentially incompatible) ways.Thus he sometimes seems to claim that running society "as an adjunct to the market" is practically undesirable, as it will lead to the "demolition of society" (Polanyi 2001, 76).In this view, "the market" and "society" have incompatible logics, so disembedding the market has deleterious consequences.At other times, Polanyi seems to suggest that the economy is always embedded in society.In this view, there is no such thing as disembedding the market; even what was called "laissez-faire" was actually "planned" and required ongoing state intervention (146-7).So understood, Polanyi's work seems to claim that disembedding the market is both dangerous (because society cannot bear being subjected to such a self-regulating mechanism) and conceptually impossible (because such a "self-regulating mechanism" is a fiction).It is not clear how these two claims are meant to fit together.
Block and Somers have helpfully explored this ambiguity, arguing that the second formulation, which they term the "always-embedded economy" thesis, is Polanyi's most insightful discovery.In their view, Polanyi's earlier formulation arises "from the Marxist conceptual framework that had been the initial organizing framework for his book" (Block and Somers 2014, 81).Over the course of writing the book, his investigations caused him to move beyond this perspective and develop the always-embedded economy thesis.Because he did not have time to revise the book's argument before publishing, the text retains both perspectives, despite their incompatibility (Block 2003, 284-5;Block and Somers, 2014, 82).
Despite the thoughtfulness of this intervention, I suggest that a similar (though somewhat more subtle) ambiguity remains even in the analysis offered by Block and Somers.This ambiguity is evident in their explication of the always-embedded economy thesis.When this thesis is first introduced, it is said to involve the observation that "even 'free' market economies consist of cultural understandings, shared values, legal rules, and a wide range of governmental actions that make market exchange possible" (Block and Somers 2014, 9).This, they claim, is a fundamental reality that "[e]conomists deny" (9).Thus all claims to "disembed" the economy in the name of the free market are deceptive.Although "market ideologues" may claim that they are "setting the market free from the state," in reality they are merely "re-embedding it in different political, legal, and cultural arrangements, ones that mostly disadvantage the poor and the middle class and advantage wealth and corporate interests" (9; italics in original).
This persuasive view of embeddedness, however, is modified in their subsequent exposition.In their later analysis of "the concept of the always-embedded economy," Block and Somers describe this thesis as entailing "that market societies must construct elaborate rules and institutional structures to limit the individual pursuit of gain or risk degenerating into a Hobbesian war of all against all" (Block and Somers 2014, 94).The next sentence reaffirms this viewpoint: "In order to have the benefits of increased efficiency that are supposed to flow from market competition, these societies must first limit the pursuit of gain by assuring that not everything is for sale to the highest bidder" (94).Notice the subtle shift in meaning that the concept of embeddedness undergoes in this passage.Initially, the embeddedness thesis involved the claim that all market arrangements (including those advocated by free market ideologues) involve different forms of embeddedness.Yet in the second formulation, embeddedness is identified specifically with rules and institutions that limit or constrain profit-seeking activity.This reintroduces a normative component into what was previously a descriptive claim.It is not simply that markets are always embedded (in the sense that they rely on rules and institutions) but that this embeddedness must constrain profit-seeking activity (because the preservation of society requires it).
This second formulation is in clear tension with the first.The first presentation of embeddedness suggests that there is no such thing as "the market."Markets always interact with legal and cultural institutions that give them a particular shape.Thus, as observed above, even "free market ideologues" are actually proposing alternative forms of embeddedness.Yet throughout their analysis, Block and Somers also speak of embeddedness as serving to protect society from "unmediated market forces" (2014, 10) and "unfettered markets" (201).If markets are always embedded, however, such formulations are incoherent.They imply that a disembedded market is a real possibility (albeit an undesirable one).
This ambiguity is a legacy of the Polanyian framing with which contemporary political theory needs to contend.Those drawing on Polanyi, inheriting this unresolved tension, tend to shift between these two different meanings of embeddedness.They sometimes argue that markets cannot, as a factual matter, be disembedded (because they are always governed by rules, institutions, etc.) and thus that the very idea of free markets is incoherent.At other times, however, they suggest that we should be wary of "unmediated market forces" because they are incompatible with the flourishing of society (a claim that relies upon the idea of an essential market logic that the first thesis disavows).Clarifying the political nature of markets, and the meaning of their embeddedness, is thus crucial to gaining clarity on contemporary debates about their role in contemporary politics.

MATERIAL MARKETS AS SOCIOTECHNICAL AGENCEMENTS
Much of the debate over market forms in contemporary life has been conducted in terms of advocacy of or opposition to the free market.In this view, markets naturally operate according to their intrinsic, impersonal logics.Critics suggest that government intervention should "tame" these markets, regulating away their most objectionable features or restraining them to carve out a space for politics.Advocates hold either that market logics spontaneously emerge in the absence of political interference or that political intervention is necessary as a supplement to ensure that markets function according to their true competitive logic.Both approaches, however, participate in the fundamental framing of markets contested in this paper.They suggest that a natural or pure form of the market exists, prior to and outside of political meddling. 5They then envision forms of politics that can either restrain or preserve the natural "logics" possessed by market forms.Yet, as Block and Somers affirm, what is called the free market by both apologists and critics is in fact constituted by a whole host of political choices, conditions, and enabling devices. 6The very designation free market is an attempt to occlude this reality, framing the "neutral" market position as if it were apolitical and technical and then opposing this to an "interventionist" position in which politics contaminates or alters the logic of the market.It is this very notion of an originary, natural market, freed from the constraints of political judgment, that the perspective developed here seeks to contest.
A sophisticated and productive line of thinking about this problem has been pursued in a lineage originating with Michel Callon's 1998 work The Laws of Markets.There and in much subsequent work, Callon develops a view of markets that depicts them not as abstract mechanisms but as "collective calculating devices."7In this view, markets are collective material "devices" that furnish agents with capacities they would otherwise lack.Markets establish equivalences and make possible calculations that enable coordination among agents.In this work, markets are analyzed not as rigid or timeless abstractions but as concrete material "apparatuses" that differentially empower agents, equip them with new capacities, and reshape the world by establishing equivalencies between and among what were previously heterogeneous and qualitatively distinct entities.The upshot of this view, he claims, is that it is "wrong to talk of laws or, worse still, of the law of the market.There exist only temporary, changing laws associated with specific markets" (Callon 1998, 47).This implies, as a subsequent paper clarified, that "more than one 'logic' exists for markets and even for capitalism" (Callon andMuniesa 2005, 1243).
The view he develops avoids both the "naturalizing" tendencies of neoclassical economics and the theories of "social construction" which assume, in essence, that economics is purely ideological, akin to a modern religion.8As Callon states of his initial findings, "we have to be wary of the catch-all that socioeconomics likes to use as a rallying cry: the market is socially constructed.What is under construction is precisely this heterogeneous collective, populated by calculating agencies.Society is not a starting point, a resource, or a frame; it is, along with the market, the temporary outcome of a process in which social scienceseconomics in this case-are the stakeholder" (Callon 1998, 30).This observation runs like a red thread through the subsequent research program inaugurated by the application of ANT to the study of market processes.Markets will no longer be considered as natural entities with rigid and timeless laws, but neither will they be treated as merely fictitious or ideological.Rather, markets will be seen as participating in the ongoing "performation"9 of the world: as sociotechnical agencements that create equivalences, provide the capacity for "collective calculation," and differentially distribute agency, power, and resources (see Caliskan and Callon 2010;Callon and Muniesa 2005).
This research program has been developed during the past two decades by a wide array of scholars with varied and fascinating results.This vast literature is too expansive to be reviewed in detail here. 10Although theoretical disagreements certainly abound, there are several shared convictions that have emerged in this literature and are relevant to the present discussion.Most centrally, studies carried out on markets in their material diversity have led scholars to conclude that thinking in terms of "the market" is inherently inadequate for engaging contemporary political concerns over the growing role of economics, finance, and markets in collective life.Central to the tradition associated with Callon and ANT is a concern with understanding markets not as ideological abstractions but as "sociotechnical agencements" (Caliskan and Callon 2009;2010;MacKenzie 2009).This means that markets are material assemblages composed of humans and nonhumans; they link people, technologies, knowledge, and goods in vast collectives that format the world and create novel linkages among "actants." 11Markets, in this view, are resolutely "material," and whatever calculative capacity they possess is a result of their having been formatted in a particular manner.They are devices that make possible distributed forms of agency, calculation, and exchange (Callon and Muniesa 2005).
The notion of agencement has a complex lineage and lacks a single, clear-cut definition.Caliskan and Callon define the term by stating that "action, including its reflexive dimension which produces meaning, takes place in hybrid collectives (Callon and Law 1995), hereafter called sociotechnical agencements (STAs).Sociotechnical agencements are comprised of human beings (bodies) as well as material, technical and textual devices" (Caliskan and Callon 2010, 9).Though "agencement" has often been translated as "assemblage" in academic literature, MacKenzie points out that this effaces a helpful play on words present in the original French.Agencement literally means "arrangement" and thus emphasizes questions of material and technical configuration.On the other hand, however, the term agencement possesses an obvious etymological link to the notion of "agency" (MacKenzie 2009, 21; see also Callon 2008, 38).An agencement, in other words, is a particular material configuration (a "network") of humans and nonhumans that possesses unique capacities to act in and with the world.Seeing the agencement as the unit of agency helps us to focus on the relationships among actors, competences, technologies, and knowledge that make complex forms of distributed agency possible.Viewing markets as particular types of "sociotechnical agencements" thus directs our attention to all of these dimensions: to the materiality of markets; the way they rely upon combinations of technologies, discourses, and knowledges; and the way they facilitate and embody novel forms of action.This notion suggests that markets, like other forms of sociotechnical life, are more open, polyvalent, and flexible than traditional critical discourse has allowed. 12It also means that their construction, their maintenance, and the capacities with which they equip agents are matters of deep political consequence.It renders the development of any form of "economization," including marketization, intelligible as an open process that is "historical, contingent, and disputable" (Muniesa, Millo, and Callon 2007, 3).Caliskan and Callon (2009;2010) have pursued some of the implications of this way of thinking in their account of marketization.For Caliskan and Callon, economization (of which marketization is a species) represents an open-ended and experimental process rather than the imposition of an abstract logic.Resisting both "formalist" (neoclassical) 13 and "substantive" (Polanyian) readings of the "the economic," these thinkers foreground "processes of economization," highlighting the processes through which things become figured or qualified as economic (Caliskan and Callon 2009). 14There is here no essence of "the economy" or "the market" whose progressive application (or resistance) we could hope to track (or endorse).Rather, their analysis encourages us to ask a different set of questions, attending to the different ways in which domains may be regarded and configured as economized or marketized.There is not a singular logic to the market but a cluster of different practices and technologies through which domains or objects can undergo marketization.This involves a variety of diverse ways of thinking and acting-with "action" here including the mobilization of vast, heterogeneous, sociotechnical collectives.
Thus, as they state early in their path-breaking article on economization, the term is used to denote the processes that constitute the behaviours, organizations, institutions and, more generally, the objects in a particular society which are tentatively and often controversially qualified, by scholars and/or lay people, as "economic."The construction of action (-ization) into the word implies that the economy is an achievement rather than a starting point or a pre-existing reality that can simply be revealed and acted upon.(2009, 370; emphasis added) In making this claim, they emphasize that "one cannot qualify an economic situation without at some point mobilizing a theory that defines what is meant by economy" (Caliskan and Callon 2009, 371).They also clarify that any persuasive attempt to carry this project through will require acknowledging "that the establishment of an economy involves institutional arrangements and material assemblages, without which nothing economic could exist or be sustained" (371).
According to these authors, marketization is a distinct phenomenon best regarded as a subtype of economization.Though markets exhibit a wide diversity of configurations and features, it is suggested that under current conditions markets form a relatively coherent set sharing something like a Wittgensteinian "family resemblance" (Caliskan and Callon 2010, 2-3).Taking this observation as a point of departure, the theorists define markets "as sociotechnical arrangements or assemblages (agencements) which have three characteristics" (3).These characteristics are that they organize the production and circulation of goods, involve technologies and rules, and "delimit and construct a space of confrontation and power struggles" (3).The second characteristic is the particularly relevant one from the perspective developed here: A market is an arrangement of heterogeneous constituents that deploys the following: rules and conventions; technical devices; metrological systems; logistical infrastructures; texts, discourses and narratives (e.g., on the pros and cons of competition); technical and scientific knowledge (including social scientific methods), as well as the competencies and skills embodied in living beings.(3) This discussion challenges much critical and "social constructionist" engagement with markets that tends to treat them as solely the result of economic interests and other "social" factors.Such approaches result, first, in a neglect of the material and technical dimensions of markets (Caliskan and Callon 2010, 4).Second and relatedly, such analyses display a dismissive relation to economic discourse, which is conceived as "nothing but an ideological endeavor or a false science" (5).Additionally, and crucially for the analysis that follows, such social constructionist positions tend to take "society" as a static and preformed entity whose constituent actors already possess the stable identities, capacities, and interests that shape the markets in question.In the ANT view, however, this neglects the extent to which society itself is an ongoing construction.Society does not construct markets; they are coconstructed simultaneously. 15dopting the "performativity" paradigm, Caliskan and Callon (2010) conceive of markets as assemblages, networks of humans and nonhumans that allow for the circulation of products and the facilitation of exchange.
Markets do, indeed, format the world and effect alterations in things and actors, but not by means of the imposition of a predetermined market logic.What logic a market will come to have is a question to be determined by its sociotechnical constitution, not presumed a priori.Differences in equipment, technology, laws, and norms will lead to different types of market arrangements operating according to different laws and entailing different distributive consequences (23-5).Markets can be established in a diversity of ways; there is no paradigmatic instance.
This complication of the typical framing of markets does not by any means suggest that marketization is an unproblematic panacea for solving society's complex problems.There is no doubt that market mechanisms can differentially distribute capacities for calculation and negotiation, resulting in important asymmetries of information and power. 16Marketization, in its responsibilizing vein, can also neglect structural injustice and consequential differentials in access to resources.Indeed, Caliskan and Callon suggest that one benefit of studying marketization from an ANT perspective is the way it draws our attention to the various forms of "equipment" that differentially empower the calculative capacities of agents (2010,12; in general, see Callon and Muniesa 2005 on "collective calculating devices").The upshot of such research, with its attention to the diversity of market arrangements and their inherently political dimensions, is not that markets are uniformly beneficial, efficient, or self-regulating.Rather, it turns our attention to the fact that it is often the "details" and "technicalities" of market design rather than the "macro" questions of state versus market orientations that are of tremendous political import (MacKenzie 2009, 182).This research tradition thus calls into question many of our assumptions regarding the "stable scale" of action (33).It is often the case that relatively minor and technical differences in market design and implementation have enormous consequences for determining which agents are empowered by a given market and which are dominated.
In this view, actors possess differential abilities to participate in markets depending on the "equipment" (technological and cognitive) and resources that they possess and on the overall structure of the market environment with which they interact.This makes the provision of such resources and the refashioning of such structures an eminently political issue (Callon 2008). 17Far from amounting to an unbridled celebration of markets, studies of marketization discover an ambivalent reality: "it is by affirming the autonomy of calculating agencies that markets are able to conceal and to legitimately impose the asymmetries that develop out of the achievement of calculative capacities" (Caliskan and Callon 2010, 13).Thus Caliskan and Callon claim that "studying the shaping of market agencies is promising and fertile.It enables us to grasp and to document the sociotechnical diversity of agencies and of the forms they take, the complexity of their calculative capacities and relations of domination which develop between agencements" (2010,14).Detailed analysis of markets as STAs allows theorists to "renew the analysis of the power struggles at the heart of any market" (12).

POLITICS AND THE MARKET
Donald MacKenzie, in concluding his detailed study of the material aspects of markets, summarizes well the central implication of this rich research tradition: At the heart of the social studies of finance18 is the conviction that treating "the market" as a singular entity is mistaken… If this conviction is correct, and multiple forms of markets with diverse characteristics and substantially different consequences are possible, then politics divided between "pro-market" and "anti-market" (both in the singular) is wholly impoverished, and even a "third way" that seeks simply to position itself between the two is insufficient.(2009,182) 19The construction and maintenance of markets, in this view, is always a political matter: not simply or even primarily because it is a matter of collective concern whether we embrace or reject the market (the more traditional view of market politics) but because there is always negotiation, contestation, and differential valuation-in short, politics-involved in the very establishment or configuration of a market. 20Trapped within a frame that requires us to be "for or against" the market, we have neglected what MacKenzie calls the "subpolitics" of market construction (2009,183). 21ecognizing markets as politically polyvalent devices shows up the vacuity of a politics centered around the "state versus market" binary and inclines us to attend to the more nuanced and complex question of the types of markets we wish to construct and the forms of agency they enable (or foreclose).
To briefly illustrate the significance of this view, consider MacKenzie's analysis of the development of markets in emissions (a crucial element of many attempts to combat climate change).Describing the construction of the US market for SO 2 emissions, he emphasizes the complexity of the "metrological" issues involved (MacKenzie 2009, 18).A matter as seemingly mundane as the method of measuring SO 2 output, for instance, turned out to involve an amalgam of material, technical, and political considerations.The question of which technology to deploy in measuring emissions affected both the willingness of some environmentalists to endorse the trading scheme (as they feared that normal reporting methods would allow corporations to game the system) and the willingness of traders to participate in the market (as, without real-time emissions disclosures, potential information asymmetries would put them in a disadvantageous position).Additionally, it was the presence of the relatively neglected "ratchet" mechanism (which capped total emissions over time and required pro rata reductions from all parties if the cap was exceeded), MacKenzie argues, that ultimately made the market effective (as theoretically optimal methods for avoiding initial overallocation were not politically feasible; 145-48).As this simple example illustrates (and as MacKenzie's subsequent discussion of the European Emissions Trading Scheme for carbon emissions confirms), simple pro-and antimarket positions are inadequate in such circumstances.It is impossible to be meaningfully proor antimarket in such cases until one has a tolerably specific account of the material configuration of the market mechanism(s) in question.Political power, economic interest, scientific knowledge, and technological capacity intermingle in the contestatory process of marketization.
Taking these insights on board offers us a new vantage point from which to evaluate contemporary debates regarding the extent and desirability of the marketization of our lives.Most centrally, the perspective elaborated above highlights the need to reframe the conversation about markets and their logics.The approach developed by ANT scholars instead enjoins us to inquire into the kinds of market arrangements that might be conducive to the pursuit of emancipatory projects.Thus "which markets we have matters, and that is a question not simply of their overall characteristics but of the details of their design, the technological infrastructures that support them, and the way economic agents in them are constructed" (MacKenzie 2009, 182).This does not rule out large-scale considerations of the general features associated with market or market-like approaches.But it suggests that, more often than we generally acknowledge, the devil will be in the details.The question of "what kinds of markets" (i.e., of material design and implementation) will often be at least as important as the issue of whether market mechanisms are appropriate in a given domain.
To demonstrate the usefulness of this approach, I turn to recent debates within political theory about the possible role of market mechanisms in securing republican freedom.I hope to show that the ANT perspective can clarify the ongoing debate between market-friendly republicans and their critics.As will be argued below, republican advocates of market mechanisms tend to adopt a framing that accepts some version of the divide between the economic and the political.This has allowed Polanyi-inspired critiques to suggest that the republican advocacy of market mechanisms as a means of curtailing domination are rooted in a denial of what Polanyi called "the reality of society."The ANT framing developed above allows us to see both sides of this debate as advancing insightful arguments: arguments that are not necessarily incompatible with one another.The ANT framework enables us to endorse the central insights of the Polanyian critique (i.e., that markets are political institutions with no determinate logics and that power is an abiding feature of the world) while also endorsing the possibility that suitably constructed market mechanisms might be beneficial tools that help us to secure the conditions of undominated agency.
Robert Taylor (2013;2017) has articulated a neorepublican vision that endorses markets and "marketlike" mechanisms as effective means of realizing the republican goal of "nondomination" in various spheres.Republicans, Taylor argues, have been far too committed to institutions that exclusively emphasize "voice" in reducing conditions of domination for the most vulnerable.Inattentive to the degree to which state institutions often risk giving rise to their own forms of domination, republican theorists have tended to advocate traditionally "political" mechanisms-those empowering voice and deliberative decision making -for curtailing domination.Against this trend, Taylor encourages republicans, often skeptical of the market, to take a more "celebratory" position toward market arrangements (Taylor 2013, 594).He thus advances what he terms an "economic model" of republicanism (Taylor 2017, vi).
For Taylor, such an approach entails advocating policies that equip agents with the opportunity to "exit" dominating relationships, as opposed to furnishing them with the "voice" to critique or deliberatively reshape them (Taylor 2017, 3).Drawing on the notion of competitive markets developed in neoclassical economics, Taylor suggests that when the conditions for perfect competition are met (or closely approximated), domination is effectively reduced.Taylor shows "how economic competition restrains-and in the limit, eradicates-market power and how such restraint helps us realize 'market freedom,' i.e., freedom as nondomination in the context of economic exchange" (Taylor 2013, 594).For Taylor, suitably resourced exit provides a powerful tool for realizing nondomination that does not carry the risks of creating new forms of state domination associated with attempts to empower bureaucratic, collective bodies to regulate economic or social life.This is no endorsement of laissez-faire capitalism, however.Taylor commends what he calls the "Anglo-Nordic" model in which substantial state provision ensures that exit is a live option for citizens in various arenas where domination might arise (the workplace, the family, and the locality, for instance). 22his entails extensive state provision of such things as basic income guarantees, capitalist "demogrants," mobility vouchers, and antitrust enforcement.The state here has a constructive role to play in securing (or closely approximating) the conditions of true "competition" in various arenas.
Taylor argues that the structure of a competitive market is such that it, like constitutions and the rule of law, can provide a kind of domination-curtailing structure that eliminates possibilities of arbitrary interference.His project thus advocates facilitating different kinds of competitive markets… .What links these domain-specific markets together is a single, powerful idea: namely, that the effective ability to pick and choose among a variety of potential partnerproviders and to exit relationships and reenter the marketplace if and when those relationships prove unsatisfactory is the best way to protect participants from arbitrary power.Extensive, empowered choice in the service of non-domination is the strategy at the heart of this economic model of republicanism.(Taylor 2017, xvii) As mentioned, however, Taylor is clear that market mechanisms of the correct type do not develop automatically.True "[m]arket freedom requires effective competition, and only the state can secure many of its regulatory and institutional preconditions" (Taylor 2013, 594).As noted above, these conditions involve a generous social safety net, guaranteed minimum incomes supports, demogrants, relocation vouchers, etc. Simply introducing so-called competitive arrangements, without suitable attention to the "resourcing" of vulnerable individuals, is highly likely to increase domination (see Taylor 2013, 601).Taylor's claim, however, is that acknowledging the potentially dominationinducing effects of market arrangements is not an argument against markets as such.Rather, it is a reminder that the details of market design matter.
In endorsing market mechanisms, Taylor does not neglect the importance of political voice in establishing conditions of nondomination.He quite explicitly argues that securing opportunities for exit can strengthen voice; those in dominating positions, he argues, are far more likely to be attentive to the voice of the potentially dominated if the latter have a credible threat of exit.Resourcing exit can thus support voice indirectly by empowering those whom the dominant would otherwise ignore with impunity (Taylor 2017, 12).In many cases, he suggests, exit will not need to be exercised, as its very possibility empowers the voices of those contesting relations of domination across multiple spheres of social and political life.Taylor's argument is not that "voice" is inconsequential but that mechanisms focused on securing the option of exit can often more effectively achieve nondomination both by offering direct means of escape from situations of domination and by providing indirectly for the empowerment of vulnerable voices through the reserved threat this exit option creates.
There is much that is persuasive in Taylor's analysis.Drawing from the ANT perspective articulated above, however, I question Taylor's framing of his position as an economic view of republicanism in opposition to a traditionally political one.I do not mean to split hairs in criticizing this formulation.Taylor understands himself to be highlighting the benefits of a way of thinking associated with the discipline of economics for designing institutions that lead to reductions in domination for society's most vulnerable. 23This is, he clearly affirms, an eminently political goal (Taylor 2017, vii).Nevertheless, I worry that his deployment of the "economic versus political" dichotomy in elaborating his position diminishes the persuasiveness (and distorts the central insight) of his project.
This problem is highlighted by Steven Klein's Polanyi-inspired critique of the approaches of market-friendly republican theorists (including Taylor), which holds that their perspective is insufficiently attuned to the necessity of contestatory politics.Klein describes such projects as attempts to preclude the need for political deliberation and contestation by relying instead on "quasi-natural" background structures such as markets and constitutions (Klein 2017, 854).Klein persuasively shows that the valorization of "neutral" background structures that disperse social power risks occluding the necessary role that contestatory (and class-specific) political movements and institutions play in consolidating the political power necessary to defend vulnerable classes from predation by the rich (857-61).Advocates of market mechanisms, he suggests, neglect the fact that market arrangements will necessarily "empower some classes at the expense of others" (861).By failing to account for the necessity of "collective agents" that can resist domination in the economy, they risk undermining the very institutions that are essential for equalizing market power (861).
Klein's insightful intervention is concerned primarily with the problematic "social theory" that he takes to undergird neo-republican endorsements of market mechanisms.Klein's concerns involve the claim that republican advocates of market mechanisms take the ideal of "fully dispersing social power" as a primary desideratum and thus assign political agency a "residual" role in correcting for market distortions and failures (Klein 2017, 855-6).This frame sees markets as neutral, "agentless," and apolitical institutions that occasionally require, as a "second-best" or "residual" alternative, political "intervention" from outside to correct failures or distortions .This leads to a neglect of the necessary role that "collective agents" play in rectifying imbalances in economic power.Klein explicitly draws on Polanyi to develop an alternative perspective that sees markets as "real, historical institutions" and foregrounds social power as an ineliminable feature of market politics (853).This perspective, he explains, highlights the fact that attempts at marketization will spark social resistance that demands "the (re)organization of economic relationships on the basis of nonmarket, political concerns" (859).
Klein's analysis is insightful, and I am deeply sympathetic to its overall thrust.Certainly, any vision of market coordination that emphasizes markets as impersonal mechanisms with abstract logics insulated from politics is inimical to the perspective I have been articulating and defending in this paper.Insofar as market-oriented neo-republicans rely on such a framing, I wholly endorse Klein's critical corrective to their views.However, the ANT analysis of markets developed above allows one to share Klein's basic worry (about an advocacy of neutral market mechanisms that leads to an eclipse of politics or "the social") without reproducing the problematic framing of markets as inherently antipolitical (as Klein's reading of Polanyi risks doing).
Klein's intervention critiques Pettit as belonging to the "formal" school of economic thinking, whereas Klein endorses Polanyi's "substantivist" position.The ANT perspective, as noted above, explicitly questions the adequacy of this dichotomy (Caliskan and Callon 2009, 373-8).The ANT perspective allows us to endorse Klein's view that "there can be no market outcomes as such" (2017, 857), without accepting his further claim that the aspiration to curtail domination through market mechanisms necessarily entails an "apolitical" and "agentless" conception of their nature (a view that seems to rely implicitly upon the notion of a "market as such" that his argument has just disavowed).In other words, Klein's critique seems to reproduce the Polanyian tension, combining the (correct) claim that social power and contestation are ineliminable features of society with the (in my view, mistaken) claim that an endorsement of market mechanisms as political means for distributing power and agency cannot be squared with this reality.
Thus, although I agree with Klein's insightful argument that problems in the neo-republican framing derive from issues with the social theory in which they are embedded, I believe that the ANT perspective on markets-which sees them as political mechanisms implicated in the creation and distribution of agentic capacities and social power-provides the most satisfactory frame in which to theorize such questions. 24rom the perspective of the ANT frame developed here, Klein's argument is best understood not as showing the necessity of elevating politics from a "residual" to a "primary" place in opposition to a "neutral" market but as highlighting the necessarily political nature of any market agencement and the essential role of politics in constituting and sustaining it.The ANT perspective suggests that there is no inherent incompatibility between a belief in contestatory political institutions (or movements) and commitment to market mechanisms as a means of achieving widely shared social and political goals. 25The fact that contestatory political movements of some kind are required to secure the conditions of social solidarity by no means indicates that market mechanisms should not play a significant role in achieving the solidaristic outcomes such a democratic politics will (hopefully) want to pursue. 26 Thus, I suggest, the ANT perspective articulated above makes it possible to defend both Klein's and Taylor's most perspicuous insights.It is evident, despite some terminological confusion, that Taylor's economic model is itself eminently political.It not only, as he clearly states, employs economic mechanisms (i.e., markets and market-like arrangements) in the service of the "distinctively political goal" of nondomination (Taylor 2017, vii).It also requires significant and sustained political activity to facilitate and maintain the kind of liberating competitive conditions that the model seeks to secure. 27Taylor's model should not be understood as carving out a space of economic mechanisms that are isolated from political meddling; rather, it demonstrates the inextricably political nature of the very constitution and maintenance of market mechanisms themselves.Taylor's interventions show that market arrangements are political arrangements and that they can be deployed in the service of a wide variety of political projects and be compatible with a wide range of political orientations and commitments.There is thus, I suggest, no necessary connection between endorsing markets as political mechanisms and the belief that they "[stand] above political conflict" or are "depoliticized" institutions (Klein 2017, 854).
Truly empowering markets require political intervention, according to Taylor, as "exit costs are not brute facts of nature, outside our control; rather, their level is partly determined by public policy, and legal and policy reforms can therefore potentially lower them" (2017, 15).Liberating markets are not "natural" or "free," but they can be produced and sustained by political initiative.Taylor argues that, in a wide range of cases, political intervention to resource exit and facilitate the creation of competitive market structures reduces domination more fully than addressing the same structures of domination through traditional channels of direct state supervision and bureaucratic direction.Thus market mechanisms are by definition political, and their implementation requires attention to issues of agency and social power.Though Taylor articulates the goal of competition as that of "purging power from economic relations" (49), it is evident that the condition so described can only be realized through a wide variety of political interventions that rebalance social power by providing genuine alternatives to dominating relationships.Taylor can describe the space of competitive markets as free from power only because he distinguishes sharply between the market itself (the arena in which voluntary exchanges are made) and its various institutional supports and prerequisites (which are figured as preconditions of the market rather than crucial components of its configuration).
However, it is not clear that this distinction can really be maintained.Taylor argues with respect to labor markets, for example, that interventions creating universal exit opportunities will realize (or at least approximate) the conditions of perfect competition.In such conditions, both buyers and sellers of labor are "pricetakers" (Taylor 2013, 596).Laborers in such a situation receive the "competitive wage," which is equal to the marginal revenue product of labor (596).Stating the problem in this way creates the illusion that this wage is a static fact of the market, which "external" political interventions facilitating exit can help us to realize. 28et, in reality, things are not so simple.According to this model, the equilibrium wage is determined by the intersection of the supply and demand curves for labor.But it is hard to imagine that some of the more radical interventions Taylor mentions-such as providing a guaranteed basic income-would not shift the supply curve of labor leftward.Such a shift in the supply curve for labor would, ceteris paribus, increase its equilibrium price.Thus the competitive price for labor is itself an artifact of political choices.No individual market actor can manipulate the equilibrium wage, but its actual level is not an economic fact insulated from politics.The structure of the labor market has an ineliminably political character. 29nterestingly, this observation also complicates Klein's critique of basic minimum income policies on the grounds that they focus merely on increasing the "individual bargaining power" of labor market participants (Klein 2017, 861).In fact, as just noted, such a policy would affect the entire structure of the labor market.Thus it might better be thought of as an indirect 25 Indeed, in some Nordic social democracies, it is precisely contestatory bodies like unions that have, after some creative recomposition, managed to represent the interests of workers and the "excluded" while working with rather than against the grain of liberalization in the labor market (see Thelen 2014).  2This is also true of the parallel case.Pettit, for instance, is clear that constitutionalism as a means of securing nondomination requires a "contestatory citizenry" to ensure that democratic politics continues to track the people's interests (see Pettit 2012). 27Taylor is clear about this aspect of his project: "Indeed, the very political framework that makes the economic model possible can only be established through the exercise of voice at the national level: for instance, the mobility vouchers I have relied upon to make my case have to be a product of both political entrepreneurship and coalitionbuilding in national politics" (Taylor 2017, 91).
form of collective economic governance: one that is (as Klein explicitly advocates) "inclusive."The universal scope of a guaranteed minimum income means that it would have significant structural effects.Indeed, it seems clear that the presence of a basic income to fall back on would significantly increase the bargaining power of all workers (and especially of those at the lower end of the income distribution). 30Advocating such a mechanism by no means implies the belief that competitive markets "transcend conflict" (861).Rather, it highlights the degree to which their particular configuration is always a result of political conflict and coalition-building. 31 Markets conceived as sociotechnical agencements thus do not provide an economic alternative to politics but highlight the extent to which putatively economic mechanisms are themselves always already political.Thus, although Taylor's approach offers many important insights, it is not best characterized as being celebratory toward markets in general.Rather, we should understand his contribution in light of what MacKenzie suggests: that a politics that is either pro-or antimarket is misguided because markets are not a unitary thing to be embraced or rejected.Taylor shows that, over a range of important sociopolitical issues, there are liberatory deployments of politically constructed competitive market mechanisms.The construction of such markets involves deeply political choices regarding the forms of agency they enable."Apparently minor matters-'technicalities', often technicalities little understood by nonparticipants-can have big effects, for example, giving advantages to some actors and some strategies and disadvantaging others" (MacKenzie 2009, 33).So understood, Taylor's analyses help us to discern some of the conditions under which "market-like" mechanisms facilitate more empowered forms of agency for those in vulnerable social and political positions.The "agentic" component of the agencement is what helps us to evaluate when market forms of intervention might be domination-reducing rather than domination-enabling.

CONCLUSION
The paradigm developed in the ANT tradition encourages us to see market mechanisms as a set of polyvalent tools (Caliskan and Callon 2010;Callon 1998;MacKenzie 2009).When we face disputes over whether, how, or to what extent to marketize a given domain, we should conceive this not as a battle between irreducible and incompatible logics but rather as a choice among novel "sociotechnical worlds that are struggling to exist" (Callon 2007, 341).The performativity paradigm developed by these scholars helps us to see that what kinds of politicoeconomic solutions are possible in our world will depend, at least in part, on the theories of the economy or the market that we mobilize in constructing and evaluating them.Our own theoretical framings are, to a significant degree, implicated in the kinds of sociotechnical worlds we allow ourselves to construct.Realizing this implies an obligation on our part to consider whether and to what extent our own inadequate framings of markets might cause us to overlook alternative deployments of market mechanisms that might facilitate the realization of progressive political goals.It may well be that, as Taylor suggests, working "with the grain" of market mechanisms, under suitably (and politically) constructed conditions, can highlight unexpected possibilities for building a more just and emancipated world.
Although there is hubris to the right-wing obsession with the "perfect" or "self-regulating" market,32 there is something generative in the conception of markets as assemblages with their own unique forms of agency.Conceptualizing markets as STAs brings forward their (potential) role in facilitating the emergence of novel forms of political agency and social action.Such agencements might well play a role in a radical politics that seeks to overcome domination without embracing centralized forms of power.This view of market agencements suggests a politics that focuses on possibilities of shaping markets politically from within rather than insisting on alternative forms of political intervention that check impersonal "market forces" from outside. 33his returns us to the critique of Polanyi with which I began.The view of markets developed here suggests that debates about the political role of markets are not well understood in a frame that opposes a preexistent society to an essentialized market.What Polanyi's work so brilliantly chronicled was the brutal attempt to impose a certain form of market organization on an unwilling populace.The form of market order imposed in this project was radically disruptive to existing moral economies and gave rise to widespread degradation, poverty, and misery.Society reacted against these ravages by imposing restrictions and regulations that "tamed" such markets and subordinated them to other imperatives.From the perspective developed in this paper, however, we should view this not as a confrontation between the incompatible logics of society and the market but as a political dispute regarding the form that market mechanisms would take in developing industrial society.So understood, this episode exemplifies the claim of ANT scholars 30 As Gourevitch has argued in his explication of labor republicanism, the lack of a basic income on which to fall back is one significant source of the "structural domination" faced by wageworkers as a class (see Gourevitch 2014). 31Indeed, one additional advantage of the ANT perspective is that it sees the construction of market actors as a component of market politics.Thus how we understand the market mechanisms we deploy (and ourselves as agents within them) is itself (on this view) a contingent outcome of political judgment and struggle.This is another implication of the ANT view that society is itself an ongoing construction.In this sense, the Polanyian emphasis on the "reality of society" risks reifying what is a contingent (and revisable) achievement.
that society and economy (including various markets) are simultaneously coconstructed.Markets are not external abstractions at war with society but part of the material configuration of society itself.Society does not need protection from the market.Rather, we should be attentive to the variety of sociotechnical arrangements by means of which the hybrid collectives in which we live our lives can be inflected and formatted along more emancipatory and egalitarian lines.
The issues involved here are admittedly complex.I reiterate that I am by no means endorsing the view that markets are always and everywhere the most desirable solution for any political problem.Indeed, the analysis presented above renders such a claim utterly untenable; because there is no market in general, no necessary logic associated with marketization, it would be senseless to advocate market mechanisms, in the abstract, as a general solution to all political or social problems.Rather, what this analysis suggests is that significant political questions are often far less abstract and principled than familiar debates over the state versus the market imply.There will certainly be many cases when proposed market solutions are politically undesirable, but this will be due to the empirical details of the particular situation (including the availability of plausible alternatives) and not to an essential flaw inherent in the dehumanizing logic of the market as such.Theorists who want to think seriously about problems of inequality and oppression must be willing to engage in the subpolitics of market design, acknowledging that the devil is often in the details and that suitably constructed and regulated market mechanisms might provide outcomes preferable to those available via any alternative arrangement.
In closing, it should be acknowledged that the perspective articulated here does not provide any clear normative criteria for the evaluation of markets.What the ANT perspective on markets provides is a social theory that offers enhanced clarity on debates about the relationship between markets and politics.It thus offers, I hope, an improved way of thinking about the realization of our desired normative ends, whatever those ends might be.The perspective's only immediate normative (or, perhaps, meta-normative) implication, however, is negative; by showing that the opposition between political and market logics is spurious, it rejects any argument that opposes or embraces a specific institutional recommendation simply on the grounds that it is (or is not) market deploying.Beyond this, all the ANT perspective can offer is an insistence that details and technicalities matter and that markets are always material and sociotechnical mechanisms facilitating diverse forms of agency.Therefore, the implications for normative theorizing mirror those of classical pragmatism, which eschews abstract oppositions (such as that between the "public" and the "private" or the "individual" and the "social") in favor of the conviction that "it is all a question of what kind of procedures the intelligent study of changing conditions discloses" (Dewey 1935, 228).
In this way, the ANT perspective is somewhat deflationary.Its central demand is that normative debate over market forms engage with the details of institutional design, specifying concretely the mechanisms and arrangements that might bring about (or approximate) desired normative goals.I have illustrated the value of such an approach with respect to debates over markets in republican political theory.However, the same approach could be applied to the analysis of issues ranging from climate change (where disputes over the appropriate role of market mechanisms currently rage) to public school reform in American education (where debates about "school choice" are often framed in abstract pro-and antimarket terms).Both of these cases, as well as many others, would benefit from consideration in light of the ANT framework.Seeing markets as polyvalent sociotechnical assemblages makes them matters of collective concern, appropriate objects of democratic reflection and contestatory political debate-but not, I argue, of abstract denunciation or celebration.Only once we deconstruct the false binaries that typically animate debates over market politics can we begin the work of critical experimental inquiry: investigating the types of markets we can construct to endow agents with the tools, capacities, and resources necessary for formulating and pursuing their desired life courses.Markets, when viewed as collective devices that enable and configure agency, become material objects of political concern that might be deployed in service of diverse political projects.They thus serve, at least potentially, as aids in elaborating novel, plural, and egalitarian forms of sociotechnical life.