1. Pettit on Freedom and Markets

Although Pettit has made a number of refinements and restatements over the past two decades, his account of freedom as nondomination has remained essentially constant.Footnote ¹⁰ For Pettit, mere noninterference (which he associates with a liberal conception of freedom) is inadequate if explicating a satisfactory conception of political or social freedom. Instead, agents need to be secured in terms of nondomination. According to Pettit, domination occurs when an agent's choices are subject to interference by another agent, which interference is not under the control of the agent interfered with. Pettit formerly referred to this as “arbitrary,” but has preferred more recently to render it in terms of “uncontrolled” interference.Footnote ¹¹

On Pettit's account, merely being left uninterfered with is not sufficient to secure freedom, if such noninterference is granted only by the whim of a potential dominator, or is merely improbable, rather than being made subject to robust checking mechanisms. This can be brought out through the paradigmatic (for republicans) example of the slave with a benevolent master: insofar as the master continues to act benevolently, the slave may be able to fulfill all of their preferred choices without being interfered with, but nonetheless they remain in an important sense unfree insofar as the master could interfere with the slave without regard to their avowed interests. As Quentin Skinner has elucidated, the point is not simply that those in the positon of the slave will typically be forced to act in cowed and toadying ways to ingratiate themselves to masters, forcing them into a life of servile behavior and psychological deformity. The point is “existential”: that even if the master remains entirely benevolent and never interferes, and the slave thus never engages in psychologically and behaviorally servile ways, the slave nonetheless remains in a position of unfreedom as such.Footnote ¹²

On the other hand, Pettit's view of republican freedom as nondomination means that freedom is not necessarily lost even if interference does occur. This is because interference that is appropriately forced to track the interests of the agent interfered with does not count as a violation of freedom if the agent has himself authorized such interference, or has consented in appropriate ways to be part of a wider structure of checking mechanisms designed to prevent such interference being uncontrolled.Footnote ¹³

With this framework of freedom as nondomination in place, where does that leave the possibility of interference generated through market exchanges? Pettit insists that the market is no special source of uncontrolled interference when it comes to social choices. Of particular importance are two interconnected arguments. The first is that inequalities arising from fair market exchanges made against a background of just initial holdings do not in and of themselves compromise republican freedom. Assuming that there is a system of legally enforced property holdings, but one that excludes directly domination-generating property claims (e.g., slavery), and where exploitative practices are prohibited and prevented (e.g., predatory pricing, insider trading, etc.), and that in turn social agents are permitted to engage in free property exchange, it is unavoidable that inequalities will arise. In turn, the capacity to exercise social choices will be different for different agents, depending on their levels of property holdings. But this, in and of itself, does not mean that some ought to be considered socially less free than others. This is because the unequal distribution of property conditions but does not compromise the choices of agents.Footnote ¹⁴ For Pettit, unequal holdings which are the outcome of legitimate market exchanges are comparable to the natural environment. As he puts it: “The property regime can have the aspect of an environment akin to the natural environment. Like the natural environment, it will certainly affect the range or the ease with which people enjoy their status as undominated agents, and it may warrant complaint on that account, but it will not itself be a source of domination.”Footnote ¹⁵ Pettit does not deny that contingently inequality of holdings may generate domination, that is, uncontrolled interference, and thus compromise republican freedom, for instance if those who become better off via market exchange “lord it” over the poor, using their wealth to engage in arbitrary interference.Footnote ¹⁶ In which case, the republican will be in favor of interventionist measures (where feasible) to reduce domination via increasing levels of material equality, thus restoring (or at least improving) freedom as republicans understand it. But this is a separate issue from whether markets as such generate uncontrolled interference, and therefore domination. Insofar as Pettit compares unequal property holdings that are the outcome of market exchanges to the natural environment, he maintains that they merely condition, but do not compromise, social freedom, and are thus not in and of themselves of special concern to freedom as nondomination.

Pettit's second argument is that uncontrolled interference will only count as domination if it is intentional. As he puts it:

Pettit's requirement that uncontrolled interference be intentional to count as dominating is, it seems, connected to his claim that unequal property holdings are akin to the natural environment when assessing the effect of markets upon freedom. After all, if only intentional interference counts as compromising freedom, but unintentional interference merely conditions it, then the interference you experience as a result of my market activities will presumably be akin to the effects that you suffer from adverse weather. For example, if my market behavior raises the price of some commodity that you could previously afford, or removes from circulation some scarce good that you previously relied upon, so long as I did not undertake such activity with the specific intention of interfering with your choices, then you have no specific complaint in freedom against me. My cornering the market is like a lightning strike: bad luck for you, perhaps, but merely conditioning, not compromising, your social freedom.

Although he does not state it explicitly, by claiming that only intentional interference counts, Pettit blocks the claim implicit in Gerald Gaus's argument that republican freedom as nondomination entails the counterintuitive result that because we are all permanently at the mercy of interference produced by disparate market actors in diffuse webs of exchange interactions, we must therefore always be ipso facto subject to uncontrolled interference, and therefore extensively unfree.Footnote ¹⁸ By contrast, if only intentional interference counts, the Gausian point will not hold. In the next two sections, however, I call into question the plausibility of Pettit's position.

2. Mountains vs. Markets

Is the “property regime” (as Pettit puts it) akin to the natural environment, merely conditioning, but not compromising, our social freedom? Pettit's remarks on this matter are brief, and it is not clear how strict he intends this comparison to be. Nonetheless, he ought to be wary of resting much weight on it.

A central contrast is that the natural environment is in an important sense just there, regardless of what we do to it, whereas a property regime is not. Certainly we could attempt to bulldoze the granite monolith that stands between us and the next village, and indeed with sufficient (demented) political will it might be possible to literally move mountains. But if we do not apply such political will, then the mountain shall continue to be there: it was around long before we were, and will likely remain long after. Similarly, if we all died out tomorrow, the weather would continue unabated in some form or another, even if an end to anthropomorphic climate change meant different patterns compared to what we are presently helping to bring about.

This is manifestly not the case with (for example) a market economy as a particular configuration for organizing a property regime, precisely because the market is purely a matter of collective human agency, whereas our interaction with the natural environment is never just that. For a start, a well-functioning market economy is dependent upon wider background human institutions that enable it to operate: at a minimum, an effective coercive apparatus to enforce property rights, recognized institutions of trade that facilitate exchange at all levels of society, bodies of law that cover different forms of commercial transactions with courts to adjudicate decisions, some sort of banking system that issues and guarantees specie as a mechanism of exchange, and so forth. Furthermore, while different institutions and laws can be adopted (and, indeed, have been adopted) in different times and places to facilitate different forms of resource exchange, including attempting to do largely without markets, the economy is importantly dependent upon the continued psychological willingness of its participants to go on adhering to a large range of conventions that cannot be simply enforced by a central authority, but must be spontaneously rendered. The situation here parallels Hume's observation that all government is founded upon the “opinion” of the ruled.Footnote ¹⁹ If everybody decided to stop respecting private property rights, or to no longer place confidence in the value of paper money or believe that banks will guarantee deposits, then the economy would collapse. In other words, the economy is artifice through and through. Opting for a specifically market-based system of property allocation is an aggregation of institutional practices, historical experiences, and psychological dispositions, and is not a necessary condition of social existence (although it may feel like that to us now). While in modern conditions large-scale societies that have tried to do without market economies have been facilitators of far greater misery than those opting for some form of market-based system, they nonetheless illustrate the point that a property regime operating along market lines is something sustained by human activity, and not just there in some form or another whatever we may do.

This contrast matters, because it means that the normative attitudes we can appropriately evince towards a property regime and the natural environment are different. Part of the point here is to restate G. A. Cohen's observation that being unable to do something in a market economy is often not simply a question of lacking resources, but of being prevented by a coercive apparatus that will be activated in response.Footnote ²⁰ Whereas if I try to walk directly to the next village the mountain may prove impassable and I will be forced either to turn back or to die of exposure, if I try to board a plane to Australia without a ticket what will ultimately prevent me from succeeding will be the airport security services, backed by the coercive apparatus of the state.Footnote ²¹ In the latter case, there is a specific and identifiable structure of human agency that prevents me from exercising the choice I had hoped to make. This will be so in many situations where my lack of resources in a market economy forecloses the possibility of my exercising a social choice: what prevents me from doing what I want is not simply the lack of resources, but the lack of resources coupled with an effective enforcement mechanism.

In turn, the attitudes that individuals can appropriately hold towards the natural environment versus a property regime are importantly different. One way to bring this out is through the aptness of resentment.Footnote ²² If I discover that I cannot walk to the next village because a mountain blocks my way, it would be at the very least inapt to resent the mountain.Footnote ²³ To do so would be to direct a complex evaluative attitude reserved for interactions with agents at an inappropriate target. By contrast, resentment against the airport security staff would not be in the same way a category mistake. Insofar as these are ultimately human agents who are preventing me from getting on the plane, they are appropriate targets for my resentment. More abstractly, one could resent the entire system of private property rights and coercively enforced maintenance of such. For example, I do not even bother turning up to the airport, because I know what will happen if I try to board the plane without the ticket—but I may nonetheless resent the entire system at work, precisely because I know that it is a product of specifically human actions, intentions, and complicities. I may aptly resent those, and their aggregate consequences, and perhaps not just because I want to go to Australia, but because of what I believe their wider effects on human flourishing, equality, justice, and so forth, to be.

It is important to stress that others may quite reasonably view my resentment in this case as unjustified. For example, those who think that enforcing a private property regime is broadly defensible, and that coercive guarantees are justly used to prevent collective action problems from sabotaging the perceived wider benefits brought about through market exchanges, will want to say that I am making a mistake in resenting the wider private-property and market-exchange system in the way that I claim to. And they may be right. But even if so, there remains an essential point of contrast here vis-à-vis the mountain case. To resent the natural environment is to make a category mistake about how to apply a normative emotion and corresponding evaluation. By contrast there is nothing inherently inappropriate about resenting the market system, even if others think that such resentment is unjustified, precisely because the market system is a product of collective human agency. Whether one views such resentment against the market system, or even just some particular outcome it has produced, as either justified or unjustified will be a function of normative views one holds about wider political and economic regimes—that is, it will be a function of political ideology. You may well be right to think that the anarcho-syndicalist is making a mistake when she expresses resentment at the entire system of private property and its effects. But it is not because she has made a category mistake about the application of resentment. It is because you and she disagree politically.

What this means, however, is that it is too quick to say that markets merely condition our social choices, but do not compromise them. On the contrary, insofar as market exchanges are the outcome of aggregated human agency, sustained by a coercive state structure and upheld by the ongoing “opinion” of participants, their effects upon social choices are candidates for being considered relevant to judgments regarding freedom, and will be assessed accordingly, depending on where one stands politically. One cannot declare that the market and its effects upon social choices count only as conditioning, but not compromising, unless one begs the question in favor of an ideological outlook that treats a given property regime as though it were simply there. Alternative ideological outlooks are wholly entitled to point out that this is simply not true, and in turn to argue that, for example, market outcomes can in some cases compromise, and not merely condition, freedom. Whether this is so remains an open question, up for contestation, and any conclusion has to be argued for in reference to wider political commitments, and not simply asserted as a neutral fact. These points are worth bearing in mind in what follows.

3. From the Monongahela Valley

As already noted, Pettit thinks that because “social freedom is what is at issue, I have to represent an obstacle of a distinctively human, interpersonal kind and this means, in effect, that I have to be intentionally or quasi-intentionally obstructive.” The reason for this seems to be that for Pettit intention is what distinguishes an interference as of concern to social freedom: “I do not interfere with you just through happening, like a natural obstacle, to be in your way or just through doing something that has the unforeseen effect of hindering you.”Footnote ²⁴ This, however, cannot be right.

The problem is that although it is plausible to think that interferences relevant to social freedom need to be distinguished from those that are not, this cannot be done by simply focusing on intention, and declaring that interferences that are not the result of intention are therefore on a par with “natural obstacles,” and thus not relevant to social freedom. This is because there are cases of interference—especially arising from market activity—that cannot reasonably be declared not to count as interference simply because they are not the product of intention directed at those interfered with.

Take as an example the protagonist of Bruce Springsteen's song “Youngstown,” which recounts the changing fortunes of Rust Belt America, from the heyday of producing the cannonballs “that helped the Union win the war,” to a postindustrial decay where “that yard's just scrap and rubble.” In the song's climax, Springsteen's steelworker states:

One of the points being conveyed is that the economic devastation experienced by American industrial sectors in the latter part of the twentieth century was very much the product of human agency.Footnote ²⁵ Decisions to close factories and relocate production to cheaper bases overseas were made by human beings in boardrooms, not chemical processes in the natural environment. Were the social choices of formerly employed steel and coal workers interfered with by the decisions of the bosses to relocate production overseas? Undeniably so. Importantly, we can remain neutral at this point on how such interference is normatively salient—the only thing we need to agree upon is that interference occurred. But crucially, such interference need not have been intentional on behalf of the bosses. As Springsteen's steelworker puts it: “Once I made you rich enough / Rich enough to forget my name.” It might have been that the bosses who made decisions to relocate overseas never gave a thought to the impact this would have on working-class Americans. Let us suppose that the decisions were made purely on the basis of attempts to maximize profits, without it ever being the intention of the bosses to render thousands of people unemployed in the process. Certainly that was a foreseeable consequence of their actions—but it does not follow that they intended it. However, to say that the interference duly rendered should not count as socially relevant—that it is akin to the effects of a natural obstacle—simply because it was unintentional, cannot be right.Footnote ²⁶ Springsteen's steelworker is not making a mistake in believing that the actions of the bosses have interfered profoundly in his life, and that this is qualitatively different from the impact of a hurricane. Whether the bosses intended to wreak this devastation might increase, or alter, the force and nature of any resentment that the protagonist feels (and it is worth noting that the bosses simply being unaware of the devastation they have unleashed need not necessarily remove the steelworker's resentment, merely change its shape). But by itself this cannot mark the difference as to whether socially relevant interference has actually occurred. It manifestly has—and what normative attitude we take towards that interference is a question downstream of its existence.

Perhaps, however, the case of Springsteen's steelworker might be said to fall under the heading of intentional interference, insofar as the steelworkers feel that (as Pettit puts it) “blame is in order” vis-à-vis the bosses. Yet here we must note that the fact of socially relevant interference is not demarcated according to whether blame is thought to be appropriate. This can be seen as follows. If the steelworkers wish to claim that the bosses are causally responsible for their unemployment (i.e., a loose sense of “to blame”), then that is surely right. But if they want to say that the bosses are blameworthy as such, then that is a much stronger claim which requires taking a position—ultimately, a political position—on the question of the legitimacy of the bosses moving production overseas. Imagine, however, that we conclude against the steelworkers: that in fact blame is not in order, for instance, because we believe it to be legitimate in a market economy for producers to seek the lowest-cost production bases, wherever located. It surely cannot follow, however, that therefore no socially relevant interference occurred. It patently did. Lack of blameworthiness does not, in and of itself, mark the relevant distinction, even leaving aside the fact that an evaluation of blame will itself inevitably draw, once again, on the wider ideological commitments that given agents bring to bear.

Pettit ought not, then, to invoke blameworthiness as relevant to deciding whether freedom-affecting interference has or has not occurred. Indeed, in the original formulation presented in Republicanism, Pettit appeared to want to say simply that domination must be a result of human action, and that intentionality is a way of picking that out, in turn delineating human-generated interference from that occurring as a result of, for instance, natural forces or phenomena, the former but not the latter of which are relevant to social freedom.Footnote ²⁷ However, merely focusing on intention cannot do the work required. This can be seen if we turn to another example, where neither intentionality nor blameworthiness features, but socially relevant interference is indisputable. Imagine that by opening a series of low-cost car factories in Taiwan, a Taiwanese manufacturer undercuts the selling power of an American car firm, which duly goes bust, and lays off its workers. There is no suggestion here that the Taiwanese firm intended to interfere in the social choices of the American factory workers. And although the American workers might correctly identify the Taiwanese firm as the cause of their unemployment, it is implausible that we ought to conclude that the Taiwanese firm is to blame, at least where that implies that it has done something illegitimate. (If we accept the market as a legitimate way to organize a property regime, then we must accept that some agents will outcompete others; in this case, she who wills the end also wills the means.) Similarly, the impact upon the social choices of the American factory workers is indubitably a result of human action—but intentionality is not what marks that out to be so. Thus, to say that the Taiwanese firm is blameworthy because it outcompeted the American firm is incorrect, while focusing on intentionality is to draw the net too narrowly when it comes to socially relevant interferences. Yet from the perspective of a now unemployed American factory worker, what matters with regard to their social choices is not whether the Taiwanese firm intended to interfere with these, nor whether third parties consider such economic activity to be blameworthy, but that the firm has indeed interfered. And once again, the situation is very different from one in which the American factory had been destroyed by an earthquake, and likewise what normative attitudes we take towards this kind of case will be downstream of the fact that socially relevant interference has occurred.Footnote ²⁸

It is, I suggest, arbitrary to rule out unintended interference as irrelevant to assessments of impacts upon social choices; furthermore, socially relevant interference cannot be picked out by invoking the idea of blame being “in order.” There are cases—such as Springsteen's steelworker, or the Taiwanese manufacturer—where socially relevant interference occurs, even if this is not the product of an intention to interfere. Yet these cases are not akin to “natural obstacles,” because they remain a function of the economy as a wider system of human agency and artifice. However, admitting that nonintentional interference may count when it comes to assessing claims in social freedom poses serious problems for Pettit's position. For once this is permitted, the net will be cast too wide.

Consider the following example. Imagine that tomorrow the Organization of the Petroleum Exporting Countries (OPEC) decided to unilaterally raise the price of oil by $1000 a barrel. The impact this would have upon the wider global economy would be dramatic: jobs would be lost overnight; governments would have to drastically alter their tax-and-spend outlooks and undertake emergency measures to keep themselves solvent; financial markets would go into meltdown; in short, general economic chaos would ensue. It seems undeniable that interference would occur in the lives of many—in fact, billions—of people. What difference can it make that the OPEC cartel does not (let us suppose) intend to interfere in the lives of those billions? If OPEC raises the price of oil by $1000 a barrel, I will have a fair claim to say that they have interfered extensively with my social choices. What difference can it make that they do not intend to do so? What is relevant to me is that they do. And again, this will not be like being hit by a hurricane. Even if OPEC no more intends to devastate my life than a storm does, in the former case I can point to a structure of human agency and decision making, against which it is entirely appropriate for me to target my resentment, and not just at OPEC, but at the wider economic system that permitted such an outcome to even be possible, let alone occur. Whether I do express such resentment—and perhaps in turn want to mount my grievance as one in terms of a cost in freedom—will again depend on wider commitments that I hold, and how I view the normative significance of the interference I experience. But wherever I stand in those regards, the situation is not akin to the effects of severe weather.

The reason these considerations pose an acute problem for Pettit's theory of republican freedom, however, is that on his account, mere absence of uncontrolled interference is not enough to secure freedom, but we must be in possession of robust guarantees that no such interference will be able to occur, or else be possessed of a checking mechanism allowing us to retaliate and thus remove the effective power of a potential dominus, who is instead forced to take our interests properly into account. The problem generated by the OPEC example is that although we have good reason to believe that OPEC will not, as it happens, drastically raise the price of oil tomorrow, there is nothing we can do to guarantee that they will not do so. We live, as it were, at their whim, making OPEC the master, and the rest of us the slaves. But this has the dramatic implication of rendering everyone in the world, perhaps with the exception of the wealthiest billionaires, to that extent unfree, because liable to uncontrolled interference, and therefore (as Skinner would put it) in the existential condition of slavery. This in itself will strike many as a suspect conclusion: that we are unfree simply because OPEC could raise the price of oil, and thus exercise uncontrolled interference in our lives, even though we are very sure it will not, will be to many an indication that things are going wrong with the account of freedom.Footnote ²⁹ This impression will deepen when it is considered how many similar cases there must be, each implying that we are all (or at least, all of us except the uberrich) multiply enslaved by multiple masters. Imagine the levels of uncontrolled interference were the Chinese government to unilaterally dump all its US government bonds; or if Google began charging exorbitant rates for its currently free services; or if major investment banks refused to underwrite the liabilities of European mortgage lenders? None of these possibilities are subject to robust checking mechanisms by parties affected by such actions, and thus on Pettit's view of freedom as noninterference—if allowing (as we ought) that unintended interference ought to count as relevant to social choices—all are at present sources of unfreedom. But that means that we must all (the uberrich possibly excepted) be at present extensively unfree, living as we do in a world comprehensively structured by multiple interdependent markets and market actors, any number of whom have the potential to arbitrarily interfere with the social choices of millions, if not billions, of people. Unless, that is, one asserts that only intentional interference counts when it comes to social choices—but as we have seen, there is good reason not to accept such a claim.Footnote ³⁰

In the next two sections I consider some potential republican responses, but a further comment is worth making. This is that although I believe many will find it implausible to conclude that the vast majority of people right now are unfree just because OPEC could drastically increase the price of oil, it does not necessarily follow that we would want to say that if OPEC did take such dramatic action, the result would necessarily be a loss of freedom to those affected. Similarly, I take it to be an open question in the case of Springsteen's steelworker, or the Taiwanese manufacturer, whether the result of economic hardship caused by the loss of working class industry is a loss of freedom. In all cases we can grant that socially relevant interference occurs, but reasonable parties may disagree as to whether this in turn constitutes a cost in freedom. How one views such cases will depend on wider commitments and beliefs about the extent to which freedom is tied to economic independence; for example, whether one can meaningfully be said to be subject to freedom-relevant interferences if one experiences withdrawal of (e.g. economic) opportunities, as opposed to experiencing the provision of incentives where they did not exist before. Or the extent to which freedom has to be understood in a specific context, for instance, if one thinks that the most relevant fact regarding the freedom of steelworkers is precisely their relationship to the bosses (as Marxists might maintain), or whether freedom can be comprehended abstractly in terms of identifying whether interference has or has not occurred, without it being relevant what the local circumstances are (a position more often favored by theorists of negative freedom, in one strand of the liberal tradition). And so on. In other words, whether interference counts as freedom violating is not a free-standing conclusion to which all parties might be reasonably expected to agree if debating in good faith, but is a function of the wider political views that differing parties bring to bear when they make claims about whether a particular sort of interference ought to count as freedom violating. Furthermore, participants in any political contest over freedom will themselves be invoking that term not simply to describe an agreed set of circumstances, but to attempt to influence perceptions not only of what is normatively at stake, but also regarding how the facts of the matter are themselves to be interpreted. Judgments of freedom will—like instances of resentment against the outcomes, or even bare existence, of market economies—be extensively conditioned by existing ideology. To hope that this can be bypassed by focusing on interference, and ascertaining whether such interference can be said to have occurred or is to be permitted to count, is to fail to see that all such judgments will already be drawing upon a range of wider commitments that are themselves controversial, and not necessarily shared by political opponents. Once those commitments are acknowledged, however, we see that disagreement over freedom is inevitable in a pluralistic society characterized by differences of political opinion, and cannot be legislated away via technical definitions focused on purported facts of interference.

4. Republican Responses

Might Pettit not respond that the above arguments simply miss the point? After all, his claims about interference were focused upon a much narrower target than I have been interested in, namely, the market as a price-allocation mechanism understood in terms of relationships between buyers and sellers. By contrast, the examples employed above are better described as interferences generated through the economy, rather than the market specifically. It is true that the economy is in a sense the aggregate sum of market transactions, but these are distinct targets of analysis. Might it not be replied that Pettit's point was simply to distinguish republicanism from more radical positions—for example, communist and socialist—by insisting that the republican need not claim that market transactions are in themselves compromising of freedom? If the market is merely a price allocation mechanism, so long as parties are of relatively equal bargaining power and transactions preclude inherently dominating practices, it follows for the republican that no problematic arbitrary interference will be generated. When it comes to the economy, things may be different: unequal bargaining powers that have accrued from a multitude of market transactions may (it might be granted) generate considerable sources of arbitrary interference, and these are of concern to the republican as losses of freedom. But those are byproducts of repeat market exchange, rather than features of the market (a price-allocation mechanism) itself. Have I not, therefore, missed the point?

I believe not, because if Pettit restricts his scope only to (as he sees it) well-functioning markets where buyers and sellers in one-to-one relationships do not face significant disparities of bargaining power, then the focus is so narrow as to be uninteresting. We can happily grant that under idealized circumstances mere market exchange in and of itself need not generate arbitrary interference (e.g., if buyers and sellers can undercut prices in turn, retaliating through the market itself so as to protect their own positions). But those are not the market circumstances that the vast majority of human agents find themselves in, and thus not the circumstances relevant to assessing the extent to which individuals really are subject to (the mere possibility of) arbitrary interference as produced by markets. To be told that markets do not necessarily generate such interference is not very interesting if what we want to know is what to say about situations where they do—as appears frequently to be the case.

Similarly, to complain that the examples used in the previous section are cases of malfunctioning markets (because characterized by e.g. uneven bargaining power), and thus not damaging to the republican position, is to fail to see that given that the economy is systematically characterized by the effects of “malfunctioning” markets, it is precisely such cases that the republican needs to say something about. Yet, as I have tried to show, real-world markets pose considerable challenges to the republican view that freedom is lost by the mere possibility of arbitrary interference, because real-world markets generate such possibilities on a massive scale. Similarly, if it is argued instead that the problematic cases highlighted above are not really about markets in the narrow sense of a label for the price-allocation mechanism, but about lack of control over, for instance, corporate decision-making by those subject to their effects, the point may be granted, but nothing of substance changes. If what is required to prevent arbitrary interference generated by the decisions made by OPEC is effective checking by those subject to its power, then it must still remain that in the absence of an ability to force OPEC to track the interests of those potentially interfered with by its decisions, then those affected are thereby rendered unfree—and this is precisely the conclusion that I suggested is implausible. Whether we say that such interference is a product of a market, or caused by lack of effective control over a corporate actor that is operating through a market, is in the end not important to the substantive question about freedom.

This brings us to a second potential line of republican response. A recent literature has attempted to articulate what a specifically republican political economy might look like.Footnote ³¹ An often-suggested feature is the provision of a universal basic income (UBI), providing citizens with some level of economic security, envisioned as a way of insulating employees from the ability of employers to dominate through the control of terms and conditions of labor.Footnote ³² In fact, as Rob Jubb has shown, it is not straightforward that republicans must favor a right to an unconditional basic income, owing to the differing capacities of individuals to exit domination-engendering employment arrangements as generated by the provision of UBI itself.Footnote ³³ Furthermore, republicans are divided on whether UBI should merely provide enough to secure citizens in basic economic necessity, or whether it should be so high as to enable exiting the labor market entirely. But I here put these more specific issues aside, and concentrate on whether provision of a UBI could address the problems raised in the previous section.

My response is that it cannot, for the simple reason that if the republican claims that in order to avoid being dominated what is required is that all citizens receive UBI (at whatever level stipulated), then it must remain the case that in the absence of such a regime of economic guarantees, we remain extensively unfree. There may be excellent reasons to think that a UBI would be a good thing in terms of reducing domination across a range of metrics, and thus that there are strong reasons for favoring such a policy (and one need not be a republican to agree). But in terms specifically of freedom, if the absence of a UBI means that we are unable to control arbitrary interference, the Pettit-style republican must remain committed to saying that we are now essentially unfree, because at the mercy of uncontrolled interference. Indeed, pretty much everybody who has ever lived in a market economy, and who is not wealthy to the point of being thereby insulated from the impact of market forces, must on Pettit's view be likewise essentially unfree. But this is the same dramatic conclusion reached in the previous section, and indicates a much more radical implication regarding the relationship of freedom to markets than the “complacency” Pettit claims.

The same can be said of Robert Taylor's suggestion that republicans be “enthusiastic” rather than merely “complacent” about markets, and should outright champion them because the more that a market system approaches conditions of perfect competition, the more it restrains and eventually eradicates market power, and thus “helps us to realize ‘market freedom,’ i.e., freedom as non-domination in the context of economic exchange.” Taylor argues that perfect competition ought to act as a “regulatory ideal,” no different from the rule of law, which although imperfect in practice “the closer we can approach them, the closer we will be to achieving nondomination in economic and political life.”Footnote ³⁴ In fact, the idea of perfect competition is very different from that of the rule of law: the former is a hypothetical modeling device used by economists to work out the implications of a set of constrained assumptions and explicitly not intended to explain real-world economics, the latter a set of principles as well as practices evolved in real institutional settings in complex, historically variable attempts to administer ongoing affairs of state. But in any case, as regards the question of freedom it must remain that in the absence of Taylor's envisioned realm of perfect competition, wherein no citizen must be a price taker and no employer or corporate actor a price giver—that is, in the world we currently occupy, have occupied for some time, and will go on occupying for the foreseeable future—we will remain subject to uncontrolled interference generated via market activity, and thus extensively unfree. In other words, insofar as we do not live under conditions of perfect competition—indeed, far from it—the republican conclusion Taylor urges must in fact be that we are all (save the uberrich) to that extent currently in the dominated position of slaves.

In the following section I address the potential republican reply that we should bite the bullet: that if our intuitions about freedom balk at the conclusion that we are all now extensively unfree because of the widespread existence of markets and the potential for uncontrolled interference they generate, then that is the hard truth we must accept and reconcile ourselves to. But before turning to that, it is worth considering the position of the nineteenth century “labor republicans” as reconstructed by Alex Gourevitch.

It is not possible here to do justice to Gourevitch's recovery of the movement he calls labor republicanism, dating from the nineteenth century, and thus neglected by those neorepublican theorists focusing on pre-eighteenth-century sources. But pertinent to the present argument is his reconstruction of the idea of “structural domination.” Gourevitch argues, on lines similar to the above, that Pettit's intentionality requirement is unduly limiting in insisting that domination must be a one-to-one relationship whereby a specific master intentionally dominates a specific slave, and he expressly draws a distinction in turn between Pettit's neorepublicanism and the “labor republicans” of the nineteenth century.Footnote ³⁵ Gourevitch reconstructs the claim of labor republicanism that because workers were forced to sell their labour to some employer in order to survive, they found themselves in positions of domination, upheld by a wider economic system whereby productive assets were controlled by a few, who were in turn able to exercise arbitrary power over workers, even if no specific employer intended to act as a dominus over any specific employee. This is a situation, however, which not only individuals, but entire communities, cities, and even countries (e.g., with regard to powerful corporations), can find themselves in.Footnote ³⁶ Gourevitch is in turn alert to precisely the problems for a republican account of freedom articulated above, and recognizes that Pettit's debarring of unintended interference from counting as domination is crucial to the latter's position: “If we expand our definition of domination to include other kinds of agents beyond personal masters then we run the risk of collapsing the distinction between unintended consequences and intentional interference. That, in turn, threatens to transform our anonymous interdependence, our very social existence, into domination. Then nobody could be truly free.”Footnote ³⁷ According to Gourevitch, although this dramatic conclusion threatens Pettit's position, he believes the labor republican view avoids it by training its response on unequal control over productive assets. What is of concern to labor republicans is not simply our dependence upon all anonymous others, but our dependence upon some employer or other, in a situation of structural domination. The labor republicans called in turn for radical rearrangement of the way that productive assets were organized in society, so as to eliminate structural domination as generated by extensively uneven market bargaining power. Hence, it is unsurprising to find that labor republicanism “goes in the direction of socialism.”Footnote ³⁸

The point for present purposes is to note the concomitant radicalism of the labor republican position, which calls for large-scale restructuring of control over productive assets. This stands in marked contrast—as Gourevitch points outFootnote ³⁹—to Pettit's assurances that his neorepublicanism is only mildly revisionary, not fundamentally at odds with the broad institutional realities of present Western societies.Footnote ⁴⁰ On the contrary, if we take seriously the claim that freedom requires removal of the mere possibility of arbitrary interference, once we admit that economic relationships generate such interference, then modern market economies constitute massive sources of unfreedom. Short of a radical program for transforming them—for example, some sort of socialized control of the means of production—they must, on the republican view, render most of us unfree. In other words, Pettit can have a relaxed view of the conditions of market modernity, or a theory of freedom as the absence of arbitrary interference, but not both.Footnote ⁴¹

5. Accepting Modernity

What might be said to a hard-headed republican reply, one that concedes that markets generate extensive uncontrolled interference, but accepting the counterintuitive conclusion that most of us are therefore extensively unfree (perhaps insisting in turn that it is our intuitions, not the republican account, that need to be revised)? The first thing to ask is why we ought to continue to support republican theory so construed if it yields such counterintuitive responses. (This is particularly acute for Pettit, who has claimed that the method of reflective equilibrium supports his arguments for liberty as nondomination).Footnote ⁴² I take it that part of the initial appeal of the republican theory is that it can make sense of our intuitive reaction to cases whereby freedom seems lost even when interference does not actually occur, as in the case of the slave with a benevolent master. Whereas negative liberty accounts seem unable to explain the apparent lack of freedom in such cases, Pettit's republicanism can. However, the recent proposal of an alternative understanding of liberty as “independence”—advanced by Christian List and Laura Valentini—indicates that republican liberty is not unique on this score. List and Valentini's account, centred on “the robust absence of constraints” (deliberately leaving open how to specify what counts as appropriate robustness as a matter of judgment), persuasively explains the intuitive unfreedom of the slave with a benevolent master, but without endorsing the republican emphasis on arbitrary interference.Footnote ⁴³ Alternative reasons for insisting on the primacy of the republican theory are therefore required.

List and Valentini also point us in a further helpful direction in our reply to the hard-headed republican. As part of their mapping of the conceptual space of different accounts of liberty, they suggest as a necessary desideratum that any satisfactory account of freedom “displays an adequate fidelity to ordinary-language use,” insofar as “a plausible conception of freedom should correctly capture at least those instances of social freedom and unfreedom about which competent speakers agree, namely, ‘paradigmatic cases.’”Footnote ⁴⁴ List and Valentini argue that republican liberty fails this test by generating conclusions about freedom—such as that the justifiably imprisoned criminal is not rendered unfree if her imprisonment is the result of an appropriately democratic political and legal procedure—that run too manifestly counter to ordinary language usage. Although it must be emphasized that this is not List and Valentini's main argument against republican liberty (which is that a Pettit-style conception is unduly moralized because of its emphasis on arbitrary interference and what counts as such), it is nonetheless worth taking notice of.

In the first place, the case that I have urged above is the mirror of List and Valentini's. Whereas they suggest that republicans must end up claiming that manifest cases of unfreedom—like the justly imprisoned criminal—are no such thing, I suggest that the hard-headed republican who insists that billions of people are unfree simply because OPEC could dramatically raise the price of oil is making a mistake in the other direction: claiming unfreedom in cases where competent language users would deny this to be appropriate.

A connected point cuts deeper, and relates to why we should care so much about what competent users of language have to say when it comes to political philosophical debate. After all, is it not the task of the philosopher to identify the truth, no matter how uncomfortable or strange common sense may find it? Here, however, we do well to recall Williams's insistence that “various conceptions and understandings of freedom, including the ones we immediately need for ourselves, involve a complex historical deposit, and we will not understand them unless we grasp something of that deposit, of what the idea of freedom, in these various connections, has become.”Footnote ⁴⁵ In particular, as Williams emphasizes, any meaningful account of freedom will have to be a value for us—and although we may disagree extensively about how and why that is so, if we hope to be able to make sense of what freedom has come to mean for people like us, arguing about it here and now, then we will have to accept that “modernity is a basic category of social and hence political understanding, and so a politically useful construction of liberty for us should take the most general conditions of modernity as given.”Footnote ⁴⁶ A central component of modernity (at least in Western societies) is that market economies intimately structure and condition the ways that we must now live. But as a result, an account of liberty that balks at the present widespread existence of markets as generating extensive unfreedom is failing to take as given the basic conditions of what freedom could mean to people like us, and what (as Williams puts it) the value of freedom has to do for us—and thus will not constitute a helpful way of attempting to understand what is at stake. This is not to deny that “there is room for much argument about what the conditions of modernity are,” and in turn that there will be vibrant debate as to when and how liberty is compromised under such arrangements.Footnote ⁴⁷ But, as noted above, these will be thoroughgoing political disagreements, over a range of contested issues, and not simply disagreements about facts (or otherwise) of interference. The point, then, is not that republican liberty is somehow unintelligible in conditions of modernity (clearly that is not so), but rather that when we see what it must actually entail in a world of markets, it becomes deeply implausible—and hence we are better off looking elsewhere.

As I have emphasized throughout, when it comes to judgments of freedom we do not make these from a freestanding basis, but already suppose a set of background ideological commitments which may be more or less hostile to various features of modernity—including especially markets and their effects—and in relation to which judgments of freedom are necessarily structured. Yet republican liberty as nondomination, as well as being implausible with regard to how it must construe freedom in a world of markets, is also badly equipped to make sense of this central feature of disagreement regarding the content of liberty. Furthermore, the centrality of disagreement to debates over freedom—because dependent upon wider political ideology, and drawing on complex and contested historical deposits—indicates that Pettit's early hope that republican liberty would prove both an overarching and ecumenical single social idea, around which many different political outlooks could converge, was bound to be futile—as indeed it has proved.Footnote ⁴⁸