2800 words, 15 minutes’ reading time.
For part 1 of this series, go here. For part 2, go here. For part 3, go here.

4. Pettit on domination
I started out by saying that I have a conflicted–with any luck, instructively conflicted–view of Pettit’s republicanism, and of its application to employment-at-will. I wish I had a single snappy way of describing my conflict, but I don’t. At one level, I agree in principle with Pettit; at another level, I don’t. At one level, I agree with Pettit’s critique of employment-at-will; at another level, I don’t. Let me work through some of this out loud in the hopes that my tangles are instructive to others.

First, some theoretical agreements. Unlike Pettit, I incline as a first approximation toward liberalism rather than republicanism. But like Pettit, I find liberal politics defective in certain key respects. So like Pettit, I’m on the lookout for remedies for those defects. One real problem with liberalism is its tendency to reduce all questions of justice to questions of wrongful interference. It’s a trivial truth that for Smith to act unjustly vis-à-vis Jones is for Smith to wrong Jones, but on a familiar liberal view, for Smith to wrong Jones is no more and no less for Smith to interfere in Jones’s rightful zone of immunity. The implicit assumption here is that whatever Smith or Jones do within their respective zones of immunity is their exclusive, discretionary “business,” hence not a candidate for further moral assessment. This isn’t (I think) so much a logical implication of liberalism as it is a long-standing (and unfortunate) tendency among liberals, but in any case, it involves a problematically narrow conception of moral assessment, and in my view at least, requires correction of some kind.(1)

Pettit’s account of domination makes a valuable contribution here. I agree with Pettit that “domination” names a distinct, significant injustice not reducible to interference in the liberal sense, one that requires prevention and/or rectification of some kind. I also agree that domination can in principle occur in the absence of an overt act. You can, at the very least, dominate someone by possessing power sufficient to dominate them, at least if you indicate a will or desire to dominate them, whether or not you take concrete steps toward doing so. Pettit, of course, wants to go farther than this, but my point is that for present purposes, at least that much is common ground.

And yet I have some misgivings even at this very abstract level with Pettit’s account. Here are five, in admittedly telegraphic form:

4.1  Freedom as non-domination
I said above that the liberal conception of freedom needs supplementation of some kind, but Pettit’s conception of domination is meant not to supplement but to supplant it. It seems to me that Pettit lacks an argument strong enough to do this. To satisfy the relevant burden of proof, Pettit would had to have laid out and refuted the best arguments for the liberal conception of freedom, and then have shown in a systematic way how the republican conception improves on them. As far as I can see, he neither does this nor makes the attempt. Given that he doesn’t, Pettit offers no clear refutation of the liberal conception of freedom. A fortiori he fails to show that freedom as non-domination successfully supplants (as opposed to supplementing) freedom as non-interference.(2)

4.2  The arbitrary
Despite its centrality to his conception of domination, Pettit lacks a defensibly clear account of “the arbitrary.” The book’s index lists about 20 entries for “arbitrariness” and “arbitrary power,” but none of it, in my view, adds up to a fully satisfactory analysis of the concept. I think it’s fair to say, then, that the concept functions throughout the book as something of a semantic black box.

4.3  State power as domination
Though Pettit is admirably sensitive to the ways in which domination operates in private and commercial life, I find him problematically insensitive to the State’s capacity for domination. On a standard definition, the State wields a monopoly on the use of force in a given territory. A monopolistic force-wielder is by definition dominative, arguably the most dominative institution in human existence. In particular, if the State has a monopoly on the possession and use of the most powerful weaponry out there, and successfully regulates the possession and use of weaponry among non-State actors, the State becomes an unregulated regulator of the use of weaponized force. If it loses control over the use of force, it can dominate with impunity without being checked by any approximately similar counter-force.(3) This explains why the State has by far the worst historical track record of any known human institution when it comes to domination. No known entity has ever dominated human beings (or other living beings) with the same scope and intensity as the State. Whatever good it’s done has to be weighed against this fact.

To be fair, Pettit is aware of the dangers of State power, suggesting on consequentialist grounds that while (apparently?) dominative, the domination of a State is better than domination under anarchy, and further suggesting that democratic checks and balances would serve to temper the State’s (apparent?) tendency toward domination.

But when all is said and done, I don’t think Pettit comes clean on a basic semantic issue: does the State dominate us or not? On one interpretation, Pettit seems to be saying that the State dominates, but is permitted to dominate because there would be more domination under anarchy than under the State; in other words, the State dominates, but dominates less than the only other alternative, anarchy. On a second interpretation, Pettit seems saying that the State doesn’t dominate because when we face options X and Y, and X beats Y on consequentialist grounds, X’s beating Y entails that it can’t dominate: no option that is the best of the available options ever counts as dominative; the less dominating of two apparently dominating options doesn’t dominate. Since the State beats anarchy, and State or anarchy are the only options, the State doesn’t dominate. Pettit never directly addresses the issue, but it seems to me worth addressing. Do we live in a world in which domination is inevitable, or one in which it’s legislated away by consequentialist calculations?(4)

Further, despite the detail (and cogency) of his discussion, I don’t think Pettit ever responds directly to the would-be critic who sees greater threats from a force-wielding monopolist than from any lesser source. It’s not obvious that State control is always better than anarchy, even by the standards of republican freedom. Few Palestinians want to be ruled by Israel or the Palestinian Authority or Hamas–and with good reason. Forced to choose between rule by the dominant candidates for political authority or life under anarchy, many Palestinians might well choose the latter. Many Kashmiris might say the same about being ruled by India or Pakistan. Until very recently, the Zapatistas opted for a form of anarchy rather than rule by Mexico. And I’m sure there are other examples of this type. In quarrels over one-state and two-state solutions, there’s something to be said for no-state solutions.

While I agree with Pettit that corporate power is dominative, the fact remains that State power is more intensively so. While employment-at-will could get me fired without reason or cause, it still leaves me free to get another job and start anew. False prosecution and imprisonment do not. My boss qua boss cannot kill me because she’s having a bad day, but a cop qua cop could. The State of New Jersey has a more activist government than the State of Idaho, judging by the amount of government regulation that prevails in the two places. But it’s an empirical matter whether there is more domination in unregulated Pocatello, Idaho than in heavily regulated Princeton, New Jersey, and it’s surely possible that Pocatello beats Princeton in the non-domination derby. I don’t see how to integrate such considerations into Pettit’s theory. In the absence of any bona fide consequentialist calculations–and there are none in the book–Pettit’s biases in favor of an activist State seem, well, arbitrary.

It strikes me as problematic that Pettit’s State has no in-principle limits, and that its constitution contains no Bill of Rights. Further, Pettit explicitly gives the republican State substantial powers of what he thinks of as non-arbitrary “discretion.” Ultimately, I found it hard to be sure where this discretion ended and how non-arbitrary it was. You don’t have to be an anarchist to demand more stringent and determinate limits to State power than Pettit’s account yields. I’m not, but I would.(5)

4.4  Coercive paternalism
In the same vein, Pettit seems very insensitive to the paternalistic implications of his conception of State power. On Pettit’s view, the State can engage in drastic interferences into peoples’ lives without courting any loss of freedom, as long as it does so in a non-arbitrary fashion, and tracks what the State regards as the interfered-with person’s interest in not being dominated by someone else. Given the sketchy nature of his account of “the arbitrary,” of the relevant interests, and of precisely how the State determines when an interference would track someone’s interests, it seems to me that Pettit leaves the door wide open to a large number of highly contestable (and objectionable) interferences. It seems to me a bug rather than a feature that Pettit’s account so easily transforms paternalistic interferences into acts that can be engaged in without loss of the subject’s freedom.

Granted, two sections of the book suggest a vaguely anti-paternalist stance (Republicanism, pp. 158-63, pp. 183-87), but ultimately, I found Pettit’s formulations here so weak and abstract as to be essentially toothless. Put crudely, Pettit’s view seems to be that if State interference into your life non-arbitrarily tracks your interests, and you’re given room to contest the State when it decides to interfere, paternalistic interferences are unobjectionable. This seems to imply that if I want to commit suicide, and the State regards my death as contrary to my interests, and happens to be right,(6) it is obliged, at best, to give me a chance to make my case. But if I fail to persuade, it can indefinitely confine me to a psychiatric ward and indefinitely keep me alive contrary to my desire to die (and contrary to medical practitioners’ wish not to have to keep me alive). This strikes me as a reductio. In general, Pettit has little or nothing to say (in Republicanism, at least) about the value of individual autonomy, which strikes me as an overall liability for his view.

4.5  Power as actualization
Finally, Pettit’s equation of domination with the capacity to dominate, minus any other consideration, strikes me as implausible. In general, it seems a metaphysical or logical mistake to equate the sheer possession of a capacity with its actualization, however short the distance between the two things.(7) So it seems a mistake to equate the sheer possession of a power sufficient to dominate with domination itself. It seems more plausible to say that the possession of a power sufficient to dominate is a necessary condition of domination, and that this necessary condition becomes sufficient only after the addition of a further factor that generates an actual tendency for the capacity to be actualized. Absent any such factor, it seems a non-sequitur to insist that the sheer having of a power sufficient to dominate constitutes domination on its own.

Put it this way: Suppose that God exists and is all-knowing, all-powerful, and all-good (on a humanly-recognizable conception of goodness). If so, it strikes me as a reductio to say that the conjunction of God’s omniscience, omnibenevolence, and omnipotence entails that God dominates us. If domination is unjust, then it’s bad. If it’s bad, and God is all good, God would never engage in it. But if God is all powerful, He obviously has power sufficient to dominate us. Yet if He’s all good, and domination is bad, He can’t in fact be dominating us. My point here is not theological but logical or conceptual. The God example serves to show that sufficiency to dominate can’t be sufficient for actual domination.

Pettit’s view entails that God’s having the power sufficient to dominate us entails that He is dominating us, even if (and when) God is perpetually actualizing a goodness that’s incompatible with domination. That can’t be right.

It clarifies things, perhaps, to think of domination in tandem with threats and suspicions. It would be absurd to say that a person “was threatening” simply because he had the capacity to threaten or constitute a threat, or “was suspicious” simply because it was possible for him to do suspicion-worthy things.

Suppose that I’m standing on the platform at the train station waiting for the train. It is certainly possible for the person nearest me to push me onto the tracks as the train approaches. If my back is to him, he clearly has the asymmetric power to do so. But there is no plausible sense in which a random person’s merely standing there is a threat or is suspicious. To license the use of those terms, we would need to introduce a further factor that implied that the capacities for threat or suspicion were doing more than merely lying dormant. If we fix dormancy as the baseline, there has to be some tendency above this baseline toward suspicious or threatening behavior that licenses the predication of “suspicious” or “threat” to a given person. Qua dormant, the capacity has no particular normative significance. The same, I think, applies to domination.

Notes

1. For a useful corrective, see Roderick T. Long, “Why Libertarians Should Be Social Justice Warriors,” in Bissell, Sciabarra, and Younkins ed., The Dialectics of Liberty: Exploring the Context of Human Freedom (2019).
2. Roderick Long and I discuss this in more detail in the comments on part 2 of this series. I’d considered integrating that discussion into this post, but decided to leave it for another occasion.
3. I’ve discussed this dynamic in “The Contested Legacies of Waco,” Reason Papers 36:1 (July 2014).
4. An analogous issue comes up in Mill’s discussion of the indefeasibility of justice in the last few paragraphs of Utilitarianism. Some version of this problem arises for virtually any consequentialist theory.
5. This seems to me a problem common to civic republican theories as such. I can’t think of a prominent civic republican with a crisp account of the limits of the state, but can readily think of civic republicans who seem to be endorsing a limitless one. For a very clear and representative example, see Alexander Hamilton’s discussion of the powers of the federal government in Federalist #34: “There ought to be a CAPACITY to provide for future contingencies as they may happen; and as these are illimitable in their nature, it is impossible safely to limit that capacity.” Hamilton seems not to have worried about the contrary danger posed by a government without limits. As Patrick Deneen correctly points out, the underlying view of politics comes from another civic republican, Machiavelli (see Patrick Deneen, Why Liberalism Failed [2018], p. 167). That said, the criticism Deneen makes of liberalism seems more appropriately aimed at civic republicanism.
6. “Happens to be right” is ambiguous, actually. If I want to commit suicide, and the State wants to stop me from doing so, the State can “happen to be right” in the sense that the reasoning for my wish is in error. For instance, I may wish to commit suicide because I wrongly regard my situation as more hopeless than it really is. If the State objects, then the State “happens to be right” in the sense that it has a more realistic picture of my situation than I do. But morally speaking, suppose that I have the (moral) right to make my decision whether I am in error or not. If so, then it’s the State that is in error when it interferes with my erroneous choice. There are two errors in the example, not one. I don’t see that Pettit’s theory has the resources to make these distinctions, or to grant that people have the moral right to act on false beliefs.
7. Cf. Aristotle, Metaphysics V.12.