Domination The Art of Sex
Part 3
Suppose that for various historical, economic, and cultural reasons, one group in some society manages to acquire a preponderance of social power,
which it wields over the other groups in that society directly and without constraint, much to its own benefit.
Since the subordinate groups are in no position to challenge directly the preeminence of the powerful group, they instead demand only that the various rights and privileges of the latter be written down, codified, and impartially enforced by independent judges.
In time, let us suppose, the powerful group accedes to this demand, on the view that since the rules will be designed to benefit them, after all, there will be no significant cost in their doing so. (2012: 147)
If power is organized and systematized in law, it does sound odd to call it arbitrary. After all, it may be entirely predictable in such circumstances.
Lovett tacks closer to the commonsense idea that what is arbitrary is unpredictable, unreasoned, and unfettered by rules.
He even claims that “real-world systems of domination” like Jim Crow and South African apartheid might cease to be dominating in possible worlds where they managed “strict adherence to explicit rules and procedures” (2010: 101). That such regimes would remain oppressive is obvious, but he believes this should be explained by the presence of other political evils than domination.
Instead of a substantive account of arbitrary power—where power dominates when it isn’t forced to track the will of those subject to it—he recommends proceduralism: power dominates because it isn’t “reliably constrained by effective rules, procedures, or goals that are common knowledge
to all persons or groups concerned” (Lovett 2010: 96–97; 2012; for more treatment of the debate between Pettit and Lovett about the nature of arbitrary power, see the section on “What Counts as Arbitrary Power?” in the entry on republicanism.)

Non-Moralized, Norm-Dependent Theories
When we take a hard look at the most obvious cases of domination—masters, tyrants, patriarchs—we’ll probably notice that they all claim authority. The Paradigms all typically think of themselves as the ones who make the rules, and that their subjects have an obligation to comply.
To say that dominators, as such, always claim this kind of authority is to endorse a norm-dependent theory: that
A dominates
B by virtue of having a kind of socially legitimate power over them.
Not all social power operates in this way—or doesn’t seem to. The neighborhood mob boss who comes by your shop demanding protection money almost certainly does not expect you to think of his command, “Pay up or else!”, as an expression of legitimate authority backed by sanctions. Plausibly, the typical mob boss cares only that you recognize him as a source of credible threats and comply.
According to most norm-dependent theories, this means that the mob boss does not dominate. Domination is always power “under color of right”. The mob boss acts under no such colors.
The patriarch, by contrast, believes that his commands create obligations for those within his household; and this – not merely his power to interfere with choice—is why he dominates.
Here, norm-dependent theories part company from their norm-independent competitors: the latter will tend to see both the mob boss and the patriarch as sources of domination.
As canvassed above, norm-dependence and moralization are separate issues. We don’t need a theory of the right or the good to tell us whether someone has power legitimized by local social norms or laws.
Of course, we do need a moral theory to tell us if social norms or laws are right or just or truly legitimate. As we will see in the next section, the most influential norm-dependent theories of domination tie it to a specific failure to respect the actual moral authority of the dominated, and so offer a moralized as well as norm-dependent theory.
Why prefer norm-dependence to norm-independence? Often, the move to norm-dependence is motivated by familiar worries about over-generalization.
The world may be full of uncontrolled powers of choice interference, but it is less common for one agent to regard someone else as a source of binding norms. I may have an uncontrolled power to interfere with your choices whenever I can take the last seat in a café before you can get to it; this obviously has nothing to do with any kind of authority over you. The latter power seems less ubiquitous and more serious.
Also, norm-dependent, non-moralized conceptions are perhaps best equipped to diagnose domination where the dominated have internalized justifications for their status. Neorepublicans often highlight how victims of domination don’t have enough power to contest their position, but what about cases where victims of domination are unwilling to contest because they accept a worldview that justifies their domination?
Michael Thompson calls such unwillingness a manifestation of constitutive domination: the waynorms, institutions, and values of the community shape the rationality of subjects to accept forms of power and social relations and collective goals as legitimate forms of authority. (2018: 44)
In so doing, he resists standard neorepublican talk of “arbitrary” power. Domination is constitutive just because it is not regarded as arbitrary by those under its sway. Instead, he claims, domination is the expression of widely rationalized and internalized norms. Certainly,
if we agree that all power necessarily involves legitimacy claims (see Azmanova 2012: 49–50), it makes sense to think that intense varieties of social power, like domination, will go hand-in-hand with thoroughgoing acceptance of legitimacy claims.
According to Marxist and Marxist-inflected theories like Thompson’s, neorepublicans are misled by their paradigms. Neorepublicans get their touchstones of domination from slavery and traditional despotism—varieties supposedly more prevalent in pre-modern societies,
where the threat of violence was the primary currency of domination. Instead, according to Marxist theories, we should see domination in the ideologies that make violence and coercion less necessary:
for example, when the redistribution of pubic goods to private elites is widely accepted as natural or as the result of free exchange, even by many who suffer as a result.

Of course, for a non-moralized theory, identifying domination as anti-democratic must involve a non-moralized account of democracy.
This is not necessarily a problem, given that we can describe political institutions as “democratic” in a merely descriptive sense. (e.g., someone who says,
“The expansion of the franchise leads to a more democratic society” might simply mean that a larger percentage of the population gets to vote and think that is a terrible idea and a reason not to expand the franchise.)
What is more difficult is showing how even broadly democratic states are non-dominating if we think the absence of domination involves literal control of the state by those it governs.
It is clear that the vast majority of individual citizens do not control their state in any meaningful way; given the enormity of contemporary states, it is unclear how they could.
We might say that they should rest content with a fair share of control, or with a fair chance at control. But to say that we avoid domination when we have a fair share of control looks very much like a way of moralizing the theory, since fairness is clearly a moral notion.
(Pettit’s primary attempt to explain non-domination in terms of democratic political arrangements can be found especially in the final three chapters of On the People’s Terms.
For more interpretation and criticism of Pettit’s idea of control, see Arnold & Harris 2017, Mayer 2015, Schink 2013, Sharon 2015, Simpson 2017, and Kolodny forthcoming.)
Frank Lovett argues that avoiding domination does not require democracy, but instead subjecting the powerful to reliably enforced and widely known rules.
Perhaps democracy does, in fact, most effectively reduce domination, but this should follow from substantive argument, not from the mere analysis of concepts (Lovett 2010).
Also, there is reason to think that subordinate groups are less dominated whenever their overlords must abide by reliably enforced and widely known rules, even when those rules do not express the will of the subordinated in any way. Lovett uses the following case to make his poin