From the anti-vigilante principle to authority (a good intuitive argument for authority)

In Ch. 8 of DEMOCRATIC AUTHORITY, David Estlund argues for a certain kind of political authority on a purely intuitive basis (as a run-up to a more-principled or intuition-vindicating defense of political authority).  His argument starts with the intuitive (and Lockean) anti-vigilante principle (AVP):

when there is a system that serves the purposes of judgment and punishment without private punishment, then private punishment is morally wrong  

The idea here is that the obligation not to engage in relevant sorts of private punishment (even when the public verdict is known to be wrong) is generated by the system of public justice forbidding private punishment or vigilante behavior.  Since forbidding-generated as well as command-generated obligation (to obey) suffices for authority, what we have here is a kind of political authority. (Notice that, despite my language here, the system need not be public in anything like the governmental sense. The system could be privately-run but dominant in a geographic area.)

One might, of course, reject AVP.  More plausibly, one might accept a version of AVP that is consistent with skepticism about political authority because one’s obligation not to punish privately runs out whenever the system of public justice fails (or reasonably seems to fail).  But the stronger version of AVP is highly intuitive.

One might object that an adequate system of public justice does not forbid one from seeking private justice (after all, there might be no direct command issued by anyone).  In which case, perhaps we have a rather interesting case like that of generating obligations to stop in drivers by stepping out in front of their cars, but we don’t have authority.  But, if the norms and expectations in a group come to ‘you are not to do X’, this comes to the group forbidding X. So dominant-collective-attitude-style forbidding does not and need not involve any order-giving or commanding.  (Alternatively, one might argue that what we need for political authority is not forbidding, but commanding or order-giving. But forbidding seems to be in the right family of social-regulatory actions and attitudes.)

Estlund worries that the relevant actions or states might cause the obligations in a merely-accidental way (as a mere side-effect of action, as when I walk out in front of your car), but I don’t see why this is a worry.  Once it is granted that the group forbids one’s X-ing, it is a mere philosopher’s quibble to worry that, perhaps, despite this, the obligation – of precisely the right content – is not generated in the right way. It is pretty clear that, in any realistic scenario, if you get the obligation not to be a vigilante from an adequate-enough public system of justice forbidding it, the obligation is generated, if it is generated at all, by the forbidding and in the right way for political authority.  

(Estlund’s language of ‘the moral power to forbid/require’ – as against, say ‘causing there to be a moral requirement by requiring’ – does not make clear that the obligation-generating action is forbidding/requiring descriptively considered.  In line with this, the relevant distinction for him is that between obligations or their generation being accidental or non-accidental relative to the relevant institution or practice. I don’t think this is the best way of framing things. However, I also worry that there is something subtle in Estlund’s treatment that I am missing.)

This is a pretty good intuitive-level argument in favor of at least a limited kind of political authority! Maybe it is still a puzzle how this could be – maybe, in the end, the relevant intuitions and hence the conclusion cannot be vindicated in a deeply explanatory/justificatory way – but that this is seems pretty intuitive once we accept AVP and some ancillary premises and think things over in a logical way.

(The philosophical anarchist would deny that the obligation here extends to not punishing privately in relevant ways when the public system fails to punish a guilty party.  The point here is simply that, in the case of public criminal justice, what philosophical anarchism says is, for many people, not very intuitive. Of course, substantive intuitions being what they are, mileage will vary.)

19 thoughts on “From the anti-vigilante principle to authority (a good intuitive argument for authority)

  1. The anarchist can accept something like the Lockean obligation to avoid, where possible, being a judge in one’s own case, by taking it as a requirement to submit one’s disputes to a neutral third-party arbiter. But given that (barring a fallacy of composition) this doesn’t entail — and in fact RULES OUT (a point that Locke misses) that there be any ONE third-party arbiter (be it a person or an organisation) to which everyone submits their disputes, this is going to involve only horizontal and not vertical appeals.

    More concretely, if there are just three people, A, B, and C, then any dispute between A and B must be submitted to C; any dispute between B and C must be submitted to A; and any dispute between A and C must be submitted to B. So this may give each one a sort of authority relative to the other two, but the authority is limited, temporary, and reciprocal. And even that authority dissolves once there’s more than three people; if there’s four people, for example, then any dispute between A and B must be submitted to C or D, so neither one has any authority until chosen by the disputants (or by one of them, if the other is refusing to submit to arbitration).

    What if there are only three people, but C rejects the obligation to submit her own disputes to a third-party arbiter? In that case, I would say that C has disqualified herself from serving as such an arbiter, in which case A and B no longer have a obligation to submit their disputes to C. That implies that the state, by claiming a territorial monopoly over the relevant service, is disqualified, in which case disputants have no obligation to submit their disputes to the state even if it is the only available public system.

    I’ve put all this in terms of dispute resolution rather than punishment, since I take punishment to be a disproportionate response to wrongdoing, and thus to be unjust. If in fact the state uses punishment, then that’s a further reason for the state to be disqualified; if C is known to use unjust procedures systematically, then A and B have no obligation to submit their dispute to C.

    Liked by 1 person

    • By the way, I don’t know why my name is now showing up as “berserkrl” instead of “Roderick T. Long.” But I’m used to WordPress being randomly evil.


    • Thanks, Roderick, for the compelling anarchist model in the general ballpark of AVP. Great to get my more-libertarian neural pathways going again.

      I take it that, on your view, the basic obligation in the logical environs of AVP is the obligation that we owe to each other to settle disputes (and potential rights-violations) by submitting to the verdicts of a neutral third party (we might call this the neutral arbiter principle or NAP). Since the obligation here is an obligation to consent, not an obligation to obey, NAP does not establish any sort of authority at all (at least not without something like Estlund’s idea of normative consent). However, if one agrees to have one’s disputes and potential rights-violations settled in such a way, one is, at least temporarily, under the authority of the arbiter. So far, so good?

      You say AVP is not true because a de facto monopolist arbiter (like the government) might take actions to enforce its monopoly – and this would disqualify it from being a neutral arbiter. That seems a little quick. I could be biased in favor of myself, but be a great, impartial judge in cases that do not involve me. Or maybe the idea is that, in enforcing its monopoly, such an arbiter would be initiating aggression or force and this is morally disqualifying in some more general way? Given your assumptions, I’m sympathetic to your point here, but maybe what is well-supported is something a bit weaker.

      Changing gears: I think the distinction between theories that accept and theories that reject society itself (relevant social conditions) as a source of original obligation (obligation that is not generated merely via prior obligations and instrumental or instantiation relationships) is fundamental. Implicit in Estlund’s approach (and his – and I think the natural or intuitive – interpretation of AVP) is the idea that social conditions themselves generate original obligations to obey. In a different part of Ch. 8, Estlund gives a more-detailed account of what these conditions are in the case of obligations to obey or authority (and why actual consent is not required for authority). Unsurprisingly, the story he provides falls short of a general theory of what original obligations are and how they are generated – something we would want to really get to the bottom of the horizontal/vertical dispute or battle-of-intuitions.

      Though I certainly feel the intuitive pull (and appreciate the theoretical simplicity of) the view that all original obligations are generated and owed to individuals (on the model of something like respect or dignity due) or the purely “horizontal” model, I also feel the intuitive pull of the idea that some obligations are generated by and owed to society as a whole or the partially “vertical” model (authority is just a special case of this – and perhaps one that requires special conditions that address the value of or the obligation to respect individual autonomy). Of course, any such story should be told without any reference to society as some mystical entity or literal agent. I suspect that there are both “horizontal” and “vertical” strands of obligation (and authority) to weave together and square with one another.


      • I don’t have a problem with the notion, as such, of obligations to society; that’s not what I had in mind by vertical. But I think it’s a long way from there to justifying the authority of some people over other people.


        • Yes, I think this is right. That we owe things to society is not so controversial or distinctive (even perhaps if the obligations here are original or basic; the vague thought here was that this would yield a kind of normative collectivism that is at odds with anarchism – but this is totally wrong as we can imagine a world without authority in which is it primarily groups or collectives that owe things to each other). Maybe more important is the idea that: (a) individuals in a society owe things to society that they don’t owe to any individuals or to any private, non-society-like groups or institutions and (b) society has obligations and rights that are different from those of individuals or private, non-society-like groups and institutions. In this sense, society is special! This does not get us authority (in the traditional “vertical” sense), but it needs to be true if we are to have such authority and might be one of the more crucial differences between the pro- and anti-authority (in the traditional “vertical” sense) outlooks. There, I think that is better.


          • Individuals’ having duties to society not straightforwardly decomposable into duties to individuals is fine (at least potentially; it depends on the details).

            But society having duties not straightforwardly decomposable into duties of individuals is trickier, since society isn’t an agent, at least in any standard sense.

            Also, what are the boundaries of “society” here? Is there any reason to treat them as corresponding to membership in the same nation-state, as most theories of authority seem to want?


          • Agreed on first point. And agreed on second point as well – but I’m not inclined to deny such reduction (more inclined to deny that such a reduction constitutes an elimination rather than a vindication/revision mix, to use some old terminology). Finally, agreed regarding your concerns about whether the membership to a nation-state corresponds to membership to a society. I suspect that for this and other reasons, any true theory of political authority will be at least somewhat revisionary and uphold something more complex than what our folk theory of political authority would uphold.


  2. Michael, I’d think that if there is a prima facie presumption against the use of force, then one has a strategic obligation to show to potential intervening third-parties that one’s use of force is just in each retaliatory or compensatory use of it. That is, the obligation is from smart strategy among individuals and moves upward to mutual compact, explicit or tacit, then stands in the community. It need not originate radically from any top-down or group-down authority.


    • Thanks, Stephen. Seeing this through the lens of the obligation not to coerce (the relevant sorts of punishments being kinds of coercion) is quite helpful. One implicit assumption in Estlund’s argument is that, prior to the system of public justice being in place, one is permitted to punish in at least some coercive ways (Roderick disagrees). This normative permission (itself an exception to the anti-coercion principle) disappears once we have an adequate system of public justice in place – this is part of the pro-AVP intuition.

      It seems to me that this “negation of the justified-private-punishment exception to the obligation not to coerce” would be grounded primarily in something like vitally important improvements to collective states of a group (avoiding chaotic violence, say), not in the individualistic/dyadic respect-for-persons considerations that ground the defeasible prohibition on coercion. Maybe, at the end of the day, you need implicit or explicit consent to justify the obligation not to punish privately when there is an adequate public system of justice (or analogously for a similar principle concerning only judgment and compensation not punishment), but, for me anyway and I think for most of us, this “individualist” stance is more theory-driven than intuitive.


      • “One implicit assumption in Estlund’s argument is that, prior to the system of public justice being in place, one is permitted to punish in at least some coercive ways (Roderick disagrees).”

        To clarify: I disagree only if “punish” is understood in the ordinary sense. But if we translate “punish in at least some coercive ways” as “use coercion in defense of rights,” then I agree with Estlund (on THAT point, if on very few others).

        (I take it to be crucial to coercive punishment that the coercion be more than is needed simply to protect or restore the victim, and that the suffering of the perpetrator is an aim of the coercion and not merely a consequence.)

        Liked by 1 person

        • More a bibliographical note than a substantive contribution to the discussion: I don’t know if either of you ever saw Reason Papers’s mini-symposium on Ayn Rand on punishment. Here’s David Boonin’s contribution:

          Click to access rp_351_6.pdf

          And here’s mine, nominally a response to Boonin, but a very long one, intended as a sketch of Rand’s “theory,” as I saw it:

          Click to access rp_351_7.pdf

          I basically agree with Roderick on punishment, at least on retributive punishment as commonly conceived. I don’t agree with Boonin’s claim that the concept of punishment refers to retributive punishment as such, where retributive punishment requires the intentional infliction of harm, and don’t agree with his historical claim that there’s an overwhelming consensus in favor of this understanding of the concept of “punishment.” The position I end up attributing to Rand (and regard as plausible, though in need of further clarification and argument) is that punishment is a form of debt collection on a revisionary conception of “debt,” “payment,” and “trade.” But abstracting from the details (which I don’t cash out anyway), I agree that punishment as intentional infliction of harm for past wrongdoing is itself unjustified.

          Incidentally, I got a lot of shit from the Objectivist crowd when I defended my account of “Rand on punishment,” the claim being that by taking the intentional harm out of punishment, I was being soft on crime. No one seemed moved by my response: nowhere in the official Randian corpus does Miss Rand ever say that punishment must involve the infliction of intentional harm. Nor does intentional infliction of harm have a clear rationale in any identifiable Objectivist principle. Whereas (I argue) debt collection clearly does–in the trader principle.

          The other objection I got was that conceiving punishment as debt collection was inhumanly reductionist, i.e., didn’t reflect respect for the moral gravity of criminality because it reduced things to mere monetary considerations. But it’s Rand who “reduces” human interaction to “payment” (on a revisionary understanding of “payment”), and my point was that debt collection has to be understood in a similarly revisionary way. That didn’t cut much ice, either. This was back in the day when it was a thought crime of sorts to sound overly libertarian. But I wasn’t trying to sound libertarian. I was just trying to articulate what I thought her theory implied.

          A puzzle, or maybe just a query (mostly intended for Roderick): suppose that we’re living under consent-based political arrangements, whether anarchist or involving consent to a limited government. In that case, we all agree to be “punished” according to specified procedures for a specified set of rights violations. In that case, isn’t it fair to say that the punishment in question isn’t coercive? Ex hypothesi, we’ve agreed to it. So if I violate your rights, and the “authorities” (or whatever we want to call them) come to arrest me, we need a way of registering the fact that they compel me in the sense of flouting my desires without coercing me in the sense of violating my rights. Right? I mention the point because I’ve always found phrases like “coercive punishment” or “the coercive nature of punishment” confusing in contexts like these. (Roderick used the first phrase, but the second phrase is commonly used.)


        • A more substantive point: I’m inclined to think that the issue of referring disputes to third parties is a bit of red herring. Both of you (Roderick and Michael) write as though we have some generalized obligation to refer disputes to third parties, but I don’t see why we would, and it seems obvious to me that we often don’t.

          Take any dispute you have with any institution (putting aside the governmental ones for present purposes)–your workplace, a corporation, an HOA, a restaurant, a shop. Most institutions have internal dispute resolution services. These services or offices are designed to look like third parties, but legally speaking, they’re not third parties; they’re parts of, and beholden to, the institution that created them, where that institution is a party to the dispute. The assumption we make is that we can trust certain institutions to practice a certain degree of impartiality when it comes to disputes. I’m inclined to think that the vast majority of disputes are handled in this way, a fact that few find all that problematic. Indeed, people are conventionally condemned for being “overly litigious,” i.e., for demanding third party resolution to disputes that could have been handled by the parties themselves, i.e., by each party acting as judge in his own case.

          One way of looking at this practice is to say that we can trust these institutions to carve themselves up so that disputes are settled by “third parties” and the initial disputant is not literally “judge in her own case.” E.g., if I have a dispute with my Dean, it’s resolved by HR, where the director of HR is not my Dean, so that neither the Dean nor I are “judge in our own case.” But you could with equal validity say the reverse: if I have a dispute with my Dean, it’s resolved by HR, but both are members of the Felician University’s administration. So there’s a sense in which a dispute between me and Felician University’s administration is resolved by Felician University’s administration, which is judge in its own case, and has the final say over the resolution of the dispute. The difference between the two descriptions seems a matter of preference, not of substance.

          Granted, if internal dispute resolution doesn’t work, either one party has to give up on his claims, or the two parties have to appeal to some third party to resolve the dispute, but my point is, I don’t see that we have a general obligation to refer disputes to third parties.

          Liked by 1 person

          • Welcome back, Irfan! What you say seems right about the obligation to have disputes (or potential rights-violations) settled by third parties (who aim at justice) not being as general as Roderick perhaps makes it out to be. Perhaps the more general obligation is something like that of settling disputes in ways that count as just and that are acceptable to each party (if each party is reasonable enough and conditions are adequately propitious). This principle, unlike Roderick’s, can apply to governments consistent with AVP because it is left open how this obligation is cashed out for governments (when in dispute with citizens or institutions within its jurisdiction, other non-competing-jurisdiction governments, competing-jurisdiction governments, etc.). If this is right, we get a helpful framework for specifying just how and why the institution of government (and perhaps others) might be special in the ways required by AVP (and also various principles expressing sorts of authority that are special to government).


  3. Pingback: Nightcap | Notes On Liberty

Leave a Reply to mlyoung57 Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s