Talk of reparations has come back into common currency in American political discourse–meaning reparations to African Americans for the wrongs done to them since the beginnings of slavery. I don’t have a fully considered view on reparations (many of the arguments both for and against strike me as one-eyed), but I’ve both been surprised (and in another sense, not surprised) to hear libertarians insist so adamantly that libertarianism rules out reparations. Anyone who thinks this owes it to himself to read or re-read Robert Nozick’s Anarchy, State, and Utopia, if not cover to cover, then through the end of Part I, as I did on a recent plane ride.
I take it to be uncontroversial not only that Nozick is a libertarian, but that his view is paradigmatic of a certain brand of minimal state libertarianism. It’s also a matter of plain textual fact that Nozick defends a theory of property that not only includes but insists on a rectificatory component (“the property issue”), along with a theory of risk impositions that requires compensation for the imposition of unconsented-to risk (“the risk issue”). Put these two things together, and you get a view that’s surprisingly reparations-friendly, a fact that Nozick himself explicitly acknowledged, at least on the property issue (see the skeptical but still favorable reference to Boris Bittker’s The Case for Black Reparations on p. 344n2 of Anarchy).
There’s already a literature on the property issue, so I won’t belabor it (too much), but it’s relatively straightforward. On a Lockean-style entitlement theory of property like Nozick’s, justice in holdings is historical; one’s right to hold a piece of property depends on how one came to acquire it, up to and including morally relevant processes in the relatively distant past. If and to the extent that that process involved theft, that theft will affect the legitimacy of current holdings. If so, there is at least a pro tanto obligation to rectify the effects of past injustice by returning stolen property or compensating for ill-gotten past gains (see Anarchy, pp. 150-53).
Obviously, when it comes to large-scale historical injustice in the distant past, there are very difficult issues here that I don’t mean to minimize, and that may ultimately lack any feasible means of resolution.* That said, no one committed to an entitlement theory of property can deny the legitimacy of rectificatory justice as such; no one rejects it, or at least coherently could reject it, in the case of garden variety crimes, torts, and contract violations. In other words, if you accept the legitimacy of ordinary rectificatory measures, like replevin actions, civil damages, or compensation paid to victims of crime, you’re already committed as a matter of principle to rectificatory justice; the rest is a matter of haggling over the price and logistics. It probably goes without saying that a conception of law that lacked ordinary rectificatory measures would make little sense. (That said, I do think that Americo-centric theorists would benefit from taking a look at non-American cases before pronouncing too confidently on the merits or demerits of large-scale rectifications in the American case. But that’s in passing.)
Less commented on is the case of compensation for risk impositions. I won’t rehearse Nozick’s argument here, because it takes up maybe a hundred pages of the book (most of Part I, up to p. 146), and can’t easily be summarized, at least by me. It is, in fact, at the heart of his case for the minimal state. Granted, you may not accept the premises and/or conclusion of the argument, and may be tempted to regard it as, at best, pseudo-libertarian. But I suspect that many (maybe most) of the people who call themselves libertarian accept some version of a minimal state, and if so, either have to work within a Nozickian framework or explain why they don’t. This last point may end up being true of anarchists as well.
Anyway, the conclusion of the argument comes in the very last section of Part I, called “Preventive Restraint.”
The fact that the public must compensate persons it preventively restrains for the disadvantages it imposes upon them in those cases (if any) where it legitimately may so restrain them would presumably act as a serious check upon the public’s imposing of such restraints. We may condemn immediately any scheme of preventive restraints that does not include provisions for making such compensation in adequate amount. When combined with our conclusions in the preceding paragraph, this leaves little, if any scope for preventive restraint (Anarchy, p. 144).
The principle is: if and when you preventively restrain people through the coercion of legalized law enforcement, you impose certain un-consented to risks on them; if and to the extent that you do this, you must compensate them for your having done so. The only way to avoid the burdens of compensation would be to avoid preventively detaining people in the first place.
It’s a little late for avoiding the burdens of compensation by avoiding preventive detention: preventive detention is established law and a fact of life. And it’s a fact of life that falls disproportionately on some rather than others. Take the most minimal of preventive detentions in American criminal procedure, the so-called Terry stop, or reasonable suspicion stop. A Terry stop permits a law enforcement officer to detain someone on the basis of a reasonable, articulable, individualized suspicion that the target of the stop has committed a crime or crime-like offense.**
All that’s needed for a Terry stop is some factually and testimonially credible, articulable (not actually articulated), better-than-merely-hunchbased suspicion that a given person has committed a legal offense within the jurisdiction of the detaining officer. With this much evidence in hand, an officer is permitted to stop, question, and cursorily search the subject (“frisk”) for as long as it takes to dispel or confirm the suspicion that led to the stop. So construed, the stop can also permissibly serve as a pretext for further searches and seizures intended to establish probable cause, arrest, indictment, and/or prosecution. So a Terry stop can either mean that you’re detained for a few minutes, or detained for the rest of your life. It depends.
It’s a little known fact, but one well-known to people “in” law enforcement (whether on the enforcement side or the receiving end), that the “reasonability” involved in the Terry standard is highly and explicitly relativistic. For one thing, it’s relativized to the training and experience of law enforcement officers: the “reasonable” in “reasonable suspicion” is what’s reasonable to someone with the authority, training, and experience of a law enforcement officer, not what’s reasonable to the generic “reasonable person.” Beyond this, “reasonability” is relativized by place, as well. Act X, enacted in a “high crime area,” furnishes reasonable suspicion for a stop. The same act, enacted perhaps a mile or two away (outside of that area), does not. The two relativizations work hand-in-hand: law enforcement officers decide what sorts of neighborhoods qualify as “high crime areas”; law enforcement officers then decide what counts as “reasonable suspicion” within a high-crime neighborhood. In short, where you live determines what rights you have.
The law enforcement policy known as “Broken Windows Policing” was (and is) an attempt to institutionalize and intensify this sort of relativism. The point of it was (and is) to insist that reasonable suspicion not be treated as a single, consistent, uniform legal standard applicable anywhere and to anyone, but be treated as a matter of non-rational, non-codified “jhunchment” made at the discretion of a given officer at a given time and place, with no expectation of consistency across officers, times, or places.*** Whatever one thinks of it, Broken Windows Policing was and in many places remains extremely popular–not just in the applause-getting sense, but in the policy implementation sense: many people support it, and it’s been in place awhile.
Suppose we focus on the places where Broken Windows Policing has been implemented, by definition the places with the most aggressive approach to policing the lowest-level offenses. Almost (but not quite) by definition, this means that we’ll be focusing on Broken Windows Policing in high crime areas. Now suppose ex hypothesi that these “high crime areas” were and are largely black, and are relatively segregated from places where Broken Windows Policing is not applied. Suppose ex hypothesi that the brunt of this style of policing–of Terry stops treated as pretexts for arrest, whether misdemeanor or felony– falls disproportionately on black people. Then, if you accept Nozick’s Principle of Compensation, the predominantly black (legally innocent) targets of these preventive detentions should all be compensated. And “compensation” strikes me as a pretty close synonym of “reparations.” What’s in a word?
This isn’t some weird, convoluted implication of some peripheral and easily-ignored part of Nozick’s theory. It follows almost trivially from the principle I quoted above, and follows from a part of the theory that’s essential to its overall structure. Recall that Nozick goes out of his way to “condemn” any system of preventive restraint that doesn’t involve compensation. So the clearest case for reparations on libertarian grounds seems to come from one of the founding fathers of libertarianism.
How do we count the number of people to whom compensation is owed? And how do we calculate the amounts owed them? I don’t know. But I think it’s safe to say that both numbers are high–higher in the highest-crime areas and/or where Broken Windows Policing has been adopted, but high just about anywhere where Terry stops are performed and the subjects involved turn out to be innocent.****
How do we know that the persons owed reparations are all or even mostly black? Well, I can’t say that I know. You’d have to ask an expert. But I think it’s pretty plausible to think that black people are disproportionately numbered among those preventively detained and owed compensation. In fact, advocates of Broken Windows Policing often go out of their way to insist on its value to black neighborhoods, the claim being that those neighborhoods need it most because they suffer disproportionately from crime. So while Nozickian compensation wouldn’t precisely be “reparations to black people” under that description, it would amount, extensionally, to reparations to (lots of) black people and more to black people than to others. This might end up being a feature rather than a bug of Nozickian reparations, at least for people who get irritated at the idea of reparations specifically for black people under that description.
“What about the price tag? Where is this money going to come from, for God’s sake?” I figure it’d come from the same source as paid for the preventive restraints, but let’s slow down a bit. I haven’t put a price tag on these reparations because I haven’t even signed on to them in the first place. All I’ve said is that they follow from the adoption of Nozick’s brand of libertarianism. If you dislike the conclusion, feel free to evade it by rejecting the argument for it. In fact, I’m obliging enough to get you started.
One way to avoid the conclusion might be to argue that even if reparations are owed, they’ve all been paid out via the improved crime rate brought about by Broken Windows Policing. So maybe the preventive restraints have already ended up paying for themselves.
Maybe. But if you go this route, you’ll have a lot of conceptual and empirical work to do. For one thing, you’d have to monetize the lower crime rates brought about by Broken Windows Policing, and show that they balanced the would-be reparations payments you would otherwise have paid out. But then, advocates of Nozickian compensation have to monetize the harms/risks of preventive detention for purposes of compensation as well, so maybe this isn’t such a big deal (or maybe the two impossible tasks cancel one another out).
A bigger deal is that you’ll have to ignore the fact that the beneficiaries of the lower crime rates and the would-be beneficiaries of compensation aren’t necessarily the same people. It’s the latter who are owed compensation on Nozick’s view, not the former, and treating them as fungible or interchangeable is not exactly consistent with Nozick’s insistence on the separateness of persons or the dignity of the individual. So this is something that’ll somehow have to be resolved or else swept under the rug.
A yet bigger deal is that you’ll have to show that lower crime rates really did come about through Broken Windows Policing, a highly controversial claim that’s plausibly contested by lots of smart, highly qualified critics.*****
To my mind, though, the really fundamental problem is that at least in its original incarnation, Broken Windows Policing was explicitly conceived as a form of policing that superseded legality, constitutionality, and individual rights. There’s no getting around the fact that at least in its original version, Broken Windows Policing was explicitly intended to bypass the First and Fourth Amendments.
At root, Broken Windows Policing involves a crude, extra-legal form of consequentialism. Stated as a principle, it says:
Disorderly behavior is subject to preventive detention in a given community if and only if it is regarded by that community as signaling criminality.
There’s a lot buried in that formulation, but for now it’s enough to note that Broken Windows Policing explicitly criminalizes constitutionally protected behavior on the premise that a community’s associating a certain behavior-type with criminality de facto gives that behavior-type crime-like status, regardless of whether or not the behavior actually is a crime.
That may sound (or look) like a strawman, but it isn’t one. It really is what the architects of Broken Windows Policing were saying, what they meant, and often enough, how they were interpreted by police chiefs, police departments, and street-level patrol officers. In an astonishing passage, George Kelling and James Q. Wilson, the architects of Broken Windows Policing, write that
The prospect of a confrontation with an obstreporous teenager or drunken panhandler can be as fear-inducing for defenseless persons as the prospect of meeting an actual robber; indeed, to a defenseless person, the two kinds of confrontation are often indistinguishable [!].
Kelling and Wilson are literally saying that the two sets of offenders against public order ought to be treated in the same way. If defenseless people are prospectively as afraid of teenagers and panhandlers as they are of robbers, well then: all three groups should equally be stopped and frisked. Presumably, if this happens, fear will dissipate and order will prevail, as teenagers and panhandlers are taught the lesson that the police can permissibly violate the law in order to assuage the fears of non-teenagers and non-panhandlers.****** That teenagers and panhandlers might themselves qualify as “defenseless persons” and/or might feel fear, at least when confronting officers armed with clubs, revolvers, and live ammunition, does not seem to come up. On Nozick’s view, of course, those prospectively dangerous teenagers and panhandlers would be owed a wad of cash, paid for by the defenseless people who called the cops on them.
It’s an open question whether real-world Broken Windows Policing has consistently exemplified what its original architects envisioned for it, but then, it’s also possible that real-world Broken Windows Policing has exceeded the (rather hazy) limits they had in mind. A smart-ass might say that while Broken Windows Policing doesn’t precisely entail constitutional violations, it sure signals that some are in the neighborhood. It’s certainly more “suspicious” in that way than a lot of the people stopped and frisked according to its dictates.
So yes, you could try to argue that Broken Windows Policing cancels out the need for reparations on the grounds that it lowered the crime rate and thereby canceled out the would-be reparations. The hidden cost to this approach is that in making it, you have to treat rights as violable on crudely consequentialist grounds. You have to be willing to treat “disorderly” but otherwise non-criminal activities like begging, loitering, gambling, obstreporousness, being drunk, smelling bad, sleeping on park benches, teenagerhood and so on as predicates for Terry stops–stops designed for searching and seizing suspected criminals. You have to be willing to think and say that Terry stops apply as much to suspected criminals as to people not suspected of any criminality but just loosely associated with a conception of “disorder” that is itself loosely associated with fears of criminality. In short, you have to be OK with the idea that some people’s freedom ends where other peoples’ free associations begin.
This is a price that no libertarian, Nozickian or otherwise, could pay. Arguably, it’s a price that no one ought to pay, Nozickian or not, libertarian or not. If the choice we face is rights-subversive consequentialism on the Broken Windows Model or reparations, reparations start to seem like a bargain. As I read him, Nozick’s claim is that fundamentally, that is the choice we face. Reparations are the price we pay to avoid Broken Windows-type consequentialism.
Another way of evading reparations-on-Nozickian grounds might be to reject Nozick’s whole approach to things, and with it, the Principle of Compensation.******* Fair enough, but you can’t abolish the problem Nozick faced by rejecting his solution to it. Either you accept preventive detention or not. And either you accept individual rights or not. If you accept preventive detention and individual rights, and you reject Nozick’s solution to that problem, you’re still left with the problem of how to reconcile preventive detention with individual rights. You could, I suppose, accept preventive detention and reject rights, but that seems like an express train to totalitarianism. You could reject prevention detention and accept rights, but then you’d have to reconcile that combination with a plausible conception of policing, which I find hard to envision.
You could reject both preventive detention and rights, or perhaps reject the terms of the problem altogether, but though logically possible, this option leaves you with a lot of explaining to do (which I mean as an observation, not an objection). You could, finally, weaken your conception of rights so as to make their exercise perfectly consistent with the imposition of preventive restraints. But again, this solution would require a lot of explanation. Taken one way, it’s the key to all normative mysteries. Taken another, it’s the worst of all normative worlds. The point is, there are no easy solutions here.
What you can’t do is insist that reparations are un-libertarian–unless you want to say that Nozick circa 1974 wasn’t a libertarian or Anarchy, State, and Utopia is an un-libertarian book. Even as I write that, it occurs to me that someone will say both. Which brings to mind a different principle of compensation: if I had a nickel for every absurdity ever uttered, I’d be a rich man.
*E.g., Tyler Cowen, “Discounting and Restitution,” Philosophy & Public Affairs 26:2 (Spring 1997), pp. 168-85.
**Meaning, essentially, any legal offense enforced by the police, including crimes, non-criminal traffic violations, and municipal offenses or petty disorderly persons offenses. Nozick quasi-defines preventive detention as “all restrictions on individuals in order to lessen the risk that they will violate others’ rights” (Anarchy, p. 142). I assume that the “restrictions” in question are all rights-infringing boundary-crossings. I also assume that risk-minimization can permissibly sit alongside other aims consistent with risk-minimization. The point of a Terry stop is to minimize risk while facilitating an investigation into the subject’s suspicious content. While Nozick doesn’t mention the investigatory aim, I assume that Terry stops, arrests, and the like are still preventive detentions within the scope of of his discussion.
***”The scheme of the Fourth Amendment becomes meaningful only when it is assured that, at some point, the conduct of those charged with enforcing the laws can subjected to the more detached, neutral scrutiny of a judge who must evaluate the reasonableness of a particular search or seizure in light of the particular circumstances. And in making that assessment, it is imperative that the facts be judged against an objective standard: would the facts available to the officer at the moment of the search and seizure ‘warrant a man of reasonable caution in the belief that the action taken was appropriate’? Anything less would invite intrusions upon constitutionally guaranteed rights based on nothing more substantial than inarticulate hunches, a result this Court has consistently refused to sanction.” Terry vs. Ohio (1968), p. 22 in original pagination, internal footnotes and citations omitted.
****I don’t mean, of course, that Terry stops are the only kind of preventive detention that take place. I focus on them because I suspect that they’re the most common sort of preventive detention, and because they’re the easiest to discuss. For a striking discussion of the misuse of Terry stops in black neighborhoods, it’s worth plowing through the U.S. Department of Justice’s consent decree regarding the city of Newark, New Jersey, which found that fully three-quarters of the pedestrian stops done by the police on the city’s streets were unconstitutional, overwhelmingly imposed on its black male population. Also useful on the same topic (focused, again, on Terry stops of black males in Newark) is this PBS Frontline episode, “Policing the Police,” along with this companion piece in The New Yorker by Jelani Cobb.
*****Embarrassingly, in the paper embedded in the second hyperlink before the asterisks, the authors quote George Kelling, one of the architects of Broken Windows Policing, as admitting that the idea was merely a “speculation,” rhetorically shrugging his shoulders as to the empirical evidence behind it (“God only knows the truth”). The speculative quality of the hypothesis should have been clear from the way in which it was stated in its original formulation: the authors tell us explicitly in their 1982 Atlantic article that their claims not only flouted the law but flouted the very experimental evidence they used as a starting point for discussion. The puzzle here is how a hypothesis that did both managed to acquire so grand a reputation for symbolizing both the rule of law and crime reduction.
******Of course, the convenient, all-purpose tool of the “disorderly persons or petty disorderly persons offense” suggests that municipal ordinances can almost always be passed to give a legal rationale for these stops.
*******As per Eric Mack’s “Nozick on Unproductivity: The Unintended Consequences,” in Reading Nozick, ch. 9. Mack rejects Nozick’s Principle of Compensation, not his “whole approach to things.” But there are precedents for the latter move.