Start

LIBERTARIAN PAPERS VOL. 1, ART. NO. 21 (2009)

“TRIUNE”PROTECTION AND ITS IMPLICATIONS FOR

THE MINIMAL STATE

J*INGLEI HU

I 1N ANARCHY, STATE AND UTOPIA ( ASU ), Nozick proposes his model of the morally-justifiable minimal state. In the Preface to ASU, Nozick defines this state as “limited to the narrow functions of protection against force, theft, fraud, enforcement of contracts and so on,” and says that “any more extensive state will violate persons’ rights not to be forced to do certain things, and is unjustified.”

Given this definition, understanding the concept of “protection” is key to determining the boundaries of the minimal state. To elucidate this concept, this paper will examine individual rights as inviolable “constraints,” triune protection, and the resulting tension between the state’s protective power and individual rights.

Individual Rights as Inviolable Constraints

In Nozick’s libertarian theory, individual rights, whose legitimacy is presumed and indisputable, may not be violated. This is an imperative reiterated throughout Nozick’s book—from the Preface, “individuals have rights, and there are things no person or group may do to them (without violating their rights)” to the penultimate page, “the minimal state treats us as inviolable individuals.”

*Jinglei Hu (jh2807@columbia.edu) is a Ph.D. candidate in Political Philosophy at Peking University. He is currently a Visiting Scholar at Columbia University. He would like to thank the China Scholarship Council for the research grants, and Professors Shuifa Han, Thomas Pogge, and David Johnston for their kind arrangement for the visiting and their precious advice. The author is also grateful to Ms. Felicia Zimmerman’s much—valued suggestions on wordings and phrasings.

CITE THIS ARTICLE AS: Jinglei Hu, “‘Triune’ Protection and Its Implications for the Minimal State,” Libertarian Papers 1, 21 (2009). ONLINE AT: libertarianpapers.org. THIS ARTICLE IS subject toa Creative Commons Attribution 3.0 License (creativecommons.org/licenses).

1New York: Basic Books, 1974. All further references to this work appear parenthetically in the text.

1

To Nozick, the inviolate nature of individual rights is rigorous2—a “constraint” to and not merely a “goal” of the minimal state: “the side-constraint view forbids you to violate these moral constraints in the pursuit of your goals; whereas the view whose objective is to minimize the violation of these rights allows you to violate the rights in order to lessen their total violation in the society” (p. 29).

In contrast to Nozick’s view, if individual rights were “goals,” the result would be a “utilitarianism of rights.” A classic example of this phenomenon is the “trolley dilemma”: If the driver of the runaway trolley can avoid running over five people by switching tracks and, thereby, crushing only one individual, should he do so? A utilitarian of rights would answer, albeit begrudgingly, “Yes, the driver must minimize the number of deaths.” But this response signifies that “sometimes violating an innocent person’s right is justifiable.”

For Nozick, individuals are distinct agents and should be treated as separate ends.3 Consequently, sacrificing one person for the larger social good cannot be justified (pp. 31–33). Anticipating the potential loophole of utilitarianism of rights, Nozick proposes a strictly deontological framework for protecting rights, in which violations are not justified under any circumstances: “don’t violate constraints C. The rights of others determine the constraints upon your actions” (p. 29). The implication is that one’s rights should not be sacrificed for those of others, that is, rights should not be violated even for the sake of rights.

2I agree with Jan Narveson’s interpretation of Nozick’s definition of right, that “if doing x would deprive A of something to which A is entitled by right, then there are absolutely no circumstances in which it would be right to do x.” (The Libertarian Idea, Broadview Press, 2001, p. 54) However, since Nozick also evaded this hardly-defensible interpretation by saying “[T]he question of whether these side constraints are absolute, or whether they may be violated in order to avoid catastrophic moral horror, and if the latter, what the resulting structure might look like, is one I hope largely to avoid” (p. 30, footnote), I here use the term “rigorous” instead of “absolute” to anticipate possible objections. Indeed, as Narveson emphasized in his book, “Judith Jarvis Thomson points out that Nozick himself doesn’t really believe this [the absolute inviolability of right in any possible circumstances—J.H.], since he wonders, for example, whether we couldn’t maybe ‘inflict some slight discomfort’ on someone to save ‘excruciating suffering’ in 10,000 cows, leaving the reader with the impression that the answer is in the affirmative” (ibid.). However, I would echo Narveson’s complaint: “[I]t is hard to see how one could avoid that question [whether these side constraints are “absolute”], though; for we simply do not know what the view is until we have an answer to it” (ibid.).

3“Social atomism” in Karen Johnson’s words. “Government by Insurance Company:

The Antipolitical Philosophy of Robert Nozick,” The Western Political Quarterly, 29, no. 2. (June 1976), pp. 177–188.

“Triune” Protection—Retribution, Preemption, and Prevention

Given the inviolability of rights, the most basic function of minimal state is to protect these rights. But what does “protection” entail?

Using Nozick’s definition of “individual rights” as inviolable “side-constraints” and “trespassing” or “border-crossing” as overstepping these “side-constraints,” state “protection” could translate into an array of measures, namely, prevention (from potential trespassing), preemption (from imminent border-crossing), and retribution (after violations of rights). Because retribution is the least extensive and most justifiable, I will analyze this form or measure of protection first.

Retribution: After the Fact, Passive, and Inadequate

Nozick redresses the violation of rights through compensation and punishment. Accordingly, his model is “after the fact” because victims are compensated and trespassers are punished only after rights have been violated. This approach is also passive because it offers no preemptive or preventive measures to forestall such violations. Because of these attributes, this form of protection may be akin to retribution. As David Miller has summarized Nozick’s model, “the value of its services depends on the speed and efficiency with which it is able to track down rights-violators and oblige them to make restitution or provide compensation to their victims.”4

To many it is self-evidently right to punish trespassers and claim compensation. As Nozick points out, “in a state of nature an individual may himself enforce his rights, defend himself, exact compensation and punish (or at least try his best to do so).”5

While I fully embrace Nozick’s view that retribution is a legitimate way to redress a rights violation, I also share his conclusion that retribution is an inadequate form of protection. In this regard, he points to two drawbacks. First, some victims, such as those who are murdered, maimed, or psychologically threatened,6 can never be fully or appropriately compensated because, to Nozick, there is “incommensurability” between compensation and rights, and retribution or compensation relies on measurement. Second,

4David Miller, “The Justification of Political Authority,” in Robert Nozick, p. 20, edited by David Schmidtz, Cambridge University Press, 2002.

5And there is another sign of Nozick’s equating enforcement of rights with punishment and compensation also in p. 12, “in a state of nature a person may lack the power to enforce his rights; he may be unable to punish or exact compensation from a stronger adversary who has violated them.”

6Section “Fear and Prohibition,” pp. 65–70.

“retributive theory seems to allow failures of deterrence” (p. 61). In other words, rights are not protected before they are violated, only afterward.7 Relative to how a citizen of a minimal state can protect himself, Nozick replied with the query, “Must one wait until afterwards?” (p. 55).

Recognizing that such a limited form of protection is hardly “inspiring,” Nozick broadened the concept of protection to forestall acts posing an imminent threat8 to rights. Now, the protection Nozick is going to offer is well beyond David Miller’s summary.

Preemption

Nozick discussed “preemptive attack” at length.9 In doing so, he tried to find a justification for this brand of protection: “perhaps the principle is something like this: an act is not wrong and so cannot be prohibited if it is harmless without a further major decision to commit wrong (that is, if it would not be wrong if the agent was fixed unalterably against the further wrong decision); it can only be prohibited when it is a planned prelude to the further wrong act” (p. 127). In turn, Nozick envisioned three levels of preemption:

a stringent principle (S-principle hereafter) would hold that one may prohibit only the last wrong decision necessary to produce the wrong … more stringent (M-principle) yet would be a principle holding that one may prohibit only the passing of the last clear point at which the last wrong decision necessary to the wrong can be reversed. More latitude is given to prohibition by the following principle (hence it is a weaker principle against prohibition, W-principle hereafter): Prohibit only wrong decisions and acts on them (or dangerous acts requiring no further wrong decisions). One may not prohibit acts which are not based on decisions that are wrong.

The paragraphs that follow will explore the three levels of preemption in detail.

M-principle: Hard to Identify, Impossible to Accomplish

While theoretically plausible, the M-principle, the “more stringent” preemptive concept, cannot be applied in reality. This is because the “last clear point” of “the last wrong decision” is impossible to identify objectively

7Even if the punishment and compensation scheme could be designed to be so harsh as to deter all potential rational criminals, the “irrational” terrorists, like the Joker from Batman, who kills for fun and never cares about “pay-off,” would not be scared off.

8Here the border has not yet been crossed.

9Part I, Chapter 6, starting at p. 126.

and, even if it could be identified, “the passing of the last clear point at which the last wrong decision necessary to the wrong can be reversed” would be impossible to “intercept.” For example, the passing of the reversible “last clear point” of a shooting might be the moment just before the gunman pulls the trigger. Evidently, acting at this precise moment to stop the gunman—or any similar split-second movement—is unrealistic. However, the law enforcement agent has to “wait” for that particular irreversible time point to justify his preemptive attack. Even so, if the gunman is poised to shoot, the agent, most likely, will be unable to “intercept” this act. Moreover, such preemption would imply perfect knowledge and certainty of a potential or future violation. That is to say, it is difficult, if not impossible, to predict the “last clear point” at which the “last wrong decision” could be reversed.

Nozick might address this issue by saying that the last clear point is a point at which “the wrong can be reversed” and the preemption can be effective. If, as discussed in the preceding paragraph, an agent cannot intercept at “the last point,” this split second is not the “irreversible” moment defined by Nozick. However, if this is the case, how can the agent (or state) identify the last reversible point? Does a rights violator act according to an operational flow chart? Or do trespassers all act in a certain way? Let’s say that there are two enforcement agents, A and B, who, despite undergoing the same training, perform differently due to the nuances of their physical attributes and perspectives. Consequently, even if they are facing the same gunman, the last reversible point for agent A would be different from that for agent B. Agent A might be able to intercept the gunman’s pulling the trigger, but Agent B could only stop the gunman from drawing the gun. In other words, the two Agents would have different judgment of “the last irreversible point.” Even though an individual agent’s discretion is critical to M-principle preemption, Nozick have not addressed these issues.

In any case, if preemption at “the last point” is improbable, if not impossible, “waiting for” that point is not justified. Equally important, if the “last point” cannot be identified, any act involving this point cannot be considered legitimate.

S-principle: The Unknown Last Wrong Decision

Because of problems of objectively identifying and acting at the “last moment,” we can rule out applying the M-principle in the minimal state. Now, let us consider the “stringent” candidate, which justifies prohibiting “only the last wrong decision necessary to produce the wrong.”

Nozick did not state clearly whether the “last wrong decision” is a decision in the trespasser’s mind, which cannot be known by others, or a 6 LIBERTARIAN PAPERS 1, 21 (2009)

decision manifested as an act. If it is the former, the law enforcement agent cannot justify his preemption without proving the inscrutable psychological process of the trespasser in that particular split-second. If it is the latter, the agent would need to know if a particular act or decision is, in fact, the “last” (or nearly so), so that he would have time to intercept it. Because Nozick does not differentiate “wrong decision” from “wrong act,” I will do so. “Wrong decision” is the violator’s intention that can lead to a “wrong act,” while a “wrong act” is the actual action of violating a border or right.

As was true in M-principle preemption, in S-principle preemption, an agent would need to know in advance that an act, which is “necessary to produce the wrong,” is, indeed, the “last.” If this were possible, Nozick’s agents would be like the characters in Spielberg’s Minority Report. The “precogs” can see into the future and predict crimes, allowing the elite law enforcing squad, “Precrime,” to use this knowledge to take preemptive action.

Is this sci-fi scenario what was in Nozick’s mind? Hardly. If foreknowledge is required to justify preemptive attack, then perhaps in the next millennium a minimal state with preemptive capacity will still be a sweet dream. And even after that millennium, discretion would still be necessary to a unit like Precrime.

W-principle: Conceded with Qualifications

To this point we have examined the S-principle and the M-principle and rebutted both because they are difficult to define and impossible to achieve. Only the W-principle (“weak” principle) of preemption remains. This principle involves prohibiting only wrong decisions and acts on them.

Let us make a comparison. The S-principle reads “one may prohibit only the last wrong decision necessary to produce the wrong,” while the W-principle reads “prohibit only wrong decisions and acts on them.” It appears that the latter legitimizes prohibiting the wrong decisions and acts on them even if they are not the last one(s). Accordingly, the penultimate wrong decision could be justifiably prohibited.

However, the W-principle of preemption raises three questions. First, if we do not know—have no foreknowledge about—the last point in a decision or act, how can we define the penultimate point or, for that matter, any earlier point? Second, how can we identify a “wrong decision and act on it”? In this regard, Nozick stated: “an act is not wrong and so cannot be prohibited if it is harmless without a further major decision to commit wrong; it can only be prohibited when it is a planned prelude to the further wrong “TRIUNE” PROTECTION/IMPLICATIONS FOR THE MINIMAL STATE7

act” (p. 127). That is to say, “a planned prelude” or “wrong decision” signifies intent to commit a “wrong act.”

And third, how can we prohibit intent (“intentions of border-crossing”)? Even good people harbor such “intentions,” like robbing a bank, or even killing some despicable people. But, such thoughts rarely lead to “wrong acts.” In any event, absent “thought police,” how could we ever possibly know beforehand whether a wrong thought could “lead to” a bad act? Furthermore, the introduction of such intrusive thought police would be incompatible with Nozick’s minimal state.

On the other hand, in support of the W-principle, people sometimes voice their “wrong intentions,” and therefore intentions become known and subsequent acts can be forbidden. Indeed, terrorists sometimes make videotaped public statements before they launch the attacks. However, once “wrong intentions” are publicly verbalized, they become “border-crossing acts.” That is because such an intention of threat violates others’ peace of mind and also challenges the public order. According to Nozick, even if the announcer of these threats was to compensate others for the fear caused by his announcement, the fear-producing acts should be “prohibited.”10

The W-principle should be conceded11 as the only potentially viable preemption concept. But, to do so, “imminent”12 should be used in lieu of “last” to describe the emergency against which a preemptive attack would be justified.

“Imminent” (or “impending” or “immediate”) defines an approximate perception, unlike the term “last,” which defines an accurate fact. As a result, unlike the S-and M-principles of preemption, the W-principle would provide for agent discretion, without which preemptive attack cannot be conducted or justified. However, due to the challenges of “when” and “how” and

10“…even some acts that can be compensated for may be prohibited. Among those acts that can be compensated for, some arouse fear” (p. 66); “Some things we would fear, even knowing we shall be compensated fully for their happening or being done to us. To avoid such general apprehension and fear, these acts are prohibited and made punishable” (p. 66); “the actual phenomenon of fear of certain acts, even by those who know they will receive full compensation if the acts are done to them, shows why we prohibit them” (p. 69); also compare to the text on p. 68.

11After all, his purpose was to rule out the possibility of exploiting “preemptive attack” as a seemingly legitimate excuse to launch an aggressive attack against others who are not posing immediate threat (p. 126, 1st paragraph; p128, 2nd paragraph). And this purpose remains largely unchallenged.

12Or “impending” or “immediate,” terms that are more frequently used to discuss “preemptive attack” in the discipline of International Relations, and will be briefly revisited soon.

variability among agents, even discretion is unreliable as a source of protection. Hence, Nozick envisioned stronger protective measures— ”preventive restraints”—to compensate for the inadequacy of preemption.

Prevention: Extensive Protection

Nozick devoted one section (Part I, Chapter 6, Section 6) to “Preventive restraints.” Though he touched on “prevention” earlier (Chapter 4 & Chapter 5), here he focused on this approach to protection, which is more proactive and precautionary than preemption.

Nozick defined this notion as follows on page 142:

the notion should be widened to include all restrictions on individuals in order to lessen the risk that they will violate others’ rights; call this widened notion ‘preventive restraints.’ Included under this would be requiring some individuals to report to an official once a week (as if they were on parole), forbidding some individuals from being in certain places at certain hours, gun control laws… [P]reventive detention would encompass imprisoning someone, not for any crime he has committed, but because it is predicted of him that the probability is significantly higher than normal that he will commit a crime.

Prevention provides more extensive protection than “preemption” because it takes into account risks that do not pose an immediate threat. For example, prevention would involve imprisoning someone who has a significantly higher than normal probability of committing a crime, such as a repeat offender (like the Joker).

While retribution is an after the fact and passive approach to protection, the distinction between prevention and preemption is less clear. Both approaches are beforehand efforts to avoid boundary-crossing acts from happening. In this context, Nozick admits, “I do not know if preventive restraints can be distinguished, on grounds of justice, from other similar danger-reducing prohibitions which are fundamental to legal systems. Perhaps we are helped by our discussion early” (pp. 142, 143). To differentiate preemption from protection, we must analyze the different phases of the violation of a right: preparation, initiation, and execution.13

13Thanks to Neil Levine for his suggested wording and clarification. The analysis of phases of right-violation was also based upon the analogy to the definition of a “crime against peace” in the discipline of International Law, according to whose rule of thumb, a crime against peace refers to “planning, preparation, initiation, or waging of wars of aggression, or a war in violation of international treaties, agreements or assurances, or participation in a common plan or conspiracy for the accomplishment of any of the “TRIUNE” PROTECTION/IMPLICATIONS FOR THE MINIMAL STATE9

Preparation covers all the activities preceding the final decision to violate a right; initiation is the final decision to commit a violation and the subsequent acts before the rights have actually been violated; and execution is the last phase and, as such, the completion of a violation of a right, regardless of whether or not the result of this act satisfies the trespasser’s intent.

A case in point might be helpful. Imagine, in the Wild West of the minimal state, an aggressive cowboy draws his six-shooter, aims it at you, and then pulls the trigger in order to kill you. The cowboy’s act of drawing his gun—along with purchasing and loading it—would be considered part of preparation.

At this point, he still has a choice of withdrawal from aiming at you, but if he chooses to aim at you, he is initiating the attack, which then can be completed by pulling the trigger, namely, execution. Whether you are shot to death or the cowboy missed his target is irrelevant here, for either way, your rights have been violated as a result of the execution.

With the preceding example in mind, we now can distinguish prevention and preemption: the former refers to intervening in an individual’s preparation to violate a right, while the latter means intercepting this person’s initiation or execution of such an act. This preceding distinction has a profound implication: if “preventive restraints” that involve intervening during the preparation of a right’s violation are justifiable, preemptive measures, which involve intervening in the later phases of a right’s violation, do not require separate justification. In fact, providing such justification would be redundant.

Returning to the Wild West example, using a preventive approach to protection, the cowboy can be prohibited from owning a gun (even be detained)14 if he is identified (by empirical evidence, probability calculation and discretion) as a probable violator of rights. In contrast, a preemptive approach only stops the cowboy from shooting another person. It would not prevent him from carrying a gun.

As mentioned already, if “preventive restraints” are justifiable, preemptive measures, which involve subsequent intervention, do not require separate justification. But, even though providing such justification would be

foregoing,” Nicolas Werth, Karel Bartošek, Jean-Louis Panné, Jean-Louis Margolin, Andrzej Paczkowski, Stéphane Courtois, The Black Book of Communism: Crimes, Terror, Repression, Harvard University Press, 1999, p. 5. And Nozick’s elaboration of preemptive attack also partially relies on the analogy to preemptive wars between countries, see page 126.

14Even in an extensive state like United States, qualified individuals are able to legally carry firearms in many areas.

redundant, Nozick does so, possibly because he did not originally plan to justify preventive protection. He may have thought that preemption would suffice but later concluded that, in some cases, prevention would be necessary.

From another perspective, Nozick tried to justify preemption because he recognized the inherent tension between active preemption and passive self-defense. As a result, he strove to prove that preemptive attack is not threatening side-constraints but protecting them.

Lastly, even in its broadest interpretation (the W-principle), Nozick limited preemption significantly. Consequently, he may have realized the need for both a separate justification as well as more extensive measures.

The Tension Between the State’s Protective “Power” and Individual

“Rights”

The final section of this essay will discuss how the minimal state provides protection. Accordingly, we will first examine preemption, then prevention, and lastly “triune” protection.

The Tension Between Preemption and Side-Constraints

Earlier we conceded that Nozick’s W-principle might be the optimum choice among the three forms (S, M, and W) of preemption. Now we will reevaluate this decision. W-principle only justifies the preemptive attack against imminent border-crossing acts, with the interpretation of “imminent” being left to enforcement agent’s “discretion.” Such discretion might be based on intuition, instinct, a rule of thumb, or actual experience. Regardless of the basis of an agent’s discretion, individual rights are subject to his judgment and interpretation of state power and, therefore are not as “inviolable” as Nozick claimed them to be.

Nozick promised to treat rights as inviolable side-constraints that predetermine the scope of state power. In fact, Nozick viewed rights as so rigid that they could not even be considered goals that could be maximized or subject to negotiation. Moreover, rights are “strongly discretionary,” as Leslie Pickering Francis and John Gregory Francis emphasized, “by which we mean that someone who possesses a right to something may do with it what he sees fit, except for using it to violate the rights of others.”15 Karen Johnson echoed this sentiment: “it seems that I ought to be able to do with it as I

15“Nozick’s Theory of Rights, A Critical Assessment,” Leslie Pickering Francis and

John Gregory Francis, Western Political Quarterly, Vol. 29, No. 4. (Dec., 1976), pp. 634–44. “TRIUNE” PROTECTION/IMPLICATIONS FOR THE MINIMAL STATE11

please—to use it, give it away, sell it, destroy it” and “once someone has a right, the claims of others are curtailed.”16 But with the introduction of the W-principle of preemptive attack, side-constraints rights or side-constraints become subject to the “discretion” or interpretation of the state power. And in turn, the state power become the “constraints” of the “side-constraints,” and curtail the rights-holder’s claims. 17

Nozick could argue that individuals have no right to risky or dangerous acts and that agents preempt risky acts. Even so, whether or not a future act is risky depends on the agents’ discretion. Thus, there is tension between preemption (as exercised by an agent using his discretion) and individual rights.

Moreover, as we have already established, “discretion” is not always reliable nor fair, as an individual may be attacked by an agent whose sole justification is his “right to judge.” According to Nozick, “a person may resist, in self-defense, if others18 try to apply to him an unreliable or unfair procedure of justice”(ASU, p. 102), but “the dominant protective association may19 reserve for itself the right to judge any procedure of justice to be applied to its clients”(ASU, p. 101). That is to say, whether a preemptive attack is legitimate or not would be up to the state. As a result, there is also tension between the state’s “right to judge and apply” and the individual’s right to resist.

Although the inviolability of rights implies that there is a “boundary”

between these rights and state power, we do not know the “width” of this boundary. Nonetheless, the boundary is not determined by individual rights alone, but rather by individual rights and state power acting concurrently or, if the “power of discretion” is abused, by state power acting alone. But, even without such abuse of power, there is substantial tension between the state’s preemptive protection and individual rights.

As discussed in the next section, Nozick’s “preventive restraints”

further expands the concept of “protection” so that the distinction between the minimal state and a minimally extensive state is blurred.

16Ibid.

17Despite Nozick’s promise that “no moral balancing act can take place among us; there is no moral outweighing of one of our lives by others so as to lead to a greater overall social good” (p. 33).

18Since this is a general claim, here the “others” should include the agents, who by nature cannot overcome the limitation of knowledge and moral misjudgment. And this interpretation is also in harmony with Nozickean conception of individual rights.

19David Miller interprets the “may” as “is morally permitted to,” which I think is pertinent (p. 24).

The Tug-of-war Between Restraints and Constraints

Nozick’s attitude toward preventive restraints is variable. For example, regarding these restraints, he anticipated the objection that the restraints “prohibit before the fact activities which though dangerous may turn out to be harmless” (p. 142). But elsewhere, he emphasized, that even if the prohibition is permissible, a compensation must be paid to those prohibited (p. 143), and “[I]t will be almost impossible for the public to provide compensation for the disadvantages imposed upon someone who is incarcerated as a preventive restraints” (ibid). However, he then “mentions”20 a potentially acceptable compensation plan (“setting aside a pleasant area for such persons”) but chooses to evade the question, “I do not discuss here the details of such a scheme” (p. 144). Given Nozick’s apparent volatility, he may or may not advocate preventive restraints, although he does envision the possibility of a tentative prevention plan with a compensation scheme (p. 145), and argues for it (pp. 145–46), claiming “it seems to be the correct position that fits the (moral) vector resultant of the opposing weighty considerations” (p. 146).

Notwithstanding Nozick’s variability on preventive restraints, they are, evidently, broader in scope than preemptive attacks, in that preventive restraints even forbid acts when individuals are not thought to be imminent danger but merely acts that might (or might not) increase the risk21 of potential border-crossing. For example, any man could be a potential rapist and, therefore, could pose a “risk” to women. In a large society, it is unclear which men have a higher likelihood of raping than the “standard-rape-probability.” For the sake of argument, let us assume that the probability is known and that agents can easily restrain those posing the highest risks. The principle of “preventive restraints” prescribes such action “in order to lessen the risk that they will violate others’ rights.”

This “legitimate” enforcement raises two fundamental questions. Let us put the easier one first: if law enforcement against individuals is justified even when they pose no direct threat to side-constraints, how inviolable are these individuals’ rights?

20He was so conscientious as to refuse to use the word “propose,” p. 144, “[F]or I mention resort detention centers not to propose them, but to show the sort of things proponents of preventive detention must think about and be willing to countenance and pay for.”

21“Risk” is a tricky term, to which Nozick devoted many pages. “TRIUNE” PROTECTION/IMPLICATIONS FOR THE MINIMAL STATE13

Question 1: How Inviolable Could Rights Be?

This question is closely linked to the tension between preemptive attack and side-constraints discussed in the previous section. This tension reflects that the justification for preemptive attack has bent the inviolability, softened the constraints, and forced the presupposed rights to share with state power some preoccupied supremacy. In other words, preemption predicated on the discretion of an enforcement agent has challenged the inviolability of side-constraints, which was stipulated in first several chapters of ASU. Broader preventive restraints further diminish the inviolability of side-constraints.

While preventive restraints are applied in the real world—even in the most liberal states—these states do not presuppose rights as rigidly inviolable side-constraints as Nozick did. Rather, these states might well be proponents of the “Utilitarianism of Rights,” which Nozick firmly opposed. Moreover, as Karen Johnson pointed out, due to the “wholly coercive” nature of state, a minimal state empowered with prevention restraints could make rights even more vulnerable, “its unity might make it potentially more dangerous than a more complex state. This would be especially likely in the context of Nozick’s social atomism. Even if people would not voluntarily cooperate with an aggressive agency, they would be unlikely to take concerted action to protect themselves against it. Being already divided, they would probably be easily conquered.”22

That Nozick “mentions” compensation for the victim of preventive detention shows that he realizes that, should restraints be permitted, the victim would incur a “loss.” If a person’s being physically detained can be justified due to his higher-than-normal probability of committing a crime and if Nozick’s seemingly uncompromising deontological view holds, we would be in an predicament: protecting some people’s some rights requires violating others’ other rights. If preemptive attack has challenged the side-constraints, preventive restraints are pushing the resilience of rights’ to their maximum.

Question 2: The Difference Between Constraints and Goals

Legitimate enforcement raises a second question, which is more difficult than the first. In this regard, let us compare the two sentences in ASU:

1. “[T]he side-constraint view forbids you to violate these moral constraints in the pursuit of your goals; whereas the view whose objective is to minimize the violation of these rights allows you

22Ibid.

to violate the rights in order to lessen their total violation in the society.” (p. 29)

2. “[T]he notion should be widened to include all restrictions on individuals in order to lessen the risk that they will violate others’ rights; call this widened notion ‘preventive restraints.’” (p. 142)

Considering the words in boldface, what is the difference between “lessen violation” and “lessen the risk of violation”? At first glance, the former appears to mean reducing the “quantity” of violations, while the latter may connote reducing the “probability” (ranging from 0 to 1) of violation.

A violation can also be defined as “the fact or reality of a violation.” Hence, when the probability of violation reaches 1 (or 100%), the potentiality of violation becomes the reality of violation. Regardless of how violation is differentiated from the probability of violation, the potentiality of violation is the phase of “violation development” prior to the fact or reality of violation.

Let us imagine an axis with increasing probability of violation from left to right or 0 to 1. “To lessen the risk of violation” is even more ambitious a goal than “to lessen the violation,” for the risk of violation cannot be reduced without first reducing the reality of violation from the far right of the axis.

In sentence 1, Nozick makes a significant distinction between side-constraint view and goal-directed view of rights and advocates the former. However, in sentence 2, Nozick implies the possibility of essentially maximizing rights and rights-protection as goals. Given this bias toward “goals” over “rights,” prevention restraints is both a violation-minimization scheme as well as a risk-reduction scheme. Is this not “utilitarianism of rights”?23 “[F]or suppose some condition about minimizing the total (weighted) amount of violations of rights is built into the desirable end state to be achieved. We then would have something like a ‘utilitarianism of rights’; violations of rights (to be minimized) merely would replace the total happiness as the relevant end state in the utilitarian structure“ (p. 28). Sensing an even worse version of “utilitarianism,” Nozick did not expound upon how to legitimize preventive restraints.

To justify preventive restraints, Nozick would have to broaden the concept of legitimate proactive protection to include not only preemptive attack but also precautionary measures, such as quarantining people prior to

23Serious analysis of sections like “risk,” “the principle of compensation,” “fear and prohibition,” and “why not always prohibit?” in Part I might well reveal that they could be consequentialist argument for rights or rights protection. But this is beyond the scope of this essay.

wrongdoings or setting up segregated habitats according to people’s inclination to cross borders.24 We have witnessed both quarantines of “dangerous people,” even in liberal countries, and also habitat segregation. But, if Nozick has modeled a “minimal state,” why would he condone preventive restraints?

Prevention, Prohibition and Triune Protection

Even though Nozick evades the legitimacy of preventive restraints, he implies that a minimal level of preventive restraints is inevitable within the state’s protective services.

First, he admitted that “preventive detention” or “preventive restraints” is related to the extensive protection that the ultraminimal state must provide, “even for those who do not pay.” (p. 142) The extensive protection “required” by the ultraminimal state (U-state) is designed to prohibit risky and unreliable private enforcement.25 Accordingly, such extensive protection will also be incorporated in the minimal state, which has more expansive powers than the U-state. Namely, prohibition of risky acts is an extensive protection that the minimal state must provide.

Now, what is the difference between preventive restraints and prohibition of risky acts? As long as both refer to blocking acts that are inclined to border-crossing, they are only different in terms of degrees. Sometimes, preventive restraint is treated as a subcategory of prohibition: “even if preventive restraint cannot be distinguished on grounds of justice from the similar prohibitions underlying legal systems, and if the risk of danger is significant enough to make intervening via prohibition permissible, still, those prohibiting in order to gain increased security for themselves must compensate those prohibited (who well might not actually harm anyone) for the disadvantages imposed upon them by the prohibitions” (p. 143; emphasis in the original). On other occasions the former is defined by the latter, “if such preventive restraints are unjust this cannot be because they prohibit before the fact activities which though dangerous may turn out to be harmless” (p. 142), or the latter derived from the former, “for an enforceable legal system that includes prohibitions on private enforcement of justice is itself based upon

24On pp. 142–44, Nozick mentions “forbidding some individuals from being in certain places at certain hours,” “imprisoning someone…because it is predicted of him that the probability is significantly higher than normal that he will commit a crime,” and he also suggested “curfews upon some persons and specific restrictions on their activities” or “setting aside a pleasant area for such (dangerous) persons” as “detention center” to prohibit dangerous people “from living among others in the wider society.”

25 Pp. 88–90, pp. 110–13, pp. 96–110, pp. 73–87.

preventive considerations” (ibid.; emphasis in the original). Indeed, even Nozick himself is not sure if he can differentiate the two: “[A]re there grounds for condemning preventive restraints as unjust that do not apply as strongly also to the prohibitions upon private justice that underlie the existence of every state’s legal system? I do not know if preventive restraints can be distinguished, on grounds of justice, from other similar danger-reducing prohibitions which are fundamental to legal systems.” (pp. 142–43)

What we can be sure about is:

1. Some prohibitions are necessary and permissible to the minimal state; and

2. Preventive restraints cover a wide range of prohibitive measures, some of which are virtually indistinguishable from some permissible and necessary prohibitions.

It follows that some preventive restraints are not only permissible in but also essential to the minimal state.

Why would Nozick permit some extensive approaches to protection and believe they “are fundamental to legal systems” (p. 143)? Because “risky actions” cover activities ranging from driving a car to making an atomic bomb, and Nozick is well aware of that, “Since an enormous number of actions do increase risk to others, a society which prohibited such uncovered actions would ill fit a picture of a free society as one embodying a presumption in favor of liberty”(p. 78). As a result, Nozick divided risky actions according to their degree of risk, thereby allowing some of these actions (like driving) but prohibiting others. However, even among the actions that would be prohibited, there are significant differences. For example, prohibiting jaywalking is very different from prohibiting nuclear aggression. After the fact punishment or forestalling on the spot might be sufficient to deter potential jaywalkers and recover the sense of security of the society, but that could never be enough to “prohibit” a nuclear threat. Along the same lines of classifying the risky actions, he developed (he had to) a wide spectrum of countermeasures that are categorized into retribution, preemption, and preventive restraints (with compensations26). The resulting

26The classification could be a daunting task: “to arrive at an acceptable principle of compensation, we must delimit the class of actions covered by the claim” (p. 81). Nozick then proposed a principle of classifying actions. But he immediately recognized “this possibility of diverse descriptions of actions prevents easy application of the principle as stated” (p. 82). On the one hand, it is absurd to forbid an act as “risky” as “driving”; on the other hand, it is ridiculous to compensate a person for prohibiting him from making atomic bombs.

structure, pyramid-like “protection,” has retribution as its base, preemption in the middle, and the rare preventive restraints at the top.

Only this triune form of protection can address the issue that retribution “seems to allow failures of deterrence” (p. 61); only this protection package could act beforehand against extremely dangerous terrorists such as Osama bin Laden; only this protective scheme could provide enough sense of security for the citizenry of minimal state, thereby being “inspiring.”

However, while triune protection seems to be able to ward off risky actions taken by individuals, it involves risky actions taken by the state that might itself cross the border. As Karen Johnson noted on ASU: “if the essence of the state is force, then governmental action is by definition coercive and, unless kept within very narrow limits, an invasion of individual rights: a form of aggression against persons whom the state is supposed to protect.”27 As our analysis shows, Nozick, albeit reluctantly, appears to endorse preventive restraints, hence he would open the door for the intrusion of state power’ (even if it is limited and compensated) to rights, in the name of “protection.” This way, Nozick seems to have offered two guarantees at the same time, one to individuals, the other to the state, and none is more legitimate than the other. As distinct theories, Nozick’s concept of “right” and that of “protection” are inspiring. But because he rules out “the utilitarianism of right,” which could serve to reconcile the conflicts between right and power, he cannot make these concepts work in tandem.

In the beginning, Nozick defines the minimal state as “limited to the narrow functions of protection against force, theft, fraud, enforcement of contracts and so on,” and “any more extensive state will violate persons’ rights not to be forced to do certain things, and is unjustified.” However, Nozick’s protection concept is more extensive than he intended, in that it not only guards against border-crossings but also against impending border-crossings, and even against the non-imminent risky actions that may or may not lead to an impending threat to rights. Given its expansiveness, the triune protection also defines the limits of rights, from which it is supposed to derive. In sum, the minimal state’s role of protection was much broader than Nozick originally envisioned and, therefore, in conflict with the rigid individual rights he presumed.

27Ibid.

About this edition

This EPUB was produced automatically from the PDF named below. The text is the publisher's; the layout was reflowed for screen reading, so page breaks, line breaks and hyphenation differ from the original.

Title: “Triune" Protection and its Implications for the Minimal State

Author: Jinglei Hu

First published: 2009

Series: Libertarian Papers, volume 1

Licence: Licensed under Creative Commons Attribution 3.0. https://creativecommons.org/licenses/by/3.0/

Source file: Triune_ Protection and its Implications for the Minimal State.pdf

Source page: https://freecapitalists.org/books/triune-protection-and-its-implications-for-the-minimal-state/

Converted: 2026-08-07 by The Liberty Archive (freecapitalists.org), converter v5