Sunday, June 04, 2006

Positive and Negative Rights

[A draft of my latest essay, minus the footnotes. Comments/suggestions welcome!]

Rights claims may be classified according to whether they make ‘positive’ demands on another’s actions, or – in case of ‘negative’ rights – merely require others to abstain from harmful interference. Human rights are often considered to be purely negative rights against political oppression. Refraining from oppressive actions thus suffices to satisfy the corresponding negative duties. On this “libertarian” view, human rights simply ask that you leave the rights-bearer alone. However, the desperately poor do not obviously suffer from a lack of being ignored. Some may require positive action on the part of others in order to secure their vital interests and continued survival. A purely negative interpretation of the “right to life” thus begins to look like an empty formalism, insufficient to protect the vital interests of persons. Motivated by this problem, the present essay examines the distinction between positive and negative rights, and explores some considerations that may recommend its deflation.

The duties associated with negative rights are typically thought to be far less demanding than positive ones. A negative duty acts as a side-constraint on action, effectively picking out one act and saying to the agent, “you may do anything else, but not that.” A positive duty, by contrast, would seem to impinge much more on the agent’s liberty by insisting that they carry out some particular action. (“You must do this, and not anything else instead.”) They also seem more costly. Intuitively, it doesn’t cost the government anything to respect our civil rights – they merely forsake the opportunity to benefit from exploitation. But if ‘subsistence rights’ require governments to provide for the basic needs of rights-bearers then that would seem a more significant burden. Further, human rights are typically considered to properly constrain government power. Critics may argue that subsistence rights would instead serve to increase the power of governments against their subjects, and are thus antithetical to the guiding ideal of human rights.

Matters are not so simple, however. To stand idly by as a murderer attacked would disrespect the victim’s (negative) right to life. If negative rights are to be significant, then there must be a positive duty to protect and uphold them. Applying this insight to the political sphere, Thomas Pogge argues that official indifference towards private rights violations is one way for a state to exhibit the “official disrespect” constitutive of a human rights violation. A just government cannot neglect the business of law enforcement, and so publicly funded institutions – such as a police force and courts of law – are needed to protect the rights of citizens. Hence Holmes and Sunstein observe that “all legally enforced rights are necessarily positive rights.” They entail duties of positive action, and not mere non-interference. So we are naturally lead from a minimal commitment to negative rights to a more substantial commitment to positive duties. This does not entirely collapse the distinction, however, for as Gewirth notes, “the ground or justification for the positive assistance in question is to see to it that potential offenders refrain from the prohibited actions.” It remains to be seen whether the libertarian is further committed to recognizing subsistence rights, though at least we have begun to narrow the gap.

We can make further progress by noting that we do not live in isolation, and that almost all of our actions affect – or “interfere with” – others to some extent. This means that focussing solely on negative duties will not actually restrict the scope of our obligations all that much. Compare J.S. Mill’s famous harm principle: “the only purpose for which power may be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.” The obvious objection arises that any action can be broadly interpreted as causing some harm to another. This is no great problem for Mill himself, as he can appeal to indirect utilitarianism to ground the notion of “certain interests, which… ought to be considered as rights.” But deontic libertarians have no such foundation upon which to make a pragmatic distinction here. The position under discussion instead treats the notion of ‘coercive interference’ as fundamental. They rule out positive rights as a matter of principle, so it suffices for my purposes to show that the distinctions they wish to draw here are not deeply principled ones. In particular, I will argue that initial acquisition of property, the ongoing enforcement of property rights, and our participation in the global institutional order all involve coercive imposition of harms, thus violating what libertarians should acknowledge as negative duties of non-interference. Importantly, libertarians hold that a breach of negative duty gives rise to a positive duty of rectification. So if there are negative rights that we cannot help but violate, then we likewise cannot avoid the burden of compensatory positive duties. I will argue that this state of affairs obtains, and effectively entails a duty to institute positive subsistence rights.

Consider the libertarian problem of ‘initial acquisition’: whenever one claims a property right over previously unowned resources, they are peremptorily excluding others from its use. Indeed, though this is too rarely recognized, libertarian principles of autonomy and non-interference would seem to require that affected individuals “have a veto over appropriations which exclude [them] from the commons.” By depriving others of what they would otherwise have access to, the appropriating action constitutes a kind of harmful interference, and so – by the libertarian’s own lights – the actor ought to recompense them accordingly.

This problem becomes all the more pressing when we consider intergenerational justice. There is no way that appropriations of unrenewable natural resources – including land itself – can meet the Lockean proviso of leaving “enough and as good” when we take into account indefinitely many future individuals. Hence the negative duty not to harm others will require one to either abstain from appropriating natural resources, or else adequately compensate those who are – or will be – left propertyless. The most plausible way to achieve such ongoing large-scale rectification would be through institutions of redistributive taxation, such as a basic income guarantee. Such positive welfare schemes are justified – indeed, mandatory – on libertarian grounds because they compensate individuals for the harms imposed on them by their initial exclusion from the appropriation and distribution of common natural wealth.

Even after the establishment of property rights, their enforcement constitutes a continuing form of interference. Not all interference is unjustified, of course. One may rightly pin down a violent offender in order to protect the innocent, for example. So we will need to return to the question of whether this particular intervention is justified. But first we must establish that interference does indeed occur, so that poverty is a form of negative unfreedom, and thus warrants the libertarian’s prima facie concern.

On the one hand, it is perfectly obvious that the enforcement of property rights involves intervention. If a poor man tries to take a loaf of bread from a store, the security guard will step in and physically prevent him. As a brute fact, described in neutral, non-moral terms, the enforcement of property rights conflicts with the negative liberty of others to not be interfered with as they take the goods in question. Again, one may consider the interference to be justified – a question we will return to shortly – but there’s no denying that it is physical interference.

On the other hand, right-wing rhetoric concerning property rights is so ubiquitous that it may be difficult to overcome one’s deeply ingrained assumptions and recognize the aforementioned brute fact. Poverty is commonly viewed as a kind of natural and merely ‘unfortunate’ lack, like lacking the strength or intelligence that could improve one’s opportunities. We thus (mis)conceive of the poor as lacking the positive ability to achieve their ends, as if their misfortune were a natural rather than social imposition. An illustrative thought-experiment may help us to see things in a new light.

G.A. Cohen asks us to imagine a society where people are issued with legal “tickets” specifying their liberties, i.e. what actions they may perform. Armed officials intervene to thwart attempts to do something not licensed by one’s tickets. Cohen continues:

“But a sum of money is nothing but a highly generalized form of such a ticket. A sum of money is a licence to perform a disjunction of conjunctions of actions – actions like, for example, visiting one’s sister in Bristol, or taking home, and wearing, the sweater on the counter at Selfridge’s.”

A poor person has the capacity to approach and board the train to Bristol. But security guards would intervene to physically prevent this, were she to attempt it. Thus, “as far as her freedom is concerned, this is equivalent to ‘trip to Bristol’ not being written on someone’s ticket in the imagined non-monetary economy.” This illustrates how poverty is a socially imposed unfreedom. The poor do not suffer any natural lack or inability. Rather, our institutions are such that poor people will be physically prevented from performing actions that would otherwise be open to them.

Having established the fact of interference, we can now address the moral question of whether it is justified. Too sweeping a negative answer would not be plausible, of course. A universal right to take another’s holdings would be self-defeating – as they could simply take it back in turn – and lead to chaos. But there are less foolish proposals worth considering. I will focus on J.P. Sterba’s more careful contrast between (i) the liberty of the rich to use their surplus resources for luxury purposes without interference; and (ii) the liberty of the poor “not to be interfered with in taking from the surplus possessions of the rich what is necessary to satisfy their basic needs.” A right to this second liberty is perfectly universalizable, and so escapes the standard objection mentioned above. We thus find ourselves with a genuine conflict between negative liberties, each of which warrants our prima facie concern and consideration.

To resolve this conflict, Sterba appeals to The Conflict Resolution Principle: “moral resolutions of interpersonal conflicts of interest cannot be contrary to reason to ask everyone affected to accept”. In other words, the demands of morality must be reasonable demands. This principle is surely beyond dispute. The idea of an unreasonable moral obligation is incoherent – if a demand is unreasonable, then one isn’t obliged to meet it.

To apply this principle to the present discussion, note that it would be patently unreasonable to ask the poor to sacrifice their liberty in (ii) above – and hence their basic needs – for the sake of the rich man’s luxury liberty in (i). As Sterba notes, “[i]n the extreme case, it would involve asking or requiring the poor to sit back and starve to death.” The Conflict Resolution Principle thus establishes that the poor cannot be morally required to make such a sacrifice. Sterba further suggests that it would be equally unreasonable to ask the poor to accept any resolution that fails to explicitly favour (ii) over (i). For example, if we held that the two liberties were equal or incommensurable, and that the conflict should be resolved by a struggle between rich and poor, this would be practically equivalent to a ruling against the poor. Importantly, while the rich might not like being required to sacrifice their liberty in (i) for the sake of (ii), it is not such an unreasonable ask, so they cannot offer any parallel objection. Thus, if there is to be any resolution at all, it must be to favour (ii) over (i).

But must there be a resolution at all? This presupposes that there can be no genuine moral dilemmas, or situations in which it is impossible to avoid treating someone unreasonably. Consider a gladiator fight in Ancient Rome between two mismatched slaves that are forced to fight to the death. Once in the arena, what are they to do? If they refuse to fight, we may suppose that both will be killed in punishment. A moral resolution asking the stronger slave to lay down his life would be unreasonable to ask him to accept. But any other alternative – including the “might is right” default struggle – would seem unreasonable to ask the weakling to accept. So the assumption of possible resolution may be a flaw in Sterba’s general argument. Let me offer an alternative.

Note that, by modus tollens, the rich cannot have any absolute right over their holdings that would entail the aforementioned unreasonable duty on the part of the hapless poor. [Footnote: To make my argument explicit: Right -> Duty, ~Duty, / ~Right.] To institute a defensible system of property rights, then, will require pre-emptive measures to ensure that no-one will be in a position to claim the liberty in (ii). In particular, it must be that no-one is left in a position whereby appropriating another’s holdings is the only means available to them to meet their basic needs. Our institutions must guarantee them some other means to welfare. This is necessary to ensure that the rich may justly defend their post-tax holdings against would-be appropriators. We thus find that subsistence rights are a precondition for a legitimate system of property rights.

This ties in nicely with Pogge’s claim that human rights are “moral claims on the organization of one’s society… The normative force of others’ human rights for me is that I must not uphold and impose upon them coercive social institutions under which they do not have secure access to the objects of their human rights.” This understanding provides us with yet another route from negative to positive rights. By constraining ourselves to only negative duties, we find that “human rights give you claims not against all other human beings, but specifically against those who impose a coercive institutional order upon you.” But this is close enough to be practically the same thing. We are all participants, contributors, upholders, and hence imposers of the current global institutional order. Our actions thus cause harms to those who suffer unjustly under this order. It is wrong to impose such harms – a violation even of merely negative duties – so we have a corresponding duty to recompense the victims accordingly.

The arguments of this essay establish that the purported distinction between positive and negative rights cannot do the job its right-wing proponents might ask of it. It cannot justify neglecting the basic needs of the poor, or favouring civil rights to the exclusion of subsistence rights. Once negative rights are granted, positive ones are sure to follow. Most obviously, there is the positive duty to protect negative rights – though this might still be purely ‘civil’ in nature, and so does not suffice for the more radical reconception of rights recommended above. More significant progress is made by realizing that our interference in others’ lives is far more wide-ranging than typically recognized. Since we cannot avoid imposing certain harms on others, a positive duty of rectification is owed, according to the libertarian’s own theory of justice. In acquiring property, recompense is owed to other individuals – present and future – for depriving them of the material resources one appropriates for oneself. Further, I argued from Sterba’s incontrovertible Conflict Resolution Principle to the conclusion that instituting subsistence rights is necessary in order to ensure that enforcement of property rights is legitimate. Finally, Pogge’s institutional conception of human rights reinforces the central argument of this essay: that our actions affect others, and that even the libertarian must thus concede that we owe them recompense for the harms thus imposed. The gap between negative and positive rights is thus bridged.


Cohen, G.A. (1995) Self-Ownership, Freedom, and Equality. Cambridge: Cambridge University Press.

Gewirth, A. (1996) The Community of Rights. Chicago: University of Chicago Press.

Holmes, S. and Sunstein, C. (2000) excerpts from ‘The Cost of Rights: Why Liberty Depends on Taxes’ in Steiner, H. and Alston, P. (eds) International Human Rights in Context: Law, Politics, Morality. Oxford.

Kymlicka, W. (2002) Contemporary Political Philosophy (2nd ed.). New York: Oxford University Press.

Lippke, R.L. (1995) ‘The Elusive Distinction Between Negative and Positive Rights’. Southern Journal of Philosophy, 33, 335-346.

Mill, J.S. (1859) On Liberty.

Nozick, R. (1974) Anarchy, State, and Utopia. New York: Basic Books.

Pogge, T. (2002) ‘How Should Human Rights be Conceived?’ World Poverty and Human Rights, chp 2. Malden, MA: Blackwell.

Steiner, H. (1981) ‘Justice and Entitlement’ in P. Jeffrey (ed.) Reading Nozick. Totowa, N.J.: Rowman & Littlefield.

Sterba, J.P. (1998) Justice for Here and Now. Cambridge: Cambridge University Press.

Van Parijs, P. (1992) ‘Introduction’ in Van Parijs, P. (ed.) Arguing for basic income. London: Verso.


  1. I daresay you start to veer off the tracks when you say "Hence Holmes and Sunstein observe that “all legally enforced rights are necessarily positive rights.”" They claim this, but to say that they observe it suggests that it is a fact, when this has certainly not yet been established.

    Legally enforced rights are not positive rights just because they are enforced. Force is used to protect negative rights, otherwise the state fails to protect the rights of its citizens. Enforcement is the only positive thing it does, according to the libertarian view that favours only the recognition of negative rights. The right to not be killed is still a negative right because it is still the right to not be imposed upon. In that libertarian view, the state merely represents the people in protecting those negative rights.

    For a philosopher like Holmes or Sunstein to say that a legally enforced right is not only a positive right, but "necessarily" a positive right is hardly to be expected. The acts of the state are seen as a duty, namely a duty to uphold negative rights. But such hardly converts those rights into positive ones.

    Further problems rear their head when you discuss "first aquisition." - "By depriving others of what they would otherwise have access to, the appropriating action constitutes a kind of harmful interference, and so – by the libertarian’s own lights – the actor ought to recompense them accordingly." But this is like saying that since you have fifty dollars which, under different circumstances, I could have had, you are depriving me by having it. But what are you interfering with? An state of affairs that never existed, it would seem! And how can such a thing be interferd with? It's a bit like accusing a person of stealing an imaginary porsche. had things turned out differently, I cold have been a major shareholder in Microsoft. Does it follow - by libertarian lights - that the person who is the major shareholdr in Microsoft owes me something?

    I'd suggest a much more accurate representation of libertarian philosophy on your part. It's not that I care for libertarianism, but I do care about presenting a case as best you can, and presenting charicatures is never a persuasive way to go.

  2. Hi Beretta, thanks for the feedback.

    You'll notice that I accommodate your first point after introducing H&S's claims by qualifying them with the clarificatory quote from Gewirth (“the ground or justification for the positive assistance in question is to see to it that potential offenders refrain from the prohibited actions”). So I do acknowledge that H&S merely establish the positive duty to protect negative rights. That's why I need the arguments in the rest of my essay ;-)

    I may need to clarify my section on initial acquisition. Note that libertarian entitlement theory depends upon having a story about how unowned resources mightly justly become owned in the first place (however far back in history this may have occurred). Otherwise there is no just title for the present owners to inherit. It would be like owning stolen goods -- even if the original thief was someone else, the illicit origin still undermines your present claims of rightful ownership. Now, I argue that there can be no such "story". Any act of original appropriation would harm others. How? By taking away what they had access to. Simple. I don't see your Microsoft analogy -- nobody stopped you from becoming a shareholder, I take it? But if I claim all the food on our island as my own, then I have stopped the other inhabitants from using what they previously had free access to. They may well starve to death. Clearly, I would be harming them by doing this! If it weren't for my appropriating action, they would have had access to the food. Because of my acquisitions, they instead don't. Isn't that then a clear case of harmful interference?

  3. 1) [ J.P. Sterba] technically - can you use terms like "poor" and "rich" on opposite sides of universilizable statements?
    Maybe this sort of nit picking can destroy the whole principle.
    2) And the other question I guess is how does one determine "surplus" and "basic". careful defining could get any result you wanted.
    Maybe it is just colloquial.

    > The Conflict Resolution Principle thus establishes that the poor cannot be morally required to make such a sacrifice.

    this becomes part of your conclusion I would have thought. (ok it probably undermines the confidence of the article but personally I like lacking confidence)
    given one accepts "The Conflict Resolution Principle" then...
    The problem being many people don't seem to care about that principle. And that may even "work" for example it is in a sense unreasonable to expect a man to go quietly to the lethal injection chair - but many people do expect that and many prisoners do it.

  4. Yes, your initial acquisition argument does need some clarification. As it stands, somebody has stopped me from becoming a major shareholder in Microsoft. Bill gates aint selling! Any person possessing something is stopping me from possessing it, and the only way to show that they are harming me and owe me redress is to first show that I had a positive right to the thing they possess, which is the one thing the libertarian is very unlikely to accept.

  5. "the only way to show that they are harming me and owe me redress is to first show that I had a positive right to the thing they possess"

    No, it suffices to show that they performed an action which made you worse off than you would have been had they refrained from so acting. That's a plain definition of "harmful interference" -- no need to presuppose "positive rights" here.

    Anyway, misreadings aside, I take it you accept the actual argument as clarified in my above comment?

  6. I've never understood how a positive right isn't just a renaming of a duty, an attempt to translate deontology into rights-based talk.

  7. Yeah, I treat rights and duties more or less interchangeably. My core argument is essentially that duties of non-interference entail duties of positive rectification for the inescapable harms we all do.

  8. Hi Richard -- see I am posting on your blog. All in all your essay reads very well. Where you have:

    > So we are naturally lead from a minimal commitment to negative rights to a more substantial commitment to positive duties.

    You mean 'led' not 'lead', right? I imagine you might have already picked that up.

    I can't really think of many right-libertarian counterarguments to this. It seems to be the point where somebody like Nozick resorts to economic history and the failure of most redistribution schemes. Perhaps I am unjust to him. Most libertarians would probably like the idea that you have just enough redistribution so that nobody has to exercise their freedom to be Robin Hood.

    Mightn't there might be a problem in your conflict resolution scheme, though? You say that subsistence rights are a precondition for a legitimate system of property rights. But 'subsistence rights' mean we have to provide the poor with enough to make it unreasonable for them to think it just to steal. Some might say: How much poverty is it reasonable to endure without resorting to theft? Could the ignominy of poverty be near to unbearable for somebody? Could you imagine somebody perfectly reasonable who was simply biologically or psychologically consituted so as to find being even a little less rich than his or her peers an unendurable torture? You might then say you have to leave out differences in subjective experience and base the system on what the average reasonable person could expect. And libertarians might set the benchmark for what the 'average' person should expect in welfare very, very low indeed.

  9. Oh sorry, by the way, it's me, Alex

  10. Hi Alex! I'd missed that embarrassing typo -- thanks.

    I find it pretty hard to imagine the person who finds a merely relative lack of wealth to be "unbearable torture". But I certainly do think that allowing people to escape the experience of unbearable torture is pretty important! So if allowing someone to have relatively unhindered access to material property really was the only possible way to achieve this, then that would seem pretty important too. (They should try to get a job before taking from my holdings, of course. But if all else fails, I could hardly begrudge them for taking those steps necessary to end their torture. The harm done to me is minimal in comparison.)

    The CRP is fairly vague, as you note: exactly what counts as "reasonable" here? There will certainly be unclear or even indeterminate cases, as I conceded to Fiona in class. But so long as the extreme case of basic subsistence is clear enough, that should suffice for my core argument. It might not establish that everyone has the right to be joyously well-off. But even a positive right to the most basic subsistence is a pretty significant result, it seems to me. So I'm happy enough to establish that for now.

  11. Do we provide great effort to make that person's life better? what if (lets say) upon becoming richer he sqanders his wealth and then returns to unbearable torture?
    Or is the solution to teach him to be reasonable?

  12. I guess the negative rights argument doesn’t rest on that sort of point.

    If you instead work in the other direction and say "to improve society we need rights", "we need the minimum amount of incidences of enforcements of rules" and “we need simple rules” you will probably end up with almost all negative rights automatically. (Or would you?)
    Then as an indirect philosopher you could argue that the fundamental principle is “negative rights” and any diversion from that will most probably be you just misjudging (and screwing up the lovely world).


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)