Tuesday, January 6, 2009

The Non-Aggression Principle and the Pauline Principle

There is no one non-aggression principle. Rather, “non-aggression principle” names a family of norms precluding the initiation of force against others. One version, familiar to many readers, is the one found in the Libertarian Party’s membership pledge, which rejects “the initiation of force.”

One deontic cousin of the NAP is the Pauline Principle, which occupies a central place in natural law theory.

The Pauline Principle (I believe we owe the label to Alan Donagan in The Theory of Morality) gets its name from St. Paul’s characterization as justly damnable those who maintain that it is morally acceptable to do evil that good may come. Rejecting the possibility of doing evil to bring about good is sometimes understood to mean declining to violate a range of prohibitions, not necessarily connected but understood as absolute. But the new classical natural law (NCNL) theorists, who have done the best and most extensive contemporary work on this principle, have argued for a more general understanding, on which I will focus here.

The NCNL version of the Principle amounts to something like, “Do not cause harm purposefully.” On the view that doing something for the purpose of achieving some other end unavoidably involves identifying with one’s chosen means, so that an instrumental harm is a purposeful harm (“Whoever wills the end wills the means”), the NCNL theorists suggest that the Principle also includes the requirement: “Do not cause harm instrumentally.” (The NCNL theorists treat the Golden Rule’s requirement of fairness a a distinguishable norm which would surely preclude imposing risks of harm on others which one wouldn’t want imposed on oneself or one’s loved ones. So accepting the Golden Rule means accepting a further constraint, not itself part of the Principle but closely linked with it, which might be summarized as, “Do not cause harm recklessly.”)

There are interesting questions about the derivation and justification of the Pauline Principle, but I will not focus on them here. (Cp. Mark C. Murphy, Natural Law and Practical Rationality, and John M. Finnis, Fundamentals of Ethics.)

“Harm,” for the NCNL theorists, has a quite specific meaning. Something causes harm if it impedes, frustrates, inhibits, attacks, etc., someone’s participation in a basic aspect of human welfare. There are alternative lists of these basic aspects of welfare; they include life and bodily well being, speculative knowledge, practical reasonableness, friendship, æsthetic experience, and play. (Other items on some lists include religion, parenthood and family relationships, and self-integration.) These basic aspects of well being are non-fungible (so that one instance of friendship, say, isn’t strictly substitutable for another) and incommensurable (so that there’s no common scale on which whole categories of well being—or individual instances—can be measured and compared). Non-fungibility and incommensurability rule out consequentialism and related views in principle, since these views require something like maximizing or optimizing the overall good in the universe (or some more limited context), and, since the various aspects of well being aren’t reducible to some common substrate, the notion of “the overall good” turns out to be meaningless, vacuous, nonsensical.

It’s important to see what the Pauline Principle doesn’t preclude. By ruling out purposeful and instrumental attacks on basic aspects of well being, it does not rule out the use of force in defense of oneself or others. For, say the NCNL theorists, one can use force, including, if necessary, lethal force, defensively while intending only to stop an aggressor (or quasi-aggressor: the moral culpability of the person being stopped isn’t an issue, so the Principle wouldn’t preclude using force if necessary against, say, a sleepwalker with a gun), not, per se, to kill or harm the aggressor. The harm to the aggressor done by a defensive act isn’t purposeful: one’s goal isn't, as such, to harm; but it’s not instrumental, either: one’s purpose isn’t to stop-by-harming. Rather, the harm is a foreseen but unintended by-produce or side-effect of one’s stopping the aggressor.

There are all sorts of interesting things to be said about the Pauline Principle, its justification, consequentialism, etc. But I want to focus myself, and to focus our conversation, on the ways in which this norm, carefully explicated and rigorously defended, might be thought to overlap with and differ from the NAP.

One crucial distinction: many versions of the NAP understand “the initiation of force” to include the initiation of force against property. Since a person’s property is not among the basic aspects of well being recognized by NCNL theory, the Pauline Principle, ruling out attacks on these dimensions of human welfare, does not, per se, preclude attacks on or violations of people’s property.

This does not mean, of course, that NCNL theory does not take property seriously. Most importantly, the NCNL theorists see the Golden Rule’s requirement of fairness (that one ought not to make arbitrary distinctions between persons and that one ought to treat others as one would want to be treated oneself, or would want one’s loved ones to be treated)—taken together with various general facts about human beings and their circumstances—as grounding the basic principle that there ought to be property rights; and this same requirement would obviously constrain interference with other people’s property rights. But the requirement of fairness, though real and substantial, does not provide support for a prohibition on all purposeful or instrumental interference with anyone’s property.

This seeming disagreement regarding property is important, but perhaps not as important as it might seem. For some versions of the NAP, including the one contained in the LP pledge, make no specific reference to property. And it would not be clear just how much even a version that did refer simply to property generated different conclusions from the Pauline Principle. To know just how much conflict there was between the Pauline Principle and the hypothetical version of the NAP, it would be necessary to know just what definition of property was presupposed by or explicitly incorporated in the version of the NAP in question. There are, after all, vastly many accounts of property rights, and simply to refer to “property” is not to answer the question which of these theories should be endorsed.

Still, it is clear that many versions of the NAP would lead to more extensive moral restrictions on actions affecting the property of others than would the Pauline Principle as understood by the NCNL theorists. That’s one key difference.

John Finnis maintains that the Pauline Principle rules out slavery in principle. But it’s not altogether clear how. Nor is it obvious how the Principle excludes confinement and other restrictions on personal liberty.

That doesn’t mean, of course, that the NCNL theory lacks resources for dealing with such restrictions. To the extent that one wouldn’t like being subjected to them oneself, one has good reason, in accordance with the Golden Rule, to avoid imposing them on others. But the wrongness of slavery seems more unequivocal than this: someone who wouldn’t mind being enslaved still does wrong by enslaving. It is clear, by contrast, how the NAP rules out slavery (I ignore the issue of voluntary slave contracts here).

The NAP and the Pauline Principle lead in practice to many of the same outcomes. But the implications of the Pauline Principle as a safeguard for liberty remain unclear and deserve more study.


quasibill said...

Even though I'm not much of Rand fan (Long and Sciabarra deserve a lot of credit for me being able to mention in any sort of positive light), I'm kind of partial to the following formulation of the NAP:

No person is entitled to use another person merely as a means to her own ends.

With a Randian understanding of what "own ends" means.

Gary Chartier said...

A lot of these principles live relatively proximity in what (in an attempt to sound clever and learned) I’ll call normative space. How would you understand this formulation to be similar to and different from the second version of Kant’s categorical imperative? (The translation in the Wikipedia article on the CI reads: “Act in such a way that you treat humanity, whether in your own person or in the person of any other, always at the same time as an end and never merely as a means to an end.”)

The NCNL theorists certainly understand the Pauline Principle as expressing an insight very similar to Kant’s (for them, the biggest problem with Kant’s view is his failure [as they would see it] to acknowledge the diversity and incommensurability of the various aspects of human well being). I’d be interested in knowing not only what you take to be Rand’s understanding of “own ends” but how the version you’ve articulated might differ in its implications from Kant’s.

quasibill said...

I take Rand's understanding of "own ends" to be very inclusive. For example it does not exclude formulations that are purely altruistic. This helps avoid the "burned the village to save it" problems that otherwise crop up.

I'm not real deep in philosophical realms, so I can't say that I know what possible implications arise from my formulation versus the CI (this is probably the first time I've ever encountered it - I don't remember if I did in undergrad). I do think that the "merely" in my formulation is necessary, because we always use each other to some extent as means; the problem is when you use another only as a means and don't respect them as their own end in themselves.

I recognize that it's far from perfect (if there was a perfect, pithy formulation, someone far smarter than I would have probably come up with it already!), but it seems to me to be a very good way to start a conversation on the NAP without goring any sacred cows from the get-go. After a few minutes discussion, most people accept it wholeheartedly in the abstract. The implications, on the other hand...

Gary Chartier said...

My guess is that Kant and the NCNL theorists would both agree with you that “merely” was important. I take it that Kant includes it in his formulation for the same reason, in general, that you do.

And I quite agree that it matters not only what the content of my ends is but also that they are mine and not yours.

I think what seems most obviously different between the NCNL formulation on the one hand and, on the other, both your proposal and the second version of the CI is perhaps that the NCNL theorists would rule out causing harm to someone as a means to some end, as well as causing harm to someone purposefully, but don't seem per se to include reducing someone's zone of autonomy as a cognizable harm. Harm, on their view, means attacking a basic good, and the basic goods are quite diverse—life, knowledge, practical reasonableness, play, friendship, æsthetic experience, etc. But autonomy isn't, as such, a fundamental aspect of well being; on their view, it's derivative. In many cases, there would be no difference: your view, Kant's view, and their view would all rule out purposeful killing, and their view, as they understand it, would, like Kant's, rule out purposeful deception. But it still feels to me (and this is something I find frustrating about their view) as if there's an insufficient appreciation for the value of freedom as such on their part.