The Libertarian Idea: Part One, part one
We’re back! Sorry for the delay. Confidential, as Dan Savage likes to say, to the readership: posting may be a bit spotty (once a week) for the next month or so. (The problem is a semester’s worth of work that needs to get done in three weeks that they don’t pay me around here. While this is anecdotal evidence that cost reflects real value, it also makes for a shoddy employee.)
Last post dealt with the broad philosophical notions that motivate The Libertarian Idea. The explanation was meant to situate The Libertarian Idea relative to other works of, and traditions within, libertarian theory, and thereby to show how strange and counterintuitive a work The Libertarian Idea is. This post focuses on the first part of Narveson’s book, which aims to defend the coherence of libertarianism. The idea in the first part is not to show that libertarianism is true, but to show how it could be true – hence Part I’s title, “Is Libertarianism Possible?”
Libertarianism, Narveson tells us, is the view that protecting or respecting “liberty is the fundamental and only legitimate concern of any just society,” so each of us has a “fundamental right to liberty” (Narveson, 13). Three concepts thus need explication. (1) To know what it is a just society is supposed to respect we need to know what liberty is. (2) To know how to respect liberty we need to know what it is to infringe another’s liberty. (3) Finally, to know what it is to respect liberty we need to know what a right is (because the concept of a right falls out of the other two concepts, I will say very little about Narveson’s treatment of it - and reserve what I do have to say for next post).
The danger to libertarian theory – which Narveson hopes to avoid – is that in providing an account of these concepts one might end up with a theory that doesn’t deliver core libertarian judgments: for instance, about circumstances in which coercion is permissible.
Here’s an example of how this might happen. Let’s say, as we will see Narveson does, that liberty is a scalar concept – one can have more or less of it, up to a maximum and down to a minimum. Thus, A has unlimited (or maximum) liberty just in case A can do whatever he wants to do. A has more liberty when A is able to do more of whatever he wants to do; less liberty when he is more constrained. Finally – you get the idea! – A has no liberty when A can’t do anything he wants to do.
On this conception of liberty, it is obviously possible (I would say: probable) that a given distribution of goods – food, for example – can have an affect on the extent to which different people have liberty, or are free. Say that A has all the food and doesn’t want to share it with anybody else. Then B and C are likely more limited in what they can do than if A, B and C have food in equal shares. But then we might say that the first possible distribution of food – where A has it all – constrains B’s and C’s ability to do what they want more than does equal distribution (one is limited in what one can do when one is starving); Distribution One interferes with B’s and C’s liberty more than does Distribution Two. But, then, if interfering with B and C’s ability to do what they want to do disrespects their liberty – violates their right to liberty by reducing them to a condition where they cannot do much of what they want – it follows that there is a libertarian case, albeit not dispositive, for redistribution! (And note that the food example generalizes to any good.)
This is the sort of conceptual vortex Narveson wants to avoid. There are a few available strategies. One is to say that a person can still has maximum liberty even if he is starving. To this end, some libertarians explicate liberty as follows: A has liberty just in case A is not coerced by another agent, i.e. just in case A is not forced to do what another wants him to. Coupled with the (deniable) premise that declining to provide another with food does not amount to forcing him to forbear doing what he would do if he had sufficient sustenance, this conception of liberty implies that a mere distribution of goods (distinguished from the history of that distribution) does not diminish others’ liberty, and so cannot violate their rights to liberty.
Narveson declines to follow this strategy because he thinks how much liberty a person has depends not on the source of whatever affects one liberty, but on what affect that source has. As he puts it: “it is difficult to discern a difference, literally, in the meaning of ‘liberty’ between cases where your liberty is affected by lack of means on the one hand or by the presence of inhibiting factors on the other; or by natural causes on the one hand or human interference on the other” (Narveson, 29-30).
Narveson thus accepts that A’s liberty is a function of the extent to which A can do what he wants to do. If one distribution of goods thwarts A’s ability to do what he wants more than another then A has less liberty on the one than on the other. But it does not follow, Narveson wants to insist, that the distribution interferes with A’s liberty. Narveson argues this is so by introducing a distinction between interfering with A’s liberty and failing to promote A’s liberty. “For a person, A, to “interfere with” the liberty of some other person, B, is for A to bring it about that B is unable to do what B would otherwise be able to do even without A’s assistance.” By contrast, “For A to promote B’s liberty is for A to bring it about that B can perform at will some action or actions that B would otherwise be unable to perform” (Narveson, 30-31). To summarize:
Interference with liberty, in short, requires positive action on the part of the interferer. The mere nondoing of what would, if done, promote your ability to do what you want is not sufficient. The “positive action” in question may be somewhere in the background: if on Tuesday I promise you that on Wednesday I will assist you in moving some furniture and then fail to show up, my inaction on Wednesday is indeed part of a train of action on my part that does interfere with your liberty. But my nonappearance on Wednesday, in and of itself, is not what has thus interfered (Narveson, 31).
Armed with this distinction, Narveson is able to say that a person may be less free on one given distribution of goods instead of another – so his liberty might be promoted by shifting from Distribution One to Distribution Two – but it isn’t necessarily the case that preserving Distribution One (more precisely, declining to shift to Distribution Two) interferes with his liberty. When A declines to promote a shift from D1 to D2, A fails to bring it about that B can perform at will some action that B would otherwise be unable to perform, but he does not prevent B from doing what B could do even without A’s help.
There are two residual problems for Narveson explication of libertarianism, which I will address next post. This will bring our discussion of Narveson’s explication of libertarianism – his account of how it is coherent – to a close.
The first problem, familiar to libertarians, concerns what it is for A to “bring it about that B is unable to” do something. Obviously if you tie someone up you bring it about that he’s unable to move. But what if you tell him you’ll kill him if he doesn’t stay still? What if you threaten to subject him to withering criticism – and you are a withering critic! – if he moves? Threaten to withhold your love? A solution libertarians pursue in practice is to develop a list of the sorts of things that can interfere with liberty – physical violence and threats of it – and say that other things don’t interfere. But that isn’t going to fly on Narveson’s view. A wants to do x. B threatens to mock A mercilessly if A does x. Because A can’t handle B’s mockery,A develops an aversion to doing x that trumps A’s desire to do x. B has brought it about that A is unable to do what he wants to do – he has inculcated an aversion in A that is stronger than A’s still occurrent want to do x. This is an interference on Narveson’s view, but it is also a paradigmatic example of the exercise of free speech.
The second problem, less familiar to libertarians because more closely connected to Narveson’s contractarian commitments, concerns who can have liberty. Narveson wants to say that rational agents are the only sorts of things that can have liberty. But are rational agents the only sorts of things that can have wants? If not, what makes rational agents special? For that matter, specialness has the ring of an evaluative, moral property – somebody’s special when he has distinct value, right? But Narveson can’t say this. If who counts as a rational agent depends on who has value then to decide who the bargainers in the contractarian situation are requires making a moral judgment. But if contractarianism presupposes moral judgments then contractarianism cannot be the criterion of morality.
Next post we will see how Narveson deals with these problems. Then we’ll turn to contractarianism, and how it is supposed to ground libertarianism.
Fn1: This is not an uncontroversial conception of liberty. Notice, as Narveson does, that it allows for the possibility of a free slave (Narveson, 27-28). The kooky case is that of the servile slave who has one desire: to serve my master. A less kooky case is the stoic slave, who is largely indifferent to what he does. Maybe he wants food, shelter and water, but that’s it. Thus, as long as his master feeds him and provides shelter, enslaving the stoic slave does not interfere with his liberty (on Narveson’s conception thereof). This even if the stoic slave is under threat of coercion should he disobey a directive.
Narveson is willing to bite the bullet in the case of the stoic and servile slave. They have liberty, he says, but they are such unusual agents that it isn’t clear slavery would be an interference of liberty in a society with enough such agents wandering about. But there is a rejoinder: can’t we come to acquiesce to a condition of slavery under the combined influence of an absence of alternatives and a pro-slavery ideology (much like some critics of the free market speculate the working class has acquiesced to capitalism - see footnote 8 of the linked essay)? Don’t we want to say of agents with this kind of history that they lack liberty even though their desires have come to conform to their plight? They may be satisfied, but they are nonetheless dominated. Considerations of this kind have led to an alternative conception of liberty – propounded by Philip Pettit – according to which one has liberty just in case one is not subject to the arbitrary will of another.
How to adjudicate between these competing conception of liberty – and there are yet others, as well – is a complicated question, to which there may be no answer. (There is no answer if our concept of liberty is sufficiently inchoate that it can bear any number of different incompatible explications.) My sense – a somewhat pessimistic take if we hope for a neutral concept of liberty – is that people choose their conceptions of liberty largely based on what their moral or political goals are. Libertarians who define liberty as the absence of coercion are attracted to that conception by the ease with which it can be pressed into an argument for a night watchman state. Pettit-style republicans define liberty as the absence of domination because they can invoke the value in an argument for a more egalitarian distribution of goods. And it should be quite clear to readers of my first post why Narveson prefers his conception of liberty to the alternatives. If we define liberty in terms if wants then it immediately follows that contractarian agents – agents bargaining for moral principles – will care about liberty. When you are bargaining for the best deal you can get you will bargain for a deal that allows you to do more of what you want.
Fn2: What makes physical violence relevantly different from psychological violence?
Update: Fixed some formatting and added some links.