Ends, means, and liberty

In the previous post I mentioned a Megan McArdle item that gave rise to a few lines of further thought -- this next one results from following a link to a link and has little to do with the original topic: Jim Manzi's post from a while back, "The Paradox of Libertarianism". Manzi starts by distinguishing two varieties of libertarianism -- "liberty as goal" and "liberty as means". The first seems pretty clear, but the second is more complicated, and it seems to me that Manzi is confusing two different kinds or levels of "means" here, which leads him into a false "paradox" at the end.

He begins simply enough, by expanding the term "liberty as means" -- it's a libertarianism that "takes liberty to be a means to the end of discovery of methods of social organization that create other benefits". This, as he says, makes it appear as "practical, inductive and experiment-based" in contrast to "liberty as goal", which appears "idealistic, deductive and theory-based". He admits these are "cartoon terms", by which he means, presumably, that they're overly simple descriptions, but they'll do for now. Here, though, is where a real "paradox" of sorts appears almost immediately: if liberty itself isn't a goal but only a means, what is the goal? In other words, what are those "other benefits" that liberty is supposed to aid us in achieving? And in what sense are those goals or benefits themselves not based upon idealism, deduction, and theory? Only, I suppose, in the sense that such goals remain implicit rather than explicit, and unconsidered rather than thought through, but that hardly gives them any particular advantage or deeper basis than liberty as an end in itself.

More to the point, keeping your goals unstated or unexamined doesn't, in itself, make you any more practical, inductive, or experiment-based. That is, while the distinction between viewing liberty as a means to some other end and liberty as an end in itself remains a meaningful one, it's quite different from the distinction between the practical and the ideal, or between experiment and dogma. And Manzi compounds this confusion by seeming to contrast experiment with theory rather than with dogma, as though experiment and theory weren't mutually dependent. These multiple confusions, then, are what lead him in the end to his false "paradox", in which, in the name of "experiment", he asserts that "a liberty-as-means libertarian ought to argue, in some cases, for local autonomy to restrict some personal freedoms". But this, then -- given the undefined "other benefits" that constitute the real goals here -- just devolves into the sort of unprincipled pragmatism that characterizes contemporary liberalism generally.

What Manzi is missing is the real alternative that fuses principle and practice, theory and experiment. To take his example of the prostitution issue, this alternative would want to look beyond a simple legal prohibition (the conservative position), beyond a simple legalization (his notion of liberty-as-goal libertarianism, but is better characterized as dogmatic libertarianism), and beyond simply passing the buck to local jurisdictions to "experiment" with (his notion of liberty-as-means libertarianism). Instead, it would assert the principle of liberty while recognizing that current realities always constrain immediate choices, and that any such change always involves unintended consequences -- so it would look at incremental changes and experiments, guided by the general goal or principle, but ready to be practical and flexible in implementation.

Now, there's actually one more level of instructive confusion in this post that's quite important -- but it deserves a title of its own, so I'll get to it next time.