I've really been enjoying all the terrifically smart comments in the last two posts, which have been really helpful to me.
Maybe one way to frame my argument is that I'm not arguing for “big-government libertarianism,” but for “limited-government liberalism.” (I see that this may confuse my prior attempts to distinguish “small” and “limited,” but bear with me as I circle and recircle looking to illuminate some things that have long nagged at me.) My sense is that most people don't see a difference and what I've been doing is groping for a way to make the distinction between the two plausible and clear.
One way to think of it is to consider where you're starting from. I am, more or less, starting from a fairly typical liberal account of the state and it's aims. I'm then arguing that those aims are best achieved through appropriately limited government. I believe this is a principled and coherent stance that stands or falls with the background arguments about the legitimacy and desirablity of liberal democracy. If you reject the basic argument for the liberal democratic state, then you're not going to be impressed. But if you accept the desirability of liberal democracy, you have every reason to listen.
On the other hand, if we start from traditional minimal-state libertarianism, we're starting with a deep suspicion of the possibility of anything more than governance of very low quality; it's just that the anarchist alternatives (which may be feasible and worse, or just infeasible) are worse. So, to this sort of libertarian, “big-government libertarianism” sounds like one of three things: (1) A libertarianism that can't possibly work–a hopeless oxymoron. The argument for keeping the state strictly minimal is that a more-than-minimum state cannot be kept from bloating into a rights-stomping superstate. You've got to choose between almost nothing or almost everything, but big-government libertarianism tries to have it both ways. No dice. (2) A strategy for getting liberals to trust libertarians enough to listen to libertarian policy arguments. Some libertarians seem to think this is worth trying (how could we do worse?) while others seem to think liberals are too hopelessly statist to be worth bothering with. But the real problem with (2) is that liberals will plausibly suspect libertarians of indifference to their conception of basic liberal values, and so libertarian policy advice dressed in liberal language looks like insincere Trojan Horsing on behalf of the powerful. Why listen? (3) Libertarian non-ideal theory. One might think minimal state libertarianism is correct, as a matter of ideal theory, but see that it is politically infeasible, and so conclude that a libertarianism that makes some concessions to welfare state liberalism is second-best, and steps in that direction are small movements toward the ideal. One problem here is: How do you tell the difference between someone who actually thinks mandatory retirement accounts, for example, are the best policy for limited-government liberalism (that's me, on days when I'm convinced that means-tested alternative can't work), and someone who argues for mandatory retirement accounts as a small step down the road to libertopia?
So, “limited-goverment liberalism” starts with liberalism and then argues that libertarianish policies and institutions will best secure liberal aims. “Big government libertarianism” starts with traditional minimal state libertarianism, but moderates it to make marginally more libertarian policy politically feasible.