Concepts such as that of "government," like those of "democracy," "law," "justice," "freedom" and "love," to cite just a few, is what W. B. Gallie called "essentially contestable" (see his "Essentially Contested Concepts," Proceedings of the Aristotelian Society, Vol. 56 [1955-56]). I heard the characterization from Alasdair McIntyre back in the mid-70s at the Creighton Club, the New York State philosophical society, although not applied to "government" but to a slew of other concepts that are constantly being debated.
Of course, from within specific philosophical positions these concepts are pretty firmly defined, so that, say, in classical liberalism "freedom" is usually defined to mean "absence of coercive force" or "not being subject to initiated force" while from within Marxism it's taken to mean "absence of necessity."
Within libertarianism, though, the concept "government" is still unstable. Anarcho-libertarians, who argue for something they dub "competing legal systems" or "competing defense organizations," claim that the concept "government" means, essentially, "a monopoly of legal services over a given territory." This isn't as clear-cut as one might wish. Are they talking about legally protected monopolies or monopolies plain and simple, which could mean very competitive organizations, indeed – for example, a department store sitting on a large piece of private property that has no competitor right then and there but is amply competed with by stores in the nearby vicinity? Yet where it stands, it's a monopoly, in a sense. Or an apartment house – it too stands alone and to rent a competitor's dwellings, one needs to move.
There are libertarians called minarchists, with whom I am usually linked − along with Ayn Rand, John Hospers, the late Robert Nozick and, during the last few years of his life, Roy A. Childs, Jr. (although he also penned a famous piece, "The Contradiction in Objectivism," back in 1968, for Rampart Journal, in which he announced his dissent from Rand's minarchist position). I disagree that governments may not compete and may coerce anyone. To be fair, neither did Ayn Rand agree that governments may coerce anyone − she, for example, denied that taxation is permissible while also claiming government is, thus disowning the characterization of government by perhaps the most famous anarcho-libertarian, Murray N. Rothbard.
But as Gallie's point makes clear, this debate as to what is the most sensible, reasonable definition of "government" is likely to continue for a long time, if not indefinitely. In my own view, for example, the institutions anarcho-libertarians support are governments in every important respect − they are administrators, maintainers and protectors of bona fide law within human communities. What critics claim is that such administration, maintenance and protection do not require contiguous spheres of jurisdiction but could work as a sort of crisscross system.
From a few historical cases, in which such a system had been in place − in ancient Iceland, for example − these disputants conclude that as a general rule governments could operate quite happily, smoothly, with no judicial failures − as inability to arrest and prosecute criminals or to render effective service when citizens (or clients) seek police protection − serving crisscross localities. Okay, so this is an interesting debate and worthy of pursuit. Either way we could get to government, however.
My one beef with many who reject this idea is that they refuse to admit that "government" need not involve coercion at all. They could just as easily dispute that the crisscross system involves law, properly understood perhaps only as various rules or edicts or policies. And even more problematic is their all too frequent use of the concept "state" as a substitute for government.
For example, in a recent letter to Liberty magazine, Professor Roderick Long of Auburn University sets out to take issue with Bruce Ramsey's claim that Hernando "de Soto's work . . . shows that a healthy economy crucially depends on property titles, identity records, and other institutions of formal law" and is thus "a standing refutation of libertarian anarchism."
As Long proceeds in his letter, however, an interesting switch takes place. He contends that "as the research of scholars like Bruce Benson, Tom Bell, and others has shown, history is filled with examples of legal systems that were perfectly formal − complete with official procedures, court records, and the rest − and yet private, competitive, and non-governmental." He states that "in late medieval Europe . . . the commercial law known as the Law Merchant outcompeted the government legal system . . .." And then, from this, he jumps to the following conclusion: "Hence the state is not necessary for formal law."
I don't know about Bruce Ramsay but I certainly would not conclude from de Soto's work that the state is necessary for anything, although I would agree that governments may well be. Because what Long and all those other scholars show, as far as I am able to discern, is that in medieval Europe there were different kinds of governments, some of them coercive and others not.
Okay, so what's wrong with this conclusion? I assume critics would now claim that I am twisting the concept "government" to suit my goals, namely, to defend governments as quite possibly a just institution administering, maintaining and protecting bona fide law. I dispute this. I claim that they are wrongly claiming that governments must be unjust and so the concept ought to be abandoned by all right-thinking folks. But one way they support this is by equivocating between "government" and "state."
It is well known that the concept "state," especially as it figured in the writings of Hegel and Marx, is not the same as "government." It is, instead, the entire organized community, akin to what Aristotle meant by "polis." The state does, then, quite sensibly call to mind a fully coercive leviathan, a pyramid-shaped, top-down system of coercive regimentation of nearly all facets of human community life (apart from those deemed not essential, although even those would be subject to regimentation if the agents of the state so chose).
Now, I am not going to resolve any of the main disputes here but I wish to make just one little final point. To equivocate between "government" and "state" is wrong and even dirty pool. It would be similar dirty pool if those critical of anarcho-libertarians referred to what the latter advocate as "chaos," recalling not the arguably esoteric conception of anarchy individualist and libertarian anarchists have been developing but the position of those old-fashioned, classical anarchists who meant by the term "lawless society."
Of course, when emotions run high − as they tend to do in discussions among people who are nearly in full agreement and know that they are more likely to be able to land a blow at those in close range than at those who don't even pay attention to their views − it's tempting to engage in hyperbole.
Labeling an allegedly "near pure" libertarian opponent a "supporter of the state" or "a statist" does carry a painful sting. One would hope, however, that just this temptation is resisted by serious scholars.