The entry on contractarianism of the Stanford Encyclopedia of Philosophy begins thus:
"Contractarianism" names both a political theory of the legitimacy of political authority and a moral theory about the origin or legitimate content of moral norms. The political theory of authority claims that legitimate authority of government must derive from the consent of the governed, where the form and content of this consent derives from the idea of contract or mutual agreement. The moral theory of contractarianism claims that moral norms derive their normative force from the idea of contract or mutual agreement.
It is pointed out also that "There is no necessity for a contractarian about political theory to be a contractarian about moral theory, although most contemporary contractarians are both." The fact that some contractarians are anarchists - like the contemporary moral and political philosopher Jan Narveson - illustrates the logical independence of the two ideas.
The political social contract is supposed to ground the legitimacy of political authority. The parties to this contract are thought to be the subjects of political authority, on the one hand, and the wielders of that authority, on the other hand. That is, the citizens of a polity and their government or state. The legitimacy of the latter is thought to derive from the (explicit or implicit, actual or hypothetical) consent of the former. Like many critics have pointed out, as a matter of history, this is wildly inaccurate. Here is Anthony de Jasay making this point:
It must not be forgotten that contrary to the underlying postulate of [political] contractarianism, in reality no state has ever been formed at the unanimous wish of its prospective subjects. The origin of states has always been the assumption of power over the prospective governed by the prospective governors, usually by foreign conquerors, with support by a part of the governed purchased with the resources transferred to them from the rest of the governed. It is this standard mechanism that social contract theory seeks to represent in the metaphor of a voluntary contract of exchange between governors and governed.
By contrast, non-political or "moral" social contract theory not only has a completely different aim, but the parties to the contract in question are completely different. The parties here are every individual person with every other, or each with all the rest. David Gauthier is a representative of this kind of contractarianism. In his book Morals by Agreement, it is argued that it is rational for each person to adopt certain constraints on her behaviour in relation to her fellows on condition that they respect the same constraints on their behaviour in relation to her. It is the conditional feature that makes the terms "contract" or "agreement" suitable despite the fact that it is not a matter of any actual contract. "Contract" or "agreement" are metaphors or models for explaining the mutual advantage of accepting a set of interpersonal rules of conduct provided that others do so as well. Whether these rules should be thought of as being moral rules is, I believe, a further question that I have addressed in an earlier post.
Both political and non-political social contract theory make use of the idea of the "state of nature", but this imagined pre-agreement situation is very different in the two kinds of theory. In trying to understand why we would want something (government or social norms, for example) it is often a very good idea to try to imagine how it would be without it. Hence, the political contractarian imagines a situation without government (but with social conventions and the like intact) and asks how that situation would be like and whether establishing government could be seen to improve on that state. John Locke is a case in point. The non-political or moral contractarian asks instead about a situation without social rules regulating interpersonal relationships; without constraints on the pursuit of individual utility maximisation. Thomas Hobbes imagined that such a state would be one of war of all against all. As Jan Narveson points out in his very helpful Contractarian Theory FAQ,
The example of Hobbes, however, brings up the need for an immediate distinction, for in Hobbes' masterpiece "Leviathan", his avowed main purpose is to justify the State. To do so, he appeals to a moral theory, which he applies to the political problem. Ever thereafter there has been confusion, usually quite thoroughgoing, between Hobbes' moral theory and his political theory. . . . The question to ask about Hobbes's political theory is whether it is a correct application of the moral theory. I believe that it is not. For that reason, if the perceived problem with Contractarianism is that its political consequences are awful, then we must at least consider the possibility that the fault might lie, not in the contractarian moral theory itself, but in the attempt to extract political blood from contractarian turnips.
Some critics have (wilfully or not) conflated political and non-political contractarianism and taken themselves to have refuted the latter while in fact having offered arguments only against the former. Michael Huemer, for example, in his book The Problem of Political Authority, successfully demolishes political social contract theory and then later in the book misleadingly asserts that his arguments also undermine the idea (which Huemer explicitly attributes to Narveson) that "moral principles are determined by a hypothetical social contract". Huemer writes that "I reject all forms of social contract theory, for reasons discussed in Chapters 2 and 3", but turning back to those chapters one finds criticisms only of political social contract theory; criticisms with which the non-political social contract theorist could (and Narveson, I believe, would) readily agree. This is disappointing coming from Huemer who should know better. That Huemer himself sees no need for social contract theory - since he believes that our moral intuitions are a reliable guide to moral truth - is no excuse. If one, like me, however, thinks that moral intuitionism is ontologically and epistemologically hopeless, then contractarianism (of the non-political kind) remains an attractive approach to social rules.