Over on Positive Liberty, Jason Kuznicki raises some interesting concerns about the theory of constitutional legitimacy I develop in Restoring the Lost Constitution. I thought I would offer some clarifications of my approach which might address some of the issues he raises.

(1) The concept of "legitimacy" I am considering concerns whether laws that are imposed on the people create a prima facie (i.e. rebuttable) duty of obedience because they are laws, or (another formulation) are presumptively binding on conscience.

(2) The concept I am considering is objective, not subjective. By this I mean that I am not discussing the perception that there is a prima facie duty to obey the law, but whether any such perception is valid or justified.

(3) One route to legitimacy is the consent of the governed, but in the absence of the sort of unanimous consent enjoyed by such institutions as schools, churches and other voluntary associations, the consent of some can never explain how nonconsenting persons come to be bound in conscience. This is the challenge to constitutional legitimacy made famous by Lysander Spooner in his No Treason: The Constitution of No Authority. If it counterintuitive to you, then I urge you to read Chapter 1 of Restoring the Lost Constitution where I explain this challenge in detail. (Don't write to me about this particular issue unless you are contesting the analysis I present in this chapter, rather than merely this brief summary paragraph.)

(4) In Chapter 2 of Restoring the Lost Constitution, I try to meet Spooner's challenge by positing an alternative route to legitimacy. (a) No one can complain about the imposition of a law if it does not violate his natural (prepolitical) rights; (b) because one has a duty to respect the rights of others, one also has a duty to obey a law that is necessary to defining and protecting the rights of others. A coercive command that meets these two requirements is "just" and binding in conscience. Therefore (c) a law is legitimate (in the sense defined above) if it is produced and enforced by procedures that make it more likely than not that it (a) respects the rights of the persons on whom it is being imposed and (b) is needed to protect the rights of others. In this sense "constitutional legitimacy" is procedural in nature, though the procedures must be assessed with a background theory of substantive rights in mind.

(5) By this account (and here I am addressing Jason's concerns), legitimacy does not depend on people's perception of their rights, which people may disagree about. Agreement on rights is not necessary for legitimacy. Legitimacy (in the objective sense) is a means of analyzing particular legal systems. Each of us may disagree on this assessment, of course, but we then must debate the claims underlying our different assessments. As with any other moral disagreement, the existence of disagreement itself does not imply that neither of us is correct, or that there is no right answer. That would be a position of moral skepticism that, while a serious argument, would need to be addressed in other ways. So while different opinions about rights will surely exist, legitimacy (like rights claims themselves) will depend on who is correct in their opinion about rights and also about whether adequate procedures exist to protect them. The existence of disagreement itself, however, does not prove that no one is correct.

(6) Legitimacy is a matter of degree. Given that is it based on the likelihood that a law is just (in the sense defined above) this likelihood can vary greatly depending on the procedures in place, from no likelihood, to barely more likely than not, to strongly likely. This is in sharp contrast to consent theories of legitimacy which yield all-or-nothing conclusions.

(7) Laws can be legitimate for some and not for others. Unlike consent theories, by this approach a law could be legitimate for one person or group while illegitimate for others. The original Constitution provides an example of this. Laws enacted pursuant to its procedures could have been legitimate for white males (perhaps only white property-holding males) while completely illegitimate for slave. Woman and nonvoting men might be somewhere in between. The absence of the vote does not go to lack of consent, but lack of assurances that the laws imposed upon the unrepresented group does not respect their rights because they were not able to protect themselves effectively in the political process. If for example, the procedures that were good enough to protect male voters was equally applied to nonvoting males, but not to women, then these laws could be legitimate for all males, though not for women. In addition, some of the laws enforced by a legal system could be legitimate and others of the procedures in place clearly permit certain types of unjust laws, while protecting against others. In practice, then, assessing the legitimacy of a real-world nonconsentual legal system will be pretty messy.

(8) A further advantage of this approach is that, assuming that a procedure has been devised that imparts legitimate laws on one group, legitimacy for others can be obtained by extending the same protections equally to other groups as well, whether or not these groups participated in the original framing of these procedures.

There is much more that can be said about this approach to legitimacy, and I say more about it in my book where I consider, for example, the objection that this form of legitimacy is far too narrow or confining. Part of my response is that unanimous consent governing institutions like schools, churches and other voluntary associations can impose many more restrictions on their consenting members than can a government legal system that seeks to impose laws on nonconsenting persons.

But this post is already much longer than I intended. It provides a perfect example of what my old friend, George Smith, used to call "creative evasion": what we can accomplish when avoiding some task we would rather not do—like grading exams.