I’ve been having an extended Twitter discussion about the history of women’s liberty with Cato’s Jason Kuznicki (@JasonKuznicki), the Atlantic’s Megan McArdle (@asymmetricinfo), and others.  I find some of the issues hard to address in 140 characters, so I’m moving my thoughts here.

Many libertarians (see here and here for example) claim that women in the Anglo-American world were much less free in the Gilded Age than they are today.  I infamously disagree. (See here, here, and here for starters).  The Gilded Age was inferior to the present in many ways.  People were a lot poorer, technology was far less advanced, and values were unpleasantly stodgy.  But from a libertarian point of view, the Gilded Age also had major advantages: Far lower taxation, far less economic regulation, and never forget near-open borders

Given these unisex advantages, libertarians who want to argue that women were on balance less free during the Gilded Age need to find some large off-setting impositions on women’s freedom.  What might these be?  The most popular answer points to marital law.  Under the doctrine of coverture, a married woman was not free to own property, make contracts, or accept employment without her husband’s permission.

As a matter of legal history, coverture was already dead by 1880.  But my opponents could just push back the year a bit and resume the debate.  I challenge their basic premise.  As I recently explained, the doctrine of coverture seems no more “coercive” than the volunteer army.  If you don’t want to waive your rights, don’t accept the deal.

Jason Kuznicki’s not happy with this response.  In his view, making people choose between marriage and liberty is a “coercive choice.”  Is he right?  The path to an answer is surprisingly convoluted, but it’s worth the effort.

For starters: You could say that marriage should be entirely private, so any government involvement in marriage whatsoever is coercive.  If so, I agree.  You could add that some kinds of government involvement are more coercive than others.  If so, I agree again.  The key open question is: What are the factors that make marital law more or less objectionable from a libertarian point of view?

I offer the following factors for your consideration:

1. Freedom of contract.  If the government defines marriage in any one-size-fits-all way, Jason’s critique is worth considering.  The government is saying, “Either get married our way, or don’t get married at all.”  However, the more latitude parties have to renegotiate the standard terms, the less objectionable the government’s role becomes.  In the extreme case, the government’s definition of marriage is merely a default rule – and we’re one step away from freedom.

2. Legality of substitutes.  Suppose the government has one unalterable definition of marriage.  If all other arrangements resembling marriage are illegal – cohabitation included – this is very coercive indeed.  If all other arrangements resembling marriage are unenforceable, there’s still a serious problem.  But if arrangements resembling marriage are legal and enforceable, the government’s definition of marriage is only mildly objectionable.  In the extreme case, suppose the government rigidly defines marriage, but recognizes all “schmarriage” contracts between consenting adults.  Government is depriving people of the word “marriage,” but that’s it.

3. Degree of enforcement.  Suppose the government has one unalterable definition of marriage, and officially bans all other arrangements resembling marriage.  If it strictly enforces its laws, then this is very coercive indeed.  The less it’s enforced, the less coercive it becomes.  At the other extreme, if the laws against other arrangements are never enforced, then it’s hard to see the big deal.  Virginia still officially bans cohabitation.  But since the law is a dead letter, it’s no more than mildly coercive.

4. Customary defaults.  Default rules are far less objectionable than mandatory rules.  Still, as Sunstein and Thaler emphasize, rule-makers can use defaults to “nudge” people
in their preferred direction.  Contrary to Sunstein and Thaler, however, this
doesn’t make nudging inevitable.  The simplest alternative to nudging
is deference to ordinary language.  Unless parties say otherwise, assume
that they’re using words in the customary way.  Rothbard puts it well:

[I]f a man simply
sells what he calls “bread,” it must meet the common
of bread held by consumers, and not some arbitrary
specification. However, if he specifies the
composition on the loaf, he is liable for prosecution if he is lying.

The same goes for marriage.  If government insists on setting the
default rule, the least objectionable path is to defer to ordinary
language.  If the common definition of “marriage” includes monogamy,
favorable custody rules for women, or coverture, then that should be
part of the default.  The more marital default rules ignore ordinary
language in favor of some vision of what marriage ought to be, the more objectionable.

My ultimate goal is to compare the shortcomings of marital law in the Gilded Age and today.  Before I go on, though, what do you think about my four criteria?  Any objections?  Anything to add?  I’m happy to hear all comments, but I freely admit that this is essentially an intra-libertarian debate.  So I’m particularly interested in libertarian critiques, revisions, or additions to my proposed criteria.