Defaults, status quo, and disagreements about sex

Scott Alexander has a thoughtful piece about who gets to set the default in disagreements about what is reasonable. He describes a couple therapy session where one member is bored with his sex life and goes kinky clubbing, to the anger of his strongly monogamous partner. Yet both want to stay together at least for the sake of the kids. Assuming the answer is an either-or situation where one has to give up on their demand (likely not the ideal response in an actual couple therapy setting), the issue seems to boil down to who has the unreasonable demand.

It resonated with another article I came across in my news flow today: What It’s Like to Be Chemically Castrated. This article is an interview with a man who wanted to be chemically castrated in order to manage his sex addiction and save his 45-year marriage. Is this an unreasonable intervention?

While the man felt he did not have any control over his use of prostitutes, the clubbing man “Adam” in Scott’s example seems to have had free choice: he might desire kinky clubbing, but presumably he could avoid it if he had reason to. The man in the castration article instead felt that his only chance of having his higher order desires win would be to have chemical castration.

However, both are similar in that there is a choice between behaviours for the sake of others. As Scott points out, some demands in a relationship are unreasonable. It is easy to tell one party that even though their preferences are honestly felt, they are too demanding, outré, or otherwise problematic given the context. But the context is a cultural one: what is regarded as unreasonable depends to a fair bit on time and place. This does not rule out some things being universally unreasonable, but clearly the value of monogamy and how acceptable clubbing or prostitutes are does shift. Even if one does think something is ultimately immoral it might still be acceptable enough not to push a preference into unreasonable territory, or vice versa. Demanding that one’s partner always obeys intellectual property laws to the letter might be moral, yet it is eccentric bordering on the unreasonable.

Scott points out that the key issue is how culture sets assumed defaults:

Adam and Steve’s individual personalities and situations will help resolve their conflict, but the tiebreaker vote is always going to be cast by the culture around them.

By setting default culture does not necessary force a couple to act in a certain way. If both want to be monogamous in a swinging culture they are fine to be, but if they disagree with each other the surrounding culture will influence what choice is the least unreasonable one.

So, what about the chemically castrated man? His wife is a conservative Catholic, apparently more concerned with fidelity than sex in the marriage, and he seems to share many of her pro-monogamous values. The couple seems to come from a more traditional subset of US society than Adam and Steve. So here the case is more like adjusting one’s own preferences to conform to higher order preferences set by culture. What makes the case interesting is that the means to do so are biomedical rather than just trying to act in a certain way.

Access to chemical castration is tightly controlled. One could argue that medical gatekeeping ensures that the cultural defaults are upheld: it is conservative, accountable, and integrated with the larger community institutions. This is not a problem when the preferences in a couple disagree and one member decides to move their preferences towards the default position (e.g. someone wanting medication to keep a relationship working). But it is a problem if the preferences disagree and one member decides to move away from the default position. If we imagine the conservative wife or the monogamous Steve wishing to be more tolerant of their partner’s philandering ways they might want a treatment for that (say a psilocybin dose), yet it seems likely that the default societal view is biased against this.

One might argue that this is fair: in a liberal culture couples should be allowed to make joint unreasonable choices (within some limits set by the harm principle) but society has no obligation to give positive support for this. Yet much of the aims of modern psychiatry and medicine are about helping people live lives that function for them and allow them happiness and self-determination, even if they are unconventional. As long as they can participate in society without too much encumbrance to either side, they are healthy. Default-supporting medical gatekeeping is bad in the same way as psychological counselling pushing people towards certain ideals is bad. If social engineering of psychological choices is bad, then social engineering of biomedical choices is at least as bad.

There is some evidence that in heterosexual relationships the interest in sex declines over time for the female while it remains steady for the male (the opposite for cuddling) and this sometimes causes friction. One can imagine couples wishing to correct this biomedically, either by “levelling up” by making the less interested partner boost their libido or by “levelling down” where the more interested partner lowers their libido. Would we say that societal defaults should bias what an individual couple wishes to do here? It seems that the primary goal should rather be whatever makes the couple happiest, and since their bedroom is private, society should be neutral (assuming both treatments are equally safe etc.).

Here the choice seems easier to leave up to the couple to settle. To me this suggests the scenarios in Bostrom and Ord’s reversal test paper. We have a status quo bias plus the cultural default bias: the person close to the default norm has an advantage, as well as the person not changing their state. Someone doing something to reach the default or shifting from the default are somewhat suspect, and someone doing something in order to deviate from the default is really suspect. This gets amplified if the means are unusual. But this is more bias than morally relevant intuitions: there is no information with actual moral content.

The societal default bias may indeed imply transition costs, which are relevant for the advisability of changing the status quo. But again this is not a moral argument, merely a practical one.

One could argue that societal defaults often represent accepted morality, while choices different from them have a higher chance of being immoral one way or another. But given past experience where we think culture has rightly evolved its moral views (often thanks to a few individuals who have differed loudly) and that we accept many forms of pluralism this does not give the defaults much high ground.

In the background we have a major cultural default in our current considerations: we are used to a culture where biomedical interventions are not the default choice. For a long time this has been reasonable since they were not even an option. That is changing in more and more fields. As costs go down, safety increases, and we know better what actually works, we should expect them to become symmetrical with non-biomedical interventions (or the default, as many who worry about medicalization point out).

Today using biomedicine to adjust sexual desires and relationships is if not unreasonable, at least unusual. There is a strong bias towards attempting a social or psychological solution. While this is not in itself ethically problematic, together with status quo bias it means that some people and some interventions will be favoured over others. That is a bias to look out for when making ethical evaluations. Especially since it may look reasonable when it actually is unreasonable.

 

  • Facebook
  • Twitter
  • Reddit

2 Responses to Defaults, status quo, and disagreements about sex

  • Paul Treanor says:

    Once again flawed because the interests of third parties are ignored. There are more than two people involved, and there are specific third parties involved, and not just a generalised ‘culture’ or ‘society’. What do the children think, for instance? What do the prostitutes think? What do feminists think of men who ‘use’ prostitutes? And what do prostitutes think of the feminists?

    And the conclusions don’t really have anything to do with the initially presented cases. In reality Anders Sandberg is concerned about obstacles to the introduction of ‘biomedical interventions’. If that’s what he wanted to write about, it would have been simpler to leave the two case examples out, since they don’t help to clarify that issue.

  • Nikolas Schaffer says:

    What I find quietly comical about the second example is the fact that Catholics are still officially supposed to balk at effective contraception, but are apparently fine with chemical castration. And we’re supposed to believe that hanging onto the faith is an example of “higher order preferences”.

    I’d tentatively suggest that the fellow should have ditched the prostitutes to save his money, and ditched his wife to save his self-respect.

Authors

Subscribe Via Email

Affiliations