The correlations are clear but causality less so. What if it’s not marriage that makes men happier, less prone to crime, and longer-lived - rather it is that women select for partners who are happier, less prone to crime, and healthier?

It's weird to try and explain this from a sterile alien-studying-humans perspective, but since that's the HN theme I'll do my best. Marriage and having dependents carries with it a lot of economic cost. By revealed choice theory, that implies it has significant benefits, or else nobody would ever do it.

I would be interested in a citation that those entering into marriages and child rearing are aware of the economic costs and are able to enumerate the benefits prior to the events, or if its look back justification after the fact. If you dig, I think you'll find most folks were not aware of the economic and opportunity costs of children, as well as the work involved in maintaining a healthy relationship with a party for an extended period of time.

(removed a bunch of pew research citations that made the comment unwieldy; happy to cite for those interested)

> It's weird to try and explain this from a sterile alien-studying-humans perspective, but since that's the HN theme I'll do my best.

This made me laugh, in a good way :) I'll approach it from that angle:

Monogamous child-rearing looks to me a successful evolutionary strategy for the human race (and keep in mind the "monogamous" part might be relatively recent). It implies less about the happiness of the parental units, though.

The Terran octopus dies off soon after giving birth to its progeny, and this is successful for octopus-kind, but results in no long-lasting happiness for the octopus mother. Likewise with many species of insects, arguably very successful lifeforms on Earth. Many of them die or are cannibalized after mating or giving birth.

Successful species propagation strategies do not necessarily make the parents live happier lives. There is an evolved reason for said strategies, but is happiness the maximized factor?

Good point. Evolution will use anything including happiness to propagate genes.

Well if we want to be all logical and science-y about this, we can't assume humans are rational beings. People obviously make bad decisions all the time.

Just because lots of people have kids, for example, doesn't mean that having kids makes people happier, and in fact studies seem to show the opposite to be true on average. That said, I'm not trying to say that having kids is always an irrational choice, and certainly it makes some people happier (or it might make people "unhappier" but lend them a greater sense of meaning and purpose that balances things out). But clearly you can't just say "well, correlation must imply causation because otherwise why would people do X if it makes them less happy?"

You might be surprised to learn (I was) that it [kids making people "unhappier"] depends on the country https://time.com/4370344/parents-happiness-children-study/

People don't necessarily optimize for maximum happiness. It's not a good target metric.

Revealed preference doesn't always work well as an explanation for why people do what they do though. E.g. if someone falls for an investment scam a-la Madoff, the person obviously didn't want to end up destitute because of that.

A closer example here might be a salesperson performing a "hard sell" on an automobile that is just at the edge of affordability for the buyer. The buyer really wants to be able to drive around in this cool looking automobile, but they end up with a lemon that they can't afford to keep drivable and sucks up all their resources. It's not so hard to draw a parallel from that to someone who has a family on accident.

This is counterintuitive. The obvious benefit of marriage is producing children.

Producing children is extremely costly for both parents. This is true both in animals and humans - you would generally not say that rearing children is good for the health of the parents. It probably brings emotional benefits to the parents (which it would have to, otherwise they wouldn't do it), but there's no reason to assume that couples are automatically better off in terms of their finances or physical health than singles.

Edit: Since people correctly pointed out that you can have children without marriage, please replace marriage with "romantic relationships that produce children".

> The obvious benefit of marriage is producing children.

No. Marriage doesn’t produce children.

Marriage can provide social obligations relating to the support of children, though. (It also provides social obligations of mutual support between spouses.)

You can produce children without marriage and a lot of people don't get married to have children.

> The obvious benefit of marriage is producing children

Have been married a long time. Today I learned from you that apparently we are missing out on an obvious benefit

Does marriage have economic costs? Dependents sure, but marriage seems to be economically beneficial: taxes, fewer bedrooms, etc.

> or else nobody would ever do it.

People often do things that are not good for them.

Or perhaps the benefits haven’t decreased but the costs have increased. Especially up-front.

To figure out the causality, you'd need an experiment where people are randomly removed from relationships though random external events.

We do have an imperfect example of this, in the case of widowers. When someone is widowed, does their happiness tend to increase, remain level, or decrease?

I think it goes both ways. Women obviously do prefer partners who are successful, higher in socioeconomic hierarchy. Yet not being able to find a partner may lower one's self-esteem, which in turn may reduce the likelihood of doing things that increase chances of socioeconomic success.

I mean, advancing in life generally requires leaving your comfort-zone, but that may be hard if you lack confidence.

In fact, AFAIK, one issue with incarceration is that you are not seen as suitable partner anymore. Turns out that both men and women tend to avoid partners with criminal record.

I think there is a really good point here. However, it should be recognized that some of those studies do not control for divorced vs never married very well and the magnitude of the benefit of marriage is over-stated. Not to say marriage does not have health/happiness/other benefits, but the effect is smaller (although the Harvard study seems to control for it certain cases, in strange ways).

There's a correlation/causation issue here.

The GP pointed out that the relationship described in this has red flags because the relationship alone is the source of happiness. It would lead to a happier, possibly longer lived, less criminal person.

You seem to be implying (and I don't think this is intentional) that close personal and romantic relationships [for straight men] == a wife. But that doesn't necessarily need to be true. Maybe for romance, but certainly not for close personal friendships.

You're correct that a person isn't an island, but the focus on single romantic partner may be to the detriment of other forms of relationship which are still hugely valuable health wise.