Marriage as a Cornerstone Rather than a Capstone

Karen Swallow Prior makes the case for getting married young and thereby having the relationship be the cornerstone of adult life rather than its capstone. Her own experience:

Looking back over a marriage of nearly three decades, I am thankful that I married before going down that road. Now as a college-educated, doctorate-holding woman, I can attest that marrying young (at age 19) was most beneficial: to me, to my husband, and to the longevity of our marriage. Our achievements have come, I am convinced, not despite our young marriage, but because of it.

I don’t doubt that she is correct about her own life. If it works out, an early-in-life marriage makes you far more likely to have lasting financial, mental and physical health.

The critical words though are “if it works out”. Divorce is typically a severe economic hit and good grief can it be nasty (Pryce and Huhne are the current leading candidates for the “most horrific marital crackup of the year” award, but don’t doubt that stiff competition will arrive soon). And of course you are more likely to endure the costs of divorce if you get married young. People face the decision about whether to “cornerstone” ex ante; post hoc everything looks simpler than it is.

Those who favor the capstone approach also have an ex ante/post hoc challenge. Someone at 22 can say “I will do this for a few years, then take a graduate degree in that, acquire this, spend a decade establishing a career in that, experience this and then get married at age 38”, all of which sounds logical, simple and appealing up front. But it can be incredibly hard to find what you consider a suitable mate once your life is fully constructed, something I have termed the “Grandma’s lamp” problem.

Capstoners are gambling just like cornerstoners, hoping that their choice will look as wise post hoc as it seems ex ante. We can all cite individuals who went down one path or the other and had a good outcome. We can do the same for individuals who made the identical choice and had a bad outcome.

Given how idiosyncratic it all is and that each of us votes not for a universal public policy on this issue but only for ourselves, it may be best to have no standing position on cornerstoning versus capstoning in the absence of personally relevant information about yourself and the person you may or may not decide to marry.

Which is another way of saying that your mother and I would like it if you went out on dates more often.

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College London. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over thirteen thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

27 thoughts on “Marriage as a Cornerstone Rather than a Capstone”

  1. Keith:

    Yes, this issue is complicated; who knows, everyone who is delaying marriage into their 30s may just be delaying divorce; it is my understanding that divorce for those 50+ over the last 20 years has really shot-up. In addition, divorces are not all created equal; a divorce when you are longer likely will have less negative financial repercussions since you have far less in assests…

    Over the last few decades, we have changed divorce to a high risk, questionable return enterprise. No wonder people are delaying it and hesitant to enter the legal arrangement…


  2. A big question is also opportunity: Had I met my husband earlier in my life, I might have married correspondingly earlier. (Or maybe not; a major reason for us to get married was the legal aspect, especially with respect to the children we wanted to have. We were otherwise perfectly content to live together without formal state recognition of our relationship.)

    I probably still wouldn’t have married at the age of 19, though; I simply wasn’t mature enough back then for such a commitment, and I was aware of that.

  3. Seems to me there is a false dichotomy here. It’s pretty clear to anyone (ahem) that committing yourself for life to the first person you seriously fall for is a dicey proposition at best. Or second or third. On the other hand, waiting till your life is set in stone has its own array of problems (especially because very few people actually have that kind of stability in their lives any more). So somewhere in between may be a plausible medium: when you know enough about yourself and your desires to have a good idea of whom you might want without having become completely set in your ways.

    But the biggest unspoken part about this is family planning. If you get married/involved early and have kids, the constraints for finding a new partner if/when you break up are incalculably greater than if you are childless — and especially so (in general) for women, on whom the primary responsibility tends to devolve. So marriage becomes a cornerstone simply because to have it otherwise involves stepping off into the quicksand. Which is another way of getting at the other big unspoken part here, namely that a wolf-eat-dog hypermobile (geographically and employment-wise) social and economic system is not a good environment for long-lasting relationships and childraising, at least if both partners’ needs are considered close to equal in priority.

    (My own anecdata involve an early partnership that lasted until she got tenure in another city; a move would have required a complete change of career for me. Both of us have subsequently found other partners with — thus far — apparent multi-decade success.)

    1. One way to approach this is via the Secretary Problem. If the number of potential mates (candidates) is known to be N (and N is large), the optimal strategy is to date N/e candidates, rejecting all. Then pick the first mate better than any of those seen to date.

      Of course, this makes the unwarranted assumption that the number of candidates is known. In the more realistic case that the number of candidates is unknown, but the arrival time distribution is known to be F, the optimal strategy is to reject all candidates until you have reached the 1/e quantile of the arrival time distribution, then choose the first to arrive who is better than any seen to date.

      It’s important to note that these strategies merely maximize the probability that the best possible mate is chosen, and there is a non-negligible probability that someone strictly following this strategy will go mateless.

      1. Both the solutions you propose involve a LONG waiting time, first to scan the candidates, then to find one better.
        Some time ago I calculated that subject to some mild (hah) assumptions you should scan, I think it was 12 partners, and then look for the best partner that was better than those twelve.

        The basic idea is you are assuming that the “desirability” (to you) of the pool of partners forms a Gaussian, and you want someone in the say the upper 5% percentile of that Gaussian. To achieve this goal, the first thing you need to do is know something about the width and mean of the Gaussian (you can’t make an informed choice about how desirable an individual is if you know nothing about the realistic range of desirability).
        You then want to trade-off a variety of uncertainties:
        – you want to scan enough possibilities that you get (with reasonable confidence) a full picture of the range of the Gaussian AND
        – you don’t want the best of your “scanning” possibilities to be so good that the second phase, searching for someone better (ie someone in the now established supposed upper 5%) never succeeds because what you’re actually looking for is someone in the upper .0001%.

        You can use ordinal statistics to establish that 12 is a good guess fairly easily, or you can do a full simulation, Either way, what you get is that if you want to be 95% confident that you’ll find someone in the upper 5 percent, 12 is a reasonable number. If you want to be more picky, the numbers grow very rapidly — so if you want to be 99% confident (ie this algorithm would succeed in 99 of 100 lifetimes) that you will get someone in the upper 1%, you now need to scan 120 partners.
        12 is practical, 120 starts to show that you are being unrealistic, especially when you remember that this is just the scanning stage; you then have to enter into the second “find someone better than this previous pool of 12 or 120 stage”.
        (To some extent how you do this also depends on what you consider scanning. If your pool being scanned is all of, then perhaps you want the upper 1%, and scanning 120 people via looking at their profiles is just fine; that’s ultimately equivalent to your pool being “people on I’ll actually invite on a date”, a pool for which the upper 5% is an acceptable segment.)

        All this is just mathematical fun of course, but there is, I think value here in terms of giving single people around us [I ran these numbers for a friend I thought was being too picky] realistic numbers — the game DOES boil down to scanning the pool of possibilities, then finding the best person given the limits of the pool. And if you insist on Mr 1% rather than Ms 5%, it really IS likely that you will find yourself like Lori Gottlieb, alone at 40+, and author of “Marry Him: The Case for Settling for Mr. Good Enough”.

        1. Maynard,

          I didn’t endorse the strategy, I merely presented it as an answer to the problem of marrying the first candidate who rocks your world.

          Incidentally, your strategy is rather different in that it attempts to establish the range of variation and assumes that the “quality” metric is Gaussian.

          The strategy for unknown N requires knowledge of the interarrival time distribution. Teens in particular are apt to have a fairly high turnover rate. If we assume that the interarrival distribution is uniform on (0, 6 months), then the waiting time to the decision point is shorter: about 2 months.

          I don’t endorse the strategy because there is a significant (about 2/3) chance of ending up with no acceptable candidate. Frankly, I like your strategy better, although I doubt that the “quality” criterion follows anything like a Gaussian distribution.

          I think the point is that sampling the available pool to get an idea of what is available is a good idea, and any “strategy” provides a means to gauge how much waiting is a good idea.

  4. Great analysis above Paul.

    I think what is missing from this analysis is some cross-national comparison. Among OECD countries, those nations with the highest divorce rates tend to be those with the highest gender equality, both formally and informatlly, those being in northern Europe and the US. Those countries around the Mediterranean tend to have both more formal and informal gender roles and responsibilities. My provactive conclusion is that more gender equality results in higher divorce.

    We should fundamentally consider that marriage in the West was always patriarchical; adding equality to the mix of the institution was probably the most radical chance to marriage, much more so than openning up the institution to interracial and gay marriage. And no amount of consuling and pshchological hymming-and-hawing will really stabilize marriage. We need to change the laws to make it a more stable institution.

    1. Frank:

      Do you have any more information beside the divorce rates? Because those are pretty close to meaningless unless we know about length of marriage, likelihood of contracting future marriages (and their lengths) and — as implicitly noted by Katja — percentage of people getting married as opposed to simply engaging in unofficial partnerships. Once again anecdata, but in my experience a substantial fraction of people who marry and then divorce fairly young appear to end up in stable second marriages/partnerships. So if you change the institutions around getting into and out of marriages you can change the stats substantially without changing anything about the underlying patterns. (Imagine, for example, that to live together you had to be “married”, but that getting a “divorce” sans kids was as simple as moving out and completing an online form. The measured divorce rate would skyrocket, but people’s habits would be just the same.)

    2. My provactive conclusion is that more gender equality results in higher divorce.

      Dunno about the rest of the OECD, but there’s certainly no evidence for that at the US state level. Divorce rates are generally higher in the less-equal South than in the more-equal North. The pattern would appear to be the opposite of what you describe.

      1. Good point J on the US South vs. non-South. My first reaction would be that in the South, it is a lot more culturally acceptable to marry young (Red State families vs. Blue State families), and thus results in the higher rates. And if you believe relative poverty results in more family instability, that could be a driver in the South.

        I think in the South, a wealthy, strong man providing for his children and wife is more of a cultural ideal than in other parts of the country. But I don’t think even Southern women or most men are down with patriarchy, formal and informal. Easier to point out Paul the Apostle attacking homesexuality than promoting patriarchy. Also, some of the patriarchy I think can only be supported with low costs of food, housing, etc.

        I also think gender norms may even be stronger in the Mediterranean than in northern Europe…

        I do need to take a look at the divorce rates again. It is my recollection that what drags down the divorce rate overall is low rates in the upper-Midwest and Northeast. Cali and Nevada, for example, have above average rates.

        1. paul:

          You are right that divorce rates is a very crude measure, sometimes concealing more than it reveals. I believe Paul Krugman once said that divorce rates between states were hard to compare and measure. My idea of changing institutions is making marraige less risky-less common property, clearer initial rules on child custody (and making them with less of a sexist lean towards women), and mandatory, comprehensive, pre nuptial agreements.

          I think in regards with pre-nupts, British Columbia and a lot of the US West Coast is moving that way, as they have on family formation policy over the last few decades…


          1. Divorce rates are difficult to compare between States because we are a fairly mobile society. My ex- and I were married in Kansas, and our divorce was granted in Illinois, a State in which I never lived. Our marriage was roughly evenly split between Kansas and New Mexico. Do you credit the divorce to Kansas (where we were married), to New Mexico (the last State where we cohabited), or to Illinois?

            Second marriages actually have a higher divorce than first marriages. Or, as Samuel Johnson put it, “Second marriages are the triumph of hope over experience.”

    3. Surely the goal is to maximize human happiness, or human flourishing or something?

      Minimizing the amount of divorce strikes me as pointless in-and-of itself, only of value insofar as it’s a good proxy for the two earlier items. And it is NOT a good proxy unless we know for a fact that the post-divorce couples (considering the whole span of their lives) were unhappier than the married-but-suffering-in-silence couples.
      All of which is to say that pondering divorce statistics between cities, states and nations, may be a fine thing for a religious fundamentalist to do, but strikes me as an utter waste of time for a policy maker, in the absence of substantially more information about happiness within marriages vs after marriage in these different cities, states, and nations.

      1. I think determining human happiness within marriages vs. after marriage would be incredibly problematic. I do know though that divorce is one of the things that most depresses people, and has high costs later in life, risking oftentimes retirement security.

        Minimizing it, and the associated depression, is a worthy policy goal, especially if it is done via preventiative, relatively cheap measures.


        1. “divorce is one of the things that most depresses people,” indeed, the causation might actually run in the opposite direction, as studies have been done to suggest that divorce is disproportionately precipitated by depression among one or both of the spouses. It’s really hard, isn’t it, to improve the odds for human happiness by prescribing any general course of action? Some people will be unhappy no matter what, those who are already unhappy are less likely to get married, or to stay married, all of which makes our ability to measure the impact of marriage on happiness quite tenuous. Marriage may be merely a proxy for people who are not challenged by particular threats to happiness, like depression or an unhappy childhood.

  5. I always wonder if early marriage advocates don’t really like sex very much or are not interested in it.

    Sex is a learned skill, or set of skills. In learning it, a person benefits from having a wide variety of mentors, so that the person gains the benefit of different experiences and learns what pleases him- or herself as well as his or her partners. In other words, if you really want to experience great sex, you need to get good at it. (I do realize that a lot of married assertedly monogamous couples who did not have a lot of premarital sex report great sex lives– this is almost certainly, however, a function of people literally not knowing what they are missing, like the 1950’s housewives who reported satisfaction with their sex lives but who didn’t know what an orgasm was.)

    Now, it’s possible to have an early marriage and still climb the sexual learning curve. But you would need some sort of polyamorous or swinger marriage to do it. And of course, the same type of people who advocate early marriage are also aghast at the idea that anyone might not be monogamous.

    If you actually care about your sex life, early monogamous marriage is simply a very bad idea. It’s likely to stick you with an inadequate partner and with few tools to improve your sex life on your own.

    1. “this is almost certainly, however, a function of people literally not knowing what they are missing”

      Or, alternatively, two people learning how to please each other, rather than how to please the average person. Developing specialized, rather than general, knowledge of the subject. It’s worth considering, anyway.

      1. Sorry Brett. That’s not how sex works. Just about everyone who is good at it learns by trial and error. The first time any person performs oral, they are terrible. Then they learn what works and what doesn’t work. And the way they learn what does and doesn’t work is because a partner who is knowledgeable tells them. You need MENTORS, i.e., people who are more experienced than you are, to tell you how things work.

        You wouldn’t learn any other skill by taking two completely inexperienced people and having them try to learn it from each other with no guidance. You wouldn’t learn how to play the piano by doing that. You wouldn’t learn how to drive by doing that. You wouldn’t train employees like that.

        The only reason anyone believes that BS about sex is because of the traditions peddled by religions that teach us that our creator is obsessed with what we do with our genitals and will torture us for eternity if we have too much fun.

        It may be nice and romantic to believe that two inexperienced people can learn everything they need to know about sex without ever getting outside help, but it is wrong, and early marriage (unless polyamorous) is wrong for this reason. (Indeed, this point is so fundamental it also disproves the religions that teach this garbage. God is not that stupid.)

        1. Dude, like, take your tantras (along with your false analogies, overgeneralizations, confidently uttered platitudes, and highly conventional attempts at shocking your audience out of their bourgeois complacency) over to Haight. There you can argue to your (heart’s?) content with the other Armistead Maupin bit players about which of you invented sex.

    2. How would anybody ever really know? And couldn’t the answer be different for different people? Couldn’t there be very shy people who will never achieve their “optimal” sexual satisfaction except with one or a few long-term partners (i.e., spouses)? And while your definition of “adventure” might be heavily dependent on the concept of novelty, clearly, that will not be true for every person.

      Whether the subject is sex or religion or marriage, the refusal to start with the individual’s need is usually a prescription for unhappiness, boredom, and dissatisfaction. It was true for 1950s housewives, and it’s just as true for the hook ups of the 2000s.

      1. How would anybody ever really know?

        We would know because (1) this is how we learn anything else– other than sex, I can’t think of a single thing where people think the best way to learn is to put a couple of inexperienced people in a dark room together with no instruction, and (2) because there are actually plenty of memoirs of people who had sexual awakenings later in life after thinking they were happy in an early monogamous marriage.

        To be clear, my claim is not that everyone needs novelty. It’s that everyone needs to learn, which generally requires multiple partners including some partners who are more experienced. Once that experience is gained, there can be all sorts of reasons to be monogamous and married rather than in other arrangements. Getting into an early monogamous marriage short-circuits that learning process, which serves the interests of people who don’t want anyone to be having sex outside of early monogamous marriages for religious reasons, but doesn’t serve the interests of anyone who actually wants to maximize his or her sexual pleasure.

        1. “because there are actually plenty of memoirs of people who had sexual awakenings later in life after thinking they were happy in an early monogamous marriage . . .”

          Data is not the plural of anecdote.

          People who leave a marriage have plenty of incentive to justify their actions to themselves and to others. It’s also the case that the sexual landscape has change a lot from even 40 years ago, with the availability of contraception. Indeed, one of the saddest aspects of marriage is when people affirmatively stop seeking sexual pleasure because of the fear of having additional children. Experimentation requires lack of inhibition whether it takes place with a spouse or anyone else.

          I am not arguing for sexual exclusivity or early marriage, far from it, I think the latter is usually a mistake and the former too likely to prompt the latter without proper reflection, but I find your reasoning to be less than persuasive.

          1. Well, I don’t know that we all need to agree on this, but. Ime, chemistry and emotion and communications/relationship skills are so important as to almost overwhelm any technique that one may or may not have. And even a measurement of a technical skill would be subjective. So, no, it’s not like learning to make an omelette, imho. And plus, depending on the type of relationships one has, it may be possible to have *too many.* One might just grow in the *wrong direction* too.

          2. NCG:

            That’s what people want to believe, because they want to think that marrying young and staying monogamous won’t lead to bad sexual outcomes. But the nation’s sex therapists’ offices are filled with couples who have fine “chemistry” but don’t know how to please each other. Meanwhile, swingers and polyamorists are having a lot of fun including with people they have no “chemistry” with. And guys (and smaller numbers of women) are hiring prostitutes/gigolos who know how to please them.

            That picture doesn’t suggest that “chemistry” and the rest of it overwhelms technique. It suggests the opposite.

Comments are closed.