Modal wives and why it is hard to marry well

I define a modal wife (or husband) as a person you would have married (could have married?) had you met them at the right time, unattached, and under normal life conditions.  The number of modal wives is typically greater than or equal to the number of real wives, although clever philosophers will recognize possible [sic] counterexamples. 

Under one view, you have hundreds or thousands of modal wives, most of whom you never meet.  (How many does the average person meet, how soon do you know when you meet one, and how confused would you be if they were all in the same room at once?)  Your correct dating strategy is to cast your net very widely, and hope to find and marry one of these people. 

Under another view, modal wives are no big deal.  Your so-called "modal wives" are no better for you than, say, the best woman you could pick out of a lot of thirty eligibles.  The key inputs for a good marriage are attitude and a minimum degree of compatibility, not search and discovery.

If this is true, searching for modal wives, or perhaps even thinking about the concept, can make you worse off.  The quest for the perfect mate makes it harder to come to terms with what is otherwise a compatible marriage.  Which perhaps is all you are going to get anyway.  Marriage is good for you, and don’t be too fussy, this is not iTunes.  Too much choice, or too much perceived choice, is problematic.

The two views offer directly conflicting advice (TC: My views are closer to the first position, although attitude remains all-important).  Yet we may be uncertain which view applies to us and to what extent.  You could put all your eggs in one basket and pursue just one strategy, but what a risk if you are wrong.  You could act upon some weighted average of the two views; I suspect this is what most people do.  But then the two strategies are constantly undercutting each other.

That is one reason why it is hard to marry well.

Addendum: Here is a good post on Deception Island, and do also read the excellent comments thread on this post.


Speaking as one who has tried both strategies, I strongly suspect that the viability of the second strategy is strongly dependent on whether the person in question adopting the strategy has traits (religious beliefs, intelligence, cultural tastes, dietary preferences, social and political views, sexual attitudes and tastes, physical appearance) close to the societal mean. (And, on another dimension, an extrovert will also have more modal wives than a misanthropist.) A 5'4" Orthodox Jew libertarian or a socialist vegan leather fetishist is going to have fewer modal wives than a moderate Christian with a 110 IQ—and, moreover, the marginal utility of a modal wives for the first two is going to be substantially higher than for the more average gentleman, since the wife selected from a pool of thirty is more likely to be close to his modal wife.

One can adopt a combination of the first and second strategies by using a pool more likely to produce a modal wife: an Orthodox Jew could restrict himself to meeting other Orthodox Jews; a genius economist stays away from the sports bars and instead trolls bookstores and academic events. But the more dimensions along which one's traits are unusual or restrictive, the harder it becomes to adopt even variants of the second strategy without coming across fatal incompatibilities. Woe is the Orthodox Jew who is only attracted to lithe blondes; a politically-active Republican vegan is likely to have trouble as well.

Separately, given the requirement of consent in modern-day marriage, there's a significant difference between "would have married" and "could have married," though "would have married" has its own ambiguities.

I think Tyler's model is dynamic enough to take into account income and intellect; it merely assumes that there exists a set of women who meet the criteria

f(w-sub-1, w-sub-2, ..., w-sub-n) > W

and that the search costs for finding a member of that set may be high enough that it is better to satisfice at a level X beneath W, because (NPV(W)-NPV(X)) < (NPV of cost of finding W - NPV cost of finding X). Nothing stops income or intelligence from being in that function. Alex does raise a different issue, which is whether the search cost is positive at all for some people. Smart successful good-looking libertine men have little incentive to satisfice below a relatively high W to the extent they derive utility from new sexual conquests. A religious person who objects to premarital sex, on the other hand, would derive additional utility from marriage the average Sex-in-the-City denizen doesn't, and we would thus expect to see what we actually observe, which is shorter search times and younger marriages amongst this population. Of course, this is also consistent with my theory that such a person is likely to have a larger set of modal wives to work with, and that satisficing involves smaller costs than for the idiosyncratic; I suspect all three factors come into play.

This brings up the so-called Sultan's dowry problem... Assuming you do meet a likely prospective spouse under the right conditions, do you marry him/her immediately, or wait and keep searching, and if so how long? The longer you delay, the greater the chance you will meet someone else whom you like even more, but at the risk of losing your current prospect.

I believe the mathematical answer is to estimate how many prospective spouses you will meet in your life, and then choose the first one better than the first 37% of them. I wonder if anyone has actually done this!

Also, if you assume there is a strong ranking between modal spouses (and not just a binary yes-or-no evaluation) then it seems essential to meet as many of them as possible. I bet the increasing use of on-line dating sites and ease of travel in modern society are leading to more well-matched marriages.

Personally I estimate there are around 50,000 modal wives out there for me.

Wait -- Why do women past 30 lose enormously by waiting?

This from a 27-year-old who doesn't consider her happy single life as "waiting" --

Actually, I personally don't thnk that is anything such as a modal wife !!
The cardinality of the modal-wife is mathmatically equal to only 1. That is , all
aspects compatablity at every attribute being met to be that persons choice is limited to only one.

But Ted also proposed a good value proposition. Still need to wrap my head around his alogrithims that he outlined.

For these purposes I believe in intelligent design, or psychological engineering. I don't expect that it's any easier to find a modal wife than to find a perfect unoccupied home. If you want something done right, you have to do it yourself. I know that everyone else says that Pygmallion is a bad idea, but they just don't know what they are talking about. OTOH, most people probably should build their own home either. If you aren't able to figure out how to make something work well, you probably should leave it to experts or to chance.

Too much thinking, try this:

ARTIST: Lovin' Spoonful
TITLE: Did You Ever Have to Make Up Your Mind

Did you ever have to make up your mind
Pick up on one and leave the other behind
It's not often easy and not often kind
Did you ever have to make up your mind

Did you ever have to finally decide
Say yes to one and let the other one ride
There's so many changes and tears you must hide
Did you ever have to finally decide

Sometimes there's one with big blue eyes, cute as a bunny
With hair down to here, and plenty of money
And just when you think she's that one in the world
You heart gets stolen by some mousey little girl

And then you know you'd better make up your mind...

Sometimes you really dig a girl the moment you kiss her
And then you get distracted by her older sister
When in walks her father and takes you a line
And says, "You better go home, son, and make up your mind"

And then you bet you'd better finally decide...

The assumption underlying this entire discussion is that economists can use such terms as search costs, satisficing and modal set in a discussion and still get a decent sample of phone numbers from said prospective mates.

So this disussion is more on the macro side of sexonomics. On the micro side, you still have to use game/IO theory to get phone numbers.

When people meet for the first time, they go through a routine: what do you do, where are you from, are you here with friends, how old are you, etc. The key would be product differentiation. As a potential mate, you would have to first assess your own value on the market and work with that. Then you have to demonstrate that value (use salesmanship)in a social situation and then differentiate yourself from the other dozen guys a girl has talked to that night. Differentiation would probably be key when it comes to getting phone numbers from which to expand your pool of modal spouses.

Does anyone remember the scene in a Beautiful Mind where John Nash and his buddies score with the ladies and he supposedly comes up with the idea for Nash Equilibria?

"The assumption underlying this entire discussion is that economists can use such terms as search costs, satisficing and modal set in a discussion and still get a decent sample of phone numbers from said prospective mates."

I believe in assigning value to things.

Hey, my personal ultramodal dropped me an e-mail to call my comments a "thing of beauty." If my girlfriend didn't appreciate these things, I'd just be satisficing.

"Sorry, you 130-IQ people, you are just not that special."

I'm thinking more of the 140-150 IQ people. If intelligence has a bell-curve distribution, and intellectual compatibility requires a mate's intelligence to be within a range of +/- n standard deviations from one's own, then the person who's three or four standard deviations from the mean has a much smaller pool than the person who's near the mean. You can perhaps disagree with the premises (if n is large enough, the effect disappears), but the math is inexorable.

"no matter which theory you believe in, if you want to find a good spouse, you are better off spending your time and money on self improvement (lose the gut! read real books! cultivate your garden!) than on mate-searching"

That's not inconsistent with my model, which acknowledges the importance of the intersection of the modal sets. Too, self-improvement has its own utility benefits outside of mate-searching. There comes a point where additional marginal time in mate-searching outweighs additional marginal time in self-improvement, however.

"IMHO, it is easier, not harder, for people with unusual preferences to find mates, at least in the Internet age."

I agree 100%, but that just increases the expected value of the small-modal-set strategy and the opportunity cost of satisficing.

"This discussion has so far ignored the big danger of the "modal spouse" theory -- self-deception."

Self-deception is one of many rationalizations for paternalistic policies that reduce utility. At the end of the day, it's extraordinarily more likely that I know what's better for me than anyone else will, even to the extent I deceive myself into confusing my potential mate's youthful complexion and callipygian figure with traits that will actually bring me more utility, such as her appreciation for and willingness to engage in abstract discussions of how different legal regimes for divorce and marriage interrelate with utility and dynamic modal sets. Otherwise, there'd be more than a fringe market for professional matchmakers, and the matchmakers would be more willing to disclose success rates and offer a pricing structure that wasn't so heavily weighted on non-negotiable up-front non-refundable payments.

But you make a strong case for having a preference for having a preference for the "satisficing" strategy. Those with larger modal sets are more likely to be happy with their mates (and for longer periods of time) than those with smaller modal sets.

Thought experiment: Would Tyler be happier if he derived great pleasure from dining at Taco Bell than he is with his smaller modal set of preferred Mexican cuisine?

NB that "having a preference for having a preference for" in the 9:41 comment was meant as "having a meta-preference" and was not a typo. There were times at my life when I had a preference for having a preference for satisficing; alas (or fortunately, as the case may be), I did not actually have a preference for satisficing, and regularly failed in attempting to exercise that strategy, sometimes multiple times with the same person.

Modelling of modal marriages must include opportunity costs of search vs. opportunity costs of marriage. Divorce must be accounted for, and that factor is missing in the cursory analysis. Lastly, the existence of multiple modal partners is a multioptima solution problem. Hence, a new economic model of polygamy may be drawn. Ph.D. proposals, anyone?

"The assumption underlying this entire discussion is that economists can use such terms as search costs, satisficing and modal set in a discussion and still get a decent sample of phone numbers from said prospective mates.""

Think of it as part of your search strategy. A woman who reacts negatively to the economic way of thinking may not be the right mate for an economist.

Voltaire also wrote: “†¦ if I had a dollar for every overweight, immature 30-ish or 40-ish male sports fan I've met who seriously expects to marry a thin, superhot blonde woman, and then wonders why he has trouble finding marriage-worthy dates.†

I wonder whether there might be some evolutionary advantage conferred to a species some of whose members price themselves out of the mating game.


If you've never had a cute girl at a party effectively end a conversation by accusing you of being a genius, good for you. High-IQ might not be very special, but it is rare (especially >140), and someone who makes you feel inferior simply by opening their mouth does NOT make a good mate.

This discussion has repeatedly touched on an issue which the theory appears to ignore: absolute constraints. A religious Jew & a religious Christian aren't going to work out unless someone changes views. In my personal case, I informed my now-wife that my children would be home schooled months before I mentioned marriage.

The vegan Republican example is not really of this class. There is plenty of evidence that conflicting political views can be part of a successful marriage. Depending on the nature of the veganism, this might not be insurmountable either.

My father was adamant that, after some minor absolute constraints, any pair can make it work. His focus is on the determination to make it work, and that, I believe, is probably a much better predictor of long-term happiness.

Could I have married "better"? Almost certainly. Am I stupid enough to let that get in the way of enjoying my marriage/family/life? Not for a million dollars.

Can't let this one go by:

From Ted:
"While John Derbyshire's infamous claim that women's peak attractiveness is shy of 20 is certainly an exaggeration for most men, a 37-year-old single male does not face the same fear of depreciation that a 37-year-old female does."

First of all, if we're talking about reproductive fitness, this is one that's likely to equalize as medical knowledge trickles down into the general population: an increasingly large number of reproductive genetic defects are known to be caused by poor sperm quality. How do you get poor sperm quality? Simple. Start with good sperm, then age 40+ years.

Second of all, an otherwise dubiously feminist set of technological advances is likely to settle the question, rarely even asked: do "older" (thirtysomething, say) women marry even older (40+) men because they want to, or they think they have to? I would argue that as women in their 30's see their attractiveness/power horizon lengthen due to better safety, affordability, and social acceptability of cosmetic surgery/cosmetic procedures, men in their 40's may be surprised to find themselves passed over in favor of virile young studs. Call it the Demi Moore effect.

Nathan Zook: my point about IQ applies to 140 as well as 130: if you have a 130 IQ, your job and social circle likely include many people with a 130 IQ and some with 140+. If you have a 140 IQ, you probably know many others with a 140 IQ and some with 150+. How relevant this to this thread depends on how we define "eligibles" -- are they a random sample of the general population, or a random sample of age/marital status-appropriate people the selector is likely to meet? I am focusing on the latter definition, since my selection bias is too great to speak to the general population.

Peter: *great* point about B-list women who think they are hot and won't talk to anyone nerdy. I would never have thought of this point, since, well, none of those girls ever talked to me.

Ted: I love your thought experiment. I intentionally make room in my life to enjoy pop culture and some trashy sci fi shows because I fear I will be unhappy if I raise my artistic standards too high. I would not like being a movie critic. But if anyone wants to buy me a $200 bottle of wine, I will take the risk of drinking it and ruining my appreciation of $10 wine forever. ;-)

I wonder how much mass media attempts to influence us into acting strategy one, when we
would be better off acting strategy two, or at least weighting towards strategy 2 on a continuum.

"The higher your threshold, the more likely you are to have overlooked some minor flaw. " That's why the only trait that really matters is ability and desire to correct flaws!

Russians are NOT happy with traditional sex roles. In fact, they aren't happy about anything, but Especially not about anything having to do with family.

Regarding IQ
a) There are no social sets with average IQ much above 140
b) Even if there were, mental abilities correlate much less well at that range. Significantly above 140 many people may have profiles of abilities that don't resemble anyone else
c) At IQs significantly above 140, gender imbalance is a major issue.
d) High IQ Jewish men may have it slightly easier than others because their mental profile favors verbal over spacial ability, making them somewhat more like women. The gender and ethnicity differences may partially cancel out.

140 IQ isn't so rare. About 1/261 have IQ 140 (or more assuming normal distribution with 15 sd). Assuming 4000000 for each age in years in the United States this is about 15000 people or 60000 for a 4 year cohort. This is about the same as the student population of ten top colleges (Harvard, Princeton, Yale, UPenn, Duke, Stanford, Caltech, MIT, Columbia and Dartmouth). The average IQ of students at these schools is probably close to 140.

150 IQ is getting rarer. About 1/2330 at 150 or above. So the average IQ of the top 10% at the above elite schools is probably close to 150.

160 IQ is pretty rare. About 1/31560 at 160 or more. So even at the above elite schools such IQs are rare probably less than 1% of the students.

Of course the above assumes IQs are normally distributed which I doubt is entirely accurate particularly at the tails.

michael e vassar, do you have a source? People with 135+ IQs are about 1% of the population and have average IQ of about 140. This is about 160000 people for a 4 year cohort. If one third of them attend the top 10 schools this will fill out the student population with average IQ of about 140. Are Yale and Princeton really far behind Harvard in average IQ?

I hate to be terribly unromantic, but I have always said that most people marry the person they happen to be going out with at the point in life they decide they are ready to get married

Ted and Coyote: great posts (the ones right above this one!)

Michael Vassar: I went to one of the top Ivy schools, and the only people I ever encountered there who disliked economics or economists were the physicists and the occasional communist! I suspect the hatred of economists you have seen has more to do with politics or tactlessness (Larry Summers, prince of willful tactlessness) than with IQ or unpopular truths. It may not be a coincidence that my school's economists were/are popular, and that our econ department is better known for contributing to the NYT oped page than to the Libertarian party.

michael vassar, 25 and 75 percentile scores for combined SAT (2004-2005 freshmen) are Caltech(1450,1570), Harvard(1400,1580), MIT(1410,1560) and Yale(1400,1560) so there is no big drop to Yale. Then there is a drop to Princeton(1370,1560) and Stanford(1370,1550), a drop to Dartmouth(1350,1530), Columbia(1330,1540) and Duke(1330,1530) and another drop to UPenn(1330,1500).

Using percentile ranks to convert median SAT scores to IQ (without regressing) then Caltech is about 140 and Duke is about 135 (estimating the median as the average of the 25 and 75 percentile scores).

OK, Yale's better than it was in 1996 when I went to school.
Note this study which might be relevant to other selective measures at elite schools, though I strongly doubt the finding of no correlation between IQ and self-discipline.

Reading these blogs beats movie entertainment. Such thoughtful and intelligent men might find greater success in finding compatible mates if they wean themselves from the breast of digital pixels and get out and meet women. I ask, when was the last time you introduced yourself to a woman who caught your eye?

If you need to think about the strategy in dating, then you are probably not attractive enough. Alpha males just get girls just like hot babes get guys all the time.

Comments for this post are closed