Marriage I: Commitment

The basic elements which give rise to (nuclear) family structure are:

  1. Mutual expectations about fidelity, financial support, and inheritance rights
  2. Valid grounds for those expectations
  3. Communal consensus on whether someone’s expectations are valid, including:
    • … consensus on what types of evidence demonstrate commitment
    • … availability of witnesses to reporting/corroborating the existence of such evidence
    • … collective judgment that X signaled commitment to Y by doing Z
  4. Communal coordination on the commitment signals
  5. Power over/responsibility for another’s commitments, in particular:
    • … decisions about resources and opportunities that X needs to have available in order to commit to Y
    • … veto over X’s ability to signal commitment to Y
  6. Communal ratification of valid mutual expectations
  7. Communal enforcement of valid mutual expectations

In the sequel to this post, I will try to lay out how these elements interact to determine the shape of courtship rituals and family life. Loosely speaking the point I hope to make is that traditional family structures have many virtues (corresponding to the various elements which provide their foundation), but to realize these virtues — indeed, to operate at all — a functional tradition relies on coordination between all of these different elements. This coordination limits the feasible permutations of the elements such that the traditional virtues they can embody cannot all be realized simultaneously. Worse still, the necessity of coordination means that none of these elements can be “restored” in isolation, and that half-forgotten standards of traditional family life are meaningless in a society where no one abides by them.

That is the destination. But first I need to explain what commitment is and how it enters into courtship.


The idea of commitment, which has a fruitful (but potentially confusing) ambiguity. Someone commits himself to a certain course of action when he has decided to execute some plan like A>B>C and then takes the initial step, A, that locks him in, forcing him to follow through with B and C. For example, a parachutist may consider a parachute landing on a certain island, and he may even talk about this plan, but he can always revise his opinion. However, once he has (A) jumped out the side of a plane with a parachute on his back, he has pretty well committed himself to (B) opening the parachute and (C) floating to a safe landing on the island below. The alternative would be gruesome.

The ambiguity arises because a “commitment” refers both to what someone has committed to do and also to the evidence which gives proof of the commitment. Thus for our parachutist, we can call the plan to open his parachute and land on the island a “commitment” (rather than a mere intention or whim) once he has already jumped and is sure to go through with it, or we can call the jump itself the commitment (the act which irreversibly commits him to his course of action).

A commitment does not have to be quite so drastic as jumping out an airplane —although in everyday life, naive young men and women eagerly rush into commitments no less drastic. In general, for a certain course of action to function as a commitment it only needs to be consistent with one of an agent’s future choices, but inconsistent with another; or at least a probable fit for one option and an improbable fit for the other.

If you see me bringing suitcases into my car and you know I had the option of driving to Québec or to Miami, and then next I drive onto the northbound on-ramp, you may well assume I am going to Québec. There are many reasons one might begin a road trip to Miami by heading north, but on the whole the likeliest thing is that I am actually heading to Canada. But if all along my plan was to head north to fuel the car at the nearest gas station before starting out, or to route around a traffic jam, then I am not set back at all when I reorient myself and finally start driving south.

On the other hand, if I drive north for three hours and then change my mind and decide to go to Miami, then I have pointlessly sacrificed three hours of my life. So getting on the northbound on-ramp is a clue that is consistent with my going to Québec, but driving north for three hours is a commitment to that option: those three hours are time well-spent if I do want to go to Québec, but time wasted if I don’t and counter-productive if I’m actually heading south.

Signals are Commitments

Compare the following:

  1. In theory, an observer seeing me take the northbound on-ramp might assume I am headed north.
  2. I notice my neighbor watching me as I drive towards the northbound on-ramp, and I know he will make this inference.
  3. I have told my neighbor that if I take the northbound on-ramp, I am going to Québec.
  4. I know my neighbor will make some choice based on whether or not I am driving to Québec, and so may be trying to figure out my destination.
  5. I have told my neighbor I will take the northbound on-ramp so that he will know whether or not I am going to Québec.

In the first case, I may not even be explicitly aware that the direction I am driving suggests certain destinations over others, although if someone called my attention to the fact it would be immediately obvious. In the final case, not only am I aware what signal I am sending, I created the signal myself. I could just as easily have arranged another signal, like a red bandana hanging in my window; whatever signal I name will become the signal my neighbor will look for.

Unless I really dislike my neighbor, I don’t want to create problems for him. If he needs to make a choice based on whether I’m going to Canada or not, I’d prefer that he have the information; certainly, I don’t to give him misinformation. If I know what sign he’ll be looking for, I may make sure he get the “proper” sign even at a slight inconvenience to myself. If he’s expecting a red bandana in the window and suddenly I decide I’d rather wear my red bandana while I drive, well, tough — unless I am willing to inconvenience him, I’ll have to settle for a blue bandana.

At times I may have to make sure he gets the “proper” sign, at a slight inconvenience to myself, even though the only reason that sign ever became a sign in the first place was because of its convenience! For example, if my neighbor expects me to take the northbound on-ramp if I’m going to Canada, but I remember that gas is 3¢ cheaper at the gas station just to the south, I may have to inconvenience myself by heading north to give him the impression that I’m taking the most convenient route to Canada.

If I know what my neighbor’s expectations are, and he knows that I know what his expectations are, then he can have even more confidence that I’ll act in accordance with his expectations — after all, I don’t want to leave him in the lurch. A further implication is that when I arrange a signal with him, arranging the signal is its own commitment.

Whether or not there is any logical connection between red bandanas hanging in windows and visits to our neighbor to the north, once I’ve created the expectation I know that my neighbor will make a misinformed choice if I signal “I’m going to Québec!” and then drive to Miami instead. And besides the respect and friendship I feel for my neighbor, if I arrange a certain signal and then act differently, my neighbor will blame me for his misinformed choice. He will be irritated, perhaps outright angry; my misleading signal may lead him to think I’m a bit of an asshole, or even to suspect that I tried to hurt him intentionally.

The more explicit and clear the arrangements for the signal are, the more careless (malicious?) it looks after I send the wrong signal. By putting neighborly harmony on the line, I create a cost where none had existed before and thus create a way to commit to the option I have chosen.

If I don’t explicitly arrange a signal in advance, but I have been driving one way to go to Canada and the other to go to Florida for years and my neighbors expectations are public knowledge, the same considerations apply — albeit with more obscurity. I can pretend that I didn’t know he was watching me drive onto the highway, or that he inferred a destination from that observation; alternatively, I can say I knew he was speculating about where I was going, but I had no idea that he was using the information to make important choices. (Seriously, why does my neighbor care so much about where I’m going on vacation?)

Nonetheless, if I knew and my neighbor knows (or suspects) that I knew what his expectations were, and I allowed these expectations to grow over years without ever discouraging them, I can predict that if I screw up the normal pattern my neighbor will make a misinformed choice and get cranky with me. And he can assume that I can predict it! So over time, going along with a fixed pattern of expectations creates an implicit signal, and allowing this signal to take shape is its own commitment.

Promising and Implying

A commitment can function in a social vacuum. Even if no one ever knows I have jumped out of a plane or driving north for three hours, those acts are still accurate guides to my motives, and I can use them in my own planning and strategy. But commitments are especially important in social interactions where each individual depends on the others. If a group of agents have little information about one another, little opportunity to communicate, or don’t trust each other, they must try to deduce each others’ future plans from past choices.

Often promises are accompanied by commitments which tie the promiser’s hands, thus reassuring the beneficiaries of the promise and giving the promise force. The pairing of a promise and a corresponding commitment may become so routine that the commitment is a promise all by itself.

For example, to wager is to promise to pay a certain sum if one loses a bet. A crooked gambler can wait until the outcome is known and then, having already lost, argue about what size the wager, or otherwise creating difficulties by refusing to pay. Thus it is a near-universal custom that when one wagers, one puts the money (or an equivalent token) “down on the table”, so that both bettors can have confidence in the honesty of the other. As a result, putting money (or tokens) down on a table is, in the appropriate situations, the same as wagering: it signals the intention to wager by committing to the stakes. If you move money towards the pot while playing poker, or put down tokens on a roulette table, do not expect anyone to believe you when you say you were just placing them there for a few minutes!

Trying the old “I only meant place my tokens on the felt for a minute, buddy” trick at a casino is a good way to get yourself into an unpleasant altercation with the bouncers. No one likes unpleasant altercations, which is another reason acts become signals of commitments, and commitments become tantamount to promises.

A commitment can function as a promise; a signal can function as a commitment; and a promise, ideally, functions as a reliable signal of the promisers’ intentions.


The readers of this blog likely understand the basic principles I am about to review, but it is important to summarize them nonetheless — both for clarity and for the benefit of the few exceptions. Voilà, the sociobiology of human kinship structures:

  • Women want perfect children (translation: sex with perfect men) and security for themselves and their children. In some societies security is mainly about protection from predators, human and non-human; in others, a child’s survival and/or social status depends on the resources his mother can get from other members of the community (for example, the father).
  • Men want lots of great children (translation: lots of sex, preferably with lots of women and/or great women) and fidelity. Fidelity combines three overlapping goals.
    1. The less the women a man sleeps with sleep with other men, the more of the children they conceive will be his (which serves the lots of children goal).
    2. And further, the less they will be distracted from caring for his children by other men’s children. If a woman neglects a man’s first-born for his second child, that is bad for the first-born but good for the man’s progeny as a whole. If she neglects his first-born for any other reason, it’s a pure loss (to him).
    3. But more importantly, if he knows which children are his, he knows which to protect and which to shower with resources.

One aspect of the sexual arms-race is how organisms signal high genetic fitness (the traits that they will pass on to their children) and ability to rear children to potential mates. These types of signals cause strategic interactions between rival organisms of the same gender. But potential mates also compete with each other over mutually exclusive romantic outcomes.

A male cannot protect a female and her young while he is off pursuing a different female elsewhere, or protecting her. A male who gives some resource to one female can no longer give that resource to another. Thus there is a conflict between the female’s quest for security and the male’s quest for more mates; a similar conflict exists between the male’s quest for fidelity and the female’s quest for the perfect mate. Each partner’s reproductive success would be best served by maneuvering the other into monogamy, while remaining free to pursue more/better mates him or herself. This conflict has many possible strategies and many possible equilibria, both at the species level and at the individual level.

Here concludes our review of Sociobiology 101. Commitment becomes relevant to romantic rivalry because organisms which can commit to fidelity/security can make themselves more desirable to a mate. This (like any desirable trait) allows the organism to attract a more desirable partner; in particular, it allows the organism to induce potential mates to make themselves more desirable by making a reciprocal commitment to security/fidelity during courtship.

Romantic Retaliation

A commitment is, as noted above, a sort of signal; and signals can be misinterpreted or counterfeited. In any context where you are suspicious of your counter-party, you have adequate reason to be suspicious of his commitment-signal, too. (Sure, when he wagered $100 he put down five $20 bills; but are they real twenties, or is he trying to dupe you with funny money?)

Reciprocal commitments to offer fidelity and security, however, call for even more care. In practice, such reciprocity nearly always amounts to mutual monogamy. Promiscuity, as miserable a condition as it may be, at least allows for portfolio diversification! If Shaniqua has five children by five different men, it is likely that each of the fathers will have little inclination to provide security for her, but it is fairly unlikely that none of the five will offer anything. If she had had five children by one of the men, perhaps that particular man would have been somewhat less likely to abandon her, but if he did abandon her, she would be left with no hope of aid from another baby-daddy. A man who (believes he) has five children faces the same trade-off: if he is supporting five children born to five different mothers some of them may have deceived him, but it is unlikely that all of them are lying. On the other hand if he has five children by a single unfaithful wife, then the paternity of all five of “his” children is in doubt.

So the more the mutual attraction between two mates is fueled by reciprocal romantic commitments, the more important the reliability of those commitments (and thus, the value of further commitment) becomes. When all of your eggs are in one basket, tread carefully! Certain types of antagonistic commitment can even acquire value as a result.

Throughout recorded history and into the dim realm of myth, men have been known to fly into jealous rages and batter wives suspected of infidelity, or even kill them — especially when the wife is caught in flagrante delicto. Humanitarians and feminists may wring their hands at such un-SWPL behavior, but we should focus on a more pressing question: how can such a trait possibly evolve? Even if the man suspects future children may not be his, or realizes existing children might not have been his to begin with, he can’t change that retroactively by killing her; given some chance that her babies are his, leaving them motherless would be a genetic disaster; and worst of all, what future romantic prospects will he have as a notorious wife-murderer?

Surprisingly good prospects, actually. As a good intra-Hajnal SWPL the observation is painful to make, but game theory is game theory. Wife-murderers can be attractive to girls who really do want to be faithful. If one of a girl’s most attractive traits is the commitment she plans to make to her future husband, she wants to commit in the most reliable, unambiguous way possible… and what better way to intensify the commitment than to knowingly offer it in a situation where the stakes are so high?

Of course, this is not necessarily the cause of jealous rage. Who am I to dismiss such compelling alternatives as “toxic masculinity” and “heteronormative patterns of discourse”? But there is no biological obstacle to an evolutionary explanation: not only can an irreversible commitment to drastic retaliation have strategic value, part of this value is precisely to make the potential-retaliator more attractive to the very mates who would be the target of any retaliation.

The same logic applies, mutatis mutandis, to the female propensity to retreat into a bleak, murderous depression after abandonment. Anyone who has ever seen a mother dote on her darling baby has difficulty understanding how any mother could be cold or indifferent — and yet there are mothers who starve, strangle, drown, or pulverize their children. We tend to think of these infanticidal mothers as unnatural monsters. Monsters, they certainly are. Unnatural… hmmm.

We’ve said everything we need to say about commitment between romantic partners (except maybe what concrete steps they take to commit to one another). Now we’re ready to nest the expectations, commitments, and strategies of the partners inside the expectations, commitments, and strategies of their extended families, and of the community as a whole.


7 thoughts on “Marriage I: Commitment

  1. Decent notes, but two comments:

    Why would there have to be a clear evolutionary path for a trait so discrete as willingness to engage in crime of passion against infidelity? It could be a case of pleiotropy or even simply an extrapolation of the fact that men being more aggressive in general will be more likely to violently retaliate against someone violating an emotional commitment.

    On the other hand, I do agree that female hybristophilia can be most concisely explained under the sociobiological framework.

    Liked by 1 person

    1. Thanks. You’re absolutely right; a more thorough explanation would observe that the ability to pre-commit to narrowly-irrational levels of retaliation can be produced in many ways, so the overall expression of the trait may be polygenic and the outliers may well be over-sensitive and abusive retaliators: like the tallest men, the outlier-trait could be the result of positive selection for many different variants, yet itself be subject to negative selection.

      And of course as you say, this need not be the explanation at all. I mainly wanted to explore how retaliation fits into commitment (which will become important if I manage to write Part 2 before the heat-death of the universe). But I do think the male reaction to infidelity is much more specific and pointed than male aggressive reaction to broken commitment in general – enough so that it was recognized as a separate phenomenon in the common law! (And as far as “violent retaliation for violating an emotional commitment”, are men more violent than women in their spats with friends, neighbors, and kin overall? I think not…)


  2. Great writing Quas- but (and not that I’m trying to pick a huge fight over it or anything) it reminded me of just what it is I don’t like about Darwinian explanation in the social sciences, namely that everything seems to disappear from the explanatory picture except immediate biological reproduction. In the case of the guy reacting violently to being cucked, the whole complex of social honour more generally comes to bear on it. To be sure, having lots of kids you know are yours is absolutely central and foundational to honour, but there’s more to it than that; the guy who’s been cucked has just had his personal authority gravely insulted and compromised, and accordingly by human default will suffer extremely serious public disgrace and loss of status (and possibly, subject to violent punishment himself) if he doesn’t violently retaliate against one or both of the offenders, even if he’s not all that particularly upset about it himself. More than the simple transmission of genetic material is at stake. (Mind you, I think a strong sociobiological case could be made for deriving the key features of honour, as the nexus in which obligations of loyalty and fidelity are given as moral values, from the exigencies of biological reproduction- but here it doesn’t seem to figure at all)

    Liked by 1 person

    1. Thanks, glad you liked it.

      As I mentioned in my reply to NTSS, I was not attempting to *prove* that sociobiology explains male rage, but rather to show how this apparently unattractive trait can, within a framework of commitment, actually be explained from the fitness advantage it gives (wrt mate selection).

      Showing how this is possible, in turn, illustrates ideas about commitment that I want to apply to problems of *collective* strategy next time. (Which will be soon, hopefully!) Will I discuss the way the group interacts with the individuals specifically in terms of status, pride, and shame? Complicated to say! I want to keep the model simple enough that it works at a high level of generality.

      I have fairly strong views on the importance of pride as a romantic motive — I’ve never had reason to blog about it, but you can see the tip of the iceberg in the first few paragraphs here:

      I think I’m going to leave that topic alone in the follow-up post (let sleeping dogs lie), and I’m also going to steer clear from talking about disgrace/humiliation as a sui generis punishment (as opposed to, like, being punched in the face, or shunned by neighbors).

      On the other hand, I will certainly be talking about the way expectations become centered on signals that embody commitments, which in turn acquire the same meaning as the acts the expectations reference – so that for example a man can react just as violently to his wife eating dinner with another man as he would to seeing them kissing, if the dinner is constructed to signal romantic commitment. That, to my way of thinking, is a building-block of the “thick” concept of honor/status you’re talking about.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s