Recently, Slate published an article adapted from the economist Marina Adshade’s contribution to the symposium Robot Sex: Social and Ethical Implications. The usual inference about sexbots is that they will provide substitutes for spouses and so will lead to less marriage. As the argument goes, people (especially low-status men) who are only weakly interested in marriage will substitute sex robots, just as they already substitute pornography and fast food. Adshade argues the contrary position, speculating that robot sex and marriage will be complements: If I can turn on a sexbot, I no longer need a spouse who turns me on.
As an application of Le Châtelier’s principle, this means that marriages chosen with the expectation of relying on sexbots for sexual gratification will allow us to optimize on other dimensions in seeking a spouse and so lead to ménage à bot marriages as good or better than the more familiar ones with the quaint expectation that the spouses have sex. For instance, Adshade suggests that we might see marriages between two straight women or between a woman and a gay man, whose complete lack of sexual interest in one another is irrelevant given that they can each have mind-blowing sex with Gigolo Joe, then turn him off, fold him up, slide him back under the bed, and get back to the real business of marriage: planning next month’s trip to Orlando, going to Home Depot to pick out bathroom fixtures, getting a lower price per ounce on Corn Flakes by buying the jumbo box, and otherwise pooling consumption.
As economic logic, this is impeccable. It is also absurd when explaining marriage. Human intimacy is not like shopping for a car, where you can get an otherwise nicer model if you decide you’re willing to install an aftermarket stereo. Saying that we could have better marriages if we outsourced the wifely duties to an insatiable nymph or priapic satyr with a 5800 mAh battery and dishwasher-safe genitals makes sense if we view marriage as a set of services provided by the spouses to each other because as long as each party somehow receives these services then everyone attains equal or greater consumer surplus. However, understanding marriage as something that could be improved if we delegated sex to voluptuous appliances is like the Yiddish idiom, “And if my grandmother had balls, she’d be my grandfather.”
The fundamental problem with applied micro-economics is that economic theories that work well for explaining markets are a clumsier fit for other spheres of social life. Human beings only very vaguely resemble market actors when they are interacting in ways that lack market institutions and are not subjectively understood as markets. In particular, human intimacy has its own rules that are understood to be separate from the market, to the extent that people jump through considerable hoops and justifications when intimacy intersects with the market. It is not enough, however, to say that sex bears an uncomfortable relationship with market logic. The more important point is that relationships are more than the sum of the services they provide and removing some of these “services” doesn’t relieve pressure on the relationship but weakens it. In particular, sex is important in the formation and maintenance of romantic relationships. We form romantic relationships through sexual attraction and cement them through the act of sex. For example, in her ethnography of bottle service at night clubs, Ashley Mears quotes a night-club promoter explaining how it is that he manages to corral an entourage of models whom he pays nothing, but for whose glamorous presence night clubs pay him handsomely: “How can you convince a whole models’ apartment to come out with you at night? I’ll tell you, you find the popular girl — the most exciting popular girl in the apartment — and you f*** her.” The promoter’s mercenary motives in seducing a model is less interesting than her devotion to him that is a result.
The frequency of sex seems to have a causal effect on marital satisfaction and marital stability, a finding that dates back to the 1930s but has been replicated many times since. There is a strong association between sex and marital satisfaction for both men and women, but women are especially likely to understand sex as tightly bound up with other aspects of relationship quality. Not surprisingly, couples that have sex more often are more likely to stay together. Of course, Adshade might argue that sexless marriage is bad because it is two people who are not having sex. Presumably, so long as they each achieve x orgasms from which they derive y utility, they need not achieve this with each other. We can’t yet know if she is right about sexbots sexually satisfying partners to a marriage, but there are methods that have existed long before Westworld through which people achieve sexual satisfaction aside from with their spouses, and we can see what impact these have on marital quality. Most obviously, one can cheat on one’s spouse with another human being. As we do not yet live in Adshade’s utopia where sexbots midwife the decoupling of marriage from sex, most Americans expect marital fidelity from their partners. Infidelity leads to deteriorating marriage quality and divorce. Nor is it just that people cheat in marriages that were failing anyway since adultery is surprisingly opportunistic.
One issue with adultery is that it directly involves another human being who can sap away the straying spouse’s affection as implied by the (defunct) tort for “alienation of affection.” Presumably, the virtue of sexbots is that they will let us achieve extramarital satisfaction with an object rather than a person, and this could obviate the possibility or anxiety of diverting not only sexuality from the marital bed but other forms of attention and affection, or even leaving the marriage for the adulterous partner altogether. In this respect, the best current analogy for sexbots is not adultery but pornography. Unfortunately, what we know about porn and marriage suggests the effects are similar, but less extreme, as those of adultery.
The best current analogy for sexbots is not adultery but pornography.
Bridges et al conducted a study of 100 women whose male partners viewed pornography. There is some good news here for the idea that sexbots will strengthen marriage in that these women mostly agreed with the statement “I am relieved my partner seeks sexual gratification through pornography rather than looking to me to satisfy all his sexual needs.” However, these women were also especially likely to agree with statements to the effect that pornography use was both caused by and contributed to a failure of intimacy for both them and their partner. This mix of feelings is not exactly consistent with the speculation that technological substitutes for sex will strengthen marriage. Rather it reflects a couple de-escalating their emotional commitment to each other. The question with the very lowest agreement from women in the study? “I believe my partner cares very much for me and for our relationship.” These feelings are based on a small (n=100) convenience sample, but Perry and Schleifer used a gold-standard survey dataset and found that the onset of porn usage doubles the odds of divorce over a two-year period. Men are more likely to begin viewing porn than woman but both genders are more likely to get divorced once they start doing so. Likewise, an informal survey of divorce lawyers finds over half of divorces in their experience involve porn and even more involve adultery. If women feel alienated from their relationships when their boyfriend or husband is having sex with a laptop and in fact divorce is more likely under these circumstances, it strains credulity to think she’ll like it any better when he wheels an android Astarte out of the closet every evening.
The idea that we can simply decouple sex from romantic love so as to enjoy both bourgeois domestic tranquility and bohemian sexual ecstasy badly misunderstands the nature of relationships in treating their components as separable. In Social Structures, John Levi Martin notes that we can’t describe a relationship as the exchange of services any more than we can describe pie as the exchange of crust for filling. Rather, pie simply is crust and filling and relationships are simply mutual attention and support. Martin was discussing patron–client relationships, but the same is true of romantic love and this is true even if people do enter into and stay in relationships in part for access to the “services” they provide. If I have access to an unlimited supply of flaky pastry, it doesn’t make me enjoy pie more by obviating the need for a good crust to let me focus on the filling, it just means the filling slides off my plate and I eventually give up on pie.
As suggested by the evidence we have for marriages when sex fades away or people seek sexual satisfaction aside from their spouses, we can’t reasonably expect happier marriages once the spouses take companionate marriage to the reductio ad absurdum of best friends and roommates who nonetheless have great sex, separately, with machines. The world in which we find ourselves “bedding Taylor Swift every night inside the Oculus Rift” is not one in which we spend the refractory period pair-bonding in sexless marriages of convenience oriented around pooled consumption.