First off, though, I simply cannot resist the temptation to quote the above mentioned Countess of Champagne in full. The quote actually allows de Sousa to introduce the framework of his talk (the relationship and difference between reasons and causes of love), but it is worth reading in its entirety for its own sake.
Apparently, back in 1176, a “Court of Love” was established — presided by the Countess herself — to address the question of “Whether love can have a place between spouses,” i.e. whether the very concept of love is compatible with matrimony. Here is the verdict quoted by de Sousa:
We state and affirm that love cannot extend its power to a married couple, for lovers give one another everything freely, without obligation or any necessity; conjugal partners, by contrast, are committed to doing one another’s will and not to deny anything to one another.
So there you have it, apparently duty and inclination are not compatible in this instance, a very stern, Kantian view of things. And indeed, Kantian interpretations of love were a major target of ferocious criticism (even downright scorn) in de Sousa’s talk, which was delivered with humor and even theatricality (he quoted several passages from literature classics, such as the Cyrano, and he did so very engagingly).
Ronnie clearly distinguished reasons of love from reasons to love, as well as between positive and negative reasons, and finally between reasons seen from the point of view of the lover and from that of the loved. I do not want to get too far into this particular aspect of his talk, but some tidbits will give you an idea.
For instance, most people would not accept “I love her because she is rich” as a good reason for love, just like — and the parallel here does a good amount of work in de Sousa’s scheme — we would find weird, within the context of a work of art, if someone were to say “it’s beautiful because it’s expensive.” Or consider this: we would take it as a good reason for someone to stop loving someone else if the former object of love were to turn monstrously evil. But it is hard to imagine the traction one would get by saying something along the lines of “I love him because he is not monstrously evil.” (Though I have to admit that I’ve heard people on the New York dating scene lower their acceptance bar almost as far...)
In one of my favorite bits during the talk, de Sousa presented his version of the famous Euthyphro dilemma, which in its original Platonic context represents the most powerful argument for the irrelevance of gods to morality. The love-related version of the dilemma goes something like this: is what we love of value to us because we love it, or do we love it because it has value? Be careful how you answer it, because either horn of the dilemma carries pretty logical consequences. The first possibility is tautologically self-referential, and would lead us to admit, among other unsavory things, that it is perfectly sensible to love the above mentioned monstrously evil individual (or to love anyone, really). The second option, however, raises the question of fungibility: let’s say that I think I love someone because she is smart, beautiful, and of good character. Well, there is always the possibility of eventually meeting someone else who is even more so in any (or all!) of said dimensions. That being the case, wouldn’t it make sense to “trade up,” so to speak? Switch to the newer, better model? While a distressing number of people do in fact do so without a thought, it seems a bit callous (not to mention extremely un-Kantian!) to treat someone who we allegedly love as if s/he were a car or a television set, no?
Which brings us (skipping around the actual sequence of the talk a bit) to how de Sousa deals with the problem of fungibility. He introduces Helen Fisher’s famous studies on the cognitive science of three distinguishable kinds of love: lust, romantic love (which Ronnie calls “limerence”), and attachment. I referred to the same research in Answers for Aristotle, and I think the general findings are interesting if, of course, open to the usual caveats and potential future falsification of any neurobiological piece of research on humans.
Fisher has described three behavioral syndromes, as well as their correlated hormonal profiles, that characterize different types of “love” in humans. In some cases the three represent a temporal sequence within a given relationship, but this is by all means not a necessity. The first type, lust, is characterized by being fairly object-generic, meaning that we can lust after a (great) variety of people, though of course even lust is somewhat discriminating in its targets. Lust drives sexual intercourse, and at the hormonal level is underpinned by androgens (especially testosterone) and estrogens. It typically has a time span measured in hours or minutes...
Next we have limerence (a word de Sousa credits to psychologist Dorothy Tennov), what most of us call romantic love. Very much unlike lust, the object of romantic love is unique, indeed obsessively so. Its function is to focus one’s energy and time on a particularly good candidate for long-term attachment, and it is chemically underpinned by catecholamins, hormones such as norepinephrine and dopamine. This phase can continue for weeks or months, and up to 2-3 years, depending on the people and the circumstances.
Finally — if a couple gets that far — we have long-term attachment. This is not focused on sex, though it can in fact arouse intense emotional distress. It is what allows people to raise families, which means that it is evolutionarily crucial, in a species such as ours, where long-term parental investment is vital for the survival and well being of the offspring. The chemistry here is dominated by oxytocin and vasopressin, and the duration is indefinite, ranging from several years to a lifetime.
What has all of this to do with fungibility? de Sousa argues that the philosophically relevant differences among the three types of love are that only lust is really fungible, that only limerence seems to be exclusive, and that only attachment has a significant historical component (meaning the unique, shared history of the people involved). If he is right, what makes long-term love (i.e., the sort that leads to attachment) non-fungible is not the exclusivity of the love object (he notes, reasonably, that people are capable of feelings of attachment toward more than one person), but rather the uniquely historical, unrepeatable, series of events that characterize the relationship. From this perspective, the longer one’s history of love with another person, the less fungible that person becomes — even though there may be another individual around the corner who is smarter, more attractive, etc.
I like the general idea, though I’m not sure I buy every aspect of it. For instance, I think the object of romantic love is indeed fungible, only on a different temporal scale than the object of lust. True, romantic love is exclusive, but it may well turn out to be only temporarily so. Indeed, I think even long-term attachment is open to the threat of fungibility, as in the case of relationships that end after many years (and even a number of children), leading one or both former partners to start new long-term, and at least occasionally just as unique and valuable relationships with other people.
However, what I think does happen in the transition between lust, romance and attachment is that fungibility becomes less and less likely, varying on a sliding scale with a maximum value in the case of lust, an intermediate one for romance, and a very low magnitude (but not necessarily zero!) for attachment. Whether this should be cause for moral concern, of course, is another matter.
Ronnie concluded his talk — which is much richer than I managed to convey here — by noting the following (and I am here quoting mostly verbatim): (a) The contingency of the circumstances that give rise to relationships makes them deeply a-rational (so long, Kant!); (b) They derive from chance priority of acquaintance, pheromone compatibility, genetic fatality, transference, and habit (i.e., they have many causes); and (c) None of the above mentioned factors count as reasons. The distinction between causes and reasons in philosophy in general, and in the philosophy of human action in particular, has a long and complex history. Still, as a distinction it is both clear and, I think, useful. In de Sousa’s hands it pretty much kills any rationalist, Kantian approach to the philosophy of love. And that, my friends, is no small accomplishment.
Reprinted from Rationally Speaking, May 20th, 2013