Love is an emotion. To understand what love is for we have to place it in the broader context of the evolutionary function of emotions. One major function of emotions is to energize motivation. If we experience a strong positive or negative emotion, we become motivated to do something that's beneficial or to avoid something that's harmful. Pain exists to make sure organisms do everything they can to avoid things that can damage their bodies. If you are crazy enough to stick your finger in the flame on the kitchen stove, pain is there to protect your body and make it difficult for you to hurt yourself, no matter how crazy you are. Sexual desire and orgasm exist to make sure organisms are highly motivated to engage in sexual intercourse and produce children, regardless of their opinions on the subject. Sexual urges are so powerful that it is difficult even for priests and nuns who take a vow of celibacy to completely suppress their sexuality. Their emotions work against their conscious decisions, and the result is that some of them end up in the news for engaging in inappropriate sexual behavior. People make arbitrary decisions about all kinds of things, but matters of survival and reproduction are too important to be entirely dependent on people's conscious decisions. Emotions evolved to encourage us to do what's good for us regardless of what we think about it. Romantic love evolved, I argue, to motivate men and women to form pair-bonds.
But why the need for this extra emotional energizer? I think the answer has to do with our primate evolutionary past. In birds pair-bonding is an ancient adaptation. Birds have probably been pair-bonded organisms for millions of years. This means that natural selection has had plenty of time to sculpt birds' brains and provide the necessary wiring to support the psychological and behavioral adaptations for pair-bonding. In comparison to birds, human pair-bonding is an evolutionary novelty. It arose very recently - a few million years is equivalent to the day before yesterday on the evolutionary time scale - and very quickly in response to the rapid changes in brain size and in patterns of child development that made bi-parental care necessary or advantageous. To complicate things, humans probably evolved from a chimpanzee-like ape species whose members were sexually promiscuous, whose males did not contribute anything at all to child-rearing, and in which there were high levels of conflict between the sexes (such as male aggression toward females and sexual coercion). The brains of our ape ancestors had probably been shaped by sexual selection for millions and millions of years to support mating and reproductive strategies that did not involve pair-bonding. As evolutionary psychologist Paul Eastwick argued recently (in the article "Beyond the Pleistocene: Using Phylogeny and Constraint to Inform the Evolutionary Psychology of Human Mating", Psychological Bulletin, 135: 794-821, 2009), when the circumstances became favorable for the evolution of pair-bonding in the human lineage, natural selection had to quickly modify human brains in ways that would counteract other features that had been honed through eons of sexual selection. It wasn't an easy evolutionary step for the male brain of a sexually promiscuous, aggressive, and misogynistic chimpanzee-like ape to become the socially monogamous, female-loving, and paternal brain of a human being. The need for this rapid transformation presented a special evolutionary problem, which required a special solution. This special solution was romantic love and adult attachment. But how did natural selection find this special solution? How did it come up with romantic love?
I suggest that the evolutionary history of human romantic love may have progressed along the following lines. As human brains grew and infants became needier and more vulnerable for a longer period of development, such that the father's involvement and bi-parental care became necessary, natural selection had to come up with a way to motivate men and women to stay together for as long as it took to raise a child successfully. Now, natural selection never invents anything from scratch but rather modifies and rearranges preexisting structures. The psychological and emotional adaptations for the infant attachment system already existed in the brains of our ape ancestors and had worked pretty well to keep infants and mothers together. Natural selection tweaked this system, making it operational through adulthood, so that it could be used to bond mates to each other. Some of the neural circuits and the neurochemical substances that had been used to bond mothers and children, such as those involving oxytocin and endogenous opioids (which are also involved in regulating the body's responses to stress and physical pain), also became involved in mediating bonding between adults.
To accomplish the goal of fostering long-lasting emotional and social bonds between adult males and females, natural selection tinkered not only with the brains of our ape ancestors but with their bodies as well. The bodies of our ape ancestors were probably similar to the bodies of modern chimpanzees: well adapted for intense sexual competition and sexual conflict, but not for pair-bonding. For example, males were larger and stronger than females, had larger and sharper canine teeth, and had relatively small penises but huge testicles that produced large quantities of testosterone and sperm. Females, for their part, advertised their fertility period during their menstrual cycle through large sexual swellings to incite sexual competition among males. To foster pair-bonding and cooperative relationships between the sexes, natural selection reduced the differences in body size, strength, and weaponry between males and females. Then it eliminated obvious signs of ovulation in women and increased their receptivity throughout their menstrual cycle. This provided the opportunity for paired men and women to have sex all the time, thus reinforcing their union and increasing the man's confidence that when a child was born it was really his, which in turn increased his willingness to provide paternal care. At the same time, natural selection reduced paired men's desire for sexual variety and promiscuity by reducing their testis size and lowering their testosterone levels. Human males have relatively small testicles for their body size and produce relatively small amounts of sperm and testosterone compared to male chimpanzees. I once saw a slide of a researcher holding a chimpanzee brain in one hand and a testicle in the other; they were approximately the same size, and not because chimpanzee brains are small.
Another physiological adaptation for pair-bonding in human males is the dramatic reduction in their production of testosterone when they find themselves in committed relationships or are married with children. Lower testosterone in romantically committed men curbs their desire for other women and allows them to concentrate on their wives and children. This has been shown by many studies, including one that my colleagues and I conducted at the University of Chicago involving over five hundred MBA students Finally, various researchers have suggested that the exceptional length of the erect human penis - human males have by far the longest penis in relation to their body size of all the primates - is also an adaptation for pair-bonding. The long penis makes possible a wide variety of copulatory positions, including more intimate face-to-face, mutually ventral positions, which promotes social bonding during sexual intercourse. Ventro-ventral sexual intercourse is rare in primates but common in another species closely related to us, the pygmy chimpanzee, or bonobo; like humans, bonobos use sex for social bonding purposes. The long human penis may also increase the probability of female orgasm, which heightens the female's readiness for engaging in sexual activity, thereby strengthening the bond with her mate.
The multiple physical, physiological, and psychological adaptations that have arisen through natural selection to induce human males and females to form pair-bonds and cooperate in rearing offspring generally work very well. The most amazing psychological adaptation for pair-bonding - romantic love - creates in the human mind a longing for the desired partner and a psychological dependence not dissimilar from that existing between a young child and her mother. Successful bonds involve a profound psychological and physiological interdependence between partners such that the absence or loss of one partner can be literally life-threatening for the other. Conversely, solid and stable romantic relationships can have many positive effects on the health and longevity of both partners and their children. In short, whether you like it or not: love is good for you.