At first glance this may seem an impossible question to answer; how can we know the mind of another species? Is not love a uniquely human characteristic that divides us from the beasts?

For any evolved trait to persist in a species it must be useful, in that it improves the chances of that individual passing on its genes. A trait, such as love, that is expensive to wield and possibly counter- productive for those experiencing it (due to, say: lack of appetite, self-harm and suicide; acts of heroism; the sacrifice of financial, material and social resources) must be very useful. Love comes in two flavours. One is the love of those genetically related – generally the closer the relation, the stronger the love. The other love is between those who are not related except by usefulness, such as sexual mates and friends. Typically, the more commitment and resources expended on a relationship, the stronger the love or friendship.

Emotions and logic

Where logical thought allows us to solve problems, emotions give us direction. They give us short and long term goals; they tell us what is important, and they convince others of our intentions. Daddy telling you he is very angry that you keyed his new BMW is nowhere near as convincing as daddy glowing bright red, sweating, shouting wide eyed and generally unable to control his temper. As we know he is not in control of his anger, we know he is sincere in his objection to our misdemeanour. The lack of emotions in emails is the reason we so often misunderstand the writer's intentions.

Genetic love

For the vast majority of species, the creature with the closest DNA match to another creature is its child. The creature with the best chance of replicating those genes is a healthy child. Any parent that commits large amounts of resources in growing and nurturing its infants will evolve a mechanism to greatly value that child - humans call it love. The greater the commitment of resources, the stronger will be that love and the greater the punishment that emotion must inflict if a parent makes a mistake.

Any species which puts large amounts of resources into raising individual infants must evolve love if it is to value those expended resources. A species which wastes its resources rapidly goes extinct. One would expect the bond between an elephant and its calf to be great and, conversely, the bond between a turtle and its young to be small. Evolution dictates that a parent, of any species, who invests heavily in an individual infant must love them.

Romantic love

Pregnancy for human females is a very big gamble. This is a period when they lose the ability to mate with others (especially with those who are genetically superior to the father to be), they lose mobility and hence lessen their ability to gain food in a time when their needs are increased, and increase their chance of being something else's meal. They cannot abort the growing infant (the exception being modern-day humans) if they subsequently decide the liaison was a mistake; each pregnancy reduces their fertility and hence reduces their desirability to other males, and ultimately endangers their life during child birth. After birth, they have an extended commitment to suckling the infant and providing for it until it can fend for itself. The reward for all this risk and commitment is that 50% of the child's genes will definitely be the mother's.

For males, they donate a sequence of DNA over a few minutes and that's it. They are free to mate again almost immediately afterwards. The number of children they sire has no negative effect on their ability to sire more, in fact it can make them more desirable to other females. If they feel they have made a mistake impregnating a particular female they can just walk away. The reward for no risk and no commitment is that 50% of the child's genes will be his, if he is the father, or 0%, if it transpires that the child is not his. The father can never be sure (except for 20th century DNA tests) that the child is his.

These differences in investment in producing a child lead initially to very different mating strategies for each of the sexes. As females are the limiting factor for reproduction (they can only carry one child) they should be highly selective with their partners.

For males, they should be far less selective as they can bet on quantity of partners rather than quality. The best strategy for a female to evolve is to choose a male that has good genes, is healthy, has access to resources to support her and the child during the pregnancy and the raising of the child and, most importantly, is going to stay around after mating to fulfil these latter requirements.

The best strategy for a male is to mate with as many females as possible, unless such a strategy causes him to be rejected by all females because it conflicts with their strategy. To gain access to any, and ideally the best, females the male must convince the female of his long term commitment. The male could just lie to the female about his commitment, copulate, and be off. Female must be able to detect sincerity and males must get better at producing it. The highest form of sincerity is shown when the one expressing it can not fake it even if they want to – they are out of control.

Romantic love is such a thing. By acting against the displayer's normal interests, the receiver of the love can have a high level of confidence in her suitor's intentions because he cannot fake them.

As the male is forced, through sexual selection, to put more investment into a single female for a single child, he also needs to be convinced by the female that she will be faithful to him (the worst case is that he will unwittingly support someone else's child, passing on someone else's DNA ) while all his resources are committed. He must be convinced by her commitment. Hence romantic love must be reciprocated.

Any species in which the above disparity in commitment is shown, and the male is sexually selected on his commitment, will need to experience romantic love.