My wife and I had three biological children starting 30 years ago, and then, five years ago, we became foster parents to a young boy whom we legally adopted when it became clear that he needed a permanent home. I call him “son” and he calls me “Dad”, although others sometimes mistake me for his grandfather.

I love my new son no less, although occasionally differently, than our biological children. If I’m honest, in one way I may love him more because we became a father and son, together, of our own free will. I like to playfully remind him, when he is particularly irritated with me, that most people can only dream of being able to choose their parents, whereas I heard him tell a judge in a courtroom that he had picked my wife and me for the job. Being unmoored from the biological tie is a relief in a way, too: I don’t seek a reflection of my (neurotic) self in my boy, and no one asks him to shoulder that burden, either, although he did recently complain that he was worried my nerdiness was “wearing off” on him.

Our relationship, like the parent-child experience of millions of other adoptive families in the United States, reflects the way the biological and social meanings of fatherhood can diverge widely. As we celebrate Father’s Day, we might consider how cultural constructions demonstrate the arbitrariness of how we define a father.

To start with, in every culture and historical era, there are many ways to become a father, socially speaking, even if there is only one fundamental way to do so biologically. Adult relationships vary in form across societies and include not only the type most common around the world today — heterosexual monogamy — but also same-sex marriage, nonmarital unions, polyamory, polygyny and polyandry.

This sheer variety of human mating practices helps to explain some of the more baroque notions of biological paternity found around the world — notions that are hard to reconcile with the reality that babies are produced by sex between one man and one woman. This fact, the “doctrine of single paternity”, is understood by almost all cultures, and has been for millennia, long before the biology was worked out by modern science.

But in some cultures in Amazonia — and a few scattered elsewhere — children are believed to have multiple biological fathers. Anthropologists have described more than a dozen groups with such beliefs about what’s known as “partible paternity”, and the distinctiveness and geographic separation of these cultures suggests that this belief system was independently invented multiple times.

With partible paternity, the woman’s role in producing babies is typically downgraded, unsurprisingly, to “receptacle”. In some of these cultures, the baby is thought to be made by an accretion of semen — like a snowball. In some, the men are even felt to pay a physical price for the pregnancy. The men report that they expend so much energy to make a baby through repeated intercourse that they can become wasted from the effort.

The Tapirapé people of Brazil did not believe anthropologist Charles Wagley when he told them that a single act of sex could result in a baby, and they insisted that intercourse had to continue throughout the pregnancy, and it could be with different men. All the men who had sex with a pregnant woman were considered the biological fathers of the child — although just the mother’s mate at the time was called “cheropu” (father). Only when a pregnant woman was known to have four or more sexual partners did the Tapirapé concede that the child might have “too many fathers”.

This type of belief system about multiple fathers can be advantageous to the survival of women and children, who can benefit from the support of more than one man.

But for most human societies, the opposite is more common: knowing the identity of the biological father (paternity certainty) is a fundamental part of social organisation (and gossip). And the biological and social processes that make it easier for males of our species to acquire such certainty — including, unfortunately, cultural and religious practices that constrain and shame women — have played a crucial role in our biological and cultural evolution.

Paternity certainty was probably an important facet of the evolutionary transition from our hominid ancestors.

The traditional evolutionary argument is that men will provide women with food and protection during pregnancy and the child’s early infancy only if they can be confident that they are advancing the survival of their own offspring and not some other man’s. But some scholars have wondered how provisioning pregnant and nursing partners could have played such a singular role in the origins of pair bonding if some members of our species — like the Tapirapé — had such freewheeling ideas about paternity. It would seem logical that men would have evolved to feel a sense of threat from male competitors for their partner’s fidelity. But it turns out that is not so essential. Human culture is extremely variable.

Indeed, at the opposite end of the fatherhood spectrum from the Tapirapé are the Na people, a group of mountain farmers near Tibet whose unique sexual practices have been the subject of Chinese texts for thousands of years. Anthropologist Cai Hua’s aptly titled book A Society Without Fathers or Husbands was written in 2001, but Marco Polo apparently also found the Na noteworthy for their unusual ideas about fatherhood.

Na households are matrilineal, with women living with their siblings, mothers and maternal uncles. Men from outside the family are rarely allowed to join the household, but they are allowed — at a female occupant’s behest — to sneak into her home for sexual assignations.

People in Na society do not care who their own biological fathers are, and it’s the matrilineal men — uncles and brothers — who fill the role of adult male supervision, caretaking and instruction of children. Despite endless probing, Hua couldn’t find anyone interested in talking to her about biological paternity. Although the Na do realise that offspring can physically resemble the man who was their biological father, most Na deem it irrelevant and uninteresting, since it involves no special rights or obligations.

The Na know that a man is required for a woman to make a baby, but they believe that babies are already in women, like seeds in the ground, and are merely “watered” by men. (They say, “If the rain does not fall from the sky, the grass will not grow on the ground.”) They feel that it makes no difference who does the watering.

The Na belief that no particular man is biologically necessary for the creation of a child, and the Tapirapé belief that many men are required starkly illustrate the decoupling of the biological and social aspects of fatherhood. Even in modern Western society, where children are rightly considered to have only one biological father, they sometimes have multiple social fathers — including loving male relatives, adoptive fathers, foster fathers, godfathers, figurative “uncles”, legal guardians, even mentors and older friends.

Still, despite the diversity of fatherhood, one thing remains painfully constant: not every child has a good father figure, but surely every child deserves one.

• Nicholas A. Christakis, a social scientist and physician at Yale, is the author of Blueprint: The Evolutionary Origins of a Good Society