Why are most people right-handed and so few left-handed? The short answer is that nobody knows. But left-handers are present in all cultures, and there are always significantly fewer of them than right-handers. In western cultures, 10 to 13% of the population is left-handed; in some archaic people (the Eipo in West Papua, New Guinea, for example), the frequency is as high as 27%.

Handedness is an ancient hominid characteristic. Artifacts from the middle (425,000–180,000 years ago) and early upper Pleistocene (180,000–10,000 years ago) indicate that our Homo neanderthalensis cousins were right or left-handed. Evidence in our own species of right-left-handedness comes from studies of stone tools, hole making rotational scratches in wood, and wear marks on tools. Most intriguing is that the frequency of right-left handedness of our prehistoric ancestors can be inferred from upper Paleolithic (35,000–10,000 years ago) “paintings” of hands on cave walls in Spain and France. These images were produced by blowing pigment through a tube held in one hand to define the outlines of the other hand on the cave wall. The proportion of left to right-handers in these cave paintings is the same as it is in French university students today.

Lots of things influence and seem to effect handedness. Social customs, for instance, have a strong impact on hand preference. Western culture (and most other cultures too) has traditionally viewed left-handedness negatively. Left-handedness has been equated with evil, misfortune, awkwardness, criminality and stupidity. The bias is thoroughly embedded in our languages. Consider the Latin words for left (sinister) and right (dexter) and the words that we derive from them. Or there’s the insult disguised as a compliment—“left-handed complement”. In Europe and North America during the first half of the 20th century, use of the left-hand was discouraged in children, who were taught to write with their right-hands even when they showed a clear preference for the left-hand. This practice changed in the last half of that century, and as estimated by hand preference for such things as writing and tennis playing, the frequency of left-handers in the population rose.

People use one hand over the other for different tasks for their own reasons too. Computer mice are designed for and intended to be used by the right-hand. Go to Best Buy; all the mice are placed to the right side of the model computers. But I use a computer mouse with my left-hand, leaving my right-hand (my dominant hand) free for other things.

And obviously human physiology plays a role in handedness. Its role is seen in the structure and function of the human brain. But our understanding of the human brain function is far from complete. Much of what we do know is gleaned from inferential studies not of healthy individuals but of individuals with brain injuries or brain diseases; for the most part, direct experimentation on the human brain is not practical because it is too invasive.

We know the brain is divided into two halves that are not exactly alike (“lateralization”). Each half of the brain (hemisphere) has specialized functions whose mechanisms are localized primarily in one-half of the brain. One of the brain’s specialized functions is speech and language. More than 150 years ago, the left hemisphere was identified as the primary seat of speech and language. At about the same time, neurologists suggested that a person’s handedness was opposite from the speech-specialized hemisphere—so right-handed people usually have left-hemispheric language specialization. But this is an oversimplification. We know now that almost all people (maybe as high as 95%) seem to have left-hemispheric brain specialization for language.

Where different specialization exists (right-hemispheric brain language specialization or little or no lateralized specialization, as examples), sex hormones such as estrogen and testosterone have been implicated in differentially affecting prenatal brain development. Stresses at the time of birth such as breech birth, premature birth, and oxygen deprivation may also condition left- handedness.

Genetic background is a strong determinant of hand preference—left-handedness certainly runs in families. Right-handed parents produce left-handed children only 10% of the time. But if the father is left-handed and the mother right-handed, 17% of their children will be left-handed; if the mother is left-handed, 22% of their children will be left-handed; if both are left-handed, the frequency of left-handed children rises to 24%. In all of these combinations, there will be slightly more left-handed boys than girls. This alone does not prove that hand preference is a heritable trait; parents after all provide an environment for their children as well as a genetic background. But fraternal (dizygotic) and identical (monozygotic) twin studies have been used to tease out the relative contribution of genetics and the environment, and these show that indeed there is significant heritability of hand preference.

But how handedness is passed on is unclear; the genetic models proposed to explain handedness do not adequately fit the inheritance data. Mendel, in his 1866 paper, demonstrated the transmission patterns of simply inherited traits in pea. His experiments could be repeated (with a little luck) by most gardeners today. But unlike the traits that Mendel characterized in pea, handedness is almost certainly not a simply inherited trait, and probably does not result from a single gene, as did Mendel’s pea traits. Handedness is a more complex trait that is likely determined by several genes—and several genes in conjunction with multiple environmental and/or biological interactions, some of which in some cases may be pathological.

As a heritable trait, handedness is subject to evolutionary selection. Any trait in a population of organisms that confers an advantage over another individual who does not possess the trait would be expected to increase in the population (that trait is said to increase “fitness”); the frequency of a trait that reduces fitness in an individual (or confers a disadvantage) would be expected to decline in that population.

Left-handedness is linked to characteristics that should reduce the left-handed person’s fitness. These include reduced life expectancy, shorter stature and lower body weight, and increased susceptibility to neurological disorders and autoimmune diseases such as allergies. Yet left-handedness persists in all human populations and at a frequency that apparently has remained unchanged for some 10,000 years. This suggests that in some way left-handedness confers some fitness advantages that offset these fitness costs.

A hypothesis that explains the stable persistence of left-handedness has been proposed by evolutionary geneticists Michel Raymond and colleagues. They speculate that left-handedness is maintained by “negative frequency-dependent selection” whereby the advantages of the trait are greater when the trait’s frequency is low, and that success in aggressive interactions is tilted toward the left-hander precisely because left-handedness is a rare trait and a right-handed opponent will “not know what hit him” and thus be at a disadvantage. This is the “fighting hypothesis”.

To test their hypothesis, Raymond and his colleagues reasoned that interactive sports (tennis, baseball, fencing, boxing) in modern societies require many of the same aptitudes and skills as do individual fighting and traditional combat. When they compared the proportion of left-handers in interactive sports with the proportion in noninteractive sports (gymnastics, swimming, bowling), they found the overall mean percentage of left-handers was 32% in interactive sports compared with only 11% for noninteractive sports. They also compared the frequency of left-handedness with homicide rates in such traditional societies as the Dioula in Burkina Faso (West Africa) and the Eipo in West Papua, New Guinea. In those cultures, left-handedness was strongly and positively correlated with homicide rates; the proportion of left-handers ranged from 3% in the most peaceful to 27% in the most violent and warlike groups. So these results are entirely consistent with a frequency-dependent selection mechanism maintaining left-handedness at a low but constant level. And since males are more likely to be involved in violent or aggressive activities, it also explains why there are slightly more left-handed males.

The validity of this hypothesis is a discussion for another day; it’s quite possible that it’s correct. It’s also quite possible that it’s incomplete or that other dynamics entirely determine left-handedness. But whatever, it brings us back to my original short answer. Despite the time and efforts that have been spent, we still don’t really know why most people are right-handed and so few left-handed.