Where Do We Come From?

The idea that a single population was the ancestor of all living humans is neat and convenient, but it is not consistent with the data.

The origin of modern humans is one of the most popular and hotly debated topics in the history of human evolution research. Researchers have produced a thick literature, both scholarly and public. I want to take issue with the two statements contained within this dominant paleoanthropological narrative: first, the suggestion that there is an identifiable point in time and place to call an origin; and second, the related implication that there exists a definable entity called “modern humans.” These two statements are taken as premises and remained largely unquestioned until recently. New research and a new generation of researchers are challenging these presuppositions at the heart of the discipline, and evidence is mounting to suggest that modern humans do not have an origin. Instead, we may be looking at fuzzy boundaries and messy origins. These terms are not as clean, but they are more likely to get us closer to true story of human evolution.

Modern humans do not have an origin

We think of modern humans as a species, Homo sapiens. In modern biology, species is the only level in taxonomy and classification to be biologically real; all other levels, genus and above, subspecies and below, are understood as abstract concepts, produced by and residing in the minds of scientists. Species is empirically defined, with a clear boundary of reproductive isolation. Members of the same species can reproduce fertile offspring; members of different species cannot. See donkeys, horses, and mules.

Since we are a species called Homo sapiens, we must have a point of origin, the moment of speciation. This is when our lineage started on its own evolutionary trajectory, separate from others—H.sapiens as a new species protected by reproductive barriers.

Or so we thought.

This understanding of species and species boundaries may apply to textbook examples, but not to all species, and certainly not to modern humans.

Paleo River. Charlotte Hollands ©2018

That modern humans were descendants of previous “archaic” populations living all over the Old World, Neanderthals being one of them, was a common paleoanthropological position until the 1990s. This decade saw a shift in the prevailing narrative, with geneticists lending support to the alternative view that modern humans arose as a new species as recently as 150,000 years ago in Africa. Time and time again, genetic research showed that Africans have the greatest genetic diversity, and the level of genetic diversity was directly linked with the depth of time. Africans, being the most diverse, were taken to be the oldest; and Africa, the place of origin of modern humans. Efforts were made to pinpoint the first modern humans somewhere on the continent. Strong candidates included Jebel Irhoud, Herto, and Omo, all sites in Africa (see Stringer 2016).

New research has been accumulating at lightning speed. Extracting usable ancient DNA from hominin fossils does not make headline news anymore, barely 20 years after the first successful research (Krings et al. 1997). Recent work on ancient DNA points toward a conclusion quite different from the first generation of ancient DNA research. A breakthrough publication in 2010 from Svante Pääbo and his research team (Green et al. 2010) started a wave of new research showing an admixture of Neanderthals and modern humans. The field of paleoanthropology was quick to incorporate this “new” discovery (perhaps setting aside the substantial body of literature arguing the exact same point based on fossil data) using the concept of introgression. The discipline conceded that there was admixture between archaic populations such as Neanderthals and moderns, at a negligible and insignificant level.

We are now writing a new chapter in human evolution research and all data point toward diversity.
In actual fact, genetics research confirms that the admixture between the archaic and the moderns was rather more significant, and the genes exchanged and introgressed were functional, not trivial. Neanderthals and Denisovans strengthened the immune system of the modern humans (Abi-Rached et al. 2011). Moderns interbred with Neanderthals, Denisovans, and yet another mystery population (McCoy et al. 2017); they all interbred with one another. Introgressed genes were important for surviving the tough and challenging environments that characterized the Pleistocene.

A spiral rather than a pendulum

A cynic might observe the recent shift in thinking about modern humans and declare the pendulum swings yet once more. What we are seeing now is something more akin to a spiralling than a pendulum’s stroke.

Modern humans arose at different times in different places, not as a separate species, but rather a continuation from the indigenous populations. Modern humans did not originate as a separate species and there was no single origin of a specific time and space; there were modern traits whose points of origin sprinkled throughout various times and spaces. There was no new species because of gene flow between populations across time and space. Asking where and when modern humans originated at a global scale is not only a question without an answer—it can erase voices that need to be heard.

This idea, not a new argument, opens new ways of thinking and gives importance to new topics. Hybridization research is one such example. Given the near-sanctity of species boundaries, hybrid species has been considered a marginal theory: by definition, species do not hybridize. New research demands that we consider the possibility that hybrid species are more often found in nature than we previously thought (Ackermann et al. 2016)—and modern humans might just be one such example. Humans throughout history moved about the world. They had sex and babies. They exchanged ideas and culture.

The admixture of archaic and modern traits is often observed in hominin fossils throughout the Pleistocene. Sometimes, they have been interpreted as “remnant” Homo erectus, implicitly driven to extinction by harsh climate or modern humans. New research opens up the possibility that rather than being remnants, these were members of modern humans, interbreeding with various populations. In fact, instead of clarifying matters, the distinction between “modern” and “non-modern” could limit the argument and hamper our paleoanthropological progress.

Fuzzy boundaries, messy origins

We have long known that speciation is a messy process. Genes in a species may have different histories than the species itself (Nichols 2001). Speciation is a gradual process of solidifying reproductive barriers. There will be exchange in genes between two populations that are on the trajectory of divergence and speciation, especially during the beginning stage of a speciation. This process is called incomplete lineage sorting.

What if this kind of gene exchange is not just a temporary process that only happens for populations undergoing speciation? What if gene exchange is itself a process that occurs throughout a history of a species? That is what the research seems to be pointing towards.

In the twentieth century, we were most interested in asking if Neanderthals were ancestors to modern humans. By the end of the twentieth century, the answer seemed obvious: modern humans arose as a separate species during late Middle Pleistocene in Africa and dispersed all over the Old World. As they moved to new regions they encountered other hominin populations. The result of such encounters was replacement without interbreeding, because the two belong to two different species. At the turn of the twenty-first century, we were asking if the introgression from Neanderthals to modern humans was significant. Starting with the Neanderthal genome published in 2010, genetics research from the last decade draws conclusions that are in stark contrast to the previous generation. Research shows genetic interaction between archaic human populations; Neanderthals and modern humans did interbreed. In the twentieth century, fossil and archaeology data were messy and complicated, while genetic data were clean and consistent. Now it seems that genetics is complicated, too.

A brave new world

We are now writing a new chapter in human evolution research and all data point toward diversity. The kind of big picture drawn with a broad brush was only possible with spotty data. Did Neanderthals go extinct? In some regions, surely; in other regions they did not. Instead, Neanderthals interbred with modern humans and left a genetic legacy. Asia was considered uninhabited after H. erectus left or went extinct. Denisovans and new discoveries show Asia was continuously populated with a new depth of antiquity (Zhu et al. 2018). The absence of data turned out not to be the data of absence.

New research reminds us to confront our own biases of racism and presentism. Much discussion about how to define modern humans implicitly includes modern Europeans and excludes Neanderthals. This definition, when applied to many extant humans, also excludes non-negligible proportions of indigenous populations. Furthermore, the diminutive hominins from Indonesia challenge us to expand what we accept as a normal range of variation for modern humans or to recognize another species of the genus Homo (see Aiello 2015).

Ancient humans lived in a different environment than us, with different trees and rivers, different landscapes, and different coastlines. There were no borders, only masses of land on which ancient humans moved. Some populations went extinct, taking their genetic signature with them. Some populations interacted with new incoming populations, exchanging genes and cultures.

The idea that a single population was the ancestor of all living humans is neat, but it is not compatible with the data. Gone are the days when a hypothesis could be tested with a few fossil specimens, a few genes, or a few stone tools. Gone are the days when a global model could be applied. In a recent article, Scerri and colleagues (2018) raise the point that modern humans did not originate from a single population in Africa, a point that has been made countless times before (for one early example, see Wolpoff et al. 1988).

Whether Neanderthals and modern humans are a same species or different may never be resolved. Neanderthal genes are found in modern humans because either they are same species (gene flow), or they are different species but genetically not yet separated completely (incomplete lineage sorting), or they are established different species that exchanges genes (introgression). Perhaps it is time to say goodbye to the idea of species as “the only entity with natural boundaries.” Perhaps it is time to be interested in new questions, questions that can be answered with new data. Perhaps, as Rosenberg and Wu (2018) suggest, it is time to move on from the imagery of trees with branches as a metaphor for human evolution and consider streams and rivers instead.

Sang-Hee Lee is a biological anthropologist specializing in human evolution. She was trained in Korea, the United States, and Japan, and is now a professor of anthropology at the University of California, Riverside. She published numerous articles and her book about human evolution in Korean (인류의 기원) is now translated into several languages (English, Chinese, and Spanish).

Charlotte Hollands is an illustrator, artist, and ethnographer who is fascinated with the power of hand-drawn images to reveal and describe complex truths. She is developing new ways to use illustration within social science research and is currently working on her first graphic non-fiction book, written by Alisse Waterston.

Cite as: Lee, Sang-Hee. 2018. “Where Do We Come From?” Anthropology News website, September 18, 2018. DOI: 10.1111/AN.972


Remember how European explorers defined different races based on a few stops along African and Asian coasts – when in fact there were clines and continuous distributions? But the labels they came up with lasted long after scientists realized that skin color, for example, did not neatly fall into categories of blacks, white, red and yellow. I suspect we have been doing the same thing with human ancestors, so “Neanderthals” were a single sample of a much larger distribution of variation. As the authors point out so well, we have a non-random distribution of samples heavily weighted towards Europe and Africa, and our view of human evolution is going to change dramatically as the blank spots in time and space get filled in. This leads me to wonder at the hubris of anthropologists who name a new species – long before we know enough about the context to tell if it is really just a population. Thanks to the authors for casting light on a set of linked issues and open questions. We don’t know so much more than we know! But none of this is going to stop NatGeo from trumpeting the next fossil find as a “lost human ancestor.”

The authors begin on a good note by noting that species is the only taxonomic level with a definite biological foundation, namely the restricted ability of one species to successfully interbreed with another species. They then muddy their argument by accepting the use of the term species to designate morphological differences between one population and a descendant population (allopatric species) without noting that the non-interbreeding criterion does not apply to allopatric species, except in the vacuous sense that they cannot interbreed since they do not coexist. This difference between sympatric and allopatric species means that selection for one population to not interbreed with another population, the basis for sympatric speciation, does not apply to allopatric species. Consider the following simplified example. Suppose a species has a quantitative trait controlled by a single locus with two alleles, a and A. Assume that of the three possible values for the quantitative trait due to the three genotypes aa, Aa and AA, the aa value is optimal for the species only in ecological zone 1 and the AA value is optimal only in ecological zone 2, but the Aa value is sub-optimal in both ecological zones (hence there would be selection for aa individuals to forage in zone 1 and for AA individuals to forage in zone 2). Suppose, initially, there is a single population with access to both ecological zones. Of the possible mating types, both the aa x aa and the AA x AA mating, only giving rise to aa and AA offspring, respectively, will be optimal in comparison to all other mating types due to the suboptimal Aa genotypes produced by these other matings types, regardless of ecological zone. Hence, there will be selection for any trait that reduces the likelihood of individuals mating with individuals having a different genotype, thus, there will be selection for the population to bifurcate into two species. Obviously, the same argument does not apply to allopatric species. Differences in fossils obtained from hominin populations ancestral to Homo sapiens have been used to characterize ancestral hominin populations. These are allopatric species, hence they did not arise through selection for speciation and instead reflect morphological change in an evolving population of hominins through time. This also means that these allopatrically defined hominin species may simply be an artifact of sparse sampling due to the difficulty in discovering hominin fossils. Consider a population with a morphological trait changing continuously through time. With sparse sampling of this evolving population, the more ancient specimens will appear morphology different in comparison to the less ancient specimens and this difference may then be treated as if it were an evolutionary event localized in space and time in which one species became the other species even though there was just a single population changing continuously through time. Thus, asking for the origin of the species, Homo sapiens, as if it arose through a distinct evolutionary event is not, as the authors argue, a meaningful question. Equally, the presumed sympatric species difference between Homo sapiens and Homo neanderthalensis, needs to be shown to arise through selection for a speciation event, especially since the differences may simply be due to geographic isolation. Not surprisingly, the DNA evidence shows that the populations involved could and did interbreed when hominin ancestral populations of Homo sapiens from Africa came in contact with hominin populations of Neanderthals in Europe and Asia. The authors rightly question assumptions that require presuming that species defined allopatrically or on the basis of geographic isolation also have biological reality as sympatric species.

This is a terrific essay that helped coalesce several lines of scholarship of which I had only a peripheral understanding. Looking forward to catching up with the contributing papers and the author’s own work.

Post a Comment

Want to comment? Please be aware that only comments from current AAA members will be approved. AN is supported by member dues, so discussions on anthropology-news.org are moderated to ensure that current members are commenting. As with all AN content, comments reflect the views of the person who submitted the comment only. The approval of a comment to go live does not signify endorsement by AN or the AAA.

Commenting Disclaimer

Want to comment? Please be aware that only comments from current AAA members will be approve. AN is supported by member dues, so discussions on anthropology-news.org are moderated to ensure that current members are commenting. As with all AN content, comments reflect the views of the person who submitted the comment only. The approval of a comment to go live does not signify endorsement by AN or the AAA.