Accurate depiction of uncertainty in ancient DNA research: The case of Neandertal ancestry in Africa

Abstract

All approaches to understanding the past must work with limited data. Like many other kinds of evidence of past peoples, the relation between ancient DNA and past events is intermediated by complex models that bear many assumptions, some untested or untestable. Statements about the past from this evidence are thus accompanied by uncertainty, some quantified and some unquantifiable. Accurate communication of this uncertainty is essential to effective cross-disciplinary collaboration and public understanding. Here I examine one well-studied case of ancient DNA inference: the inference of Neandertal ancestry for today’s African peoples. In this case study, scientific predictions about Neandertal introgression and the genetic variation of all living people both gave consistent predictions before the sequencing of Neandertal DNA. Still, at the time that a draft Neandertal genome was published, a myth became established among the public that today’s Africans are different from all other living humans in that they lack Neandertal ancestors. This contribution reviews public statements, press releases, and press accounts to understand the origin of this story and why it became widespread. I review the ultimate impact of this story and the path toward correcting it. In light of this example, I provide some guidelines on how to recognize accurate depiction of uncertainty and examples of how effective engagement with content experts in archaeology and biological anthropology can lead to stronger and more easily communicated scientific outcomes.

Keywords

Ancient DNA science communication human evolution race

After the publication of the first draft Neandertal genome (Green et al., 2010), a myth arose that today’s African peoples have no DNA derived from Neandertal ancestors. This notion is incorrect. According to the best current estimates, African and African diaspora peoples today derive between 0.4 and 0.8 percent of their genomes from Neandertal ancestors (Chen et al., 2020). This is around one-third of the approximately two percent Neandertal DNA carried by many people indigenous to other parts of the world, including the indigenous populations of Eurasia, Oceania, and the Americas. The amount of Neandertal DNA ancestry in African populations is basically what would be predicted from the extent of haplotype sharing between Africa, Eurasia, and Oceania, reflecting historic and prehistoric gene flow between these regions during the last 50,000 years (Chen et al., 2020; Pickrell et al., 2014).

Yet when these estimates were published in early 2020, media reporting on the research emphasized that finding Neandertal DNA in Africans was overturning previous scientific thinking. For example, “Africans carry surprising amount of Neanderthal DNA,” reported Science (Price, 2020). The first paragraph of this article paraphrases what had become an established myth: The first draft Neandertal genome led scientists to discover DNA sequences from Neanderthals in most of the world’s peoples. “Not so in Africans, the story goes, because modern humans and our extinct cousins interbred only outside of Africa.”

This story was encapsulated by a headline from a decade earlier, “Evidence for interbreeding with Neanderthals, only Africans pure” (Camphausen, 2010). This and many other accounts of Neandertal research over the last decade have promoted the myth that African peoples have no Neandertal ancestors. This idea is a substantial misconception of the initial Neandertal genome results. As the original research (Green et al., 2010) and commentary from scientists clearly described, the science showed that many living people have more Neandertal DNA than today’s African peoples, but it could not show that Africans have zero Neandertal DNA.

Why does this matter? Many other authors have drawn attention to the ways that aDNA findings have fueled racial and nationalist ideologies (Gibbs, 2020; Hakenbeck, 2019; Harmon, 2018; Panofsky and Donovan, 2019; Wolinsky, 2019). They have discussed the responsibility of scientists to correct harmful misconceptions and misinterpretations of their research. Over the past decade, many people around the world have begun to cite the Neandertal ancestry of European populations as a scientific rationale for their belief in the racial superiority of White peoples (Harmon, 2018). Others have suggested that the identification of Neandertal ancestry of Europeans has driven scientists’ desire to “humanize” Neandertals—in other words, to rehabilitate Neandertals as ancestors of White peoples (Saini, 2019). Both these lines of thinking are wrong, and they have detrimental impacts across societies.

To stop these false ideas, scientists must work to understand how they emerge and how they spread. In this contribution, I examine how this myth arose. I present the background understanding of Neandertals, ancient DNA, and population mixture at the time of the publication of the first Neandertal draft genome in 2010. I look at some early accounts of this research to understand how scientists and journalists conveyed the emerging findings. Then I consider why the inaccurate myth seems to have captured the public imagination (and the imagination of some scientists). Throughout, I focus on the subject of communicating uncertainty.

Ancient DNA and contamination

The study of genetic material from ancient bones, teeth, and soft tissues began during the 1980s and came of age during the early 2000s. From its beginnings, the field of ancient DNA (aDNA) was the beneficiary of extraordinary public interest. People are captivated by the promise of recovering biological information from extinct species, or even using DNA to resurrect them (Cherfas, 1991; Lessem, 1993; Wade, 2008; Shapiro, 2015). Applying aDNA methods to ancient humans and extinct hominins, researchers aimed to settle high-profile unsolved problems in archaeology, especially the fate of the Neandertals (Krings et al., 1997).

Reality disappointed some of the dreams of the growing field. During the 1980s and 1990s, many high-profile results were slowly revealed to be unrepeatable, from supposed “dinosaur DNA” (Hedges et al., 1995; Woodward et al., 1994) to Miocene-era magnolia DNA (Golenberg et al., 1990; Lindahl, 1993) to insect DNA from Cretaceous amber (Austin et al., 1997; Cano et al., 1993). Some aDNA laboratories devoted their efforts to the development of rigorous methodology for reducing contamination. Researchers also worked to understand the processes of biomolecular diagenesis as a way of recognizing needles of genuine aDNA within haystacks of modern bacterial, fungal, and human contaminants. By the early 2000s, the community of aDNA research achieved some consensus about the conditions in which aDNA preservation was likely, and recognized types of additional data that help substantiate the preservation of endogenous DNA sequences (Pääbo et al., 2004). This maturation of the field was accompanied by more extensive investment in aDNA research by national funding agencies. Parallel developments in high-throughput sequencing technology and whole-genome datasets of living organisms enabled researchers of the 2010s to examine ancient nuclear genomes, greatly broadening the questions that scientists can examine using aDNA data.

This is a story of technological and methodological progress. At the same time, this history exerted marked effects on the kinds of questions that aDNA specialists saw as legitimate to ask, and the kinds of uncertainty that they encountered. From the 1980s up to the early 2010s, the chief uncertainty in aDNA research was whether the bases in sequence data could be trusted. Leading aDNA researchers worked incrementally to reduce this analytical uncertainty. In the process, many chose questions that would minimize the possibility that contamination would go unnoticed in their results. For example, Pääbo and coworkers (2004) described the difficulty of carrying out research on early Upper Paleolithic (or “Crô-Magnon”) modern humans from Europe. Most researchers expected that these fossil humans would have mtDNA sequences similar to those of living Europeans. This similarity of expected results would have made it impossible, with methods available at the time, to differentiate genuine sequence from contaminating sequence from a recent person. I am personally familiar with cases from the time in which aDNA researchers obtained sequence from Upper Paleolithic modern humans that were unexpectedly similar to living people from a different part of the world. In these cases, the contamination was tracked down to other samples that were being studied in the same laboratory. The concern with contamination was one reason why Pääbo’s research group restricted their early focus upon Neandertals. The difference that they observed in mtDNA sequences between Neandertals and modern people provided a check on contamination in results.

In 2006, Pääbo’s team released preliminary results based on approximately 1 million base pairs of DNA extracted from the Vindija 33.16 Neandertal specimen (Green et al., 2006). This result increased by nearly two orders of magnitude the amount of aDNA information known for Neandertals at that time. Comparing the results with modern human sequences led to a fascinating conclusion: When comparing both the Neandertal sequence and sequences from living humans to a chimpanzee sequence, both shared more mutations with each other than expected if the Neandertal and modern sequences had come from isolated populations. The results appeared to indicate genetic exchanges between Neandertals and modern humans. This possibility seemed consistent with statistical results from samples of living people that suggested gene flow from ancient groups (Plagnol and Wall, 2006), although the Neandertal data could not easily be compared with these results.

However, reanalysis of the draft results soon revealed that a substantial fraction of the DNA from Vindija 33.16 derived from contamination by modern human DNA (Wall and Kim, 2007). The extent of this contamination was not clear; some suggested that the data might have been nearly 80 percent modern human sequence. Whatever the level, the initial finding of shared mutations between modern and Neandertal genomes seemed in doubt. The finding prompted intensive work on authenticating ancient DNA sequence and reducing the amount of contaminating DNA from modern sources (Briggs et al., 2010; Green et al., 2008, 2009).

African genomes in early Neandertal DNA work

The prevalence of contamination also prompted intensive attention to analytical methods for testing possible population mixture with aDNA results. Two of the major researchers in these efforts have published memoirs that describe their thinking at this time (Pääbo, 2014; Reich, 2018). What both emphasize is that each strongly doubted that any mixture from Neandertals into modern populations had happened. As they describe their thinking, the pattern that emerged from the data was a surprise, one that Pääbo elsewhere described: “At first I thought it was a statistical fluke” (Boyle, 2010).

Ancient DNA presented formidable problems for statistical analysis. In the years between 2005 and 2010, datasets on human genetic variation from many populations in the world had markedly grown. These data enabled more and more detailed studies of population mixture across Africa and other continents (Li et al., 2008; Tishkoff et al., 2009). But most of the data came from technologies that were unsuitable for comparison with the fragmented bits of sequence from aDNA. For example, Tishkoff and coworkers (2009) had assembled a panel of genetic data from more than 2400 African and African diaspora individuals. But the data were from two categories of length polymorphisms, insertion-deletion and short tandem repeat loci, that were accessible to researchers working with DNA from living subjects but nearly impossible to examine with fragmented DNA from ancient bones. To compare living people with the Neandertal genome, researchers ideally needed to examine whole-genome sequences of living people, so that they could maximize the fragmented information across ancient sequences. In 2010, the sample of modern whole genome sequences was very small. Pääbo’s research team added five to this number, which they chose to represent African (two individuals), European (one individual), Asian (one individual), and Melanesian (one individual) populations.

To test for possible population mixture from Neandertals into the ancestry of these genomes from living people, the team developed a statistical approach based on a four-way comparison of sequences. Using a chimpanzee genome sequence as an outgroup, this approach simply compared the Neandertal genome to two living people’s genomes (hence, the four-way comparison), and counted the number of nucleotide positions that are the same between one human and the Neandertal, and different from the other human and chimpanzee. In a population history with no mixture of any human populations and Neandertals, neither living human should share more with the Neandertal than the other living human. But if one of the two living humans has derived more of their genome from the Neandertals, that human’s genome will share slightly more nucleotide variations with the Neandertal genome. This method of comparison became immediately widespread for the analysis of ancient population mixture, because it is very tolerant of variations and chemical damage in the ancient DNA data, although it may be affected strongly by contaminating modern sequence (Durand et al., 2011). The result of the initial Neandertal comparison was that the Neandertal sequence shared slightly more sequence variations with the European, Asian, and Melanesian genomes than with either of the two African genomes (Green et al., 2010). This result demonstrated that Neandertals had contributed to the ancestry of these three people, more than any contribution Neandertals may have had to the ancestry of the two African individuals.

This was the first result from aDNA data to demonstrate introgression from Neandertals into more recent human populations. The result was limited in scope. By itself, the aDNA showed that some living people must carry DNA ancestry from Neandertals, and that there was a difference between African and non-African samples in the amount of Neandertal DNA ancestry. As the research paper also noted (Green et al., 2010), the results could not rule out the hypothesis that some of the difference in similarity with the Neandertal genome might be explained by ancient population structure within Africa.

Understanding introgression before 2010

The methods chosen by Green and coworkers (2010) showed that Neandertal mixture was not a “statistical fluke.” In that sense, the contribution of aDNA evidence to the long-standing scientific debate about Neandertals was very important. But this contribution was one of the last of many steps toward understanding the Neandertal role in the ancestry of today’s people. Many scientists had considered Neanderthal introgression into populations of modern humans long before the sequencing of any ancient genomes.

The thinking of earlier scientists naturally emphasized an expectation that people today who live in the same regions where Neandertals once lived would have greater resemblance to Neandertals than other living people in other parts of the world. In particular, anthropologists focused upon Europe, the region where they had identified the most Neandertal fossils. Some morphological features of Upper Paleolithic European modern humans appeared to suggest a legacy of Neandertal ancestry (Frayer, 1993; Frayer et al., 1993; Wolpoff et al., 2001). Some scientists considered Neandertal gene flow into Africa (Smith et al., 1995). Still, many researchers disputed whether such morphological evidence was explained by Neandertal ancestry instead of competing hypotheses such as parallelism or retained features from common ancestors (Bräuer et al., 2004; Stringer and Bräuer, 1994).

By the year 2000, scientists had begun to identify regions of the genomes of living people with gene genealogies that included deep clades rooted in Eurasia rather than Africa (Harding et al., 1997; Harris and Hey, 1999b; Labuda et al., 2000; Parham et al., 1994). This genealogical pattern is unlikely to arise in a population history lacking genetic ancestry from Neandertals or other equally divergent groups. These observations prompted a search for additional regions with this pattern (Evans et al., 2005; Garrigan et al., 2005a, 2005b). These findings from single-locus studies were reinforced by examination of whole-genome data, where ancient structure was a strong conclusion (Plagnol and Wall, 2006). These results also prompted widespread theoretical investigation of the conditions and expectations of genetic introgression from Neandertals (Garrigan and Hammer, 2006; Harris and Hey, 1999a; Hawks and Cochran, 2006; Hawks et al., 2008; Templeton, 2002, 2005; Wall and Hammer, 2006). This research led to much greater understanding of introgression as a phenomenon and what human populations might have derived from Neandertals.

It is important to be clear that this work did not result in a scientific consensus in 2009 about whether Neandertal introgression into modern human populations ever happened. Some morphological data supported the idea, but a number of researchers asserted that the same data could be explained by alternative hypotheses such as parallelism. Some genetic data also supported the idea, but it was difficult to rule out the alternative hypothesis that ancient population structure within Africa might have left disequilibrium gene genealogies (Hodgson and Disotell, 2008). Notably, this alternative hypothesis even persisted for some time after the draft Neandertal genome showed strong evidence of greater similarity of Neandertal and non-African modern peoples (Hodgson et al., 2010). It would eventually take much more hard data from ancient bones to change many geneticists’ way of thinking.

Nonetheless, the literature on introgression and modern human DNA variation made clear that if Neandertals contributed any genes to modern people, then those genes would be unlikely to occur exclusively outside Africa. That is to say, the occurrence of Neandertal introgression was within the realm of scientific uncertainty, but ideas about what introgression would look like between Africa and Eurasia were fairly certain. The empirical data prior to aDNA genomes appeared to confirm the expectation that Africans would share some introgressed haplotypes. Nearly every candidate haplotype for introgression was present within some African samples, usually at lower frequency than in Asian or European samples (Hawks et al., 2008). It was this very observation that made credible the alternative hypothesis of ancient African population structure for these deep genealogies (Hodgson and Disotell, 2008). Some (e.g., Hawks and Cochran, 2006) predicted that evidence for introgression from Neandertals would be more salient for some loci in Africa than outside Africa, if selection had increased their frequency in Eurasia to a sufficient frequency.

These ideas about possible introgression from Neandertals fit into a broader picture of human genetic variation. The living peoples of the world share extensive genealogical and genetic connections. They do not have separate and isolated histories during the last 50,000 years. During the 19th and early 20th centuries, there were some European and American scientists who believed that today’s human populations had distinct and isolated origins. This discredited idea is known as polygenism. One polygenist idea held by some early and mid-20th-century scientists was that Neandertals were racial ancestors of more recent Europeans, and that they had no role in the ancestry of other groups, such as recent African peoples (Coon, 1962). This idea was rejected by the majority of anthropologists (Wolpoff and Caspari, 1997). Anthropologists active in human origins research between 1970 and 2000 had diverse views about the role that Neandertals might have played in the ancestry of today’s people. Those who thought that Neandertals were among the ancestors of living people also emphasized the high rate of gene flow among recent populations of the world. No one saw Neandertals as exclusive ancestors of recent Europeans or other groups.

All living humans are connected by genealogy, and the most recent genealogical ancestor of everyone today lived within the last few thousand years (Chang, 1999; Rohde et al., 2004). The human network of genealogical relatedness has become more interconnected within the last 500 years, during which great migrations and diasporas have brought together many groups that were more separated in the past. This recent high rate of migration has only accelerated a much longer process of gene flow.

Well before the recovery of Neandertal whole-genome evidence, the importance of migration from Eurasia into Africa during the last 35,000 years, and subsequent mixture and dispersal through the populations across Africa, had been documented from mtDNA, Y chromosome, and whole-genome evidence (Cruciani et al., 2002; Henn et al., 2008; Olivieri et al., 2006; Tishkoff et al., 2009). The Neandertal ancient DNA made clear that some of the earliest modern populations to enter Eurasia mixed with Neandertals. Their descendants who migrated back into Africa carried some of this Neandertal genetic heritage with them. This knowledge was well-known within the community of aDNA researchers and influenced subsequent work into identifying recent genetic exchanges from Eurasia into Africa and vice-versa (Pickrell et al., 2014; Wang et al., 2013).

The origin of a myth

By 2010, the anthropology and human genetics communities had generated much research about introgression and possible Neandertal ancestry of living people. The first Neandertal draft genome in 2010 was a highly informative data point, with some important limits. The publication of the Neandertal draft genome was accompanied by an enormous amount of press coverage, most of which was responsibly written.

I was one of those writers (Hawks, 2010). I point to this contemporary blog post to emphasize that the background that I am describing in the current contribution is the same as that described in 2010. In a section titled “Do living Africans have Neandertal ancestry, too?” I discussed the lack of power of the study to provide an estimate of Neandertal ancestry in Africans, the record of historic and prehistoric migration from Eurasia into Africa, and my guess that Neandertal ancestry in Africa probably is greater today than in Late Stone Age or earlier times. We have learned many additional details about the population history in the last 10 years, but the basics have not changed.

A TED talk by Svante Pääbo in early 2011 likewise gave an accurate description of the science behind the Neandertal DNA results. In this talk (Pääbo, 2011), Pääbo emphasized the lack of genetic variations that distinguish all Africans from all other peoples. He described the methods that demonstrated a greater contribution of Neandertal genetic heritage to populations in other parts of the world where compared to Africans. Pääbo ended this talk by emphasizing that the Neandertal mixture is not an absolute difference between Africans and other peoples of the world. He predicted that archaic humans in Africa would also be found to have mixed with the ancestors of modern people.

So how did it come to be that one of the most enduring take-aways for many people was, “Africans have no Neandertal ancestors”? An examination of the press releases that accompanied the publication of the draft Neandertal genome provides some insight into this question. The two releases, from the Max Planck Society and the National Institutes of Health, both emphasized similar language.
‘Those of us who live outside Africa carry a little Neandertal DNA in us’, says Svante Pääbo. (Press release, Max Planck Society, 2010)

DNA Signatures Found in Present-Day Europeans and Asians, But Not In Africans. (Subhead of press release, National Institutes of Health, 2010)
The idea shared in both press releases, that Neandertal DNA was found in some genomes and not others, captured one important element of the results. But as expressed, this idea was misleading. The methods used in the Neandertal genome research in 2010 could demonstrate that introgression from Neandertals was present in some genomes today, but those methods could not demonstrate that introgression was absent from other genomes. It is a subtle difference, but a real one.

The press releases were not the only source of this misconception; many of the researchers repeated similar comments to journalists for publication.
‘We do not find any evidence of Neanderthal gene flow into Africans’, said population geneticist David Reich at Harvard University Medical School, who helped analyze the Neanderthal genome. ‘What we find is shared equally by Europeans, East Asians and Papua, New Guineans’. (Hotz, 2010)
This is an accurate statement. The study did not find evidence of Neandertal gene flow into Africans. But it is also misleading. The study was not designed to find Neandertal gene flow into Africans. In fact, if the study had concluded that Africans were equally similar to the Neandertal genome as non-Africans, the conclusion would have been that no gene flow from Neandertals had happened at all.

Any scientist who interacts much with the media will sometimes see their own words used in misleading contexts. Occasionally, journalists even use such out-of-context quotes to convey the polar opposite of a scientist’s intended meaning. In my experience, such misleading quotation usually happens in spite of both scientist and journalist having good intentions. I have often spent hour-long telephone conversations with journalists explaining new research in many different ways, only to have them use a single quote on what I may consider to be a small or peripheral aspect of the study. At the same time, journalists often want to convey a full picture under pressure of a deadline, needing experts to illustrate both the promise and uncertainty of new research. All this is to say that a small number of published comments from researchers may not give a good impression of their efforts to communicate any caveats of their own research. Most journalists rely on experts outside the author list of a paper to convey a skeptical or critical point of view.

Press releases issued by institutions and journals upon the release of a scientific paper are a different story. In these cases, scientists themselves determine the quotes that appear, in coordination with a press information officer. The scientists have the opportunity to provide feedback or edit the surrounding context to convey their message with clarity. This is important because it provides a way of anticipating and correcting misconceptions. Ideally, scientists will check their quotes with other colleagues to make sure they do not induce unanticipated confusion.

In this case, the National Institutes of Health press release (2010) included two important statements reflecting the uncertainty of the results:
The team did not find traces of Neanderthal DNA in the two present-day humans from Africa. It is not known, however, whether a more systematic sampling of African populations will reveal the presence of Neanderthal DNA in some indigenous Africans.

‘These are preliminary data based on a very limited number of samples, so it is not clear how widely applicable these findings are to all populations’, said Vence L. Bonham, Jr., J.D., senior advisor to the NHGRI Director on Societal Implications of Genomics.
The first statement here gives a misleading impression of the research by using the metaphor “finding traces of Neanderthal DNA,” when instead the study tested for differences in Neandertal similarity. But it correctly emphasizes that the sampling of African populations was too sparse to test for the absence of Neandertal ancestry. Not surprisingly, most journalists omitted these kinds of cautions in their stories.

Media interpretations of the ancient DNA results

From the perspective of today, I was interested to review the diversity of ways that journalists reported the initial Neandertal DNA results in 2010. Many of them relied upon a race-centric way of describing the groups and their interactions.
So, our ancestors made babies with Neanderthals. But that’s not the whole story: Only some modern populations have Neanderthal parentage. Africans don’t seem to have any distinctively Neanderthal DNA. (Becker, 2010)

By comparing the DNA of Africans (whose ancestors could not have crossbred with Neanderthals, since they did not overlap with them) and various Eurasians (whose ancestors could have crossbred with Neanderthals), Dr Paabo has shown that Eurasians are between 1% and 4% Neanderthal. (The Economist, 2010)

In [Dr. Pääbo’s] and Dr. Reich's view, Neanderthals interbred only with non-Africans, the people who left Africa, which would mean that non-Africans drew from a second gene pool not available to Africans. (Wade, 2010)
While this kind of race-centered account was the most common, some contemporary journalists described the result accurately in a way that reflected the method’s limitations. For example, Lauren Gravitz of MIT Technology Review:
But the new research shows that modern humans in Africa have a lower percentage of the Neanderthal genome than non-Africans do–implying that the founder group that left Africa interbred with Neanderthals before moving on to populate the other continents. (Gravitz, 2010)
Others adopted a more skeptical line. The New York Times account by journalist Nicholas Wade (2010) was notable in emphasizing the possibility that contamination or under-representative sampling of modern human variation might have created the appearance of gene flow where none actually existed.

In subsequent years, the insight of Neandertal mixture with recent humans continued to command great scientific and public interest. Some of these continued the same misconception that African peoples have no Neandertal DNA. For example, in a 2011 press release on an X-chromosome haplotype found to have derived from Neandertals (Yotova et al., 2011):
Non-Africans are part Neanderthal, genetic research shows (Headline of press release, University of Montreal, 2011)
The haplotype identified in this study was actually present in sub-Saharan Africa at a frequency of 0.4 percent, by itself demonstrating the presence of Neandertal DNA in today’s African peoples and contradicting the press release headline. This aspect of the study didn’t capture any press attention.

A much later contribution helps highlight the legacy of this myth, in an article written for young readers:
Because humans and Neanderthals only had kids outside of Africa, you will not find Neanderthal DNA in present-day African people. However, in the rest of the world, people have about 3% Neanderthal DNA. (Ottenburghs, 2019)

Best practices for communicating ancient DNA research

Effective public communication about ancient DNA can reduce common misconceptions about race and human ancestry. But ineffective communication can reinforce those misconceptions, as exemplified by the widespread belief that African peoples today could have no Neandertal ancestors. Social media presents a particular problem because people are more likely to share stories that align with their preconceived ideas about race and ancestry.

In my previous work on science communication and ancient DNA, I have discussed several guidelines to prepare research groups for public release of their research (Hawks, 2019). Here, I highlight some key elements.
Be candid about uncertainty. The science of ancient DNA is still developing, and today’s results will be modified in light of future evidence. All public communication should be clear about the limitations of today’s datasets and how results may change with new data or methods.

Tell people about researchers’ ethical obligations. Ancient DNA research draws upon fields with high standards for ethical behavior and professionalism. One way of recognizing sensationalized media coverage is that it emphasizes conclusions and outcomes without discussing the basic legal and ethical standards that guide research.

Involve participants in public communication. Researchers are wise to give voice to people who belong to descendant communities, who have participated in research by providing samples, or who have carried out some of the sampling. Including these perspectives increases the effectiveness of public communication, while providing avenues for a broader array of voices to shape research questions.

Be clear about ethnic, linguistic, archaeological, and genetic observations. These categories of evidence are not equivalent, and findings that explain one of these may not determine the others. Confusing these with each other tends to fuel public misconceptions of race, ethnicity, and genetics.

Have key team members carefully vet all press releases. Research teams are transdisciplinary and trans-institutional. In particular, geneticists who lead ancient DNA studies likely come from different institutions than the archaeologists or anthropologists who may have shaped research questions and carried out fieldwork. This makes it extremely challenging for a public information officer for any university to compose a release that covers the research accurately. Researchers have a responsibility to make sure that their public releases are accurate and do not promote misconceptions.

When researchers conceive of public communication of their work, they often think in terms of a broadcast model. But “the public” is not a monolith. People who read about ancient DNA research with great interest are stakeholders in the research, including government decision-makers, participant communities, students, and voters. When these stakeholders read or see sensationalized versions of the research, their future engagement with the field will be affected. Thinking about public engagement as communication with current and future partners is a helpful way to help frame efforts within research groups or collaborations.

Conclusion

Ancient DNA once stood at a crossroads where bad results and unwarranted hype imperiled the credibility and sustainability of valid research. The field weathered this storm, guided by researchers who emphasized practices centered around understanding and quantifying contamination, so that laboratories could implement practices that minimized contaminants and quantified their presence even within valid studies. In the process, scientists learned much about the basic biochemistry of DNA diagenesis, the way that biomolecules are affected by environmental conditions, and the living microbial world within ancient fossils.

Neither the press nor the public changed their attitudes about ancient DNA instantaneously. New sequels to Jurassic Park are still being made. Journalists continued reporting on basic doubts about ancient DNA results well into the 2000s. Today, it is reported as a “surprise” when research shows that the pattern of Neandertal DNA in living people fits decade-old expectations.

In other words, today’s problems communicating aDNA science are more about mythology than technology. In human genetics and anthropology, much of that mythology has to do with race. Admittedly, Neandertals and human origins are not at the top of many people’s concerns. But race is, and any apparent difference in the history or biology of different groups of people can become part of the public dialogue on race. During the 2010s, the idea that Africans have no Neandertal DNA was one of many myths of racial purity that gained currency among the public.

Today it is a major misconception that affects people’s acceptance of evolution and of human differences.

As an anthropologist and public advocate for science, I care about how people think about Neandertals. They are an ancient people that can no longer speak for themselves. I have found it a privilege to stand in the places they walked, and to try to recover some small fragments of their voices. The genealogical network that connects us together connects these ancient people as well.

Ancient DNA research has been a fast-paced discipline for the last decade, yielding some insights that could not have been produced in any other way. But as these methods has matured, it has become more and more necessary to slow down and tell stories that convey the full degree of uncertainty in these studies. Some researchers have expressed concern that conveying the extent of their uncertainty may harm the public trust in their results. Research in communication science suggests that this concern is misplaced (Retzbach and Maier, 2015). For those of us who work to bring light to human evolution, we must understand that our research matters to policy and public trust in science. It is important to think of the way our results and interpretations will impact all people.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biography

John Hawks is the Vilas-Borghesi Distinguished Achievement Professor of Anthropology at the University of Wisconsin-Madison and Visiting Professor at the University of the Witwatersrand. His research emphasizes the connections between the fossil and molecular records of human origins.

References

Austin

Ross

Smith

, et al. (1997) Problems of reproducibility – does geologically ancient DNA survive in amber-preserved insects? Proceedings Biological Sciences 264(1381): 467–474.

Becker

(2010) Did Neanderthals mate with humans? PBS Inside NOVA. Available at: www.pbs.org/wgbh/nova/insidenova/2010/05/did-neanderthals-mate-with-modern-humans.html (accessed 10 September 2020).

Boyle

(2010) Neanderthal DNA lives on … in some of us. NBC News, 6 May. Available at: www.nbcnews.com/sciencemain/neanderthal-dna-lives-some-us-6C10404047 (accessed 10 September 2020).

Bräuer

Collard

Stringer

(2004) On the reliability of recent tests of the out of Africa hypothesis for modern human origins. The Anatomical Record. Part A, Discoveries in Molecular, Cellular, and Evolutionary Biology 279(2): 701–707.

Briggs

Stenzel

Meyer

, et al. (2010) Removal of deaminated cytosines and detection of in vivo methylation in ancient DNA. Nucleic Acids Research 38(6): e87.

Camphausen

(2010) Evidence for interbreeding with Neanderthals, only Africans pure. Digital Journal, 10 May. Available at: www.digitaljournal.com/article/291798 (accessed 10 September 2020).

Cano

Poinar

Pieniazek

, et al. (1993) Amplification and sequencing of DNA from a 120–135 million-year-old weevil. Nature 363(6429): 536–538.

Chang

(1999) Recent common ancestors of all present-day individuals. Advances in Applied Probability 31(4): 1002–1026.

Chen

Wolf

, et al. (2020) Identifying and interpreting apparent Neanderthal ancestry in African individuals. Cell 180(4): 677–687.

10.

Cherfas

(1991) Ancient DNA: Still busy after death. Science 253(5026): 1354–1356.

11.

Coon

(1962) The Origin of Races. New York: Knopf.

12.

Cruciani

Santolamazza

Shen

, et al. (2002) A back migration from Asia to Sub-Saharan Africa is supported by high-resolution analysis of human Y-chromosome haplotypes. American Journal of Human Genetics 70(5): 1197–1214.

13.

Durand

Patterson

Reich

, et al. (2011) Testing for ancient admixture between closely related populations. Molecular Biology and Evolution 28(8): 2239–2252.

14.

Evans

Gilbert

Mekel-Bobrov

, et al. (2005) Microcephalin, a gene regulating brain size, continues to evolve adaptively in humans. Science 309(5741): 1717–1720.

15.

Frayer

(1993) Evolution at the European edge: Neanderthal and Upper Paleolithic relationships. Préhistoire Européenne 2: 9–69.

16.

Frayer

Wolpoff

Thorne

, et al. (1993) Theories of modern human origins: The paleontological test. American Anthropologist 95(1): 14–50.

17.

Garrigan

Hammer

(2006) Reconstructing human origins in the genomic era. Nature Reviews Genetics 7(9): 669–680.

18.

Garrigan

Mobasher

Kingan

, et al. (2005a) Deep haplotype divergence and long-range linkage disequilibrium at Xp21.1 provide evidence that humans descend from a structured ancestral population. Genetics 170(4): 1849–1856.

19.

Garrigan

Mobasher

Severson

, et al. (2005b) Evidence for archaic Asian ancestry on the human X chromosome. Molecular Biology and Evolution 22(2): 189–192.

20.

Gibbs

(2020) Ancient DNA and modern identity: The promise and pitfalls. Leviathan 11(1): 14–18.

21.

Golenberg

Giannasi

Clegg

, et al. (1990) Chloroplast DNA sequence from a Miocene Magnolia species. Nature 344(6267): 656–658.

22.

Gravitz

(2010) Neanderthal genome rewrites human evolution. MIT Technology Review, 6 May. Available at: www.technologyreview.com/2010/05/06/203848/neanderthal-genome-rewrites-human-evolution/ (accessed 10 September 2020).

23.

Green

Briggs

Krause

, et al. (2009) The Neandertal genome and ancient DNA authenticity. The EMBO Journal 28(17): 2494–2502.

24.

Green

Krause

Briggs

, et al. (2010) A draft sequence of the Neandertal genome. Science 328(5979): 710–722.

25.

Green

Krause

Ptak

, et al. (2006) Analysis of one million base pairs of Neanderthal DNA. Nature 444(7117): 330–336.

26.

Green

Malaspinas

Krause

, et al. (2008) A complete Neandertal mitochondrial genome sequence determined by high-throughput sequencing. Cell 134(3): 416–426.

27.

Hakenbeck

(2019) Genetics, archaeology and the far right: An unholy Trinity. World Archaeology 51(4): 517–527.

28.

Harding

Fullerton

Griffiths

, et al. (1997) Archaic African and Asian lineages in the genetic ancestry of modern humans. American Journal of Human Genetics 60(4): 772–789.

29.

Harmon

(2018) Why white supremacists are chugging milk (and why geneticists are alarmed). New York Times, 17 October. Available at: www.nytimes.com/2018/10/17/us/white-supremacists-science-dna.html (accessed 10 September 2020).

30.

Harris

Hey

(1999a) Human demography in the Pleistocene: Do mitochondrial and nuclear genes tell the same story? Evolutionary Anthropology: Issues, News, and Reviews 8(3): 81–86.

31.

Harris

Hey

(1999b) X chromosome evidence for ancient human histories. Proceedings of the National Academy of Sciences of the United States of America 96(6): 3320–3324.

32.

Hawks

(2010) Neandertals live. John Hawks Weblog, 6 May. Available at: http://johnhawks. net/weblog/reviews/neandertals/neandertal_dna/neandertals-live-genome-sequencing-2010.html (accessed 10 September 2020).

33.

Hawks

(2019) How can archaeologists make a difference to media and public perceptions of ancient DNA? The SAA Archaeological Record 19(1): 39–43.

34.

Hawks

Cochran

(2006) Dynamics of adaptive introgression from archaic to modern humans. PaleoAnthropology 2006: 101–115.

35.

Hawks

Cochran

Harpending

, et al. (2008) A genetic legacy from archaic homo. TRENDS in Genetics 24(1): 19–23.

36.

Hedges

Schweitzer

Henikoff

, et al. (1995) Detecting dinosaur DNA. Science 268(5214): 1191–1194.

37.

Henn

Gignoux

Lin

, et al. (2008) Y-chromosomal evidence of a pastoralist migration through Tanzania to Southern Africa. Proceedings of the National Academy of Sciences of the United States of America 105(31): 10693–10698.

38.

Hodgson

Bergey

Disotell

(2010) Neandertal genome: The ins and outs of African genetic diversity. Current Biology 20(12): R517–R519.

39.

Hodgson

Disotell

(2008) No evidence of a Neanderthal contribution to modern human diversity. Genome Biology 9(2): 206.

40.

Hotz

(2010) Most people carry Neanderthal genes. The Wall Street Journal, 6 May. Available at: www.wsj.com/articles/SB10001424052748703686304575228380902037988 (accessed 10 September 2020).

41.

Krings

Stone

Schmitz

, et al. (1997) Neandertal DNA sequences and the origin of modern humans. Cell 90(1): 19–30.

42.

Labuda

Zietkiewicz

Yotova

(2000) Archaic lineages in the history of modern humans. Genetics 156(2): 799–808.

43.

Lessem

(1993) Designing dinosaurs: How to bring Jurassic Park to life. Omni 15(9): 50–56.

44.

Absher

Tang

, et al. (2008) Worldwide human relationships inferred from genome-wide patterns of variation. Science 319(5866): 1100–1104.

45.

Lindahl

(1993) Recovery of antediluvian DNA. Nature 365(6448): 700.

46.

Max Planck Society (2010) The Neandertal in us. Press release, Max Planck Society for the Advancement of Science Press and Public Relations Department, 7 May. Available at: www.eva.mpg.de/fileadmin/content_files/genetics/neandertal-genome/pdf/PR_MPI_Neandertal_EN.pdf (accessed 10 September 2020).

47.

National Institutes of Health (2010) Complete Neanderthal genome sequenced. Press release, 6 May. Available at: www.genome.gov/27539119/2010-release-complete-neanderthal-genome-sequenced (accessed 10 September 2020).

48.

Olivieri

Achilli

Pala

, et al. (2006) The mtDNA legacy of the Levantine early Upper Palaeolithic in Africa. Science 314(5806): 1767–1770.

49.

Ottenburghs

(2019) Why do some humans have Neanderthal DNA? Frontiers Young Minds 7: 104.

50.

Pääbo

(2011) DNA clues to our inner Neanderthal. TED Talks. Transcript. Available at: www.ted.com/talks/svante_paabo_dna_clues_to_our_inner_neanderthal/transcript?language=en (accessed 10 September 2020).

51.

Pääbo

(2014) Neanderthal Man: In Search of Lost Genomes. London: Hachette UK.

52.

Pääbo

Poinar

Serre

, et al. (2004) Genetic analyses from ancient DNA. Annual Review of Genetics 38(1): 645–679.

53.

Panofsky

Donovan

(2019) Genetic ancestry testing among white nationalists: From identity repair to citizen science. Social Studies of Science 49(5): 653–681.

54.

Parham

Arnett

Adams

, et al. (1994) The HLA‐B73 antigen has a most unusual structure that defines a second lineage of HLA‐B alleles. Tissue Antigens 43(5): 302–313.

55.

Pickrell

Patterson

Loh

, et al. (2014) Ancient west Eurasian ancestry in southern and eastern Africa. Proceedings of the National Academy of Sciences of the United States of America 111(7): 2632–2637.

56.

Plagnol

Wall

(2006) Possible ancestral structure in human populations. PLoS Genetics 2(7): e105.

57.

Price

(2020) Africans carry surprising amount of Neanderthal DNA. Science, 30 January. Available at: www.sciencemag.org/news/2020/01/africans-carry-surprising-amount-neanderthal-dna (accessed 10 September 2020).

58.

Reich

(2018) Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past. Oxford: Oxford University Press.

59.

Retzbach

Maier

(2015) Communicating scientific uncertainty: Media effects on public engagement with science. Communication Research 42(3): 429–456.

60.

Rohde

Olson

Chang

(2004) Modelling the recent common ancestry of all living humans. Nature 431(7008): 562–566.

61.

Saini

(2019) Superior: The Return of Race Science. London: Beacon Press.

62.

Shapiro

(2015) Mammoth 2.0: Will genome engineering resurrect extinct species? Genome Biology 16(1): 1–3.

63.

Smith

Falsetti

Simmons

(1995) Circum Mediterranean biological connections and the pattern of late Pleistocene human evolution. In: Ullrich H (ed) Man and Environment in the Paleolithic. Liège: University of Liège, pp. 197–207.

64.

Stringer

Bräuer

(1994) Methods, misreading, and bias. American Anthropologist 96(2): 416–424.

65.

Templeton

(2002) Out of Africa again and again. Nature 416(6876): 45–51.

66.

Templeton

(2005) Haplotype trees and modern human origins. American Journal of Physical Anthropology 128(S41): 33–59.

67.

The Economist (2010) A cave man blinking in the light. The Economist, 6 May. Available at: www.economist.com/science-and-technology/2010/05/06/a-cave-man-blinking-in-the-light (accessed 10 September 2020).

68.

Tishkoff

Reed

Friedlaender

, et al. (2009) The genetic structure and history of Africans and African Americans. Science 324(5930): 1035–1044.

69.

University of Montreal (2011) Non-Africans are part Neanderthal, genetic research shows. ScienceDaily, 18 July. Available at: www.sciencedaily.com/releases/2011/07/110718085329.htm (accessed 10 September 2020).

70.

Wade

(2008) Regenerating a mammoth for $10 million. The New York Times, 19 November.

71.

Wade

(2010) Analysis of Neanderthal genome points to interbreeding with modern humans. The New York Times, 7 May.

72.

Wall

Hammer

(2006) Archaic admixture in the human genome. Current Opinion in Genetics & Development 16(6): 606–610.

73.

Wall

Kim

(2007) Inconsistencies in Neanderthal genomic DNA sequences. PLoS Genetics 3(10): e175.

74.

Wang

Lachance

Tishkoff

, et al. (2013) Apparent variation in Neanderthal admixture among African populations is consistent with gene flow from non-African populations. Genome Biology and Evolution 5(11): 2075–2081.

75.

Wolinsky

(2019) Ancient DNA and contemporary politics: The analysis of ancient DNA challenges long‐held beliefs about identity and history with potential for political abuse. EMBO Reports 20(12): e49507.

76.

Wolpoff

Caspari

(1997) Race and Human Evolution. New York: Simon and Schuster.

77.

Wolpoff

Hawks

Frayer

, et al. (2001) Modern human ancestry at the peripheries: A test of the replacement theory. Science 291(5502): 293–297.

78.

Woodward

Weyand

Bunnell

(1994) DNA sequence from Cretaceous period bone fragments. Science 266(5188): 1229–1232.

79.

Yotova

Lefebvre

Moreau

, et al. (2011) An X-linked haplotype of Neandertal origin is present among all non-African populations. Molecular Biology and Evolution 28(7): 1957–1962.