Did Neanderthals Have Language? Part 1: Language Genes

The most cited research pertaining to language evolution and Neanderthals is almost always the case made from genetics. The reason for this is that the information we have obtained from ancient DNA is relatively more straightforward than what we can even begin to attempt to deduce from anatomy. The only problem with this research is that as straightforward as it is, it also so happens that its case right now is one of the weakest.

Introduction

• Introduction and Terms Glossary
• Part 2: The Anatomy of Speech
• Part 3: The Anatomy of Hearing
• Part 4: Speculations and Moving Forward

This is the first part of my short series on Neanderthal language. If you have not read the introduction to the series, you can follow the link above which also includes a terms glossary. Also linked above is the next part of the series.

This post is a bit dense in terms coming from genetics, so if you don’t understand something, I encourage you to follow up in the glossary. I have tried to summarize the most difficult terms as best as I can for any laypeople who I hope are reading. For our geneticists, I have left out a lot of the story to keep things short and have simplified a few things until I can think of a better way of explaining them (linkage-disequilibrium).

FOXP2 – The Language Gene

In 2001 a protein called FOXP2 was discovered which, when mutated in humans, results in severe speech deficiencies and a severe reduction in language skills. This discovery was the result of genetic sequencing of a human family in which approximately half the members expressed a complete lack of speech abilities¹. This disability caused by the mutation is characterized by the inability to perform complex motor control with the mouth in humans as well as an inability to understand the basic rules of language.

The gene coding FOXP2 (the FOXP2 gene), which is typically strongly conserved in mammals, was immediately implicated by the authors of its first study as being relevant to the evolution of language. As it turned out, compared to mice, the FOXP2 gene found in humans differ by only three amino acids². Two of these changes in amino acids occurred after the divergence between humans and chimpanzees 5 to 7 million years ago. Not only was it obvious to geneticists that it was phenotypically important, but it was also found that the FOXP2 gene underwent a >60 fold increase in substitution rates during the divergence between hominins and chimpanzees, indicating strong selection on it during that period³.

It appeared likely that changes in this single gene yielded significant phenotypic changes in some way contributing to the origins of speech and language in humans.

A radial phylogeny showing relative rates of FOXP2 evolution among mammals. The central square includes baboons and all of the African apes.⁴

The ultimate function of FOXP2 was not immediately obvious to geneticists. Shortly after its discovery in humans, FOXP2 was also implicated in the evolution of other vocal animals. In particular, one study showed that FOXP2 was essential in the evolution of vocalizations in echolocating bats and echolocating shrews, but not in echolocating whales and dolphins (as seen in the phylogeny above, it has since proved to be quite important for bats)⁴. FOXP2 was also shown to have diversified in vocal-learning species of bird⁵.

In addition to FOXP2, other genes have since been implicated in the evolution of language through studies of other people with genetically involved language impairments including SRPX2, CNTNAP2, ATP2C2, and a number of others⁶.

Neanderthals

It wasn’t long after the initial discovery that scientists at the Neanderthal Genome Project looked for FOXP2 in the Neanderthal genome. In 2006 Swedish biologist Svante Pääbo at the Max Planck Institute for Evolutionary Anthropology announced his team would attempt to sequence the entirety of the Neanderthal genome. Pääbo had previously been one of the first researchers to place the FOXP2 gene in an evolutionary context by studying its molecular evolution in respect to other mammals². By November of that year they had sequenced 3 million base pairs of a Neanderthal genome⁷. Almost exactly a year later in 2007, Pääbo’s team published a study finding that the same two derived amino acid substitutions of the FOXP2 gene that humans had differed in contrast with chimpanzees were present in Neanderthals⁸.

This finding was much to the triumph of many paleoanthropologists who had long crusaded for the similarities between modern humans and Neanderthals. Erik Trinkaus, who is arguably the world’s foremost leading expert on Neanderthals, in summarizing the team’s findings stated that the results, “address, albeit indirectly, the ways in which we conceive of the emergence of modern human biology and behavior. For this reason, if no other, it is welcome.⁹”

Of course, this triumph was not very long lasting. The first study of FOXP2 in Neanderthals had only confirmed that this gene only shared those two critical base pairs humans had substituted since the split from chimpanzees. Because the technology was not yet available, Pääbo’s team was not able to analyze the gene in its entirety, but they had found something else of interest. Four years before the start of the Neanderthal Genome Project began in 2006, Pääbo’s lab had previously analyzed the FOXP2 gene in humans to measure if it had been strongly selected for since the divergence from chimpanzees. While their findings confirmed that strong selection had taken place, one of the humans sequenced in this study, an individual from Nigeria, had a FOXP2 gene which differed from other humans in at least 16 different positions on the gene, seven of which were ancestral (found also in chimpanzees). Noticing these peculiarities, Pääbo’s team sequenced the Nigerian gene again in 2009 and this time included 14 additional Nigerians¹⁰. Shockingly, the team found that both sides of the genome next to FOXP2‘s two important substitutions had changed quite significantly, meaning that the powerful selection sweep they had detected on the gene in humans in 2002 had happened somewhere other than the site differing from chimpanzees. Other labs had come to a similar conclusion, arguing that the selection Pääbo’s team detected could have only taken place at some point after the Neanderthal-human split and suggesting that Pääbo’s team had botched their sensitive sample¹¹.

The team proposed several scenarios, one of which was that there may have been multiple instances of selection on FOXP2 leading to modern-day humans. In 2010 Pääbo’s team finally released a draft sequence of the entire Neanderthal genome, and in 2012 they published another study on FOXP2 titled, “A Recent Evolutionary Change Affects a Regulatory Element in the Human FOXP2 Gene.^12,13” This time, with the combined genomes of the Neanderthals and Denisovans, the authors were able to analyze 94.3% of the archaic human FOXP2 gene. Walking back some of the conclusions pulled from their earlier paper, the authors found that the FOXP2 gene found in Neanderthals and Denisovans was actually different from that in modern humans in at least two different positions which were found to be highly conserved across all other vertebrates. One of these positions had been the site at which the strong selection they detected earlier must have occurred. This site, position 114076877 of the genome, had been conserved across all tetrapod animals plus archaic humans for the last 700 million years but differed only in the human genome. This was likely the target for the other selective sweep that Pääbo’s lab had detected in 2002.

What Do The Language Genes Do?

What the difference between the Neanderthal genome and the human genome at position 114076877 actually means for human language is unfortunately not currently well understood, and this happens to be the case with most of the language genes. Most of our understanding of the function of these genes comes from comparative studies with other animals and laboratory studies of knockout mice whereby small changes are made to the genome and resulting phenotypes are recorded.

As FOXP2 was the first of these genes discovered, it is the one we know the most about. Knockout studies of FOXP2 in mice showed that when you knocked-out one copy of the gene, the typical ultrasonic vocalizations made by infant mouse-pups were moderately impaired in addition to their motor skills¹⁴. Mice who had both copies of the gene knocked-out were severely impaired in both of these regards. It has also been found that specialized motor movements are affected by these changes, much like the inability to perform complex mouth movements in the original family in which FOXP2 was discovered¹⁵. Another study following single-copy knockout mice throughout their development found that despite still possessing the ability to vocalize these mice did not emit the rhythmic and syllabic structures that characterize mouse courtship songs¹⁶. A similar study showed developmental issues in knockout zebra finches¹⁷. This time, the developmental issues were made in direct association with vocal learning, which is a necessary prerequisite for language acquisition. In this case, knockout finches were unable to copy the songs of other finches.

What about the site that differs between Neanderthals and humans? Well, position 114076877 is part of a small, 14 base-pair area of the genome which binds to a protein known as POU3F2. Currently the function of POU3F2 is not well understood, although changes in its protein have been implicated in association with a number of neuropsychiatric disorders and learning disabilities. Some authors have implied that POU3F2 has much to do with language development¹⁸. The evidence thus far in humans has only shown that the issues associated with disruption in the protein are obesity, bipolar disorder, and learning disabilities- not language issues¹⁹. One study, using knockout mice, found that mammalian-specific sequences in POU3F2 were associated with maternal retrieval behaviors²⁰. One might speculate that this might be important for language given some of the hypotheses regarding its evolution. Namely, Dean Falk’s “putting the baby down hypothesis,” suggested that talking to babies as a method to calm them down was perhaps an initial causal factor in the development of human language²¹. Others have suggested that song arose before language²², while still others have suggested that lullabies specifically were among the earliest forms of song²³.

With all this said, no humans have been described with point mutations in the POU3F2 14 base-pair sequence yet, and those with changes in the POU3F2 protein seem to be able to speak. If such a human were found, they would essentially possess the Neanderthal FOXP2 gene sequence. Despite whatever developmental issues such a person might possess, they would be critically informative to our understanding of what this change actually does, assuming some interaction does not make this a lethal mutation for humans. Pääbo’s team suggested in their 2012 study that the protein may regulate expression of FOXP2 as they are both expressed in the same areas of the central nervous system, but the phenotypic consequences of the change in the sequence are yet unknown and may be completely unassociated with the ability to speak.

Conclusion

What does all of this actually mean for Neanderthals? Some scientists, including the geneticist Simon Fisher who was on the original FOXP2 study lament when they hear any of these genes described as language genes. Despite this, calling these genes language genes is partially fair. It’s apparent that FOXP2 is critical for language in humans and studies in mice have shown us that this connection to FOXP2 is not arbitrary as it’s linked to both vocal production and vocal learning. Despite us not yet having found a human with a point-mutation in POU3F2 similar to the Neanderthal mutation, one can make several conclusions about Neanderthals based on what we do know.

Knockout mice contain the same base-pair change at POU3F2 that Neanderthals did, and disruption of their FOXP2 showed that FOXP2 itself is still critical for vocal development. This case was replicated across birds, and it was shown that FOXP2 has critically diversified across bats. One point that is often made regarding FOXP2 is that not only is it expressed in the brain, but it is also expressed in the lungs and other systems essential for altering vocalizations as in mice. But this gene was also associated with changes in rapid orofacial control and specialized motor movements. The fact that the two amino acid substitutions happened in both humans and Neanderthals means that at some point before their split, there was some selection pressure on a highly conserved gene associated with vocalizations and vocal learning to split. FOXP2 is still critical for language.

Moving Forward

Other genes have been found which are associated with FOXP2, and still others unassociated with FOXP2 have been found which are critical to our understanding of language development in humans. I won’t go into the details of what others have found, but the work there is important and still ongoing in our exploration of which genes affect our ability to communicate with language. What has been achieved with our understanding of FOXP2 must be accomplished with other genes, and work is in progress around the world to accomplish this.

Currently the evolution of FOXP2 and its interacting genes across primates is little understood. This was going to be my project when I set out for grad school. As far as I am aware, until a study on the variation of FOXP2 across the apes published last December (which showed it was relatively conserved except in humans and orangutans), there had been no real phylogenetic treatment of FOXP2 in primates aside from the bat study referenced above²⁴. This seems to be a pretty important piece of the puzzle, and it might be worth it to look at other acoustically adapted primates to see which phenotypic traits these changes are associated with. As we saw in the Nigerian sample, there are also population differences between polymorphisms FOXP2 in humans and the genes which interact with it. There are also a great deal of autism-associated studies. Additionally, the advent of CRISPR gives us the possibility of more easily inserting a Neanderthal FOXP2 gene, complete with its point-mutation at POU3F2 to study the effects of this allele²⁵.

Studies of FOXP2 and its associates are not worthless. Some people dismiss FOXP2 as if it is rather arbitrary and had no importance in our evolutionary past, but the literature on its function and the variation we see in humans is still moving rapidly. I’ll say here that it is rather obvious that FOXP2 was critical in the evolution of language. The picture right now on Neanderthal language from a genetic standpoint is relatively narrow and our knowledge base is quite limited. There are a number of other genes and functions we have to sort out, and while we’re not certain if Neanderthals had all of these, the science is moving!

In the future this information is more than likely going to be critical, but for now further work is necessary. Since the literature is far denser than what I have shown you here, you can probably expect me to return to this topic in the near future after I have finished the Neanderthal series. Next week, I will be shifting the conversation to talk about what we know from anatomy to answer some other essential questions related to speech production in the Neanderthals.

Like my blog and want to support my writing? Please consider supporting me on Patreon.

References

¹Lai, C.S., Fisher, S.E., Hurst, J.A., Vargha-Khadem, F. and Monaco, A.P., 2001. A forkhead-domain gene is mutated in a severe speech and language disorder. Nature, 413(6855), p.519.

²Enard, W., Przeworski, M., Fisher, S.E., Lai, C.S., Wiebe, V., Kitano, T., Monaco, A.P. and Pääbo, S., 2002. Molecular evolution of FOXP2, a gene involved in speech and language. Nature, 418(6900), p.869.

³Zhang, J., Webb, D.M. and Podlaha, O., 2002. Accelerated protein evolution and origins of human-specific features: Foxp2 as an example. Genetics, 162(4), pp.1825-1835.

⁴Li, G., Wang, J., Rossiter, S.J., Jones, G. and Zhang, S., 2007. Accelerated FoxP2 evolution in echolocating bats. PLoS one, 2(9), p.e900.

⁵Webb, D.M. and Zhang, J., 2004. FoxP2 in song-learning birds and vocal-learning mammals. Journal of Heredity, 96(3), pp.212-216.

⁶Mozzi, A., Forni, D., Clerici, M., Pozzoli, U., Mascheretti, S., Guerini, F.R., Riva, S., Bresolin, N., Cagliani, R. and Sironi, M., 2016. The evolutionary history of genes involved in spoken and written language: beyond FOXP2. Scientific Reports, 6, p.22157.

⁷Green, R.E., Krause, J., Ptak, S.E., Briggs, A.W., Ronan, M.T., Simons, J.F., Du, L., Egholm, M., Rothberg, J.M., Paunovic, M. and Pääbo, S., 2006. Analysis of one million base pairs of Neanderthal DNA. Nature, 444(7117), p.330.

⁸Krause, J., Lalueza-Fox, C., Orlando, L., Enard, W., Green, R.E., Burbano, H.A., Hublin, J.J., Hänni, C., Fortea, J., De La Rasilla, M. and Bertranpetit, J., 2007. The derived FOXP2 variant of modern humans was shared with Neandertals. Current Biology, 17(21), pp.1908-1912.

⁹Trinkaus, E., 2007. Human evolution: Neandertal gene speaks out. Current Biology, 17(21), pp.R917-R919.

¹⁰Ptak, S.E., Enard, W., Wiebe, V., Hellmann, I., Krause, J., Lachmann, M. and Pääbo, S., 2009. Linkage disequilibrium extends across putative selected sites in FOXP2. Molecular Biology and Evolution, 26(10), pp.2181-2184.

¹¹Coop, G., Bullaughey, K., Luca, F. and Przeworski, M., 2008. The timing of selection at the human FOXP2 gene. Molecular Biology and Evolution, 25(7), pp.1257-1259.

¹²Green, R.E., Krause, J., Briggs, A.W., Maricic, T., Stenzel, U., Kircher, M., Patterson, N., Li, H., Zhai, W., Fritz, M.H.Y. and Hansen, N.F., 2010. A draft sequence of the Neandertal genome. Science, 328(5979), pp.710-722.

¹³Maricic, T., Günther, V., Georgiev, O., Gehre, S., Ćurlin, M., Schreiweis, C., Naumann, R., Burbano, H.A., Meyer, M., Lalueza-Fox, C. and de la Rasilla, M., 2012. A recent evolutionary change affects a regulatory element in the human FOXP2 gene. Molecular Biology and Evolution, 30(4), pp.844-852.

¹⁴Shu, W., Cho, J.Y., Jiang, Y., Zhang, M., Weisz, D., Elder, G.A., Schmeidler, J., De Gasperi, R., Sosa, M.A.G., Rabidou, D. and Santucci, A.C., 2005. Altered ultrasonic vocalization in mice with a disruption in the Foxp2 gene. Proceedings of the National Academy of Sciences of the United States of America, 102(27), pp.9643-9648.

¹⁵Fujita, E., Tanabe, Y., Shiota, A., Ueda, M., Suwa, K., Momoi, M.Y. and Momoi, T., 2008. Ultrasonic vocalization impairment of Foxp2 (R552H) knockin mice related to speech-language disorder and abnormality of Purkinje cells. Proceedings of the National Academy of Sciences, 105(8), pp.3117-3122.

¹⁶Castellucci, G.A., McGinley, M.J. and McCormick, D.A., 2016. Knockout of Foxp2 disrupts vocal development in mice. Scientific Reports, 6, p.23305.

¹⁷Haesler, S., Rochefort, C., Georgi, B., Licznerski, P., Osten, P. and Scharff, C., 2007. Incomplete and inaccurate vocal imitation after knockdown of FoxP2 in songbird basal ganglia nucleus Area X. PLoS Biology, 5(12), p.e321.

¹⁸Benítez-Burraco, A. and Boeckx, C., 2014. FOXP2, retinoic acid, and language: a promising direction. Frontiers in Cellular Neuroscience, 8, p.387.

¹⁹Kasher, P.R., Schertz, K.E., Thomas, M., Jackson, A., Annunziata, S., Ballesta-Martinez, M.J., Campeau, P.M., Clayton, P.E., Eaton, J.L., Granata, T. and Guillén-Navarro, E., 2016. Small 6q16. 1 deletions encompassing POU3F2 cause susceptibility to obesity and variable developmental delay with intellectual disability. The American Journal of Human Genetics, 98(2), pp.363-372.

²⁰Nasu, M., Yada, S., Igarashi, A., Sutoo, D.E., Akiyama, K., Ito, M., Yoshida, N. and Ueda, S., 2014. Mammalian-specific sequences in pou3f2 contribute to maternal behavior. Genome Biology and Evolution, 6(5), pp.1145-1156.

²¹Falk, D., 2004. Prelinguistic evolution in early hominins: Whence motherese?. Behavioral and Brain Sciences, 27(4), pp.491-503.

²²Masataka, N., 2007. Music, evolution and language. Developmental Science, 10(1), pp.35-39.

²³Mehr, S.A. and Krasnow, M.M., 2017. Parent-offspring conflict and the evolution of infant-directed song. Evolution and Human Behavior, 38(5), pp.674-684.

²⁴Staes, N., Sherwood, C.C., Wright, K., Manuel, M., Guevara, E.E., Marques-Bonet, T., Krützen, M., Massiah, M., Hopkins, W.D., Ely, J.J. and Bradley, B.J., 2017. FOXP2 variation in great ape populations offers insight into the evolution of communication skills. Scientific Reports, 7(1), p.16866.

²⁵Singh, P., Schimenti, J.C. and Bolcun-Filas, E., 2014. A mouse geneticist’s practical guide to CRISPR applications. Genetics, pp.genetics-114.