After the uproar over the de-extinction of dire wolves, bioethicist Arthur Caplan asked in Plos Biology: “Should scientists be allowed to bring distant human ancestors back to life?” The Italian edition of Scientific American invited me to investigate how technically difficult it would be to de-extinct a Neanderthal and what the risks and benefits of such a project might be. Below are the statements provided by the specialists I consulted.

TOM GILBERT (paleogeneticist, University of Copenhagen) asked to give an estimate of the genetic differences between modern humans and Neanderthals: “I believe there are on the order of tens of thousands (∼12,000 in one estimate) of single‐nucleotide substitutions that are human‐lineage derived since the split from Neandertals. But of those, only ca 40 changes cause amino‐acid substitutions in humans. If one looks more broadly to non coding elements that might have regulatory changes, one might say thousands of the differences are plausible candidates for functional impact, but only a small minority of all differences are likely to have strong functional consequences. I think the technical problem therefore is not testing the function of the 40 or so that cause exonic differences, but what all the putative other ones might do.”

TOM GILBERT, again, asked about the technical hurdles and biological unknowns: “It’s a tough question as one of the challenges with ‘de-extinction’ is the very loose (if not completely absent) definition of success. At one extreme people argue a recreated form has to be 100% genetically identical to the lost form. At the other end you have people arguing that introduction of small numbers of genes that enable a modern (modified) organism to fill a niche formally filled by an extinct species, is enough. Obviously its technically a lot easier to do the latter than the former. However regardless, some key challenges would be:

Identifying what genetic differences were actually those key to enabling the lost form to survive in its environment. We are of course very biased to what we guess was important, rather than what was important. I mean obviously hair on a mammoth seems important. But a study maybe 10+ years ago showed through resurrecting mammoth haemoglobin, that probably structural differences between elephant and mammoth haemoglobin were relevant to explain why mammoths didn’t experience hypothermia. Something that would be hard to guess at unless you really know your stuff.

You can only introduce mutations you know about. The more divergent an extinct from extant form there is, the more likely you don’t recover all the genomic information of the extinct form. Especially for divergent parts of the genome. Which are probably those evolving fastest, so may well be those most important to what made the extinct form unique.

One other complication is that many genes work in concert with other genes. So just changing one may break a whole chain. So one has to have a good understanding of how genes are linked. Which may be information we don’t have (even in modern humans). For an analogy, consider taking a Fiat500. And trying to convert it to a monster truck by adding monster truck wheels. It simply won’t work as you also have to change the axels, and the wheel arches and the engine and probably the base plate etc etc etc. Obviously we know a lot about how all the car bits affect each other…but how genes interact is very very complex!

ALYSSON MUOTRI (developmental biologist working with neanderthalized mini-brains, University of California, San Diego) asked about the technical feasibility of a real de-extinction project: “Currently it would be impossible to resurrect a complete Neanderthal. You could get really close by editing the 61 protein-coding genes, but still many regulatory regions are also needed to complete the task. The only way with current technology is not by editing, but by nuclear transfer. I invite you to watch my discussion with Tim Disney about his movie William.

MARTA TOMASI (BioLaw expert, University of Trento) interviewed on the legal aspects of such an experiment: “There are numerous regulations around the world that prohibit experiments on embryos aimed at introducing genetic modifications that can be passed on to descendants, and even more that prohibit the implantation of modified embryos to initiate pregnancy. A further obstacle would arise from interpreting the proposed experiments as forms of crossbreeding between different species. If this were the correct interpretation, such attempts would encounter the limits set by international instruments such as the Universal Declaration on the Human Genome and Human Rights (UNESCO, 1997) – which considers interventions on germ cells to be potentially “contrary to human dignity” – and the Oviedo Convention on Biomedicine (CoE, 1997) – which excludes interventions aimed at introducing changes in the genome of descendants. As the preparatory work for both acts seems to indicate, they aim to protect, first and foremost, the integrity and identity of the human species as a whole (‘not only the individual but the species itself’).

ALTA CHARO (bioethicist, University of Wisconsin) asked about the utility of discussing the ethical legitimacy of ventures that are unrealistic or very far in the future: “I think discussing future applications of technology is very useful, as it allows up plenty of time to imagine both the benefits and risks of the various ways a technology may be used, and to examine whether current regulatory structures and business models are adequate for the task for balancing innovation with overall distribution of risks and benefits.  But this is a difficult task, because it can be very tempting to make unrealistic projections about the range of applications, the civilian and military appetite for those applications, and the effect the applications might have on future individuals or society as a whole.  In general, it is helpful to look back at predicate innovations that led to the technology under discussion, and take some lessons from how those predicate technologies have fared.  For example, when making predictions about the likely pattern of uses for heritable genome editing (if it is ever shown to be safe enough), it is useful to look back at whether and how older reproductive technologies (like artificial insemination and IVF and preimplantation diagnostics) did or did not expand beyond their initial uses, and where they did not, to look at the inherent limitations (biological or social) that kept them from expanding to the point of widespread uses (such as those predicted in dystopian analyses).”

ALTA CHARO, again, when asked whether she wanted to emphasize any of the usual arguments raised in the de-extinction debate (such as the impossibility of obtaining informed consent, the potential psychological suffering of a de-extinct individual, the health risks associated with extensive genome editing, and the risks to a gestational carrier if an artificial womb is not available). “The hypothetical development of a hominid species that is likely to have inferior capacities in some respects as compared to homo sapiens raises serious concerns about the moral status of these beings.  Given that they would likely be considered to be of lower moral status, more akin to non-human primates, I think the most pressing ethical concern is the risk of re-creating the worst abuses of our history (and to some extent, our current experiences) with slavery.  Slavery was a torture that violated the rights of the slaves and eroded the moral status of all who owned, used or benefited from slaves”.

You may also be interested in REBECCA WRAGG SYKES‘s opinion about the utility of a Neanderthal de-extinction effort. Here is the excerpt from a Live Science article:

“It would fail to answer virtually all of the interesting things that we would want to know about Neanderthals,” Rebecca Wragg Sykes, an archaeologist and author of “Kindred: Neanderthal Life, Love, Death and Art” (Bloomsbury, 2020), told Live Science. For example, many researchers are interested in understanding how sophisticated Neanderthal spoken language was. A combination of anatomical studies, genetic information and advanced tool technology suggests they communicated with one another, but possibly without metaphors or abstract words. Still, even if scientists could bring back a healthy Neanderthal and tried to teach him or her a modern human language, that wouldn’t provide much insight into the language they actually used, Wragg Sykes said, just like reconstructing an ancient musical instrument wouldn’t tell you what songs people played on it. “Neanderthals were their own enormous range of different populations with cultural variations, culinary and very likely linguistic diversity,” Wragg Sykes said. Since none of the natural context of Neanderthals’ development is replicable, “taking a shortcut by cloning or recreating Neanderthals isn’t going to get you there.” The same logic holds for other abilities. Teaching one re-created Neanderthal how to read or solve an algebraic equation would only provide as much information about ancient Neanderthals as testing one living human would reveal about the abilities of all humans on the planet.