Let the Dead Rest
- Alina Chen
- 1 day ago
- 6 min read
Why de-extinction is a seductive idea we should resist
There is a photograph taken in 1933 of the last known thylacine, more commonly known as the Tasmanian tiger, pacing its concrete enclosure at Hobart Zoo. It circles. It circles again. Three years later it was dead, and with it went an entire lineage of a marsupial predator, four million years in the making. When people look at that photograph, maybe they share articles, feel briefly concerned, but eventually move on. That fleeting moment of grief is real, and it matters. However, it should not be the emotion that drives our science policy. This is because when it is channeled into the dream of resurrection, it then risks becoming something more dangerous: a story we tell ourselves to avoid the harder moral reckoning that extinction imposes.
"Every act of de-extinction is, at its root, an act of substitution — and substitution is not the same as restoration."
De-extinction is the project of recovering extinct species through cloning, gene-editing, or selective back-breeding. This fascinating novel solution has attracted serious scientific and commercial investment in recent years. Colossal Biosciences, backed by hundreds of millions in venture capital, is working simultaneously on woolly mammoth, thylacine and dodo revival programs amongst other extinct mammals (Colossal Biosciences 2023). The ambition and sheer innovation and drive to do this is breathtaking. So is the confidence. Counterintuitively, it is precisely that confidence and technological surety that we can undo what has been done which deserves the most scrutiny.
Let us begin with what de-extinction actually produces. It does not produce the original animal. It produces a genetic approximation: an organism with some, or even many, of the target species' genes expressed in a living body, but without the accumulated epigenetic history. The learned behaviours, the gut microbiome, the social knowledge that every wild animal inherits is not just from DNA but from its mother, its group, its landscape (Shapiro 2015). The woolly mammoth Colossal proposes to release into Siberian tundra would be an Asian elephant carrying perhaps fifty edited genes. It would have never learned to navigate permafrost in a herd. Nor would it cultural memory of which vegetation to seek, which predators to fear, how to find water under ice. Every act of de-extinction is, at its root, an act of substitution — and substitution is not the same as restoration. We would be releasing our idea of a mammoth into a world that has spent ten thousand years learning to live without one.
This matters ecologically in ways that are easy to underestimate. Take, for example, the ecosystems into which de-extincted animals would be introduced back into. They are not the ecosystems those animals left. For instance, the Siberian tundra of today is warmer, differently vegetated, and populated by entirely different assemblages of species than the Pleistocene steppe the mammoth evolved to inhabit. Reintroducing a large grazer into a transformed habitat is not pressing an undo button. Rather, it is introducing a new variable into a complex system whose behaviour we cannot reliably predict. The history of species introductions, even deliberate and carefully planned ones, offers little comfort here. Previous examples in history include the cane toad in Australia, the grey squirrel in Britain and the Nile perch in Lake Victoria. The overarching theme was that each species was introduced with confidence, and each reshaped ecosystems in ways their advocates failed to anticipate (Clavero and García-Berthou 2005). A proxy mammoth that fails to behave as modelled because it lacks the social structures needed to graze at scale or because it carries unexpected disease susceptibilities could damage the very permafrost it was meant to protect.
But the deeper problem is not scientific. It is psychological, or perhaps moral. There is something in the de-extinction project that flatters our species enormously, and flattery is always worth being suspicious of. The narrative it offers is seductive: yes, we broke it, but we can fix it. Our ingenuity is not just the cause of the crisis but its solution. We tell ourselves that technology will redeem technology. This is a very comfortable story, and is therefore the first red flag. Comfort, in a crisis, is often a warning sign. The environmental philosopher Val Plumwood argued that one of the core pathologies of industrial modernity is what she called "the dream of total mastery", defined as the refusal to accept that some processes are genuinely irreversible and that some losses cannot be engineered away, that finitude is not a problem to be solved but a condition to be acknowledged (Plumwood 2002). De-extinction, in its most utopian form, is a manifestation of exactly that dream.
Philosopher Ronald Sandler (2014) has argued that where humans directly caused an extinction, a moral obligation to remedy the damage follows. I find this persuasive in the abstract, but less so in practice. The obligation to repair harm does not automatically license any particular method of repair, especially when that method carries its own serious risks. A surgeon who accidentally punctures a lung has an obligation to the patient, but not an obligation to attempt an experimental procedure whose failure rate is unknown. The more honest fulfilment of our obligation to the thylacine, the passenger pigeon, the dodo, is not to build a simulacrum of them in a laboratory. It is to protect, with everything we have, the species that are still here. There are currently over 47,000 species on the IUCN Red List threatened with extinction. Their situation is urgent, their habitat loss is accelerating, and they are competing for conservation funding with biotechnology projects whose appeal lies partly in being spectacular and toeing the line between ethics and the pursuit of knowledge (IUCN 2025). The opportunity cost of de-extinction is not hypothetical. In reality, it is being paid right now, in every conservation dollar that flows toward a resurrected mammoth and away from an extant pangolin.
The thylacine's extinction should make us feel the weight of what is irreversible, not wonder at our capacity to undo the past.
There is also the question of what de-extinction does to our relationship with loss and with our own culpability. Extinction is among the most morally serious things a species can cause. It is permanent, total, and visited not just on individual animals but on an entire evolutionary lineage, on all the creatures those animals would have been and the ecosystems they would have shaped across millions of years of future time. To absorb that properly requires us to sit with a kind of grief that is genuinely uncomfortable. It should hold the power to change how we act, how we legislate, and how we live through moral obligation and regret. The promise of technological resurrection offers an escape route from that discomfort. If we can bring them back, the urgency of preventing extinction in the first place is diluted, skipping our entire pathos. Minteer and Collins (2012) identified this moral hazard as one of the most serious systemic risks of the de-extinction programme. The thylacine's extinction should make us feel the weight of what is irreversible, not wonder at our capacity to undo the past.
None of this is to say that genetic science has nothing to offer conservation. Gene banking, the preservation of frozen tissue and viable cells from endangered species, is an invaluable use of biotechnological resources. It is a kind of biological insurance against catastrophe (Ryder et al. 2000). Genetic diversity analysis can inform breeding programmes for critically endangered animals. These applications are modest, grounded, and directed toward the living. They do not require us to believe that we can rewind evolution and play God. De-extinction in its fullest, most publicised form does require that belief, not because science is untrustworthy, but because the history of human intervention in ecosystems gives us every reason to approach the idea of irreversible releases of engineered organisms with caution.
The mammoth is not coming back. What is coming back, or what could be, is something new. It could be a creature shaped by our desires, our regrets and our extraordinary technical capacity, released into a world we have already changed almost beyond recognition. That is not resurrection. It is, in a meaningful sense, the creation of a new kind of wilderness: one that exists not because nature made it, but because we decided we wanted it back. Whether that is hubris or hope depends on how well it goes and how the media will advertise it. But the history of confident human interventions in complex systems suggests that we should not be gambling the world's remaining wild places on our own ingenuity — not again, and not yet.
References
Clavero, M & García-Berthou, E 2005, 'Invasive species are a leading cause of animal extinctions', Trends in Ecology & Evolution, vol. 20, no. 3, p. 110.
Colossal Biosciences 2023, Woolly Mammoth Revival Program: 2023 Progress Update, Colossal Biosciences, Dallas, viewed 23 March 2026, <https://colossal.com>.
IUCN 2025, The IUCN Red List of Threatened Species: Summary Statistics, International Union for Conservation of Nature, Gland, Switzerland, viewed 18 March 2026, <https://www.iucnredlist.org>.
Minteer, BA & Collins, JP 2012, 'Ecological ethics in captivity: balancing values and responsibilities in zoo and aquarium research under rapid global change', ILAR Journal, vol. 53, no. 3–4, pp. 439–450.
Plumwood, V 2002, Environmental Culture: The Ecological Crisis of Reason, Routledge, London.
Ryder, OA, McLaren, A, Brenner, S, Zhang, YP & Benirschke, K 2000, 'DNA banks for endangered animal species', Science, vol. 288, no. 5464, pp. 275–277.
Sandler, RL 2014, 'The ethics of reviving long-extinct species', Conservation Biology, vol. 28, no. 2, pp. 354–360.
Shapiro, B 2015, How to Clone a Mammoth: The Science of De-extinction, Princeton University Press, Princeton.
Comments