Woolly mammoths are in the news again. According to recent news reports, they are "on the verge of resurrection" and "will be back from extinction within two years." The prompt for this was a statement by the Harvard geneticist George Church, lead scientist for the "Woolly Mammoth Revival" project.
However, in the wake of the excitable reports, other commentators have been urging caution. The palaeontologist John Hawks went so far as to call the reports of the mammoth's imminent revival "fake news".
So are shaggy-coated pachyderms really poised to stampede across the frozen wastelands of Siberia any time soon? If such creatures are created, will they really be mammoths? And even setting aside exactly when this might happen, is it a good idea to bring a species back from the dead?
I am a former cell biologist and have spent the last two years researching and writing about de-extinction, the science of bringing extinct animals back to life. I have spoken with Church, as well as many of the other scientists at the forefront of de-extinction research, and two things are clear to me. The first is that a living, breathing woolly mammoth is far from imminent, and the second is that, nevertheless, the science needed to make it is progressing at quite a lick.
The woolly mammoth would not be the first species to be revived after having been wiped out. The world's first de-extinct animal was born in 2003 in a laboratory in Spain. It was a type of wild mountain goat called a bucardo, or Pyrenean ibex.
The animal only lived for a few short minutes, but since then the science underpinning de-extinction has progressed in leaps and bounds. The researchers who brought back the bucardo tell me that only two things stand between them and a whole herd of healthy de-extinct bucardo: time and money.
Church is not talking about making a living, breathing, fur-coated mammoth calf
Such revivals depend on manipulating basic biology.
An animal's life begins when a single cell splits in two, then keeps on dividing to form a miniscule bundle of cells called an "embryo". In time, the cells of the embryo divide and become more specialised. Different tissues, organs and limbs form, leading eventually to a complete living animal.
Anyone wanting to make a woolly mammoth must first create a single mammoth cell, coax it into becoming an embryo, and then nurture that embryo while it grows into an entire mammoth.
Speaking ahead of this week's annual meeting of the American Association for the Advancement of Science (AAAS) in Boston, Church said he thought he was a couple of years away from making an embryo. In my opinion, this is entirely possible, but only so long as you pay close attention to what Church is claiming.
The first thing to bear in mind is that Church is not talking about making a living, breathing, fur-coated mammoth calf. His work so far focuses solely on single cells.
For the last few years, Church and his team have been carefully placing mammoth genes into the cells of Asian elephants, woolly mammoths' closest living relatives. They are endowing these elephant cells with the "essence of mammoth": the potential to grow long, luscious locks, lay down Sumo-wrestler-sized rolls of body fat, and other characteristics that should help the animal to look and act like a mammoth.
It will be an animal whose DNA is largely elephant, but with a smattering of judiciously-placed mammoth DNA
When I spoke to Church in 2015, he had already added in the gene for mammoth haemoglobin. Haemoglobin is the molecule in red blood cells that ferries oxygen around the body. The mammoth version works even at low temperatures; essential for an animal, like the woolly mammoth, that evolved during an ice age.
It is now two years later and Church's team have made 45 changes to the elephant genome. They are moving closer to their goal of "mammoth-ifying" an elephant cell.
But they may have a long way to go. A 2015 study compared the woolly mammoth genome with that of living Asian elephants, and identified changes in 1,642 genes. Church's 45 changes amount to a small fraction of that total, but Church argues that he does not need to put every single tweak into his elephant cells to make something mammoth-like.
Church has never claimed that he is making a bona-fide, genuine, 100% authentic woolly mammoth. Instead it will be an animal whose DNA is largely elephant, but with a smattering of judiciously-placed mammoth DNA.
He talks of making a "mammoth-elephant hybrid" or a "cold-adapted elephant". I prefer to call it a "mammophant" or "elemoth".
In order to make a mammophant embryo, Church's team will also need to advance the science of cloning
Assuming Church manages to create a mammophant cell, the next step will be to convert it into a mammophant embryo. One way to do this is to insert the mammophant DNA into an egg cell – again, from an Asian elephant – that has had its DNA removed. This is the same technology used for cloning.
In 1996, Dolly the sheep became the first mammal to be cloned from an adult cell. Many different mammalian species have since been cloned, but the elephant is not among them. Cloning research suggests that, just because it is possible to clone one type of animal, it is not necessarily easy to clone another, even if they belong to closely-related species.
One key problem is that there is still a lot we do not understand about cloning. During the process, the egg somehow reprograms the newly-inserted DNA to a more youthful state so that it can drive embryonic development. It is a bit like restoring the factory settings on a mobile phone, but no one knows how it happens or how to fully control it. So in order to make a mammophant embryo, Church's team will also need to advance the science of cloning.
It is a mammoth task (sorry about that) but it is not impossible. However, even if Church pulls it off, he will still need to find a way to nourish the embryo while it grows.
When I first heard about attempts to bring back the woolly mammoth, I was not keen. There was an elephant in the room, which most people either failed to notice or chose to ignore: in order to revive the woolly mammoth, scientists would have to perform experiments on elephants.
These are not animals to be experimented with
First, elephant eggs would be needed for the cloning process. Then, surrogate elephant mothers would be required to nurture the developing embryos in their wombs, then give birth to them and bring them up. Since cloning is an inexact science that produces far more failures than live births, this would put the health of the elephant surrogate at risk.
The trouble is that there are so few elephants left. The number of Asian elephants has more than halved over the last three generations because of habitat loss and poaching, and they are classed as "endangered".
In short, these are not animals to be experimented with. I am not keen on bringing back the woolly mammoth at the expense of an animal so deserving of our care.
However, Church is unfazed. He believes he can perform his experiments without risking any elephants at all.
The elephant eggs needed for cloning need not come from the reproductive systems of living elephants. There are other ways to make them.
Animal studies have shown that skin cells can be reprogrammed in a dish to make stem cells, which can then be coaxed to form eggs. It is almost like cellular alchemy. Rather than performing invasive surgery to retrieve eggs from adult female elephants, it should be possible to make them from non-invasive skin biopsies.
The technology of artificial wombs is still, well, embryonic
Once the embryos have been made, they could be nurtured, not in adult animals, but in "artificial wombs". This may sound like the stuff of science fiction, but Church and other researchers are quietly developing the science needed to make it happen. He explained last week that researchers in his lab have grown a mouse embryo in an artificial womb for 10 days; halfway through its normal gestation period.
In theory, artificial wombs would get around the surrogate mother problem. But it will be no mean feat to scale them up from mice to mammoths.
Mouse pregnancies last for 20 days, while mammoth ones take over two years. And while a newborn mouse weighs as much as a paperclip, a newborn mammoth calf is the size of a washing machine. What's more, each species will need a different cocktail of "ingredients" to help its embryo grow.
In short, the technology of artificial wombs is still, well, embryonic.
At this point it should be clear that de-extinction is profoundly difficult, and that we are not going to have any living mammoths roaming the Siberian tundra in the next few years. But at the same time, it seems likely that if scientists like Church keep pushing, they will get there eventually.
The bigger question is: should we do it? What would be the point?
Advocates of de-extinction will tell you there are many reasons to bring extinct species back to life.
We live in the midst of a global biodiversity crisis, with up to 150 species disappearing every day. Extinction rates are currently 1,000 times higher than during pre-human times. We are responsible for this travesty. So if the tools exist to undo some of that damage, maybe we have a moral obligation to do just that.
De-extinction should be about producing populations of genetically-diverse animals that can breed and live in the wild
However, I think we need to think a little more deeply about it.
In 2014, ecologist John Ewen of the Institute of Zoology in London, UK helped design a checklist. The idea was to help researchers select the most suitable candidates for de-extinction. The paper went on to influence a set of de-extinction guidelines drawn up by the International Union for Conservation of Nature (IUCN).
The checklist essentially asks whether, if we re-created an extinct species, it could survive in our modern world. Among other things, it asks: whether there is suitable habitat to be had; if the original cause of extinction is still present; whether the animal could become a pest; and how its revival might affect people living nearby.
"Resurrecting the woolly mammoth should be more than the production of a biological orphan destined for an exhibit," says Ewen. "To embrace this technology with responsibility, we should consider the loftier goals of ecological enrichment, and assess the feasibility and risks surrounding mammoth reintroduction."
In other words, de-extinction should be about producing populations of genetically-diverse animals that can breed and live in the wild, where they can contribute to the ecosystem.
All living things provide vital services to the ecosystems they are part of. There are pollinators, seed dispersers, water purifiers, pest controllers, and much else.
An estimated 500 gigatonnes of organic carbon is locked up in the Arctic permafrost
Mammoths were gardeners. In their heyday, they trundled around eating grass, trampling saplings, and fertilising the ground with their nutrient-rich dung. When they went extinct, there was no other species to do these jobs. As a result, the rich "Mammoth steppe" turned to scraggy, mossy tundra.
If we were to plug the ecological niche left vacant by mammoths' demise, it may be that they could return the unproductive wastelands of northern Siberia to their biodiverse heyday. But no one knows for sure.
"There will be uncertainty here," says Ewen. "We need to judge the capacity for the mammoth to restore such habitat."
Similarly, Church argues that mammoths could help stop dangerous climate change.
A series of as-yet-unpublished experiments in Siberia have suggested that, where large grazers are present, soil temperatures are several degrees colder than where they are absent.
It would still take well over half a century to make a single viable herd
"When big animals graze, they trample the snow, which exposes the surface of the soil to cold air," says biologist Sergey Zimov of the Northeast Science Station in Cherskii, Russia, who conducted the research. "This helps keep the ground frozen."
An estimated 500 gigatonnes of organic carbon is locked up in the Arctic permafrost. That is two to three times as much as in all the existing rainforests combined. As our climate changes and the Arctic melts, scientists have warned that this "carbon time bomb" could go off and accelerate global warming still further.
However, if we were to repopulate the wilds of Siberia with herds of woolly mammoth, the argument goes, they might prevent the carbon time bomb from going boom.
It is an enticing argument, but not without flaws.
Suppose Church perfects all the methods that he needs and makes a viable hybrid "mammoth" embryo in the timescale he imagines. It would then take a further two years for that embryo to turn into a calf, and another 15 years for it to reach sexual maturity. Even if all the technical hurdles involved in making a mammoth were overcome tomorrow, it would still take well over half a century to make a single viable herd, which would not be anywhere enough to do the job.
There appear [to be] some uncertainties that may reduce the value of the mammoth as a de-extinction candidate
Instead, by that time, if current predictions are to be believed, the Arctic permafrost will already have melted. What's more, the Siberian ecosystem may have changed too much and may be unable to support the new arrivals.
In other words, we cannot look to mammoths to help solve global warming. The real story is that climate change undoubtedly played a role in polishing mammoths off last time round: they died out when Earth's climate warmed and the ice retreated. So if anything, climate change makes the future of any de-extinct mammoth looks decidedly woolly.
"Using the IUCN reintroduction criteria, there appear [to be] some uncertainties that may reduce the value of the mammoth as a de-extinction candidate," says Ewen.
At this point, it may seem that bringing back woolly mammoths is a hopeless prospect. The challenges are certainly formidable. But that is not to say that de-extinction is without value.
For me, the single biggest argument in favour of resurrecting extinct species is the beneficial spin-offs it will create for endangered living species.
Conservationists plan to use genome editing and cloning to help save black-footed ferrets
The same techniques being used to bring back extinct animals – stem cell technology, genome sequencing and editing, and assisted reproduction techniques such as cloning – can be used, far more easily, on threatened living species.
In New Zealand, genome sequencing is being used to help rescue the kākāpō; a large, green, flightless parrot. The species is suffering from fertility problems and inbreeding. Researchers are currently decoding the full genetic codes of all 154 living kākāpōs, and plan to use the information to help steer their breeding program. By mating only the most genetically distinct birds, they hope to produce offspring that are more healthy and fertile.
Similar plans are also afoot in North America.
Conservationists plan to use genome editing and cloning to help save black-footed ferrets. In the 1980s, these feisty little creatures were on the brink of extinction, partly because of an infectious disease called sylvatic plague. A decades-long captive breeding program has kept the species going, but there are still only a few hundred in the wild.
Worse, today's wild black-footed ferrets are massively inbred. They are all effectively second cousins. This is not good for the long-term health of the species.
If both of these projects pan out, they would make conservation history
However, stored away in vats of liquid nitrogen are the frozen cells of wild ferrets that were captured, but then died without passing on their genes. These cells are a resource waiting to be used.
In 2015, the pro-de-extinction organisation Revive and Restore submitted two ground-breaking proposals to the US Fish and Wildlife Service, the body that oversees American nature.
The first is to use the frozen cells to make new, genetically vibrant animals. These will then be able to breed naturally, giving the population back some of its lost genetic variation.
But there is no point going to all this trouble if the animals are then set free and become infected with plague. So the second proposal seeks to edit the ferret's genome to make it plague-resistant.
If both of these projects pan out, they would make conservation history. Even better, similar techniques could then be applied to countless other species.
In Kenya, the Ol Pejeta Conservancy is home to the world's last three northern white rhinos. These magnificent beasts were once widespread in east and central Africa before poaching, civil war and habitat loss took their toll.
We do not know if projects like this will work. But desperate times call for desperate measures
The last three – Sudan, Najin and Fatu – are grandfather, mother and daughter. They are too old, too ill and too closely-related to ever breed naturally. Their subspecies is functionally extinct and they are the walking dead.
However, stem cell technology and assisted reproduction could even now pull them back from the brink. Led by veterinarian Thomas Hildebrandt of the Leibniz Institute for Zoo and Wildlife Research in Berlin, Germany, researchers plan to create some truly special test-tube babies.
Using sperm from male northern white rhinos, collected and stored before they died, and eggs created by the same method that Church is developing for his mammoths, the team hope to create the embryos of northern white rhinos in a dish. These will then be transferred into the surrogate wombs of the animal's closest living relative, the southern white rhino, who will carry the little embryos to term.
We do not know if projects like this will work. But desperate times call for desperate measures.
The world is warming, habitats are being lost, many places are becoming more polluted, and the biodiversity crisis is deepening. De-extinction, for all its pitfalls, is worthy of our attention because of the potential it has to help repair ailing ecosystems.
De-extinction is not just about bringing back the dead
In short, this is a bigger issue than simply whether we could, or should, bring back woolly mammoths. They are only one species, and there are many we could restore – whether already extinct, or merely close to it.
When de-extinct animals finally blossom into healthy wild populations, they could vastly boost overall levels of biodiversity, not just by their presence, but by the impact they will have on the other species in their ecosystems.
What's more, the knowledge we are gaining from figuring out how to bring back extinct species is already having a positive effect on the wildlife that we still have. In other words, de-extinction is not just about bringing back the dead. It is also about helping the living.
Helen Pilcher (@HelenPilcher1) is a freelance science writer. Her book Bring Back the King: The new science of de-extinction was published in 2016 by Bloomsbury.
Join over six million BBC Earth fans by liking us on Facebook, or follow us on Twitter and Instagram.
If you liked this story, sign up for the weekly bbc.com features newsletter called "If You Only Read 5 Things This Week". A handpicked selection of stories from BBC Future, Earth, Culture, Capital and Travel, delivered to your inbox every Friday.