"Mass" Effects

Leonard Finkelman writes...

It is a stormy night in 2217 when Eve Zien takes on the mantle of Death. She sits at her desk in the Nordic Genetic Resource Center, staring at her hands in disbelief. NordGen is the world’s leading biodiversity bank, providing digital storage for the genetic information of over one million species. By the 23rd century, that genetic information is all that’s left of 90% of those species. Eve is one of NordGen’s software engineers, responsible for maintaining the digital ark that carries the remainder of Earth’s biodiversity. At least, all of that was true until just a few minutes ago.

Working late into the night, Eve had gotten bored and (she had to admit) a little stir crazy. She pulled up the bank’s source code. On a lark, she entered the RESET command. Then she executed it, just to see what would happen, because boredom is the bad-influence-friend of stupidity.

Now NordGen doesn’t have a biodiversity bank anymore.

Eve breathes slowly. NordGen has an offsite backup that can restore any lost data. The North American Federation has its own genetic bank. Even if Eve can’t undo her last command, one of those facilities will keep the species’ legacies safe. But what if the storm knocked out the backup facility? What if the NAF’s bank suffered its own failure? What if Eve really just killed 90% of Earth’s species?

Eve thinks: “I wish someone had thought of this contrived scenario a couple of hundred years ago, back when all of this could have been avoided.”


A few weeks ago I very nearly deleted all of my emails when, for want of something better to do, I hit the "Delete" key while my Inbox was highlighted. A few days later--after I manually restored some 9,000 messages--part of the Svalbard Global Seed Vault flooded. Everyone involved in both incidents avoided catastrophe. The whole thing nevertheless got me thinking: what if some future seed vault or gene bank vault hires an idiot like me? Forget floods--idle hands could be the greatest danger to Earth's future biodiversity.

These thoughts came back to mind this past week, when an article by Peter Brannen began making the rounds on my social media feeds. The headline asserted: “Earth Is Not in the Midst of a Sixth Mass Extinction.” There have been similar headlines (atop similar articles) in the recent past. These articles are set against contradictory (and widely publicized) claims made by scientists and science journalists.

I guess it all depends on what we mean by “mass extinction.” This is a convenient development for contributors to a philosophy blog named “Extinct.”

Image courtesy Wikimedia Commons.

Image courtesy Wikimedia Commons.

In the brief foray into science fiction given above, should we say that Eve caused a mass extinction? If so, then it would follow that we probably are in the middle of a sixth mass extinction event. If not, then maybe we aren’t.


Defining “extinction” seems like a good starting point for defining “mass extinction.” As we'll see, it might not be, really. But we'll start there anyway, because doing so will make clear why Eve's story should be a compelling one. It also helps that I’ve gone through a similar exercise before and can rely on some of that previous work.

Spoiler alert for everyone who hasn’t read that earlier post: there are lots of different definitions of “extinction.” Some of those definitions are consistent with paleontological theory and others aren’t. I previously came to the conclusion that paleontology assumes that extinction is defined by the disappearance of all the individual organisms within the species. [1] This assumption is also borne out in practice: paleontologists estimate extinction dates based on the probability of finding a specimen from a species in a rock stratum (Marshall 2010; Bradshaw, et al. 2012).

As it turns out, my reading of this and other practices in paleontology may be too weak. Adrian has already dissected Charles Marshall’s sure-to-be-influential proclamation of paleobiology’s “five laws,” but it’s worth focusing again here on the last of those laws. Marshall defines extinction as the erasure of a clade’s history, ‘making it difficult to infer ancestral character traits and its diversity dynamics from the living biota alone’ (2017, 4). In other words, extinction is a loss of information rather than the death of organisms.

Defining extinction as a loss of a species’ information is similar to what Delord calls “true” or “final” extinction (2007, 660). By this standard, seed vaults (such as the one in Svalbard) or gene banks (such as the one that Eve disrupts) stave off a species' extinction by preserving information relevant to the species' propagation. In fact, the species only goes extinct when that information becomes irretrievable. Most of the time information is lost because its organic storage medium dies and decomposes, but it's conceivable that the information could be lost by digital means (cf. Delord 2007, 660-661; Siipi and Finkelman 2016, 11-12).

What all of this should show is that it isn't unreasonable to imagine that, given an appropriate state of technology, an action like Eve's could really (rather than metaphorically) cause a species' extinction. I’ll admit that my attempt at flash fiction isn’t going to win a Nebula Award. (Maybe prove me wrong, award voters?) Its point is just to serve as a basis for conceptual analysis. Grant that Eve's data erasure causes the immediate extinction of 900,000 out of 1,000,000 species left on Earth. [2] Is that big, scary number enough to qualify her action as a mass extinction event?

Mass extinctions (by the numbers)

Proponents of the "Anthropocene" mass extinction argue that mass extinctions are all about big, scary numbers. Their arguments go like this: some minimal proportion of species, P, went extinct during previous mass extinctions and we project current species loss to exceed P, so we must be living through a mass extinction.

Mass extinctions visualized. The graph's y-axis represents the number of genera to go extinct. Image courtesy Wikimedia Commons.

Mass extinctions visualized. The graph's y-axis represents the number of genera to go extinct. Image courtesy Wikimedia Commons.

There are (at least) five undisputed mass extinctions in the paleontological record. [3] The first came at the end of the Ordovician period, when 60-70% of all species went extinct. The second happened in the late Devonian period, when more than 70% of species went extinct. The third marked the end of the Permian with a loss of 90-96% of the planet’s species. The fourth, when 70-75% of species went extinct, now defines the boundary between the Triassic and Jurassic periods. The last mass extinction famously put an asteroid-sized punctuation mark on the Cretaceous period, killing more than 75% of all species then on Earth.

Ceballos, et al. (2010) and Ceballos, et al. (2015) argue that similar big, scary numbers are here now. They cite as evidence the extinction rate among vertebrates--that is, the number of species to disappear per 100 years--which is now significantly higher than at any time since the end-Cretaceous extinction, and may in fact exceed the extinction rate during that event. Wilson (2002) estimates that half of the multicellular species extant in the year 1500 will be extinct by the year 2100, which is less than a blink of Mother Earth's tectonic eye. These numbers are big enough and scary enough to cross the threshold between "extinctions" and "mass extinctions."

Eve's science-fictional blunder also crosses that threshold. Her data erasure falls into the "mass extinction" category only if a mass extinction just is a rapid loss of a lot of species, and a 90% extinction rate would rival the end-Permian event. Eve might feel better about herself, then, if she can access what will by her time be a 200-year-old essay in The Atlantic.

Mass extinctions (Behind the numbers)

Following the lead of paleontologist Douglas Erwin--whose lead is a good one to follow, given that he literally wrote the book on prehistory's worst mass extinction--Brannen argues that mass extinctions aren't really about the numbers themselves. The big scary numbers are necessary for a mass extinction, but they might not be sufficient.

That probably sounds weird, as if Erwin and Brannen are saying that mass extinctions aren't really a lot of extinctions. And yes: that's exactly what I think they're saying. As it turns out, I don't think their claim is that weird or that crazy.

I'm willing to bet that at least a few readers don't think that Eve's actions would constitute a mass extinction, even if they were to grant that Eve had really caused a very large number of extinctions. The term "mass extinction" conjures imagery (much) more dramatic than a wayward keystroke. Brennan even uses the descriptor "hellscape." A large number of species go extinct during a mass extinction, but the numbers are a consequence of some worse-than-biblical event. What's really important is that epochal cause.

It can be tough to draw a substantive line between a mass extinction event and another kind of event that just happens to correlate with a bunch of extinctions. There have been several attempts to bring the "Big Five" events together under the umbrella of a common cause--Peter Ward's Medea hypothesis, for example, or appeals to a "Death Star"--but none are widely accepted. To show where, how, and why he draws the line, Erwin compares the end-Permian extinction with the 2003 American Northeast blackout.

Science at its worst? Image courtesy of Wookieepedia. [4]

Science at its worst? Image courtesy of Wookieepedia. [4]

In both of the events that Erwin cites, a cascade of relatively minor events added up until catastrophe struck. Brennan quotes Erwin: "everything is fine until it's not ... then everything goes to hell." The 2003 blackout was unique because a large number of power grids collapsed as a result of a single unified cause. The power grids could have collapsed independently, and while that would have been a really weird day, it wouldn't have been an historical event. Similarly, 90-95% of Earth's species could coincidentally die off around the same time in a statistical fluke--Marshall's first paleobiological law is that all species eventually go extinct (2017, 1)--but without the proverbial fire or brimstone it just wouldn't be a real mass extinction.

Erwin suggests that the essence of mass extinctions is not the cause itself, but a particular feature of the cause: unstoppability. The 2003 blackout became unstoppable once the cascade of events crossed a certain threshold, and that's what Erwin and Brennan take to be the point of analogy with the end-Permian event. We can say the same for the other four mass extinction events, too: each was a true "end-" time (read "end-hyphen," as a prefix before the appropriate geological period) because there were huge changes coming that couldn't be stopped.

The primary reason we can't be in the midst of a mass extinction (according to Erwin and Brennan) is that we still have reasonable hope to save some species from extinction. In a true mass extinction, no earthly effort could save a species in decline like the polar bear or Amur leopard. In a true mass extinction, conservation efforts would be better spent--perhaps paradoxically--on "successful" species like rats or deer (or, indeed, humans) to ensure that something makes it through the panoptic crucible (Erwin 1991; Willis, et al. 2007). Modus tollens: if bio-conservation isn't a massive waste of resources, then we're not in our own "end-" times.


Deleting a bunch of emails isn't the end of the world. This is true in a metaphorical sense (and some of us have to learn that the hard way). What normally goes without saying is that it's also true in a literal sense. It's worth making that explicit here because the same reasoning explains why we shouldn't think of Eve as an agent of mass destruction.

Eve might have killed a lot of species, but she isn't (won't be? wouldn't then be?) in any "end-" times. [5] Her action will have no substantive effect on anything other than some abstract species count. Her Earth's environments and ecosystems--such as they might be--will continue unaffected by her momentary lapse. Life goes on, as they say, unless "they" are in the middle of a mass extinction.

Are we? It's tough to answer that question on anything other than a downer of a note. I don't think that the story at the top of this post is one about a mass extinction, and that commits me to Erwin's qualitative definition of mass extinctions. The conclusion that we aren't in the midst of a mass extinction follows deductively from Brennan's other premises, but at least one of those premises is a flimsy one. We do have an active, vital conservation movement that does give some of us hope for the future. But what we hope to be true isn't necessarily what is true. Our hope--the one that supposedly belies the claim that we are in a mass extinction--may be a false one.

We can't fairly say that we know whether or not we're in a mass extinction, one way or the other. Worse, this is one case wherein "plan for the worst, hope for the best" doesn't seem like viable advice, given that the actions we'd take in the worst-case scenario (conserve the least endangered species) are diametrically opposed to those we should take in the better case (conserve the most endangered species). So it's not like we can split the difference there.

And perhaps this is one of the great virtues of a blog as opposed to a journal: I can just punt the rest of this over to you, the reader, for more discussion in the comments.


[1] Delord (2007) called this "demographic" extinction. By contrast, "functional" extinction is defined as the cessation of a species "life processes" (e.g., maintenance of a viable gene pool, participation in ecological processes, etc.). Functional extinction is consistent with extinct species having living members--my preferred example being Benjamin, the thylacine endling.

[2] You can change the details of Eve's story, if you'd prefer. If "demographic" extinction better matches your concept of extinction, then Eve can run (and ruin) a massive zoo and botanical garden. If you prefer "functional" extinction, then she can oversee a planet-sized ecological preserve and mess that up somehow. The details matter less than the basic idea: a vast majority of Earth's species disappear almost instantaneously due to one person's technological bumbling.

[3] Two other events in the geologic past might qualify as mass extinctions, although both are debatable. A period of volcanism at the end of the Capitanian age seems associated with a significant loss of species, but it’s unclear whether or not the relevant extinctions should be folded into the broader end-Permian event. The Great Oxygenation Event of the early Proterozoic eon might also qualify--if the single-celled organisms of the time could fairly be said to form species at all, which is no sure thing.

[4] For those not already in the know: this was not a joke and Wookieepedia is, in fact, a thing.

[5] One would presume that the story does mark the end of Eve's tenure with NordGen. If someone wants to pick up the sequel, I'm willing to sign the rights off.


  1. Bradshaw, C. J. A., Cooper, A., Turney, C. S. M., & Brook, B. W. (2012). Robust estimates of extinction time in the geological record. Quaternary Science Reviews, 33, 14-19.
  2. Ceballos, G., Ehrlich, P. R., Barnosky, A. D., García, A., Pringle, R. M., & Palmer, T. M. (2015). Accelerated modern human–induced species losses: Entering the sixth mass extinction. Science advances, 1(5), e1400253.
  3. Ceballos, G., García, A., & Ehrlich, P. R. (2010). The sixth extinction crisis: loss of animal populations and species. Journal of Cosmology8(1821), 31.
  4. Delord, J. (2007). The nature of extinction. Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences, 38(3), 656-667.
  5. Erwin, T. L. (1991). How many species are there?: revisited. Conservation Biology5(3), 330-333.
  6. Marshall, C. R. (2010). Using confidence intervals to quantify the uncertainty in the end-points of stratigraphic ranges. Paleontological Society Papers, 16, 291-316.
  7. ---. (2017). Five palaeobiological laws needed to understand the evolution of the living biota. Nature Ecology & Evolution, 1(6), s41559-017.
  8. Siipi, H., & Finkelman, L. (2016). The Extinction and De-Extinction of Species. Philosophy & Technology, 1-15.
  9. Willis, K. J., Araújo, M. B., Bennett, K. D., Figueroa-Rangel, B., Froyd, C. A., & Myers, N. (2007). How can a knowledge of the past help to conserve the future? Biodiversity conservation and the relevance of long-term ecological studies. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 362(1478), 175-187.
  10. Wilson, E. O. (2002). The future of life. Vintage.