Sometimes, I wonder about philosophers. I can sort of see that considering counterfactuals and even absurd premises is useful and can generate novel insights, but I wonder at how some can adopt the weirdest assumptions. As in this paper, How to Tell If Animals Can Understand Death, which begins like so:

It is generally assumed that humans are the only animals who can possess a concept of death.

Now that is just bad writing. Passive voice, broad statement as fact, no supporting evidence…but mostly, who assumes animals can’t have a concept of death? Not me. Not the general public. I doubt that most ethologists think that. There are far too many empirical observations of animals fearing death, or behaving as if grieving, or avoiding the dead. I think most mammals have an understanding of death; other groups, like birds and reptiles, probably do but are harder for people to read, and may have a very different concept of what death is.

The great mystery is whether spiders have a concept of death, or are simply content to be bringers of death to others.

Anyway, after that initial off-putting sentence, the rest of the paper is much better and more interesting, and is trying to bring some rigor to how we evaluate the conception in non-human animals. I can get into that.

It is generally assumed that humans are the only animals who can possess a concept of death. However, the ubiquity of death in nature and the evolutionary advantages that would come with an understanding of death provide two prima facie reasons for doubting this assumption. In this paper, my intention is not to defend that animals of this or that nonhuman species possess a concept of death, but rather to examine how we could go about empirically determining whether animals can have a concept of death. In order to answer this question, I begin by sketching an account of concept possession that favours intensional classification rather than mere extensional discrimination. Further, I argue that the concept of death should be construed as neither binary nor universal. I then present a proposal for a set of minimal conditions that must be met to have a concept of death. I argue that having a minimal understanding of death entails first expecting a dead individual to be alive, and then grasping its non-functionality and irreversibility. Lastly, I lay out the sort of observational and experimental evidence that we should look for to determine whether animals have the capacity for a minimal comprehension of death.

I appreciate that the concept of death is nonbinary. In particular, what comes to mind is that some religious people don’t have a grasp on it — they think the dead people are still alive in some other realm! — so if we demanded that the concept entails an absolute, definitive appreciation of non-existence, then we’d have to argue that some humans lack a concept of death. The author considers that in the paper, and I think it’s fair to say that a concept of an afterlife complicates the understanding of death, but doesn’t negate it.

Contrary to what is often assumed by animal ethicists, the concept of death should not be viewed in binary terms. Possessing a concept of death is not an all-or-nothing matter but rather something that is subject to gradation. This becomes clear once we consider the case of human children, who do not acquire a concept of death overnight. In fact, the scientific consensus is that it takes them an average of 10 years to fully master the concept of death (Kenyon 2001), but we credit them with some understanding of death before they reach this stage. If we accept a gradation in the case of human children, we should also accept it in the case of animals.

So let’s break down what a thorough understanding of what death is would entail. See, this is where philosophers shine, in organizing the ideas!

1. Non-functionality: death implies the cessation of all bodily and mental functions.

2. Irreversibility: dead individuals cannot come back to life.

3. Universality: all living things, and only living things, die.

4. Personal mortality: death will also apply to oneself.

5. Inevitability: eventually, all living things must die.

6. Causality: death occurs due to a breakdown in the bodily functions.

7. Unpredictability: it is impossible to know in advance the exact timing of death.

That’s where the nonbinariness is relevant. I would say that I personally accept all seven, probably most of you do too, but I know people who would reject many of those points, yet still understand the idea of death. They don’t reject them with good reason, of course, but because of some a priori commitment to supernatural beliefs. So the author distills this list down to a minimal collection of behaviors characters of individuals who understand death.

A creature can be credited with a minimal concept of death once she classifies some dead individuals as dead with some reliability, where ‘dead’ is understood as a property that pertains to beings who:

(a) are expected to have the cluster of functions characteristic of living beings, but

(b) lack the cluster of functions characteristic of living beings, and

(c) cannot recover the cluster of functions characteristic of living beings.

Not addressed: I wonder, then, if by these criteria someone in a vegetative state should be considered “dead”. I know, nonbinary again, but something that might be tested/applied in humans.

All right, now it’s time to apply these ideas to animals. What would we expect to see if an animal understands death? There is a long list of things to look for, but to keep it simple, I’ll just consider how they apply to spiders, as a test case.

1. Varied behaviour towards corpses

This one is a little unfair to spiders. The idea is that the subject should exhibit some confusion about how to respond to corpses, but to a spider, which is primarily a tactile animal, a dead animal more or less becomes invisible and ceases to exist. That they can still feed on immobilized prey suggests there is more complexity than that, though.

2. Unhygienic/maladaptive behaviour towards corpses

Interesting. Disposing of or burying corpses is indicative of a hard-wired response, while maladaptive diddling of dead bodies implies that their behavior is not hardwired, and suggest the possibility that they could acquire more complex ideas about death. Hmm. I’ve seen spiders drag corpses into their nests and festoon the place with the bodies of their prey, so maybe this is a positive for them.

3. Different treatment of corpses vs. asleep individuals

Recognizing the difference between asleep and dead (or comatose and dead?) sounds like an important concept. But what does a sleeping spider look like? I can’t tell. Can they?

4. Investigative behaviour towards corpses

Hey, you wanna see a dead body? When a spider finds a dead animal, and hoists it into her nest in the same way she would a bit of bark or a leaf, is that curiosity? I don’t know.

5. Aggressive behaviour towards corpses

Nope, I don’t see that in spiders. If they can’t eat it or mate with it, there’s no reason to fight it.

6. Caring behaviour towards beings with limited functionality

In spiders? Ha ha, no.

7. Mourning behaviour towards corpses

This one is tricky. The author cites returning repeatedly to corpses, or in the case of primates, carrying dead infants around. My spiders don’t exhibit parental care (other than the fact that they don’t eat newborns, which is a generous notion of “care” — I was a good daddy because I didn’t gnaw on my children, either), but sometimes their nests look like graveyards.

8. Eventual ignoring or abandoning of a corpse

No fair! It’s a sign of understanding when they haul a dead baby around, but also when they throw the dead baby aside? OK, spiders are very good at ignoring corpses.

9. Age or experience-relative difference in behaviour towards corpses

As animals mature and gain experience, their treatment of the dead may change, indicating that their behavior is learned. I haven’t seen that in spiders — the babies are ravenous little beasts, the adults are ravenous big beasts. They’re also so short-lived and usually solitary, so they don’t have the opportunity to learn.

While I appreciate the effort, I think one problem here is that the author is more familiar with domesticated mammals than more alien species. Can you compare, for instance, the concept of death between carnivores and herbivores? How does a vulture view death? How do you practically assess the concept of death in a solitary ambush predator that doesn’t engage in the kinds of social interactions we take for granted? Maybe spiders are acutely aware of the meaning of dead vs alive, in the sense that it is their business to make live things dead, but they don’t have the kind of behavioral repertoire that would allow us to assess that fact.

I guess I’m just going to have to ask them this morning.