The Elephant’s Child & Supernatural Punishment

Here is an explanation, courtesy of Rudyard Kipling, for why elephants’ trunks are so long. One day a curious young elephant (with a short nose) wondered what crocodiles ate for dinner. After some adventure – and, oddly, quite a lot of spanking – the elephant found a crocodile, and put him the question. When the elephant bent down to hear the crocodile answer, the crocodile grabbed the elephant’s nose. Retreating, with nose held by the crocodilian, the elephant found his nose stretched, much in the fashion of a rubber band.

Now, this is an explanation for why one particular elephant, previously possessed of quite a short nose, now had quite a lengthy one. Now, true, as it stands, it’s pretty difficult to see how one might falsify this explanation – it makes no particular predictions that I can detect – so it seems one would be justified in denying that it’s a scientific explanation for why all elephants, today, have long trunks. At the very least, to try to elevate it to this level, you would have to add something in about the transmission of traits from an ancestral elephant to descendants, and that sort of thing.

Now, it is, to be sure, a truly and absolutely horrible explanation. (A horrible explanation if one’s interest is scientific. For entertainment value, I know of none better.)  The explanation is obviously and palpably wrong, and horrendously so. For the explanation to be right, many facts would have to be true that we know are false, not the least of which is the elasticity of your basic proboscis. Then there is the matter of the details of descent; the story makes an implicit commitment to a pretty strong and peculiar version of Lamarckianism.

From this, we see that Kipling’s crocodile-stretched-elastic-nose explanation for the length of the elephant trunk is both a Just So Story – you can tell because it appeared in a book entitled, Just So Stories – as well as an explanation (that is neither scientific nor correct).

Here’s another explanation for why elephants have long trunks. Once upon a time, there were elephant ancestors with short trunks. Then, as a result of random mutations, some elephants had longer trunks than others, and because longer trunks were more effective in circulating blood than short ones, elephants with longer trunks enjoyed greater reproductive success as a result of the process of evolution by natural selection.

Now, that story doesn’t have Kipling’s ineffable style, but it’s also a story as well as an explanation. It is, like The Elephant’s Child, a very bad explanation, but as a scientific matter, perhaps less bad than Kipling’s. For one thing, the story doesn’t require a lot of things we know to be false to be true. That’s, you know, a pretty big plus for an explanation. Also, this explanation makes predictions, namely the properties of the trunk. Claims of function – circulating blood, in this case – are strong claims, requiring evidence, specifically evidence that the trait in question has properties that contribute to the putative function. These claims should both be consistent with existing data and point to avenues of new exploration. Once you have the blood-moving function in mind, you’d have to go check to see if the trunk has a chamber that contracts in order to move liquids around, a mechanism such as valves to prevent back-flows, and that sort of thing. A small amount of this enterprise would probably be sufficient to cause one to rapidly lose confidence in the trunk-as-pump explanation. The crucial piece is, of course, that functions explain biological traits because of the way that natural selection works.

Why am I writing about this just now? Because Denis Alexander, who introduces himself in this Guardian article as “an evolutionary biologist,” criticized Jesse Bering’s piece (also in the Guardian) about his book, cautioning the audience in his title to “beware evolutionary ‘just-so’ stories.” Now, don’t misunderstand me. To the extent I understand Bering’s argument – I haven’t read the book – I disagree with one part of it for reasons I’ve written about before. Briefly, he thinks that (false) beliefs about supernatural punishment reduce selfish behavior which improves one’s reputation. My worry is that individuals who inhibit selfishness because of incorrect beliefs are worse off than those who don’t – one’s best bet is to compute the expected value of any candidate (selfish) act, taking into account the probability of detection, benefit of the act, and cost of being seen –  so false-belief-generating systems will be punished for lost opportunities.

But that’s all beside the point. Is anyone else just really tired of this accusation of just-so story-telling? Alexander writes, “the psychologist Jesse Bering makes up a wonderful just-so story about “selfish behaviours” being “punished by supernatural agents” thereby promoting “prosocial reputations.” Well, who knows, there just isn’t any evidence either way.”

Yet again a hypothesis about function is confused with one about history – Alexander says that “we have virtually no firm knowledge of the details of religious beliefs prior to the invention of writing about 5,000 years ago.” But of course Bering’s view, because it commits to a function, commits to design features. If Bering is right, for instance, then there ought to be lawful relationships between supernatural beliefs and the probability of doing reputation-damaging, selfish things. I’m looking forward to reading the book – available next month from Amazon (U.S.) – to see the evidence he marshals.

The rest of Alexander’s article is not, in my opinion, worth the read, but there are times when, for some reason, I do sort of want to believe that there might be some supernatural punishment in the afterlife…

07. January 2011 by kurzbanepblog
Categories: Blog | 4 comments

Comments (4)

  1. Someone got a nook for Christmas…

  2. Robert,
    you write about the improbability of evolving a mind generating incorrect beliefs, as you write:
    “…so false-belief-generating systems will be punished for lost
    but isnt the existence of the vast amount of religious beliefs – and especially beliefs of supernatural beings – a proof of the existence of such a mind by natural selection? But we could also see this not just as a deficit of the mind, but rather as a by-product of an otherwise enough reliable epistemic processes (our information filtering ) of the mind which create a susceptibility to certian kind of misinformation (e.g. comnipotent supernatural entities), which is an issue also of your article about the gullibility of people.

    I would like to put your attention to a paper which deals with these issues in a more detailed way:

    Dan Sperber. 2009. Culturally transmitted misbeliefs (Commentary on Ryan T. McKay & Daniel C. Dennett, “The evolution of misbelief”). Behavioral and Brain Sciences 32, 534-535.

    available here:

    • Thanks for the note (and I hope you don’t mind my slight edit). My claim is not that one can’t explain false beliefs within a functional framework. My claim is about this particular function, inhibiting selfishness. Sperber’s byproduct claim is one possibility. The McKay and Dennett piece discusses other possibilities. I’ve considered a different function (having to do with commitment.) My claim in the part you quote was a narrow claim about the specific function at issue here.

Skip to toolbar