[SEQ RERUN] Three Fallacies of Teleology
Today’s post, Three Fallacies of Teleology was originally published on 25 August 2008. A summary (taken from the LW wiki):
Theories of teleology have a few problems. First, theories of teleology often wind up drawing causal arrows from the future to the past. It also leads you to make predictions based on anthropomorphism. Finally, it opens you up to the Mind Projection Fallacy, assuming that the purpose of something is an inherent property of that thing, as opposed to a property of the agent or process that produced it.
Discuss the post here (rather than in the comments to the original post).
This post is part of the Rerunning the Sequences series, where we’ll be going through Eliezer Yudkowsky’s old posts in order so that people who are interested can (re-)read and discuss them. The previous post was Magical Categories, and you can use the sequence_reruns tag or rss feed to follow the rest of the series.
Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day’s sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.
This doesn’t require one to draw a causal arrow from the future to the past, at least no more so than does the chicken-egg problem. For Aristotle, the end for which organs develop is the mature life of the animal, its form. Animals inherit form from their male parents, so it’s a nice, straightforward past to present causal arrow. But that’s a little misleading, since for Aristotle there Is one and only one temporal arrangement of causal relationships: simultaneity. So no future to past arrows, and no past to future or present arrows either. This gets him in some serious trouble, and he recognizes the fact, when it comes to billianrd ball mechanics.
And to answer the bit at the end of the quoted section, Aristotle believed that the efficient cause of the animal is also its form, though by way of its male parent, in the way the art of statuemaking is the cause of a statue by way of a particular artisan. But Aristotle is pretty clear that the efficient and final cause of an animal, or organ, are extensionally identical. Which isn’t a great move, so far as providing explanations goes, but there you have it.
The water flows downhill not because it wants to reach the center of the Earth, but because water flows downhill. There is a telos there- water doesn’t flow uphill in all cases where flowing uphill results in flowing downhill later. Things in general do not plan, even when they have a simple purpose.
The purpose of teeth aren’t to chew in the future, but to chew what was available in the past. Mary isn’t driving to where the grocery store IS, she’s driving to where the grocery store was the last time she saw it or (more generally) where she thinks the grocery store is. Her belief in the grocery store is the immediate cause of her navigation. Backwards causation isn’t inherent in teleology: water flows downhill because it has the purpose “flow downhill”, not the goal “reach the bottom”.
The anthropomorphic flaw also isn’t in teleology, it’s in the execution of teleology; if you ascribe the correct purpose to every event. That becomes complicated when we ask what the purpose of life in general, or of the universe as a whole is: is it as complicated as driving to the grocery store, or as simple as running downhill? The bronze statue tarnishes, but it would be madness to claim that the purpose of the statue was to tarnish. All life which hasn’t died out has avoided dying out, and it would be similar madness to claim that is evidence that the purpose of life is not dying.
To say that motherly love is inheritable and was pro-survival is one thing; to claim that is has the specific external purpose of being pro-survival is quite different. Capture is also an error used by agents, not by the process.
Those fallacies are worth cautioning about. However, they aren’t fallacies of teleology but fallacies of most teleologists.
A different way of summarizing telos that may be helpful when discussing this topic is that an object or agent fulfilling it’s telos is supposed to be the most fully itself. So, for Aquinas and Aristotle, because the quality that most clearly sets off the category ‘human’ is “a reasoning animal” actions that interfere with that part of human identity are unnatural and interfere with telos.
This framing is perhaps a little easier to tie back to the last two rereads on Unnatural Categories and Magical Categories since then telos is linked to a partitioning of conceptspace.
One of the most interesting places I’ve seen “the x-iest x” outside of telos discussions is in Turing Tests, where one of the human controls wins a prize for “the most human human” (i.e. the human that was most frequently rated as human by the judges). There’s a book out by one of the contestants, who set out to win that prize. Apparently, one fairly successful strategy is being belligerent. Not the most encouraging result, but interesting.
When Charlton Heston was on the Planet of the Apes and he found that human beings were no longer differentiated by their reasoning powers (which were sub-par) but by their hairlessness, should he have devoted his life to keeping exceptionally well-shaved?
(this question brought to you by my continuing confusion with teleology)
I posted answers, so far as I have them, to your questions in the linked discussion.
Very many thanks!
I might end up saying that reasoning was still the ne plus ultra and that apes and humans had a lot of overlapping telos. Apes and humans might end up like men and women or like Einstein and normal people; there are other salient differences, but the ability to reason would still be the thing you’d pare off last. (People would be more likely to recognize a hairy human as human than one that didn’t claim to be conscious).
The reason for this is because of the etymology of the word “telos”. It has a “common ancestor” so to say with the word teleos which means perfection. The background for this is that we are dealing with words that had a particular meaning in pre-Greek language/culture. A culture that for some reason linked those words. Fast forwarding 3,000+ years that inference doesn’t make sense to modern readers.
Their flaw is in concluding that because a feature is distinctive about an object or agent, it is the most inherent thing about that agent. They never ask why reasoning is more human than emotion, or metabolism.
Yes they do. Nichomachean Ethics I.7, for example. EDIT: but this is also, in part, what the whole of the NE sets out to show: that the best life, and the end of all our activity is a life of the intellect, be it practical or theoretical. So if you’re looking for a short argument in the mode of establishing a premise, you’re right that there’s nothing very satisfying in Aristotle. But the question of what the highest human aim is is hardly a question he ignores. Its the whole subject of his ethical philosophy.
He asks the question of what the human purpose is. He never addresses why a given purpose is the chief purpose, only that some intermediate goals are intended to further other goals. Aristotle doesn’t ask why heavier objects want to fall faster than lighter objects, he simply observes that they do and makes the claim that they do so because they are supposed to.
To be honest, every time I try to read Aristotle, I end up drawing a triangular diagram and labeling the three points ‘agape’, ‘hatred’, and ‘apathy’, (in response to every time he uses the concept of “contraries”), and the fundamental theorem of calculus, if I’m reading the Physics. In general, though, I tend to underline phrases such as “it is evident that”, annotate “nope!” in the margins, and try to provoke a counterargument from the text. So far, my text has not been forthcoming.
I didn’t peruse the entire NE to see if he addresses every possible answer, because I know well enough that “it will be evident that” his pet reason will be the One True Answer, and he will not address in enough detail why that is evident.
I’m not sure I understand what you think the gap in his argument is (which is not to say it’s not gappy). In I.7 he says specifically that human beings have a variety of aims and activities, but that a life in accordance with reason is the chief aim because it’s the activity we pick out when we say someone is a good human being (as opposed to a good carpenter). So adresses, if inconclusively, the question ‘Why is this our chief aim?’ by pointing out that when we call a carpenter good, we mean that he’s good at the activity of carpentry...in other words, the thing we picked out when we called him a ‘carpenter’. In ethics, we’re concerned with being a good human, so what activity do we pick out when we call someone simply a ‘human’? Well, no one calls someone a good human being (in the ethical sense) for metabolizing or running well. We call them good for leading a life of reason and virtue. That’s why a life of virtue is our chief aim: it’s what we’re good at when we’re good at being human.
Also note that the whole run of I.7 has a hypothetical, provisional sort of character. He’s clear that he thinks this whole ‘chief aim of human life’ thing is a serious and difficult problem. I.7 isn’t a great argument, but he is at least taking the problem seriously.
I think a triangular diagram will be misleading in understanding Aristotle’s theory of contrariety. A better one might be a continuous line, with a positive end and a negative end. Every value on the line is always expressed as a predicate of some subject characteristic of the contrariety. So the range of color, has as its characteristic substratum ‘surface’, and every color value on the line is predicated of a surface.
‘It is evident that’ is often a translation of a more colloquial and less committal segway. Aristotle’s Greek diction, however horrible, is never as pompous as translations make it out to be.
Is this argument circular? Or at least a disguised appeal to intuition?
“When we think of what a good human is, we think of a rational human. Therefore, what sets humans off is being rational. Therefore, the final cause of humans to be rational. Therefore, a good human is a rational human.”
Couldn’t you skip straight to “If we think X is good, then X is good”?
Sorry, my summery left some important points in the dark. Aristotle is saying something like this:
When we say something is good, we say that it is a good X or Y, not good simpliciter. When we say someone is a good flute player, we don’t mean that they are good and a flute player, but that they are good qua flute player. By picking someone out as a good flute player, we are already deploying some idea of what it is that distinctively they do and what it is that they aim to do qua flute player.
Now, we’re presently asking ourselves what the good life is, and generally in ethics, we’re concerned with what it means to be a good human being. So, when we say ‘good human being’, what idea are we deploying about what human beings distinctively do and aim at qua human beings? Well, it looks like the thing human beings distinctively do is reason, and the thing they do well when we say they are good qua human being is live a life of virtue in accordance with reason.
And the rest of my book is supposed to explain and elaborate on that.
This is kind of what he’s saying. Aristotle isn’t playing the “lets do ethics from scratch” game. He says explicitly that he is counting on his and his student’s intuitions to get the whole thing going. Ethical philosophy, he says, is only possible for people who were raised well and generally have a pretty good idea about right and wrong and stuff. Aristotle’s ethics is not about making people decent. You need that just to get started. Aristotle’s ethics is about making people extraordinary. It’s about greatness.
Really? I’m reading Ed Feser right now, and he’s arguing very strongly that the only reason we don’t have objectively grounded ethics right now is that we’re not Aristotelians, and that the ability to objectively ground ethics is one of the biggest advantages of Aristotle over everyone else. He specifically attacks utilitarianism from an Aristotelian viewpoint for giving a moral criterion but not being able to prove from first principles that one should follow it.
Is he operating outside the Aristotelian mainstream?
No, I kind of agree with that, though I don’t think he knows what he’s getting himself into. An appeal to intuitions doesn’t take Aristotle out of the objective morality game, and the ‘function argument’ is pretty preliminary.
Aristotle’s ethics is monumentally, catastrophically evil. I think it’s also the perhaps the only real ethical theory we’ve ever come up with, which is a problem. Aristotle isn’t okay with slavery, he positively argues for it. He would say that if we didn’t have slavery, we should start, because slavery is a good thing. He thinks infanticide is a reasonable way to deal with population issues because children are ethically valueless. He is super evil.
Is the life of reason and virtue the end in itself? If so, why do some people choose to live lives that are not of reason and virtue- what is their end? Could it be that virtuous people and non-virtuous people share a common end, but differ in how they seek that end? What is the purpose of virtue? It can’t be turtles all the way down, but each turtle does need something to stand on.
Oh, I understand the theory he has. I want to shove that diagram in his face, say “Here is an example of three things, each of which is contrary to the other two, existing on a continuum. Your one-dimensional continuum model is proven to not always be the case- now every time you use that model to show a specific case, you need to show that it accurately models the specific case you are using it for.” I can forgive him for not being able to determine the limit as dT->0.
Perhaps my translation adds to my issues with the work; learning ancient Greek is out of the question for me; is there an English translation that you would recommend?
Yes. And right, people who choose to live lives of viciousness and ignorance have the same end, they just radically misunderstand how to get there. The purpose of virtue is good action, a life of which is the happy life. And happiness is an end in itself. No more turtles after that, at least that’s what NE X says. You might not like his answer (there’s a lot to object to), and you might object to my specific reading, but it’s not as if Aristotle just didn’t address it.
He raises that worry himself in Metaphysics Iota 5, with the triad ‘equal’ ‘less’ and ‘greater’. So if you shoved that in his face, he would probably admit that it’s a real puzzle for his view. His explanation in Iota 5 is confusing, but so far as I can tell he would say that apathy is a privation of love and hate alike, and therefore a kind of intermediate:
,
I think he would have been very interested in the development of calculus, but you’re right that this was not in his arsenal. I don’t think he really contradicts that though. Much of the argument mid-Physics is centered around the claim that continua cannot be composed of extensionless points. I think he would have accepted the idea that continua can be constructed from an infinity of infinitesimals, since these just have infinitely small extension, not no extension.
We would now contradict Aristotle about the construction of a continuum from extensionless points however, since we can construct the real numbers by Dedekind cuts. But its not clear to me Aristotle would even have found this an unwelcome modification.
Really, if you wanted to freak Aristotle out, you’d just show him that we’ve traveled past the moon. Or evolution. Or you could just point out that slavery is super, super evil, and not good like he thought.
I don’t think it’s worth being picky about translations. There’s only so much precision to be had in a 2300 year old text, transmitted by a 2000 year long game of telephone. Just remember that Greek temples and statues were painted in garish, silly colors, and Aristotle spoke with a lisp because he wanted to fit in with the Athenians. And that he probably married his slave after his wife died because he loved her, and that he wrote Nichomachean Ethics for his nephew, because he wanted him to become a good man, and that he probably thought he would never be as smart as Plato.
To which I can respond that love and hate can be combined in the same person (along the base of the triangle, at their most intense), while having neither would be at the top. Measuring it along one dimension fails to differentiate between people I never meet and people for whom I have strong and mixed feelings. I can generalize that to a tetrahedron, if there are four mutually contrary descriptions, and I think to an analogous geometric construction in n-1 dimensions for n independent mutually contrary things (Concave, each point is the same distance from all other points)
If they’re not mutually exclusive, then they’re not opposites.
I never said they were opposites- I said that they were contraries; opposite corners of a continuum.
The continuum between love and hate does not pass through apathy. There is also continuums between love and apathy and between hate and apathy; those two pairs are also pairs of contraries.
In all fairness, teleology can be prompted by mathematical models. For example, the principle of least action (also known as the Fermat’s principle in geometric optics) makes it look like a certain inanimate object has a purpose or even can predict the future: “light travels between two given points along the path of shortest time”. Of course, if you dig deeper, the next-level model says that all paths happen, but only the one with the least action happens to survive their mutual interference, negating any teleological ideas.