While many ID proposals are based on introducing teleonomy into evolution, I wanted to ask the question as to whether or not evolution, even by a Darwinian definition (i.e., natural selection and materialism) was already teleonomic.
The reason I ask this is because all sorts of things that Darwinian evolution has trouble explaining gets thrown into the basket of “sexual selection”. Basically, the reason why an organism evolved feature X was because that feature was selected by mating. In other words, the other organisms appreciated feature X, and therefore copulated and reproduced more with organisms showing more and more of feature X.
I find this interesting, because, especially if taken materialistically, this gives a teleonomic direction to selection, something that Mayr attempted to rule out.
Think of it this way. If one is a materialist, then what is determining the desires of the organism? It is the organism’s genetics! If the organism is desiring a mate, that’s because its genetics is telling it to do so. If an organism sees mates with feature X as being more desirable, that means its genetics are telling it to do so. Therefore, the organism’s genes are, in a very direct way, directing the selection process themselves.
Mate selection, under materialism, seems to me to definitely fall under the umbrella of teleonomy. And, since it governs a large component of the evolutionary process, it seems that one must then say that to a large extent the evolutionary process is teleonomic, even under Darwinian terms.
I’m curious to your thoughts on this. I am not aware of this idea being discussed in the literature, but if someone has papers or links to other discussions of this, I would love to see them.
Point of fact: the term “teleology” was coined in 1728 by a German philosopher Christian von Wolff. He did so by building on Leibniz, who I myself find severely under-appreciated in contemporary debates about teleology in science. Although Leibniz’s mature metaphysics is, shall we say, bizarre, his reasoning behind it contains very deep criticisms. Leibniz argues at length that merely “mechanistic explanations” — meaning, explanations couched in terms of 17th-century calculus applied to empirical measurements — are insufficient for understanding life and mind. Wolff and then Kant, Hegel, Coleridge, and many others follow Leibniz in this.
The alternative tradition within modernity arises from Spinoza, who argued at length (see the Appendix to Ethics I) that our belief in teleology arises from our ignorance of efficient causes. But on this basis Spinoza also denies free will and the immortality of the soul. It was widely believed at the time (as it still is in some parts of the world) that public morality and civil order required that we believe that we had free will and that we would face judgment after death. Hence Spinozism was regarded as a threat to the public good, and Leibniz responds to this threat by reviving teleological explanations.
The point of this history lesson is that the contrast between teleological and mechanistic explanations has two significant dimensions: (1) that it relies on an outdated and superseded conception of “mechanistic explanations” and (2) that it has cultural and political importance only on the basis of other assumptions that are highly suspect.
phoodoo,
Same answer. But genotypes that don’t exist have no evolutionary significance, so I don’t know why you’re gnawing on that particular bone. Do dice that don’t exist yet have an expected value for their mean? They are drawn from the space of all possible dice, so they do in the same sense.
Any genotype has an ‘expected value’ (qv) for its replication. For most genotypes in the space of all possible genotypes, it will be zero, because it can’t even exist. Once it does, you have a mean probability of producing a particular number of offspring in a given environental context. That does not mean you WILL produce that number, even as an average in that run – selection is not deterministic, and populations are necessarily finite.
Alan Fox,
Right Alan. And thus by your bizarro definitions of fitness, ANY new genotype which comes into existence has ZERO expected value, thus has ZERO fitness.
Because Allan has just stated that the definition of fitness is the expected value of a type. There is no expected value for a type that has arisen yet. So every novel feature had zero fitness, and yet survived.
Unless of course you want to continue along the lines of, well, it survived, thus has fitness, which is of course how you have to retro-fit all of your definitions for them to mean anything.
Allan Miller,
The only method you have for determining that expected value Allan, is by looking at past results Allan. Thus all you are saying is, those that survived before will probably survive again. Those that survived abundantly in the past will probably survive abundantly in the future.
Thus you are not saying anything about true fitness, other than it got lucky before, let’s see if it does again. You are back to the same old problem that you always deny. Its an after the fact circular logic.
phoodoo,
And said that this ‘expected value’ exists before it has reproduced at all, yes …
Christ on a bike.
Genotypes that do not yet exist can still have a notional ‘expected value’ for offspring number. It would be a bit academic, but one can say that if that genotype cannot exist, its expected value can only be zero.
If the genotype can exist (the only kind of interest to evolution), it still might be zero.
If it can exist and reproduce, its value clearly is not zero, even when it has yet to reproduce.
How do we get from that to “every novel feature had zero fitness, and yet survived.”? A type doesn’t start off with zero fitness. It’s not the mean of offspring it has produced any more than the mean for a weighted die is the average of throws you have made.
phoodoo,
It can only be measured after the fact, but can have a value before it. Say I have just shaved a corner off a fair die. Do you deny that it has an expected value for the mean (which is no longer 3.5) before I have thrown it? Do you know what ‘expected’ means?
phoodoo,
Why does that matter? Suppose a collection of gas accreted into a star. I had no means of determining what the mass of that star would be in advance. I could only check ‘after the fact’ Have I thereby disproven gravity, and relativity to boot?
I’ve tried this before: One can take two bacteria A and B of completely unknown provenance and growth rate. Maybe Venter made them in his lab. Put them together in a medium, and you find that A consistently grows faster than B – a ‘circular after-the-fact’ determination. If it’s only determined after the fact, how come A grew faster than B?
“Expected value” is a theoretical term. What you can practically determine from past results is the mean value. The theoretical “expected value” is what the mean should be in completely idealized circumstances. The actual observed mean will differ from the expected value.
Neil Rickert,
Pay attention Neil. Try to keep up if you can. If you can’t understand it, read some books, then come back.
phoodoo,
What did Neil say that was incorrect? Have you read up on expected value yet?
He was fit.
Mung,
Not what I heard.
Not sure what any of this has to do with the OP, but …
The way I understand what’s been said is that every genotype that actually exists can be said to have an expected value but that value is not based upon the history of that genotype [the ev doesn’t refer to the previous generation] but rather the future of that genotype [the ev refers to the next generation].
Exactly, because in this context, “expected” = “predicted”.
Mung,
That’s pretty much it.
Kantian Naturalist,
Sort of – ‘predicted’ as in the mean we would get if we did an infinite number of independent trials.
But I can hear the rumble of a can of worms being opened up … ! All finite trials depart from the theoretical mean to some degree, and a succession of generations is a series of finite trials. The winner in any round, and in the series as a whole, is not always the fittest. It’s a tendency, not an absolute.
Allan Miller,
I have read your expected value, and it has nothing to do with the fitness of an organisms for its environment, despite you saying that it does.
But what is worse, is that you can’t seem to admit to the fact that in order for you to come up with any expected values, you first have to decide what survived before, then if it survived abundantly in the past you simply predict it to survive abundantly in the future. That has nothing to do with fitness.
AND because you can only come up with expected values based on previous results, EVERY novel supposed beneficial mutation becomes unfit by your definition, because it has no prior history to give it a high expected value. Thus by definition it defeats the theory of a fit new mutation. But nevermind, you say evolution happens anyway. The problem isn’t whether it happens or not, the problem is your definition of fit being the expected value negates the entire theory.
Every new mutation has low fitness, until it doesn’t. That’s stupid.
Mung,
Yes, it refers to the next generation, in the sense that, if that generation did well, then we predict it will do well the next generation. As simple as that. It is zero about fitness, it is about what survived survives, what didn’t must not be fit. It is not predictive, it is a simple assumption that history repeats itself. Until it doesn’t.
It tells you nothing at all about why genotypes spread. It tells you nothing at all about luck as a factor in survival. It tells you nothing at all about why if beauty is selected by nature, ugly organisms still thrive. It tells you nothing at all about why 13 percent of the population is gay, always has been, always will be. It tells you nothing at all about how a new genotype becomes prevalent, if it starts of as being less fitness (based on history) and then becomes more fit (based on history).
It simply says, ____ % have this type now, we predict, it will have that % later, unless it changes.
Meaningless. Inaccurate. and Illogical (since by defintion all new mutations have low fitness). Other than that, just more evolution stupidity.
How did a peacock get its tail? It has nothing to do with their silly definition of fitness.
Looks like not even the evodice thread can help poor Phoodoo.
phoodoo,
I see that you are still very confused.
Neil Rickert,
How long did it take you to compose that post Neil? Did you have to reference any source material for guidance?
Do you feel even a little bit shameful for using a public space for such empty drivel of zero intellectual value? Is your expected intellectual value in the future now zero?
Neil Rickert,
The difference between wanting to know and wanting to be right….
I did extensive research.
That is to say, I read your post (to which I was replying), and saw that it was nonsense. People, particularly Allan, have been explaining this to you, so you should have been able to do better.
Neil Rickert,
Your Expected Intellectual Value is still falling rapidly Neil. Can an EIV fall below zero? Apparently it can.
Rather than mock, why not try and understand? Expected value is used across multiple disciplines.
Richardthughes,
Yes, I agree with you Richard, why doesn’t Neil just try to understand?
I suppose its the difference between wanting to know and wanting to be right? Or maybe he is just incapable, but he still wants to type? Sort of like you?
Who knows?
‘Expected value’ has been around long before Neil, or you, or I. You’re the one with his eyes shut, Phoodoo.
phoodoo,
Yes, it’s stupid, but then it is not what is being said. That you continue to say that it is what is being said, despite repeated and patient correction … well, let the reader decide on your comprehension skills vs mine (and Mung’s for that matter) of explication.
A new mutation has an expected value for fitness. That is the mean of offspring of carriers that would be expected on a very large number (really, an infinity) of trials. This says nothing about whether that expected value is low, high or somewhere in the middle. And of course if you have not even reproduced the genotype once, you don’t know what the value is. This does not mean it does not even have a value.
So why do you persist in this mischaracterisation, that my words somehow mutate into ‘every mutation has low fitness (until it doesn’t)’? Do you get extra credits at the Pearly Gates for not getting it?
Does a brand new not-quite-fair die have an expected value for the mean of infinite rolls? Or is this something that only appears when you have rolled it an infinite number of times? What’s the fundamental difference between an unrolled die and an unreproduced genotype, probabilistically speaking?
phoodoo,
Yes it does. Genotypes with higher mean fitnessess (hypothetical mean values from infinite trials) are more likely to outcompete those with lower in a real competition (actual mean value in a finite trial).
You seem to be saying that a type producing more offspring (on an assumed, unmeasured infinite average) will not outcompete that with a lower on any trial, irrespective of that differential. That’s just bollocks. You persist in this fantasy that a differential can’t exist until you measure it.
Bollocks again. It is the mean of a probability distribution, so ‘Luck’ is deeply embedded in it. Smaller populations lose more alleles to ‘luck’ than larger for example. This has been well investigated and mathematically characterised.
On some trials in real populations, the fitter type will not succeed. It must be so, else it would not be probabilistic. You can’t get a distribution at all without visiting the regions of relative ‘unlikelihood’ from time to time. Expected value is not the result of one finite trial. It is the notional result of infinite trials.
Wha?
Wha?
[eta – of course it is not a complete causal account of the reason for the differential in every conceivable case. It’s not meant to be]
The distinction between teleology and teleonomy seems clear enough. Teleology affirms a purpose behind processes in terms of intent or wilful consciousness. Teleonomy admits that processes seem to lead up to or end up somewhere, but it’s only seeming or apparent. Teleonomy does not admit any real conscious intent involved.
It should be uncomfortable when somebody denies consciousness, awareness, will, things like that. Philosophically it’s reductive and unscientific, and in real-life terms it’s unrealistic, plain idiotic. That’s why it should be uncomfortable.
Of course (y)our particular definition of causation depends on the model you (we) are employing and also on the aims of what we are trying to achieve with the modelling. This does not change whether causation is and what it is.
Causation is multi-faceted. For example Aristotle identified four types of causes. The kind of causation you find empirically depends on what kind of process you are studying and what kind of model you are employing for it. Should be blindingly obvious.
As to the opening post, there’s no reason to deny teleonomy. It would be blatantly unscientific (unrealistic and unempirical) to deny it. The problem with teleonomy is that it’s insufficient to explain reality properly. You need concepts like intent and will, i.e. full-blown teleology, for the complete explanation.
Are you saying that it would also be unscientific to deny teleology?
And is teleonomy (and/or teleology) empirical? Can you back that up if that’s what you meant?
Of course.
Without teleology, a concept such as “empirical” does not even begin to make sense. In order for any concept to make sense, you need to distinguish it from something else. In order to distinguish one thing from another, seeing or feeling is not sufficient. You need to either comprehend the difference with intellect or have a purpose to distinguish between them. This is particularly so with concepts. You don’t see “empirical” – it’s a mental concept, but you cannot deny its purposefulness and meaningfulness (and thus necessary existence), because otherwise you would be talking self-refuting nonsense.
There’s this example in the opening post, “If the organism is desiring a mate, that’s because its genetics is telling it to do so. If an organism sees mates with feature X as being more desirable, that means its genetics are telling it to do so. Therefore, the organism’s genes are, in a very direct way, directing the selection process themselves.”
This seems to give a clear loop of directed or oriented behaviour, but it’s insufficient. The “organism” may desire a mate, but the desired mate may belong to another “organism” who has an intimidating character. Thus the first “organism” must postpone its desires. But it has been claimed that we are genetically hardwired to mating, so it should be impossible to postpone the desires. Yet we observe in real life that it’s possible to postpone the desires. Thus genetics is not everything. We are not causally hardwired to our genes. We are conscious beings who can figure out the balance and meaning of things. We can prioritize things and genetically wired desires often have a rather low priority for good reasons. And that’s reasons, not genes.
So it’s all about assigning a purpose after the fact, in hindsight?
What does that add to the explanation? Nothing if you ask me.
The same could be said of anything one observes, is the purpose of rain to fall from the clouds?
dazz,
I suspect we are into causes here. Aristotle and Aquinas, perhaps.
Oh my. I just read the last paragraph…
So empirical and scientific, no doubt about that :/
Talk about bollocks! There is zero way to measure the role of luck in populations in the wild. Zero! That is the entire problem with your whole concept. You have no way of knowing why some individuals survived better than others. You have no way of knowing why one mate was chosen over another. You have no way of knowing why one animal was eaten and another wasn’t.
Because EVERYTHING your theory of population does is measured after the fact, and then retro-explained. Whatever survived best is the most fit. Survival and fitness mean the same dam thing. It is not a prediction of what is fit, it is an analyst of what survived, and thus labeled fit.
Why is 13 percent of the population gay Allan? Well, according to the great materialist geneticists it must be because being gay at a rate of 13 percent is helpful to the population at large. We already have the figures, now we just need to make up the story about why. That is how every one of your just so stories is created. See what survives, call it fit, then make up a story about why. Why do some people go bald? Must be beneficial. Make up the story why. Why are some people fat? Well, it was beneficial in the Serengeti. Think of a reason.
Bollocks and bullshit!
Erik,
Not only that, but we also know that our thoughts can physically change our brains. So our brain, which is hardwired to be a certain way, can decide if its hardwired the way it prefers, and hardwire itself another way. Is it hardwired to decide to change?
You never managed to wrap your mind around the concept of teleology, did you?
Teleology is not just the end result of things. It’s also the intended result. Intention or motive is there before the process begins. The process points to and ends up somewhere that may or may not harmonize with the original intent or the master plan. Teleology involves all that, causality in the widest sense.
Cool story bro. So your “scientific” hypothesis is that there’s “intention” in life and that can’t be explained by genes alone. Waiting for the details, operational definitions and the evidence.
Why are you waiting, if there’s no such thing as intention? The only way you can sensibly wait is because you want to, i.e. waiting presupposes intention.
On my part, I am waiting for a sound denial or refutation of intention.
I never said there’s no such thing as intention. You said all this was scientific and empirical stuff, so I’m asking for operational definitions and a hypothesis with some level of detail that can be tested.
Do you… intend… to provide those or can we move on already?
Ah, so you were just playing games. I knew it, but it’s nice of you to admit it.
Thus we agree that there is intention. And this implies teleology, not mere teleonomy.
As to scientific and empirical stuff, provide your definitions of science and empiricism and we can move on. It’s already clear that it’s a game of definitions with you. I have no intention to talk past you.
Lurker here, I just created an account because I just cannot understand why so many people here think that teleo-* is a feature of living things.
As I see it, goals and purposes do not belong to things themselves (other than creatures that are capable of conceiving goals and purposes), not even to things that are genuinely designed, but to our conceptions of things. You could say that the purpose of a car is to move people around, or to release CO2 to the atmosphere. Either purpose changes nothing about the way the car is constituted and the way it operates. (Of course, we make cars with a distinct purpose, but the purpose resides in us. How on Earth could it be “transferred” to the cars themselves?)
I’ve seen KN bring up autopoiesis a few times. My thinking on this is pretty much my reading of Maturana and Varela. I wonder about his opinion on this bit:
http://topologicalmedialab.net/xinwei/classes/readings/Maturana/autopoesis_and_cognition.pdf
Relevant pp: 77-87
I agree there’s intention in our actions, and in actions of animals, there’s teleology in that, of course there is, but you are saying that somehow suggests genes are not enough to explain life if I got it right, correct me if I’m wrong. This thread is about teleonomy/teleology in evolution, so you are not really discussing the topic at hand.
So if your hypothesis is that teleological behavior of individuals suggests something else must exist, or whatever, I just want to know how does that develop into a testable hypothesis since you claimed it was “empirical” and scientific
Right. And the reason why genes are not enough is seen from the example in the opening post. It says, “If the organism is desiring a mate, that’s because its genetics is telling it to do so.” But then there are instances when the “organism” is not desiring a mate. Is that also because genetics is telling so? If yes, then genetics is telling contradictory things to the “organism” and you need some other thing than genetics to clear up the confusion. If no, then you have already silently presupposed another thing than genetics that is telling things to the “organism”.
Wrong. Instead, I claimed that it’s unscientific and unempirical to deny teleology. Because to deny teleology is to deny causality. Would you say that it makes sense to have a science that has no concept of cause and effect, that denies laws of nature? Is it a science worth the name?
Just an observation having no philosophical pedigree or credentials:
Intention and purpose are uniquely properties of evolved systems, systems that change their behavior over time as a result of feedback.
Now back to the serious discussion.
Sorry but I find the argument extremely naive.
First off, there’s no reason to believe that genes “tell” organisms to do something. Those decisions work at a different level even if they may be influenced by inherited factors, there’s no reason to believe “something else” is needed to explain living forms are capable of making decisions. You almost seem to imply there must be a gene for every potential choice or else, genes can’t explain such behavior. But the best part is the claim about contradictory decisions. Why would be contradictory to make different decisions in different situations?
What? Are you saying there’s intention and purpose in every cause-effect event?
Moving right along, intention and purpose are labels applied to dynamic systems that learn via feedback.
As such, a robot that learns has intention.
Life, considered as a dynamic system, learns via variation and selection, and has intention. Not only does it change, but it can improve its mode of change. evolvability can evolve.
If we start with this operational definition of intention and purpose, we can move on to discussions of how it works, and how different systems exhibit intention, and at what level of behavior.
There’s a rather important difference between systems — say plate tectonics or the hydrologic cycle, that cannot learn, and system that can learn. Until recently, only living things could be observed to learn. Now we have electronic devices that learn.