http://mashable.com/2016/06/02/elon-musk-simulated-reality/#sdLXHm2_jsqB
2,657 thoughts on “Elon Musk Thinks Evolution is Bullshit.”
Leave a Reply
You must be logged in to post a comment.
The Skeptical Zone
"I beseech you, in the bowels of Christ, think it possible that you may be mistaken."
keiths,
I sort of agree with this, but I would say the use of WHY here is more like short for “why should it be that…” , in other words if we know that organisms evolve simply as ways that make them better at surviving, there is no reason to assume that this would mean evolving more intelligence across the board, as Glen suggests is what has happened to all the different clades. Its more like asking, what sense does that make.
Simply put we have no reason to believe that intelligence is a better means for surviving than a whole host of alternative means. It would be like if every clade of animal over time developed large feet. What is so special about large feet.
Whereas from a religious perspective it makes perfect sense that animals developed intelligence, because that is the only way to have a divine intuition of the gods.
BTW, my sport is not lacrosse. But if I had happened to grow up in Maryland or Virginia, who knows.
It took me a while to settle this. But I’ve come to the conclusion that “information” should mean Shannon information. That is, a sequence of symbols used as a message.
ETA — deleted, posted too soon. See next post.
As I understand him, Shallit thinks weather parameters like wind direction constitute Shannon information that does not depend on a person being involved, eg in measuring them. I’m thinking of posts like these (also see the comments he writes in them):
Meyers bogus information
Reply to Paul Nelson
As far as I can see, he never says in so many words that weather provides Shannon info, but I take that as implied by his comments that Shannon info and Kolmogorov complexity are the only concepts of information formally defined satisfactorily and then by his calculation of the number of bits in some weather parameters.
(But I’ve never seen him mention probability in the same context, so that is one reason to think I may be misreading him).
Bruce,
Right. The defining characteristic of Shannon information is the reduction of uncertainty at the receiver. Whether the sender is an intelligent being (or an artifact) is irrelevant.
The signals we receive from reality reduce our uncertainty about reality. The dapples of sunlight on my blinds indicate that it is not an overcast day out there.
phoodoo,
The arguments are similar for both, but we started out talking about perception, not intelligence. Let’s stick to perception for now.
Can you see the advantage of a perceptual system capable of distinguishing nutritious foods from toxic foods?
Interesting that God granted us an intelligence sufficient to converge on a worldwide consensus regarding physics, but not religion.
I gather that you’d prefer not to say what it is?
I’m not quite sure of your point. However, I don’t seem to have any disagreement with those Shallit posts.
Neil,
The point is that there’s Shannon information in them thar signals, so your earlier statement doesn’t make sense:
Keith found the quote I was thinking of (with his magic google skills, I imagine). I read Shallit as saying there is info without us putting it there (but no measurements without us making them).
I think it’s kind of funny that this ancient realism v. idealism question is here put in terms of Shannon Information as if that clarifies rather than confuses matters.
FWIW, I found the Shallit posts Bruce linked to profoundly unhelpful. They seemed like good examples of what happens when those in other fields and little background in philosophy leap into philosophical waters without thought-preservers.
Shallit quote:
Shallit mentions “wind speed” and “wind direction” as information.
Shallit does not assert that the wind itself is information. The natural signal is the wind itself.
If Shallit had said that the wind itself was information, I would have disagreed.
I was confused by Neil’s saying there was signals in the world but information was not there unless we put it there. Quoting Shallit for me was making the point that if you were going to say signals were in the world, there was no reason not to say information was as well.
Now I understand that Neil believes there are only an unnamed, uncountably infinite number of things in the world without us (or something like that), so in that sense I guess both signals and information need us to exist under those names.
It is the implied asymmetry of the quoted text on signals that bothered me.
Or maybe that bother is just a carryover of my annoyance at Neil’s earlier dismissal of Putnam’s MTA argument as “absurd”.
I don’t think Shallit was attempting any philosophy nor do I think it is needed for his goal of commenting on the correct technical usage of “information” in mathematics and biology.
I was using his quotes in a way that Shallit was not, at least not directly.
That was precisely the Shallit quote I found confusing and unhelpful.
Professor Shallit has posted here on the odd occasion. I’m sure he could be persuaded to chip in if someone gave him a headsup.
So the wind is there in the world without us as a signal but it does not have properties like speed or direction (the information) until we put them there?
Right, but then there’s still going be the further question whether the perceptual and conceptual abilities necessary for distinguishing edible from poisonous foods really have all that much in common with the perceptual and conceptual abilities necessary for astrophysics, quantum mechanics, or evolutionary theory.
If the answer to that question somehow turns out to be “no,” then we wouldn’t be justified in thinking that we have the ability to construct increasingly accurate models of the universe at multiple scales of spatio-temporal resolution just because we inherited from our hominid ancestors the ability to recognize unripe fruit or poisonous snakes.
There’s a deep philosophical problem here that can’t be evaded: how much continuity or discontinuity is there between (1) our basic mammalian/primate cognitive capacities that allow us to perceive affordances in the ecological niches in which we evolved and (2) our historically contingent and culturally constructed ability to use technology and mathematics to construct, test, and improve models of the structure of the universe?
If there’s not much continuity between (1) and (2), then my general thesis about animal cognition certainly won’t help anyone who wants to defend scientific realism.
Apologies to Neil. I haven’t been reading your (sometimes Laconian) comments closely enough. Now you are now making perfect sense.
Knowing some of his posts here, I could guess his attitude towards considering the philosophical implications of his post as I have (mis?)used it.
But there is!
Does scientific realism need defending?
I tend to agree with you there. I don’t think it was helpful in his disagreement with Meyer. But I think it gets to the core of where people are arguing about whether signals are information.
keiths, to phoodoo:
KN:
Sure, but first things first. I want to establish whether phoodoo understands the value of accurate perception at a very basic level — distinguishing things that are good to eat versus from those that aren’t — before tackling some of the more abstract applications of perception and intelligence.
That depends on whether you are a realist or a nominalist about properties.
I tend to take properties as real. But there are uncountably many of them, of which we single out a few to be identified and named.
But I think you have completely missed the point.
Sure, you can say that speed and direction are in the wind (seen as a signal). But they are there implicitly. Our important uses of information (logic, computation, communication) require that the information be explicit. So having it there implicitly is of little value. We have to make it explicit before we can count it as information. And making it explicit is the kind of thing that cognitive systems do.
To see where Neil’s idea goes badly astray, think back to our infamous discussion of heliocentrism vs geocentrism.
Neil made the claim that heliocentrism was no truer than geocentrism, because both are merely ideas that we impose on reality. He also argued that the scientific consensus in favor of heliocentrism amounts to nothing more than a pragmatic preference for one coordinate system over another.
That’s utterly false, of course. The choice of model isn’t constrained by the choice of coordinate system. Regardless of which coordinate system you choose, the heliocentric model comes out the winner.
Reality favors heliocentrism over geocentrism. The choice is imposed on us, rather than vice-versa. It’s due to information that we gain from the signals we receive.
Neil’s position is untenable.
Putnam’s argument depends on looking at a natural language as a syntactic structure, with meanings seen as an interpretation of that syntactic structure.
I look at a natural language as primarily a semantic structure, with syntax as an addon — something like a protocol — to aid communication and to disambiguate. Looked at this way, the model theoretic argument seems a complete misfit.
I’m confused about who this post is addressed to.
I was also confused about “Laconian”, but after a bit of googling, it turns out that perhaps I have been speaking in Laconian without knowing it. If it was me you meant and not Neil.
(Yes, I had heard of “laconic”, but not its etymology).
That’s more than your Laconian “absurd”, so thanks for your empathy.
Bruce, to Alan:
Better to be Laconian than Lacanian, that’s for sure.
That’s not what Derrida said.
Too bad.
Glen Davidson
I think you have made an enormous, unnecessary leap here. What most would call the dumbest of the dumb creatures can decide what they should eat and what they shouldn’t based on using their basics senses. Plus, why is it that most animals can drink just about any water you put in front of them, and yet humans need everything boiled or sterilised for them. Animals pretty much get used to eating what they eat with barely any perception at all.
In fact in that respect, of all the animals on the planet, humans are probably the absolute worst at knowing which plants they can eat and which they can’t. So we got more intelligent and along the way it made us need a book, or a giant corporation to tell us what is safe and not safe to eat.
http://www.geoengineeringwatch.org/wp-content/uploads/2015/06/dead-sea-lions-california-2yroebigz8gapac6g799my-450×209.jpeg
keiths, to KN:
phoodoo:
Hence my question. Do you understand the value of that very basic perceptual ability?
Fair enough, but Putnam’s argument was directed against metaphysical realists who he thought must hold a different view of reference, truth, and knowledge than you do.
I think you have to consider what an argument is directed against to evaluate the merits of that argument.
Ignoring issues of realism versus anti-realism:
As best I can tell, you think information must involve us whereas Shallit thinks information, in particular Shannon information, is out there in the world regardless of us.
Further, Shallit wants to claim this to counter arguments from Meyer and his supporters that information requires a mind.
SEP has some interesting comments. As I read the article, it agrees Shannon information is out there in the world. But it also says that when it comes to biological explanation there is a philosophical controversy about whether a richer, semantic type of information is needed.
Although the article does not get into this issue, once semantics is brought into the picture, we have the issue of derived versus original intentionality. If one accepts this distinction, it would support in general your view that we need to be involved, at least for that type of biological information.
keiths,
Do YOU understand that there is nothing particularly intelligent about being able to detect what food is safe to eat and which is poisonous?
Why do you think it is that it only seems to be humans that don’t have this ability without being told?
I have no idea what to do with a comment this obviously stuoid, however sincerely intended.
Ah, well, thanks for a brief chuckle there.
Kantian Naturalist,
It’s not so much stupid as blatantly goal-oriented, IMO, like nearly every post by the religious faction(s) here. For phoodoo, only human beings can be intelligent, because only they are made in God’s image. Other creatures must be provided for in other ways. Thus, they can ‘tell’ when water is poisonous without the aid of intelligence, while humans lack THAT facility.
Kantian Naturalist,
Really? So if I put a big bowl of poisonous mushrooms in front of you, you would know not to eat them because they were poisonous huh? You can do this with many plants can you?
What great nose you must have, you really are smart KN. I bet you could give truffle pigs a real run for their money.
You must be a Platina-ist.
Shallit’s post was vague enough, that I can’t be sure of his view.
To me, information is abstract. A physical signal might be a physical representation of the abstract information. But it is not itself the information.
Shallit is a supporter of AI/computationalism. And it is common for such folk to hold that physical signals are information.
All philosophical arguments are written by aliens from outer space, and addressed to other aliens from outer space.
Or, at least, that’s the way that it often seems from this side of “The Two Cultures”.
What sticks out like a sore thumb, is that philosophers use logic in a way that I would never think of using it, and in a way that I would not expect to get useful results. Putnam’s argument is just an extreme example of this.
phoodoo,
You’re avoiding my question.
Again, you wrote:
And I responded:
If we can agree on that, then we can move on to discussing more complicated examples of perception, and then on to intelligence.
Neil,
How can a physical signal be a representation of information if the information is not “out there” in the world?
Remember, you claimed:
But as the geocentrism/heliocentrism debate shows, there is information in the world, and that information is what enables us to favor heliocentrism, rejecting geocentrism as crackpottery.
keiths,
What I don’t understand is why you are calling it a basic perception when humans don’t even have it (Well, except special individuals with a heightened pig gene, like KN of course).
phoodoo,
Humans can’t see in the ultraviolet, either. Does that mean that bees’ ability to do so, which helps them to locate nectar sources, is not a basic perceptual ability, in your opinion?
I’m not seeing a problem.
Mathematical objects, such as numbers, are not out there in the world. Yet we can have physical representations of them.
Which raises the question whether that which is being represented exists. If it’s not out there in the world, then in what way, for a physicalist, does it exist? Or what are the considerations that make it worth to be represented?
Neil,
To reuse my example from yesterday, the dapples of sunlight on my blinds tell me that the sky is not overcast. They deliver that information to me.
I am not first deciding that the sky is overcast, then observing the dapples and imposing that information on them.
The information comes from “out there”, not from “in here”.
No, they don’t.
Your highly trained perception, making use of the evidence of the sunlight, tells you that.
Neil,
Before I look, I don’t know whether the sky is overcast — that is, I’m uncertain about it. Then I look at the blinds and see dapples of sunlight on them, allowing me to conclude that it is not overcast. My uncertainty is reduced.
Reduction of uncertainty is the defining characteristic of Shannon information.
As I said:
Or my personal favorite, the mantis shrimp.
Whoa. Sixteen cone types??