24
Feb
2019
This Is Setting Off My BS-Pattern Detector
This is a really smart guy; my son and I read (a lot of) his book Our Mathematical Universe. But around 14:15 when he starts asking questions, instead of saying, “Nothing,” I can think of a different one-word answer that’s far more accurate.
I think the most obviously wrong step he takes, is to say that equations describing how matter behaves, are the same thing as matter itself. This would obviously be a bogus move anywhere else in science, so why allow it when trying to explain the origin of our physical reality?
(E.g. if you’re holding a map of France, that’s not the same as holding France. If you write out Newton’s laws describing the path of a baseball, that’s not the same thing as a baseball in flight.)
You MUST feel strongly about this because you broke your rule about what you post about on Sunday.
BTW, great interview with Alex Tabarrok. I’m not usually that patient but I didn’t need to be. It held my interest throughout.
David wrote: “You MUST feel strongly about this because you broke your rule about what you post about on Sunday.”
Except that my answer to his questions is, “God.”
Oh thanks David for the comment about the interview.
Of course, “mathematical patterns” only exist in… consciousness!
I am not sure that “accurate” is the right word. 14:15 is the bit about life.
I presume you have no objection to the first one about the moon? There is no extra explanation required for a celestial object than a terrestrial object.
Is your one word answer that is more accurate “something?”
There is no logical requirement for there to be anything extra required for the life question. This is not something we can prove at present. But then again we can’t prove that everyone else is not a zombie. However Occam’s Razor allows us to prefer the explanation with the fewest assumptions, and “nothing” is the fewest we can have.
Even if we accept this it does not preclude the existence of a supreme being, just that perhaps he works through arranging the patterns. So I don’t see any need to suppose the existence of an extra parameter.
I would love if people stopped misusing Occam’s Razor.
Please read:
http://rc3.org/2011/05/23/on-the-misuse-of-occams-razor/
It does not help your argument here since you cannot test any hypothesis. Occam’s Razor does NOT tell you what is right, it is an efficient problem solving method. That is all.
I was quite careful. I said it allows us to prefer the explanation with the fewest assumptions. I was careful not to say it proves that hypotheses or anything like that.
Your linked article says “is a principle that generally recommends selecting the competing hypothesis that makes the fewest new assumptions, when the hypotheses are equal in other respects.”
This is exactly what I did – I selected the hypothesis with the fewest new assumptions. I expressed this recommended selection as a preference. Your article supports my claim.
You suppose that this only applies when you can test these competing hypotheses, but that is not the case. Without testing we cannot be certain, but it still allows us to have a preference.
As an example, the vase has been broken and little Jimmy was playing football in the hall. We have two competing hypotheses.
1) Johnny broke the vase.
or the alternative offered by Johnny.
2) A pixie flew in (or an intruder came in) and pushed the vase over.
We cannot test either, yet we prefer the former. That is because introducing the pixie (or the intruder) is an unnecessary assumption. It does not tell us Johnny did break the vase, but it allows us to select between competing hypotheses.
Sometimes we get new evidence that means our originally preferred hypothesis no longer explains the observations. The CCTV camera recorded an intruder, perhaps. We then need to revise our preference as the intruder is no longer an assumption we have unnecessarily introduced.
I too am irritated by misuse of Occam’s razor, but I don’t think I misused it here.
In my experience Occam’s Razor mostly is (mis)used in discussions that are very abstract with little to no existing fact basis. Like religious discussions or some events in the past with basically no public records, usually called conspiracy theories. And then for one side, sometimes even for both!, Occam’s Razor is called upon as prove that your side is the one who makes the least assumptions, or at least less outrageous assumptions. (Obviously also the quality of the assumptions needs to be compared not just the quantity, so you open a total can of worms there). It is used as a tool to win the argument by trying to show who is more reasonable. In the end it is a void discussion. Because obviously both sides think they are more reasonable…
But as soon as you discuss abstract and very often loaded issue like that, what actually is the side that has less (outrageous) assumptions? E.g. it is not clear that the Atheistic side makes less (outrageous) assumptions than the other side for me.
Also your analogy with Johnny doesn’t work, because first it is very concrete one with well-established facts (broken Vase, Johnny playing football, he was present etc..) And secondly you ask someone who is conflicted, so both hypotheses are not equal in other respects, because one explanation means Johnny avoids being guilty of causing damage and possibly being punished for it, while the other would likely cause troubles for little Johnny.
So in my view, Occam’s Razor should not be used in any highly abstract discussion with little or no fact basis at all, but only in very fact based and testable scenarios, where assumptions are a small amount compared to what is known, and where is little to no ambiguity in terms of judging the quality of different assumption so that you are actually able to really make an apple to apple comparison for different assumptions.
“. And then for one side, sometimes even for both!, Occam’s Razor is called upon as prove that your side is the one who makes the least assumptions, ”
Yes, this is irritating.
Your objection to the Johnny example is countered by a simple tweak. Instead of asking Johnny you ask your partner “What do you think happened?” They respond with the pixies argument. Except of course they don’t, because they have informally applied Occam’s Razor to eliminate hypotheses with unnecessary assumptions.
In the situation here, I think a reasonable counter to my use of Occam’s razor is that I was only considering a simple question. We have consciousness (assuming we can agree in this). Given *no other information* is it better to assume it arises out of existing stuff arranged in a particular way, or that an *extra* undefined something is required to produce consciousness? Occam’s Razor suggests the former. However, this can be countered with extra information, analogous to the CCTV footage, or perhaps strong evidence that Johhny had stopped playing football at the time the vase broke. Or by pointing out where a hidden assumption had been smuggled in, which is all to easy to do accidentally.
So someone may say “God is not an assumption because of evidence x, y and z”, which would lead to endless further discussion.
We always have incomplete information. As long as we understand the limitations Occam’s Razor is useful even in untestable and highly speculative situations.
Physicists posit a Rube Goldberg machine with a random explosion at one end and conscious, intelligent human beings at the other. Many refuse to consider that a supernatural Rube Goldberg might have devised the machine.
The only modest defense I can offer them is that they are assuming that our present reality is a necessity, because it exists, and trying to reverse engineer reality based on that assumption. In that light, “nothing” is not the final explanation for where this whole universe came from. It’s merely the list of explanations that they have come up with to date.
‘I think the most obviously wrong step he takes, is to say that equations describing how matter behaves, are the same thing as matter itself. This would obviously be a bogus move anywhere else in science, so why allow it when trying to explain the origin of our physical reality?’
What about the example he gives of a future virtual reality game where some of the characters are conscious ? In this game both “matter” and “consciousness of matter” (as experienced by characters in the game) are just features of the way the game is programmed (its “equations”). The physical implementation of the game could be on any bit of computer hardware capable of executing the appropriate code. So in this case it seems that not only matter but consciousness itself are , in the context o the game game, the same as the equations describing them.
I suspect that physicists such as the one in the posted video think the universe itself is (at some deep level) somewhat like this game and that its the equations describing the universe that “create” the matter of which we appear to be conscious and that therefor matter and the equations describing matter are one and the same.
(comment getting rejected so adding an additional line)
Bob, hope you don’t mind me linking these, but Landsburg has a couple of interesting posts about this which deal with Transformer’s points, although he does not entirely agree with Tegmark.
First one here
http://www.thebigquestions.com/2010/09/30/the-mathematical-universe/
Some interesting discussion in the comments here.
http://www.thebigquestions.com/2014/04/01/many-many-worlds/
The whole ’emergent properties’ thing bugs me. He talks of ‘wetness’ as if it isn’t inherent in a molecule of water. My ability to detect things is finite and I may need a lot of water. But the emergent properties seems more to me like categorization and how we attempt to simplify the world in modeling. There is stuff we leave out because we focus on what matters given the context.
So are there really emergent properties? or just methods of classification we use when dealing with molecules present in different orders of magnitude?
I care because people belief in emergent properties is one justification for the state. “I don’t have the ability to initiate force morally, but if we all agree then we get to re-write moral law”
Ok, maybe you didn’t like his wetness example but the idea of emergence has been around and accepted since ancient times. Life is an emergent property of chemistry. Salt is salty even though sodium or chlorine atoms are not. The metals that make up a computer can’t information process even though the computer can. Cells themselves can’t do the function of the organs they make up. If you don’t recognize the existence of emergent properties then you’ll constantly be making the fallacy of division. This is when you take the properties of the ‘whole’ and conclude its ‘parts’ have those properties as well since in your mind emergent properties of the whole don’t exist.
Do you have a time stamp of when he says equations that describe matter is the same as matter itself? I’m not arguing for or against that idea at the moment. It’s just I listened to the video twice, but I seemed to have missed any direct reference to that. I think the closest he gets to saying that is when he says consciousness is a mathematical pattern, and thus some mathematical patterns simply are conscious. Is that what you are referring to?
I don’t think he uses those words in this talk. I suspect it arises out of Bob having read a lot of his book, which presumably goes well beyond this 15 minute talk.
Compare this to Steve Landsburg’s claim in the first link I posted above.
“2. Every modern theory of physics says that our universe is a mathematical object, and that we are substructures of that object. Theories differ only with regard to which mathematical object we happen to be a part of. Particles, forces and energy are not just described by equations; they are the equations (together with abstract, purely mathematical relations among those equations).”
I think you have to read the whole thing for it to make any sort of sense.
We have made progress with computers that can play chess, but we have not made much progress in natural language processing (I’ve been involved in selling “sentiment analysis” systems that process bulk text and give information about whether the general sentiment is positive or negative … they are better than nothing, but they absolutely should not be oversold as being able to understand what people are talking about). We are at the very start of “computers that can drive cars” … barely scratched the surface IMHO.
The “arrangement question” has nothing to do with consciousness. Why is it difficult to live inside a river bank made of clay, but it’s quite easy to live inside a brick house (made of the same clay)? Why is a mountain of iron ore not a fork? Almost everything that humans do involves rearranging matter … so get that concept sorted out before even approaching concepts such as consciousness.
On the “complexity question” … how can something as complex as Windows 10 be explained by 1’s and 0’s with some handful of simplistic logical truth tables? Well, complex things are complex. I you want to be reasonable about it … the miracle is that anything can be simplified.
Once you accept that arrangement is important, you automatically admit that we don’t have all the necessary pieces to work with … because some of those pieces are themselves products of statistically unusual arrangements (e.g. if you give some student information about basic Carbon chemistry they are unlikely to come up with the formula for DNA). Chemistry is merely “arrangements” of physics, and biology is merely “arrangements” of chemistry, and sociology is merely “arrangements” of biology. I might glibly point out that the entire discipline of Engineering consists of taking what comes out of science and tweaking the arrangements. None of those disciplines are entirely obvious based on knowledge of the more fundamental discipline.
On “waves vs substrate” … the wave contains energy and most of physics depends on the concept that “energy is a thing” … indeed, some would argue that’s the very essence of physics. Computational substrate is different in as much as all computations destroy exergy (i.e. convert energy into an irreclaimable form, since energy cannot be destroyed). I agree with his basic concept of substrate independence but these substrates are fundamentally different. Here’s another example: photons are waves in the electromagnetic field … in theory you could have photons on any substrate with identical properties to the electromagnetic field, but photons are also a particle and therefore a physical “thing” (like a gamma ray can hit you and cause cancer), and at the same time they are information (like the microwaves out of a cell phone tower). The point of dualism is that sometimes there’s more than one way to look at the same entity … not the claim that there’s additional forces coming out of nowhere.
Modern physics does not suggest that the entire universe is perfectly described by physical laws. They are struggling with the invention of copious amounts of “dark matter” depending on an unobserved entity to explain observable phenomenon. There’s deeper unanswered questions about physics, e.g. https://en.wikipedia.org/wiki/Hierarchy_problem
We could imagine, perhaps in future physics will improve and describe more of the universe, and quite possible computing will also improve to make a better effort at achieving intelligent behaviour. We could imagine a lot of exciting developments. I’m not against his final conclusion about the importance of arrangement, but that hasn’t brought anyone in the audience closer to an understanding of consciousness.
“The point of dualism is that sometimes there’s more than one way to look at the same entity … not the claim that there’s additional forces coming out of nowhere.”
I think the dualism he is talking about is precisely different “forces” coming out of nowhere.
“The “arrangement question” has nothing to do with consciousness. ” I don’t understand what you mean here, probably a deficiency on my part. All your questions could be answered by “because of the arrangement of matter”. Why could not he question “why is this thing conscious and this thing not?” also be answered the same way?
“Once you accept that arrangement is important, you automatically admit that we don’t have all the necessary pieces to work with … because some of those pieces are themselves products of statistically unusual arrangements.” I don’t get this either. You would not arrive at DNA through random efforts, but guided efforts may lead you to something similar. A house is a statistically unusual arrangement yet we have no problem with people building houses. Statistics doesn’t really come into it, beyond the idea of entropy.
“None of those disciplines are entirely obvious based on knowledge of the more fundamental discipline.” This is very clear. We do not discuss someone moving their arm as a series of physical processes of atoms or particles. We understand that this is the underlying mechanism but we call the assembly of molecules “an arm”. We don’t need to track every molecule to talk usefully about the arm. Whilst it is possible to explain the mechanisms that operate the arm in terms of chemistry, and the interactions of atoms in broad terms in terms of physics, once we have the arm available to us, it is not easy to arrive at “an arm” based on physics”. Is this your point?
Suppose we find a physicist and offer them a hypothetical choice between two rooms. In one room, brand new hunting knives come raining down at you from 30m above in large numbers. In the other room brand new soup spoons come raining down in the same numbers from the same height and the spoons are made from the same weight of the same steel with the same kind of plastic handle.
What’s the difference? Same atoms, same mass, same kinetic energy, only a small rearrangement. Does this have anything to do with solving any questions of consciousness?
OK, you might slightly prune the pool of physicists if you find a few of them decide to choose the room of knives … thus leading to deeper theories about how Darwin might be involved in the gradual evolution of human intelligence. Let’s leave those indirect associations aside for the moment and ask whether understanding this simple problem about the arrangement of matter gives any direct insights into the nature of consciousness. I argue it does not.
” Does this have anything to do with solving any questions of consciousness?”
Only in the very broad sense that it demonstrates that rearranging the same molecules in different ways gives the collection of molecules different properties – in this case the property of “sharpness”. It is consistent with this finding that certain arrangements of matter may have the property “consciousness” whereas different but quite similar arrangements may lack this property.
“Self-Understanding”, activity of brain patterns moving this way and that, is more than just the patterns themselves.
Speaker giving incomplete case.
There is a philosophy of science called structuralism in which it is wrong to say that it is matter which exists, but the patterns describing it which exist. From Wikipedia:
Structuralism asserts that all aspects of reality are best understood in terms of empirical scientific constructs of entities and their relations, rather than in terms of concrete entities in themselves. For instance, the concept of matter should be interpreted not as an absolute property of nature in itself, but instead of how scientifically-grounded mathematical relations describe how the concept of matter interacts with other properties, whether that be in a broad sense such as the gravitational fields that mass produces or more empirically as how matter interacts with sense systems of the body to produce sensations such as weight.
Sean Carroll had philosopher James Ladyman on episode 33 of his podcast, and it was a very thought provoking interview.