Monday, April 17, 2017

Uncommon Descent Lies Again

The blog Uncommon Descent is, of course, one of the two main propaganda arms of the intelligent design movement --- the other one being "Evolution News". Since the ID movement is essentially based on religious dogma and deception, it's no surprise that these blogs have a large amount of fake content. But I am always surprised at how shamelessly they mislead.

One of the recent entries at Uncommon Descent is a good example. They refer to a 1980 article of Hamming entitled "The Unreasonable Effectiveness of Mathematics". It's not hard for anyone to verify that what I just wrote is the correct title of Hamming's article; indeed, the full text of the article is easily available online.

I am not going to criticize Hamming's article in much detail here. There is much that is good in it, but I feel his final conclusion is unmerited. On what rigorous basis can we measure how effective mathematics is, and on what basis are we allowed to conclude that the effectiveness we observe is "unreasonable"? It seems purely a matter of personal taste.

My own personal taste is that mathematics is remarkably ineffective, because the vast majority of events that we see in the physical world are quite difficult to model accurately. If we release a single tritium atom in a lecture hall at 10:00 AM, where will it be at 11:00 AM? No physicist in the world can tell you with very much precision.

Similarly, Hamming asks, "How can it be that simple mathematics, being after all a product of the human mind, can be so remarkably useful in so many widely different situations?" Well, lots of mathematics is not remarkably useful. Much of what I personally do has little real-life application. So how can we measure, in a precise way, when that usefulness is "remarkable" and when it is not? Hamming does not tell us. My own personal view is that humans tend to use what is effective and discard what is not. If, for example, dancing were more effective in describing the physical world, scientists would be ballerinas.

In any event, Hamming's observations are not my main point. My main point is that, at Uncommon Descent the title of Hamming's article has been altered from "The Unreasonable Effectiveness of Mathematics" to "The Unreasonable Effectiveness of Mathematics vs. Evolution". Whether this change is a matter of deliberate deception or pure incompetence, I am not certain. But it is part of a larger pattern that we see repeated.


"Shecky Riemann" said...

With scientists attempting to maintain some level of rigor/accuracy/integrity, it’s hard to see how one even combats the deliberate lies, distortions, misinformation to which the other side is willing to stoop. And now we have a compulsive liar as President drawing a devoted fanbase. The near-term future, frankly, seems bleak indeed. :(

Neil Rickert said...

"Whether this change is a matter of deliberate deception or pure incompetence, I am not certain."

I'm pretty sure that was a post by O'Leary. So I'm inclined to go with incompetence.

Lee Witt said...

"Whether this change is a matter of deliberate deception or pure incompetence, I am not certain."

My first thought - consider the source, it's deliberate deception. But when I look at how blatant the error is

A friend draws attention to an old paper, “The Unreasonable Effectiveness of Mathematics vs. Evolution” (The American Mathematical Monthly Volume 87 Number 2 February 1980)

and then see an immediate link directly to a pdf version of the paper, it seems (to me) to be stronger: it is a deception given with the certainty that accompanies the knowledge that the people intended audience don't care that the thrust of the article has been misrepresented. Sad.

(And apparently they've back-linked this article to some problem(?) about mathematicians not yet knowing whether pi is normal.
I thought anti-vaccinationists were dense and dishonest -- these folks may surpass them).

scientious said...

There is a topic that I wouldn't mind discussing if I can find anyone who wants to discuss it and is able to. There is a theory that has been researched for the past several years. It was hoped that it could be published in 2018, but Trump's election removed any urgency in that respect. It is now not expected to be published before 2021, assuming that Trump does not get re-elected. The theory description would be a book although an outline might be submitted to a scientific journal before that. The theory is about cognition. It explains in detail how intelligence and consciousness evolved from roundworms to fish to mammals to humans. It explains in detail how consciousness works and explains what would be necessary to replicate it in machine form. I suppose it isn't surprising that when you have a large theory there are also side-effects.

As the theory developed, it matched up well with evidence but it lacked a firm foundation. Unfortunately, there didn't seem to any previous work in the various fields that provided this. So, it was necessary to create an entirely new knowledge theory to provide the base. With this grounding, the theory appears much more solid and therefore more likely to have some substance.

Some of the implications seem likely to attract perhaps rabid counter-attacks. For example, it has refutations of Idealism and the notion of a soul or spirit. It details limitations with organizational leadership such as CEO's of large corporations or presidents of larger countries. It shows the limitations of search algorithms and market models such as those used by Google.

The theory seems to close up one of the last remaining gaps in evolutionary theory. I mention this in particular because not only is this a common defense by religious apologists such as Berlinski and Deepak Chopra, but it has also been mentioned by more serious scientists such as Jerry Coyne. You get the impression that consciousness is mysterious and perhaps unexplainable, and this provides one more refuge for those who desperately want to believe in the supernatural and life after death. I don't know of a theory that could disprove God, but if the soul is disproved then it wouldn't seem that God would serve much purpose.

The theory not only suggests how to build a human-equivalent machine intelligence but opens up a larger area of computer architecture that is not being used today. This could prevent computer development from grinding to a halt as we become unable to maintain process shrinkage. It looks like 14nm is approaching that limit. It is not known today if Intel can go any smaller. And, since we've already reached the practical limits of data width (64 bits), pipeline depth, and clock speed there doesn't seem to be much left to keep things moving forward. This theory could provide a path that allows that.

A disturbing aspect is that it does not appear that consciousness is computable. In other words, you wouldn't be able to run it on a Turing Machine. On the other hand, consciousness is not mysterious or supernatural so it could be replicated on finite machine hardware. This suggests that Church-Turing is a subset of a larger class of operations for which there is no good label. If Church-Turing is computation then what would the larger group be? CTx perhaps. Maybe but what would you then call the operation if not computation? Supercomputation, Intercomputation, Cognitive computation -- none of these exactly roll off the tongue.

The theory seems to do what you would expect it to. It appears able to fully deconstruct the Chinese room and Mary's room. It seems able to answer both questions of the binding problem and seems able to give a dramatically more robust solution to the Frame problem. It provides a refutation to the computing theory of mind and likewise refutes Harris' and Coyne's claims about free-will.

I've wondered what the result would be of publication. It seems like it could give an unprecedented boost to the economy. It could also be highly disruptive.

scientious said...

The past responses have been varied if not helpful:

Please post the algorithm for consciousness. -- Consciousness requires several entire systems, none of which are based on a single algorithm.

Prove that such a theory exists and is correct and then we can discuss it. -- The only way to prove it would be to publish it. If it was then correct then it wouldn't be discussed on an internet forum or blog. I assume they would have conferences to discuss it and it would become part of the college curriculum. It also wouldn't make sense to publish something to have a discussion about the implications of publishing something. The implications would be moot at that point.

Such a theory can't exist because it makes me feel jealous and stupid. -- Well, that's too bad but theories have been invented by people for some time now. Probably someone felt jealous and stupid when Archimedes figured out density. Nevertheless, density has been a useful concept. There have also been many other, very clever people. Figuring out the diameter of the Earth by measuring the angular variation down a well, remarkable. Figuring out the speed of light without modern measuring devices, amazing. Figuring out the weight of the Earth using a very sensitive scale, impressive. Calculating Avogadro's number, brilliant. You'll grow old and die before you get around to being jealous of everyone who has ever thought up something impressive; I don't recommend it.

No one cares what religious people think; that shouldn't be important. -- My mother listens to various religious broadcasters/charlatans. They regularly denounce Evolutionary theory as a tool of the devil and even suggested that Obama was possessed by demons. Jim Bakker has made the extraordinary claim that he personally saved Donald Trump from demon possession and that Trump's election was blessed by God. My mother is still convinced that God would have destroyed the US if Hillary had been elected. The truth is that religious broadcasting is worth a lot of money; it's in the hundreds of billions. They aren't going to give up without a fight and I can't imagine what type of rational argument would hold sway in that kind of environment. Think of the right to life movement and then multiply it by 10.

If this theory were true then it would have been the result of experimental evidence. -- This isn't true of every theory; sometimes theory leads and other times it lags experimental evidence. Special Relativity was ahead of experimental evidence. The Ideal Rocket equation wasn't proven until Goddard's experiments. Sometimes leading theories are later proven and sometimes they are later disproved.

That theory can't be correct because I like IIT or neural network theory. -- Well, you could be right. Which one has a definition of understanding? Which one agrees with evolutionary theory? Which one can explain what hardware would be required to make a machine version of human comprehension? IIT also claims that a diode has a little bit of consciousness. It is an emergent theory but can't explain when this emergence might occur.

At what point does consciousness emerge? -- It never does; this concept is incorrect.

If it was a real theory it would be published in a journal, not in a book. -- In hardcover form, it would probably run about 500 pages. This would seem to be too long for a journal article. Darwin published an outline in a journal but gave the lengthy details in a book.

scientious said...

It seems to me like there would be a strong backlash against a theory like I described from fundamentalist religious people and those who like the idea of the supernatural, people who like ghosts or telepathy or universal consciousness. Then you have the televangelist charlatans, groups like Transcendental Meditation which even sucked in people like Doug Henning, and people pushing concepts that your brain has limitless potential. Again, a lot of these people make money off of this so it isn't difficult for me to imagine that they will take refuge in idea of an attack on religion. And it isn't difficult for me to think that the party that takes a pilgrimage to Liberty University to kiss the ring of the president to prove their religious bona fides will strongly back these people.

I see multiple forces at work and it seems rather complicated. If the theory was right then I assume it would be strongly backed by scientists although to be honest it is difficult for me to imagine scientists being gracious after seeing their life's work demolished. Surely there would be some hard feelings and arguments against it, at least until there was supporting evidence. Corporations might be at something of a loss since they might see a strong profit motive but many are also very conservative and their CEO's might very much resent the implications about leadership. Common people might be torn between the promise of jobs and the loss of religion. Then on top of this you would have competition from other countries. Russia, China, Indian, South Korea, and Japan wouldn't care about the religious aspects; they would simply charge forward and try to gain a technological advantage. They would be followed by western Europe so we would probably end up with a greater competition than the US/Soviet space race. I just don't know how all of this would shake out but I'm pretty sure that it would be better if the administration was both competent and not beholden to religious fundamentalists.

JimV said...

I liked this rebuttal article by D. Abbott in an IEEE paper (PDF)

"The Reasonable Ineffectiveness of Mathematics"

philosopher-animal said...

Does the article by Hamming actually discuss evolution? If so, I might be willing to give them the slightest tiniest benefit of the doubt. If not, I'd go with dishonest, since they are that too.

Mikkel Rumraket Rasmussen said...

The point about the subjectivity of the effectiveness of mathematics is further undermined by the fact that the vast majority of human beings who ever lived never learned anything more advanced than the barest of basics of arithmetic. If at all.

Most people either don't have the time to get competent at a high level of mathematics, or are incapable (or so bad at it it would take them most of their life).

JimV said...

One of my personal ideas is that math is thinking, thinking is math. When I have three errands to do, at different locations, and take a moment to think what is the best order to do them in, I am doing math. (There is good math and bad math - good and bad thinking - of course.)

So yes, to the extent that we understand the universe, it is math that gives us that understanding. The deeper question is, how is thinking/research/design accomplished? Another of my personal ideas is that thinking, design work, and scientific progress are done by a process of evolution (trial and error, with selection criteria and memory). If I am correct, instead of contradicting the biological theory of evolution, our mathematical accomplishments are more evidence that evolution works.