Three Flukes Made Modern Science Possible

Roger Schlafly
What does this have to do with your essay? I am refuting what you say in your essay.

I have just explained why there are no numbers in computers, and why computers are not doing addition, and I might well have added that “logic gates” are not doing logic. “Logic gates” / transistors are useful because they can be used to symbolise logic, if you place them in the right setting of circuits and voltages, and with the right computer program.

“Everybody knows about number representations”

Do they? What are numbers? The real world, including bodies, brains and the physical correlates of consciousness, is built out of real-world categories (examples would be mass, energy, position), real-world lawful relationships, and real-world numbers. Numbers representing the real world are obtained from measurement of the real world (e.g. by physicists). People use symbols to represent these real world numbers, but the real world numbers are not in any way implied by the symbols that are used to represent them (except from the point of view of pre-existing educated human beings), and the symbols of numbers don’t reverse-engineer themselves and transmogrify into real-world numbers.

The voltages, circuits, and transistors in computers, that might well be used to represent real-world numbers, real-world categories, and real-world lawful relationships (possibly including the physical correlates of consciousness), don’t reverse-engineer themselves and transmogrify into real-world numbers, real-world categories, and real-world lawful relationships, and real-world consciousness.

Roger Schlafly
Re your own essay:
I am objecting to your mindless propagation of the following sort of nonsense:

“They may even get to the artificial intelligence singularity, and people might become slaves to super-intelligent robots that take over the world. The laws of physics would become implicit in the trillion-parameter models being used. The computers could deduce everything we can with our theories and much more, but the poor mortals would have no conceptual understanding of how the world works.”

In my above reply, I explained that computers/ AIs merely symbolically represent numbers, addition, and logic; computers/ AIs are not actually performing addition or logic. The symbolically represented numbers in computers are not like the numbers that the real physical world is made of. The same goes for the symbolically represented categories and relationships in computers: they are not like the categories and relationships that the real physical world is made of.

Re my own essay:
As I have noted above (and I said more or less the same thing in my essay):

“The real world, including bodies, brains and the physical correlates of consciousness, is built out of real-world categories (examples would be mass, energy, position), real-world lawful relationships, and real-world numbers. Numbers representing the real world are obtained from measurement of the real world … People use symbols to represent these real world numbers …”

There is a difference between the real physical world (e.g. a real-world number associated with relative position) and the symbols that people use to represent the world like a number symbol. Although symbols may be made out of measurable physical matter, the symbol aspect of the matter only exists from the point of view of people.

Contrary to what you say, there is no “trivial distinction between numbers and representations of numbers”. There is no “trivial distinction” between the real-physical-world categories, the real-physical-world lawful relationships, and the real-physical-world numbers that the real physical world is made out of, on the one hand, and the mere symbols that people use to represent these real-world categories, real-world lawful relationships, and real-world numbers on the other hand. There is no “trivial distinction”, there is an enormous distinction.

Roger Schlafly
Yes, so that is why your:

“They may even get to the artificial intelligence singularity, and people might become slaves to super-intelligent robots that take over the world. The laws of physics would become implicit in the trillion-parameter models being used. The computers could deduce everything we can with our theories and much more, but the poor mortals would have no conceptual understanding of how the world works.”

is complete nonsense. I object to your propagation of this complete and utter nonsense.

Roger Schlafly

Hi MustardGorilla, I now read your essay once again.

You nicely and reasonably worked out the first fluke, although I would suggest that even before human beings began to systematically examine the mathematical precisions of the heavens, they must have had a deep intuition that the world is, in certain areas, deeply ordered. For example humans self-evidently know that their bodies usually follow their inner intents (surely within the limits of what can be achieved by human bodies), for example lifting an arm etc. This is also a kind of order, although not a mathematically predetermined one, and not subject to rigorous proofs, since bodies can also misbehave (illness etc.).

Since mathematics starts with counting, it is no wonder that it exists, since in everyday life, things are countable. The wonder, at least to me, is that it can be extended to ever more abstract objects in the minds of human beings, like for example an uncountable infinity. Here I would be cautious whether or not such an uncountable infinity has any counterpart in the natural world, although that infinity can be proven to be consistent with the basic assumptions of mathematics. This surely poses the question whether or not a consistent mathematical object and its constituents must have any counterparts in the physical world, or are just a concatenation of logical operations in the human mind.

The next fluke you rightfully mention is solar and lunar eclipses:

“What if we had no eclipses? And if the sky were cloudy all the time, we would have no Astronomy. No stars, no moon, no planets, and no celestial mechanics. Where would we even get the idea of a clockwork universe?”

Very good question in my opinion and a remarkable fluke indeed. Someone like me is tempted to say that it looks a bit like fine-tuning for the purpose of discovering the wonders of creation for everybody to see if one is willing to look.

Your third fluke is also surely remarkable. In the dark ages, there were no bibles written in other languages than Latin or Greek around in Europe, this had to wait until Guttenberg came along. Only the curches had complete scripts of the holy texts. People must have relied on what had been preached from the pulpit. Only after Guttenberg people where able to study the holy texts and think about them. This must have been a real step towards proving and / or disproving what was said from the pulpit. Unless this happened, people were already creative in developing ever more sophisticated tools for everyday life as well as ever sophisticated theories about the regularities in the world. After the bible was accessible for everyone, civilisation refined its system of justice, based on what has been found in the bible and on what has been found out about logical deduction / induction. We therefore should do a little justice to christianity, since Newton and surely others were inspired by the bible to blieve that within creation there must be deep order. If they did not believe that, they rarely would have figured out what we know today.

Your next assumption, the inevitability of computers and even AI, can really be seen as the consequence and the success of a deterministic, clockwork world view that had its beginnings with the ancient astronomers, and its subsequent refinement towards an exclusive worldview of mechanistic causality: Aristotle's final causes more and more lost their meaning within science. The more astonishing it is for me that with the new era of AI, these final causes seem to have a revival in somewhat new clothes: AI is supposed to deliver the final answers to questions that human beings haven't been able to answer yet. AI itself is seen as somewhat the teleologically pre-determined end point of fundamental scientific research. But that is only a belief, based on the successes of the clockwork world view. However, according to the bible, there exists another mode of prediction, it is called prophecy. There are many prophecies that already fullfilled. To name only one, the one that made christianity possible in the first place, was discovered by a retired Scotland Yard Inspector. His name was Sir Robert Anderson, and he deciphered a prophecy in the book of Daniel, Chapter 9, 25-27, so that we can determine the day Jesus Christ has ridden into Jerusalem on a donkey. We therefore can also deduce when Jesus began to preach, he was 29 years old (in the epistles it is stated that he was around 30 years old when he began to preach).

This is only one of many examples which, together with biblical archeology and language examination show that the book of Daniel must have been written long before Christ was born. Understanding Darwinian evolution without genetics is a matter of fact, but we should not forget that is also a matter of fact that we have no clue how the first self-replicating thing (something very, very complicated which must be stable and reliably copy itself for Darwinian evolution to at all take place) came into the world.

“What if Science had developed without reliable mathematics and a belief in universal laws? … Eventually, they would get computers and neural nets and scale up to large language models. They would make science problem-solving engines where they would feed in data and get valuable predictions.”

No, I think this is fantasy. To faciliate a computer, to produce semi-conductors and all the rest, you surely need some reliable mathematics and moreover a kind of conceptual framework how nature works, at least in its coars-grained appearance.

“If we had gotten to large language models before classical mechanics, we might have developed a very different idea of science. Science would be all about feeding large datasets into black boxes for artificially intelligent predictions.”

Again no, since you need tested, reliable theories to build something that complicated like a computer array. In other words, you need a sufficiently accurate model in your head how nature behaves, otherwise building a computer like we have today counterfactually makes no sense at all.

All in all I found that your essay expresses some memorable flukes, worth thinking about in my opinion!

Roger Schlafly

Thank you also MustardGorilla for evaluating my comments.

Best wishes
AquamarineTapir

Global externalities and Black Box Science happened in spite of the flukes that made modern science possible. Some interesting points regarding math, astronomy and Christianity. The latter did promote the nuclear family but don't see it cultivating the individualism you mention. Not much room in our essays to make definitive connections. Seems natural for the ancients as well as the moderns to be drawn to astronomy and its wonders, mysteries explained with religion and finally science. Soft science is with us but also a bastardized science with a business agenda, robbing science of a more purist focus and replacing it with a bias for the commercial. That's the distortion I see that needs correcting.

Roger Schlafly You searched for historical reasons for the development of our science. I think your explanations are good, and even if they are not, they are a base for further improvements.

You made even some predictions, i.e. about GPT10. We will see. Only your Black box science is too simplified, but this essay is short, and time is limited for such research. I would add here that guessing of physical formulae also belongs to this black box and p-value. Although we do not know p-values at such guessing, we can estimate it and later we can improve it. Namely, I guessed formulae for the elementary particle masses in past, and they are evident in intranet. Best wishes