The "no miracles" argument against scientific realism

[RexAllen]

Let's assume that our best scientific theories tell us something true about the way the world *really* is, in an ontological sense. And further, for simplicity, let's assume a deterministic interpretation of those theories.

In this view, the universe as we know it began ~13.7 billion years ago. We'll set aside any questions about what, if anything, preceded the first instant and just draw a line there and call that our "initial state".

Given the specifics of that initial state, plus the particular causal laws of physics that we have, the universe can only evolve along one path. The state of the universe at this moment is entirely determined by two, and only two, things: its initial state and its casual laws.

But this means that the development of our scientific theories *about* the universe was also entirely determined by the initial state of the universe and it's causal laws. Our discovery of the true nature of the universe has to have been "baked into" the structure of the universe in its first instant.

By comparison, how many sets of possible initial states plus causal laws are there that would give rise to conscious entities who develop *false* scientific theories about their universe? It seems to me that this set of "deceptive" universes is likely much larger than the set of "honest" universes.

What would make universes with honest initial conditions + causal laws more probable than deceptive ones? For every honest universe it would seem possible to have an infinite number of deceptive universes that are the equivalent of "The Matrix" - they give rise to conscious entities who have convincing but incorrect beliefs about how their universe really is. These entities' beliefs are based on perceptions that are only illusions, or simulations (naturally occurring or intelligently designed), or hallucinations, or dreams.

It seems to me that it would be a bit of a miracle if it turned out that we lived in a universe whose initial state and causal laws were such that they gave rise to conscious entities whose beliefs about their universe were true beliefs.

A similar argument can also be made if we choose an indeterministic interpretation of our best scientific theories (e.g., quantum mechanics), though it involves a few extra steps.

[Twiglet]

Quantum physics suggests that the universe is inherently random. I'd be interested to see what line of reasoning leads you to believe that we are "preprogrammed" to discover anything.

[FBM]

Is it not acknowledged by scientists that scientific knowledge is incomplete, therefore, inaccurate? Until we get it ALL figured out, we're in one of those deceptive universes, eh?

[Xamonas Chegwé]

Interesting first post, Rex. Welcome to Ratz.

I have never been much of a fan of determinism, personally. There are too many interlocking systems that are too finely dependent on initial conditions for me to swallow a complete lack of randomness in the universe.

Even if we accept determinism, your 'honest' and 'dishonest' universes don't follow, in my view. The point at which you state (with no justification) that

For every honest universe it would seem possible to have an infinite number of deceptive universes that are the equivalent of "The Matrix"

is where you lose me completely. You step right out of physical science and into metaphysical sci-fi here - sorry.

Besides, the human race alone has developed an enormous number of theories about the way that the universe works - everything from tribal beliefs in anthropomorphic beasts to quantum physics - and more emerge almost daily. Assuming we hang around for a little longer, we will almost certainly develop more theories and refine those that we already have. We cannot rule out the possibility of other intelligent races in the universe - in fact, its sheer size makes them almost inevitable - and have no way of knowing what ideas they have hit upon regarding the nature of reality.

So how can you say that there is a single, hard-coded theory in each universe - by your own argument, all of the above are hard-coded.

[RexAllen]

Quantum physics suggests that the universe is inherently random. I'd be interested to see what line of reasoning leads you to believe that we are "preprogrammed" to discover anything.

Well, first, there are deterministic interpretations of quantum mechanics. The Many Worlds interpretation and the De Broglie-Bohm interpretations are both deterministic. Also Gerard 't Hooft (1999 Nobel Prize in physics for work in electroweak interactions) has proposed a deterministic interpretation. AND, the "Consistent Histories" interpretation is agnostic with respect determinism vs. indeterminism.

But, setting that aside, even in an indeterministic universe there are still initial conditions and there are still laws - the laws just have an intrinsically probabilistic aspect.

In the indeterministic case, the physical laws of our universe are like the rules of a card game that includes a certain amount of randomness...for instance, requiring a random shuffle of the deck after each hand is played. But the number of cards, the suits, the ranks, and the rules themselves are not random...those aspects of the game are deterministic.

In quantum mechanics using the Schrodinger equation, the evolution of the wavefunction describing the physical system is taken to be deterministic, with only the "collapse" process introducing an indeterministic aspect.

Another example of using randomness in a deterministic framework are "randomized algorithms", like the Randomized Quicksort. To some extent you can think of the unsorted list as the "initial conditions" and the algorithm as the "causal law".

So, in the case of the quicksort algorithm, no matter which pivot values end up being randomly selected the algorithm is still going to correctly sort the list. The randomness is constrained by the fact that it is used in the context of the deterministic algorithm. Because of the deterministic aspect of the algorithm, it will always be a sorting algorithm...never a Pac-Man game.

Returning to my original point: The same goes for our universe in the indeterministic case. Quantum randomness can only increase the probability of life discovering something true about the underlying nature of the universe IF the initial conditions and the non-random aspects of the causal laws allow for this to be the case.

So, in the indeterministic case our discoveries aren't preprogrammed, but they are still entirely a function of:

1) Initial conditions
2) Causal laws (which have a probabilistic aspect)

Did I get that across clearly? Or does it sound like gibberish?

[RexAllen]

Is it not acknowledged by scientists that scientific knowledge is incomplete, therefore, inaccurate? Until we get it ALL figured out, we're in one of those deceptive universes, eh?

Not necessarily. Even incomplete theories could still tell us something true about the underlying nature of reality, even if they weren't correct in every respect. See "Structural Realism".

The issue is that science relies entirely on our observations. But it requires a leap of faith to assert that our observations tell us something true about how the world really is. For instance, if we were in The Matrix, our observations (and any science based on them) would only tell us about The Matrix's simulated version of reality, not about how the universe outside the Matrix *really* is.

Any scientific experiments you performed when you were in the Matrix would only tell you about the rules of the simulation. The experiments would never reveal anything about the hardware the simulation ran on, OR the physical laws of the universe that contained the hardware.

Any set of observations that matched the observations we have, would lead one to derive the laws of physics that we have, EVEN if you are in a simulation running on an alien supercomputer in an alternate universe with completely different physical laws (laws which still allow for implementations of Turing machines).

[Trolldor]

Even incomplete theories could still tell us something true about the underlying nature of reality

No.

[Twiglet]

Did I get that across clearly? Or does it sound like gibberish?

I think you were pretty clear, but it does seem like you are trying to plug interpretations of quantum theory which are deterministic, and those haven't gained very much traction in the scientific community. There is no experimental evidence to support the many worlds theory.

Also you take on the Schrodinger equation is a little puzzling. The Schrodinger equation describes a wavefunction which is essentially a set of probabilities for the outcome of a measurement, which summed over the wavefunction - come to 1. It isn't deterministic. Simplistically, it says if you roll the dice, the number will come up between 1 and 6, with equal probability. When you roll the dice, the outcome you get has a probability of 1 and all other possibilities "collapse" to zero.

Your argument about physical laws being non-random is fine, but the trouble is, if the universe is inherently random, then the chances of being able to predict from the initial conditions what things will look like a year later are vanishingly small. Even with pre-defined physical laws.

[RexAllen]

Interesting first post, Rex. Welcome to Ratz.

Thanks!

I have never been much of a fan of determinism, personally. There are too many interlocking systems that are too finely dependent on initial conditions for me to swallow a complete lack of randomness in the universe.

It sounds like you are describing chaotic systems, not random systems.

You can write computer programs that exhibit chaotic behavior. If you change the starting data just a little, then the final results will be vastly different. BUT, you can run the program over and over with the same starting data and you will get exactly the same results each time.

Even if we accept determinism, your 'honest' and 'dishonest' universes don't follow, in my view. The point at which you state (with no justification) that

For every honest universe it would seem possible to have an infinite number of deceptive universes that are the equivalent of "The Matrix"

is where you lose me completely. You step right out of physical science and into metaphysical sci-fi here - sorry.

Hmmm. Well, what is your position on consciousness? Probably we should clear that up first.

Do you believe that an accurate computer simulation of a human brain would be conscious in the same way that you are?

What do you think of functionalism and multiple realizability?

Are you familiar with David Chalmers' Facing Up to the Problem of Consciousness?

Also, you sound a bit dismissive when you say "metaphysical sci-fi", but note that scientific realism/physicalism involve metaphysical commitments as well.

"Scientists sometimes deceive themselves into thinking that philosophical ideas are only, at best, decorations or parasitic commentaries on the hard, objective triumphs of science, and that they themselves are immune to the confusions that philosophers devote their lives to dissolving. But there is no such thing as philosophy-free science, there is only science whose philosophical baggage is taken on board without examination." -- Daniel Dennett

So how can you say that there is a single, hard-coded theory in each universe - by your own argument, all of the above are hard-coded.

I never said that there was a single, hard-coded theory in each universe. I said that (assuming scientific realism) the universe's initial conditions and causal laws determine everything that follows.

So for any particular event that happens today, theoretically you should be able to trace the chains of cause and effect from that event *back* to the initial conditions at the first instant of the universe.

And the reason that you could do this is because (according to scientific realism) there are causal physical laws that govern how the state of the universe changes over time.

So, ultimately what explains the occurrence of that particular event? Answer: The initial conditions of the universe *plus* the causal laws of physics as applied over ~13.7 billion years.

Right?

[Xamonas Chegwé]

I have never been much of a fan of determinism, personally. There are too many interlocking systems that are too finely dependent on initial conditions for me to swallow a complete lack of randomness in the universe.

It sounds like you are describing chaotic systems, not random systems.

You can write computer programs that exhibit chaotic behavior. If you change the starting data just a little, then the final results will be vastly different. BUT, you can run the program over and over with the same starting data and you will get exactly the same results each time.

This is true of a purely mathematical, chaotic system - ie. one that exists solely in the realm of number. I am not convinced that it is true of any physical, chaotic system - once we get down to quantum scales, there appears to be inherent randomness built-in from what I understand (which is not a huge amount - but then, who does understand QM?)

Even if we accept determinism, your 'honest' and 'dishonest' universes don't follow, in my view. The point at which you state (with no justification) that

For every honest universe it would seem possible to have an infinite number of deceptive universes that are the equivalent of "The Matrix"

is where you lose me completely. You step right out of physical science and into metaphysical sci-fi here - sorry.

Hmmm. Well, what is your position on consciousness? Probably we should clear that up first.

We probably should. Perhaps it's relevant. I don't really see how though.

Do you believe that an accurate computer simulation of a human brain would be conscious in the same way that you are?

Yes. Any construct that is capable of doing exactly what my brain does, would be worthy of the accolade 'conscious' IMO.

What do you think of functionalism and multiple realizability?

I would say that I am functionalist in approach to human consciousness, in that I am of the opinion that there is no part of 'me' that is separate from my physical structure. I am mostly in agreement with Dennett's ideas. I am not familiar with multiple realizability - I will go and look into it.

Are you familiar with David Chalmers' Facing Up to the Problem of Consciousness?

No. But I have heard of him. He is essentially a dualist, is he not? Skimming that paper, he seems to imply that there is 'something else' in consciousness beyond the physical brain - I would dispute that (without absolutely ruling it out - I just feel that such an 'extra' is not required to understand consciousness.)

I have answered your questions on consciousness but I fail to see their relevance. My complaint was not against the possibility of Matrix style universes per se but against your huge non sequitur in claiming that there are possibly an infinite number of such universes for every 'honest' one. It is the leap from hard science to untestable, metaphysical speculation that made me go, "Woah there!" Do you see my point?

Also, you sound a bit dismissive when you say "metaphysical sci-fi", but note that scientific realism/physicalism involve metaphysical commitments as well.

"Scientists sometimes deceive themselves into thinking that philosophical ideas are only, at best, decorations or parasitic commentaries on the hard, objective triumphs of science, and that they themselves are immune to the confusions that philosophers devote their lives to dissolving. But there is no such thing as philosophy-free science, there is only science whose philosophical baggage is taken on board without examination." -- Daniel Dennett

I was being dismissive - I found it a huge leap into speculation. It is one thing to admit that there is a philosophical element to scientific thinking (as in the Dennett quote) and quite another to accord speculative metaphysics an equal standing with physics.

Dennett does not specifically mention metaphysics in his quote but philosophy in general. I assume (although without reading the quote in context, it is hard to say for certain) that Dennett is referring to the epistemology of scientific thought - the underlying philosophy of the scientific method - and to its central tenet of physicalism. Knowing Dennett, I doubt that he was going further and welcoming 'alternative views of reality' into the scientific fold. Please provide the context of the quote if I am wrong here.

So how can you say that there is a single, hard-coded theory in each universe - by your own argument, all of the above are hard-coded.

I never said that there was a single, hard-coded theory in each universe. I said that (assuming scientific realism) the universe's initial conditions and causal laws determine everything that follows.

I would rephrase you here and say that, "the universe's initial conditions and causal laws determine everything that can follow." With that caveat, I agree. Note that I am allowing for a possible random element after the initial conditions. I concede that it is not currently possible to say for certain whether the universe is 100% deterministic or not, however, I feel that the balance of probability lies in the 'not' camp.

So for any particular event that happens today, theoretically you should be able to trace the chains of cause and effect from that event *back* to the initial conditions at the first instant of the universe.

Given time and resources far in excess of those available in our universe, such a theoretical exercise might be feasible! Travelling 'back' is one thing though, travelling 'forward' and extrapolating future events from the initial conditions is only possible if you can be sure that there is no randomness in the universe.

And the reason that you could do this is because (according to scientific realism) there are causal physical laws that govern how the state of the universe changes over time.

Agreed. However, these laws may contain/allow for randomness.

So, ultimately what explains the occurrence of that particular event? Answer: The initial conditions of the universe *plus* the causal laws of physics as applied over ~13.7 billion years.

Right?

Right, up to a point. If there is a random element to the universe then the initial conditions of the universe *plus* the causal laws of physics as applied over ~13.7 billion years determines not what will happen but the full range of what could happen.

It all comes down to whether or not you hold to determinism. You may have noticed that I don't.

[AshtonBlack]

erm... is it 42?

http://www.weirduniverse.net/blog/comments/3968/


Welcome Rex!

[RexAllen]

So, ultimately what explains the occurrence of that particular event? Answer: The initial conditions of the universe *plus* the causal laws of physics as applied over ~13.7 billion years.

Right?

Right, up to a point. If there is a random element to the universe then the initial conditions of the universe *plus* the causal laws of physics as applied over ~13.7 billion years determines not what will happen but the full range of what could happen.

It all comes down to whether or not you hold to determinism. You may have noticed that I don't.

I think that first we should decide whether the claim in my original post is correct with respect to a deterministic universe. IF we can agree on that, THEN let's debate the indeterministic case.

Hmmm. Well, what is your position on consciousness? Probably we should clear that up first.

We probably should. Perhaps it's relevant. I don't really see how though.

Conscious experience is where everything starts, right? I think, therefore I am.

However, as Descartes noted, the existence of hallucinations and dreams shows beyond a doubt that our conscious observations do not necessarily tell us anything about how the universe actually is.

But since science is based on those conscious observations, science does not necessarily tell us anything about how the universe actually is either. Science only tells us about our observations.

So we start with our observations, and then we construct plausible narratives that are consistent with what we have observed. These narratives may be useful in analyzing recurring patterns in the records of our past observations, but that doesn’t necessarily mean that they are true of anything that exists outside of our observations.

The possible existence of matter in the form of quarks and electrons (or strings, or quantum fields, or whatever) is consistent with our observations, but obviously we have no direct knowledge of quarks and electrons or the rest. Their existence, and the physical laws associated with them, are only inferred from our observations.

Even something right in front of me, like my chair, I still only know through my conscious experience. I see a chair here, but I don’t know that the chair actually exists. I could be dreaming, for instance, in which case the chair exists entirely within my mind.

SO...knowing where you stand on consciousness is key to understanding your position on everything else.

Do you believe that an accurate computer simulation of a human brain would be conscious in the same way that you are?

Yes. Any construct that is capable of doing exactly what my brain does, would be worthy of the accolade 'conscious' IMO.

Good, good. Then you believe in multiple realizability. This becomes important in a bit.

My complaint was not against the possibility of Matrix style universes per se but against your huge non sequitur in claiming that there are possibly an infinite number of such universes for every 'honest' one. It is the leap from hard science to untestable, metaphysical speculation that made me go, "Woah there!" Do you see my point?

This actually boils down to pretty straightforward computer science.

Okay, so the Matrix scenario is basically a computer simulation of our universe. As you said above, any humans that were part of that simulation would be conscious in the same way that you were conscious.

Any program that can be run on a computer can be executed by a Turing machine. This is the Church-Turing Thesis:

The Church-Turing thesis (formerly commonly known simply as Church's thesis) says that any real-world computation can be translated into an equivalent computation involving a Turing machine.

Further, there are infinitely many Turing machines that compute any such real-world computation:

Indeed, we can prove that there are infinitely many Turing Machines that compute any Turing Machine computable functions.

So:

1) We assume that our universe is an "honest universe".
2) The Matrix is a computer simulation of our universe...humans inside the simulation live in a deceptive universe.
3) Any program that can be run on a computer can also be run on a Turing machine.
4) There are infinitely many different Turing machines capable of running any given program.
5) We can conceive of a universe containing nothing but an implementation of such a Turing machine.
6) We can conceive of infinitely many universes, each containing a different Turing machine that runs the same "Matrix"-style program.
7) There could be infinitely many of these deceptive universes running "Matrix"-style simulations of our "honest" universe.

So. There you have it. There is no huge non sequitur in my claim.

Note that on step 5, I'd also claim that there are an infinite number of possible physical implementations for any given Turing machine.

Setting aside limits on physical memory, all modern "digital" computers are equivalent in terms of what they can compute. And there's no limit to what you can build a computer out of.

I could say more, BUT, we digress.

Returning to the main topic: As I said, first I think we should try to agree about whether my original post is correct with respect to the deterministic case. And if we can agree on that, then we should move on to the indeterministic case, which is a bit more complicated.

[Twiglet]

Your initial post essentially says that in a deterministic universe, our understanding of the world is "pre-baked". Well, that completely stands to reason, the initial conditions entirely determine absolutely everything that happens from point zero to time n, where n as anything you like.

I'm not sure that the rest of your arguments (about the miracle of consciousness etc) follow from that though.

There could in principle be heaps of possible ways for consciousness to arise. Even in an entirely deterministic universe, why the need for a "hidden hand"?

Plus, how would we know (in either a random or deterministic universe) whether our science is "true" - you assumed that our ideas are "honest" but you can't use your own assumptions as proof of the outcome you derived from them. Thats a tautology. You arrived at that outcome by assuming it in the first place.

_____

Further, taking what we know of our history so far, why is it (in our miraculous universe) - that there have been such a vast number of misunderstandings?

We know at a core level, even now that General Relativity and QP suffer from irreconcilable differences. It's one of the reasons coming up with ideas like String theory. Our scientific method may be honest, but let's not kid ourselves, we don't for sure know we have the "right" answers just yet. Only good working approximations (you made an assumption about that too).

Moreover, there have been numerous instances of theories which were held true for hundreds of years being ditched. In a deterministic universe, that would be "baked in" too, wouldn't it?

[Lion IRC]

I don't find anything particularly "miraculous" about things we don't yet understand and I dislike scenarios which are based on assumptions that there was nothing before 13.7 billion years ago.

It’s a bit like saying… suppose you opened your eyes and found yourself seated in the cockpit of a Learjet a few thousand metres above the ground and calling that your “initial state”.

Are you going to let someone tell you that your “scientific theories” should remain strictly limited to everything that happened AFTER you opened your eyes?

Failure to speculate and investigate about the goings-on outside the cockpit might have unpleasant consequences.

The mind of the prudent acquires knowledge. The ears of the wise seek it out.

Lion (IRC)

[Xamonas Chegwé]

So:

1) We assume that our universe is an "honest universe".

A reasonable assumption. It consistently appears to be so and there are (apparently) potentially dire consequences for assuming that parts of it are illusory.

2) The Matrix is a computer simulation of our universe...humans inside the simulation live in a deceptive universe.

The Matrix is a fairly good Sci-fi film that postulates such a system - let's not lose sight of that. A true Matrix, capable of simulating the entire universe would need to be unfeasibly large and complex. It would need to contain every piece of information in the universe in order to replicate it - making it far larger than the universe itself!

3) Any program that can be run on a computer can also be run on a Turing machine.

The Turing machine does not actually exist, you know? For a start, it requires a tape of infinite length to function!

4) There are infinitely many different Turing machines capable of running any given program.

You are fond of infinite numbers of things. Exactly where are you going to put them in our finite universe?

5) We can conceive of a universe containing nothing but an implementation of such a Turing machine.

You can. I can't! Since it does not and cannot exist!

6) We can conceive of infinitely many universes, each containing a different Turing machine that runs the same "Matrix"-style program.

More infinity. We can also conceive of infinite reasons why this is fanciful nonsense.

7) There could be infinitely many of these deceptive universes running "Matrix"-style simulations of our "honest" universe.

Not from my point of view, no.

So. There you have it. There is no huge non sequitur in my claim.

I'm afraid there is. You stray from hard science to thought experiment to fanciful speculation.

I have no truck with metaphysical thought experiments and 'what if' scenarios - they are fun and can lead to interesting conversations. I am a little perturbed, however, when an attempt is made to justify those scenarios scientifically. Science is based on a physicalist standpoint and deals with what we can know about the universe as it is (or as it appears to be.) If the universe is dishonest, fake, a construct, or contained within an infinite number of infinitely powerful Turing machines, it doesn't matter to science because we can't tell. Sure, everything I have ever experienced might have been a dream but until that is proved to me, I will treat it exactly as it appears to be.

Note that on step 5, I'd also claim that there are an infinite number of possible physical implementations for any given Turing machine.

There's that 'i' word again. There is a lot of room in your universe.

Setting aside limits on physical memory, all modern "digital" computers are equivalent in terms of what they can compute. And there's no limit to what you can build a computer out of.

Equivalent in what way? Certainly not in terms of memory capacity, processor speed, etc. And self-amending, 'neural-network' computers are capable of performing tasks that your PC is not. Furthermore, there are many restrictions on what you can build a computer out of - try making one out of nothing but hydrogen and see how far you get!

I could say more, BUT, we digress.

Yes. From science to fantasy.

Returning to the main topic: As I said, first I think we should try to agree about whether my original post is correct with respect to the deterministic case. And if we can agree on that, then we should move on to the indeterministic case, which is a bit more complicated.

I have another criticism of your OP.

Even if we concede the deterministic POV and accept that the universe can only unfold in a single manner, based upon its initial state and the laws of physics, that is a long way from knowing how that universe is going to unfold. It might be that the only way to find out how the 'program' is going to run is by running it. By the determinist argument, the weekend's lottery numbers are already determined - but the processing power necessary to predict those numbers based on the position of every particle in the universe at the point when the button is pushed is unimaginably huge - especially when compared to simply pushing the button!

So, perhaps it is a little disingenuous to suggest that our theory of the universe is hard-wired, as it presupposes an entity or machine capable of not only setting up the initial conditions just so, but also of the 'infinite' processing power necessary to predict the outcome of running that universe.

I prefer to see the universe as just being exactly what it appears to be - it is incredible and mind-boggling enough as it is without needing to add an extra layer of woo.