Cosmic fine tuning

[AndyT_81]

Are you thinking of a modern example where cardinality is considered more a threat to parsimony than multiplication of types? I can’t think of any. I won’t say it doesn’t happen, but it’s foreign to my experience if it is used that way. Can you think of an example?

The kind of Occam’s Razor formulation that I was considering as “modern” is probably a less formal type. For instance, I was thinking something along the lines of Richard Swinburne’s “definition” or formulation of the Razor:

“…other things being equal – the simplest hypothesis proposed as an explanation of phenomena is more likely to be the true one than is any other available hypothesis, that its predictions are more likely to be true than those of any other available hypothesis, and that it is an ultimate a priori epistemic principle that simplicity is evidence for truth”

Such a definition doesn’t seem to inherently include concerns regarding cardinality and multiplication of types. Rather it seems to be aiming at a more intuitive understanding of complexity or simplicity. However, marrying this with a formal tool such as K-Complexity was probably a blunder on my part! (see below)

Allright, we have a problem at this point. Kolmogorov Complexity has no idea how to measure such a description — “how a life friendly universe could come about”. Maybe you are thinking of another form of formal description? This leaves me scratching my head. But you can bank on this (go Google it if you’re dubious): K-Complexity is the measure of the shortest program needed to perfectly recreate the a universe at any given “snapshot” in time (and yes, that’s problematic itself, as there is no such thing as “instantaneous” across space that would make sense of the word “snapshot”). Your concern with describing “how” and “why” represent important questions, but are totally unrelated to Kolmogorov complexity.

Ok, yes I understand your issue now. I was mis-interpreting the formality required by K-Complexity. I was attempting to use the principle behind K-Complexity to measure the complexity of the explanation or hypothesis by the amount of information it contained, not the complexity required to reproduce that which the explanation or hypothesis was attempting to explain (i.e. the universe). Thanks for picking this up.

I won’t address the rest of your comments because I think the problem with my K-Complexity approach has now been identified. It probably is too formal a tool for the use I was looking for. Anyway, I look forward to any additional comments you may have.

 

[Touchstone]

Hi AndyT,

No, I wasn’t thinking of entities that way. I was considering the universe generation mechanism as a new type of entity. This is perhaps an incorrect way of thinking of things though with Smolin’s idea of the multi-verse, i.e. universe genesis from black hole singularities. However for other theories such as Linde’s eternal inflation model, whereby individual pocket universes nucleate out of this eternally inflating region, I think it would be a stretch to say that this eternally inflating mechanism is somehow of the same type as this universe.

Oh, I’m sure you’re right. My point was that a “self-similar”, or “recursive” model stands up there, too, and in terms of parsimony, is type-economical in a way that the God hypothesis is not, and even other meta-cosmologies are not. Some of the models that are conjectured which are not “fractal” in nature nevertheless make their new types out of the “raw materials” of physics of this universe; quantum fluctuations, as the “mechanism” for uncaused or uncorrelated events, for instance, draws upon a familiar, empirically solid understanding of our own universe. An “outer context” whereby universes “pop” due to quantum perturbations is just an extension of what we understand locally to be a fundamental dynamic of our reality.

We can’t test that, in the direct, empirical sense, but the “metaverse” built with that idea is extruding known, observed principles into that outer context, rather than positing something alien to it. Those other types look more… natural, because they are built from concepts drawn from nature. That may be incorrect, but doesn’t offend the explanatory economy the way a “personal God” does.

I suppose you could say the God hypothesis is similarly made out of “natural materials”, in the sense that we know humans design and create things, and the God hypothesis is a bit of “metaphysical anthropomorphization”. That places primacy on mind however, on the will, on a subjective reality, rather than an objective one. That’s OK, just something quite apart from what I’ve understood “natural” or “physical” to mean, here.

Well, I would again say that certain multi-verse models do contain mechanisms which are arguably of different type than this universe, such as an eternally inflating region. I would also disagree that God, at least in the traditional way of understanding Him, is a completely new type, but more on this later.

Agreed. My point was that some don’t, and suppose little, if any, type revisions to what we have for our local universe. Just that as a hypothesis, even if its not the only one (and as you point out, and I affirm, it’s not), that seems a problem for the God hypothesis in terms of parsimony. It doesn’t have to be proven or established, but merely capable of qualifying for “all other things being equal”. Since metaphysics is inscrutable that way, I think we have to admit it’s just as plausible as the God hypothesis, or any of the other conjectures. But it’s also more parsimonious, just do its ontological simplicity - the way a fractal set is simple, compared to a random string.

Ok, but that is only one multi-verse model (and as far as I can gather, not the most popular amongst physicists).

That’s right, and if you ask physicists, parsimony as the recommendation mechanism doesn’t get a lot of support. Parsimony is a heuristic, and a valuable one, but it’s predicated on some empirical underpinnings that obtain in our local experience, but we have no reason to expect of a metaphysic. I remember Brian Greene being asked about this once in a Q&A after a speech I attended, and his reply was halting: *why should we think parsimony is useful at all in thinking about metaphysics?
*
That’s a powerful, profound question. We apply the principles of economy in thinking about natural explanations, because we have a wealth of emprical evidence that nature has a strong disposition toward economy – see the Principle of Least Action, for example. But that’s nature, not meta-nature. It’s a mistake to just assume that parsimony is useful at all in thinking about metaphysics.

Which, when you think about it, may be a good thing for the God hypothesis!

At any rate, if parsimony is your yardstick, it’s hard to see how the God hypothesis competes well.

continued…

 

[Touchstone]

… continued…

I’m not sure how this follows, perhaps there is a misunderstanding here. My argument is that God is not a completely new type, and shares characteristics with existing types, i.e. other persons, such as a free will, the ability to cause things freely, an intellect etc. It is not logically necessary that whatever mechanism that produced the universe be ontologically self-same with said universe, but it doesn’t then follow that whatever caused the universe (which isn’t ontologically self-same) cannot be of a similar type of entities within said universe.

Those isomorphisms may well exist; I suggest that the concepts of “free will” for an omnimax God and “free will” for a human being are much more dissimilar than similar, but won’t drag the discussion down that rabbit hole just here.

But whatever similarities or characteristics you may connect, there’s this “universe making faculty” to address, and that’s the big kahuna. That’s a capability that seems quite foreign to humans, something not distilled in us as part of our imago dei. God is impassible that way, at least as I understand Catholic/Orthodox/Christian thinking.

I’m not sure if God could be classed as an entirely new type, of the form “a new, fundamental, and yet unobserved, unadmitted entity”, as He shares significant characteristics with an existing type, namely, persons. While God may be “more” of a new entity than similar universes in the recursive structure type of multi-verse, I wouldn’t consider Him to be “more” of a new entity than the eternal inflation regime in other multi-verse theories.

more to follow…

You can take a pan(en)theist line here; if so, then you are right, God becomes ontologically blurred with his creation, the further you push on that axis. My understanding in this conversation is that God is ontological distinct from his creation; he may be immanent, but being immanent is not being the thing. To the extent that God is distinct from his handiwork, I think that necessarily introduces a new fundamental type of entity. If not, it’s de facto pantheism – God as “Mama Bear” giving birth to our “Baby Bear” universe, instance of type arising from another instance of that type. I can’t think that’s where you’re heading with this, but maybe so?

-TS

 

[Touchstone]

The kind of Occam’s Razor formulation that I was considering as “modern” is probably a less formal type. For instance, I was thinking something along the lines of Richard Swinburne’s “definition” or formulation of the Razor:

Such a definition doesn’t seem to inherently include concerns regarding cardinality and multiplication of types. Rather it seems to be aiming at a more intuitive understanding of complexity or simplicity. However, marrying this with a formal tool such as K-Complexity was probably a blunder on my part! (see below)

I would say that the assessment of cardinality and multiplication of types inheres in phrase “other things being equal” from Swinburne’s quote. I understand the idea that “all explanations are equally unverifiable”, but I don’t think that’s what Swinburne means, or if so, I think it’s an error on his part. The “bill of materials” required for an explanation is what that “other things” refers to.

As an example, if we have a couple of competing hypotheses addressing some phenomenon, and Hypothesis A offers this very complex set of interactions between gravity and matter, and Hypothesis B offers a simpler set of maths, but invokes “dark matter” as a key element of the explanation, not only is Hypothesis B not at parity (“other things being equal”) with Hypothesis A, or in an advantageous position, it’s in a hole – assuming that “dark matter” is a new principle, something we do not have in our settled lexicon of explanatory resources.

Hypothesis A is doing an intricate ballet (not simple), but the characters involved are “off the shelf”. Hypothesis B brings a “hey-we-aren’t-comfortable-that-dark-matter-is-coherent-here-let-alone-effective” problem to the table.

Looking at A and B, if we just wave our hands and say they are “equal”, then B becomes the more attractive. But, at least in terms of science and rational analysis, that is fallacious – “other things” are NOT equal, in that case.

Ok, yes I understand your issue now. I was mis-interpreting the formality required by K-Complexity. I was attempting to use the principle behind K-Complexity to measure the complexity of the explanation or hypothesis by the amount of information it contained, not the complexity required to reproduce that which the explanation or hypothesis was attempting to explain (i.e. the universe). Thanks for picking this up.

That’s a very hard problem you are addressing, then. Kolmogorov is oppressively boring by comparison, because it’s so formal – just an enormous amount of work. But what you trying to capture, somehow, and in an informal fashion, is a “score” for complexity, based on… narration. That seems to preclude any answer but God, because an omnimax God is the theoretically simplest explanation – in informal terms – for anything. I don’t think you are trying to “rig” things with language tricks, which is why I have said what I did.

Consider this: a bunch of ancient Greeks are musing about the cause of thunder, and the hypothesis put forward by Alexander of them is “the gods are bowling!” (nevermind the anachronism of “bowling” in ancient Greece, please!).

Now, what chance does Brutus have with “Well, I think it has to do with wind, and air temperature, and convection, and electricty, and this idea I have about something I call ‘evaporation’…” against Alexander’s explanation?

In informal terms, Alexander’s idea (“the gods are bowling”) is invincibly simple, compared to any “meteorological” explanation Brutus or anyone else might offer. But that “simplicity” is deceptive… a kind of… cheating on Alexander’s part, at least so long as we understand that these Greeks can’t empirically validate these “gods” or activities we might call “bowling” in the skies.

Parsimony only works when the competitors pass some level of diligence in terms of substantiating their explanatory resources. That is why Kolmogorov would be very powerful if it could be applied. It’s totally impractical, but I think you can see that having to provide a formal description of God – the ability to encode God so that He can be perfectly reproduced by a replay – demonstrates the maximal complexity of God, in formal terms.

I won’t address the rest of your comments because I think the problem with my K-Complexity approach has now been identified. It probably is too formal a tool for the use I was looking for. Anyway, I look forward to any additional comments you may have.

Fair enough! It’s totally impractical to apply, anyway, right? That seems a good reason to try to find another approach to this question. But whatever you end up coming to here, I wonder how you expect to ward off the “the gods must be bowling” trap, the problem that arises from “God” being just a simple three letter word, but a proxy for a maximally complex and intractable resource to invoke in any explanation?

Thanks for the high quality thought and comments you invested in this conversation.

-Touchstone