Jump to content






Photo
* * * * - 1 votes

On the Cosmological Argument, Part 4

Posted by Michael S. Pearl, 20 March 2010 · 101 views

On the Cosmological Argument, Part 4


Continued from here

The Cosmology of Stephen Hawking


As discussed previously, the Hawking-Penrose theorems appear to be wholly compatible with there being a first context as per the Cosmological Argument, but Hawking's imaginary time-based “no boundary” condition for the space-time universe would seem to effectively do away with the Hawking-Penrose singularity which is something other than – or outside – the universe and is, thereby, consistent with the Cosmological Argument notion for a first context other than the universe itself.

The imaginary time, no boundary condition still posits a beginning to the space-time universe, but that beginning would seem to be a coming to be without there being any context other than the universe itself. Such a sort of beginning would effectively eviscerate the Cosmological Argument, because this type of beginning would have the universe as not only uncaused but also as coming to be without there being a sequentially prior context.

In fact, Stephen Hawking and James Hartle, authors of the Hartle-Hawking unconditional probability wave-function for the universe, have gone so far as to claim that, based on their model, the universe appeared uncaused and from nothing.

As it turns out, however, Hartle-Hawking does not appear to be consistent with that conclusion put forth by both Hartle and Hawking.

This is to say that even in light of Hartle-Hawking, the Cosmological Argument apparently succeeds. There is a first context in which the universe begins or comes to be. Furthermore, Hartle-Hawking provides a basis for describing or depicting something about at least one characteristic of this first context.

A brief foray into what has been be referred to as the Hartle-Hawking unconditional probability wave-function for the universe along with some consideration about the nature of events suggests an essential characteristic regarding the first contextinto which (or from which) the universe appeared. What is particularly interesting is that this characteristic of the Cosmological Argument first context might also be more substantially an aspect of space-time than the most conventional understandings about science can – or are willing to - accommodate.

According to Hartle-Hawking, there is a non-zero probability that the universe (space-time) came to be out of nothing. The problem with this claim is that it pertains as much to the nature of probabilities as it does to the alleged nature of the coming to be of the universe. All probabilities are possibilities, and, by their very nature, probabilities depend upon some sequentially prior state, condition, or context in order to provide for sequentially subsequent possibilities put forth in terms of probability. This is to say that the very claim of there being nothing prior makes the probability claim seem not only extremely dubious but even incoherent.

Much the same point is made by Graham Oppy14, who also quotes Hartle and Hawking as saying:

One can interpret the functional integral over all compact four geometries bounded by a given three geometry as giving the amplitude for that three-geometry to arise from a zero three-geometry, i.e. a single point. In other words, the ground state is the amplitude for the Universe to appear from nothing.



To this Hartle-Hawking claim, Oppy responds quite correctly, saying, “a single point is not nothing”. Quentin Smith seeks to counter Oppy by saying, in effect, that, this “single point” is most appropriately regarded as among the “timeless abstract objects ('mathematical spaces') rather than physical existents”15. Smith notes that in an attempt to rectify the problem of having apparently identified the single point with “nothing”, Hartle has subsequently written that “the 'nothing' is not realized as a physical state”, and Smith says that Hartle's and Hawking's “misleading statement about nothing being a physical state, a 'single point' should be omitted.” According to Smith, “Hawking also recently emphasizes that the universe 'would quite literally [come to exist] out of nothing: not just out of the vacuum, but out of absolutely nothing at all, because there is nothing outside the universe.'”

The fact is, however, that omitting Hartle's and Hawking's reference to “a single point” is not sufficient to produce a justification for the claim that the universe appears from “nothing”. All that Hartle and Hawking have done is insist: 1) that anything which “is not realized in a physical state” is “nothing”, and 2) that there are no physical states other than those which constitute the universe, but this still leaves the Hartle-Hawking model dependent upon an initial something – even if it is a strictly non-physical something -  in order for there to be any non-zero probability at all.

Smith says that this initial something is an abstract something; Smith would likely call it an “abstract object”. Yet, the only way that such abstract things can be not just Hartle's “nothing” but also Hawking's “absolutely nothing” is via the notion that only physical states (and/or their constituents) are things. One could then quite correctly say that the Hartle-Hawking model itself is based on “absolutely nothing”, but the most correct expression of the conclusion that Hartle and Hawking could actually derive from their model (assuming that model is adequate and accurate) would be the statement that the universe comes to exist from no other, or no prior, physical state or physical thing.

Hartle-Hawking provides no reason to deny that the universe follows from the Hawking-Penrose singularity even if Hartle-Hawking provides some (even if merely a semantic) basis for saying that there is “no physical law that … connect[s] the singularity to” the universe16. Accordingly, it may well be that the supposed nature of the singularity might provide some indication about just what characteristics can be expected of the thing upon which the Hartle-Hawking model is based.

As noted previously, the singularity can well be regarded as the intersection of “every past-directed spacetime path”; however, this intersection is only approached asymptotically17 via such a space-time regress inasmuch as the singularity, in addition to being this intersection, is also devoid of the laws of physics which constitute or describe space-time.

According to Hawking18, the singularity represents an actual break-down of “the classical concepts of space and time” which is to say the “laws of physics”; this break-down is regarded as mind-independently actual, because, as put forth, it is not a conceptual limitation resulting from ignorance. This means that, in essence, there are no restrictions on what can follow from the singularity19 and this is to say that the singularity is most correctly characterized as constituted by possibilities that are presumed to be mind-independent.

'On the Cosmological Argument' concludes with Part 5, 'The Nature of Possibilities'
--------------------------------------------------------------------------------------------------------------------------------------------------
14 Oppy, Graham, “On Some Alleged Consequences of 'The Hartle-Hawking Cosmology'”, Sophia, 36 (1997), 1, pp. 84-95; also available at http://www.infidels....ppy/smith1.html

15 Smith, Quentin, “Why Stephen Hawking's Cosmology Precludes a Creator”, Philo, Volume 1, No. 1, 1998, pp. 75-94; also available at http://www.qsmithwmu...s_a_creator.htm

16 Smith, Quentin, “Atheism, Theism, and Big Bang Cosmology”, Australasian Journal of Philosophy, March 1991, Volume 69, No. 1, pp. 48-66; also available at http://www.infidels..../cosmology.html

17 See http://plato.stanfor...l-argument/#4.4

18 As noted by Smith in “Atheism, Theism, and Big Bang Cosmology”.

19 Smith, Quentin, “The Uncaused Beginning of the Universe”, Philosophy of Science (1988), Vol. 55, No. 1, pp. 39-57; also available at http://www.infidels....h/uncaused.html





davidm
Mar 20 2010 04:07 PM
Before commenting specifically on parts three and four of this analysis by Michael, I just wanted to make a couple of other general observations about the cosmological argument. These comments may not pertain directly to the argument that Michael presents here, but they were inspired by reading his posts, which prompted a certain amount of reflection. I’ll comment specifically on the Hawking stuff a bit later. If one were to argue (and I realize this is not specifically Michael’s argument) that the cosmological argument provides at least some evidence for the possibility of a creator/God of the classically theist sort, a problem crops up, even if we accept that the argument in its strongest Kallam version has true premises and that the conclusion follows from the premises — a notion that I contest (see my previous post) but which we can accept here for the sake of argument. The basic idea is that everything that begins to exist is caused; and since the universe itself began to exist, it was caused by an uncaused first cause. Assuming this uncaused first cause (necessarily having always existed) is God, and that this God is the God of classical theism, then this God supposedly cares about human free will.But is free will itself actually compatible with the Kallam cosmological argument, assuming that that argument is itself successful? It seems not.For if it is true, as the argument holds, that everything that begins to exist has a cause, then a so-called “freely willed choice” cannot even begin to exist unless it has a sequentially prior cause. But if a “freely willed choice” has a sequentially prior cause, then it cannot really be a freely willed choice. It must be determined — determined by a sequentially prior cause. Hence there can be no free will.If, on the other hand, one wants to argue that freely willed choices really are free either because they are self-caused, or that they are uncaused effects, then the very first premise of the Kallam argument, that everything that begins to exist must have a cause, is invalidated.Of course, one might argue that a self-caused event is consistent with Premise 1 of the Kallam argument, because that premise merely states that everything that begins to exist must have a cause. It does not necessarily hold that the cause to every event must be sequentially prior to it, or outside of the effect itself. One could construe Kallam to admit of self-caused effects.But, in that case, if Kallam can be read to admit of effects that are self-caused, it seems the need for God (for those using Kallam to point to God) is vitiated. For if effects can be self-caused, and do not require a cause outside of themselves or sequentially prior to them, then plainly the universe could have been self-caused. Of course, an uncaused effect would also invalidate the standard cosmological argument, but also would not comport well with free will. For an uncaused effect, as applied to free will, is no better for free will than determinism. For while determinism holds that free acts aren’t really free because they are pre-determined by a chain of prior events (stretching back to the beginning of time?), an uncaused effect would by definition happen for no reason at all. That is, it would be random, and if you act randomly you are no more exercising free will then you would be if your acts were actually pre-determined. This concern goes back at least to Hume.And this is why (as I tried to indicate in my previous post on this subject) that God might well have created, not a universe directly, but a probability wave function in which the universe with its known properties had a 95 percent chance of appearing. I argued that God, in creating a probabilistic universe, was leaving wiggle room for free will. On further reflection, though, I’m not sure this argument holds, for it must be remembered that under quantum theory, events are probabilistic but nevertheless are intrinsically random in their occurrence. And if this randomness applies to the acts by humans that we would like to believe are freely willed, then we are right back to the problem described just above: randomness is no better for authentic free will than is determinism.The upshot is: it seems that for any act to be authentically freely willed, it must be self-caused. It cannot be caused by a sequentially prior event, or by an event outside itself; and it cannot just happen randomly; i.e. for no reason at all. If authentic free will requires these conditions, than both the Kallam cosmological argument and quantum mechanics are incompatible with human free will. So the theist who hopes to vindicate the classical conception of God by appeal to the Kallam argument seems to left with the conundrum that if his appeal works, human free will is ruled out. But since human free will is traditionally thought to be among the aims of the classically theistic God, that’s a real problem for the theist who uses the Kallam argument. Kallam and free will do not seem compatible.There is yet a further issue: What, exactly, is cause and effect? Michael has touched on this, but for now I’ll just observe that if eternalist conceptions of the universe are correct, then cause and effect (just like change and, for that matter, motion itself) seems to be an illusion. If the whole universe, extended in space and time, just is — if it is never becoming, but merely in an eternalist state of being — then it makes no sense to say that something causes something else to happen. For something to cause something else to happen, then the universe must be in a condition of constant change: That is, it must be in a state of becoming. But the eternalist conception has no place for “becoming.”Those are some thoughts inspired by these blog entries. As I say, I’ll talk more specifically about the Hawking interpretation later.

Michael S. Pearl
Mar 20 2010 05:12 PM
davidm, even if there is no becoming in an eternalistic universe, is there a sequence to what is?Michael

davidm
Mar 20 2010 05:33 PM
Well, this is an interesting question.Yes, there is an apparent sequence. But what conclusions or inferences should we draw from this?If you consider a simple act like striking a match, we would say that the striking of the match causes the fire. We even understand the underlying physics/chemistry of how and why this occurs.But if you now take the untensed “block world” theory of space time we get a series of events, which for the sake of simplicity we can boil down to three. Time 1: Match is struckT2: Chemical reaction occursT3: Fire ignites.And I’m saying, if we should happen to take the ontology of Minkowski spacetime literally, and if we believe the Times 1, 2, and 3 all exist independently of one another — are all “ontologically on a par,” in the same way that three different locations all exist (just not at the same place; that’s why they are different) — then it is very difficult to see exactly how the event at T1 causes the event at T2, and how, in turn, the event at T2 causes the event at T3. These events, though sequential in the sense that they instantiate a logical order, also seem (if Minkowski ontology is literally true) completely independent of one another. These events can’t cause the succeeding event, because the succeeding event, like all events, timelessly exists. And my intuition is that for events to cause other events, the sequentially prior event must cause the successive event to be brought into existence. But if all events in spacetime actually exist, this condition for causation cannot be meant. Hence it might be that cause and effect is an illusion; the Kallam argument would fail because: nothing ever begins to exist. Everything just does exist, at its specified location in spacetime.

Michael S. Pearl
Mar 20 2010 05:42 PM
davidm, instead of speaking in terms of causes, if we speak in terms of a type of necessity, we could say that A or A & B, for instance, are necessary for C to be as it is (including where it is in the sequence).So, given such a manner of speaking without reference to causes, do you see any problems with there being sequence in an eternalistic universe?Or, to put it another way, do you think that there are mind-independently actual sequences in the proposed eternalistic universe.Michael

davidm
Mar 20 2010 06:11 PM
Yes, well, this is where I find severe difficulties, either ontologically or epistemically or both.I do not understand what it means to say that either A, or perhaps A & B, are necessary for B to be as it is, including where it is in the sequence. Isn’t this just another way of saying that A + B causes C to occur (including its occurrence in the specific place in the sequence)?If it’s saying something different, I’d like to know what it is.This problem is very unsettling. If the eternalist version of reality is true (and I’ve argued elsewhere, in the good company of many philosophers and scientists, including Einstein himself it would seem,  that modern physics gives us good grounds to believe in eternalism) then I really can’t see how anything is necessary for anything else. If, as we speak, for instance, events in the year 3030 exist, then how does anything I do now influence what just does exist in 3030? And how, to get closer to the cause/effect riddle, can anything in 3029 influence what just does exist in 3030?The other riddle is, if the eternalist picture casts doubt on cause and effect, then why do we observe these logical sequences at all; sequences that seem to validate cause and effect? And then, if this question is valid, perhaps it’s the same sort of question that makes eternalism so counterintuitive: if eternalism is true, then why does time seem to pass, when in fact nothing is passing at all? And the answer to one question might provide the answer to the other, if there is an answer to be had.Now, the real question might be as you put it: Are their actual mind-independent sequences? Perhaps sequentiality itself is a kind of illusion or a misinterpretation of  reality?It should be noted that sequentiality seems to depend crucially on the so-called arrow of time, the impression (even if illusory) that time is ordered sequentially, from past to future. And there are a number of physical reasons given for this arrow: the second law of thermodynamics, for instance, and also the expansion of the universe. But then again, it must also be noted that at the microphysical level, this arrow, or these arrows if you will  (there are about seven or eight of them identified) simply disappear, and the microworld is held to be time symmetric. In addition, as I understand it, some constructions of quantum mechanics allow for actual backward-in-time causation.How to bring all these ideas together into a coherent interpretative whole is rather daunting, but I should mention in passing the views of the physicist Julian Barbour, who holds that time does not exist at all. He claims that what we call sequences of events, and the arrow of time, and indeed history itself, are explained by so-called time capsules, in which “observers” just happen to have, via quantum mechanical explanation, memory “traces” of a linear history that did not in fact exist. That’s how I understand Barbour’s argument. It sounds like a physicist’s version of Last  Thursdayism.

Michael S. Pearl
Mar 20 2010 08:19 PM

davidm, on 20 March 2010 - 07:11 PM, said:

This problem is very unsettling. If the eternalist version of reality is true (and I’ve argued elsewhere, in the good company of many philosophers and scientists, including Einstein himself it would seem,  that modern physics gives us good grounds to believe in eternalism) then I really can’t see how anything is necessary for anything else.
Would it be apt to describe the noted unsettledness as epistemic? If so, doesn't this sort of unsettledness generally indicate an inadequacy in how we are thinking -- and speaking, since thinking and speaking are so tightly intertwined? Epistemic or not, it might be worthwhile to see whether there are alternative ways of expressing what we think. For instance --

davidm, on 20 March 2010 - 07:11 PM, said:

If, as we speak, for instance, events in the year 3030 exist, then how does anything I do now influence what just does exist in 3030? And how, to get closer to the cause/effect riddle, can anything in 3029 influence what just does exist in 3030?
Within an eternalistic context, the term influence suggests a dynamism which is not (supposed to be) actual. A less dynamic term than influence could be some form of depend, and the above could be reconsidered in terms of whether any "events in the year 3030" in any way depend on "anything I do now".

davidm, on 20 March 2010 - 07:11 PM, said:

I do not understand what it means to say that either A, or perhaps A & B, are necessary for [C] to be as it is ... Isn’t this just another way of saying that A + B causes C to occur (including its occurrence in the specific place in the sequence)?
I would say that speaking in terms of either the sort of necessity previously presented or in terms of dependence as described in this response -- rather than speaking in terms of causes -- seems at least expressively more compatible with an eternalistic universe lacking actual dynamism as well as becoming. Even so, there is still the following matter which you raised:

davidm, on 20 March 2010 - 07:11 PM, said:

Perhaps sequentiality itself is a kind of illusion or a misinterpretation of reality?It should be noted that sequentiality seems to depend crucially on the so-called arrow of time, the impression (even if illusory) that time is ordered sequentially, from past to future ... at the microphysical level, this arrow, or these arrows if you will (there are about seven or eight of them identified) simply disappear, and the microworld is held to be time symmetric. In addition, as I understand it, some constructions of quantum mechanics allow for actual backward-in-time causation.
Note that the alleged symmetry of time does not eliminate sequentiality -- meaning that there is some sort of dependence even if there is some sort of perspectival difference. This, in turn, means that, despite any epistemic unsettledness that may come of thinking in terms of these as-yet-unexperienced alternative perspectives, the one thing we are left with is sequentiality, which itself indicates ordered-ness and not randomness.

davidm, on 20 March 2010 - 07:11 PM, said:

physicist Julian Barbour, who holds that time does not exist at all. He claims that what we call sequences of events, and the arrow of time, and indeed history itself, are explained by so-called time capsules ...
The thing is that so long as these "capsules" (about the only things to which Barbour seems not to deny existence - Heh) present in sequences (which is to say with order), so long as science does not see sheer randomness, then it seems that we and (metaphysical) science must at least admit to there being mind-independent sequences of dependent relations; there are actual sequences, and there is actual dependence.That being the case, then with regards to (let's just call them) scientific cosmologies which put forth (as covered in these CosmoArg blogs) a beginning upon which the rest of the universe is dependent relationally (even if there is no time passage), then the main points of the Cosmological Argument still seem not only legitimate but also basic, we might say even essential, and always present (even if in the background).Or, to put it another way, speaking in terms of dependence, context, and sequence appears to be invariantly viable.Michael

davidm
Mar 22 2010 02:19 AM
Well, I’ll have to come back to this in greater depth. For now I’ll say: we do, no doubt, observe sequence and context. Whether we also observe something called dependence seems open to challenge.On a side note, this eternalist picture of the universe was the subject of a lot of discussion here with respect to free will. I took the position that free will was possible even in an eternalist universe, a position that may be hard to defend, but I think it is defensible. And a successful defense of free will in a universe that is eternalist might also amount to a successful defense of dependence, or even of cause and effect, in such a universe.The problem of free will in an eternalist universe seems to be similar to the problem I’ve noted here of cause and effect. If everything just is, how can anything change? The gist of my defense, which I’ll just allude to here, is that if eternalism is true, there is a timelessly true description of all that has ever happened, or will ever happened (we just don’t have access to it.) This description is made to be true by the acts, including the free acts, that happen within the world. This reverses the usual concern: which is that if there is a timelessly true description of all that happens, has happened or ever will happen, that this description forces these things (including human acts) to happen. This intuitive idea, which lies at the root of fatalism (including logical fatalism) gets the truth-making flow backward, as I’ve argued. This is why, for instance, logical fatalism, going back to Aristotle, can be shown not to be a concern.I think this idea has some bearing on causation, or dependency and context, if you will, which I’ll try to flesh out later.

Michael S. Pearl
Mar 22 2010 09:09 PM

davidm, on 22 March 2010 - 03:19 AM, said:

For now I’ll say: we do, no doubt, observe sequence and context. Whether we also observe something called dependence seems open to challenge.
I am interested in how the notions put forth about dependence are "open to challenge". After all, this dependence has been discussed in terms of things (or even events) being as they are (including whether in time or simply in sequence) because of, or relative to, or depending on other things (or events). Even in terms of what we might call a reverse-sequence perspective (otherwise referable as a backwards arrow of time, probably even backwards causation), the relative sequential positions do not really change. That is to say that instead of describing anything from just one perspective, a fuller, more correct or accurate description would be one that included the multiple perspectives. As cumbersome as that would be, such a mode of presentation would make it quite clear that a thing is at least descriptively dependent on other things for its being as it is and/or is where it is sequentially.

davidm, on 22 March 2010 - 03:19 AM, said:

The problem of free will in an eternalist universe seems to be similar to the problem I’ve noted here of cause and effect. If everything just is, how can anything change? ... I think this idea has some bearing on causation, or dependency and context, if you will, which I’ll try to flesh out later.
I think that one way in which an eternalist account might account for the possibilities which are necessary for there to be free will is to regard possibilities as or in terms of event termini (discussed in Part 5, if I recall correctly) and to regard these possibilities as veritable bridges to dimensions which are in addition to the fourth dimension (time). If things are to be located or identified with four dimensions, and no two non-identical things can have precisely and only the same four dimensions, then an alternative which shares the first four dimensions must have some other sort of dimension in order to distinguish it. This would be a way in which one could put forth and describe an eternalist multi-verse.But, all that aside, I would like to know whether (and how) an eternalist account of the universe (meaning, of course, the physical universe, since one claim of support for the eternalist universe is relativity physics) is compatible with Hawking-Penrose and Hartle-Hawking, especially as discussed in this blog series.Michael

Big Blooming Blighter
Mar 22 2010 10:15 PM

Quote

For if it is true, as the argument holds, that everything that begins to exist has a cause, then a so-called “freely willed choice” cannot even begin to exist unless it has a sequentially prior cause. But if a “freely willed choice” has a sequentially prior cause, then it cannot really be a freely willed choice. It must be determined — determined by a sequentially prior cause. Hence there can be no free will.
I'm not sure how you came to such a conclusion, particularly given your advocations of Compatibilism in the past, David.In any case, the first-cause issue does not pose a threat to believers in free will, whether or not they accept Compatibilism. A quick and simple way out is simply to realise that a single cause can have multiple effects (which, of course, would be the answer of Cosmological Argument proponents to the multiverse theory.) Hence, while our making a choice at all may be determined, it is still ultimately not evident that the choice made is determined.

Big Blooming Blighter
Mar 22 2010 10:47 PM

Quote

There is yet a further issue: What, exactly, is cause and effect? Michael has touched on this, but for now I’ll just observe that if eternalist conceptions of the universe are correct, then cause and effect (just like change and, for that matter, motion itself) seems to be an illusion. If the whole universe, extended in space and time, just is — if it is never becoming, but merely in an eternalist state of being — then it makes no sense to say that something causes something else to happen. For something to cause something else to happen, then the universe must be in a condition of constant change: That is, it must be in a state of becoming. But the eternalist conception has no place for “becoming.”
The Eternalist does not have to deny the concept of 'becoming' in order to be consistent with his/her position as an Eternalist.The A-theorist, for example, might propose that 'becoming' represents the processes that bring future events into the present (since, according to the A-theory, there is a non-indexical present, and time flows. This is entirely compatible with Eternalism), or actualises future potentials.For the B-theorist, the problem is not so easily solved. For one thing, the B-theorist cannot contest that future states, even if existent, are ontologically distinct from present states. It would be interesting to see a B-theorist response to your 'becoming' charge, David.One thing the B-theorist might argue is that, if the first cause is beyond space-time, or Block time, then it does not matter that there is no becoming in a B-theoretic universe, since the incompatibility of becoming and Block time only has ramifications for things contained within the Block time.

← May 2013 →

S M T W T F S
   1234
567891011
12131415161718
19202122 23 2425
262728293031 

Categories

  1. The Galilean Library
  2. → Community Blog
  3. → Michael S. Pearl's Blog
  4. → On the Cosmological Argument, Part 4
  5. Privacy Policy
  6. Site Rules ·