In my previous posts about the Carroll-Craig debate, I've been skirting around the edges of an important claim by Carroll which summarizes the main reason he doesn't buy the Cosmological Argument. That's because I was focusing on the question of whether the universe had a beginning. Now I want to grapple with his more philosophical claim.
Carroll says this:
So, I think I can make these points basically by following Dr. Craig’s organization starting with the kalam cosmological argument, and unlike what he said I should be doing I want to challenge the first of the premises: If the universe began to exist it has a transcendent cause. The problem with this premise is that it is false. There’s almost no explanation or justification given for this premise in Dr. Craig’s presentation. But there’s a bigger problem with it, which is that it is not even false. The real problem is that these are not the right vocabulary words to be using when we discuss fundamental physics and cosmology. This kind of Aristotelian analysis of causation was cutting edge stuff 2,500 years ago. Today we know better. Our metaphysics must follow our physics. That’s what the word metaphysics means.
[Well, many scholars think that Aristotle's book on causation/God was called Metaphysics because it was the book immediately following his Physics in the traditional order of the canon. But perhaps Carroll knows this and is simply introducing his thesis in a joking way...]
And in modern physics, you open a quantum field theory textbook or a general relativity textbook, you will not find the words “transcendent cause” anywhere. What you find are differential equations.
[Of course, because QFT and GR concern the interactions between various material entities (excitations of fields). Their interactions (i.e. their causal relationships) are indeed described by differential equations. And therefore the subject matter does not concern transcendental or ultimate causes, any more than economics or psychology textbooks discuss the weak force. What he really ought to say to make his argument, is that the term "cause" does not appear in many modern physics textbooks (apart from the use to mean "logically implies") due to its lack of usefulness in specifying the precise form of the laws of physics.
It is, however, a separate question whether concepts of causation are implicit in the fact that some particular differential equation holds. It does not seem outrageous to state that the state of the fields at a given point are caused by the state of the fields just beforehand in the past lightcone; indeed this concept is called causality in contemporary physics. This concept may indeed be different in some ways from traditional philosophical ideas of causality, but it is not sheerly different, and there is room for argument about the precise relationship between these ideas of causation. Carroll continues:]
This reflects the fact that the way physics is known to work these days is in terms of patterns, unbreakable rules, laws of nature. Given the world at one point in time we will tell you what happens next. There is no need for any extra metaphysical baggage, like transcendent causes, on top of that. It’s precisely the wrong way to think about how the fundamental reality works. The question you should be asking is, “What is the best model of the universe that science can come up with?” By a model I mean a formal mathematical system that purports to match on to what we observe. So if you want to know whether something is possible in cosmology or physics you ask, “Can I build a model?” Can I build a model where the universe had a beginning but did not have a cause? The answer is yes. It’s been done. Thirty years ago, very famously, Stephen Hawking and Jim Hartle presented the no-boundary quantum cosmology model.
[See here for my discussion of the claim that the Hartle-Hawking model leaves no room for a Creator.]
The point about this model is not that it’s the right model, I don’t think that we’re anywhere near the right model yet. The point is that it’s completely self-contained. It is an entire history of the universe that does not rely on anything outside. It just is like that. The demand for more than a complete and consistent model that fits the data is a relic of a pre-scientific view of the world. My claim is that if you had a perfect cosmological model that accounted for the data you would go home and declare yourself having been victorious.
Carroll seems to be saying that the ability to construct a mathematically precise and consistent model of the universe is the only important criterion for the fundamental laws of physics, aside from fitting the data. Presumably he would add something about simplicity if we were considering two rival models which both accounted for the same data, but I think he doesn't mention that since his primary concern is saying what counts as a possible model according to the contemporary set of informal procedures that cosmologists use, rather than asking how we decide between models once we have them.
This is probably the most important claim that Carroll made during the debate. I appreciate the fact that he was able to state his view in such a clear and accessible way. Now, there is one important point in which I agree with Carroll, and one important way in which I disagree.
I agree that Modern Science has involved some shifts in how we think about causation in the physical universe. This shift, which occurred sometime around the beginning of the 20th century (give or take a couple hundred years) involved a move away from mechanistic notions of causation, towards more abstract mathematical models. As a characteristic example, in the 19th century it was presumed that light waves couldn't exist unless they were supported by a medium, called the luminiferous aether. When Einstein developed special relativity, he showed that no such aether was necessary. If you can write down a satisfactory equation for the way the electromagnetic field propagates, there's no need to ask questions like "In what medium does it propagate?" (Of course, this philosophical point could have been made in the time of St. Maxwell as well, but it wouldn't have been as compelling then, since before Einstein's theory of relativity an aether was needed to preserve the principle of relativity).
Another example concerns normal forces of the sort that prevent two solid bodies from occupying the same space. A premodern physicist would have said that solid bodies have a property of solidity or massiness which, using our everyday intuitions, prevents two bodies from occupying the same position. In Lucretius' atomic theory, the atoms were in solid shapes like billiard balls (but not necessarily round) which excluded each other due to their solidity. Very different was St. Roger Joseph Boscovich's 1763 proposal that atoms are merely mathematical points exerting force on other mathematical points.
In modern physics, there's nothing inherently wrong in general with two objects occupying the same position, since you can write down equations in which this is allowed. The default, in the absence of interactions, is for two bodies to pass right through each other. That this does not occur for real solid bodies is due to a combination of electromagnetic forces and the Pauli exclusion principle for electrons.†
So I agree with Carroll that there have been changes in our attitude to causality in Physics. Now let's see whether this should change our views about Metaphysics. As a reminder, Metaphysics is the branch of philosophy which studies the fundamental nature and relationships between various beings (abstracting from the particular descriptions of the particular things as studied by more specialized fields like Physics or Biology). It asks questions like these:
Do individual things have a particular essence or nature which does things?
In what sense does causality exist, and if so what kinds are there?
Is there really a physical universe?
Are there really minds?
Can everything be reduced to physics, or can everything be reduced to mental properties, or are these two independent realities, or are they both dependent on something else?
In what sense do mathematical entites like "7" or "isoscoles triangles" exist?
In what sense does morality exist?
Does some type of God exist, and if so what is he/she/it like?
What are the relationships between these different types of existence?
Do the "holes" in doughnuts really exist, or are they merely the absence of doughnut?
Are the questions I just asked even meaningful, or are some of them silly word games?
(The subset of questions about what exists are also sometimes also called Ontology. In case you're wondering, analytic philosophers really do debate the one about holes. Personally I consider the one about God to be more important. But I figured that given my last question, I needed to include at least one question which I consider to be a silly word game...)
Now, these questions raise some difficult and contentious issues, and philosophers have been debating them for thousands of years. But rather then tackle them head-on, let's ask to what extent developments in Physics and other Sciences help answer these questions.
There are two opposite extreme approaches one might take:
1. One is the view that our Metaphysics should essentially follow immediately from our Physics. Once we learn from experiment what specific Laws of Physics are true, that tells us everything we need to know about Metaphysics. People with this view tend to be scientistic and dismissive of philosophical arguments. (You might think from the quotation above that this is Sean Carroll's view, that once we work out the correct physical model there are no more interesting questions to ask. But in fact his view is more subtle, since he acknowledges that there are interesting philosophical questions not obviously resolved by Science.)
2. The opposite extreme would be to say that developments in Physics have little or no bearing on Metaphysics. For example, there are modern day followers of St. Thomas Aquinas (typically but not always Roman Catholics) who believe that something similar to Aristotle's metaphysical views follow necessarily, given that the world is rationally explicable at all, and that the very possibility of doing Science depends implicitly on accepting them. In this view, the basic concepts of metaphysical reality such as substance/accident or act/potency are written in stone, although identifying the particular "substances"‡ which actually exist, and their properties, depends on empirical Science. One such philosopher is St. Ed Feser, whose blog is on my sidebar. (I recently read his book Scholastic Metaphysics in which he presents his arguments for this position. On a first pass, I found it serious and interesting but ultimately unconvincing, though giving my reasons would take this post too far afield.)
But it seems to me that the correct view is in the middle, that Physics has some bearing on Metaphysics but it doesn't fully determine it. Physical models and metaphysical views have, not a 1-to-1 relationship, or even a 1-to-many relationship, but a many-to-many relationship!
People who agree on the Laws of Physics (to the extent that we are able to discover them at present) can still have radical disagreements about Metaphysics. Of these metaphysical views, some seem irrational and silly, like the Monadic view that all distinctions are illusory and that only one thing exists. But it seems to me that there are several possible reasonable (i.e. non-crazy) views. So even after we agree on a single physics model, multiple metaphysical views can be reasonable. The variety of interpretations of Quantum Mechanics is one prominent example. Choosing between these interpretations requires philosophical arguments; doing an experiment is not enough.
On the other hand, it seems rather unlikely that the philosophical considerations by themselves are sufficient to pin down an exact Metaphysics either. That would be to say that every single view about the nature of Being, besides one, can be decisively refuted by philosophical arguments. That would seem to me almost as surprising as a claim that the Laws of Physics can be deduced from pure Reason. There are just too many possible sets of belief, and logical consistency plus noncraziness is too weak of a constraint. (Especially since our notion of noncraziness needs to be flexible enough to include things like interpretations of QM and the mind-body problem, where nearly all views seem "crazy" in one way or another.)
Given that there are multiple possible views about Metaphysics, we need all the help we can get to choose one. So it would be absurd not to rely on Physics to some extent, especially as it impinges on relevant questions. How could anyone have an informed position on the nature of time without thinking about what General Relativity says about spacetime? But we also need to recognize that Physics is not enough, and that abstract arguments about what "makes sense" are also called for. Even together these are not sufficient to bring all reasonable Philosophers into agreement.
That means that Carroll is moving much too quickly when he says that "This kind of Aristotelian analysis of causation was cutting edge stuff 2,500 years ago. Today we know better. Our metaphysics must follow our physics."
If there are multiple possible metaphysical views, then we can't just dismiss the views of Aristotle (or whoever) just for being old, without carefully thinking about their compatibility with Modern Science. Maybe people switched to modern views because of updates in physical knowledge (some of which, like Determinism, may themselves be out of date!). Or maybe they changed because of philosophical arguments, in which case we need to check whether those arguments are good or bad. Or maybe the old views simply became unfashionable. Most likely, it was some combination of these four causes, and we need to be careful! As St. Lewis writes concerning Owen Barfield:
In the first place he made short work of what I have called my "chronological snobbery", the uncritical acceptance of the intellectual climate common to our age and the assumption that whatever has gone out of date is on that account discredited. You must find out why it went out of date. Was it ever refuted (and if so by whom, where, and how conclusively?) or did it merely die away as fashions do? If the latter, this tells us nothing about its truth or falsehood. From seeing this, one passes to the realization that our own age is also a "period", and certainly has, like all ages, its characteristic illusions. They are the likeliest to lurk in those widespread assumptions which are so ingrained in the age that no one dares to attack or feels necessary to defend them.
While I don't accept St. Feser's arguments that Aristotelianism is compulsory, I think that studying it is a useful antidote to thinking that there's only one possible view. Certain aspects of modern thought might be unreasonable inheritances from the Enlightenment, or from Positivism. The easiest way to notice that we are making unnecessary presuppositions, is by comparison to what previous thinkers have thought. If we dismiss previous metaphysical views out of hand, that makes it hard to notice our own modern blindness towards certain issues and problems.
We need to be especially careful given the tendency for theories to be expressed in the language of the existing philosophical structures. This can cause us to think that those theories depend on a certain Metaphysics, when really they don't. Just because theologians expressed Trinitarian doctrines using Aristotelian terminology, or early 20th century physicists expressed QM using Positivist concepts, doesn't necessarily mean that these ideas can't be transplanted to certain other systems.
Aristotelianism became unpopular long before QM was discovered. But suppose that somehow the medieval European or Muslim world had discovered QM. I bet that their "interpretations" would have been radically different from anything we have now, even if the mathematics which they developed ultimately turned out to be equivalent to ours (as happened to Heisenberg and Schrödinger)!
None of what I have said here is a positive argument for Theism. It is merely my attempt to sweep away Carroll's strong claim that once we have a physical model, all of our work is done and there is no place to ask further metaphysical questions. On the contrary, we can continue to ask questions, and we can judge possible answers to those questions by whether they make logical or intuitive sense to us.
Perhaps at the end of the day, a person might conclude that the certain questions about the origin of the Universe are meaningless, or that that they are meaningful but that Theism is not the correct answer to them. But that has to come after considering the merits of different rival views, not beforehand.
Notes:
† The Pauli exclusion principle forbids two fermions from occupying the same state, which sounds a bit like the principle that solid objects cannot occupy the same space, except that:
(i) ``state'' refers not just to position but rather to all properties including both position and velocity (to the extent that both can be simultaneously measured in Quantum Mechanics).
(ii) It follows from a much more abstract mathematical principle, which I will cite without explaining, that identical types of fermions (e.g. electrons) have antisymmetric wavefunctions. This means that the complex number describing the quantum mechanical amplitude for one electron to be in position 
and the other in position 
is minus the amplitude for the first to be at and the other at . This is called Fermi-Dirac statistics.
(iii) It doesn't apply to bosons, which have symmetric wavefunctions. Therefore you can have a bunch of identical bosons in the same state, as in lasers or Bose-Einstein condensates.
‡ "Substance" is a technical term in Aristotelian metaphysics. It means an individual entity which possesses existence independently, as opposed to an "accident" which "inheres" in a substance and can be changed without affecting the underlying nature of the thing in question. Sort of like the difference between nouns and adjectives. Not to be confused with the more recent chemical notion of substances. This confusion of terminology causes great misunderstanding when Roman Catholics try to explain their doctrine of transubstantiation to modern people!
Great post, and let me say that I greatly appreciate your blog in general. To my knowledge you have the best reflections on the interface between truly cutting edge theoretical physics and theology.
I'm trying to suss out your exact stance on Scholasticism, though. Are you saying that you accept it but just think it can't be conclusively proven, or that you don't accept it but think it's at least somewhat plausible?
Personally I was convinced by Ed Feser's case for Scholasticism, so I'd love to hear your thoughts on his case. I also don't think it really does justice to the method of (at least classical) metaphysics to say that it reduces to the criterion of "logical consistency plus non-craziness". Classical metaphysics, at least, does not merely conjure up logically consistent views out of thin air and refine them so that they hang together, but begins from certain empirical, albeit very general, observations and attempts to make sense of them.
Welcome to my blog, JD Walters, and thanks for your question.
I don't accept Scholasticism as a whole, but I also don't reject it as totally worthless. I think that there are some things which it gets "righter" than other views, but there are also specific distinctions it claims are necessary to thought, which I don't find to be necessary. For example, I very much doubt that there is a absolutely sharp distinction between substances and accidents. It seems to me arbitrary to say that certain properties of a thing are essential to it and others are inessential, since it seems to me that any kind of thing falls into multiple different categories, some broader and some narrower, and I don't see why just one of those categories should necessarily be privileged over all the others. I agree that "rational animal" is a better description of human beings than "featherless biped", but I suspect that judgements like this are a matter of degree rather than a matter of kind.
It seems to me that in some ways Aristotle was confusing the way we think grammatically with the way things have to be metaphysically. Perhaps for this very reason, it is a highly useful langauge for expressing ideas and making important distinctions. When it comes to Scholastic proofs of e.g. the existence and properties of God, I think there is definitely "something to" them; they should certainly not be rejected out of hand. But I also recognize that they depend on metaphysical assumptions which I don't think can be proven with certainty (even if, once you accept those assumptions, you can make a demonstrative proof).
Of course, I recognize that St. Feser gave very detailed arguments, and I can't pretend that these brief reflections grapple with what he had said in anything like sufficient detail to refute him. That project would require another careful re-reading of his book, and several very long blog posts.
I agree that "logical consistency plus non-craziness" isn't how we should go about selecting the most probably true metaphysics; I was instead proposing it as a way of identifying a metaphysics which might possibly be true, i.e. to delineate the space of possibilities. To choose which of these possibilities to accept, we should indeed "[begin] from certain empirical, albeit very general, observations and [attempt] to make sense of them". But I suspect that even this process has a nonunique conclusion, for us limited and fallible human beings. If there are a large number of possibilities, then even the most likely possibily to be true might be probably false. I suspect that the real answer is as thick and rich as Scholasticism, but also as weird and counterintuitive as Quantum Mechanics, and probably we cannot really grasp the foundations of the world any more than Job could. Metaphysics is a hard subject!
One reason that this does not lead me to despair is that I don't think we need to know the "literal truth" about how things work in order to have many true ideas about the universe, since metaphor and analogy are also very powerful means of communicating ideas. In the Bible, God communicated to the Israelities using the vehicle of their premodern scientific and philosophical ideas, and he can use our own wrongheaded conceptions as well.
I hope it doesn't get tiresome reading compliments for your blog, but your posts have helped me come to understand a great many things, and I thank you for that. Now to my question... In the debate Craig seems to have responded sufficiently to Carroll's rebuttal to the first premise (appealing to his semi scientistic metaphysics of causality) by saying that this new formulation of the argument avoids the presupposition to Aristotelian causation. he states " I protest, not at all, there is no analysis given of what it means to be a cause in this first premise. You can adopt your favorite theory of causation or take causation to be a conceptual primitive. All it requires is that the universe didn't pop into being uncaused out of absolutely nothing." My two questions for you Dr. Wall are: Did you find this response insufficient to defeat Carroll's rebuttal? and if so what are your reasons for not thinking they are sufficient. Sorry if I have many grammatical errors in my comment I'm ESL.
Welcome to my blog, John.
I'll stop appreciating compliments about my blog just as soon as I stop being vain, which is unlikely to occur any time soon. :-) In the meantime, the closest I get to humility is praying that I can appreciate the accomplishments of others just as much as my own.
I was responding to Carroll's comments directly, not saying anything about whether St. WLC's response was or was not sufficient. But St. Craig was of course completely correct that there are lots of different models of causation out there, besides the Aristotelian one, and that quite a few of the products on the market, many of which are Enlightenment rather than pre-scientific, wouldn't allow the universe to just come into being without any explanation. An important exception would be the Humean view that causation is nothing other than constant conjunction of events, but I have difficulty accepting this view since if there is no explanation as to why the constant conjunctions keep occuring (not even to say that there's some mysterious reason which I don't understand, but there isn't any reason at all!), then it seems rather surprising, to say the least, that the constant conjunctions continues to occur in such a regular way.
However, I took Carroll to be attempting to articulate his own view of causation, when he said that:
So ultimately, Carroll is trying to defend his own view of causation, in which he claims that causality is a result of the laws of physics and of the arrow of time, and that without these two things causality doesn't make sense. To ask, "What is the explanation of this fact that there are these unbreakable patterns", is apparently to him a meaningless question which has no answer. Although apparently he does not regard the question "Why is there an arrow of time" as similarly meaningless, since he's devoted quite a bit of time and energy to trying to explain this... So you could also ask whether St. Craig successfully refuted Carroll's specific model of causation, and I'm not sure that he did. He made some comments about how if the universe could pop into existence, then so can "bicycles and Beethoven and root beer", but I'm not sure he really took Carroll's ideas seriously enough to make a refutation that might convince some of the people who think in Carroll's way.
But where Carroll is certainly wrong, is when he claims that this view is some sort of inevitable consequence of the state of cosmological understanding today. If even Aristotle can't be dismissed so easily with that sort of progress-of-science triumphalism, than a fortiori the other models of causation can't be dismissed that easily either.
Thanks Dr. Wall for your reply.
Responding to what you said about St. Craig not taking Carroll's view on causation seriously: I believe the reason St. Craig doesn't take this view of causation seriously is for 2 reasons. Firstly, he states that causation must follow the metaphysical first principle which states "being does not come from non being." And secondly, to adopt Carroll's view would be committing what Dr. Alexander Pruss calls the taxi cab fallacy (to arbitrarily stop at the universe when riding the causality principle). With this knowledge, should we take such a view seriously? I do agree with you when you say Carroll is certainly wrong when he claims that his view is some sort of inevitable consequence of the state of cosmological understanding, but if St. Craig's appeal to his a priori analysis of causation and your claim that the laws of physics haven't shown that causation only works within the arena of he universe are true, then that means he has neither any a priori ground for his metaphysics of causality, nor any evidential ground for his views. So how can Carroll's view even amount to a reasonable objection to be taken seriously? It would be great to read your reply. Thanks again!
Professor Wall:
I learned many years ago that a model is a representation of reality, the latter consisting of a multitude (but finite) number of parameters or variables. So right away, to speak of a perfect model strikes me as an oxymoron. Is there such a thing as a perfect model? It’s like asking: find the limit of a function but, in fact, the limit does not exist.
As model builders, we (a) make certain (simplifying) assumptions, (b) identify the key variables and parameters, (c) work out the cause and effect relationships through the system of equations, (d) review the model for mathematical soundness, and (c) test the model’s predictions against the observations.
It seems to me that we fall into the philosophical trap of believing that ultimate truth lies in mathematics alone (hence the models). This mindset is dismissive of metaphysical questions (z) that are part of the equation: y = f (x + z). In a perfect model, z would be zero and we might as well consider f an identity, or a calculator that always gives an exact and deterministic answer from the inputs.
1) Do we have such models in physics?
2) Do you [see] theoretical physics embracing the sort of metaphysical questions (especially the ontological subset on God) stated in the blog post.
Would appreciate to have your thoughts.
Many thanks.
TY
[correction in brackets---AW]
One of the things that came up in the Carroll-Craig debate was the correct interpretation of quantum mechanics. I think W. Craig and J. Sinclair support bohmian mechanics, Carroll support Everett's and Collins, may support the Copehagen (I don't have much to base that last judgement on), but out of curiosity, do you have a preferred interpretation? I'd be interested because some of the arguments that you've talked about already, like Smith's self-caused universe (and perhaps arguments you might talk about in the future), assume a certain interpretation.
John,
I'm sure Carroll would not agree that his view can be characterized as "being [coming] from nonbeing". He certainly claimed that the phrase "popping into existence" is a misleading characterization of his view since, granting for the sake of argument that the universe had a first moment of time, there was no time before that, during which it would be true to say that there was "nothing". Regarding the so-called "taxicab fallacy" (which seems part of a general trend in atheist/theist arguments of turning anything annoying the other side does into a fallacy-I-just-made-up), I imagine that Carroll would say that he gave a specific reason why the causal principle should stop at the universe, rather than God. Maybe you don't think much of his reason, but he did try to give one.
But presumably somebody who agreed with Carroll's views would be able to defend them more vigorously than I would. Carroll himself said this in his Post-Debate Reflections:
TY,
I agree that we should not regard our models as the full reality, at best our theories of physics are isomorphic to physical reality. As Bertrand Russell pointed out, they don't say what are the actual things that have the relationships described by the laws of physics. This seems to leave some room for discussion of metaphysics in addition to physics.
I'm not sure what the verb was supposed to be in your (2), but I'm sure you know that there is no "theoretical physics" model for God. As for whether theoretical physicists are likely to embrace such metaphysical questions, I think most have very little patience for the subject, but that may be in part due to a mistaken belief that their own way of looking at the world is, in principle, capable of answering all meaningful questions.
PS I don't see any reason why a physical model couldn't have an infinite number of parameters, so long as you only need to know a finite number to answer any particular question... something like this situation comes up in the "effective field theory" of "nonrenormalizable quantum field theories", where there are an infinite number of free parameters but, below a given energy scale, each successive parameter become less and less important.
Charlie,
I don't have a "preferred" interpretation of QM; I think the subject is a very difficult one. I'm sympathetic to something called "Quantum Bayesianism" but I'm not sure it's logically consistent. I think the universe is probably really nondeterministic. I dislike the Many Worlds Interpretation, since I don't think it's as well-defined as people think it is, nor do I think it explains the Born rule.
Prof Wall,
The missing verb in the second sentence was "see": "Do you see theoretical physics embracing the sort of metaphysical questions (especially the ontological subset on God) stated in the blog post." You correctly guessed.
In reply to Charlie, you said, "I think the universe is probably really non-deterministic:, which is the inescapable conclusion of the blog post. The indeterminsm notion leads to a related topic I've been following I've yet to see a well-reasoned essay or article : Does QM make theism credible and in what ways? Subject for a future blog post?
Many thanks.
TY
Thank you, Aron.
Carroll says
"But crucially, both of these features of the universe that allow us to speak the language of causes and effects are completely absent when we talk about the universe as a whole."
Question is whether a physicist that lacks a metaphysics can talk about "the universe as a whole"?
The physicist may talk about behavior of this and that object or about geodesics and space curvature around this star or black hole but what in the terms of physicist is
"the universe as a whole"?
I have a solid modelling approach to Metaphysics. It is not parametric in the usual Cartesian sense. It is compatible with the "Greek Thomist " way of seeing the world but it is terribly analogical. I might characterise it as an Escher Moebiusted kind of analogicalypsomatic (g)approach.