The Quantum Thomist

Musings about quantum physics, classical philosophy, and the connection between the two.
Why Is There Something, Rather Than Nothing? (Part 5)

Why Is There Something, Rather Than Nothing? (Part 6)
Last modified on Tue Feb 5 23:13:14 2019

I have been sporadically writing a series on a recent article by Sean Carroll. This is taking me several posts to respond to it fully, so I am going one section at a time. My first post gave an introduction to the topic, and covered Carroll's own introduction. My second post discussed his first main section, where he established his definitions and described the scope of his discussion. My third post discussed the definitions of "something" and "nothing". Carroll was a bit more reasonable here than many of his fellow atheists, who managed to get something out of nothing only by redefining the word "Nothing" to mean "Something." The fourth post discussed whether the universe could come into existence without something causing it, or could continue indefinitely. Carroll argued that both were plausible, and this is where his article started going wrong. In the fifth section, Carroll asked "Why this particular universe," both in terms of the fine-tuning coincidences and the structure of physical law. But he wasn't really able to give a satisfactory answer.

Now we come to the final section, entitled "Why does anything exist at all?" Here Carroll offers his conclusions. And as with the earlier part of the article, they are something of a mixed bag.

Early on in his article, Carroll offered five options to explain why there is something rather than nothing.

  1. Creation. There is something outside physical reality which brings it into existence, and/or sustains it. This is often identified with God or some sort of necessary being.
  2. Metaverse. Reality is just one member of a wider series of structures. For example, one might have pocket universes emerging from a larger structure, or a bouncing universe where one universe dies and then another begins, stretching back to infinity, although Carroll intends his concept to be more profound than this. This wider structure will be similar to our own reality (for example governed by similar rules, so it is distinct from the idea of a creator), but larger and at a more fundamental level.
  3. Principle. Reality is the way it is because of some underlying principle, perhaps simplicity or beauty.
  4. Coherence. The concept of "nothingness" is incoherent, so anything except a universe consisting of something is non-viable.
  5. Brute Fact. Reality is the way it is, and that's it. Eventually, the chain of explanation will reach a point where we can't go any further. The is no underlying reason why that terminus has to be what it is -- it could be different -- but it just happens to be that way, and we have to live with it.

As I mentioned earlier, I agree that these are the only real options. The only alternative I can think of is to combine principle and coherence; to say that any universe except the one we live in is logically inconsistent in some way. However, that flies in the face of modern physics: by tweaking the laws of physics, we can come up with numerous other possible universes which are equally self-consistent. That option doesn't seem to be open to us. I'll discuss metaverse, principle and coherence first, since here Carroll and I are in agreement. I'll then discuss his preferred option of a brute fact and mine of a creator.


By this he means a set of non-interacting distinct universes, with different laws of physics, and not coming from a common past.

One problem with this is that we cannot by definition have any observational evidence for this, either direct or indirect by reasoning from what we do observe. It could provide an explanation for the fine tuning coincidences: the metaverse samples numerous different universes, so some of them will be in the window where life can arise. Of course, we would find ourselves in one of those.

However, this explanation is entirely ad-hoc. There is no obvious reason why there should be a metaverse. Indeed, if the universes aren't connected by some principle or outside agency, then we can't really use another universe which has no causal, historical, or any other connection to our own as an explanation for what is going on here. In other words, the metaverse fails to answer the question. None of the alternative universes can explain why, in this universe, there is something other than nothing. Thus neither can the ensemble of universes as a whole.

Plus, of course, we still have to answer the question. The metaverse is something. It could not be there. So we ask "Why does this universe exist rather than nothing?" The answer comes, "Because it is part of a metaverse." So then we ask "Why is there a metaverse rather than nothing?" And it is clear not only is there no evidence and can there be no evidence for a metaverse as Carroll defines it, but its existence doesn't even answer the question.

This, more or less, is Carroll's argument, and I agree with it. The only caveat I might have is that his definition of the metaverse is stronger that the multiverse which is usually posited. This arises from some interpretations of string theory or inflation. Here the different "universes" are connected to our own, either in a larger multi-dimesnional space, or from different inflationary bubbles expanding in a larger subspace. I'm not sure, however, if they are connected to our own universe, these can necessarily answer the fine tuning argument (since there would still be some over-arching laws governing the multiverse, which might need fine tuning). It certainly doesn't answer the question of why there is something rather than nothing. Could such a multiverse have lasted indefinitely into the past? The problem here is from the second law of thermodynamics: one would have to show that the multiverse can't reach an equilibrium state. If it had, and extended infinitely in the past, then we would be in that equilibrium state. That's problematic since our solar system is not in equilibrium. The second law is derived from a small number of axioms, such as locality, which is necessary for the causal structure of the universe and thus that the universe is rational. Thus we can expect it to be satisfied over a wide range of possible fundamental laws of physics, and perhaps all of them.

Thus the metaverse and multiverse can be discarded as explanations of why there is something rather than nothing, because they don't answer the question.


Is there some principle that picks out this universe? Carroll answers that we don't know what such a principle could be. He hopes that future scientific discoveries might provide an answer.

I agree with him that science does not currently give us any hints as to such a principle. There are aspects of our best formulation of scientific law which seem to be devoid of any higher scientific principle. Appeals to symmetry have got us a long way. But the next question we have to ask is why those symmetries? And that is simply not a question that science can answer. Experiments can measure the value of parameters, notice regularities that can only be explained in terms of symmetry laws. But that is all. We cannot explain the existence of symmetry laws by appealing to a symmetry law. Yet that is pretty much the only tool that theoretical physics has had for the past hundred years.

Thus whatever this principle is, it will not be scientific. It will have to be philosophical. Science discovers what the law is. It cannot explain why it is. Philosophy obviously has a disadvantage over science in that it is one further step away from experiment. So the hope of finding some underlying principle that explains not only the nature of the universe but the existence of the universe seems to be fairly slim. And, of course, that principle would not only have to say why this particular universe, but why it exists at all. This is a much harder task. Why? Because whatever principle we suppose, it would just be an abstract concept. For example, we might search for the simplest possible universe, or the prettiest universe. But then we have to define what we mean for the universe to be simple or pretty, and that definition will be either subjective or cast in abstract terms. An I think this is true for any principle we can think of that might explain the universe. But an abstract principle cannot give rise to a concrete universe, as ours seems to be, just as acknowledging that 2+2=4 can't put four pounds in my pocket. Therefore this principle, if it exists, it must be an existent being.

However, we can still induce some attributes of this underlying principle based on the basic facts of nature. It's rationality, for example (as exemplified in substance causality), and regularity. And this is what theologians have done for the past several thousand years. So it is not as though these issues haven't been thought about. Of course, one can argue that now we have a better knowledge of science, we need to rethink things. Fine. But even so, there is a vast treasure trove of work that has already been done on this subject. We would be fools not to consult it, and adapt it as required to fit contemporary physics.

So the principle argument and the creator argument are not so dissimilar. Either the principle is solely abstract, in which case it cannot explain existence, or it is at least in part residing in some existent being, in which case it can explain existence, but collapses into a creator outside physical reality. Either way, we need not think further about this possibility.


Is the statement "Nothing exists" in itself a coherent idea? Does it entail a logical contradiction? Carroll first of all mentions the idea that we don't see a logical contradiction in saying that any typewriter (for example) doesn't exist. Had circumstances been different, then that typewriter might not have been created. If it is suddenly erased from having ever existed, then there not obviously any logical contradiction. So why should the universe be different?

This direct argument from analogy is problematic, as Carroll notes, because the universe is not like any other thing. The universe, after all, has been defined as the entirety of physical reality. It is a stretch from saying that because the typewriter need not have existed that physical reality need not have existed.

But then, neither is there a good reason for saying that reality must exist (if we leave out the observational evidence, which is what this possibility asks us to do: it states that we can deduce that things exist from first principles, without referencing any observation). And I don't think there can be. In physics, we map reality (or at least part of it) to an abstract representation. Physical space and time are mapped to a geometrical space. Physical potentia are mapped to the set of eigenvectors of an operator acting on a Hilbert space. Our knowledge of which state is occupied is mapped to a quantum amplitude. Once we have this representation, we can perform calculations and make predictions. If we are to think about an understandable physical universe, then the possibility of a mapping from physical reality to abstract representation is a necessary element of it. Every possible understandable physical universe will have an abstract representation. Given that the mapping is two way, it also seems reasonable that every abstract representation satisfying a minimal set of conditions corresponds to a possible physical universe. There is nothing incoherent about an empty set.

Indeed, there is a link between this argument and the most disputed argument for the existence of God. I personally believe that God's existence is not logically necessary, but if anything else exists (and the universe is rational) then God must also exist. But if nothing else exists, then there is no logical reason why God should exist either. However, the ontological argument attempts to demonstrate the existence of God from first principles, without reference to the existence of other things. It attempts to argue that God's existence is logically necessary. This argument is rejected for many reasons, which I don't need to go into here.

I don't, however, see why an argument that the universe (or matter, or something of an unspecified nature and state) is logically necessary (and don't forget: we are talking about a purely abstract level of discussion here since we are discussing logic; we cannot bring in observational evidence in this discussion) should fare any better than an argument that God is logically necessary. This is particularly true since the argument implies certain attributes of the logically necessary being. For example, such a logically necessary being would also have to be a necessary being (in the weaker sense of not depending on anything outside itself for its existence), and consequently cannot come into or out of existence (if we are to accept the premise that the universe is rational). Now change implies that the being has several different potentia or states. One of them ceases to exist and another one moves from potential existence into actual existence. Thus a necessary being cannot change, since change involves it ceasing to exist in one particular state. Any universe capable of change (which obviously includes our own) thus cannot be logically necessary (by change I mean in any respect, for example either in space, or in time, or not being perfectly homogeneous). And if this is true (and I have only sketched out the argument here), then the question of the universe's existence cannot boil down to it being incoherent that it doesn't exist.

Theologians have deployed the ontological argument. It's not an argument I use myself, but nonetheless we should consider what comes after it. Because they then go on to demonstrate that the end result of the ontological argument must have the traditional attributes of God. So if the ontological argument is correct, and there is something out there whose existence is logically necessary, then it can only be God. These follow up arguments are frequently neglected by the philosophical community, but they are still there to be made and answered. So the problem for any atheist who wants to demonstrate that the non-existence of stuff is incoherent is that he would also have to show the falsity of the arguments that the logically necessary thing he ends up with would have to be God (which are considerably stronger than the ontological argument itself).

Carroll's response to this question is that even if we do have such a principle, it cannot explain why the universe has this specific form. Thus an attempt to argue that it is incoherent that no universe exists doesn't answer the question we are most interested in: why does this universe exist.

Brute Fact

So now I come to Carroll's preferred solution, the brute fact. This was defined earlier in his work twice

  1. Brute fact: Reality itself simply exists, in the way that it does, without further explanation.
  2. At the other end of the spectrum, explanations might bottom out with a brute fact: something that simply is the case, without further reason, even though it didn’t necessarily have to be that way.

However, the first definition is simply to say that there is a terminus to the chain of explanation. Note that the brute fact was defined in terms of reality, while the universe was defined as "the entirety of physical reality." With "No judgment is implied about whether things other than physical reality can be usefully said to exist," implying that there might (or might not) be parts of reality not included in physical reality. Now, the theist would claim that the terminus of the chain of explanation is God (in fact, a theist might define God to be this terminus). And this would fit this definition of the brute fact if reality includes the supernatural as well as the natural. However, Carroll also places the brute fact and God (or creator) as distinct answers to the question. Since the second definition avoids this conundrum, I will take that to be my starting point. Thus the first distinction between brute facts and the creator is that a brute fact doesn't necessarily have to be in the particular state it is in. It is the same type of thing as the beings or explanations which are derived from it. The second distinction (taken from the definition of the creator) is that a brute fact exists as part of physical reality, while the creator is outside it.

And that makes sense. Traditionally God is viewed as being unchangeable, immaterial and perfectly simple. In particular, this makes God a being of pure actuality (in Aristotelian terms). For example, the hydrogen atom is described by a series of states, corresponding to the energy levels of the electron. So we can imagine the electron continually bouncing from one level to another. This sequence can either continue indefinitely, or it would terminate in some particular starting state. But whichever state the electron started in, it would be arbitrary. We have no good reason to arbitrarily select one state over another as a starting point. It just happens to be that way. That would be a brute fact.

The alternative is that the sequence of explanation terminates with a being that has only one possible state. This would not be a brute fact, because we are not selecting one state arbitrarily as the starting point. The being started in that state because it has to exist in that state, if it exists at all. And this is the starting point that theologians use when thinking about God. God is the terminus of explanation, but to avoid having to resort to a brute fact, we say that that terminus is a being of pure actuality. That means that it must be simple (because if it were composed of separable parts, those parts would be prior to it in the chain of explanation), unchanging in either time or space, unique (because two such beings would share the same attributes but be separable, violating simplicity), immaterial (since no material substance satisfies the previous attributes), omnipotent (it must be able to influence matter, since it is the start of the chain of explanation, but its unchanging nature means that it is not just the first matter that is influenced by it, but it must interact with all matter no matter the time or place in the same way), contain a will (because it is an immaterial substance which nonetheless directs other things), an intellect (because it grasps the forms of those things it directs without becoming them), and free (from the indeterminacy of physics). Obviously I have here tried to sketch out in one sentence what I should spend at least a chapter on. But the argument above is fleshed out in all the standard textbooks on scholastic theology. But I hope that I have done enough to suggest that the alternative to a brute fact (other than an infinite regress) is none other than our old friend God.

So what reasons does Carroll give for thinking that the brute fact is a the correct solution? Primarily because he believes that he has eliminated the alternatives.

Every attempt to answer the question "Why is there something rather than nothing?" ultimately grounds in a brute fact, a feature of reality that has no further explanation. The universe is not unique, and there are no necessary beings; even if we decide that the concept of nothingness is incoherent, at least some properties of our particular universe are ultimately contingent. By the standards of modern science, it is extremely hard to see what could possibly qualify as a final and conclusive "reason why" the universe exists.

I note that Carroll means by necessary being what I refer to as a logically necessary being, and I will discuss what he has to say on this subject in the next section. I note, however, that his discussion leaves out an alternative: a terminus of explanation that is neither a brute fact nor a logically necessary being.

Note his appeal to the "standards of modern science." This is something I would agree with; but where I would disagree is that in this matter science presents the final word. All science can do is formalise the laws of physics and describe the initial conditions of the universe. But it cannot explain them, which is what we are trying to do. To explain them, we need to turn to various philosophical world-views.

Carroll states that his conclusion shouldn't be surprising. He appeals to an analogy from astronomy. Kepler tried to fit the orbits of the then known planets into some geometrical pattern, from which one might try to deduce some reason for their orbits. But we know that he failed; the distribution of the astronomical bodies is random.

Random? Yes and no. It's random in the sense that there is no underlying pattern or principle behind the orbits of the planets. But we also know that they can be explained in terms of a sequence of causes and effects stretching back to the birth of the solar system and beyond that to the explosion of whichever star or set of stars from the previous generation of stellar bodies in which the iron and other elements that form the core of the planets was fused together.

Carroll also discusses the fine structure constant (which defines the strength of the electromagnetic force). People searched for patterns to explain its values. Now we know that the quest is hopeless; it just seems to be an irrational number that doesn't depend on any others. In this case, Carroll is begging the question: the fine structure constant is one of the philosophical principles we are trying to explain. So it can't be used as an illustration of a brute fact, since that is what is under dispute.

So what arguments can be used against the nature of the universe being a brute fact? Firstly, it is unsatisfying. It is the easy way out. It is similar to people saying that Aristotle's physics was a brute fact, so there was no need to investigate further. In short, it is refusing the question before people have had a chance to examine it.

Secondly, it cannot be proved that whatever you identify as a brute fact is, in fact, the terminus of the series of explanations. Whenever you get to what seems to be an end state of the chain of explanation, if it is a brute fact rather than something which can only exist in one state, it is always possible that there is another explanation behind it. You can never show that the state of affairs you have is truly the end. The only way you can be sure is if you end the chain of explanation with something like the theist God, the unchangeable, uncreatable being of pure actuality.

This leads us to the favourite new atheist objection to God: if God is the explanation of the universe, then what explains God? The theist answers this by saying that his definition of God is the being which cannot have a prior explanation without implying a contradiction (and then what this entails in practice is worked out in detail, and the attributes of such a being are found to coincide with the traditional attributes of God). But this question can also be turned against Carroll's brute fact. If a set of brute facts are the explanation of the universe, then what explains those brute facts? And here the one who believes in brute facts cannot turn to the theist answer, since a brute fact is defined as something which could in principle have a deeper explanation, but happens not to. The brute fact is defined as being at the end of the chain of explanations. But it is also defined as being of the type of things that need not be at the end of the chain of explanations. Clearly something is amiss here.

Now could we argue that the universe could not have a deeper explanation, but the chain of explanation stops with the laws of physics and initial conditions, and can go no further in principle as well as in practice? But, of course, there are those who claim that the universe can be explained in terms of deeper principles, for example the theologians. The theologians' explanation might not be correct, but that it can be articulated shows that deeper explanations than the supposed brute fact are possible in principle.

Then we can argue that saying the explanation of the universe reduces to a brute fact ultimately means that the universe is irrational. Reason, of course, is the movement from premise to conclusion, or, in terms of the physical world, from cause to effect. If that sequence is broken, and there is at any point an effect without a cause, then the rational universe and thus science is a fraud. One of the goals of science is to make predictions about future events from knowledge of the present. These predictions are expressed in terms of probabilities. But all probabilities are conditional, in this case on the initial state and the underlying laws. If effects can happen without a cause, then all that logic goes out of the window. One cannot express a probability for something with no connection to what came before it. As soon as you find a rule that states there is a certain probability for the effect to appear in certain circumstances, then to define that probability you have to parametrise the effect occurring in terms of some model. But if it is described in terms of a model, then it does not appear for no reason; the model itself encodes a reason.

So we cannot have something happening for no reason at any stage of the causal sequence if the universe is to be understandable (and if the universe is not understandable, then it is difficult to account for the fact that we do understand it very well already). "Any stage" includes the start of the sequence. There is nothing special about the start of the sequence of causes, unless we suppose it to be fundamentally different from all that comes after it, as those who argue for the creator do. We can't say that once the universe has began, we can't have an effect without a cause, in any of the causal sequences, but we can right at the start. Because that would imply that the occurrence of the causeless effect depends on the external circumstances. And as soon as you start saying there is some dependence on something else (even if it is a prevention of the effect), you are saying that the effect does have a cause. In the brute fact model, of course, the starting point is of the same sort of thing as everything that follows it, so one can't use the theologian's escape.

So, in terms of causality, the brute fact model undermines science. It leaves us with no confidence that scientific predictions will be successful, because if a brute fact happens once it can happen again, and if it happens during our experiment, then we will get an inexplicable result. Our whole extrapolation back to the origins of the universe and to the most fundamental scientific law is based on the premise that all across those chains of reasoning there are no brute facts introducing new features into the universe. If we suppose there are brute facts, we undermine the process that was used to describe the initial state that we (if we accept that it was a brute fact) we believe can't be explained further.

Of course, Carroll isn't discussing sequences of physical causation, but a sequence of explanation. Does this make a difference to the argument? Not enough to undermine it. For a chain of reasoning is undermined just as much if we introduce an element of irrationality -- something which does not follow from what preceded it -- at an intermediate step. So if we feel that we can introduce brute facts at the beginning of the chain of explanation, then why can we not also introduce them later, and destroy whatever logic led to the purported brute fact in the first place?

Of course, every rational argument depends on at least one axiom which cannot be proven within the context of the underlying logic. So one might argue that every rational argument contains at least one brute fact. But the universe is not a rational argument. We can apply reason to understand it, at least partially, but there is some physical basis that underlies the logic. So the terminus of explanation we are looking for is something in reality, not something in abstract thought.

And this undermines the idea that the laws of physics could be the sole ultimate explanation. The laws of physics (including the mysterious parameters) are an abstract description of physical processes. They themselves cannot be an explanation for reality because they are not part of reality, but only a description of it (or part of it). Expressed as a mathematical law, the laws of physics cannot cause anything. All they can do is describe a relationship between different existent beings. Just as my drawing lines and angles on a piece of paper doesn't cause light emerging from a lamp through a slit onto a screen. If we are to explain the universe, we have to ultimately come to some being or set of beings. Which brings us back to causality, and the assumed absence of brute facts in causal systems.

So if the sequence of explanation has to terminate with a being or set of beings, the next question must be what attributes must this starting point have so that it could be a reasonable end point. This is the question which Carroll does not discuss. But it is one that a great many people have thought about, and which we should think about.


So now we come to the final option. The earlier definition was the following:

Creation: There is something apart from physical reality, which brings it into existence and/or sustains it. This hypothetical entity is often identified with God in the literature, but there is not necessarily any strong connection with a traditional theistic conception of the divine.

Carroll lists two main reasons why people might argue for a creator. Firstly by an analogy with complex mechanisms in our experience; we know that they only arise through the work of somebody who designed them. Secondly, the idea that the "laws of nature" are inexplicable without a law-giver, just as human laws require a parliament or king to fashion them. Carroll dismisses both these analogies as not really relevant, with some justification. Arguments from analogy are only good if there is a strong resemblance between the two things; and clearly there is little resemblance between a pocket watch and the universe. However, I will come back to the legal argument, because I am not sure that needs to be presented as an argument from analogy.

Next he turns to the classic new atheist argument: if we invoke the creator to explain the existence of the universe, then we have to then explain the existence of the creator. Carroll is at least aware of the response: the creator is different from the universe, and explains His own existence. But why then can't we just say that the universe explains its own existence?

There is a footnote which Carroll provides here which is worth citing in full:

All else being equal, a self-explaining and necessary universe would be a simpler overall package than a self-explaining and necessary creator who then created the universe. But to most advocates of this general strategy, necessity seems like a more natural property to attribute to a supernatural creator than to the natural universe.

Carroll notes that he argued earlier that the idea of a necessary being is incoherent, although as I commented at the time, he here confused the two notions of logically necessary (i.e. whose non-existence involves a contradiction), and necessary as in not contingent (i.e. which doesn't depend on anything outside it). He argued against a locally necessary being, but what we need to avoid the explanatory regress is something necessary in the second sense. While every logically necessary being is necessary, it is not so clear that every necessary being is logically necessary. At least, Carroll hasn't proved this.

So, Carroll concludes, invoking a creator doesn't solve the problems of the brute fact; we just terminate the sequence of explanations one step further than we need to. So why not just say that all of reality is the end point rather than invoking a creator?

I'll finish by quoting his conclusion.

The existence of a creator of the universe should be judged on ordinary empirical grounds (does it provide a useful explanatory account of observed features of what we see?), not on a priori arguments for its necessity.

I immediately note the false dichotomy (particularly since he means logical necessity when he writes necessity). I also note the lack of justification for this statement. Why should we restrict ourselves to one type of reasoning, if others should prove useful?

So what do we make of this?

I'll start by addressing his summary of the reasons why appealing to a creator seems appealing. I note that he just lists two relatively modern arguments: the argument from design, and the idea that laws need a lawmaker. I think that he is right in saying that as expressed these reduce to arguments from analogy, which are only rarely useful. The second flaw of these arguments is that they are based on a mechanistic philosophy of physics, which a theist has little reason to accept, and which is very much out of line with contemporary physics anyway.

However, I will dwell on the lawmaker argument a little longer. Historically, the idea of a lawmaker came first (albeit not expressed in those terms), and it was only much later that the idea of laws came after that. The reason that laws were supposed was that God was rational, regular, constant in time and space, and actively sustains all the matter in the universe. God's actions in sustaining the universe would therefore be rational, regular and constant in time and space, and apply equally to all matter. They could thus be described without direct reference to God (although, obviously, they indirectly depend on God). From this, the notion of the laws of physics arose in the late medieval period. Clockwork had just been invented, where the movements of the gears and weights followed regular and predictable patterns, and it was suggested that the motion of the planets could be thought of in terms of similar principles. This analogy was then taken by the late renaissance and early modern mathematical scientists and natural philosophers, and a new philosophy of physics drawn from it. Because the laws so formulated don't directly reference a creator, people started to wonder whether the creator might not be redundant, and thus atheism was born. [Apologies for being over-simplistic: I will claim that this is one reason behind the birth of atheism. It will be too bold to claim that it is the only reason. For example, another factor that led to atheism was the recasting of God as a rival explanation to science rather than the more traditional thinking that God is the explanation of science.]

The laws of physics are just a mathematical description. It is not Newton's law of gravity which causes objects to fall; it describes how they fall. Nor does the curvature of space time and the geodesic equation cause objects to fall: if anything, this is even more abstract than Newton's theory. An abstract description cannot cause a physical object to do anything. The concepts of geodesics and the space time metric are all notions that arise out of a geometric representation of reality. But they are not themselves direct representatives of physical things. So we are left with the question of why matter moves as though it were forced to follow geodesics of a space time curved by matter as described by Einstein's field equation. On the table we have the original reason proposed by the medievals who first launched modern physics. What is the atheist response? Carroll seems to prefer the brute fact answer "they just do," which just declares the question unanswerable. Surely we should not be satisfied with that?

Then, of course, there are alternative reasons why people believe in God. I'll skip over the teleological argument to be briefer, and focus on two. The first is the experiential argument. People have experienced what they claim to be the presence of God as something tangible. By itself, of course, this might not mean much -- just a misfiring neuron. But when it is backed up with something such as a prophecy or physical sign, then we have to take more interest. Obviously, atheists dismiss miracles using various arguments, none of which are actually much good. [All rational arguments are based on premises; in this case premises taken from a philosophy of physics, together with a definition of a miracle that equally only makes sense in the context of their own, or perhaps a related, philosophy. Those who wield such arguments assume that their philosophy of physics is correct. But what if there is an alternative philosophy of physics that explains the physical data equally well, or perhaps even better, but which lacks the key premise that their anti-miracle argument relies on? What the argument against miracles then reduces to is a dismissal of observational evidence because it disagrees with a pre-conceived philosophy, which is asserted without evidence to be better than the theist philosophy of nature -- how unscientific is that?] This, I can say, is probably the primary reason why people believe in God. They find the historical evidence for whatever miracle is at the basis of their own preferred religion to be compelling, and it is also backed up by their own personal experience, or miracles they witness themselves.

Secondly, we have the traditional philosophical arguments, and in particular I want to focus on the cosmological argument, or the argument from efficient causality. I know that atheists have their objections to this argument, but I have discussed those elsewhere and don't intend to do so here. Efficient causality links one being in a particular state, with either the same being (i.e. same form and matter) in a different state or a different being (different form or matter). It may be that an effect has several causes, or a cause several effects, but we only need focus on one of them. At each junction, we need only take one of the paths, and we ultimately finish in the same place. Each link in the chain represents either one individual being or a small number of individual beings. There is then an argument that the chain can't continue indefinitely or be circular, and we are left with an initial member of the chain which is usually identified with God. But here I will just call It the First Cause.

Every member of the chain of causality is an individual being. That includes the First Cause. When discussing the part of an explanation that is dealt with by efficient causality, the First Cause is the terminus of the chain of explanation. In other words, it is the subject of Carroll's discussion. Carroll asks why the universe itself couldn't be the First Cause. The answer is obvious. The First Cause is a single being. The universe is not a single being. The laws of physics are not a being at all. Therefore neither the universe nor the laws of physics can be the First Cause.

The next question is what attributes the First Cause must have. Clearly there must be something different about it. We know that everything else in the chain can come into existence, or can change its state. If it is possible for something to happen, then given enough opportunities it will happen at least once (that's the definition of possible). So the First Cause must be different, something which logically cannot come into existence or change its state. We immediately have that it possesses the divine attribute of immutability. Neither can it change its state across space, for similar reasons, so it possesses the divine attribute of omnipresence. Nor can it be composed of parts, because those parts would be prior in the sequence of explanation. We now have simplicity and unity. No material being satisfies these properties, and so we conclude that it is immaterial, that is outside the scope of physical universe. It nonetheless can generate change in objects outside It, and in an immaterial being we call the power to do that a will; which must be coupled with an intellect, because It is able to grasp the forms of that which is generated. Being outside of space and time, It cannot be biased towards the matter at the start of the universe, and so has to relate to all matter whenever it is. Thus it sustains the matter in the universe, and directs it (albeit through secondary causes).

One can object that this is based on the argument from efficient causality, and Carroll doesn't phrase his paper in these terms. But it is still relevant. The question is "Why is there something rather than nothing?" The question "Why?" was asked in two different ways, and one of these (Carroll's section 3) refers to the efficient cause of the earliest things in the universe. So although Carroll doesn't name efficient causality, that is still the type of explanation that he spends at least some of his time considering.

The other question (Carroll's section 4) asks about where the laws of physics originate from, which is not related to efficient causality. The concept of "the laws of physics" arose in late Medieval Europe. Thus there is no direct correlation between this question and standard Aristotelian thought. The role played by the laws of physics was taken on by the final causes and powers of individual beings. A contemporary Aristotelian would describe the laws of physics as a summary of these total effect of these final causes of individual objects. And, of course that's where the teleological argument (which is not the argument from design) comes into play. It demonstrates (or attempts to demonstrate) that the existence of these natural tendencies in material objects implies the existence of an immaterial Will. And, of course, one then builds on this to show that this Will must possess the other traditional divine attributes.

Carroll once again favours the idea that laws of physics are either a brute fact, or can be explained by more fundamental principles which ultimately end in a brute fact. For example, classical electromagnetism and optics is the small ℏ limit of quantum electrodynamics. We could have taken classical electromagnetism as a brute fact, but now we see that it arises from a more fundamental theory. Similarly, I don't think that Carroll would object if the standard model of particle physics, which incorporates quantum electrodynamics, arises as a limit of a more fundamental theory, and that theory as a limit or application of something more fundamental, and so on. [Or the standard model is just an effective field theory derived from something more fundamental.] But he argues that the sequence of explanation should end in some theory which is a brute fact; it could be different, but just happens to be the case.

The medieval philosophers asked themselves the same question. Obviously, their physics was different, but there is a similar process of explaining ideas in terms of more fundamental theories. They concluded, like Carroll, that there has to be something at the end of that chain of explanation. But, unlike Carroll, they then asked themselves what attributes such a termination should have if it were to be the sort of thing that might terminate an explanatory series. And, as in the argument from efficient causality, these turned out to be the traditional divine attributes.

Finally, I need to address that final comment of Carroll's. Does the idea of a creator provide a useful explanatory account of observed facts? Obviously it does for miracles, but that's another issue.

So let us think back to the birth of modern science, in the late medieval period. Experimental and mathematical science were at that time seen as rival approaches -- it wasn't until Galileo a bit later that they were brought together. But there was a still a big step made in formulating the basic assumptions of physics that we now take for granted: that the laws of nature exist and can be studied in themselves; that they are unchanging in time or space; that they can be understood objectively in terms of (mathematical or other intellectual) abstractions; that everything obeys the single principle, and so on. These are not obvious ideas. It is very difficult to find a philosophical framework in which they are justified. The Chinese and Indians, for example, were in many ways more advanced than the Western Christians of the time. They had some great and highly intelligent and creative thinkers. But they never came close. Modern atheists have inherited this tradition, but didn't originate it, and it is difficult to justify these principles from an atheistic world-view (most atheistic world-views instead presuppose them).

The people who set upon this path were influenced by three sources; firstly Christianity; secondly the Greek Socratic philosophical tradition (of all the Greek philosophical traditions, the one with the closest affinity to theism), as mediated through early Christians such as Augustine and Boethius, and Jewish and Islamic commentators; thirdly the Greek and Indian mathematical traditions as combined and systematised by the Arab scholars. Aside from the mathematics, they were motivated by theism. They adopted these assumptions because they viewed science as a description of God's actions and it would thus reflect the divine attributes. If they were wrong about God's nature (and, say, the Chinese were right), then there would be no reason why the universe would have such properties, and modern science would be a failure. Perhaps they just hit on the answer by luck. But that doesn't mean that theism ceases to be a valid explanation of these aspects of science.

Modern science is what grew out of their thoughts. It has taken seven centuries and still going, but the success of science is their vindication. But every time a scientific breakthrough is made, confirming these basic assumptions of science, it is a further explanatory success of the theistic outlook.

I would, of course, go further than this, and have argued that not only the broad outlines but many of the finer details of contemporary physical theory can be deduced from principles drawn from theism. (I can't get everything; I need to draw in some experimental data such as that there are three space and one time dimension, and that the physical universe exists). I won't say that I have the final word; no doubt my arguments need to be strengthened and improved. But I have certainly tried, and, by doing so, done more than the atheist.

So when Carroll states that the idea of the creator needs to have explanatory success, there are three responses. Firstly, that explanatory power is not the only way in which we can deduce that an idea is sound. The most important historical philosophical arguments for God are deductive. The most important practical argument for God is from direct observation of God's special effects. Thus even if there were no explanatory consequences of God's existence, we would still have reasons for believing in Him. Secondly, the idea of a creator does have explanatory power. Many of the fundamental assumptions behind modern science were drawn from it. The other assumptions of science don't contradict it, at least not the God of classical theism (although they might contradict other notions of god). And if my own arguments are correct, then it can explain far more than this, but many of the details of the standard model. Thirdly, we should put atheism to the same test. Can an atheistic world view explain the laws of physics? In particular, can the brute fact approach do so? It seems strange to demand a stringent condition for one hypothesis, but leave it off for the others. At the very least, can it even match the postulates of those original theist founders of modern science (as something derived from some form of atheism rather than something assumed by it), and explain them in terms of some more fundamental principle?


I have come to the end of this series now, and I find myself agreeing with much of what Carroll writes in his article. Only when he discusses God do I find myself in disagreement with him, and that's mostly down to the fact that we have different understandings of and use different definitions of God. Like most of the new atheists, Carroll doesn't understand much about theism, and it shows.

So my main criticism of Carroll is not that he is going in the wrong direction, but that he doesn't go far enough. He reasons that the sequence of explanations must have an end point. But he doesn't ask what would make a reasonable end point. What attributes it must have that make it stand out from the rest of the sequence. In terms of efficient causality, this is clear enough. Every being in the sequence beyond the First Cause can have a cause. If the First Cause were a similar type of being, then it too could be caused. So if it could be caused, and there was an infinity of time (or whatever plays the same role as time in whatever lies beyond the big bang), then it would have been caused, contradicting that it is a first cause. The only alternative is to say that the First Cause is something different, something that by nature can't change. And pondering on this leads us to the classical notion of God.

So Carroll's argument was generally good, but his problem was that he didn't think through its implications enough. If he did, he might have found himself retreading the paths of the medieval philosophers and theologians.


But what is the difference between a brute fact as a final explanation, and God as a final explanation? I had a comment to my last post, which is worth addressing here.

Aquinas writes in De Potentia 3.17 co, when we ask why the world came to be when it did, or why it is of its size or location, the answer is not in necessity or in some inherent "best" but simply in God's will ("non potest huius ratio reddi nisi ex voluntate producentis") and wisdom.

I do not see how this functionally different from saying "God did it." When we get to God's will as an explanatory principle, we admit that our understanding of the world reaches a point beyond which is inscrutability, since God’s will is not known to us.

Theists maintain that an appeal to God (what I consider an appeal to God's will, which implies of course God's intellect) is different from saying that we arrive at brute fact. Because God exists necessarily and His essence is identical with His existence, etc., appeals to God are not appeals to brute fact.

Since we have no direct access to God's essence, being forever on the near side of the dividing line of analogical predication, it's not clear to me how God don't cash out as brute fact appeals.

Aristotle on the other hand accomplishes the same without needing to give his god a will. The upshot is that in Aristotle, the kind of necessity by which nature exists and operates is not the hypothetical necessity needed by Thomas, but absolute necessity: cf. Metaphysics Lambda 7 1072b10-13. More economical. A consequence down the road is that Aristotle does not need to distinguish between eternity, "aevum" (Aquinas' mode of measuring duration of existence of separated, immaterial substances) and time. Aristotle only needs time, which is eternal, the principle of prior-posterior, in a universe that is eternal and by necessity (and, for him, "best"). More economical!

But WHY is the universe this way and not some other way? One person says, "it is God's will." Another says, "it's brute fact." Maybe the focus of difference will turn out that at least on Aristotle's picture (and in most of ancient dogmatic philosophy except possibly Plato), there is no possibility in reality, only in imagination, that there could have been a different universe, even if on a day to day level there can be automatic events here and there. The Thomist on the other hand seems to be able to allow that in reality, not only in imagination, there could have been a different universe. I'm not sure whether it's an interesting gain to preserve for God the option of having created different universes than this one, since this one is the only one we have. It doesn't seem to further scientific inquiry to maintain that God could have created a different universe - it only serves theological ends. But if we're debating whether God exists, then we can't take it as given that the principle subject of theology exists.

First of all, I should note that Christians (at least Western Christians, Roman Catholics and classical Protestants) regard God's intellect as having priority over His will. There is also a distinction between God and the will of God. God is a single, being. The will, on the other hand, proceeds from God. So while it is an element of God, it shouldn't be equated with God as a whole (we are getting into Trinitarian doctrine here, and the distinction between the different persons of the Trinity). So equating God with God's will is problematic.

Secondly, the claim is that we have no direct access to God's essence, as we are on the wrong side of the analogical divide. Now I agree that analogy is important in scholastic thinking. When we say that God has power and we have power, we are arguing analogously. Power in God is not the same thing as power in us; but there is enough of a resemblance that we can talk about the power of God analogously to human power. But not everything understood about God is done so using analogy. As I argued, we deduce God's simplicity and so on from His being the terminus of the chain of explanation. The chain either has to continue indefinitely, or be a brute fact, or a simple being. This is not an argument from analogy, but an argument from deduction.

Next, I ought to discuss the final paragraph. There is no possibility in reality, only in imagination. It doesn't seem to further scientific inquiry to maintain that God could have created a different universe. Firstly, physicists don't use their imagination to write down their theories, but their intellect. If we were restricted to what we can imagine, we would never have come up with quantum mechanics. The whole point of the mathematical structure of theoretical physics is to put aside the limitations and vagaries of the imagination. Secondly, it is now part of scientific inquiry to consider other possible universes. It is not only theology that leads us to that, but science itself. Indeed, theology is here more restrictive than science, since theology tells us to expect a fine tuned universe and science doesn't. The only principle of logic that might constrain the physics of the universe is the principle of non-contradiction. And that is simply not sufficient to remove all possibility. Absolute necessity might be more economical, but it is not the universe we live in.

Hume defined a cause as a necessary connection between two things, say A is the cause of B. My objection here is to the word necessary. I certainly think it necessary that (unless B is God) there is a connection. But, like most physicists, I regard the universe as fundamentally indeterminate (meaning only that even if we fully understand the present, it is not possible to make definite predictions about the future). I don't think there is any other reasonably conclusion that can be drawn from quantum physics. So what this means is that given A we can't predict through deduction and calculation that B will result, or when B will result. It need not have been B. It could have been C. But nonetheless, if B results, then A would be the efficient cause of B.

Now I should emphasise that this relates to our ability to make predictions. If it is possible to understand what is to us the future by some means other than making predictions, then one can have a definite knowledge that B followed A. When we look back at past events, we know what happened, even though they were still governed by quantum indeterminacy. That's because we don't know what happened through knowledge of A and a prediction, but because we know that B happened by either remembering or checking some historical record. In the same way, God, sitting outside of time, sees all events not in succession but together. So God knows that B followed A because He sees both of them together, even though knowledge of A by itself doesn't necessarily entail that B will follow.

I maintain that substances have their own final causes. A could lead to either B or C or D or .... But it is not the substances themselves that cause the change to happen. A by itself is not a sufficient reason to explain why B emerged at that particular time. Now, I personally accept that everything needs a sufficient reason (obviously this statement is controversial, but to discuss this controversy here would be a digression). So this means that A alone is not enough to explain why it was that B and at that moment of time. We need something else in addition to A. Clearly, that something else can't be physical, since if it were it would sit in our equations somewhere and physics wouldn't be indeterminate. Thus we are looking at a non-physical agent, and the obvious candidate is God (I recognise that I ought to justify that statement, but I don't want to here as it would be a digression).

Now, that God might choose either B or C or D, and the precise moment of actualisation, indicates that God's will is not determined (with the same meaning that even with complete knowledge, we cannot predict the outcome; however, we can observe the outcome through means other than prediction). In other words, God has a free will. [In fact, I find it hard to combine the principle of sufficient reason and thus a rational and understandable universe and physical indeterminism without postulating a free will outside the material universe which directs matter.]

The comment I quoted above created a juxtaposition between Aquinas, who argued for God's free will in contrast to Aristotle who didn't. Aquinas argued that the universe could be a different size, and Aristotle implied that it had to be this size out of necessity. Famously, Aristotle also argued for an eternal universe, which Aquinas rejected (on theological rather than philosophical grounds), and we reject today on scientific grounds. So the type of necessity by which the universe operates is a hypothetical rather than an absolute necessity, in support of Aquinas and against Aristotle. If the operation of the universe comes down to a hypothetical necessity, then it seems likely that the origin of the universe would be similar. In other words, God could have created a different universe, with different symmetry laws and a different number of dimensions, or chosen to not create a universe at all. It just happens that God choose this one.

We can formulate numerous different possible and self-consistent laws of physics. Each of these corresponds to a different possible universe. We have no good reason to prefer one over the other, except through experiment and observation. If we suppose that the universe was created by God, and God is only constrained by logical consistency, then God certainly could have created any of these universes. Thus we find ourselves in agreement with Aquinas' view that the universe is only hypothetically necessary, only in many more different ways than Aquinas realised.

Of course, this is where the fine tuning argument comes into play. For we need only suppose that God's intention was to create rational animals such as ourselves, and His freedom in how to create the universe is drastically curtailed. So rather than a massive choice over physical parameters and symmetries, God only has a single decision to make, and then a much more limited choice of physics consistent with that decision.

Now come to what I think is the main point of the comment. It argues that the difference between a series of explanations terminating in a brute fact and a series of explanations terminating in God is that a brute fact is defined as something which could in principle be different but happens not to be, while God has to be the way that He is. But if God had a choice about which universe to create, does that mean that God's will could manifest itself in different ways, one of which results in this universe, and another in that universe, and another in no universe at all, and hence that God could manifest himself in different states? Thus there is no difference in principle between the God explanation and the brute fact explanation. Or, alternatively, if we argue that the chain of explanation must terminate with a simple entity, then that entity cannot be God. God is at best an intermediate step.

The first question is whether an indeterminate (in the sense of unpredictable) explanation is still an explanation. Here we are discussing the results of a free will. For example, will I end the next sentence with the word axolotl or the word diplodocus? And even after analysing my style, and pouring in great depth over everything that I write, I think predicting what the word will be is a feat beyond any man, woman or diplodocus. So that's an example of free will in action. So is there an explanation for why I wrote diplodocus there instead of axolotl (or, indeed, some other cauliflower)? Yes: the explanation resides in my free choice, and ultimately myself. I am the explanation for each of those words. So it was impossible to predict what I wrote, but looking back at it, we can see the word and how it followed from my choices as I typed. Now the deed is done, the choice is fixed. There is only one me, a multitude of possible choices, but only one actual choice. My free will is not inconsistent either with there being many choices in principle, nor only one choice in actuality. So it is a indefinite connection in the sense of prediction (going from myself to the word), but an absolute connection in the sense of reading it now it has been typed (going back from the word through my choices to me). The future is indeterminate. But now that I am in the future (compared to when I typed that last sentence), I can see that there was only one thing that happened in practice.

So whether whether a choice is absolute or indefinite depends on how we look at it. This isn't to say that it could be two contradictory things simultaneously, or there are two separate worlds, one where it is one thing, or another where it is another. It only means that we need to be careful with our definitions.

Obviously, the analogy with either time or composition has difficulties with we try to apply it to the timeless God. But, I think there is a resemblance. We are searching for an explanatory principle. That is to say, what we know about is the end of the chain of reasoning (analogous to a later time), and what we are searching for is the start (analogous to an earlier time). God has free will, and could have created numerous different universes. So, just as there is one present and multiple possible futures, from knowledge of God we can't deduce which of the universes will be created. However, knowing the universe, we can induce the existence of the single God. Note also that God shouldn't be identified with His Will. The Will of God is something that proceeds from the Godhead and God's logos (intellect is prior to Will).

So I don't accept that because God could have created different universes, that means that God could exist in different states. Unpredictability does not mean that it is indeterminate in terms of its state.

But how does this relate to the explanatory power of God? If God could have created numerous different universes, how can knowledge of God allow us to deduce that this universe can emerge?

Of course, one does not need to be precise to have explanatory power. Even if we can make some statements about the universe, even if not completely determine it, that is still some progress. Secondly, we need not just postulate God, but also assume some characteristics of God, i.e. a God who wanted to relate to rational animals. Thirdly, the nature of God, and the need for the universe to be consistent and consistent with God's nature, also constrains the possible universes that could have been created. So it is not a complete free-for-all, where anything imaginable could have happened. Thus the possible universes that could arise given that theism is correct is smaller than the total set of all possible universes. God still has a choice to make, but it is more restricted than we might find in an atheist metaphysics. And thus theism still has some explanatory power.

Plato and why Democracy is doomed to fail

Reader Comments:

1. Scott Lynch
Posted at 06:27:43 Thursday February 7 2019

Thoughts on God’s Free-Will

Dr. Cundy,

Another great post. I largely agree with everything you say. I like your take on indeterminacy with regards to God’s free will. I would like to comment on that and also point out a few things you said that I believe Aquinas would disagree with or at least consider ambiguous.

You say:

“First of all, I should note that Christians (at least Western Christians, Roman Catholics and classical Protestants) regard God's intellect as having priority over His will. There is also a distinction between God and the will of God. God is a single, being. The will, on the other hand, proceeds from God. So while it is an element of God, it shouldn't be equated with God as a whole (we are getting into Trinitarian doctrine here, and the distinction between the different persons of the Trinity). So equating God with God's will is problematic.”

Aquinas would disagree with this. With regards to will, Aquinas would say that there is will “in” God (Summa Theologiae Part 1, Question 19, Article 1). Furthermore, Aquinas’ doctrine of Divine Simplicity (ST Part 1, Question 3) necessitates that God’s will is identical with His essence. This follows from the cosmological arguments that demonstrate that God is Pure Act and non-composite. If God’s Will (qua will) proceeds from Him, then he would be composed of two parts, a thinking part and a willing part, or His Will would be some sort of Platonic emanation, which would make His Will a creature. Aquinas would probably see this as a form of Arianism.

Now with regards to God proceeding from God in the Holy Trinity, you are right about that. And the Augustinian psychological analogy typically portrays the Holy Spirit as proceeding from the Love between the Father and the Son. But proceeding from the Love of the Father and the Son is very different than “being” the Love between the Father and the Son. Aquinas would say that all three Persons of the Holy Trinity are consubstantial, which means they all have the same essential attributes (existence, intellect, will, goodness, simplicity, etc.). Aquinas basically argues (following somewhat in Augustine and Boethius’ footsteps) that within the inner life of the Trinity, there is a relationship between the infinite Divine Essence and itself. The infinite Divine Essence is relational, and Aquinas treats relation as a sui generis concept that is different from nature or essence. The most relevant point with regards to the subject at hand is that the relations of the Trinity are necessary (God cannot not be a Trinity) whereas the relationship of creatures to God is contingent. Furthermore, the relationship of creatures to God is a relationship of dependence to the Divine Essence. But the Divine Essence is common to all three Persons of the Trinity. Hence, we cannot reason a posteriori to the Trinity. Therefore, we can be metaphysically certain that the Trinity is a proper mystery (it cannot be metaphysically deduced from natural reason). This is all to say that the doctrine of the Trinity will not be of help when dealing with explanations of our contingent world.

Now there is a loose sense in which one could say God’s will towards the created order is contingent and “proceeds” from God. This is so because the objects of God’s will are contingent (not necessary). But in reality, this contingency or indeterminacy is a property of the thing willed and not God’s will itself. (I think I am right to characterize your use of the word indeterminacy as a special case of contingency with respect to time and initial conditions). I have encountered the argument that you respond to from atheists before. It seems that many atheists either want brute facts or a Leibnizian rationalism (a mechanistic determinism). Either the terminus of explanations for this world is a brute fact or it is something that necessitates the existence of this particular world (in this particular state and at this particular moment). Queue Best Possible World theology. This failure to see the middle ground comes from a lack of nuance as well as a forgetfulness of what got us to God’s existence in the first place (the contingency of the world). The nuance comes from the difference between absolute and suppositional necessity. Dr. Feser gives an excellent explanation of the concepts here:

The basic atheist objection is this. God is identical with His will. God is necessary. Therefore God’s Will is necessary. Therefore anything God wills is necessary. Therefore this world is necessary. But this world is not necessary (according to our observations), so God’s will must not be necessary, so God must not be necessary, so God must be a brute fact. So God can be eliminated by Occam’s Razor.

The problem child in this objection is the sentence “Therefore anything God wills is necessary”. The problem here is that God wills things to be what they are and to have a given nature. Therefore, God does necessarily Will His own Good, for God is absolutely necessary. However, God contingently wills the existence of contingent objects. This defect of contingency is not a defect in God’s Will, but rather a defect in the objects willed. This is where the language of suppositional necessity is helpful. God absolutely necessarily wills Himself, but with regard to contingent things, he necessarily wills them only on the supposition that He wills them. This suppositional necessity comes from the eternal immutability of God’s will (which makes His will such that it cannot be otherwise) and the contingency of the objects willed (which makes it so that His Will is not absolutely necessitated by these objects).

I think it is helpful to start with God only having His absolutely necessary Will towards Himself and then giving Him some contingent objects to will (or not will). What are His options?

1. He wills nothing to exist.

2. He wills every possible thing to exist (an infinity of universes).

3. He wills only the Best Possible Universe to exist.

4. He wills any arbitrary universe to exist.

5. He wills a universe that results in rational creatures that can love Him and appreciate their existence.

We can note that the first four options are consistent with a God who does not have Free Will (although they are not inconsistent with Free Will). Only the fifth option is inconsistent with God not having a Free Will. These first four attempts neglect that crucial distinction between absolute and suppositional necessity. Many people in failing to see this distinction have posited three of these options in order to maintain the Principle of Sufficient Reason (no one really argues for option 1 because it is obviously false).

Let’s therefore examine these three options and see how they fair. Options 2 and 4 do not seem possible because they do not account for teleology and fine-tuning. Of all of the possible worlds that exist, there are far more (infinitely more) chaotic worlds than orderly ones. Therefore, it is a statistical certainty that we should be in such a world and thus we should see all of the laws of physics scramble to unintelligibility (leading to the immediate destruction of the universe) in any given second. The fact that we have been lucky so far does not make a difference just as the fact that we won the lottery a thousand times in a row does not increase the odds of us winning the next lottery (which is one in 300 million assuming it is a fair game). Likewise, even if a random TV signal gives us five hours of the Pride and Prejudice mini-series, it does not change the fact that it is infinitely more likely to produce an unintelligible mess of pixels in the next second rather than the next scene of Pride and Prejudice.

That only leaves option 3 left. Dr. Feser does a great smack down of the Best Possible World logic in Five Proofs of the Existence of God. The basic refutation is that better and best imply Goodness, so these can be rephrased as more good and most good. But goodness and existence are convertible. So the only thing that could be the most good would be God Himself who is Goodness itself (and hence the most good). Any contingent thing or set of contingent beings will necessarily be in potency to be better. A universe with one trillion perfect, intelligent, sinless humans and angels with no fallen angels or humans is still far worse than a universe with one quadrillion perfect humans and angels. We could go on ad infinitum. The basic principle is that no finite number is greatest. Hence no finite good is the best possible good.

But if God is going to create at all (which He obviously has), He has to start somewhere. And that leaves us with option five (which is your conclusion) that God chooses some world that leads to rational creatures that can appreciate their existence and love God. Now there is a sense in which from the creature’s subjective point of view, we can talk about best possible worlds. For instance, any world in which I personally exist and have free-will is best in the sense that I would not experience it at all if contingent things had not happened that led to my existence. Some of these contingent things were undoubtedly awful (perhaps an ancestor from 3,000 years ago conceived another one of my ancestors from an act of rape). Had all of these contingent events not occurred (including the terrible ones), I would not exist today. If we look at the perfect sinless world example, well that is a world where I, a sinner, would probably never have been created.

So from God’s perspective, He could have done much better. But there is a sense in which, from the creature’s perspective, He could not have done better. But God does not need us at all, so He is not creating us for His sake but for our sake.

And God’s selfless love is really at the crux (pun intended) of Christian philosophy. What is the explanation for God’s fine-tuning the universe? He wanted to create rational creatures. Why did He create a universe that allowed me specifically to exist?

"Before I formed you in the womb I knew you, before you were born I set you apart...” - Jeremiah 1:5

We were all specifically chosen by God’s unconditional love for us out of an infinity of possibilities. I will take that over brute randomness any day of the week.

2. Ficino
Posted at 15:29:27 Thursday February 7 2019

Thanks for this post, Nigel. It will take a while to digest it all.

Since I just wrote up some passages where Aquinas says that God's will is His essence or being (essentia or esse), I will supplement Scott Lynch's references now and come back later for the rest.

God’s will and knowledge are distinct only logically, not as a real distinction. Just as God’s knowing is identical with His essence (“sua essentia est suum intelligere,” ST 1a 87.3 co, cf. SCG I.45 et alibi), so God’s willing is His essence or being (“sicut suum intelligere est suum esse, ita suum velle,” ST 1a 14.1 co., SCG I.73, et alibi). God’s goodness is the same as His will in reality [i.e. distinct only logically] (In I Sent. d. 45 q. 1 a. 1 ad 3). God’s action or operations in turn are identical to His willing (“eius velle est eius agere,” SCG II.31.5, In II Sent. d. 15 1 q. 3 a. 1 ad 3, et alibi). In fact, God’s action is not an accident but is His substance, De Potentia 3.17 ad 30. The principle is this: If God is not identical with His acts, God must be related to the act as potentiality is to act. But then there will be potentiality in God as well as actuality, which is impossible, SCG I.45.5.

3. Nigel Cundy
Posted at 20:44:21 Thursday February 7 2019


Thanks both of you for your comments and corrections, and reminding me to be more careful.

4. Scott Lynch
Posted at 03:19:53 Friday February 8 2019

No problem! I don’t think many people are expecting analytical precision in an addendum to a blog post. But I thought I would point it out just for good measure.

These posts at very helpful for me. I recently got your book and hope to start reading it in a few months.Keep up the great work!

5. Ficino
Posted at 18:58:56 Friday February 8 2019

Nigel, I'll try to respond more to yours above.

"As I argued, we deduce God's simplicity and so on from His being the terminus of the chain of explanation. The chain either has to continue indefinitely, or be a brute fact, or a simple being. This is not an argument from analogy, but an argument from deduction."

1. I referred to analogical predication, not to arguments from analogy, when I said we do not have direct access to God's essence. You mention God's simplicity. It is already a doctrine of Thomism that God's simplicity is unlike ours, since there subsist real distinctions with the God who is "entirely simple". In a human there is no direct correlative. So even based on what is asserted in Thomism, that example exemplifies our lack of direct apprehension of God. To have a valid demonstrative proof, according to Aristotle (AnPo), each term must be predicated univocally. That doesn't happen with "God" terms. It's not clear that theology should even aspire to being a science in the A-T sense, since it's demonstrations employ what Aristotle calls "pros hen" predication, which is not univocal. I understand that Scotus and others, unlike Aquinas, said that [some?] predicates are univocal when applied to us and God, but I haven't studied Scotus.

2. I don't press the term, "imagination." Another word may better express my view that a number of ancient Greek philosophers held that the universe is as it is by necessity. So another kind of universe, or an absence of a universe, would not in their view be a real possibility.

3. "Thirdly, the nature of God, and the need for the universe to be consistent and consistent with God's nature, also constrains the possible universes that could have been created. So it is not a complete free-for-all, where anything imaginable could have happened. Thus the possible universes that could arise given that theism is correct is smaller than the total set of all possible universes."

I'm not sure I see this above as a gain unique to classical theism. The PNC and other axioms rule out universes that admit of defeat of the PNC, for example. And we can be confident of less about God's nature than it may seem, given both the problem of predication and at least the Thomistic tendency to make transcendentals inter-entailing - so that existence and goodness are convertible perfections, such that ANY state of affairs, if it exists, is also good at least qua existing. I am not at all sure that a universe, say, in which far more gratuitous suffering occurs than in our own cannot have been created by the God of theism. Or a universe populated only by immaterial substances. Or ...

I appreciate your point that perhaps the motive for belief is not in order to get an Ur-explanans of everything. Yet, VERY many arguments I see for theism, when pushed, seem to end with a claim that if we don't accept theism, then our world-view can only be radically irrational, no science is possible, etc. So the need to have an Ur-explanans outside the universe does seem a major motive for belief.

And I'm no further ahead than I was when writing the comment on your last blog post. Suppose lightning strikes one out of a grove of trees and burns part of the wood. Why THAT tree and not any of the others? Scientists can talk about the structure and properties of lighting and wood and fire and whatever else is in play, and if someone had enough data, perhaps that person could trace the efficient causality to explain why the lightning hit that tree and not another tree. But today's scientist wouldn't be invoking final causality to explain the efficient causality. What would the Thomist add by invoking "aimed at the determinate end"? Beyond the Aristotelian's talk of lightning and tree and fire as actualizing their forms (the tree getting rather a rum do), the Thomist is compelled to explain, not only the generic operations of the natural things, but those operations as directed to this singular end, as caused by a directing intelligence (cf. e.g. SCG III.24, esp. 24.5). Yet, to the question, "but why was THAT tree burned by lightning?", the answer from the POV of *final* causality cashes out as "because it was God's will." Am I wrong? Certainly with his doctrine of divine providence, the Thomist can't admit only an Aristotelian intrinsic final causality of typical operations, on which particular outcomes are beneath the notice of the Unmoved Mover.

So I'm still not seeing the outcome beyond "God did it." Maybe that's the best place anyone can end up, but since "God did it" doesn't answer those pesky "But WHY did God do that???" questions except with a kind of tautology (e.g. whatsoever God does exemplies His goodness), I'm not seeing the payoff as worth the cost of positing various speculative entities not invoked by chaps like Carroll. You can't know, for example, that God won't cease to conserve some entities in existence without warning... (cf. De Pot. 5.2).

I'm used to being told that I'm equivocating or just fail to understand Aquinas (heh heh), so perhaps some day the light will dawn.

6. Ficino
Posted at 15:37:17 Saturday February 9 2019

Just came across this by Peter van Inwagen and thought it relevant:

"One of them is the love of one person for

another, and a second is two or more persons working together in love to give a gift to

someone whom they both love. This fact does, incidentally, provide Jews and Muslims

with an answer to the question, “Why did God create a world?” Trinitarians must regard

this as a mystery. To the question, “Why would God bother to create anything?”,

Trinitarians have no answer. For Trinitarians, a possible world in which God creates

nothing is as good a world as any other world."

7. Scott Lynch
Posted at 18:16:17 Saturday February 9 2019

Response to Ficino


With regards to analogical predication, you can make inferences about divine attributes because all analogical predicates are partly univocal and partly equivocal. John Duns Scotus basically agreed with Aquinas, but he did not like to use the word analogical. It is similar to how Aquinas does not use the term “formal distinction”, but his doctrine in real distinctions includes the concept of a formal distinction.

So for example, we cannot know God’s intelligence in the sense that we can do psychoanalysis on Him and perform brain scans on Him. God’s intelligence is not like that. However, we can know that, whatever God’s intelligence is, He can grasp universals and individual objects (at least well enough to create them and sustain them in their particular way of being). In order to allow the Principle of Sufficient Reason, we have to allow for something like intelligence in God.

As far as “Why God would make anything”, the Classical Theist answer would be the same regardless of belief in the Trinity. All classical theists (Jews, Christians, Muslims, philosophical theists) believe God is totally self-sufficient and does not need anything or anyone else. Creation is totally superfluous from God’s perspective. You are right that a created world with nothing but inanimate objects has some goodness, but none of the inanimate objects are aware of their existence, so there is no point to their existence. From God’s perspective, their existence is worthless, but also from their own “perspective” (since they have no perspective), their existence is worthless. At least a world with only irrational animals would have some worth from the perspective of creatures. However, the greatest universe (in kind if not necessarily in degree) is a universe with rational animals and rational spirits that can appreciate their own existence and have immortal life. So as Dr. Cundy says, Theism does not make this universe necessary, but it does explain its existence and it also gives us a decent idea (even if not a complete idea) as to why our particular universe exists. In order to know exactly why our particular universe exists out of the infinite alternative universes with rational creatures, we would have to comprehend God’s essence as well as all of creation (i.e. be omniscient). Aquinas’ doctrine of analogical predication guarantees that that will never happen.

Posted at 19:51:43 Saturday February 9 2019

@Scott Lynch: this may well be quite familiar, but I write it for completeness' sake. Aquinas' analogical predication of names of God is not constructed on the lines of Aristotelian proportional analogy. Proportional analogy is often exemplified so: the pounce is to the cuttlefish as the spine is to the fish as the bone is to the land animal (e.g. AnPo II.14, 98a20-23). We can note that, strictly speaking, it is the things that stand analogically to each other; the predications name the things.

Aristotle discusses another way in which something may be "said in many ways," i.e. "pros hen" predication, called "focal meaning" since Gwil Owen. These are cases of equivocation, but not of homonymy as with the referents of "dog, dog fish" and "Dog Star." The standard examples of pros hen are: "healthy" in reference to the animal, urine, and a drink; "medical" in reference to the doctor and to his instruments and works (EE 1236a15-b26, Meta. e 1030b2). In pros hen predication, there is a primary analogate and secondary analogates. The relation between the primary and secondary analogates is non-reciprocal. They contain its meaning, but it doesn’t contain theirs; its significata are logically prior to theirs. Doctors are logically prior to every other “medical” thing. “... a medical instrument is one that a medical man would use, but the definition of the contained is not implied in that of ‘medical man’.” This is why pros hen predication is a kind of equivocation.

What would happen if we did not know what a doctor was, but only were told that some instruments or works are 'medical'? Or if we did not know what health in an animal is, but are only told that some urine and a drink are 'healthy'? I don't think we could have certainty of propositions about the primary analogate in itself if we can only try to construct knowledge about it from the secondary analogates. As Being itself, God must be the primary analogate. But if we're arguing from some F predicated of a secondary analogate, we can't say that a conclusion about God as F has been demonstrated with certainty.

This is why I said for a deductive system to be a valid demonstration, it has to employ terms univocally. If 'good' in an argument about God is not predicated univocally, then there is danger of a quaternio terminorum, and we can't have confidence that its conclusion has been demonstrated with certainty as true (cf. AnPo I.11 77a5-9). Things that are analogical will have a middle term that is analogous, but this is on proportional analogy; there is no middle term if terms are equivocal (AnPo II.17 99a7-17).

I don't find that Aquinas' discussions of analogical predication create confidence that despite the above, his arguments using analogically predicated terms demonstrate their conclusions with certainty. Cf. e.g.:

ST 1a 13.5 co: Aquinas justifies analogical predication of names of God by saying that without it, if predication were equivocal, we could not know God nor demonstrate things about God from creatures. This argument from consequences pretty clearly begs the question, as is clear from St. Thomas' evidence: the fact that philosophers and the Apostle Paul say that truths about God can be demonstrated!

ST 1a 13.10 ad 4, Aquinas says that Aristotle accepts equivocation ‘largo modo accipit aequivoca, secundum quod includunt in se analoga.’ Here Aquinas admits, using the indicative, that analogically predicated terms are a kind of equivocation, 'aequivoca.'

Obviously, this is a huge topic with a long history. I can't see, however, how analogical predication can be squared with claims that A-T arguments demonstrate truths about God with certainty. Since we don't know the primary analogate in advance, we can't evaluate whether the "partly univocal" elements secure a middle term in a deductive system that uses pros hen predication.

9. Scott Lynch
Posted at 21:22:55 Saturday February 9 2019

Second Response to Ficino


You are right that in general you would need to know the definition of a term in order to make an a priori analytical deductive proof. Proofs for God’s existence are not like that, however. Proofs for God’s existence are a posteriori deductive arguments. Essentially, you have data that you start with (thoughts, perception, change, subsistence, unity, composition, repeatability, etc.) that cannot be coherently denied.

Then you have to posit an entity with certain bare minimum attributes that will explain said data. So in order to explain change, you must posit something that is Pure Act. In order to have a contingent, non-necessary universe, you have to posit some non-contingent entity with something like what intelligence and free-will is in us.

Just because you cannot provide a strict definition, does not mean you cannot know that something has a bare minimum attribute that at the very least encompasses certain finite attributes.

For example, if a 100 kilogram iron kettlebell floated from the ground to one meter above the ground and then stopped, you would not know the cause, but would at least know that whatever the cause was had at least 9800 joules of energy stored in it (perhaps it was an out of sight electromagnet hooked up to a battery). Of course that does not preclude the magnet-battery system from having much more potential energy at its disposal (stored in its battery). It may not even be an electromagnet, perhaps it is an angel acting on the weight, but whatever it is, it must at least have the power to exert 9800 joules of energy.

Similarly, we cannot know exactly what God’s intellect is like, but we can know that, whatever it is, it must at least be able to grasp enough universals to create and sustain a particular non-chaotic universe out of an infinity of logically and metaphysically possible chaotic universes.

Since these are metaphysical deductions, and not merely logical deductions, they are in one not as logically airtight as logical deductions. But once you accept certain metaphysically self-evident premises such as the Principle of Sufficient Reason and the reality of change, they do become logically airtight deductions.

The problem many people have is accepting the metaphysical premises. However, these premises can be shown to be self-evident by retorsion and the principle of explosion. That is to say, if these premises are deemed false, then any truth system is immediately undermined.

10. Ficino
Posted at 11:43:15 Sunday February 10 2019

@Scott Lynch:

SL: "You are right that in general you would need to know the definition of a term in order to make an a priori analytical deductive proof. Proofs for God’s existence are not like that, however. Proofs for God’s existence are a posteriori deductive arguments."

The question, whether a term in a deductive system is predicated univocally, equivocally or analogically (in the scholastic sense, i.e. as "pros hen" predication) in different premises, concerns any syllogism, whether its premises are arrived at a prior or a posteriori. I think you acknowledge this implicitly when you refer to "a posteriori *deductive* arguments." Any deductive argument has to have a distributed middle term in order to go through. Pros hen predication creates a problem, because the middle term will not have the same sense in every premise.

Deductive syllogisms play a major role in the structure of a science as Aristotle conceived of it, and as Aquinas did also, following Aristotle. Aquinas wrote a commentary on the Posterior Analytics, in which the saint expounded the role of the syllogism in a "demonstrative science."

SL: "Since these are metaphysical deductions, and not merely logical deductions, they are in one not as logically airtight as logical deductions."

I don't understand this. Deduction is an operation in logic, whatever the intellectual discipline that is employing logic in its argumentation. A metaphysical system, whose conclusions are reached by fallacious or unsound logical arguments, will have conclusions that are vitiated by faults of logic. At a minimum, its conclusions cannot justifiably be presented as certain.

On retorsion: it's generally maintained in the literature that retorsion arguments against a thesis do not establish the truth of one's own position. A retorsion argument against, say, "scientism" a la Richard Dawkins doesn't establish the truth of the opponent's own system. As I said above to Nigel, I think you too are right to suggest that the PSR is a place where many of these discussions wind up.

My point about analogical predication was not by itself to deny the PSR but to deny that Thomistic arguments, in which terms are predicated analogically of God and creatures, should properly be considered demonstrations, in the classical sense, that yield conclusions known with certainty.

11. Anthony
Posted at 15:25:17 Sunday February 10 2019

a question on necessary existence

This question might be a bit off topic, but nevertheless I want to ask it. Also, excuse my lack of knowledge regarding contemporary physics. The issue I have is with regards to God's necessary existence. Since God's essence and existence are identical, He exists necessarily. But the same thing could be said of matter as well, as it seems to me. Why could it not be the case that some fundamental layer of physical reality necessarily exists aka that it cannot fail to exist? Suppose that we are dealing with some form of "atomism". How do we know that some kind of physical particles or fields don't have necessary existence? Existence could be a PART of particles/fields or some other kind of fundamental physical reality. When I say that existence could be a "part" of particles/fields, I mean by this that existence could be a "part" of the ESSENCE of particles/fields. What it means to be a particle/field (to exemplify properties that are essential to particles/fields/some other kind of fundamental physical reality) ENTAILS that the particles/fields/(fundamental physical reality) EXIST. Because, there could be a set of determinable physical properties that simply could not fail, in reality, to be instantianted out of ontological necessity. Hence, a particle/field would have to exist necessary in order for there to be anything in reality at all. Then, what MEANS TO BE a particle/field ENTAILS, among other things, that this particle/field EXISTS.

In order to exist necessarily a thing would then have to be a kind of thing that has certain determinable physical properties (aka charge or mass or whatever...) and those properties then would be properties that necessarily have to be instantiated, because something HAS TO exist necessarily (since I believe that NOTHINGNESS as such is metaphysicaly incoherent). So, why not think that there is a set of determinable physical properties that out of ontological necessity cannot fail to be instantiated?

Even if hylomorphism is true, and anything physical is a composit of matter and form or an actualised potential of prime matter, or better to say, physical reality is INFORMED matter, then why not think that there are particles/fields or some other kind of physical reality that NECESSARILY cannot fail to EXIST aka that there is some ontological necessity that there has to be INFORMED matter in the form of particles/fields? Just because something is a COMPOSIT that doesn't mean that it can FAIL TO EXIST as it seems to me at least, I don't know. Because I don't see a reason why there cannot be certain composites that have to exist necessarily? Why can it not be that some kind of composition has to be necessarily instantiated?

Hence, are there any good reasons to prefer God as a necessary being aka a being who's essence and existence is identical and who cannot fail to exist, to some kind of physical reality? Why can it not be that some kind of physical reality necessary has to exist? How can we argue against the view that atheists present to us, that there is some kind of fundamental layer of physical reality, some physical thing with physical properties, that has to exist necessarily?

Again, I am sorry for being tedious with my questions, but I simply don't know whom should I ask and what to read because nobody deals with this question when it commes to arguments for a necessary being. Also, I hope that my question was clear, I tried to explain the problem as best as I could. I would appreciate any help regarding this issue. Thank you very much!

12. Anthony
Posted at 15:27:51 Sunday February 10 2019

It seems that something went wrong with my previous comment, but I hope that it is still intelligible

13. Nigel Cundy
Posted at 22:41:13 Sunday February 10 2019


Sorry for not joining in this discussion. Real life has intervened, and robbed me of my weekend, so I have fallen behind schedule. I have a few other things to deal with first, and then I'll come back to this discussion.

14. Scott Lynch
Posted at 07:30:35 Monday February 11 2019

Third Response to Ficino


When I said metaphysical deductions, what I meant was that these are a posteriori deductive proofs. That is to say, they start with a conclusion and end with a premise. This would be opposed to an a priori (purely logical) deduction that reasons from premise to conclusion. Both ways of reasoning are valid. I really do not understand your issue. Are you saying that it is impossible to make any kind of logical argument unless we fully comprehend everything about the terms we are using? If I know all men are mortal, and Socrates is a man, I know that Socrates is mortal regardless of whether or not I know if men are rational or how much Socrates weighs.

Take the following argument:

Axiom 1: We perceive data that both changes and maintains some consistency (repeatability, form, subsistence, teleology, take your choice of words).

Axiom 2: PSR is true.

Premise 1: Since the data from Axiom 1 is contingent, it must be ultimately caused by something necessary (from PSR/Axiom 2).

Conclusion 1: So the cause of our initial observational data must be a necessary being that is able to cause change to and impart teleology to our data.

Premise 2: But the only thing that could be necessary and impart change and teleology is something that is Pure Act (I am leaving out the details of essentially ordered series etc. for brevity) and that has the ability to impart specific forms and teleology to objects out of an infinity of possible forms and teleologies (or no teleology at all).

Premise 3: But the only thing that can impart said forms and teleologies is something like a mind and a free-will.

Conclusion 2: Therefore, the cause of our data is a necessary being with an intellect and free-will that is also Pure Act (from C1, P2, and P3).

Now you may disagree with the premises, but it will not due to say that the logic is invalid. The key is to rule out other factors (such as premise 3). Once all of the premises are defended as being self-evident or reducible to self-evident principles, the argument is shown to be sound as well.

Just because I cannot start with a definition of God and reason to the created order does not mean I cannot start with the created order and reason to a cause with certain minimum attributes.

As for retorsion, it only does not affirm the truth of your position if you are committing a bifurcation fallacy. If there are genuinely only two options, then disproving one proves the other (by the Law of Excluded Middle).

I think it is something to consider that so many atheists deny PSR. If the logic of Thomistic and other classical Theistic arguments for God’s existence were so tenuous, don’t you think atheists would be picking that apart rather than denying things (like PSR) that allow for ridiculous conclusions?

15. Scott Lynch
Posted at 07:48:38 Monday February 11 2019

Reply to Anthony


I would look up Dr. Edward Feser’s article on Existential Inertia and the Five Ways online or in his book Neo-Scholastic Essays.

Basically, he argues that any basic material constituent is not going to be necessary because of its ability to change and its finitude (e.g. three-dimensional as opposed to 27 dimensional or one dimensional, or having charge as opposed to color, etc.)

Dr. Feser would argue that all of these attributes are only explicable as being a composition of form and matter. However, barring that, a fundamental constituent is still going to have some teleology or “way of being”. And unless we are going to deny the Principle of Sufficient Reason and day that these objects’ ways of being are brute facts, then we need an explanation for their finite behavior. Basically only the infinite and the simple, the Actus Purus is the sort of thing that can in principle be necessary. Everything else is either dependent on such an entity or else is a brute fact. Then ultimately we get into the debate on the PSR.

16. ficino
Posted at 11:15:07 Monday February 11 2019

@Scott Lynch:

An example of a deductive argument involving pros hen predication would be this:

"For all liquids, if the liquid is healthy, the liquid is good to drink.

This animal's urine is healthy.

Therefore, this animal's urine is good to drink."

"healthy" has a different signification in the major from that in the minor. In the major, "healthy" means "promotes/produces health in the drinker." In the minor, "healthy" means "a sign that the animal is healthy." You could represent "healthy" in the major as H1 and in the minor as H2. The deductive system is vitiated by an equivocation fallacy and does not go through, because H1 and H2 have different significations.

It may not be the case that this animal's urine is harmful to drink, but the conclusion that it is good to drink has not been demonstrated by this argument.

I picked healthy drink and healthy urine because these are classic examples of pros hen predication in both Aristotle and Aquinas. It can be discovered whether an animal's urine is healthy for (humans) to drink, because a lot is known about the primary analogate, health in the human. So there can be other ways, apart from reasoning w/ pros hen predication, of determining whether urine called "healthy" because it's a sign of the animal's health can also be called "healthy" because it will promote/produce health in one who drinks it. The meaning of the primary analogate, "health in the animal" is known.

The problem with pros hen predication of names of God is that we know what the terms mean - despite some arguments over semantics - when they describe creatures. But we are then told that God is so transcendent that His attributes are fundamentally unlike ours, so that, we are told, terms for those attributes are not predicated of God and of creatures univocally. "Good" will not have the signification when predicated of God that it has when predicated of creatures. A deduction in which one premise contains a term with the value G1 and another premise contains a term with the value G2 will lack a distributed middle term if the middle premise contains G1 or G2, and it can't employ simply G as though G1 and G2 are univocal.

The conclusion of an argument, whose terms are not predicated univocally, may not be false. But we are not warranted to say that the truth of that conclusion has been demonstrated with certainty by the argument. So we are not warranted to say that a discipline that makes heavy use of non-univocal predication of terms is a demonstrative science.

Some people say the above is not really a problem for Aquinas' system because Aquinas is giving us an analogy of being (e.g. Thos. Joseph White). But that reply would misconceive the issue, which is how we try to demonstrate the truth of statements about things that exist. Aquinas himself says that the order outside us is followed by the order in our knowing of things, Comm. in Meta. V l. 1 C759.

17. Scott Lynch
Posted at 18:37:37 Monday February 11 2019

Fourth Reply to Ficino


The problem with your example is that you are treating your terms as univocal as opposed to analogical, and so your argument fails. When you are making arguments with analogical predicates, you have to move from the specific to the general or the general to the specific. But being good to drink is too specific of a predicate to apply to health in general (obviously healthy human beings are not good to drink), so your first premise is not sound. Take this argument instead.

Premise 1: For all liquids, if the liquid is healthy, it has been produced by something with enough relevant health to produce it.

Premise 2: This animal’s urine is healthy.

Conclusion 1: Therefore said animal is healthy enough (proper kidney function, etc.) to produce healthy urine.

Notice I had to say “healthy enough” because healthy urine is not enough to know whether an animal is healthy in general. But it is at least enough to know that the animal is healthy enough to produce it (by the Principle of Proportionate Causality and PSR).

Obviously an animal could have a brain tumor and still produce healthy urine.

But arguments for God’s existence are sufficiently general to be valid in this way. Thomists are not saying that God produces intelligible effects, His intelligence is exactly like this or that. All Thomists are saying is that God produces intelligible effects, therefore God must have some attribute that explains said effects (if PSR is true). Now human intelligence (being able to grasp universals and have specific determinate intentions out of an infinity of possibilities) coupled with free-will is the closest thing to being able to produce intelligible effects, so God must have something at least as capable as the human intellect (but in reality much more capable considering the scope of His causal power) in order to produce intelligible effects.

I just feel like you are not understanding the way analogical predication is used by Thomists. I have never heard this objection that you raise before, so I wonder if this is a common objection. Do you know if any literature that gives objections and rebuttals from Thomists and non-Thomists on this point?

18. Ficino
Posted at 20:05:35 Monday February 11 2019

@Scott Lynch, this is all fascinating stuff, and the more we try to say about it, the longer the posts we may have to post! And I have work I am supposed to be doing, as I'm sure you do, too.

1. I proposed an example of an argument rendered invalid by a fallacy of equivocation that arises from pros hen predication. You offered a different argument, in which pros hen predication does not arise. So I don't think your example is apposite. Your example is also not apposite because, like valid arguments about urine that I allowed can be made, your example supposes that we know what an animal's health is. That is the primary analogate, with reference to which "healthy" has some connection when applied to urine.

But we do not start off knowing God's nature as we do knowing health. That is the heart of the matter.

2. You say that when making arguments with analogical predicates, we have to move from the specific to the general or from the general to the specific. Perhaps you are thinking of proportional analogy? That is different from the pros hen relation. As I noted higher up on the thread, the classic example of proportional analogy is pounce is to cuttlefish as spine is to fish as bone is to land animal. The genus is something like "internal supporting bodily structure". The three terms are species of that.

The pros hen relation is unlike. Is it not the case that a proportion obtains among healthy animal, healthy urine and healthy drink (or among surgical man, surgical instruments and surgical operations). Healthy urine and drink are healthy in a secondary way, as predicates that are parasitical upon "healthy" as applied to the animal. You have to know what health in an animal is before you can understand the sense in which we can say its urine is healthy or that a drink is healthy for it. The secondary analogates are secondary because their meanings are linked to and dependent on a core notion. Pounce, spine and bone, on the other hand, are all equally species of "internal supportive bodily structure."

3. I also don't think that movement from general to specific is a feature at all in propositions about God. God is not a mega-genus under which there are species of anything. God is in no genus at all, as we both know, in Thomism. The key feature in propositions about God is that His perfections control what is true of us, as the root concept, healthy animal, controls derivative senses of 'healthy'. Maybe the health-urine example doesn't quite align with God's goodness vs our goodness, but it is exactly the example Aquinas when he discusses analogical predication of names of God in ST 1a 13.5: "“the name which is spoken of in many ways signfies different relations to some one thing; just as ‘healthy’, spoken about urine, signifies a sign of health of the animal, but spoken about medicine indicates a cause of the same health.”

I know what Thomists are saying. I'm questioning whether they are justified in proposing their system as a demonstrative science when it does not predicate key terms univocally in many of its arguments.

From the literature, I think a pretty clear statement of the problem is in this article comparing Maimonides' negative theology with the less negative theology of Aquinas, by Isaac Franck, “Maimonides and Aquinas on Man’s Knowledge of God: A Twentieth-Century Perspective,” RevMeta 38.3 (1985) 591-615 at 595, re Maimonides’ view, “(1) For an attribute X (e.g., "good") to be predicated analogically of two entities, e.g., God and man, it would be necessary first to know that the attribute under discussion as it applies to one of the entities in the analogy is the same as, similar to, or overlaps with the attribute as it applies to the other entity in the analogy. There can be no analogy unless at least such similarity or over lapping exists. However, since we cannot possibly know what the attribute X would mean in its application to God, we have nothing on which to base an analogical ascription of attribute X to both God and man...” 596 “attributes are predicated of God results precisely because the radical otherness of God makes all comparisons between God and other things utterly impossible, and any term used both about God and about other things therefore cannot be used univocally. There indeed is no middle ground here between univocity and equivocity.”

Obviously, Aquinas tried to cut a middle ground!

Some other literature I've read:

Austin Farrer, Finite and Infinite (London: Dacre Press, 1943)

Ralph M. McInerny, The Logic of Analogy: An Interpretation of St. Thomas (The Hague: Martinus Nijhoff 1961)

Anthony Kenny, The Unknown God. Agnostic Essays (London/New York 2004)

On my list:

David B. Burrell, Analogy and Philosophical Language (New Haven: Yale 1973)

Humphrey Palmer, Analogy (London: Macmillan 1973)

I have read a review article saying that Burrell defended Thomistic analogical predication of names of God, and Palmer attacked it.

19. Anthony
Posted at 21:51:25 Monday February 11 2019

Reply to Scott Lynch

Thank you Scott for your reply. I thought also something among the lines of theleology and the PSR. Maybe there are also other arguments against the necessary existence of material things, and maybe there are even some usefull insights from physics regarding this issue, I don't know

20. Scott Lynch
Posted at 04:44:21 Tuesday February 12 2019

Final Reply to Ficino


Thank you for your comments and literature recommendations. I will do some more reading on the subject.

I will let you have the last word (I also cannot spend too much time engaging in a blog discussion).

I would just ask, can we predicate existence of God (assuming First Cause arguments are successful)? To say we cannot predicate existence of God is to essentially say that First Cause arguments fail. I would argue that all cosmological arguments for God’s existence are ultimately reducible to the Principle of Non-Contradiction and the Principle of Sufficient Reason (and you concede these principles at least for the sake of argument).

So if we can predicate existence of God, how can we do so? Well it cannot be an equivocal predication, for that would simply mean that God does not in fact exist. However, it also seems that it cannot be a univocal predication, for God’s existence is necessary, self-sufficient, and underived whereas our existence is contingent, participatory, and derivative. So it seems that the only thing left is some sort of middle ground. The First Cause has existence of a sort, but it is not quite like existence in us.

If existence can be analogically predicated of God, then certain other attributes should also be applicable. The trouble is that we have to be careful in considering our attributes, because not all attributes can be analogically predicated in the sense we need (for example triangularity could not be analogically predicated of God). Such attributes must be sufficiently general and not inherently limited. I would argue that existence, intellect, will, and goodness are not necessarily limiting terms.

21. Scott Lynch
Posted at 04:48:09 Tuesday February 12 2019

Second Response to Anthony


I think that is the crux of Dr. Cundy’s work. His point is that there are many other logically coherent physical theories. This means that our current universe is not intrinsically necessary and hence needs an explanation.

22. Red
Posted at 12:27:05 Tuesday February 12 2019


This article might be of interest to you.

23. ficino
Posted at 12:39:19 Tuesday February 12 2019

@Scott Lynch:

just two quick expressions of opinion (feeling guilty because I'm supposed to be reading about Aristotle's Meta. Lambda, heh heh):

1. as far as I know, existence is not a predicate.

2. re your penultimate post: to me at this stage in my thinking, it doesn't seem that Aquinas can have all three of the following: analogical predication by the "via eminentiae"; God not in any genus; convertibility among all the transcendentals.

24. Anthony
Posted at 15:26:54 Tuesday February 12 2019

response to Red

ouh I didn't know that Rasmussen wrote about this topic... I will check it out later, thanks.

25. Nigel Cundy
Posted at 22:30:52 Thursday February 14 2019

Can something material be a necessary being?

Thanks to all of you for the good discussion. I'm writing a quick reply to Anthony's comment first; I'll need to think a bit more about Ficino's questions. I note that the thrust of the argument below is similar to Dr Feser's, namely material beings are capable of change while a necessary being isn't. My reasons for getting to that point are different. Apologies that I have drafted this quickly, and that it is not as well-thought through as I could make it had I a bit more time.

  1. The first thing to do is to define the terms "material being" and "necessary being".

  2. A material being is composed of form and matter, and exists within space and time (plus any additional dimensions that might be out there). It thus cannot be captured solely by an abstract concept (although it can be partially represented abstractly). In particular, material beings are the objects studied by (mathematical) physics.

  3. A necessary being is something which, given its nature, is incapable of

    having an efficient cause or something outside itself to explain it. This is the opposite of a contingent being, whose existence or explanation depends on some other factor outside itself.

  4. All direct interactions in physics are local, in that they occur in one particular point in space and time. (Of course, indirect interactions need not be local, for example if there is a particle which mediates the interaction). This is certainly true of all current theories of physics. It (or rather, its consequences) has also been tested to a very high precision experimentally.

    It is, however, more general than just our current theory. We begin by mapping from a physical space time to a geometrical and then a coordinate representation of the space time. As we do so, we introduce redundancies, such as the origin of the coordinate

    system, the direction of the coordinate axes, and the scale used to measure length. None of these redundant features of the coordinate system can play a role in predicting physical events. The absence of any natural origin to the universe means that the interactions between

    material beings cannot depend on an absolute location in space or time (i.e. distance from an origin). That means that interactions can only depend on the relative distance between particles. However, it equally cannot depend on the distance alone, because distance is not a dimensionless quantity; one would need to introduce an absolute measure of distance to compare it against. The only exception to this is if the distance is zero, which implies a local interaction.

    Now, obviously, historically, there have been physical theories which seem to imply action at a distance, such as Newtonian gravity. However, these have been

    superseded by field theories, where an intermediary is placed between the two

    bodies, namely the field.

  5. The principle of locality coupled with the universality of physical law

    implies the conservation of energy and momentum. [In classical physics, the conservation of energy and momentum arises as a consequence of the translation symmetry in the universe; in quantum physics it arises from locality directly. It's difficult to explain this without going into the mathematics. The mathematical description can apply to any theory which allows transitions from one state to another.]

  6. This means that for any material object to interact with something, it must change its state. [The exception would be something with infinite energy, but

    infinities are problematic in physics, especially since it would generate an infinite gravitational force.]

  7. Therefore any material object must either be capable of change or incapable of interacting with other beings.

  8. If it is incapable of interacting, then we have no way of observing it or being aware of its existence, and it cannot be the terminus of the causal chain (which by definition interacts with the next member of the chain).

  9. Change is the movement from one state to another. Changes in state for a being which involve a change in energy or momentum require interactions with something outside itself (either creation of that second being, or its destruction, or some other change of state).

  10. Every material being carries energy, and therefore its creation involves a change in energy.

  11. Every being must exist in some state. Every state that has been changed is explained in terms of something outside itself (for example a previous state of the same being, coupled with whatever it interacted with to produce that change).

  12. A necessary being is incapable of having a dependence on a being in some other state.

  13. Therefore a necessary being is incapable of change.

  14. Therefore no material object can be a necessary being.

  15. God obviously escapes from this by not being a material being, neither bound to time and space, nor describable by physics.

    It might be objected that my analysis here depends on our current understanding of physics. Could there not be a necessary material being described by some other very different physics? I think this objection fails, for two reasons. Firstly, the assumptions I used are very general, and taken from the possibility of an algebraic representation of the material universe. Secondly, we do live in a universe governed by the physical laws that satisfy the properties I need, so since it is this universe we are trying to explain, it is not unreasonable to assume these axioms to be true.

26. James
Posted at 06:36:48 Friday February 15 2019

@Nigel Cundy

This is off topic, and I believe you have talked about this in the comments of another post before, but do you believe that B time theory poses a threat to a belief in causality and actuality and potentiality?

27. Scott Lynch
Posted at 18:05:02 Friday February 15 2019

Generation vs. Creation

Dr. Cundy,

I notice that in your blogs, you use the word “creation” and “annihilation” regularly. I know that this is the standard particle physics jargon, but in Aristotelian philosophy, this would be generation and corruption. I guess my point is that this could lead to confusion, so I may be a good idea to clarify those terms when you use them in arguments. Just a suggestion.

28. Nigel Cundy
Posted at 00:26:58 Sunday February 17 2019

B time theory

Is B time theory incompatible with causality, actuality and potentiality?

Please note, that I am approaching this question from the perspective of a physicist rather than a philosopher. I am not aware of all the nuances of the philosophical debate, and there might well be something I am missing. Please let me know what it is.

In my view, it depends in part on which version of B time theory you favour. Obviously many proponents of the block universe theory see it as implying that the universe is essentially static, which would deny the process of actualisation and thus efficient causality. I have never really understood this, though (which possibly means that I am missing the point).

First of all, I relate the potentials with the possible eigenstates of the Hamiltonian operator describing whatever it is you are looking at (in other words, the quantum states, or super-positions of quantum states). [More precisely, I view the eigenstates as a partial representation of the physical potentia.] This doesn't depend on one's view of time. So that's OK in either the B theory or the A theory.

Now we come to the process of actualisation. If actualisation is defined as saying something that previously existed potentially now exists actuality, I think the main issue for the B theorist is in the word now. In the B theory, every moment of time can be said to be real. The A theory states that only the present is real. Thus in the B theory there is no objective sense of the word now, and that definition of actualisation makes no sense.

However, if we define actualisation as saying that something which existed potentially at one moment of time gained actual existence at the next moment in time, then I don't see that there is any problem for the B theorist. The B theorist still acknowledges the dimension of time and the possibility of change in time. That things can be at one moment of time in one state, and the next moment of time in another state. This describes precisely the same process as in the A theory.

The only difference that I can see is that the A theorist can say that something exists actually or something exists potentially (since they believe that only things in the now can exist in reality there is no need to qualify the statements). The B theorist has to qualify these statements, saying that something exists actually at a given time or something exists potentially at a given time. In the A theory, we discuss something as being actualised (in the sense that it is happening now). In the B theory, we can say that it was actualised at a particularly time and place, or simultaneously (in some reference frame) with something else (say a particular thought or perception). The B theorist still observes the passing of and changes in time; he just views each moment in time together rather than successively. But he is still aware of the relationships between one moment of time and the next, and the direction of time (from a state of low entropy to the state of high entropy). Obviously, if we have the process of actualisation in B theory, then we can also have efficient causality: we can say that object Y which existed at a given moment of time was generated from object X which existed at the previous moment of time.

Expressed like this, there are no practical consequences for the physicist arising from the two different views of time. Which is why we have two different ways of expressing the theory,

the four dimensional Lagrangian approach and the 3 dimensional Hamiltonian approach. Both are equivalent to each other, and both can be mapped onto each other. Thus if actualisation and causality are useful concepts in the 3-dimensional view (A theory), they have equivalents which are equally useful in the 4-dimensional view (B theory).

I find the two pictures to be complementary. The A theory gives a human-eye view of the universe. The B theory gives a God's-eye view of the universe.

29. James
Posted at 22:46:26 Thursday March 7 2019

@Nigel Cundy

Thank you for the response!

However, I have another question I was wondering if you could answer.

How does QM (and you view of it) play into how we view/understand free will? Does it enforce it, or not allow it, or decrease it, etc?

30. Nigel Cundy
Posted at 18:11:28 Saturday March 16 2019

QM and free will

Dear James,

This is an interesting topic; but I don't know that much about the philosophy of the mind, so I am likely to upset people with my approach.

I start with a couple of definitions:

Physical Indeterminism is an ability to precisely predict the physical states that are the outcomes of an initial state. Given a known initial state A, then it will evolve in time. A system is physically deterministic if we can confidently predict with 100% accuracy that it will be in state B at the end of the time period. (Obviously there are complications due to experimental imprecision; I am assuming the utopian picture of perfect experimental precision. Experimental uncertainty also leads to imprecision in our predictions, but it is from a different source and can thus be distinguished from physical indeterminism. Experimental uncertainty can be reduced by more careful experiments; physical indeterminism implies a hard limit coded into the universe which we can't beat no matter how hard we try). A system is physically indeterministic if the best we can do is predict that the state will either be B, or C, or D, or ... (note that this is not the same as a free-for all where anything can happen, and we can also compute how frequently each outcome would occur should the experiment be repeated). The success of quantum physics strongly implies that the universe is physically indeterministic.

I define free will in a similar way, i.e. given even complete knowledge of a mental state at a given time, it is impossible in principle to predict what the mental state will be at a future time; this means also that it is impossible to predict somebody who has a free will to do.

With these definitions, free will and physical indeterminism are clearly closely linked together.

If we are to assume that changes in mental states are wholly reducible to physical processes, and that physics is fundamentally deterministic, then obviously free will is ruled out. I think that is why many people have chosen to reject the idea of free will. A common approach to defend free will is to challenge the first of these assumptions, either with Cartesian dualism or Aristotelian hylomorphism, or perhaps something else. What I would suggest is that the second of these assumptions is flat out wrong. Of course, one can also challenge the first assumption as well (and I have argued that physical indeterminism implies some agency that influences the universe but is not describable by the methods of physics; one can apply a similar argument for mental indeterminism.

Does quantum physics enforce, not allow, or decrease free will? Given my answer above, I would answer none of these. It could still be that the brain is at a level where the classical approximation is valid; i.e. quantum indeterminacy is simply never observed. However, that is an additional assumption, and I am not sure how well it would stand up in a detailed science of the brain (indeed, given the observation that we do have free will, I think that the brain is very likely to be based on quantum processes). Quantum physics allows free will, even if mental processes are entirely reducible to physical processes. It falls short of enforcing it, since there is still that get-out. But it still makes free will very likely. So the answer is somewhere between allow and enforce, and I would say closer to enforce since I can't imagine that quantum effects wouldn't be important in neuroscience.

31. Anthony
Posted at 23:30:12 Friday April 5 2019

Short question

Dr. Cundy

What is your opinion on the quantum switch and could you maybe explain this phenomenon a bit further? Does a quantum switch realy entail an indefinite causal order? Could you comment the claims of this short paper?

32. Nigel Cundy
Posted at 17:07:00 Saturday April 6 2019

Quantum Switches

Dear Anthony,

Thanks for the question.

I do have thoughts on quantum switches, but they are a bit too long and involved to describe in a blog comment. I'll try to convert my notes into a suitable format and post it as a full blog post sometime soon (with the word soon meaning when I find time to do so).

33. Anthony
Posted at 17:13:36 Saturday April 6 2019

Thank you very much dr. Cundy for taking this issue into consideration! I would realy be interested in the relation between quantum switches (as described in the paper) and aristotelian aetiology.

Post Comment:

Some html formatting is supported,such as <b> ... <b> for bold text , < em>... < /em> for italics, and <blockquote> ... </blockquote> for a quotation
All fields are optional
Comments are generally unmoderated, and only represent the views of the person who posted them.
I reserve the right to delete or edit spam messages, obsene language,or personal attacks.
However, that I do not delete such a message does not mean that I approve of the content.
It just means that I am a lazy little bugger who can't be bothered to police his own blog.
Weblinks are only published with moderator approval
Posts with links are only published with moderator approval (provide an email address to allow automatic approval)

What is 6-9?