A, B, and Aristotelian Theories of Time

[Jac3510]

So the terms "A-Theory" and "B-Theory" (of time) get thrown around often enough that it's probably a good idea to have a formal discussion on the matter. Further, I would contend that an entirely separate theory exists in the Aristotelian/Thomistic view of time that fits neither of the previous two. As I've said elsewhere, I'm concerned that people look at the A/B Theory debate as a strict either/or, in which time is either understood on A-Theory terms or it is understood on B-Theory terms, and that those two theories thereby exhaust the options. A similar mistake, I think, is made in the traditional and never to be resolved Calvinism/Arminianism debate, in which proponents of both see a strict either/or relationship (e.g., you either believe in conditional or unconditional election; there is no third option). It's worth noting that if it turns out that an either/or situation does exist, then proponents have two means by which they can defend their view: 1) prove their own, which necessarily negates the other; or 2) disprove the other, which necessarily proves their own. That's just a standard disjunctive syllogism:

• Either A or B
  ~A
  :. B

As such, people tend to think that just by disproving B (or A) theory, they have thereby sufficiently proven their own view. This would be true if A and B Theory exhaust the options and are contraries. That's why I suggest that there is a third view.

So, first, a quick overview of A and B Theory and the problems I have with each. I'll conclude by briefly explaining the Aristotelian view and why I think it is correct.

A Theory

"A Theory" takes its name from McTaggart's Unreality of Time, a paper published in 1908. He argued that there are two ways to understand temporal relationships: 1) according to tensed properties (e.g, yesterday, right now, tomorrow), or 2) according to positional properties (earlier than, at the same time as, later than). Series of the first kind make up an A-Theory of time. It requires little explaining primarily because it is (arguably) the common sense view. The past no longer exists; the present does; the future does not yet. McTaggart, though, pointed out what I think is a devastating problem, namely, as intuitive as the theory may be, it turns out to be self-contradictory and thus incoherent.

The reason is that no event can be simultaneously past, present, and future.* Yet A-Theory seems to require just this. We naturally think of any given event as having a certain temporal property, and that that property changes with time. Thus, if we were to talk about Dom's first reply to this thread, we would say it has the property of "being a future event," and we would describe that by saying something like, "Dom will reply to this thread sometime in the future." Once he does, however, his reply will no longer have the property of being future; it will change to having the property of "being a past event," which we would describe by saying something like, "Dom replied yesterday."

On closer examination, though, it appears that on A-Theory any event (say, Dom's reply) has all three tensed properties; that is, on this view, all events turn out to be past, present, and future, insofar as all events have all types of properties. That's easy enough to see when we consider the fact that Dom's reply may now be future (so it has the future property), but it also will be present once he replies (and thus will have the present property), and will also be past a few moments later (and thus will have the past property). Of course, the immediate objection is that this is nonsense, because no event has all three properties at the same time. Rather, each event has its tensed property at a moment in time, and that property changes with time. So, again, Dom's reply right now (in the present) is future; Dom's reply will be (in the future) present; Dom's reply will be (again, in the future) past.

But the problem with this is easy to see. A-Theory is supposed to explain what time is (a sequence of temporally ordered tensed relationships). And yet this reply appeals to time. Notice the italicized words above. Each of those events, McTaggart points out, themselves are temporal events and have all three temporal properties. To resolve the problem by again pointing to the tenses of each is just to set up an infinite regress, so nothing has been explained.

Frankly, I don't see how to get around the argument. I regard it as decisive. But there is also the serious problem that an A-Theory poses to theology. Craig has properly and correctly demonstrated that on an A-Theory of time, God is temporal. But if God is temporal, then He changes. And if God changes, then He is mutable. Not only is He mutable, He cannot be said to exist a se, nor is He absolutely sovereign. Moreover, God is in some sense dependent on us (insofar as His temporal relationship with us is dependent on our position in time). But if God is in any way dependent on us, then He is not the First Cause. So not only does an A-Theory experience a significant philosophical challenge, it presents a very significant theological challenge.

B Theory

One can avoid both challenges above by accepting a B-Theory. On this view, time is not a temporally ordered series of tensed relationships, but rather a temporally ordered series of tenseless relationships. Thus, to return to our example, Dom's reply to my thread is later than this post; this post is earlier than his. Note that this is true whenever it is stated. Suppose I say, "My post is earlier than Dom's reply," and I say that yesterday. It is a true statement. Now suppose I say the same sentence now. It is still true. And now suppose I say it a week from now. It is still true. The same cannot be said about the sentence, "Dom will post tomorrow." That will only be true on the day before he posts!

Now, the big point here is that, on this view, change is rather illusory, and time is really a mind-dependent event. The past still exists, as does the present, and the future already does, too. All of our tensed language can really be reduced to tenseless, B-series language. Clearly, then, there is no self-contradiction, because strictly, there is no such thing as temporal properties at all. They're just conventions we use to describe a static existence of tenseless relationships. But the idea that the past still exists, while not philosophically absurd, certainly denies common sense! And the idea that the future already exists also just seems silly to us. And is time really nothing more than a figment of our imagination?

One of the serious problems with B-Theory for me is that it seems to suggest fatalism. Consider the following argument:

• 1. There exists a set of propositions that accurately describes every event that will happen in the future
  2. If there exists such a set of propositions, then whatever will happen in the future is unavoidable
  3. Therefore, what will happen in the future is unavoidable (that is, fatalism is true)

But this seems hard to accept. Yet on B-Theory, it seems the argument is right. A-Theorists can get around it by saying that there is no such set of propositions, since future events have not yet happened, and therefore, it is neither true nor false to say that "Dom will post tomorrow." That will be true or will be false once he posts (or doesn't). Until then, it's simply indeterminate. Yet since everything is already real on B-Theory, it's rather obvious that (as it is already real) then it is unavoidably real! So fatalism and B-Theory seem to go hand in hand. As I reject fatalism, I think B-Theory has to be rejected.

So what alternative do we have?

Aristotelian Theory

The Aristotelian view defines time as the just the numbering of events before and after a change. So this thread now exists, but Dom has not yet posted. When he posts, though, it will have changed, and that change marks the passage of time. Of course, there will be time that passes between the creation of this thread and Dom's posting in it, but that is only because lots of other changes will happen between now and then. For instance, Dom will first have to read the thread, so it will change from having not been read by him to having been read by him. Likewise, the earth will have experienced a change in location, and millions of other changes will have happened. All that is the movement of time. Dom's post is understood as after some changes and before others. That's all time is.

What really sets this view apart from the other view is that it doesn't see time as an independently existing reality into which events "fit." See, on A-Theory, time exists in and of itself as an objective part of the universe. Changes happen in time. On A-Theory, it's perfectly coherent to imagine a world in which all change ceased for a period of time and then later resumed. This is because the temporal properties really exist as constituents of a thing. Dom's post (right now) really is future. That is today, the event called "Dom's first reply to this thread" has the property of "being future." That property, though, refers not just to its relationship with THIS post (e.g., "being future" just means "stands in a 'later than' relation to this post"), for if it did, we would just have a B-Theory. No, on A-Theory, the temporal properties refer to a part of reality we call "time."

It's the reality of the temporal properties as real constituents that gives rise to the self-contradiction the theory faces. Aristotelianism, though, doesn't face that problem because it denies said reality. Change is real (unlike on B-Theory, in which it is entirely illusory), but the "before" and "after" relationship is assigned by the rational mind. In this way, Aristotelianism falls between A and B Theory.

I think this is rather evidently true. To me, it is completely incoherent to talk about the passage of time without change. On the other hand, it's obvious to me that where there is change, there is time.

This view does not face the philosophical challenge of incoherency that A-Theory does. Nor does it face the theological challenge, since time is not an objective part of the universe, there is nothing for God to "relate to." Therefore, no change is posited in God, allowing Him to maintain His immutability, etc. (that is to say, God is not temporal, contra Craig). Moreover, it does not face the problem of fatalism that B-Theory does, since future events do not yet exist, and so propositions regarding them are indeterminate.

Now, much, much, much more could be said on this subject, and there are many other objections we could raise about both A and B Theory. But this is long enough as it is, and I'd like to give whoever a chance to respond before going any further.

---------------------------
* Note the word "simultaneously" here is not intended in a temporal sense, but rather in a logical sense. That's why I used this word rather than the phrase "at the same time."

[narnia4]

Niiiice. Just a few general, mostly non-philosophical comments. Hopefully someone adds some thoughts to this. At this point I'm pretty open minded as far as what theory is correct, although intuitively B-theory doesn't sit with me that well.

I looked at a survey of philosophers a while back and A and B theory were neck in neck (and from a couple other I've read it does seem like philosophers are fairly evenly divided on it). But if I remember correctly only about 15% picked each position (it was a pretty big survey). I'm not sure if everybody else just didn't have an opinion, didn't know, didn't care, or if a few even held to an Aristotelian theory of time. My guess would be that most just don't have expertise in the area, but it was interesting.

If you go from popular culture, most people would go with A-theory. In tv series like Star Trek or Stargate, if there's some sort of "time stopper", people stop moving and everything freezes.

[Danieltwotwenty]

I remember an episode of Red Dwarf where they came to a planet where time ran backwards, it was very humerus especially when one had to go to the bathroom.


Dan

[Reactionary]

Would it be possible for someone to argue that, under the B theory of time, the Universe is actually eternal and no First Cause was needed?

[narnia4]

Would it be possible for someone to argue that, under the B theory of time, the Universe is actually eternal and no First Cause was needed?

If you're talking about the Aristotelian point of view (you weren't I know, but bear with me), there are four types of causes- Material, formal, efficient, and final. If you talk about the efficient cause (actualization), it doesn't rely on time because on the A-T view the cause is simultaneous with the effect.

Now with B-theory, I don't know much as much about it but the Kalam Cosmological Argument, for example, does seem to depend on A-theory being true. But I don't know if you could get away from there being a "First Cause" in the logical, sequential sense, if you can then it seems to me that you could throw doubt on the entire principle of causality.

[Jac3510]

Would it be possible for someone to argue that, under the B theory of time, the Universe is actually eternal and no First Cause was needed?

If you're talking about the Aristotelian point of view (you weren't I know, but bear with me), there are four types of causes- Material, formal, efficient, and final. If you talk about the efficient cause (actualization), it doesn't rely on time because on the A-T view the cause is simultaneous with the effect.

Now with B-theory, I don't know much as much about it but the Kalam Cosmological Argument, for example, does seem to depend on A-theory being true. But I don't know if you could get away from there being a "First Cause" in the logical, sequential sense, if you can then it seems to me that you could throw doubt on the entire principle of causality.

I agree with what Narnia has said here, but I just wanted to add a couple of things.

The Kalam argument as rendered does seem to rely on either an A-theory or an A/T theory, since it is predicated on the reality of a temporal series of changes. With that said, I think a B-Theorist could still salvage a version of the Kalam. The First Cause would just be a First Cause in the Aristotelian sense. The way you would do it would be something like this:

Let T⁰ = "now"
Let T^-1 = "one moment ago"
Let T¹ = "one moment from now"
Let X^o@T⁰ cause X¹@T¹, etc.

While there would be no way on a B-Theory to posit an absolute temporal beginning on the basis of the absurdity of X⁰@T⁰ having been caused by X^-1@T^-1 which was caused by X^-2@T^-2 ad infinitum; you could still argue that such a causal chain cannot be infinite since X remains in existence at every temporal point (T^-1, T⁰, T¹, etc.), and therefore X⁰'s causing X¹@T¹ means that if X⁰ does not exist, then neither does X¹. But that means that an finite number of causal events must be presently existing. But if an infinite number of simultaneously existing causal events are in existence, then there is no First Cause actually "powering" the causal chain to "begin" with (where "begin" here is not a temporal beginning, but rather a logical beginning).

By way of analogy, imagine a chain link hanging in front of you. What is holding it up? The link it is attached to just above it. And what is holding that up? Yet another link. Now, in each case, the link is being held up by the link above it; but that explanation is insufficient, for the entire chain link must be being held up by something concrete (pun intended). The chain couldn't go upwards forever! Or, we could say that the chain could extend upwards forever, but we would still require some other source holding up the chain as a whole (for instance, perhaps there is a magnetic field holding it in place; but that just raises the question, what is holding in place that field, etc.). Now, notice the present tense of the verbs in these sentences.

The important thing on a B-Theory Kalam would be the fact that time would require an Aristotelian "First Cause." It could be called a Kalam only because it is dealing with the reality of time, in which each moment in time causes the next, and there could not be an infinite regress of temporal events. But that impossibility would not be based on an accidentally ordered series (as is currently done with A-Theory), but via an essentially ordered series (as is done in Thomistic arguments).

So even here--as Narnia noted--the universe requires a First Cause, at least in a logical sense. It just can't be demonstrated that this cause must be first in a temporal sense. Then again, though, as I've argued before, I don't even think you can prove the universe requires a temporal first cause assuming A-Theory (without resorting to some argument similar to the one I just made above).

Against all this, if you assume a B-Theory and defend the second premise of the Kalam on purely scientific grounds and forget arguments about the impossibility of an infinite regress, then the Kalam would still work, since if there is an absolute beginning (as per science), you would still need to have an efficient cause of that "first" moment. So I don't really think that the Kalam is affected one way or another by this whole debate. I do, though, think that the argument I've laid out and that Narnia alluded to is much stronger and ought to be used more than Kalam is.