Usually, we content ourselves in saying that the past “did exist,” that something “was” in the past, and the future “will exist.” But that’s actually tricky—what do “did”/”was” and “will” actually mean? Are they special classes of existence? Are they like somewhere in between existence and non-existence? We think of them is if their meanings are clear, but if you really think about it, “did exist” and “will exist” boil down to nothing more than the grammatical conventions we use exclusively in reference to the past and future respectively, so they really tell us nothing, and it’s not clear what other kinds of “exists” there could be than the normal one or if such a consideration is even logical.
So, it would seem more sensical to assume that the past and future either don’t exist in any sense, and effectively never “did” or “will,” or that they still and already do exist “somewhere.” There’s no way even in principle to measure or observe the past or future, because any measurement must be done in the present, hence whatever it measures can only be a part of the present. Everything we know about the past is mere inference from what we know about the present. For example, we know about dinosaurs because their bones exist in the present. We know about WWII because of the various documents, war planes, stories passed down to the present generation, etc. that exist in the present. And all our memories of the past, including what happened 5 seconds ago, apparently exist as current neurological patterns in our brains.
Similarly, everything we know about the future is obviously merely an expectation or anticipation existing in our imagination, deduced from what we know of the past and present (where “the past” is, again, nothing more than an aspect of the present).
You could even argue that assuming a past or future therefore violates Occam’s razor, which is the principle that the theory that explains the phenomenon in question while introducing the fewest necessary (or unnecessary) external elements is the most likely correct one. (Occam’s razor makes sense if you consider that each element has a certain probability of existing, and the probability of all of them existing, which would be necessary for the theory to be correct, is the product of all the individual probabilities, which diminishes with the number of probabilities being multiplied.) But it’s not really clear whether the assumption of a past violates Occam’s razor, because what would it mean to explain the present if you can’t appeal to past states as a cause? The assumption of a future, though, may more clearly violate Occam’s razor.
So, there is somewhat more reason to think that the past exists than that the future does: if the past “didn’t” (i.e., doesn’t, somewhere/somehow) exist, then why does the present exist as it does, with all of the indications of a causal history—for example, our bodies, all other plants’ and animals’ bodies, their fossils, etc. that seem to indicate a long-running process of biological evolution?
Though there is some reason to assume that the future “will exist” (i.e., does somewhere/somehow exist): we experience the “now” as ever moving from past times to more future times (or at least our memory, which exists in the present, tells us that), or in other words, we experience the past as ever-unfolding into the future (at least all the way back as far as we can remember). Therefore, it makes sense to assume that the “now” marker “will” (whatever that means) continually “move” from the past to the present to the future, or that the now “will” unfold into the future, which “will” then be the current state of existence.
As another answerer went into, the relativity of “now” in the general theory of relativity (which has been verified in every observation we’ve ever made to test it, and is currently used in practical technologies such as GPS) seems to indicate that a “block universe” in which past, present and future always-already exist is the only logical possibility. “Now” would therefore be an illusion, perhaps one that’s useful for the purposes of survival and such as driven by the forces of biological evolution.
Though it raises the question, if all of time already exists, why is there any coherence/order between past and future in accordance with the principles of causality (which is what makes biological evolution even a thing) at all?
But I guess you could make the same argument for a universe in which the past really does continually unfold into the future, as in, why doesn’t it do so completely chaotically? David Hume showed that there’s no logical necessity for the principles of causality—i.e., why one event should follow another. In other words, we can’t prove, ultimately, why it does.
But it’s worth considering that if there were no causal coherence between the past and the future, there’d be no basis for the perception of time to begin with (nor even any organisms that could perceive it), and therefore it’s arguable whether time would actually even exist in that case. Time seems to be an emergent property of the laws of physics. For example, some scientists believe that there was no initial moment of the universe, even though the universe didn’t extend infinitely into the past (as according to the big bang theory), because at the beginning of the big bang, causality/fundamental mechanics worked such that time “curved around,” in a way analogous to how the curvature of the earth means that you can walk forever straight across its finite surface without ever actually escaping it or going beyond Earth’s diameter.
As another example, there’s a theory that once the heat death of the universe sets in, approximately a googol years from now, and the universe is completely homogenous, it’ll be functionally the same as the singularity at the beginning of the big bang, and hence the universe will start over again, at least from a certain perspective (and really, nothing ever has any meaning independently of any perspective; there is no such thing as a “view from nowhere”). This is supposedly fully consistent with known physical theory, and it’s something I’d personally strongly suspected even before I’d heard of the theory.
In his books, Julian Barbour gives an interesting explanation of how the illusion of now and of the progression of time could emerge from a universe in which the past, present and future all always-already exist.
I personally think that the randomness of quantum mechanics might be what allows for an actual gradual progression of time, because if reality were deterministic, then future states would amount to nothing more than a mathematical transformation of past states (and probably vice versa), and therefore there would be nothing to stop time from progressing from the initial moment of the universe to the last moment all in an instant, thereby making any experience of the gradual flow of time impossible. I envision quantum mechanics to entail a kind of ongoing “dissonance” between the past and future. This would seem to imply that there’s a lot more to the “random events” of quantum mechanics than there appears to be. I wrote a little bit more about the significance of random events under the third bullet point of my essay https://philosophy.inhahe.com/2016/12/13/notes-on-free-will/.
