How Time Really Passes

We all experience the passage of time. This is not an illusion. We agree on the serial delivery of moments, and it is very hard to dislocate us from this agreement. However, the core of this experience can be located as a feature of our physical theories. These theories do not replace our notion of passage. Rather, physics provides a reliable account of the regularities underlying our common experience of time passing. It does this by providing us with a concrete relationship between any given now, together with its future and past.

1. We Experience Time Passing

We can apprehend the passage of time through the succession of three mental states, in this order:

  1. the expectation of an event;
  2. the experience of the event; and
  3. the memory of the event.

We prepare for lunch, we eat lunch, and we remember how tasty it was afterward. Moreover, we all agree that this is the way it works — the order doesn’t change.

This is an account of passage that John Norton has proposed. As John suggests, this agreement shows all the signs of not being an illusion. There’s no known procedure that eradicates the experience of passage. There’s no obvious underlying mechanism that reduces passage to an artifact of our perception. The best explanation of the reliable succession of these three mental states is that they reflect an objective regularity in the world. Passage is real. It produces the serial delivery of moments to our minds. We become aware of it through the succession of these three identifiable mental states.

2. Physical Theories Describe Passage

In a physical theory, passage appears as a relationship between matter at any given moment (the ‘now’) and its future and past developments. This relationship is determined by Cauchy evolution. The regularities appearing in Cauchy evolution correspond precisely to the regularities we experience with passage. This gives us good reason to suspect they reflect the same objective feature of reality.

We have said that the core feature of passage is the serial delivery of moments: a future event becomes present, and then recedes into the past. Another way to describe this situation is to say that, once a now has been specified, we are guaranteed a definite future and a definite past. These categories never get mixed up: the past cannot arrive after now; the future cannot arrive before now. There is some real regularity that guarantees future, present, and past will be delivered to us in the right order.

The regularity underlying our experience of passage reflects the same reality as the regularity underlying the way physics describes change. In physical theories, this change is called evolution. Here is how the regularity appears.

By specifying a matter configuration ‘now’ (indicated as a gray circle), we determine a local past and a local future.

We can input a configuration of matter representing ‘now’ into a physical theory. These are called (local) initial conditions. Once now has been specified, the theory determines (in the most common physical situations1.) a definite future and past. The equations of motion of our theory determine the moments in the past from which our matter evolved, as well as the moments in the future to which it will evolve soon. This separation into past, present, and future states of matter matches the key feature of passage: the serial delivery of moments is fixed once we specify our ‘now.’

For example, I might input my ‘now’ to be 10am this morning. My matter configuration at 10am is determined (according to some hypothetical theory) to have evolved out of my bed at around 7am, and is determined to evolve to eat lunch at noon. When I set my ‘now’ to be noon, then everything changes in a regular way. I am determined to have evolved from my 10am configuration, and determined to wash my dishes soon. And so on.

This is a striking match of our experience of the serial delivery of moments. At 10am, I am in a state of remembering my bed, and a state of expecting to eat lunch. This is followed by my noon state of remembering my 10am state, enjoying my lunch, and expecting to do the dishes.

Craig Callender (2000) argues that Cauchy evolution is what makes time informative. Here, we have reached a simpler conclusion: Cauchy evolution is how physics reflects time’s passage.

3. Experience and Theory Reflect the Same Thing

The human mind apprehends the passage of time through a change of state: we expect, we experience, and then we remember. This is a fact about our human experience. The steadfastness of this experience suggests it reflects an objective feature of the world.

Physical theory describes passage by describing a state’s evolution: given an initial state, we determine the past states as well as the future ones. This is a fact about our physical theories. The empirical confirmation of these theories suggests that they too reflect something objective about the world.

This does not mean that our apprehension of passage can be reduced to the physical description of passage, or vice versa. Rather, our conclusion should be that both the experience and the physics are very likely reflective of the same phenomenon. Physics is not devoid of passage. It is even more evidence for its existence.

1. This is not to ignore the difficult problem of specifying, and justifying, the conditions under which the Cauchy problem is well-posed. There are many well-known cases in which initial conditions fail to settle future/past evolution. However, it is sufficient for us to note that in a great many cases, and in almost all common applications, initial conditions can be chosen so as to guarantee a well-posed Cauchy problem.

Soul Physics is authored by Bryan W. Roberts. Thanks for subscribing.
Want more Soul Physics? Try the Soul Physics Tweet.

12 thoughts on “How Time Really Passes

  1. Baptiste

    Hi there.
    It seems to me that the problem consist in the specification of initials conditions. Ok, there is genuine passage in our experience of time. Ok there is passage in physics described by Cauchy evolution. But the passage in physics is derived of a first step : the fixation of “now”. If you have to localise the “now” before you can speak of a passage of time, can you really speak of a passage as an objective feature of the world ?
    You can specifies a lot of initials conditions and so a lot of different passages. If there is so different passages can you say it’s objective ?
    If i’m preparing for lunch, this lunch is in my future. But I can find initials conditions for it’s in the past of the system.

    My idea is more something like that : there is only genuine passage in experience of time. Objectively what there is, are ordered facts, successive facts (what are called B-relations in metaphysics). Subjective passage is constructed by the perspective of on the eternity, the bloc universe described by relativity.

  2. Bryan

    Thanks for the note Baptiste,

    I agree that physical theory apparently does not provide an objective account of ‘now.’ However, in many cases it does give us the universally quantified sentence, ‘for any now, there exists a definite past and future.’ So, if we can think of that proposition as capturing the essential feature of ‘passage,’ and if we take the theory as a whole to be describing reality, then we have an objective account of passage.

    I’m sympathetic to what you’re saying, that genuine passage is an artifact of experience. But as John argues (I think rightly), it’s very hard to call our experience that ‘time really passes’ an illusion, or even a feature of the way we perceive.

    Kant declared that Euclidean geometry is an artifact of the way we perceive (in the jargon, it is the transcendentally ideal a priori form of sensory intuition). This turned out to be wrong. If we want to avoid making the same mistake twice, we should perhaps take suggestions about what’s ‘built into our perception’ with a grain of salt.

  3. Bryan

    Noah: I just mean the way the passage of time is made available to our senses.

    We seem to apprehend time in a ‘serial’ way — that is, we go through a particular ordered succession of mental states. When this happens, we are picking up on a particular feature of time, which is made available to us (I used the metaphor ‘delivered’) by the real world.

  4. Baptiste

    Hello again. Please forgive my bad English language but I would like to say one or two things.

    In metaphysics there are two theories about passage. According to theory A there is a genuine passage of things in time. Things are future then they are present and finally they are past. On the contrary theory B says that there is no passage of things in time. Things are what they are for eternity (that’s why theory B go with a perdurantism account of identify throught time).

    The point is that you have to choose between A and B theories. There is no middle hybrid solution. Physics seems to give us a B-theory of time because as you say ‘for any now, there exists a definite past and future.’ The expression ‘for any now’ shows that past and future are in physics a question of perspective.
    I do not understand why you are saying “it’s very hard to call our experience that ‘time really passes’ an illusion, or even a feature of the way we perceive.”
    I agree that passage of time is not an illusion. But it doesn’t imply that passage of time is not a feature of the way we perceive : it’s the only thing it can be in a B theory.
    Compare that with experience of space. In B theory “now” is just like “here”. You know that you are here just watching at your desk for example. You could be in another place right ? Just like you could be in another time if you accept a classic B theory of time account. Then you move, you go in other place. Surely you can say that space has passed but intuitively we prefer to say that is it us who have changed of location. It’s the same in time : we change of location (but independently of our will in the case of time).
    We pass in the time (or time pass, never mind) between two instants in the sense that our temporal localisation change between that two instants. Passage is profoundly linked to our perspective and in that sense is mind-dependant (always if you accept a B theory).
    So I do not understand in a B theory background what could be a genuine physical passage of time, except a change of location from a space-time slice to the successive hypersurface.

    Maybe I’m wrong and you will just answer me saying that philosophy of physics do not defend a B theory of time. But if B theory is wrong A theory A is right. In that case physics lacks the specificity of “now”, the fact that present is not a mere perspective on eternity but an ontological feature of the universe. Then you can not say that physics reflects time’s passage : it does not because nothing in physics show the existence of a privilege time, our time, the present as required by A theory.

    I know it’s a blog of philosophy of physics so don’t hesitate to tell me if I’m boring you with metaphysical considerations.

  5. Bryan

    >> The point is that you have to choose between A and B theories. There is no middle hybrid solution.

    Come now! Can’t we compromise? ;-)

    Actually, I really am arguing for A-series passage here, not B-series. But more importantly, I’m saying that many have been too quick to assume that physics provides us with nothing more than a temporal ordering (which is what B-theorists settle for). I think physics provides us with an A-series as well.

    Physics does indeed lack a ‘now’ — I agree that’s provided by one’s personal psychology. But what physics does have is the capacity to take any given ‘now’ as input, and provide a future and a past as output. I argue that this is enough for us to say that physics provides an A-series ordering.

  6. Tim Nash

    The problem with basing an A-series ordering on given NOWs is that these NOWs are localized just like Time is.

    In the famous twin paradox, the faster moving twin (as well as his spaceship) is experiencing a ‘fatter’ NOW. His year is his twins decade, his second is equal to his twin’s entire day, etc. This means that NOW does not extend into infinity, it is localized. It also means that there can be no Universal time line because the localized time Arrows do not necessarily line up.

    One more problem is that if you line up a ‘fat’ NOW with a thin one, you have a time arrow being overlaid upon a moment (point vs line). So for any given ‘moment’ for which you want to calculate output there already exists a faster moving particle who has already put such a moment in his rear view mirror.

    Julian Barbour has really got me wondering. I’d like to see more understanding of what NOW is.

  7. Bryan

    Actually, I like the idea that “NOW” should be localized to finite regions of spacetime. After all, our experience of the present is purely local. Shouldn’t our description temporal passage be local as well?

  8. Tim

    The perception of infinite “NOW”, the collapse of the “wave function of the universe” is more pervasive than any other time experience. And a quantum computing model of the universe also seems to argue for a universe with only one wave function. So even if NOW seems localized, it really cannot be. Where does this NOW end and a different one begin? It is a shifting frame of reference as you travel from point to point. Each point sharing the same instance, the same collapse of the wave function.
    All these calculations remind me of the effort that went into adjusting planetary motion to place earth at the center. A lot of effort to preserve our notions of time. If Sean Carroll’s new book represents the current view I think it is too conservative. Here is my favorite line so far. It seems to argue against the main thrust of his book: “the final theory of everything is likely to exhibit non-locality in some very dramatic ways” page 249
    This may be a nod to B-series thinking.

  9. Bryan

    Tim: You’re adopting a very natural assumption about “now”: that if (1) a theory tells us that two events happen at the same time; and (2) one of the events has the property of being “now”; then both events have the property “now”.

    However, since I think the property “now” can only be defined in a local region of finite volume, I have to deny this assumption. I think that an event A can count as “now” and not an event B, even though A and B happen at the same time. Yep. It’s unusual. But we local-now types have to bite the bullet here.

  10. Tim

    I agree that that is a biting of the bullet and a rejection of an important “objective feature of the world”.

    I am curious, do you think any of these are equivalent to the other?:
    1) simultaneous events A and B
    2) “NOW”
    3) collapse of wavefuntion of the universe.

    On a side note, I hope B- series ideas are a funded area of research.
    A- series types won’t care much, but some Series B models bring up the question of why we all share a NOW of the same duration.

  11. Bryan

    If “simultaneous” means “lying on the same (infinite) Cauchy surface,” then I’m saying: 1 and 2 are not equivalent. I also think 2 and 3 are not equivalent. The reason is, a simultaneity surface and a wavefunction extend all the way to infinity. But there is simply no evidence that “now” extends all the way to infinity, this being quite clearly outside the realm of anyone’s perception.

    To put it another way: “now” is a concept that appears in theories of perception. Nothing about these theories suggests that “now” can or should be understood non-locally. Contrast this with “simultaneous,” which is a concept that appears in physics. Spacetime physics suggests the concept “simultaneous” can and should be understood non-locally. Conclusion: “simultaneous” is a non-local concept; “now” is not.