Once Present, Now Past – David Sanson (Illinois State)
Those of us who think that some of the fundamental facts that constitute reality are temporary, and so subject to change, must account for a corresponding contrast between past and present. That contrast cannot be captured by pointing to the facts that constitute reality; it requires us to point to the facts that once constituted reality too. So, if the fundamental facts are temporary, then the past is explanatorily ineliminable, but it is not real. This gives us a reason to be Presentists rather than Moving Spotlight Theorists or Growing Block Theorists, and it gives us a reason, as Presentists, to reject the demand that truth be grounded in being.
Full paper available here.
Because the papers are works in progress, they have been password protected. To register for the conference, and get the password, please contact Graeme A Forbes at G.A.Forbes@kent.ac.uk
26 comments
Skip to comment form
What do you make of my claim that to describe a scenario in which the once present and now past diverge requires resort to ‘hypertime’, which then turns out to be time itself, undermining the scenario? To be sure we can then construct a scenario in which the hyper-once hyper-present diverges from the hyper-now hyper-past. But that requires hyper-hyper-time.
Author
So, for the sake of argument, let’s accept that the Growing-Block is committed to havingt reality point beyond itself. (When I say the sun will rise tomorrow, I’m saying something even though there exists no day of which I’m offering a discription.
Might there not still be merit in defending the Growing-Block view, because it gives us an ontological asymmetry in the future. Sure we have to answer questions about the necessity with which reality consistently points beyond itself such that times aren’t skipped, and they all once were present, and so on (See my Topoi piece, where I press these very points). But we might think that the ability to refer de re to past things but not purely future things captures a nice asymmetry about the extent to which reality can point beyond itself.
Thanks for the interesting article! I have probably misunderstood something important here, but I was wondering: why does thinking about the past as what was “once present” amount to a contrast between what constitutes REALITY and what once did? Doesn’t this beg the question against the growing block theorist, who says that something once was at the leading edge of the growing block and now isn’t, but it is still part of reality? Isn’t the above argument assuming presentism, and restricting reality TO what the growing block theorist would see as merely the “leading edge” of the block? And, if the answer is that the growing block just adds something which is unnecessary in explaining how we think about the past, there is Graeme’s thought above that it’s not unnecessary because it gives us an ontological asymmetry between past and future.
For example, if we say that things are “once present,” but are no longer part of reality, wouldn’t we have to say something similar about specific future events that are also not a part of reality -something like that they are “to be present?” We would be as committed to their coming into existence as we would be to the fact that specific past events were once present. The growing block theorist seems to have an explanation for the ontological asymmetry, which is that the past is part of reality. Thanks again!
Thanks for the interesting paper! Naive question, but why think that saying ‘it is built into the past-oriented property, having been President, that nothing can have it unless it once had the property, being President’ means that the Presentist account of the past is ‘expanatorily parasitic upon an account in terms of the once-present’ (p. 10)? All we’re required to postulate is a necessary connection between the properties, the order of explanation is still up for grabs.
After all, one plausible view – priority presentism – flips the direction of explanation and holds that the two properties are connected because e.g. Bush’s presently possessing the past-oriented property grounds/metaphysically explains his past possession of the present-oriented property. According to this view, our account of the once-present is ‘explanatorily parasitic’ on our account of the past (i.e., in terms of the properties things have now).
Additionally, do you think this view escapes the ‘pointing beyond’ charge, since all it ever needs to postulate are presently existing things?
Sorry for the delayed replies to everyone. Monday was a long teaching day for me. Today I can focus my energy on this conference. So before I tackle my assigned paper, and then the rest of the papers, let me briefly reply to your thoughtful questions.
First, in reply to Peter:
I need to add a discussion of hypertime to the paper. I am not a fan of hypertime, and I don’t think the examples depend on it. Or, if they do, I think the problem is caused by the theories I am criticizing, not a problem with the critique. But I need to say more to explain why I think this.
As I am understanding the situation, the Growing Block Theorist posits an existing ordering of events, and claims that the existing ordering accurately reflects how the block developed. I want to say that time, properly understood, is about how the block developed, not about *any* existing ordering of events. So that’s the wedge I’m trying to drive in between two things that might count as candidates for being “time”.
You might think that such a wedge depends on a distinction between two existing orderings of things that coexist: one, the ordering of events in the block; the other, the ordering of successive blocks of different sizes. Since I don’t think that time ever involves an existing ordering of things that coexist, I don’t think this. The examples are meant to motivate the claim that the real temporal action, in the Growing Block Theory, is the growth of the block, not the existing ordering of events that it leaves behind.
Still, I do need the distinction between the existing ordering of events (that I don’t think is time) and the successive development of the block. And you might see in this a distinction between two “times”, one “ordinary time” and the other “hypertime”. And so then my claim is that the “hypertime” is the real time.
But, first, I doesn’t seem right to me to say that the existing ordering of events is “ordinary time”. It is not where the real temporal action occurs in the theory. It is just a trace that the real temporal action is supposed to leave behind. And, second, if one wishes to insist that the existing ordering of events is a time, and is also committed to the growth of the block, then I think you have to explain how the two are connected (which is what people do when they describe the view: they describe the growth, and the describe what it leaves behind). The examples trade on that distinction. If that distinction already involves hypertime in some objectionable sense, then the Growing Block Theory already involves hypertime in an objectionable sense.
But maybe this is begging all sorts of questions! I’m not sure. Note that the view that I prefer embraces the kind of time that looks to you like a “hypertime”, and says that it is the real time, but it also jettisons the kind of time that looks to you like ordinary time. So the view I prefer is not committed to two layers or levels of time.
Graeme:
Thanks for the comments. I need to read your Topoi piece. Like I say in footnote, this is an old paper, and needs to updated to reflect more recent developments.
The short answer is that I don’t think the relevant asymmetries between past and future should be understood as ontological asymmetries. I don’t see any reason why we cannot have de re reference to future things as well as past things. And I don’t think the modal asymmetry between past and future is properly captured by the claim that the past exists but the future doesn’t. Instead, I’d be inclined to tie it to an asymmetry in the sorts of changes that we posit when we posit temporary facts.
But this relates to a whole bunch of other things. I prefer an ontology of persisting individuals over an ontology of events or perduring individuals. And I want to say that those individuals come into and go out of existence. So for me, another important reason to say that the past does not exist is to avoid contradiction without relativizing properties or states of affairs (I follow Hinchliff’s 1996 “Puzzle of Change” here). And that reason applies to both past and future equally.
Here is one way to think about it. If your reason for saying that past things exist is that you think we can refer to them de re, and you are willing to allow that new things will come into existence, then you are committed to saying that, if (perhaps per impossible), a past thing was annihilated, we would no longer be able to refer to it de re. And don’t see any reason to be confident that such annihilation never happens. But it seems to me that our ability to refer to past things de re should not depend in this way on their continued existence. So I don’t the kinds of causal historical relations that we stand in to past objects depend on their existence.
Author
Thanks for the reply. I think I am committed to saying that, if per impossible, a past thing was annihilated, we would no longer be able to refer to it de re. There would be no such thing.
The motivation for this is drawn out nicely in George Orwell’s Nineteen-Eighty Four. The attempt to change the past by removing the traces of it in the present is futile, because the thing still would have happened.. Having an ontological commitment to the past seems to be the obvious way to capture the fact that regardless of how the present is, the past is real (in the sense of mind-independent), and all the people that did things in history, who are now dead, are also real—there are mind-independent facts about them, which they themselves ground. I think my Granddad has a reality that my children don’t. I have no children, but my Grandad, someone whom I can remember (and memory is factive) is a part of the Universe whether he is preserved in the causal traces, or memories or not. My memories, if they are to be memories, and not confabulations, depend on how things were with him, rather than than how things are to me. How can some facts depend on how things were with a person if no such person exists? If my granddad is to ground facts about himself, it sounds a lot like I’m ontologically committed to him. (Or else I’m not sure what it takes to be ontologically committed to something.)
Right!
If you were convinced by my arguments about how we need to understand the past, then you would be convinced that we need to say both that your Grandad is explanatorily ineliminable AND that he doesn’t exist. I agree that this causes trouble for ordinary understandings of ontological commitment.
So I think I am committed to saying that we can de re refer to nonexistent things. That either forces me to become some sort of Noneist (“∃x(x doesn’t exist)”) or it forces me to make complicated use of tense operators (“PAST∃xNOW(‘Socrates’ refers to x and ~∃y(x=y)”) or it forces me to do something else.
These seem like bad results! So I am keen to find partners in crime. I think anyone who admits that they need to “point beyond reality” is a partner in crime 🙂
Author
Now we’re getting somewhere!
So I am committed to a pointing beyond reality, but think that I’m not a partner in crime. I think that fundamental modality points beyond reality, bringing events into existence, but doesn’t allow/require referring de re to anything I’m not ontologically committed to. It is that latter thing that is the crime, not the dynamism, which is grounded in whatever it is that makes the laws of nature constrain the future.
Good. I need to read your Topoi paper, obviously. So I’ll do that and then respond in more detail. The easiest way to make you a partner in crime would be to convince you that (a) facts exist and (b) when you point beyond reality, in the way needed to capture the sense in which reality is dynamic, you need to point to facts that once existed but don’t exist, or will exist but don’t exist. (Does the same thing apply in the case of modality? I’m not sure: I am inclined to think that *change* makes this all more pressing in the temporal case, but maybe that is wrong.)
I don’t take this to be an especially powerful argument, since both (a) and (b) can be denied. But I think that even if (a) is denied, something like (b) remains true, and I think that even someone who makes a sharp distinction between ontology (what there is) and facts (the ways things are) needs to appeal to the ways things were, and needs to understand that appeal in a way that makes it close kin to, for example, de re reference to nonexistent objects.
Author
My Topoi piece doesn’t go very far by way of defence of my response to the worry. My papers with Rachael deal with it slightly more, but it’s something I’m going to want to say things about in future work.
I’m going to be committed to facts, I’ll grant you. But I think they all need to be made true by things we can quantify over.
The point you’re pressing is a good one, and I feel the force of the worry. The idea is that I’m committed to facts about how the Universe used to be, and how it will be. Even if I try and make them true with stuff that I can quantify over, I need to be able to guarantee that the stuff in my ontology does relate to how things were and how things will be in the right ways, such that it can make them true. I still need the current stage of the Growing-Block to point beoynd itself, such that it is a block that has grown, and a block that will grow.
The solution I hit upon is to use the ‘nomological package’ as Rachael and I call it. Whatever it is that grounds the laws of nature has to be the sort of thing that can point beyond itself anyway. Saying a glass is fragile points beyond itself in that it tells us about what the glass would do counterfactually. If the laws of nature can ground counterfactuals, they can ground past claims and future claims, and, the very growth of the block itself. So, yes, I’m committed to reality pointing beyond itself but counterfactuals, laws, dispositions and the like seem to involve reality pointing beyond itself in any case.
Emily:
Thanks your thoughtful comments!
To your second comment first (and see my response to Graeme above):
I do want to say the same things about the future! I don’t think there is an ontological asymmetry between past and future. I would instead try to locate relevant asymmetries in the asymmetries found in the underlying changes. For example, I am open to the possibility that there is an asymmetry in determinate identity, grounded in the nature of generation, of the sort Geach defends in chapter 3 of Providence and Evil. But I don’t think this sort of asymmetry gives us a reason to think that past objects exist!
As to your first worry that I am begging the question. I worry about that too. Worst case scenario, this paper explains why, from the perspective of a Presentist, the Growing Block Theory, Moving Spotlight Theory, and many forms of Presentism, look weird. That might be connected to the origin of the paper: it began as an argument against other Presentists; I brought in the other views because it seemed easier to draw the distinctions I wanted to draw when considering views that offered up a lot of ontological and ideological structure.
But let me try to defend myself! I take it that the Growing Block Theorist thinks that things that were once at the top of the block—say, the death of Elvis—continue to exist, intrinsically just as they were, and undergo the extrinsic change of becoming earlier than other things. The argument does not depend on denying this. Instead, the argument supposes that this is something that needs to be said. Simply saying that the death of Elvis is intrinsically a certain way, and that it is earlier than some other things, does not yet establish that it is intrinsically just as it was, when it was at the top of the block. The thought experiments are supposed to show that, unless more is said, the the two could come apart.
If you grant that distinction, and grant that this needs to be said, then the intuition I would invite you to have is that, when we ask about the relationship between what the death of Elvis is like now, and what the death of Elvis was like when it was at the top of the block, the answer to the first question should be controlled by the answer to the second, because the answer to the second gets at what we really care about, when we care about the past.
Does that still seem like it begs the question?
Thanks very much for this, David. Your distinction helps, and I can see how it links into the “once present” fact being the more important and necessary to explain that things in the block are intrinsically just as they were when at the top of the block. Perhaps it might be helpful to emphasise that your argument doesn’t depend on you assuming that things in the past aren’t a part of REALITY…which was what got me worried. Thanks again!
Nathan:
Thanks for the question! I think the paper lost something important in revision that needs to get put back in.
If you grant the distinction, and the need to express the necessary connection, you can indeed still assert that the priority goes either way. I find the claim that the priority lies with the present instantiation of past-oriented properties extremely implausible, but I don’t think I say enough in the current draft about why.
Granting the necessary connection, the deviant examples are all per impossible. But thinking about per impossible examples can help tease out priority intuitions. So here is my intuition: if Bush now has the property *having been President*, but did not once have the property, *being President*, then, really, he never was President. The property *having been President* is, as far as I can see, a property that he now has in virtue of once having the property *being President*. (Ben Caplan and I present some arguments to this effect in our “The Way Things Were” (2005), but I don’t know if you will find those arguments any more convincing!)
I think it’s a bit of a silly view myself, but know a few people who half-heartedly prefer it. That being said, I don’t quite follow your point in the final paragraph. Roughly, the view I had in mind goes something like this: in e.g. 2002, Bush has the property of being president, and this grounds the fact that, in 2015, Bush has the property of having been president (so, the grounding/explanation goes in the direction that you – and I, for that matter – find plausible). However, in e.g. 2015, the grounding/metaphysical explanation has, if you like, flipped: Bush’s having the property of having been president grounds the fact that, in 2002, Bush has the property of being president. So, speaking now, Bush was President, though what grounds this fact changes with each moment (at least in token, though not necessarily in type).
Regardless, I’ll definitely check out the paper you mentioned – sounds interesting!
I know some people who seem to prefer it to! I just think it is seriously wrong-headed 🙂
It’s the flip that seems wrong to me. Speaking now, nothing about how Bush presently is grounds how he was in 2002. A property like having been President is only suited to do that work if we suppose that it is not a property he has in virtue of how he is, but a property he has in virtue of how he was, back in 2002.
David
We agree that when so-called ‘hypertime’is posited it is then the real time and the time-like dimension of the back block is quasi-time or as I call it hypo-time.
You talk of the past ‘continuing to exist’. That is hyper-presentism. At lest on my version of the growing block the past exists simpliciter. I find it helps to think of God in these cases (but maybe that is because I believe in Him/Her!,). Our specious present is only a minute long. Think of God knowing the whole back block as Her/His specious present . It is all there, and real in just the way the presentist says only the present is real.
Peter
Thanks Peter. I do talk about the past ‘continuing to exist’, but I don’t mean anything by that except that it exists simpliciter. If something did not continue to exist, that would mean that it stopped existing simpliciter. It seems to me that we both agree that existence simplicter is not permanent, in the sense that you think that ever more things will come to exist simplicter, that did not used to exist simpliciter. I agree, and also think that some things that exist simplicter will cease to exist simpliciter. So I don’t intend these tensed ways of talking about existence to imply the existence in question is anything other than existence simpliciter.
I’m not sure what to say about how all of this relates to God. On my preferred reading of Boethius, for example, God knows both past and future times as eternal because he knows things not according to their nature, but according to his nature, and his nature is eternal. So it doesn’t follow, from the fact that he knows them as eternal, that they are eternal. God’s view, on this sort of account, does not reveal the true nature of created world; it simple reveals the true nature of God.
Hi David, thanks for your paper! So, here’s an all too brief summary of what I took to be the thrust of your paper—please correct me if I’m off here!:
All dynamic theories of time require that present events be fundamental; and therefore past events, on such theories, must be analysed in terms of your ‘once-present’. Given this, any dynamic theory is going to have to ground present truths about the past in facts that no longer exist. This levels the playing field, with respect to the grounding objection, between presentism and other dynamic theories. Furthermore, given that we don’t, fundamentally, need the ‘now-past’ to explain or ground the past, we should feel less pressure to reject presentism in favour of one of the other dynamic theories.
Assuming that’s right, here are some thoughts:
I think I agree that the standard renderings of the growing block and moving spotlight theories require your ‘once-present’ in accounting for the past (incidentally, I think that this leads to incoherence for those renderings), and I agree that this levels the playing field with respect to the grounding objection. A possible exception might be Tooley’s version of the growing block, since it admits the concept of truth simpliciter, and therefore can plausibly ground truths about the past in facts that are part of reality (though I still think Tooley has an incoherence problem).
But I’m wondering whether this is as welcome a result for the presentist as you imply. I guess you’re going to have to endorse something like Merricks’ view, according to which truth depends on the world, but not in any substantive way. Rather than biting that bullet, shouldn’t we hold out for a theory of time which avoids both the grounding objection and incoherence?
I should note, for myself, it was never the problem of grounding past truth that led me out of the presentist camp, but the problem of grounding temporal asymmetry in a way that respects the dynamic nature of time. Even acknowledging your arguments here, I think the growing block theory, with respect to that problem, still has the upper hand (I argue for this in my 2005 AJP paper).
Thanks again!
Thanks Joseph. Your summary gets it right.
This paper came out of my dissertation (2005), which included a separate chapter arguing that Tooley’s view was incoherent, but for different reasons, so I think we are on the same page about that.
I think the result should be welcome. The point is supposed to be that, if you were attracted to a dynamic theory to begin with, this commitment was already implicit in the ontological attitude you wanted to take toward change (or, at least, toward some changes). It feels to me like, when this commitment is made explicit, a lot of people want to run away from it. But I think we should instead try to work with it, and see where it takes us.
I agree that grounding temporal asymmetry is a different project. I take it that the asymmetry you are most keen to capture is an asymmetry in determinate identity: past things have thisnesses; future things (as it were) do not. As I mentioned in my comments on your paper, I am still not a big fan of thisnesses. But I am a fan of this asymmetry. I just don’t think it is an ontological distinction.
More specifically,
1. I don’t think it gives us any reason to suppose that past things exist (because I take it that, in virtue of being committed to a dynamic theory, we are committed to explaining truth by pointing to nonexistent facts, and I don’t see a big gap between that and explaining reference and determinate identity by pointing to nonexistent objects).
and
2. I don’t think it gives us any reason to suppose that we can’t now make true claims whose truth depends on the not-yet determinate identities of future objects. That is, I think we can now express true singular future contingent propositions (“Antichrist will come”, to use the medieval example, or “Newman1 will be a girl”, to use Kaplan’s example), and I think that both their truth and singularity depend on, and can be explained by, facts and things that do not yet exist and do not yet have determinate identity.
Holding off on replying to these further comments until I finish reading and replying to Joseph’s paper!
Thanks! My two pennies’ worth in the meantime:
I’m really sympathetic. But the way I see it, your admission that (2) on p. 11 can be analyzed in the same way as (1), which I think is right, amounts to the admission that ‘once-past’ doesn’t force the intended interpretation either. And as you say, we can take tense operators to capture it, but we needn’t. That means the distinction in question is quite elusive. Moreover, even if an A-theorist allows for two senses of ‘what was the case’, why couldn’t they just hold that deviance is metaphysically impossible, which is to say it’s metaphysically necessary that what was the case (in the first sense) is such that it was the case (in the second)? On p. 4, you seem to consider such a move and reply that if that’s so, we can’t tell by considering something like Fig. 2, whether E0 is really past. But isn’t that then just because we can’t tell whether Fig. 2 is an accurate description of reality, i.e. whether the snapshot it depicts really is the way things are (now)? That seems ok?
I’m inclined to think that the need to recognize the distinction you have in mind is closely related to the need Kit Fine feels to recognize non-standard realist positions.
Thanks Natalja.
I agree that the distinction is elusive. I think I am committed to saying that there is no clear way to make it in terms of what we say, since what matters is how we link what we say to the underlying metaphysics.
I agree that the A-theorist should say that such deviance is metaphysically impossible. But I think that can be so in at least two ways: first, it can be trivial, because she refuses to allow any way of expressing the “once present”, so the necessary connection is just between two ways of expressing the “now past”. Or, second, it can be substantive, because she allows the distinction, and asserts a necessary connection between the existing facts (e.g., that some event is earlier than a present event) and the facts that once existed but don’t (e.g., that that event was present). If she goes that way, then I want to encourage her to think that the the existing facts are not the “real” past, but only a trace of it. Once she admits that, I think I have established what I want to establish.
I don’t think nonstandard tense realism helps here: it places everything on a par—either all fragments of the same incoherent reality, or all real relative to some reality-external perspective. It seems to me that the kind of change posited by a dynamic view requires the standard tense realist idea that only one “perspective” is real, but also requires the recognition that what was once real (or will be real) continues to be important to explanation of what has happened to reality as a whole…
I agree that the A-theorist should say that such deviance is metaphysically impossible. But I think that can be so in at least two ways: first, it can be trivial, because she refuses to allow any way of expressing the “once present”, so the necessary connection is just between two ways of expressing the “now past”. Or, second, it can be substantive, because she allows the distinction, and asserts a necessary connection between the existing facts (e.g., that some event is earlier than a present event) and the facts that once existed but don’t (e.g., that that event was present). If she goes that way, then I want to encourage her to think that the the existing facts are not the “real” past, but only a trace of it. Once she admits that, I think I have established what I want to establish.
Ok.
I don’t think nonstandard tense realism helps here: it places everything on a par—either all fragments of the same incoherent reality, or all real relative to some reality-external perspective. It seems to me that the kind of change posited by a dynamic view requires the standard tense realist idea that only one “perspective” is real, but also requires the recognition that what was once real (or will be real) continues to be important to explanation of what has happened to reality as a whole…
As far as I understand it, Fine’s external relativism doesn’t say that different sets of tensed facts are ‘real relative to some reality-external perspective’; I take it to say that there is no atemporal perspective from which to assess what’s real. But I agree it doesn’t help in the end, I just think that the impulse to endorse it is just the same as the one you’re describing, since it has to do with not just the tensed facts that obtain now composing reality, but also the tensed facts that obtain at other times. (One thing you might protest about here is why there is apparently no room in this schema for facts having composed reality.) So I think the omission Fine thinks he’s spotted is very similar to the one you’re describing, and similarly elusive.
One of my reasons for preferring a growing block (strictly a growing hyperblock) theory was to capture the intuition that the past is fixed. I can see that your distinction between the now now and the once true explins the fixity of the past provided we grant that the once once is once.
I have another reason. Part of the case for a dynamic theory of time is based on the phenomenology, an so we dynamists, of whatever sect, should grant the reality of more than the knife-edge present. Not only can we surmise that the specious present could be rather long (maybe if I fall into a coma for a year the whole year is art of the specious present that includes falling into it and coming out of it.) but it could be much shorter (Tourette’s? Non-human animals?). The only princioled ways of extending reality from the knife-edge present are to whole Block or something like the Growing Block or Falling Branches etc.
I think I’m with you on this Peter, though I have not been so worried about phenomenology in particular. It seems like the Presentist ends up being committed to supposing that everything is instantaneous. And my sort of Presentist seems to be committed to reducing everything to a succession of instantaneous facts or events. But there are good reasons for thinking that we cannot do this.
Freddoso, in his “Accidental Necessity” paper, argues for what he calls the “Primacy of the Pure Present”, and seems to imagine that the Pure Present will be momentary. While I am very sympathetic to much of what goes on in that paper, I agree that it is wrong to suppose that we can reduce all of what happens over time to facts about what happens at each instantaneous moment, and relations among those facts.
I’m not sure where this leaves the Presentist. I am inclined to move toward a view that allows that there are entities (intervals, temporally extended events) that never wholly exist, and are more fundamental than their momentary parts.