The Growth of the Growing Block Theory of Time: Early Twentieth
Century Roots and Reactions – Emily Thomas (Groningen)
The growing block view of time holds that the past and present is real
whilst the future is unreal; as future events become present and real,
they added on to the growing block of reality. Surprisingly, given the
recent interest in this view, there is very little literature on its
origins. This paper locates those origins in three early twentieth
century British works: Samuel Alexander’s (1920) Space, Time, and
Deity, C.D. Broad’s (1923) Scientific Thought, and Hilda Oakeley’s
(1931-2) “The Status of the Past”.
In addition to providing a general history of the view’s
origins, I advance two specific theses. First, I argue that although
C. D. Broad is the first advocate of the growing block theory, fellow
emergentist Samuel Alexander first articulated the view. Second, I
argue that it was Broad’s relationism about time, coupled with his
newfound conviction that time has a direction, that led him towards
the growing block theory. By way of tying these theses together, I
argue that Broad’s views on the direction of time – and possibly even
his growing block theory – are sourced in Alexander.
Full paper 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
Most interesting! I have one question concerning Alexander’s complaint that the Growing Block is not compatible with a smooth dynamics. Is one premise in his argument the thesis that time has the ordinal structure of the natural numbers, which I hold is true of hypertime?
Thanks Peter! Unfortunately, Alexander doesn’t make that argument (although it would have made a neat contemporary link); instead, it’s closer to the thesis associated with Descartes that the world would be recreated from moment to moment.
A very interesting paper which, among other things, reminds me that I really should make another attempt to read Alexander!
One curious thing is that while you quote Broad as citing books by Alexander and Whitehead as the source of his view, there’s no discussion at all of Whitehead in the paper. My assumption for awhile had been that Broad basically got the GBT from Whitehead. One finds, for example, the following passage in Whitehead’s Enquiry Concerning the Principles of Natural Knowledge (Sect. 14.3):
“Events never change. Nature develops, in the sense that an event e becomes part of an event e’ which includes (i.e. extends over) e and also extends into the futurity beyond e. Thus in a sense e does change, namely, in its relations to the events which were not and which become actual in the creative advance of nature. The change of an event e, in this meaning of the term ‘change’, will be called the ‘passage’ of e; and the word ‘change’ will not be used in this sense. Thus we say that events pass but do not change. The passage of an event is its passing into another event which is not it.”
It may be that Whitehead is influenced by Alexander here as well (this was published a year before ST&D, but after the Gifford lectures were delivered, so I’m not sure if he would have had access to the material). I kind of suspect Bergson may be in the background here as well, but I don’t know him well enough to say for certain.
Thanks David, that’s very kind. You’ve also highlighted an issue that I wondered about and ultimately set aside (for this paper, at least). I read both bits of both of Whitehead’s pre-1921 texts – Enquiry and The Concept of Nature – but ultimately was unsure how to take them. Whitehead’s discussions about the future weren’t explicitly about their reality, but rather about tangential issues. So, although I definitely think that Whitehead plays a role in this story (even if, as you say, there’s an earlier line of influence from Alexander to Whitehead) I found that role much murkier than Alexander’s.
David’s question was similar to my own.
I wonder if there is merit in including a bit explaining why Whitehead and Bergson prima facie look relevant, but are not featuring in the paper, just to acknowledge that you are aware of them. In my experience of defending the Growing-Block, the two names that come up most other than Broad are Whitehead and Bergson.
My take is that Whitehead pretty consistently holds a Growing-Block view from at least 1919 onward, but unfortunately there’s nothing near as clear an articulation of it as one finds in Broad (which is probably why everyone cites Broad as the source of the view instead of Whitehead, even though Broad only held it briefly). You certainly find a lot of Whitehead’s followers (e.g. Hartshorne) very clearly embracing it.
Okay great, thanks both. David – I completely agree with that, especially Hartshorne.
Graeme – good idea, I’ll add a note explaining this!
Hello Emily – I thoroughly enjoyed your paper – a very interesting read. You must have had some good teachers at some point in the past.
I did have one quick question regarding Broad’s relationism. The question is about temporal positions. Do you know whether Broad said anything about how an object’s temporal position changes as new layers of the block come into existence? I’m thinking here that a substantivalist Growing Blocker can say that every object has a fixed temporal position, and that as the block grows it retains that position despite the fact that new temporal positions come into existence (i.e. such things come to bear new temporal relations to the new things at the new positions, but retain their old positions nonetheless). But this view doesn’t appear to be available to the relationist, who must presumably think that spatiotemporal positions are (to use a term that seems to have dropped out of fashion) individuated in terms of their relations. [Very briefly, so you can see why I’m asking: My concern here is to do with de re reference to times. Indexicals are often thought to be devices that enable us to make de re reference to particular times – on Kaplan’s view, for example, times are loaded directly into the propositions that are expressed using token sentences containing temporal indexicals. But, if the relationist Growing Block view is true, then it seems there are no ‘stable’ entities to serve as times, and this wreaks havoc. In particular, on Kaplan’s view it is crucial that sentences like “Yesterday it was sunny” used today, and “Today it is sunny” used yesterday, express the very same proposition, but I can’t see how this can be made to work on a relationist growing block view. And so this is why I was interested to hear about Broad’s relationism, and interested to learn whether Broad says anything about temporal positions]
Hah yes, good teachers indeed thanks!
Your question is interesting but, as far as I’m aware, Broad doesn’t discuss the positions of things in time. (If anyone else knows better, please do let us know.) Oddly enough though I’m currently writing about this issue in the context of early modern relationism, and there the lack of absolute position in time is usually assumed to be one of the hallmarks of relationism.
My only thought is that I’m not convinced there’s a special problem for the relationist growing block theorist here; if an object’s spatiotemporal position can be individuated by relations at all, why wouldn’t the relations it bears to past objects be sufficient?
Thanks Emily. I don’t myself see why an eternalist relationist has to think that there are no absolute positions in time. Or, at least (given that the term ‘absolute’ is used in a number of different ways), they don’t have to be anti-realists about positions in time, they just have to think that what positions there are supervene on the spatiotemporal arrangements of objects. And so, because the spatiotemporal arrangement of objects across an eternalist universe never changes, the positions never change either. We thus have a stable set of entities (albeit supervening, and so non-fundamental entities) to serve as times that indexicals such as ‘yesterday’ can latch on to. Matters are different, though, for the growing block theorist, precisely because the spatiotemporal arrangement of objects does change. As the universe changes, so does the subvenient base upon which positions in time might be thought to supervene, and I can’t see any sensible way of identifying the positions that exist before and after such a change. (Of course, we can employ a description such as ‘the point in time such that Caesar crosses the Rubicon’ but such descriptions are de dicto and nothing guarantees that a token used before a change in the universe refers to the same thing as a token used after.) At any rate, this is my concern – I don’t expect you to engage with it as it’s not on the topic of your paper – I just thought that Broad might have said something interesting and relevant – but it seems not!
Thanks Ben. This is interesting, so don’t worry about being slightly off topic. I’ll expand a bit on what I said before, and we’ll see if it’s useful.
I think the following is true: on the pre-relativist picture, all thinkers who deny there are absolute locations in space and time are relationists. However, I completely agree that relationists need not deny there are absolute locations (in fact, I read Isaac Barrow as a modal relationist who accepts absolute locations).
On your special problem for relationists seeking location in the growing block, I believe you’re thinking something like this: for the relationist, an object such as ‘o’ is located in time via its temporal relations to objects that existed before it (j,k,l,m,n) and after it (p, q, r, s). And, on the growing block view, o’s position in time would change as it ceases to be present and becomes past, as its new relations (to p, q, r, s) come into existence. My thought is this: isn’t it the relations that o already bears to the objects that existed before it (j,k,l,m,n) sufficient to ground its unchanging temporal location, if those prior objects continue to exist?
Hi Emily. Iv’e just seen it’s the last day of the conference, so I’m replying very quickly as comments may close (I don’t know whether they will or not). In short, yes, I took this to be the natural suggestion, but I don’t think it can work. Your appeal here is to another description (‘The position occupied by o and such that the positions occupied by j, k, l, m, and n precede it’). If the other positions mentioned in this description could be individuated without the use of a description this might work, but how is this possible? And in the absence of this the description is just like the one I suggested previously (i.e. ‘the point in time such that Caesar crosses the Rubicon’). And then I ask again, what guarantees that such a description picks out the same thing before and after a change in the universe? (Indeed, I find it hard to see what could be meant by the claim that such a description does pick out the same thing before and after a change if there are no independently existing entities to serve as positions.)
Thanks! I enjoyed reading this. So I take it that there is a deep tension in Alexander’s thought, insofar as he leans towards something like the block view in general, but leans towards something like the growing block view when it comes to thinking about the nature of God? Would you call him a pantheist of sorts (since God is the world, at some future stage)?
Yep definitely – a deep, deep tension! I believe that pantheists usually identify God with the world, so Alexander is doing something even stranger by placing God within it (i.e. the world is not exhausted by God).