- What follows below is the full text of the Introduction. I have divided it into the two sections – Preamble and Summary.
- I have also added the occasional footnote1 of my own, and added links to whatever I have by the authors mentioned.
- It is also worth referring to:-
in order to see how Mellor’s thoughts have developed between the two versions of the book.
- "Mellor (D.H.) - Real Time: Introduction and Summary", in
- "Mellor (D.H.) - Real Time", as well as
- "Mellor (D.H.) - Real Time II: Preface" in this book ("Mellor (D.H.) - Real Time II")
Preamble (Full Text)
- Like its ancestor, "Mellor (D.H.) - Real Time", this book is about such of the metaphysics of time as follows from settling the basis in reality of our distinctions between past, present and future. One way and another that covers most of time’s metaphysics, including the following questions2:-
All these questions will be answered in the course of settling the status of past, present and future.
- What makes a statement that something is past, present or future true or false?
- What is it to have the thoughts that such statements express, and why do we have them?
- How do we know when they are true: what tells us whether something is past, present or future?
- Why do we only ever act and have experiences in the present, affect what is future and see what is present or past?
- Could there be exceptions to this: could a time machine3 let us see the future or affect the past?
- What, even in time machines4, makes everything outside us keep moving from our future to our past via our present, i.e. what makes time flow?
- What has the flow of time to do with change, why has it no spatial analogue, and what does this fact tell us about how time differs from space?
- There are of course other metaphysical questions about time, and three5 of them at least I shall not try to answer.
These questions are too independent of those I shall discuss, and too large, to be tackled here. (The last turns too much on cosmology, the second on how far the measurement of time depends on conventions, and the first on the logic of infinity.)
- Is time continuous, dense or discrete?
- What makes time intervals differ or be the same in extent?
- Has time a beginning or an end?
- On the questions I shall discuss I noted in my 1981 "Mellor (D.H.) - Real Time: Introduction and Summary" that the literature was then so vast and various that I could not do it justice succinctly without obscuring the argument I wished to present. The situation has not changed. If anything, the flow of philosophical works on time, if not of time itself, has grown since then. In writing this book I have taken account of it, and of my own changing views, in the ways indicated in "Mellor (D.H.) - Real Time II: Preface". But it is no more feasible now than it was then to discuss all the works on time and related matters that I have read and been affected by. So now as then I hope that the fewness of my references will not be taken as a sign of ignorance or arrogance. On the contrary, it means that I take my main points, if disputed, to be so common in the literature as to be by now public property. So as before I trust that no offence will be taken by, or on behalf of, those whose well-known work I adopt, adapt or reject without explicit acknowledgement.
- This book in short is not a guide to the philosophical literature on time. Like "Mellor (D.H.) - Real Time", it is a self-contained, argued exposition of a theory of time many of whose points are already familiar to philosophers. As before, its aim is simply to develop and defend the theory in plain terms and to show how ramified its consequences are. There if anywhere lies its claim to be read. So, to fortify that claim, I will now sketch the main points I wish to make, in case they are obscured by the detail of ensuing chapters.
Summary (Full Text)
- My answers to the main questions I tackle are as follows. There is in reality no such thing as being past, present or future. By this I do not mean that it is never true to call an event e past, present or future: that would be absurd. The question is, what makes a statement like ‘e is past’ true when it is true, namely at any time later than e? There are two answers to this question. One is that at any such time e has the property of being past. This is what, in "McTaggart (J. McT. E.) - The Unreality of Time" (1908) now standard terminology, I call the ‘A-theory’ view. My own ‘B-theory’ view is that what makes ‘e is past’ true at any time t is the fact that e is earlier than t. Similarly, what makes ‘e is present’ true at any t is e’s being located at t, and what makes ‘e is future’ true at any t is e’s being later than t.
- This B-theory entails that there are no such A-properties as being past, present or future, since e’s being past, present or future at t is by definition what makes ‘e is past’, ‘e is present’ or ‘e is future’ true at t. So if, as the B-theory says, that is not what makes these A-statements true, these A-properties do not exist. Compare for example ‘Nobody is smaller than a flea’. What makes this true is not that there is such a person as Nobody, who is smaller than a flea, but that no person is that small. So there is no such person as Nobody, since if there were all true instances of ‘Nobody is F’ would by definition be made true by Nobody’s being F. As that is not what makes any such statement true, there is no such person.
- But why is no true ‘Nobody is F’ made true by something’s — Nobody’s — being F? The reason is that to make all such statements true Nobody would need impossible combinations of properties. Not only for example would Nobody have to be smaller than a flea, he and she (Nobody is both male and female …) would also have to be larger than a galaxy6. But as nothing can be both smaller than a flea and larger than a galaxy, there can be no such person. So every true ‘Nobody is F’ must be made true, as it obviously is, by something else, namely the fact that no person is F.
- Similarly7, if less obviously, with ‘e is past’, ‘e is present’ and ‘e is future’. These A-statements cannot consistently be made true when they are true by e’s having the A-properties of being past, present and future — or so McTaggart argued. I agree with him, and will defend and update his proof of this in chapter 7 ("Mellor (D.H.) - McTaggart's Proof"). This is why I say that there are no A-properties, and that our A-statements need other truthmakers. And then it is obvious what their truthmakers are, namely those given above by the B-theory.
- Unfortunately not everyone accepts McTaggart’s proof. So to win the day as well as the argument we B-theorists must press other objections to A-theories of time. One is that important consequences drawn by many A-theorists from events being past, present or future conflict with modern cosmology, which makes simultaneity at a distance relative to an arbitrary choice of so-called ‘reference frame’. For a star like Sirius, about ten light-years away, this means that no frame-independent fact fixes which Sirian events within a twenty8 year span are simultaneous with your reading of this sentence and therefore present. This poses no problem for those A- and B-theorists who think it is always true to say of any actual event, whenever it occurs, that it exists9. But most A-theorists (and some B-theorists10) say that nothing exists until it is present11, and for them, modern cosmology poses a real problem, since no one can seriously take existence12 to be relative to an arbitrary choice of reference frame.
- Besides pressing objections to A-theorists, we B-theorists must dispose of their objections to us. One, dealt with in13 "Mellor (D.H.) - Real Time", is that A-statements cannot be translated by B-statements. Thus ‘e is past’, said at t, cannot mean14 that e is earlier than t. For if ‘e is earlier than t’ is ever true it is always true, while ‘e is past’ is true only after e; and obviously no two statements can be synonymous if one can be true when the other is false. But this does not stop the fact that e is earlier than t making ‘e is past’ true at t. Compare for example ‘I live in Cambridge’ said by any person x (where ‘x’ is a name like ‘Hugh Mellor’, not like ‘me’). This cannot mean that x lives in Cambridge, for if ‘x lives in Cambridge’ is true, it is true whoever15 says it, whereas ‘I live in Cambridge’ is only true if said by a Cambridge resident. Yet16 no one can deny that the truthmaker for ‘I live in Cambridge’, said by any x, is that x lives in Cambridge.
- We B-theorists need not then make the hopeless claim that A-statements are translatable by B-statements. Yet despite this we can still say in B-theory terms what A-statements mean. For all I need to know in order to understand and use the sentence ‘e is past’ is that it is true at any time t if and only if t is later than e. Similarly, all I need to know to understand and use ‘I live in Cambridge’ is that this is true if and only if a resident of Cambridge says it. So on the orthodox view of a sentence’s meaning, as what we must know in order to be able to understand and use it, what ‘I live in Cambridge’ means is a function, from any person x to its truth-maker if said by x, namely that x lives in Cambridge. Similarly, I say, what the A-sentence ‘e is past’ means is a function from any B-time t to its B-truthmaker at t, namely that t is later than e; and similarly for ‘e is present’ and ‘e is future’.
- This kind of B-theory, giving a semantics as well as a metaphysics for A-statements, has been called ‘the new theory of time’, to distinguish it from older17 B-theories that try to translate or dispense with A-statements. My B-theory does neither. For not only, as we have seen, does it show why A-statements are untranslatable, it also shows why we need the A-beliefs these statements express. The reason is that we need such beliefs to make us act successfully when our success depends, as it nearly always does, on our acting only at certain times. I know for example that it is no use shopping when the shops are shut. So I shop only when, and because, I believe they are open, i.e. open now. If this A-belief of mine is true, I will succeed in shopping, and if it is false I will fail. This is a quite general link between belief and truth: truth is that property of beliefs which makes the actions18 they combine with our desires to cause succeed, i.e. achieve the objects of those desires. So if an action’s success depends on when it is done, its causes must include a belief which is not always true and which we therefore try to have only when it is true. But as we have seen, being true only at certain times is the mark of an A-belief, since all B-beliefs that are true at all are true always.
- This is why no agent, human or animal, that acts on its beliefs, and must act at certain times to get what it wants, can do without A-beliefs. But then we agents must be constantly changing our A-beliefs, especially our beliefs about what is happening now, in order to try to keep them true. These changes in us, mostly prompted by our senses, are what make us think of time as flowing, even though it does not flow. For without such properties as being past, present and future, time cannot flow, i.e. make events change from being future to being present and then to being past. But time need not flow for things to change in other ways, by having other incompatible properties, like different temperatures, at different times. And that includes me changing my A-belief that e is future successively for the incompatible A-beliefs that e is present and that e is past. This sequence of A-beliefs is a real process of change — but only in me19, not in e. Changes like this are the psychological reality behind the A-theorist’s myth20 of the flow of time.
- This is why our undeniable experience of time flowing is no evidence that time really does flow. Nor does the undeniable fact that all our actions and experiences occur in the present support the A-theory. For on a B-theory it is true by definition to say, at any time t, of anything at t, that it is present. So in particular it is always true to say or think, whenever we do or experience anything, that this action or experience of ours is present. The confining of all our actions and experiences to the present is not a real constraint on their location but a mere tautology21, like my inability to be anyone but me or anywhere but here.
- This account of the presence of experience also explains how we can tell whether an event is past, present or future. No B-theory can admit that these non-existent properties of events are perceptible, and indeed they are not. We cannot for example refute someone who claims to see the future in a crystal ball by pointing to the visible pastness of the scene it shows, since there is no such thing. Whether it is past or future, the scene will look the same. Nor can the look of a supernova tell us directly how long ago it happened. That we must infer from its independently known distance and the speed of light, which tells us how much earlier it is than our experience of seeing it — and hence, since we are seeing it in the present, how past it is. Similarly with the terrestrial events which, as we see them, we take to be happening almost now, because the light we see them by takes so little time to reach us. This is what tells us that most of what we are seeing is, for all practical purposes, present.
- How can we tell when events are future? We cannot tell this by seeing them, and not just because being future is not a visible property of events. For while an event is future we cannot see it at all, since the future is by definition later than the present, seeing something (or perceiving it in any other way) is always an effect of what is seen, and causes always precede their effects. But if this link between the causal order of events and their time order stops us seeing directly that an event is future, it can enable us to tell this indirectly, as when we predict the future effects of our present actions or of the other present events that we can now see.
- The link between time and causation22 also enables B-theorists to explain why it is that, although — or rather because — we can see the past, we cannot affect it. To do so, of course, this link must be, as it is, consistent with a B-theory of time. But consistency is not enough: to be adequate, our theories of causation23 and time must between them explain why the link holds. Many philosophers explain this by using the time order of events to fix their causal order, defining a cause as the earlier of two causally related events. I do it the other way round, using an independently defined causal order to fix the time order of all events, even of those that are not related as cause to effect.
- This enables me to meet several other challenges to B-theories of time. First, it lets me distinguish time from the dimensions of space, and define its direction, not as that in which time flows but as that of causation24. For while a cause often has effects in the same place as itself, or in any spatial direction from it, this is never so in time. A causal definition25 of time order explains all this by requiring a cause’s effects to occur in the same temporal direction from it, namely later. This can then be, and I argue is, what makes time differ from space and gives substance to its direction, i.e. to the difference between being earlier than something and being later than it.
- Second, we can use the causal machinery of memory to show why the time order we perceive in the world, and in our own experiences, must generally coincide with that given by the causal order of events. By doing so I show in particular how a causal definition of time, and of time order, entails the success of Kant’s famous definition of time as ‘the form of inner sense’, i.e. as the dimension in which our own experiences are ordered.
- Third, if time is the dimension of causation26, the need for causation27 to preserve the identity of a thing as its properties change shows why, since causes and effects cannot be simultaneous, there can be no spatial analogue of change. This in turn, by explaining why only temporal and not spatial variations of properties can be changes, meets the most persistent objection to B-theories, namely that only the A-theory’s flow of time can account for the essentially temporal nature of change. That is not so: the dimension of change must also be the dimension of causation28.
- Finally, I can explain why no one can affect the past or recall the future. This might seem to follow directly from a causal definition of time order, but in fact it does not29. For the definition does not itself rule out the causal and hence temporal loops that would let us affect our past by letting us travel back in time. Suppose for example Dr Who’s time machine30 TARDIS could travel back in an hour from 2045 to 1945. Then he could recall in 1945 what he had seen and done an hour earlier in 2045, and those events could affect events in 1945, e.g. by causing a fashion for futuristic scarves of the kind he decided in 2045 to wear.
- Why can this not happen? The classic answer is that, if it could, Dr Who could, e.g. by killing his infant grandmother, cause himself never to leave in 2045 (by causing himself not to exist), thus causing a contradiction, which is impossible. Although this answer has convinced few philosophers, it is in fact correct, as I shall show in the last31 chapter. To find out why — and why I make the other claims outlined above — now read on …
Useful summary of the book's conclusions. Compare with "Mellor (D.H.) - Real Time: Introduction and Summary".
Footnote 1: Some of which may be removed or clarified if and when I read the book.
Footnote 2: This is a different set of questions to those in "Mellor (D.H.) - Real Time: Introduction and Summary".
Footnote 5: There was a 4th non-discussed question in "Mellor (D.H.) - Real Time: Introduction and Summary".
Footnote 6: Because “Nobody is larger than a galaxy”, and “Nobody” is taken to be univocal.
Footnote 7: Just what is the analogy here? Presumably “F” corresponds to “e and “Nobody” to “past” (etc.). The ordering is reversed.
Footnote 8: Why?
Footnote 9: Why is the existence of any event key here?
Footnote 10: Remember that Mellor is a B-Theorist, though presumably not of this camp?
Footnote 11: What does this mean? Just that future events do not exist, whereas present and past ones do? So, most A-theorists are supporters of the “growing block universe”? They are neither presentists nor eternalists.
Footnote 12: Agreed. This is (slightly) analogous to the “arbitrariness” – or “only a and b” – objection to “Closest Continuer” (Click here for Note) theories of personal identity.
Footnote 13: Is it also dealt with in "Mellor (D.H.) - Real Time II"?
Footnote 14: Note that this is a question of meaning, not truth.
Footnote 15: Ie. Any y can say this truly of x, if x lives in Cambridge.
Footnote 16: So, the analogy with time is … ?
Footnote 17: Who propounded these, and what were they?
Footnote 18: So, this adopts the standard Belief / Desire model of Action. Does everyone (and does it matter)?
Footnote 19: While what Mellor says is true, these changes are NOT merely psychological, but – in general – responses to the external world.
Footnote 20: Can the B-theorist really not accommodate the “flow” of time?
Footnote 21: These are indexicals, but the term doesn’t seem to be used by Mellor.
Footnote 25: As in "Mellor (D.H.) - Real Time: Introduction and Summary", Mellor “defines” time – is this possible? Also, the “definition” appears to rule out – a priori – backward causation. But Mellor later says “no” – because of the “possibility” of causal or temporal loops – though backward causation is later ruled out for logical reasons – the well-known “grandfather paradox” (though Mellor uses “grandmother”).
Footnote 29: Why not?
Footnote 31: See "Mellor (D.H.) - The Linearity of Time". Why has the argument “convinced few philosophers”?
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2018
- Mauve: Text by correspondent(s) or other author(s); © the author(s)