The Atlas of Reality
Page 73
Tensers will respond that there is no good reason to lump tense in with such indexicals as ‘I’ or ‘here’. Instead, tense is much closer to grammatical “mood”. We use the indicative mood to say what simply is the case, and we use the subjunctive mood to talk about what isn't but might have been. Similarly, Tensers can insist that we use the present tense to speak about what simply is the case, and we use past and future tense to speak about what isn't but was or will be.
The indicative mood (and related adverbs like ‘actually’) does not seem to be at all indexical. When one says that the earth actually has only one moon, one is not speaking about the location of one's utterance in some kind of “logical space”. One is not saying that in our local reality, it is the case that the earth has only one moon. One is saying that the earth has only one moon simpliciter. If one says that the earth might have had more than one moon or no moon at all, one is not speaking of something true in some other, more remote part of reality. One is saying that our reality had the potential at one point to produce a three-moon or no-moon system around the earth. If someone were to say that the earth has actually had three moons, we would suppose that what he said was simply false. We wouldn't have to check first to see who said this and where in logical space the assertion was made.
Tensers can appeal to many parallels between tense and mood. For example, we sometimes talk about timeless entities such as numbers. When we do, we almost always use the present tense. We say that two plus two is four, not that it was four or will be four. This suggests that the appropriate use of the present tense is to speak of how things are simpliciter, without qualification about temporal location. This is why it is appropriate to use it in speaking of things entirely outside of time.
Some Anti-Tensers, in particular those like David Lewis (1986a) who are Modal Indexicalists, object to this line of argument on the grounds that the indicative mood and related adverbs like ‘actually’ or ‘in actuality’ are indexicals after all! Concretists, for example, say that when we speak of how things are in actuality, we are only referring to our local universe. Somewhere out there are speakers who can truthfully say, ‘There are actually three moons around the earth’, just as there are speakers in other times who can truthfully say, ‘The Roman Empire is now intact’1 or ‘There are human colonists on Mars.’ If these Modal Indexicalists are right, then Tensers lose an important argument for their position.
20.5.3 The experience of the flow of time
Most arguments for Tensism from experience have focused on aspects of experience that appear in a single moment, but it seems that the most interesting challenges to Anti-Tensism come from considering experience holistically or diachronically.
There is a methodological assumption lurking here. We take it for granted that a metaphysical theory gains support whenever it enables us to take the world of phenomenal appearance at face value, as accurately representing the world as it is. This is in the neighborhood of Reidian Common Sense (PEpist 2). Obviously, this evidence is defeasible, since there are cases where it seems plausible at least to take appearances to be misleading (in the case of secondary qualities, for example). Defeasible evidence is still evidence, however. If Anti-Tensism is able to explain, in the sense of explain away, the phenomena of experience, this lightens somewhat the evidentiary strength of that phenomenon for Tensism, but it doesn't entirely neutralize it.
20.5.3.1 Experience considered atomistically
There is one datum concerning our temporal experience considered at a moment that seems to tell against Anti-Tensism, namely, the palpable fleetingness of the present moment. The sense we have of the present's having just succeeded a past moment and its being about to be succeeded by a new present. This fleetingness seems to represent moments as coming into and out of existence. (An excellent source for the phenomenology of time is Augustine's Confessions, Book 11. We don't know of anything better.) Anti-Tensers must attribute this sense of fleetingness to a kind of illusion. Past moments don't pass out of existence nor do future moments emerge from nothingness. They are all eternally existent. Some are merely “out of view”, so to speak.
Anti-Tensers might claim that our sense of fleetingness is simply a perception of the brevity of the present moment, but this seems to miss the mark. Even when the present moment is specious. That is, even when we seem to experience an extended duration as present, the sense of fleetingness is not attenuated. So, it seems that the sense of fleetingness is independent of the length of the present moment.
20.5.3.2 Experience considered holistically
As we saw when we considered Zeno's paradoxes in Chapter 19, there are reasons to deny that dimensionless instants of time are ontologically prior to finite periods of time. That is, there are good reasons to embrace Intervalism (19.1A). Untensed facts about time do not supervene on facts about dimensionless time-slices. In addition, there are irreducible metrical facts about the temporal distances between such slices. These metrical facts don't supervene on intrinsic facts about instants, including for example their number or ordinal arrangement. So the existence of periods of time with some sort of intrinsic, even if frame-relative, measure must be ontologically on a par with the existence of the instantaneous slices.
This metaphysical fact seems to support a phenomenological claim that Tensers could put forward: we have a genuinely diachronic experience of time. Our experience of time is not limited to our experience of the present but includes the experience of an ordered sequence of moments spread out in time. (We think the same metaphysical fact puts some pressure on a theory of persistence favored by Anti-Tensers, namely, Ramsey-Lewis-Sider Worms (24.3T.3). But we ignore this point in this context.)
Tensers could argue that our experience of time is diachronic in two senses. First, each moment is experienced as a specious present that includes some movement and change. Second, we experience time as a sequence of such specious presents. The idea of the specious present seems to be pretty well accepted by those who reflect on the phenomenology of experience, going back at least to William James's Principles of Psychology (1890). Let's focus here on the second, more contentious claim that our experience of time is as of an inexorably and really ordered succession of specious presents.
Here are two possible points of tension between that observation and Anti-Tensism:
the inexorability of the temporal succession
the appearance of real change in one's intrinsic mental state
First, the inexorability of succession. The succession moments, with earlier moments succeeded by later moments, appears real, categorical, and inexorable, as opposed to notional, hypothetical, or optional. The standard Anti-Tenser response appeals to some theory about the intrinsic directionality of time, either in terms of causal order (causes precede their effects) or in terms of increasing entropy or information. The intrinsic directionality of time is important in accounting for the fact that we remember the past and not the future and the fact that we deliberate about the future and not the past. There is, however, a very important gap between intrinsic direction and real succession. For example, the words in a well-written, meaningful book define a real direction within the text (standardly in English, this is left-to-right, top-down). This direction is intrinsic to the text as meaningful, and not merely the product of extrinsic conventions. However, this fact about the text in no way determines a real, as opposed to a notional, succession of earlier words by later ones. It is perfectly possible to read the words in such a book in reverse order or in any other order one might choose. In the same way, even if the moments of time are intrinsically directed, this by itself is not sufficient to explain the inexorable experience of real succession.
In The Time Machine, H.G. Wells toys with the idea that a well-trained mind could choose to read the moments of time in a non-standard order or with a non-standard velocity, shifting experience dramatically into the past or the future. This seems impossible, but it is unclear how Anti-Tensers could explain the impossibility.
What Tensers must claim is that we experience a unique, sui generis sort of succession in the succession of time. All other “order” on this view involves a kind of metaphorical or indirect reference to temporal succession. Without the real succession of times in the A series, we would have no concept of earlier or later, of prior or posterior. Our concept of ordinal numbers (first, second, third, etc.) are also dependent on the reality of temporal passage, since without real temporal passage, there is no real succession or direction. Mathematicians can model the ordinal numbers and the relation of succession in the intrinsically static world of set theory, with 0 modeled as the empty set, and the successor n as the union of the members of n with the set {n}. However, we can understand this construction as a model of the ordinal numbers only because we have a prior understanding of order as such, an understanding anchored in the real direction of temporal passage.
Merely identifying some static but asymmetric relation between world-states is not enough, not even if we label this relation as one of ‘causation’, ‘memory’, or what-have-you. We can always take the members of the domain of an asymmetric relation in any order: we can move up or down the set of integers, and we can always extend such a set to include members “earlier” than its original “origin”. In contrast, there could be no such thing as an ordinal number “earlier” than first (the use of ‘zeroth’ is a kind of joke), and there is something essentially backward about counting from third to second to first. Only Tensism provides a metaphysical basis for these distinctions.
Could Anti-Tensers take a Kantian line (as in Kant's Analogies of Experience in the The Critique of Pure Reason) and argue that it is impossible for a rational person to do otherwise than order his memories in accordance with irreversible causal laws, such as increasing entropy? Here we need to distinguish between long-term and short-term memory. The Kantian line seems plausible as an account of our ordering of long-term memories, which do seem something like a stock of snapshots stored in some box in the mind. Presently, however, we are talking about very short-term memory, of our experience of the passage of time as an aspect of our present, lived experience. Here it seems the Kantian line would simply be wrong. We can easily imagine experiencing counter-causal and counter-entropic sequences in short-term memory in which broken eggs become whole, circular waves converge on a point from which a pebble jumps up, and so on. Again, there seems to be a clear gap between claims about the intrinsic directionality of time and our experience of the inexorability of temporal passage.
Second, the experience of real change in one's mental states. In Logic Matters, Peter Geach appeals to the experience of real change to support some form of Tensism:
But now, quite similarly, even if my distinction between past, present and future aspects of physical things is a fragmentary mis-perception of changeless realities, it remains true that I have various and un-combinable illusions as to what realities are present. I must therefore have these illusions not simultaneously but one after another; and then there is after all real time and real change. (Geach 1972: 305)
According to Geach, our experience of time presents the self as numerically one, as an enduring entity, with mutually incompatible intrinsic states. This isn't merely notional change, as characterizes McTaggart's iron rod that “changes” from hot on one end and cold on the other. We don't experience ourselves, when we go from seeing something red to not, as having a part that sees red and a part that doesn't. Still less do we think of ourselves as a thing that sees red with a counterpart that doesn't. It is one and same thing that sees (first) red and (then) not red.
Some Anti-Tensers seem simply to deny that our experience presents reality in such a metaphysically laden way. They might plausibly claim that it is merely acculturation in a Tenser worldview that has colored our phenomenology in this way. We claim no expertise in the anthropology of time experience, but our impression is that this Tenser coloring is universal in human experience. (Some linguists in the mid-twentieth century claimed the contrary, but most later commentators seem to have dismissed these claims.)
As we said above, these phenomenological considerations are not decisive, but they must be given considerable weight. If we don't treat the appearances as innocent until proven guilty, we will quickly be driven into global skepticism. Further, the same observations may create problems for Tensism no less than for Anti-Tensism. This question is one that one must consider in order to evaluate the relative merits of Tensism and Anti-Tensism. In any case, one thing that Anti-Tensers cannot do is admit that real succession occurs in our experience, but not in the extra-mental world. If there are A-facts anywhere, whether in the mind or without, then Anti-Tensism is refuted. A hybrid view according to which there are A-facts with respect to the mind, but only B-facts with respect to the physical world, gives rise to a bizarre kind of dualism.
20.5.4 The present as a uniquely actual boundary between past and future
If we accept Aristotelian Finitism (Temporal Finitism 19.3T plus Temporal Anti-Discretism 19.4A), we have an additional argument for Tensism. Aristotelian Finitists argue that instants of time exist only as boundaries, dependent on the existence of temporally extended processes. Any process is indefinitely divisible in time, but any finite interval in the actual world can contain only finitely many actual boundaries. We could say that time is indefinitely divisible (there is no lowest bound to the temporal extent of any process), but it is not infinitely divisible: only a finite number of actual boundaries can coexist. An infinite division is metaphysically impossible.
If we accept such a view of time, and we think of the present as an actual boundary between past and future parts of occurring processes (which is how Aristotle himself suggested that we think of it), then we can construct the following argument for Tensism and for a strongly Tensist view of reality.
The present moment is a fully actual (not merely potential) boundary of all processes. (Premise)
If Aristotelian Tensism is false, then every moment contained by any process is a boundary of that process that is as actual as the present moment. (Premise: definition of Aristotelian Tensism)
So, if Aristotelian Tensism is false, then every moment contained by any process is an actual boundary of that process. (From 1 and 2)
Time is dense: any process contains infinitely many moments. (Premise: Anti-Discretism)
Therefore, if Aristotelian Tensism is false, then any process is divided by a temporally dense set of actual boundaries. (From 3 and 4)
Boundaries of potential parts are themselves potential: consequently, actual boundaries are boundaries of actual parts. (Premise: metaphysical dependency of boundaries on extended things)
Therefore, if Aristotelian Tensism is false, then every process contains temporally extended parts that are actual. That is, there are no temporally extended but actually undivided processes. (From 5 and 6)
There exists a fundamental level of temporal things, and this consists entirely of temporally extended processes. (Premise: Intervalism, 19.1A)
Fundamental intervals have only potential parts. (Premise: Temporal Finitism, 19.3T)
Therefore, Aristotelian Tensism is true. (From 7, 8, and 9)
This argument provides an argument for Aristotelian Tensism, for the view that the present moment is uniquely actual. It seems to be at least consistent with Simple Tensism and with Falling Branches and Growing Block Tensism. However, there is some tension between Aristotle's argument and Presentism. Aristotle's position entails that processes exist as fundamental, actual entities, even though their duration stretches into the past, and possibly even into the future. Presentists would have to argue that only that instantaneous part of the process that occurs now can really exist. How, then, can that instantaneous boundary be dependent on the temporally extended process of which it is a boundary? How can the existent be dependent on what is mostly non-existent?
We might consider a modification of standard Presentism, which we could call ‘Aristotelian Presentism�
��. Aristotelian Presentists hold that the only things that exist are fundamentally processes that are going on at present, along with those entities that are participants in those processes. However, any process going on at present exists as a whole, including those parts that stretch into the past. This would be something like Growing Block Tensism, except that the past is thought to exist only insofar as it involves parts of presently occurring processes. More remote parts of the past no longer exist in any sense.
20.5.5 The threat of fatalism
We defined untensed propositions as those that possess their truth-value permanently. There is an important qualification to add. We may need to treat untensed propositions about contingent events in the future as a special case. Some Tensers hold that such future contingent propositions have no truth-value. Aristotle suggested such a position in his Posterior Analytics, using the example of a sea battle that may or may not take place tomorrow. If tomorrow is 30 September 2016, then the untensed proposition that such a sea battle takes place on 30 September would have no truth-value today (i.e., on 29 September 2016). The idea that future contingent untensed propositions have no truth-value, while all untensed propositions about the present or past are determinately true or false, may be behind the attractiveness of Growing Block Tensism.