Is Time Real: The Block Universe and the Vanishing Now

At Confessions XI.14.17 the Latin runs, with one sentence of the original omitted here and marked as omitted: "quid autem familiarius et notius in loquendo commemoramus quam tempus? ... quid est ergo tempus? si nemo ex me quaerat, scio; si quaerenti explicare velim, nescio." Nothing is spoken of more familiarly than time; so what is time? If nobody asks, he knows; if he has to explain it to somebody who asks, he does not. The sentence left out says that we understand time perfectly well both when we speak of it and when we hear somebody else speak of it, which sharpens the puzzle rather than softening it. A few lines earlier in the same chapter he had already asked it in different words: "quid est enim tempus? quis hoc facile breviterque explicaverit?" The English above is a paraphrase and carries no quotation marks, deliberately. The Latin was fetched from an online archive of the text and read for this article, and it is quoted. Our own research file renders the first of those sentences in an English that matches no translation this article could identify, and that English is not reproduced here.

J. M. E. McTaggart, writing in 1908, distinguished two ways of ordering events in time. In the first, which he called the A-series, events are ordered as past, present or future. Those positions change. Yesterday was future, then it was present, and it is now past. The A-series is the ordering that builds passage into time itself.

In the second, the B-series, events are ordered by the permanent relations earlier than and later than. Those relations never change. The battle is earlier than the treaty, and it stays earlier than the treaty from whichever moment the ordering is read. The B-series is the ordering that describes time without passage.

Our file records that the argument has generated over a century of debate, and that most philosophers reject McTaggart's conclusion while disagreeing about which premise to deny. That clause is what organises the rest of this article, because it makes the live positions two ways of refusing the same argument. Deny the first premise and the result is the B-theory and the block universe. Deny the second and the A-series survives, which gives presentism or the growing block. One caution travels with it. The quantifier is our own file's and is unmeasured: most philosophers is what the file says, and it is not a percentage anywhere in that file and is not converted into one anywhere in this article.

Einstein's special relativity, published in 1905, and Minkowski's spacetime formulation of it, delivered as an address at Cologne on 21 September 1908 and published in 1909, remove absolute simultaneity. Whether two events happen at the same time is not a fact about the events alone. It depends on the state of motion of the observer asking, and two events simultaneous for one observer occur at different times for another. That is the whole of the physics this article needs, and it is not in dispute. It is stated here as the premise of an argument and nothing else.

The inference our file calls the Rietdijk-Putnam argument, after papers of 1966 and 1967, runs like this. If simultaneity is relative, then events in one observer's future are already present for some other observer. If that other observer's present is real, those events are real. Therefore all events are equally real, which is eternalism and the block universe. This is the step on which everything dramatic about the subject depends, because it is where a result in physics is made to bear on a question in metaphysics. It is an argument. It is not a finding of physics, it is not presented as one here, and our own file supplies its rebuttal in the very next lines.

Relativity does not compel eternalism, and our own file says so under a heading of its own. It names Stein, in the Journal of Philosophy in 1968, and Maudlin, in 2002. The point is precise rather than sweeping: relativity shows that now cannot be defined as a global hyperplane cutting across the whole universe, while local temporal relations and the causal structure of spacetime still preserve a meaningful distinction between past and future. Stein's reply appeared in the same journal as Putnam's argument, one year later, and it appears nowhere in our own file's bibliography. It is cited below, because an article that presents the argument without the reply is presenting one side of a live dispute. Our file gives Maudlin's year and no title, and Maudlin published more than one relevant item that year, so the paper cited below is this article's identification of what the file meant and not the file's own.

Presentism: only the present exists. Past events have happened and no longer exist; future events have not happened and do not yet exist. Our file records one strength and three challenges. The strength is that it aligns with common-sense temporal experience, which is not a small thing, because that experience is what the whole subject is trying to account for. The challenges are special relativity, which leaves no absolute present to privilege; truthmaker theory, which asks what makes past-tensed statements true if the past does not exist; and cross-temporal relations. The defenders our file names are Prior, in Past, Present, and Future of 1967, Zimmerman in 2008, and Bourne in 2006. Of those three, only Bourne's book resolved to a verified identifier for this article, and it is cited below.

Eternalism, the block universe: past, present and future events all exist equally, the universe is a four-dimensional block of spacetime, and temporal passage is a subjective illusion. Our file records one support and one challenge. The support is that this is the natural interpretation of Minkowski spacetime. The challenge is the manifest image, which is to say the pervasive human experience of flow and becoming. That challenge is doing real work and this article does not walk past it, because it is why the dispute is not closed by physics: an argument from relativity can remove the universal now without touching the experience of one. The defenders our file names are Sider, in Four-Dimensionalism of 2001, Mellor, in Real Time II of 1998, and Quine, who is named with no work and no date.

The growing block: past and present events exist and the future does not, so the block of reality grows as time passes. Our file attributes it to Broad, in 1923, and Tooley, in 1997, and supplies its challenge in a single parenthesis: why does the present edge feel special? The growing block is the position that tries to keep both the reality of passage and the reality of the past, which is what makes it the third real option here rather than a compromise between the other two.

Rietdijk's half of that hyphenated attribution is missing from our own file's bibliography, and its title is the story. The earlier of the two papers, in Philosophy of Science in 1966, is called A Rigorous Proof of Determinism Derived from the Special Theory of Relativity. Whatever is thought of the proof, the framing in the title is explicit. The argument was put in print as a derivation of determinism from relativity, not as a description of a block.

Entropy increases toward the future, and our file calls the second law the physical basis for the asymmetry between past and future. The philosophically sharp part is the sentence underneath it. The fundamental laws of physics are mostly time-symmetric, meaning that they work equally well run forward or backward. The word mostly is our file's own and it is kept here, because dropping it would state something stronger than the file does. The asymmetry everybody lives inside is not written into the laws.

The asymmetry is traced instead to boundary conditions: the universe began in a state of extraordinarily low entropy, a framing our file attributes to Albert in 2000 and Carroll in 2010. And then the file says, in its own voice, that why the Big Bang had low entropy remains unexplained, naming inflationary cosmology and anthropic reasoning as proposals rather than answers. That is the honest gap in the whole account, and it is worth stating plainly rather than at the end of a paragraph. The explanation of why time seems to run one way bottoms out in an initial condition nobody has explained.

The psychological arrow is the last link and the softest. Memories are of the past and not of the future, and time feels as though it flows from one toward the other. Our file says this is hypothesized to be a consequence of the thermodynamic arrow, on the ground that memory formation requires entropy increase. Hypothesized is the file's own word and it is kept, because an article that upgraded it to because would have invented a result. This is also where the physics hands the question back, since the experience the psychological arrow describes is exactly what presentism appeals to and exactly what eternalism has to explain away.

Henri Bergson, in Time and Free Will of 1889 and Creative Evolution of 1907, drew a line that neither the A-theory nor the B-theory draws. On one side, scientific time: spatialized, measured by clocks, represented as a line of discrete instants. On the other, duree, lived temporal experience, which he held to be continuous, indivisible and irreducible to spatial representation. This is a third position rather than a variant of the other two, because it denies that either ordering captures what time is as it is lived.

Bergson argued that the intellect spatializes time for practical purposes and thereby distorts its true nature, and that only intuition grasps real duration. Our file's note on his reception has three parts and all three belong here. He was influential on Proust, on Deleuze and on process philosophy. He was largely dismissed by analytic philosophers. And he is experiencing a revival. Dropping the dismissal would make him a settled authority he is not. Dropping the revival would make him sound finished when he is not.

On 6 April 1922, at a meeting of the Societe francaise de philosophie in Paris, the two registers met in one room. Einstein argued that Bergson's philosopher's time has no physical reality. Bergson argued that Einstein's physics presupposes lived temporal experience. Our file gives only the year, and adds that the debate echoes today in consciousness studies. The date, the venue and the standing book-length treatment of the episode, Canales's The Physicist and the Philosopher of 2015, were resolved for this article and the book is cited below. What Canales concludes about the exchange is not reported here, because the record was read and the book was not.

One paragraph, because the linear frame this article has been arguing inside is not the only frame human beings have used. Our file names Hindu and Buddhist cosmology, with its kalpas and yugas and samsara and its vast cycles of creation, destruction and recreation; the Maya long count; the Aztec Five Suns; and eternal return, in Nietzsche and in Stoic ekpyrosis. Against those it sets the Judeo-Christian-Islamic frame, in which time is a unique narrative with a beginning and an end: creation, history, eschaton. Our file's own conclusion from the comparison is the part worth keeping. These diverse frameworks challenge the assumption that any single philosophical account of time is universal, including the account this article has been conducting entirely inside one tradition. Cyclical time is a subject in its own right, with its own research file in this library, and it is not developed here.

The formalism at the centre of this is real and datable. DeWitt's Quantum Theory of Gravity. I. The Canonical Theory appeared in Physical Review in 1967, and it is the source of what is now called the Wheeler-DeWitt equation. Our file names the equation and cites nothing for it; the paper is cited below. What is carried at Tier 2 here is that the 1967 formalism exists and says what it says. What it means for the reality of time is the next block, and that is Tier 3.

In that equation the wave function of the universe satisfies an equation with no time variable in it at all, which is what our file calls the problem of time in quantum gravity. Its Tier 3 claim is that time may not be fundamental but emergent from a more basic timeless structure. That is a live research problem rather than a discovery that time is unreal, and a resolved identifier under the formalism does not launder the claim above it into a verified one.

Loop quantum gravity and some string-theory approaches suggest that time emerges from more fundamental relational structures. Our file states this at exactly that level of generality and names no specific result, so no specific result is named here either, and the hedge belongs in the sentence rather than in a note underneath it.

Julian Barbour, in The End of Time of 1999, proposed that the universe is a collection of Nows, that is, configurations with no intrinsic temporal ordering, and that time is an illusion generated by the correlations between those configurations. It is a striking proposal and it is the natural place for an article like this one to stop. It does not stand unchallenged, which is why the next block sits directly underneath it rather than at the end of the section.

The British Journal for the Philosophy of Science published a study of Barbour's book by Butterfield in 2002, running from page 289 to page 330 of volume 53. That a leading journal in the field devoted a review essay of that length to the proposal is a checkable fact, and it shows two things at once: that the proposal is taken seriously, and that it is contested. What Butterfield concluded is not reported here. The record was resolved for this article and the essay itself was not read, and the entry below is labelled as a review rather than as Barbour's book.

Goedel published an exact solution of Einstein's field equations in Reviews of Modern Physics in 1949, under the title An Example of a New Type of Cosmological Solutions of Einstein's Field Equations of Gravitation. That solution contains closed timelike curves. Our file's Tier 4 section names the Goedel universe and cites nothing for it, and this is the paper. What is carried at Tier 2 here is the existence and the content of the 1949 solution, and nothing beyond it.

Time travel to the past has not been physically realized, and popular claims that it has lack any scientific basis. Our file's wording is careful and this article keeps it. General relativity permits closed timelike curves in certain exotic spacetimes, the Goedel universe, the interior of a Kerr black hole and the Tipler cylinder among them. There is no evidence that any of those configurations exists in the actual universe. And the chronology protection conjecture, proposed by Hawking in 1992, holds that quantum effects prevent their formation. Two distinctions have to survive the compression. What the mathematics permits is not what the universe contains. And a conjecture is a conjecture, which is why this article does not say that physics has ruled time travel out.

The field was polled on exactly this. The 2020 PhilPapers Survey, published by Bourget and Chalmers in 2023, put the question to philosophers, and the wording of two of its questions was confirmed for this article from the survey organiser's own published discussion document. The options offered are the three positions section 03 describes and the two orderings section 01 describes, which is itself a checkable fact about how the discipline frames its own disagreement. The results are a different matter. No percentage from that survey appears anywhere in this article, because the research pass behind it could not reach the survey's own results: four independent routes to them refused, and a figure taken from a search summary is corroboration rather than verification. What can be said is that the field was asked, and that these were the options.