20.03.03 · philosophy / phil-of-physics

Space, time, and relativity: substantivalism vs. relationalism, the direction of time

stub3 tiersLean: nonepending prereqs

Anchor (Master): Earman, J. — World Enough and Space-Time (1989)

Intuition Beginner

Is space a thing — a container in which objects exist — or just a system of relations between objects? Isaac Newton argued for absolute space: a real, invisible substance in which all physical objects are situated. Gottfried Leibniz disagreed. Space, he said, is nothing more than the order of relations between bodies. There is no container, no backdrop, no stage. Just objects, standing in spatial relations to one another.

Newton thought he could prove absolute space exists. Hang a bucket of water from a rope. Twist the rope and release it. The bucket spins. At first the water stays flat and still. Then, dragged by friction, the water begins to spin too — and its surface curves into a bowl shape. What is the water rotating relative to? Not the bucket: the water spins with it. Newton's answer: absolute space.

Albert Einstein transformed the debate. Special relativity (1905) showed space and time are not separate. They weave together into spacetime — a four-dimensional block in which past, present, and future all coexist. General relativity (1915) went further. Spacetime itself curves and bends. Matter tells spacetime how to curve; spacetime tells matter how to move. The stage is no longer static. It is dynamic, shaped by everything in it.

The arrow of time — why time flows one way — remains puzzling. The fundamental laws of physics are time-symmetric: run them backward and they still hold. Yet eggs break but never unbreak. Ice melts but never re-freezes on its own. We remember the past, not the future. The explanation points to entropy, a measure of disorder. Entropy increases. But why did it start so low? That is the deep question.

Visual Beginner

Picture three rival conceptions of space lined up against one scene — a rotating bucket. Panel 1 (Newton / substantivalism): the bucket sits inside an invisible grid representing absolute space; the water's curvature is indexed to that grid. Panel 2 (Leibniz / relationalism): the grid is gone; only bucket, water, and rope remain, with arrows marking the relations between them; the curve is unexplained unless further bodies are introduced. Panel 3 (Einstein / general relativity): the grid becomes a curved, flexible mesh that deforms where matter sits; the mesh itself is a physical field.

Each panel answers the same question — what is the water rotating relative to? — with a different ontology of what space is.

Worked example Beginner

A bucket of water hangs from a rope. The rope is twisted and released. The bucket begins to spin. For the first few seconds the water stays flat: it has not yet been dragged into motion. Then friction couples the water to the bucket, and the water spins too. Its surface curves — rising at the edges, dipping at the centre. This concave shape is the signature of rotation.

Under Newton's substantivalism, the curving water is rotating relative to absolute space. The bowl shape is caused by the water's motion through the invisible container. If absolute space did not exist, the curve would have nothing to be relative to — yet the curve is real and measurable. Therefore absolute space is real.

Under Leibniz's relationalism, this argument fails. Rotation is motion relative to other bodies. The water curves relative to the bucket, the rope, the Earth, and the distant stars. No invisible container is needed; the relations among material bodies do all the work. The challenge for Leibniz is to explain why the curve appears when the water spins relative to the distant stars but not when it is at rest.

Under Mach's relational inertia (a later refinement), the water's rotation is measured against the fixed stars — the totality of distant matter. If those distant masses were absent, there would be no rotation and no curve. Inertia itself is determined by the distribution of matter in the universe. Einstein was deeply influenced by this idea, though general relativity does not fully realise it.

The three readings of one experiment frame the entire debate that follows.

Check your understanding Beginner

Formal definition Intermediate+

Substantivalism is the thesis that spacetime points, or the spacetime manifold itself, possess identity independent of any material or field content. In its classical Newtonian form, absolute space is a substance whose parts endure through time and relative to which absolute motion is defined. In its modern "manifold substantivalism" form, the bare differentiable manifold $M$ — prior to the assignment of any metric or matter fields — is the bearer of point identity: each point $p \in M$ is a haecceitic individual that would be the very point it is regardless of which fields are placed on it.

Relationalism is the negation: there are no spacetime points with identity independent of material content. Space (or spacetime) is constituted entirely by the network of relations among physical bodies, or among events. What exists, on this view, is the structure of relations, not a container in which those relations are embedded.

The hole argument (Earman-Norton 1987) ^{[Earman Norton 1987]} exploits the general covariance of general relativity. Let $(M, g)$ be a model of the Einstein field equations, where $M$ is a four-dimensional manifold and $g$ is a Lorentzian metric satisfying $G_{μν} = 8 π T_{μν}$ . Let $d : M \to M$ be a diffeomorphism that is the identity outside a region $H \subset M$ (the "hole") but nontrivial inside. Then $(M, d^{\astg})$ is also a model: it satisfies the same field equations and agrees with $(M, g)$ everywhere outside the hole. General covariance — the requirement that the laws take the same form under arbitrary smooth coordinate transformations — guarantees this.

The argument runs:

Manifold substantivalism grants each point $p \in M$ an identity independent of $g$ .
The models $(M, g)$ and $(M, d^{\astg})$ differ at the manifold points inside the hole: the metric value assigned to $p$ in the first model is assigned to $d^{- 1} (p)$ in the second.
Yet the initial data on any Cauchy surface outside the hole does not determine which assignment obtains.
So if substantivalism is true, general relativity is indeterministic: the same initial data admits multiple distinct futures.

Leibniz equivalence is the thesis that diffeomorphism-related models represent the same physical situation. Accepting it dissolves the indeterminism but requires giving up the haecceity of manifold points: the "shifted" model does not describe a different world, because manifold points lack identity apart from the metric structure defined on them.

Sophisticated substantivalism (Butterfield, Hoefer, Pooley, Maudlin) is the modern compromise: spacetime points exist, but their identity is fixed by the metrical and causal structure, not by bare manifold position. The metric field $g$ is physically real and does not require a pre-existing manifold of haecceitic points to be "placed on."

The block universe (eternalism). In special relativity, a spacetime event is a point $(t, x, y, z)$ . The set of all events forms a four-dimensional manifold with a Minkowski metric $η = diag (- 1, + 1, + 1, + 1)$ . Simultaneity is relative: two events $e_{1} = (t_{1}, x_{1})$ and $e_{2} = (t_{2}, x_{2})$ are simultaneous in one inertial frame and sequential in another. Eternalism is the thesis that all events — past, present, and future — are equally real; the "passage" of time is a feature of experience, not of reality.

Entropy and the direction of time. Let $Ω (E)$ be the volume of the region of phase space compatible with macrostate energy $E$ . The Boltzmann entropy is $S = k ln Ω (E)$ . The second law of thermodynamics states that the entropy of an isolated system tends to increase. The past hypothesis (Albert 2000) ^{[Albert 2000]} posits that the universe began in a macrostate of very low entropy; combined with standard statistical mechanics, this yields the observed thermodynamic arrow.

Counterexamples to common slips

"Relativity proves relationalism." It does not. General relativity treats the metric field $g$ as a dynamical, physical entity — which many read as a vindication of substantivalism (the metric is a real field, not a mere bookkeeping device). The debate survives relativity; it is transformed, not settled.
"The block universe means time is an illusion." Eternalism denies that the present is metaphysically privileged; it does not deny that temporal structure (before/after relations, causal order) is real. Time exists in the block as the ordering dimension; what is denied is objective "becoming."
"Entropy increase is a law of nature on par with Newton's laws." The second law is statistical: entropy decrease is not impossible, merely overwhelmingly improbable given a low-entropy past. The asymmetry is a boundary-condition fact, not a dynamical law.

Key argument — the hole argument against manifold substantivalism Intermediate+

The hole argument is the most influential technical argument in the philosophy of spacetime since 1987. John Earman and John Norton reconstructed a line of reasoning that Einstein himself went through in 1913–1915, when he struggled to formulate generally covariant field equations — and then abandoned, before rediscovering covariance in the final form of general relativity. The argument turns general covariance into a weapon against a particular reading of what spacetime points are.

Setup. General relativity is generally covariant: its field equations hold in every coordinate chart, related by arbitrary smooth transformations. Equivalently, if $(M, g, T)$ is a model and $d : M \to M$ is any diffeomorphism, then $(M, d^{\astg}, d^{\astT})$ is also a model. The drag-along $d^{\astg}$ assigns to each point $p$ the metric value that $g$ assigned to $d^{- 1} (p)$ .

The hole. Choose a region $H \subset M$ — the "hole" — in which there is no matter ( $T = 0$ inside $H$ ), and a diffeomorphism $d$ that is the identity on $M ∖ H$ but nontrivial inside $H$ . Then $(M, g)$ and $(M, d^{\astg})$ agree perfectly on all matter and metric values outside $H$ . They differ only in which manifold point carries which metric value inside the hole.

The three-premise argument.

(H1) Manifold substantivalism. Each point $p \in M$ has haecceity — identity independent of the fields defined on it. The question "which metric value does $p$ carry?" has a determinate, physically meaningful answer.
(H2) General covariance. Diffeomorphism-related models are each genuine solutions of the Einstein equations; the theory does not prefer one over the other.
(H3) Determinism. A satisfactory physical theory should determine the future from the initial data; a theory that admits distinct futures from identical initial data is indeterministic in an unacceptable way.

The fork. The models $(M, g)$ and $(M, d^{\astg})$ share identical initial data on any Cauchy surface outside $H$ . Under (H1), they are distinct (the metric at $p$ differs). Under (H2), both are licit. So the initial data outside the hole does not fix which future obtains. The theory is indeterministic — contradicting (H3).

To restore determinism, at least one premise must go. Earman and Norton argue that (H1) is the culprit.

Responses.

Leibniz equivalence (the relationalist response): deny (H1). Diffeomorphism-related models are the same physical situation. The "difference" between $(M, g)$ and $(M, d^{\astg})$ is a mathematical artifact of treating manifold points as individuals. On this reading, general covariance is a gauge symmetry — the diffeomorphisms are redundancies in the description, not transformations to a distinct state of affairs.
Sophisticated substantivalism (Butterfield, Hoefer, Pooley, Maudlin): accept that spacetime points exist, but deny that their identity is fixed by bare manifold position. Points are individuated by the metric structure. Since $d^{\astg}$ defines the same metrical structure (just labelled differently), the two models describe the same world. This preserves realism about the metric field while blocking the haecceity that drives the indeterminism.
Reject determinism (rare): accept that general relativity is genuinely indeterministic at the manifold-point level, but argue that this indeterminism is benign because it is not observable. Few have found this response attractive.

The modern consensus leans toward Leibniz equivalence or sophisticated substantivalism. Earman's World Enough and Space-Time (1989) ^{[Earman 1989]} remains the canonical framework; Pooley's "The Reality of Spacetime" (2013) and Maudlin's Philosophy of Physics: Space and Time (2012) survey the contemporary state of the debate.

Exercises Intermediate+

Exercise 1 (easy, classification).

A philosopher holds that spacetime points exist but that their identity is fixed entirely by the metrical structure defined upon them, so that diffeomorphism-related models describe the same physical world. Which position is this?

A. Naive (manifold) substantivalism

B. Leibniz equivalence / strict relationalism

C. Sophisticated substantivalism

D. Presentism

Hint

The view is realist about the metric field but denies bare haecceity to manifold points.

Answer

Option C. Sophisticated substantivalism keeps realism about the spacetime metric (the field is physical) while denying that manifold points carry identity independent of that metric. Diffeomorphism-related models are identified, blocking the hole argument's indeterminism. Butterfield, Hoefer, Pooley, and Maudlin defend variants of this view.

Exercise 2 (medium, argument reconstruction).

Reconstruct the hole argument as: (H1) manifold substantivalism, (H2) general covariance, (H3) determinism; therefore a contradiction. State which premise Leibniz-equivalence relationalism and sophisticated substantivalism each reject, and how.

Hint

Both responses restore determinism, but by different routes. One denies points altogether; the other keeps them but changes how they are individuated.

Answer

Leibniz-equivalence relationalism rejects (H1). It treats diffeomorphism-related models as the same physical situation, reading general covariance as a gauge symmetry. The "two models" inside the hole are one world described twice.
Sophisticated substantivalism also rejects (H1), but only its haecceity clause. It accepts that the metric field is a real physical entity; it denies that manifold points have identity apart from the metric. The two models are identified because they define the same metrical structure.

Both preserve (H2) and (H3). The disagreement is over what is left of substantivalism once haecceity is abandoned — a live metaphysical question.

Exercise 3 (medium, defence).

In four or five sentences, defend the past hypothesis (Albert) against the objection that positing a low-entropy initial condition is just replacing one mystery with another.

Hint

Consider what the past hypothesis buys: a single boundary condition that, combined with statistical mechanics, recovers the entire thermodynamic arrow. Compare this to the alternative of building the asymmetry into the laws themselves.

Answer

The past hypothesis does not explain why the universe began in a low-entropy state; it identifies the minimal additional posit that, combined with time-symmetric microscopic laws, recovers every thermodynamic asymmetry we observe.

Without it, we face a choice: either build a time-asymmetric arrow directly into the fundamental dynamics (which no known law exhibits), or accept that the arrow is a contingent boundary condition. Albert and Loewer argue the latter is the more conservative move — one boundary condition does a great deal of work, grounding not only entropy increase but also recording, memory, and intervention.

The residual mystery — why the boundary condition has the value it does — may point to cosmology (the Big Bang) rather than to a defect in the statistical-mechanical framework.

Exercise 4 (medium, critique).

In four or five sentences, critique the "growing block" theory of time (Broad, Tooley) — the view that the past and present exist but the future does not, with reality "growing" as new events come into being.

Hint

Growing-block theorists must say when the growing happens, and relative to whom. Special relativity makes "when" frame-dependent.

Summary

The growing block must specify a preferred foliation of spacetime — a fact about which hypersurface is "the present edge of being." Special relativity denies any frame-independent such surface: what counts as "the present moment" depends on the observer's state of motion.

If the growing block posits a hidden preferred frame, it reintroduces a structure that relativity was constructed to eliminate (absolute simultaneity by another name). If it does not, it cannot say where the edge of being is.

Critics (Smart, Putnam, Rietdijk) conclude that objective becoming is incompatible with the relativistic picture; the block universe is the cleaner option.

Exercise 5 (hard, comparison).

Malament (1977) proved that, given the causal structure of Minkowski spacetime, the relation of simultaneity relative to an inertial observer is uniquely determined (up to a conventional choice of one parameter). Explain why this result is taken to undermine the Reichenbach-Grünbaum thesis that simultaneity is conventional, and state one objection a conventionalist might still raise.

Hint

Reichenbach's conventionality thesis hinges on the claim that the one-way speed of light depends on how we synchronise distant clocks, and that this synchronisation is a free choice. Malament's theorem speaks to what the causal structure alone determines.

Answer

Reichenbach and Grünbaum argued that simultaneity between distant events depends on a conventional choice of synchronisation procedure for distant clocks; since the one-way speed of light cannot be measured without first synchronising clocks, the procedure is underdetermined and therefore conventional.

Malament showed that if we require the simultaneity relation to be definable from the causal (light-cone) structure of Minkowski spacetime, to be an equivalence relation, and to be preserved by the causal automorphisms, then standard Einstein synchronisation is the unique relation satisfying these constraints. This removes the underdetermination: there is no "conventional" freedom left once natural constraints are imposed.

A residual conventionalist objection (Sarkar-Stachel, Norton) is that Malament's constraints — especially the requirement that the relation be definable from causal structure alone — are themselves nontrivial and contestable. If one does not accept that simultaneity must be causally definable, the conventionality thesis survives in a weaker form.

Substantivalism, relationalism, and general relativity Master

Earman's World Enough and Space-Time (1989) ^{[Earman 1989]} set the modern framework for the substantivalism-relationalism debate by translating it into the language of differential geometry and the hole argument. The pre-relativistic debate between Newton and Leibniz (and its continuation in the Leibniz-Clarke correspondence) turned on whether absolute space was detectable and whether God would have had reason to place the material world in one location in absolute space rather than another. Earman sharpened these into a precise question: do the points of the spacetime manifold have identity independent of the physical fields defined upon them? If yes, the hole argument generates an unacceptable indeterminism; if no, substantivalism in its naive form collapses.

The dominant contemporary response is sophisticated substantivalism. Butterfield argued that one can be a realist about the metric field without being committed to bare manifold points; the field is physical, but its "points" are individuated structurally. Hoefer developed the view that spacetime is identified with its metrical structure rather than with a bare container. Pooley, in his 2013 review "The Reality of Spacetime," surveys the options and defends a point-free or structuralist reading: what exists is the metrical-causal structure, not a set of haecceitic points bearing it. Maudlin, in The Metaphysics within Physics (2007) ^{[Maudlin 2007]}, takes a stronger line: the metric field is a physical object among the objects of physics, and no separate metaphysical inquiry is needed to license its reality. The passage of time, on Maudlin's view, is likewise real and objective — a claim that puts him at odds with the block-universe consensus.

Malament proposed neoometric relationalism: the metric, not the manifold, defines all physically meaningful structure, so the substantivalism-relationalism dispute dissolves once we identify spacetime with its metric rather than its points. The debate between sophisticated substantivalism and metric-based relationalism is now largely one of metaphysical taste: both agree on the physics and on the failure of naive manifold substantivalism, but they disagree about whether the metric field counts as a "substance" in the philosopher's sense.

Baker's "Against Field Interpretations of Electromagnetism" presses a related ontological question: is the electromagnetic field a thing in its own right, or a bookkeeping device for the states of charged particles? The answer bears directly on spacetime, since general relativity treats $g$ as a field on the same footing as matter fields. If field ontology is accepted for electromagnetism, parity of reasoning supports realism about the metric — and sophisticated substantivalism becomes the natural reading. The question connects forward to the philosophy of quantum field theory 20.03.02 pending and quantum gravity 28.04.04 pending.

The block universe, simultaneity, and the passage of time Master

Minkowski's 1908 formulation recast special relativity as a geometry of a four-dimensional manifold equipped with a Lorentzian metric. World-lines of particles trace curves through this manifold; light cones partition each event into absolute past, absolute future, and an elsewhere region whose temporal ordering is frame-dependent. Simultaneity is relative: two spatially separated events that one inertial observer judges simultaneous are judged sequential by another in relative motion. The Rietdijk-Putnam argument (Rietdijk 1966, Putnam 1967) ^{[Putnam 1967]} leverages this to argue for the block universe: for any two events $e_{1}, e_{2}$ that are spacelike-separated, there exists an inertial frame in which they are simultaneous; if what exists-now is real, then by transitivity of "is real relative to," events in the distant future of one frame are real for observers in another. The conclusion — all events are equally real — is eternalism, the block-universe view.

Presentism (only the present exists) and the growing block (past and present exist; future does not) both struggle with special relativity, because they require a preferred foliation of spacetime into "now"-surfaces. No such preferred foliation is available in the theory. Defenders of presentism (Crisp, Zimmerman, Prior) have explored whether a privileged frame can be added as a hidden structure without empirical consequence; most philosophers of physics find the cost too high. The moving spotlight view (Broad, recently revived by Cameron and Deasy) accepts the block but adds an ontologically privileged present that "moves" — a view Maudlin endorses in a qualified form but that most block-universe theorists reject as adding nothing explanatory.

The conventionality of simultaneity thesis (Reichenbach, Grünbaum) held that the choice of how to synchronise distant clocks is a free convention, since the one-way speed of light cannot be measured without prior synchronisation. Reichenbach introduced a parameter $ε \in (0, 1)$ governing the synchronisation: $ε = 1/2$ corresponds to Einstein's standard convention. Malament's 1977 theorem ^{[Malament 1977]} showed that if simultaneity is required to be an equivalence relation definable from the causal structure and invariant under causal automorphisms, then $ε = 1/2$ is uniquely selected — undermining the conventionality claim. Sarkar-Stachel and Norton have objected that Malament's constraints are not neutral, so a weaker conventionalism survives. The debate remains technically active, though the majority view follows Malament.

Price's Time's Arrow and Archimedes' Point (1996) ^{[Price 1996]} argues that we must reason about time from an atemporal standpoint — outside the flow — to avoid imposing our temporal perspective on the physics. From this vantage, the block universe is the natural ontology, and the "passage" of time is a feature of embedded observers, not of the world. Dorato and Hoerl have examined whether the experienced flow can be accommodated within the block; Phillips, Deng, and others debate whether the "seeming" of passage is itself a phenomenon the block theory must explain. The tension between the block ontology and the phenomenology of becoming remains a central open problem.

The direction of time and the past hypothesis Master

The puzzle of time's arrow is that the fundamental dynamical laws are time-reversal invariant: Newton's equations, the Schrödinger equation, and classical electromagnetism all admit solutions that run equally well in either temporal direction. Yet macroscopic phenomena are overwhelmingly time-asymmetric: entropy increases, radiation propagates outward but not inward, we remember the past but not the future. How can a time-asymmetric world be built from time-symmetric laws?

Boltzmann's statistical mechanics offered the first answer. Entropy $S = k ln Ω$ measures the number of microstates compatible with a given macrostate. High-entropy macrostates vastly outnumber low-entropy ones, so a system evolving under time-symmetric dynamics will, with overwhelming probability, move from rare low-entropy states toward typical high-entropy ones. Boltzmann's H-theorem appeared to derive the monotonic increase of entropy, but Loschmidt's reversibility objection (1876) showed that for every evolving state there is a time-reversed state whose entropy decreases — so the H-theorem cannot prove a genuine asymmetry without an additional assumption. Boltzmann's response invoked the assumption of molecular chaos (the Stosszahlansatz), which is itself time-asymmetric. Zermelo's recurrence objection (1896) pressed the Poincaré recurrence theorem: an isolated system will eventually return near its initial state, so entropy cannot increase forever. Boltzmann replied that recurrence times are cosmically vast; the objection is practically irrelevant but philosophically pointed.

The modern resolution, developed by Albert (Time and Chance, 2000) ^{[Albert 2000]} and elaborated by Loewer, is the past hypothesis: the universe began in a macrostate of extremely low entropy. Combined with the time-symmetric microscopic dynamics and the standard statistical postulate, the past hypothesis entails that entropy increases away from the initial state in both temporal directions — but we, as entropy-increasing subsystems, experience only one direction as "the future." On Albert's reading, the past hypothesis functions as a law or law-like posit; Loewer's Mentaculus (named after the probability catalog in the film A Serious Man) proposes a complete physical description: the microscopic dynamics, the macro-state space, the past hypothesis, and the statistical postulate together generate all macroscopic regularities, including those underwriting records, memory, and counterfactuals.

The past hypothesis has attracted sharp criticism. Earman's "The 'Past Hypothesis': Not Even False" (2006) ^{[Earman 2006]} argues that the posit is too ill-defined to function as a law: what counts as "the macrostate of the early universe," and how is its entropy to be measured without a pre-existing statistical-mechanical framework? Frisch and Winsberg press related concerns about whether a single boundary condition can bear the explanatory weight Albert and Loewer assign it. Page and others have explored whether the low-entropy condition is anthropically selected. Callender, in What Makes Time Special? (2017) ^{[Callender 2017]}, argues that the arrow is grounded in the macro-physics of our universe rather than in a fundamental temporal asymmetry, and that over-metaphysical readings of the past hypothesis should be resisted. Wallace's "The Logic of the Past Hypothesis" (2010 onward) offers a reformulation in terms of typicality rather than probability.

Penrose's Weyl curvature hypothesis proposes that the gravitational entropy of the universe is controlled by the Weyl tensor, and that the Big Bang was a state of vanishing Weyl curvature — a gravitational low-entropy condition that would explain the thermodynamic arrow at the cosmological scale. Sklar's Physics and Chance (1993) ^{[Sklar 1993]} remains the canonical philosophical survey of the foundations of statistical mechanics. Goldstein, Lebowitz, Mastrodonato, and Tumulka have developed Boltzmannian quantum statistical mechanics using typicality measures on the wave function. Horwich's Asymmetries in Time (1987) traces the arrow to causal and explanatory asymmetries rather than to entropy alone.

Time travel, quantum gravity, and the emergence of spacetime Master

Time travel and closed timelike curves. Gödel (1949) ^{[Gödel 1949]} discovered an exact solution of the Einstein field equations describing a rotating universe in which closed timelike curves (CTCs) exist: an observer can travel through spacetime and return to their own past. The Kerr solution for rotating black holes, the Tipler cylinder, and traversable wormholes (Morris-Yurtsever, Thorne) provide further CTC structures within general relativity. The central philosophical problem is the grandfather paradox: if time travel is possible, what prevents inconsistent histories? Lewis's "The Paradoxes of Time Travel" (1976) ^{[Lewis 1976]} distinguishes personal time (the time measured by the traveller's own clock and memory) from external time (the time of the world) and argues that time travel is logically consistent provided the traveller's actions are part of the very history they "return" to — the future that the traveller experiences is already fixed. The Novikov self-consistency principle proposes that only histories consistent with the laws of physics are realised; the initial conditions of the universe conspire to avoid paradox. Deutsch has proposed an Everettian resolution: the time traveller enters a different branch, so the "grandfather" killed is not the one in the traveller's origin branch.

Earman, Smeenk, and Wüthrich have examined whether CTCs can be ruled out on physical or metaphysical grounds. Hawking's chronology protection conjecture proposes that quantum effects always prevent the formation of CTCs, keeping the past safe for historians. The status of the conjecture remains open. Hawthorne's "Before-Effect and Zeno Causation" explores backward causation without CTCs. The topic connects to the general theory of causation 20.08.03 pending.

Quantum gravity and the emergence of spacetime. The deepest current challenge to the substantivalism-relationalism debate comes from quantum gravity. Rovelli, in The Order of Time (2018) and the technical paper "Neither Almanac nor Music," argues that there is no fundamental time in quantum gravity: the Wheeler-DeWitt equation has no time parameter, and temporal structure emerges only in approximate, thermodynamic, semi-classical limits. Barbour's The End of Time (1999) develops a timeless framework in which reality is a configuration space of "nows" related by "best matching," with time an emergent feature of the matching relation. Causal set theory (Sorkin, Bombelli, Rideout) posits that spacetime is fundamentally a discrete, partially ordered set of events; the order relation is primitive, and continuous spacetime emerges as a coarse-grained approximation. This view has been applied to entropy and the arrow of time.

Loop quantum gravity (Ashtekar, Rovelli, Smolin) describes spacetime as a network of spin states (spin networks, spin foams) at the Planck scale, with area and volume quantised. String theory posits an 11-dimensional spacetime with extended objects (branes) and has produced the AdS/CFT correspondence (Maldacena 1997), a concrete instance of the holographic principle: a gravitational theory in a bulk anti-de Sitter space is exactly equivalent to a non-gravitational conformal field theory on its boundary. Van Raamsdonk (2010) argued that the connectivity of bulk spacetime is encoded in the entanglement structure of the boundary theory — spacetime emerges from entanglement. Quantum graphity (Wen) and related tensor-network models (HaPPY code) explore similar ideas. The philosophical implication, as Butterfield, Crowther, Huggett, and Vassallo have argued, is that spacetime may not be fundamental at all: it is an emergent, effective structure recovered from underlying non-spatiotemporal degrees of freedom. This reframes the substantivalism-relationalism debate: if spacetime is emergent, the question "is spacetime a substance?" may have no answer at the fundamental level.

The cosmological constant problem (Weinberg 1987) asks why the vacuum energy density is some 120 orders of magnitude smaller than naive quantum-field-theoretic estimates predict. Anthropic and landscape-based responses (Susskind) connect to the fine-tuning debates in philosophy of science. See 28.04.04 pending for dark matter and dark energy.

Discrete and digital physics. Fredkin, Wolfram (cellular automata), 't Hooft (deterministic quantum mechanics), and Lloyd (Programming the Universe) have explored whether the universe is fundamentally computational. Sorkin's causal sets provide a specific discrete model. The philosophical question — is spacetime digital or continuous at the Planck scale? — remains open and connects to the philosophy of computation.

Connections Master

Measurement problem 20.03.01 is the declared prerequisite. The substantivalism-relationalism debate runs parallel to the realism-instrumentalism debate about the wave function: both ask what in the mathematical formalism corresponds to something real. The metric field $g$ and the wave function $ψ$ raise structurally similar questions about field ontology.
Philosophy of quantum mechanics 20.03.02 pending connects through quantum gravity and emergent spacetime. If spacetime emerges from entanglement (Van Raamsdonk, AdS/CFT), the interpretation of the quantum state becomes inseparable from the interpretation of spacetime. Relational QM (Rovelli) and the substantivalism debate share a structural affinity: both ask whether the relevant structure is "in the points/objects" or "in the relations."
Consciousness 20.06.01 (pending) connects through the phenomenology of temporal passage. The experienced "now" and the felt flow of time are data that any theory of time must accommodate; the hard problem of consciousness 20.06.01 and the problem of temporal becoming share a structural shape.
Causation 20.08.03 pending (pending) connects through time travel, backward causation, and the causal arrow. Lewis's counterfactual analysis and the grandfather paradox both turn on the logic of causal order.
Thermodynamics and statistical mechanics [11.02.*] (pending) is the physics-side anchor for the direction-of-time material. The second law, Boltzmann entropy, and the past hypothesis are jointly physical and philosophical.
Quantum gravity / dark matter and energy 28.04.04 pending (pending) is referenced in the master section on emergent spacetime and the cosmological constant problem.

Cross-domain to epistemology 20.01.01: the conventionality-of-simultaneity debate invokes underdetermination and conventional choice — core epistemological notions. Cross-domain to phil-of-mind 20.06.01: the specious present, the felt flow, and the A-theory/B-theory dispute import phil-of-mind commitments into phil-of-physics.

Historical and philosophical context Master

The substantivalism-relationalism debate predates Newton. Aristotle's analysis of place (topos) as the innermost boundary of the containing body was relational in spirit; the atomists' void was substantivalist in everything but name. The modern debate crystallised in the seventeenth century. Newton's Principia (1687) defined absolute space as a real, homogeneous, infinite substance existing independent of matter, and absolute time as a quantity flowing "equably without relation to anything external." The bucket argument (Scholium to the Definitions) was offered as empirical evidence: the concave water surface is a dynamical effect of rotation relative to absolute space, not relative to the bucket. Leibniz's reply, in the Leibniz-Clarke correspondence (1715–1716), pressed that absolute space and time would violate the principle of sufficient reason — God would have had no reason to place the universe in one region of absolute space rather than another. Space and time, Leibniz argued, are relational orders: space is the order of coexistences, time the order of successions.

Mach's The Science of Mechanics (1883) reopened the debate with a relational account of inertia: the water in Newton's bucket rotates relative to the fixed stars, not to absolute space. Mach's principle — that local inertial frames are determined by the distribution of distant matter — deeply influenced Einstein, though general relativity only partially implements it. The status of Mach's principle remains contested.

Einstein's path to general relativity (1907–1915) ran directly through the hole argument. In 1913–1914, Einstein and Grossmann published the Entwurf theory, which was not generally covariant; Einstein believed he had proved that generally covariant field equations were physically unacceptable, via an argument essentially identical to the modern hole argument. He concluded that manifold points must lack independent identity — a relationalist conclusion he embraced at the time. In 1915 he reversed himself, rediscovered general covariance, and published the final field equations. The episode was largely forgotten until Stachel (1980) and then Earman-Norton (1987) reconstructed it as a live philosophical argument against manifold substantivalism.

The block-universe view emerged from Minkowski's 1908 formulation of special relativity. Rietdijk (1966) and Putnam (1967) argued that special relativity is incompatible with presentism; the argument was refined by Maxwell and others. The growing-block view (Broad 1923, Tooley 1997) and the moving-spotlight view (Cameron 2015) are the principal non-eternalist responses. The A-theory/B-theory distinction (McTaggart 1908) frames the metaphysical background: A-theorists hold that tense (past, present, future) is objective; B-theorists hold that only the before/after ordering is real.

The direction-of-time literature was transformed by Boltzmann's statistical mechanics and the reversibility/recurrence objections of Loschmidt and Zermelo. Reichenbach's The Direction of Time (1956) offered a branch-system analysis. The modern revival came with Albert's Time and Chance (2000) and Price's Time's Arrow and Archimedes' Point (1996), which between them set the terms of the contemporary debate. The past hypothesis, the Mentaculus, and the Earman-Callender critiques define the current frontier. The 21st century has seen quantum statistical mechanics (Goldstein, Lebowitz, Tumulka, Wallace) and quantum-gravity approaches (Rovelli, Sorkin) reshape the question of whether time is fundamental at all.

The contemporary literature is centred in Studies in History and Philosophy of Modern Physics, British Journal for the Philosophy of Science, and Foundations of Physics, with the Philsci-Archive at Pittsburgh hosting an open-access preprint collection. Maudlin's Philosophy of Physics: Space and Time (2012) ^{[Maudlin 2012]} is the standard intermediate-level entry point; Earman's World Enough and Space-Time (1989) ^{[Earman 1989]} remains the master-level framework text.

Bibliography Master

Newton, I. — Philosophiae Naturalis Principia Mathematica (London, 1687); Scholium on absolute space and time.
Leibniz, G. W. & Clarke, S. — The Leibniz-Clarke Correspondence, ed. H. G. Alexander (Manchester University Press, 1956).
Mach, E. — Die Mechanik in ihrer Entwicklung (1883); trans. The Science of Mechanics (Open Court, 1960).
Einstein, A. — "Die Feldgleichungen der Gravitation", Sitzungsber. K. Preuss. Akad. Wiss. 844–847 (1915).
Minkowski, H. — "Raum und Zeit", Phys. Z. 10, 104–111 (1909).
Gödel, K. — "A remark about the relationship between relativity theory and idealistic philosophy", in Schilpp (ed.), Albert Einstein: Philosopher-Scientist, 557–562 (1949).
McTaggart, J. M. E. — "The unreality of time", Mind 17, 457–474 (1908).
Reichenbach, H. — The Direction of Time (University of California Press, 1956).
Broad, C. D. — Scientific Thought (Harcourt Brace, 1923), Ch. 2 on becoming.
Lewis, D. — "The paradoxes of time travel", Am. Phil. Q. 13, 145–152 (1976).
Putnam, H. — "Time and physical geometry", J. Philos. 64, 240–247 (1967).
Malament, D. — "Causal theories of time and the conventionality of simultaneity", Noûs 11, 293–300 (1977).
Earman, J. & Norton, J. — "What price spacetime substantivalism? The hole story", Br. J. Philos. Sci. 38, 515–525 (1987).
Earman, J. — World Enough and Space-Time: Absolute versus Relational Theories of Space and Time (MIT Press, 1989).
Stachel, J. — "Einstein's search for general covariance, 1912–1915", in Howard & Stachel (eds.), Einstein and the History of General Relativity (Birkhäuser, 1989).
Sklar, L. — Physics and Chance: Philosophical Issues in the Foundations of Statistical Mechanics (Cambridge University Press, 1993).
Price, H. — Time's Arrow and Archimedes' Point: New Directions for the Physics of Time (Oxford University Press, 1996).
Horwich, P. — Asymmetries in Time: Problems in the Philosophy of Science (MIT Press, 1987).
Albert, D. Z. — Time and Chance (Harvard University Press, 2000).
Maudlin, T. — The Metaphysics within Physics (Oxford University Press, 2007).
Maudlin, T. — Philosophy of Physics: Space and Time (Princeton University Press, 2012).
Pooley, O. — "The reality of spacetime", Synthese 190, 3719–3759 (2013).
Hoefer, C. — "For fundamentality, but against 3D monism", in The Ontology of Spacetime II (Elsevier, 2008).
Butterfield, J. — "The hole argument", in The Oxford Handbook of Philosophy of Physics (2013).
Norton, J. — "The hole argument", Stanford Encyclopedia of Philosophy (2019).
Barbour, J. — The End of Time: The Next Revolution in Physics (Oxford University Press, 1999).
Rovelli, C. — The Order of Time (Riverhead Books, 2018).
Sorkin, R. D. — "Causal sets: Discrete gravity", in Lectures on Quantum Gravity, ed. Gomberoff & Marolf (Springer, 2005).
Bombelli, L., Lee, J., Meyer, D. & Sorkin, R. — "Spacetime as a causal set", Phys. Rev. Lett. 59, 521–524 (1987).
Maldacena, J. — "The large N limit of superconformal field theories and supergravity", Adv. Theor. Math. Phys. 2, 231–252 (1998).
Van Raamsdonk, M. — "Building up spacetime with quantum entanglement", Gen. Rel. Grav. 42, 2323–2329 (2010).
Callender, C. — What Makes Time Special? (Oxford University Press, 2017).
Earman, J. — "The 'past hypothesis': Not even false", Stud. Hist. Philos. Mod. Phys. 37, 399–430 (2006).
Loewer, B. — "The Mentaculus vision", in Time, Chance and Reduction, ed. Ernst & Hüttemann (Cambridge University Press, 2010).
Penrose, R. — "Singularities and time-asymmetry", in General Relativity: An Einstein Centenary Survey, ed. Hawking & Israel (Cambridge University Press, 1979).
Wallace, D. — "The logic of the past hypothesis", manuscript (2010 onward).
Goldstein, S., Lebowitz, J. L., Mastrodonato, C. & Tumulka, R. — "On the distribution of the wave function for systems in thermal equilibrium", J. Stat. Phys. 141, 341–352 (2010).
Hawking, S. W. — "Chronology protection conjecture", Phys. Rev. D 46, 603–611 (1992).
Thorne, K. S. — Black Holes and Time Warps: Einstein's Outrageous Legacy (W. W. Norton, 1994).
Reichenbach, H. — The Philosophy of Space and Time (Dover, 1958; orig. 1928).
Grünbaum, A. — Philosophical Problems of Space and Time, 2nd ed. (Reidel, 1973).
Sarkar, S. & Stachel, J. — "Did Malament prove the conventionality of simultaneity?", Philos. Sci. 66, 208–220 (1999).
Dorato, M. — "The irrelevance of the presentist/eternalist debate for the ontology of Minkowski spacetime", Stud. Hist. Philos. Mod. Phys. 37, 645–660 (2006).
Baker, D. J. — "Against field interpretations of electromagnetism", Stud. Hist. Philos. Mod. Phys. 40, 215–228 (2009).
Butterfield, J. & Crowther, K. — "Decoupling emergence and reduction in physics", in Why More Is Different (World Scientific, 2019).
Novikov, I. — "Time machine and self-consistent evolution in problems with self-interaction", Phys. Rev. D 45, 1989–1994 (1992).
Ashtekar, A. — "Loop quantum gravity: four recent results and some frequently asked questions", in Geometrical and Topological Methods for Quantum Field Theory (2011).
Weinberg, S. — "The cosmological constant problem", Rev. Mod. Phys. 61, 1–23 (1989).
Poincaré, H. — La Science et l'Hypothèse (1902); Ch. on geometry and convention.
Smart, J. J. C. — "The temporal asymmetry of the world", Australas. J. Philos. 70, 385–391 (1992).