ris3n's Apologetics Codex

Argument

Argument from the Impossibility of an Actual Infinite Past

Intro

There are ads on our codex that pay for hosting and keep the codex free. If you can, please consider whitelisting ris3n.com or allowing scripts to support the work.

Sponsored

"The universe might have just always existed. Why does there have to be a beginning?"

This argument says the past cannot stretch back forever. Not just that it didn't, but that it couldn't. If that is right, then the universe had a start, and a start needs a cause.

Try imagining a hotel with an infinite number of rooms, all of them already full. A new guest walks in. The clerk just moves Room 1's guest to Room 2, Room 2's to Room 3, and so on, and now Room 1 is open. The hotel was full, and yet it has room. In fact, it can take an infinite number of new guests just by moving everyone to even-numbered rooms. This is David Hilbert's famous puzzle. The math works in pure set theory, but when you try to picture an actual hotel doing this, it falls apart. A real fully-booked hotel cannot get more rooms by shuffling.

Now apply the same idea to time. If the past is truly infinite, then to get to today, the universe had to finish counting through an endless number of moments first. But you can never finish counting through endless moments by adding them one at a time. It is like trying to walk to the end of a road that has no end. You never arrive. Yet here we are at today. So somewhere back there, the counting started.

The argument runs in two prongs that each work on their own. The first says an infinite collection of real things produces contradictions, like the hotel. The second says even if such a collection could exist, you could not build one by adding moments one after another, which is how time actually moves.

If the past has a beginning, the universe began. If the universe began, something outside the universe started it. That cause must be outside time, outside space, outside matter, immensely powerful, and personal, because only a personal will can choose to begin something. This argument is the engine that drives the second premise of the Kalam Cosmological Argument.

The quick reply: "If the past is endless, the universe had to finish an endless count to reach today. You can't finish endless. So the past had a start."

In full

This argument is the philosophical engine driving the second premise of the Kalam Cosmological Argument, but it deserves its own page because the case for the finitude of the past is independently weighty and frequently engaged in live debate as a stand-alone argument. The case combines (a) the conceptual impossibility of a concrete actual infinite (Hilbert's Hotel, Tristram Shandy, Grim Reaper paradoxes) with (b) the inability to form an actual infinite by successive addition (Craig's "successive-addition" argument). If the past is finite, the universe began. If the universe began, it has a cause, and the cause must be timeless, immaterial, spaceless, enormously powerful, and personal. This page is structured as debate prep, each premise carries a second-order positive case, anticipated objections, rebuttals, a live-cite kit, and tactical notes.

Argument structure

# Premise
P1 An actual infinite cannot exist in concrete reality.
P2 An infinite temporal regress of past events would be a concrete actual infinite.
P3 Even if a concrete actual infinite were possible, it could not be formed by successive addition.
P4 The temporal series of past events has been formed by successive addition.
C Therefore, an infinite temporal regress of past events cannot exist, the past is finite, the universe began, and (joining the Kalam's P1) has a cause that is timeless, immaterial, spaceless, enormously powerful, and personal.

Form

Two-pronged deductive argument. The first prong (P1+P2) closes off the possibility of an actual-infinite past on conceptual grounds. The second prong (P3+P4) closes off the formation of an actual-infinite past on temporal grounds, even granting (for the sake of argument) that an actual infinite could in principle exist, it could not be reached by adding one moment at a time, which is how temporal sequences accumulate. Either prong alone suffices for the conclusion; together they constitute a fortiori case. Soundness is contemporary: the conceptual-impossibility prong is the more contested (Cantorian set-theorists demur); the successive-addition prong has narrower disputed surface and is the workhorse premise in live debate.


P1, An actual infinite cannot exist in concrete reality

Affirmative case (second-order arguments)

  1. Hilbert's Hotel paradox. David Hilbert's 1924 lecture "Über das Unendliche" introduced the thought experiment: a hotel with infinitely many rooms, all occupied. A new guest arrives. The manager moves Room 1's occupant to Room 2, Room 2's to Room 3, and so on, Room 1 is now free for the newcomer. The hotel was full and can accommodate more guests and the move involves no impossible action. Worse: infinitely many new guests can be accommodated by moving each existing guest from Room n to Room 2n, opening all odd-numbered rooms. The hotel is full, yet has infinitely many vacancies. Infinity plus infinity equals infinity; infinity minus infinity is undefined. The paradoxes are mathematically consistent in Cantorian arithmetic but violate basic concrete-reality intuitions about subtraction and combination. Hilbert himself: "The infinite is nowhere to be found in reality. It neither exists in nature nor provides a legitimate basis for rational thought... The role that remains for the infinite to play is solely that of an idea."

  2. Tristram Shandy paradox. Bertrand Russell's variant (deployed by Craig against Russell): Tristram Shandy writes his autobiography, taking one year to record each day of his life. If he has been writing for infinitely many years, has he ever finished? Russell argued yes (each day is eventually recorded). Craig argues no (at any point Tristram is infinitely many days behind his living, and the gap never closes). The paradox sharpens to a contradiction: on an infinite past, Tristram should have already finished an infinite time ago, yet he is still writing. The contradiction shows that concrete actual-infinite sequences generate genuine paradox, not merely surprise.

  3. Grim Reaper paradox (Pruss-Koons). Imagine infinitely many Grim Reapers, each scheduled to swing his scythe at Fred at a different time during the next hour, with their schedule densely ordered such that for every time t, there is some Reaper scheduled before t. When does Fred die? Each Reaper would kill Fred if Fred is still alive at his time, but for any Reaper, there's an earlier one who would have killed Fred first. Fred can't survive (someone kills him), yet no Reaper kills him (each is preempted by an earlier one). Contradiction. The paradox can be set up over an infinite past with the Reapers in the past instead of the future, generating the same contradiction. No actual-infinite collection of causally-relevant events can exist without producing this kind of contradiction. (Alex Pruss, Infinity, Causation, and Paradox, 2018; Robert Koons, in Faith and Philosophy 2014.)

  4. Library of Tristram Shandy / book paradoxes. An infinite library with infinitely many red and infinitely many black books. Remove all the black books, you have a library with infinitely many fewer books, yet still infinitely many books, and the same cardinality you started with. Remove all the books numbered 4, 5, 6,... and you've removed infinitely many books and have only three left. Subtraction-of-infinities is not well-defined, but in concrete reality subtraction is well-defined: take three books off a shelf and you have three fewer books. Concrete reality and actual-infinite arithmetic give incompatible answers. The reality side wins.

Anticipated objections

  1. "Cantor's set theory shows actual infinites are mathematically coherent, you can't ban them from reality just because they violate finite intuitions."
  2. "Hilbert's Hotel is counterintuitive, not contradictory. ZFC handles infinite cardinalities consistently."
  3. "The paradoxes assume infinite collections can be operated on (added to, subtracted from); maybe a static infinite past doesn't permit such operations."
  4. "You're privileging finite-intuition over mathematical reality, that's a pre-Cantorian prejudice."

Rebuttals

  1. Cantor's set theory accepts actual infinites as abstracta, not as concrete physical or temporal entities. Cantor himself distinguished the actual infinite in concreto (rejected by Cantor, "we have nothing in mathematics or in nature whose existence is unconditional") from the actual infinite in abstracto (his transfinite cardinals). The Kalam argument concerns the concrete case: actually-elapsed past events, actually-existing physical objects in extended infinite array. Cantor's mathematics does not underwrite the concrete claim; Cantor himself denied it. Failure mode: conflating mathematical and physical instantiation.

  2. Counterintuitive vs contradictory is the right distinction, but the paradoxes deliver contradiction, not mere strangeness. Tristram-Shandy on infinite past should-have-finished-yet-still-writing is a contradiction in the situation, not just a surprise. The Grim Reaper situation is straight-up contradictory: someone kills Fred and nobody kills Fred, in the same setup. ZFC's consistency for the cardinal arithmetic does not extend to the consistency of concrete situations described by the cardinal arithmetic; the latter generates genuine contradiction. Failure mode: conflating formal-system consistency with situation-consistency.

  3. The "static-infinite-past doesn't permit operations" move concedes the substance of the argument. If past events form a real concrete collection (which they must, on any realist view of time), then operations on that collection are meaningful even if no one performs them. The Grim Reaper paradox runs on whether a given setup can obtain, not on whether someone performs an operation. If infinitely many past events could exist as a collection, the contradiction follows from the structure of the setup, regardless of who is or isn't operating. Failure mode: special pleading, proposing exceptions to the concrete-infinite structure precisely where the argument bites.

  4. Hilbert himself was pre-Cantorian only in the sense of agreeing with Cantor about the concrete vs abstract distinction. Hilbert was the leading mathematician of his generation and the foremost defender of Cantor's program, yet he denied concrete actual infinities precisely because the paradoxes show that the concrete application is incoherent. The "finite intuition prejudice" charge is exactly backwards: it is Cantorian and Hilbertian to deny concrete actual infinity. The opponent who appeals to Cantor against this premise has misread Cantor. Failure mode: misappropriation of authority.

Live-cite kit

  • Scripture: Genesis 1.1 ("In the beginning"); Psalm 90:2 ("from everlasting to everlasting You are God", the finite past suits a beginning-having creation alongside an unbegun God); Hebrews 11.3 ("the worlds were prepared by the word of God")
  • Scholarly: David Hilbert, "Über das Unendliche" (Mathematische Annalen 95, 1926); William Lane Craig & James Sinclair, "The Kalam Cosmological Argument" in The Blackwell Companion to Natural Theology (Craig & Moreland, eds., 2009); Craig, The Kalām Cosmological Argument (1979); Andrew Loke, God and Ultimate Origins (2017); Alex Pruss, Infinity, Causation, and Paradox (2018)
  • Aphorism: "Hilbert's Hotel is the universe's polite way of saying it cannot accommodate an infinite past."

Tactical notes

  • Lead with Hilbert's Hotel in front of any audience, it is the most accessible paradox and has the highest rhetorical voltage per word of explanation.
  • Save Grim Reaper for technically-engaged opponents (atheist philosophers, analytic-trained skeptics). It is harder to explain but harder to escape.
  • Force-commit move: "Do you accept that infinity minus infinity is undefined? If you do, explain how subtraction can be undefined in concrete reality, when I subtract two events from a real sequence, I get a definite answer."
  • What NOT to defend live: the consistency of ZFC, the axiom of choice, or any technical set-theoretic detail. The argument runs on the conceptual-impossibility result; the formal apparatus is for backup, not for live deployment.

P2, An infinite temporal regress of past events would be a concrete actual infinite

Affirmative case (second-order arguments)

  1. Past events are concrete, not abstract. A past event, the assassination of Caesar, the eruption of Vesuvius, your last meal, was a real concrete occurrence in the spatiotemporal world. The set of past events is therefore a concrete collection, not a mathematical abstraction. On any reasonable realist view of time (A-theory, B-theory, or growing-block), past events are objectively real and form a real series.

  2. An infinitely-many-membered series is an actual infinite if completed. If the series of past events is beginningless, extends indefinitely backward, then the totality of past events is an actually-existing infinite collection, not a merely-potential one. The collection is complete now (it includes all events up to the present); there are no further past events to add. A complete-now infinite collection is, by definition, an actual infinite.

  3. The A/B-theory distinction does not save the day. On B-theory (eternalism), all events exist tenselessly, past events exist now in the timeless sense, and they form a literal infinite collection. On A-theory (presentism), past events do not exist as present, but the past series is still characterized by the same infinitely-many-past-events count if the past is infinite. Either way, the cardinality claim is the same: the past, if infinite, is an actual infinite of concrete events.

Anticipated objections

  1. "Past events don't exist now (on presentism), so there is no concrete actual infinite now, the series is merely abstract."
  2. "The past is a potential infinite, always extendable backward in description, but never an actual completed totality."
  3. "Time itself may be discrete (Planck time); the question of infinite past doesn't have the structure you assume."

Rebuttals

  1. Presentism does not save the opponent. Even on presentism, the number of past events that have occurred is real, the question "How many days have elapsed up to today?" has a definite answer. If the answer is "infinitely many," the count is an actual infinite, regardless of whether the events still exist. The Grim Reaper and Tristram Shandy paradoxes run on the count and ordering, not on whether the events persist into the present. Failure mode: conflating ontological existence with cardinality of occurrence.

  2. Past events are completed, not potential. That is the relevant asymmetry with future time. A potential infinite is a finite collection that can always be extended, like the future, which is always growing. The past is closed: it has happened; it cannot be added to (we can only add new past events as time progresses, but the past-at-time-t is fixed and complete at t). Therefore an infinite past is an actual infinite (completed) by definition. This is the central point: future is potential infinite; past is actual infinite if beginningless. Failure mode: conflating completed and uncompleted directions of time.

  3. Discrete-time (Planck-scale) does not dissolve the question, it sharpens it. If time is discrete with a fundamental tick (Planck time ~5.4 × 10^-44 sec), then an infinite past would consist of infinitely many discrete ticks, a literal Cantorian actual infinite. The Hilbert and Grim Reaper paradoxes run more cleanly on discrete events than on continuous time. The discrete-time hypothesis, if anything, strengthens the argument. Failure mode: mistaking continuity for the source of the paradox (it is the actual-infinite-cardinality, not the continuity, that bites).

Live-cite kit

  • Scripture: Psalm 90:2 ("from everlasting to everlasting You are God"); Hebrews 11.3 (creation by divine word, implicit beginning)
  • Scholarly: William Lane Craig & James Sinclair (Blackwell Companion, 2009, ch. 3, esp. §3.21 on the A/B-theory variations); G.J. Whitrow, The Natural Philosophy of Time (1961, 2nd ed. 1980, pre-Craig classical statement); Craig, Time and Eternity (2001, ch. 4)
  • Aphorism: "The future is potentially infinite. The past, if endless, is actually infinite, and that's the problem."

Tactical notes

  • Force-commit on the asymmetry of time: "Do you agree that the past is completed and the future is open? If yes, the past-if-beginningless is an actual infinite, by your own concession."
  • Don't get drawn into the A-theory vs B-theory dispute live. Acknowledge that the argument holds on either; refuse to defend a position on the philosophy of time as part of the argument.
  • Anticipate the "infinite is just a description" deflection. Reply: "If 'infinitely many days have elapsed' is just a description, what does it describe? Either a real count (actual infinite) or nothing, and if nothing, you've conceded the past is finite."

P3, Even if a concrete actual infinite were possible, it could not be formed by successive addition

Affirmative case (second-order arguments)

  1. Counting to infinity is impossible by addition of one at a time. Start counting: 1, 2, 3,... You will never reach infinity. There is no number n such that n + 1 completes the count. The natural numbers are potentially infinite under iteration (you can always add one), but no finite number of additions yields an actual-infinite total. This is uncontested mathematically.

  2. The series of past events is formed by successive addition. Each moment, one new event is added to the totality-of-past-events. The past at time t+1 is the past at time t plus the events at t. The accumulation is strictly additive: events one-at-a-time become past.

  3. No additive sequence ever crosses the gap from finite to infinite. From any finite starting point, successive addition yields only finite values. Therefore, the temporal series cannot have been formed by successive addition to an actual-infinite total. This is independent of the abstract question of whether actual infinites exist; it's a claim about whether they can be reached by addition.

  4. Al-Ghazali's medieval form. Tahafut al-Falasifa (11th c.): the "traversal" argument. You cannot traverse an infinite by successive steps, in either direction. If the past is infinite, then we would have had to traverse infinitely many past moments to get to the present, but no traversal of an infinite is possible. Therefore the past is finite. The argument predates and underwrites Craig's modern formulation.

Anticipated objections

  1. "Each individual past moment is finitely far from now, there's no 'infinite traversal' required; the traversal is finite at every step."
  2. "The infinite past is given, not formed. You don't need to construct an infinite series by successive addition if the series eternally was."
  3. "Time may not have a unique direction or starting point in the way the argument assumes (block universe, eternal recurrence)."

Rebuttals

  1. The objection trades on equivocation. Each individual past moment is finitely far, true. But the totality of past moments is not a single moment; it is a collection whose cardinality is the disputed claim. If every past moment is finitely far, then for each moment m in the past, the count from m to now is finite. But the total number of moments in the entire past series is the cardinality of the set, not the distance of any single moment. The objection conflates point-finitude with collection-infinitude. Failure mode: collection / member equivocation (composition-fallacy variant).

  2. The "given, not formed" move is question-begging. If the past was given (eternally was), then there is no event of its starting, and no event of any particular moment becoming past after another. But every observed past moment did become past after another. The temporal series is demonstrably additive at each transition; the proposal that it is "given" rather than "formed" requires re-describing the data, not engaging it. Furthermore, if the past was given as an infinite totality, then we never reached any particular present moment by traversal, we are at moment t, but how did we get to t from an infinite-many-moments-before-t situation? The "given" move erases the very phenomenon of temporal becoming. Failure mode: erasing the explanandum.

  3. Eternal recurrence and block universes face the same problem. On eternal recurrence (Nietzsche), the same events recur in an infinite cycle, the cardinality of past instances is still infinite, and the formation-by-recurrence requires infinitely many cycles to have been completed, which faces the same traversal problem. On block universe (B-theory), the temporal series is a tenseless 4-dimensional structure, but for any observer inside the block, the past is still characterized by the same additive structure, the observer's "now" point still has all the past-moments-from-her-perspective in the additive arrangement. Failure mode: non-engagement, these are alternative descriptions, not solutions.

Live-cite kit

  • Scripture: Romans 4.17 ("God who gives life to the dead and calls into being that which does not exist"); Hebrews 1:10-12 (creation as having a beginning, with the Creator unchanging)
  • Scholarly: al-Ghazali, Tahafut al-Falasifa (11th c.); Craig, The Kalām Cosmological Argument (1979, ch. 4); Craig & Sinclair (Blackwell Companion, 2009); J.P. Moreland, Scaling the Secular City (1987, ch. 1)
  • Aphorism: "You cannot count to infinity by counting one at a time. Time counts one at a time. So time has not counted to infinity."

Tactical notes

  • This is often the cleaner premise in live debate, it bypasses the "but Cantor!" deflection by granting (for argument's sake) that actual infinites exist abstractly. The successive-addition claim is harder to dispute.
  • Force-commit move: "At what finite past moment did the universe transition from a finite past to an actual-infinite past? If never, the past was never finite; but if always actual-infinite, how was it formed? Pick one."
  • Watch for the "given" deflection, it is the standard escape route. Pin it down by asking what work "given" does that "formed" does not, and whether the opponent can defend a non-additive temporal series given that we observe additive transitions every instant.

P4, The temporal series of past events has been formed by successive addition

Affirmative case (second-order arguments)

  1. Observation: time accumulates one moment at a time. Every second that passes adds one second to the past. This is not theory, it is the structure of temporal experience. The accumulation is additive; the series is formed by addition.

  2. Any physical-cosmological model that has a Hubble parameter > 0 has a finite past (BGV). The Borde-Guth-Vilenkin theorem (2003) shows that any universe in average cosmic expansion has a finite past, the past series did not extend infinitely. This is the empirical-cosmological corroboration of P4: even on naturalist physics, the past is finite. (See Kalam Cosmological Argument P2 for the full BGV treatment.)

  3. The growing-block / A-theoretic structure of time is the natural reading. Past events accumulate; future events are not yet; the present is the growing edge. This is the structure presupposed by P4 and supported by phenomenology, ordinary discourse, and most A-theoretic philosophy of time.

Anticipated objections

  1. "B-theory denies that the temporal series is formed, on eternalism, all moments exist tenselessly."
  2. "BGV is theorem about classical cosmology; quantum gravity might rewrite it."

Rebuttals

  1. B-theory does not dissolve the additive structure. On B-theory, all moments exist tenselessly, but they are ordered, there is a temporal succession (event A is temporally before event B). The additive structure remains in the ordering. The argument's force survives translation from A-theoretic "formed" language to B-theoretic "ordered" language: the question becomes whether the B-theoretic timeline can have infinitely-many earlier moments before any given moment, and the same paradoxes apply.

  2. The promissory-quantum-gravity move is faith, not science. No empirically-tested quantum-gravity theory exists. The objector is making the very move ("future-physics will solve this") that he accuses theists of when theists appeal to mystery. Until a quantum-gravity theory exists and survives empirical test, BGV stands. (See Kalam Cosmological Argument for the extended treatment.)

Live-cite kit

  • Scripture: Genesis 1.1 ("In the beginning")
  • Scholarly: Borde-Guth-Vilenkin paper (Phys. Rev. Lett. 90, 2003); Alex Vilenkin, Many Worlds in One (2006); Craig (Time and Eternity, 2001)
  • Aphorism: "Time was, and the 'was' has been accumulating ever since."

Tactical notes

  • This is the least-contested premise; spend minimum debate time on it.
  • If the opponent denies temporal additivity, treat as a reductio, they have abandoned the framework in which any inference (including their own objection) can be made.

Conclusion

Therefore, an infinite temporal regress of past events cannot exist, the past is finite, the universe began, and (joining the Kalam's P1 everything that begins to exist has a cause) has a cause that is timeless (because it caused time), immaterial (because it caused matter), spaceless (because it caused space), enormously powerful (because it caused all that is), and personal (because only a personal agent can act timelessly to produce a temporal effect). The argument supplies the Kalam's P2 (the universe began) by two independent prongs, the conceptual impossibility of concrete actual infinity and the impossibility of forming an actual infinite by successive addition, either of which alone suffices for the conclusion.

Master objections to the argument as a whole

  • "You're using philosophical paradoxes to override empirical possibility, but the past might be infinite, and we'd just have to accept the paradoxes as how things are." Reply: the paradoxes are not mere "surprise" but deliver contradictions (Grim Reaper, Tristram Shandy). Contradictions cannot be empirical features of reality; if a setup is contradictory, the setup does not obtain. The argument is not "the paradoxes are unwelcome"; it is "the paradoxes are incoherent."

  • "The argument proves only a finite past, not God, and a finite past could have a non-personal cause." Reply: conceded that the bare argument from finitude does not by itself entail a personal God. The further step uses (a) the timelessness of the cause (since it caused time) and (b) the puzzle of how a timeless cause produces a temporal first effect without itself being temporally prior. The cleanest solution is a personal agent who could timelessly will the first temporal effect without being temporally caused to do so. This is Craig's "personal-agent inference", supplied in the broader Kalam Cosmological Argument.

  • "Even granting your argument, the cause might be deistic, not theistic." Reply: conceded. The Kalam alone yields a personal timeless creator; specifically Christian theism comes from the cumulative case (see Cumulative Case for Christian Theism, Christian God is the Only True God).

  • "Cantor / set theorists treat actual infinities as legitimate, appeal to them." Reply: as established under P1, Cantor and Hilbert both distinguished abstract from concrete actual infinity, and both denied the concrete case. The set-theoretic appeal misappropriates the position of the cited mathematicians.

Tactical opening / closing

Opening line: "If the universe has an infinite past, then today is the day infinitely many days arrived to. But you cannot count to infinity by counting one day at a time, and that is exactly how days arrive. Therefore the universe has a finite past."

Closing landing strip: "The argument doesn't ask you to accept God yet. It asks you to accept what Hilbert and Cantor accepted: that concrete actual infinites generate paradox, and that you can't count one-at-a-time to infinity. If you grant that, the universe began. The question what caused it to begin is the next question, and it's the question the rest of natural theology answers."

Connection to Scripture

  • Genesis 1.1, "In the beginning God created the heavens and the earth." The finite past directly affirmed.
  • Psalm 90:2, "Before the mountains were born or You gave birth to the earth and the world, even from everlasting to everlasting, You are God." God's everlasting being is contrasted with the created order's beginning.
  • Hebrews 11.3, "By faith we understand that the worlds were prepared by the word of God, so that what is seen was not made out of things which are visible." Creation by divine fiat, presupposing a beginning.
  • John 1:1-3, "In the beginning was the Word... All things came into being through Him." Logos-creation; the Word as eternal predicates a beginning for the things-created.
  • Romans 4.17, "God who gives life to the dead and calls into being that which does not exist."
  • Colossians 1.17, "He is before all things, and in Him all things hold together."

Patristic / scholarly note

Classical / patristic / medieval:

  • John Philoponus (6th c.), De Aeternitate Mundi contra Proclum; first Christian theologian to develop the actual-infinite-impossibility argument against pagan eternal-cosmos. Direct precursor to al-Ghazali.
  • al-Ghazali (11th c.), Tahafut al-Falasifa (The Incoherence of the Philosophers); the medieval Islamic kalam's classical formulation. The "kalam" of "Kalam Cosmological Argument" derives from this tradition.
  • Bonaventure (13th c.), Commentary on the Sentences, II.1.1.1.2; defends the impossibility of an actual-infinite past against Aquinas's hesitancy.
  • Aquinas (13th c.), ST I q.46; held that we cannot demonstrate philosophically that the universe began (only by faith), but accepted that the actual-infinite-impossibility prong is philosophically strong. The disagreement with Bonaventure is intramural-Christian.

Modern:

  • David Hilbert, "Über das Unendliche" (Mathematische Annalen 95, 1926), Hilbert's Hotel.
  • G.J. Whitrow, The Natural Philosophy of Time (1961, 2nd ed. 1980), pre-Craig classical statement.
  • William Lane Craig, The Kalām Cosmological Argument (1979), the canonical modern treatment.
  • Craig, Reasonable Faith (2008, ch. 3), popular-level statement.
  • Craig & James Sinclair, "The Kalam Cosmological Argument" in The Blackwell Companion to Natural Theology (2009), the comprehensive scholarly statement.
  • Andrew Loke, God and Ultimate Origins (2017), recent technical defense.
  • Alex Pruss, Infinity, Causation, and Paradox (2018), Grim Reaper and related paradoxes.
  • Robert Koons, "A New Kalam Argument: Revenge of the Grim Reaper" (Faith and Philosophy, 2014).
  • J.P. Moreland, Scaling the Secular City (1987, ch. 1).
  • Borde-Guth-Vilenkin (Phys. Rev. Lett. 90, 2003), the cosmological theorem.

See also