Argument

The Seedtime Argument

Intro

Every time you turn a key, plant a seed or take a pill, you are betting that the world will behave tomorrow the way it behaved yesterday. Science makes the same bet on a huge scale. But in 1748 the philosopher David Hume pointed out a problem: you cannot prove that the future will be like the past by looking at the past. All your evidence is about yesterday. To trust it for tomorrow, you already have to assume that nature keeps its habits.

Two things make the problem sharper. First, the bet keeps paying off. Physicists can check whether the basic numbers of nature have changed over billions of years, and they find that they have not, to astonishing precision. Second, we are born making the bet. Babies a few months old are surprised when objects vanish or pass through each other.

So: we cannot help trusting that nature is steady, nature really is steady, and nothing inside nature can tell us why it must stay that way. The Bible gives a reason. After the flood God promised, "While the earth remaineth, seedtime and harvest, and cold and heat, and summer and winter, and day and night shall not cease" (Genesis 8:22). The Seedtime Argument says the steadiness of the world is what you would expect if it rests on the faithfulness of a Person who keeps His word.

◎ See this page's graph and discover linked topics →

In full

The Seedtime Argument is an epistemological argument from the problem of induction. Hume showed that inductive inference depends on a principle of uniformity, that unobserved cases resemble observed ones, which cannot be established by demonstration (its denial is consistent) or by experience (that would be circular). Yet induction is rationally indispensable, and it is spectacularly successful: the fine-structure constant, for example, has changed by less than about one part in ten million since the Oklo natural reactors ran some two billion years ago, and by no more than a few parts per million in light from a quasar absorber that left its source billions of years ago. Human infants, moreover, expect the persistence and regularity of objects before they could have learned it. A deductive spine (P1 to P5) argues that the rational trust in uniformity that induction requires needs a ground; that the main naturalistic accounts (Humean regularity, which supplies none; pragmatic vindication, which supplies no reason to expect success; and necessitarian laws, which relocate the question) do not supply an adequate one; and that a faithful Creator who sustains nature by a constant will, and who on the Christian account has promised its regularity, does. The thesis develops John Foster's argument in The Divine Lawmaker (2004), adding the developmental, physical and covenantal strands. The historical claim that early modern science drew confidence from belief in a faithful lawgiver (Whitehead, Oakley) is used as corroboration, not as a premise. The conclusion is abductive and confirmatory: the reliability of induction is more expected, and its rationality better grounded, on theism than on naturalism.

Format note

This page is a braided case in the house format of Ris3n Arguments: a valid spine whose premises are each supported by data, by a comparison of rival explanations and by agreement across fields, followed by a check on double-counting, a ledger and a defeater gate. It braids five fields. Philosophy supplies the problem (Hume's argument) and the rival solutions. Psychology shows that the expectation of regularity is in place before experience could have taught it. Physics measures how uniform nature actually is. Theology supplies the ground: God's faithfulness and the covenant with the seasons. History shows that the founders of modern science reasoned in just this way.

The spine

Validity. P1 and P2 say induction needs a ground. P3 says the expectation is both innate and borne out, so the ground, if there is one, is real. P4 and P5 compare the candidates: only theism supplies the ground. The form is an inference to the best explanation closed by the likelihood principle.

Strength of the conclusion. Abductive and confirmatory. The argument does not claim that theists have a demonstrative proof that the sun will rise, or that atheists cannot use induction. It claims that theism explains why induction is reliable and why trusting it is reasonable, and naturalism, on its main accounts, does not.

P1, Induction is indispensable

Inductive strand

1. Science is inductive. Every law of physics is a generalisation from finitely many observations to all cases, including future and distant ones.
2. Daily life is inductive. Eating, walking and speaking assume that bread nourishes, floors hold and words keep their meanings.
3. The sceptic is inductive. Any argument that induction is unreliable relies on remembering its premises and on expecting logic to work the same way at the end of the argument as at the start.

Abductive strand

Convergence

Philosophy of science (Hume, Russell, Reichenbach, Goodman), the practice of physics and ordinary life all treat induction as unavoidable. The dispute is over its justification, not its use.

Objections and rebuttals

1. "Science uses falsification, not induction (Popper)." Rebuttal. Falsification still assumes that a refuted law stays refuted tomorrow and that the test conditions will repeat. Uniformity is presupposed by testing too.

P2, Induction needs a ground it cannot give itself

Inductive strand

1. Hume's argument. In An Enquiry Concerning Human Understanding (1748), section IV, Hume argues that "all our experimental conclusions proceed upon the supposition that the future will be conformable to the past." That supposition cannot be demonstrated, since a change in nature's course is conceivable without contradiction, and it cannot be proved from experience, since that would assume what is in question. See David Hume.
2. Russell's chicken. Bertrand Russell, The Problems of Philosophy (1912), chapter 6: the man who has fed the chicken every day "at last wrings its neck instead, showing that more refined views as to the uniformity of nature would have been useful to the chicken." Past regularity alone does not guarantee future regularity. See Bertrand Russell.
3. The circle. "Induction has worked, so it will work" is itself an inductive inference. That does not make induction false; it means its rationality must rest on something else.

Abductive strand

Convergence

Hume (a sceptic), Russell (an atheist), and theists such as Foster agree on P2: induction does not justify itself. See Inductive Reasoning.

Objections and rebuttals

1. "Induction is just what rationality means (Strawson's analytic reply)." Rebuttal. That shows that inductive reasoning is called reasonable; it does not show that a world whose regularities might end at any moment is one in which reasonable beliefs tend to be true. The question is why the reasonable practice is also a reliable one.

P3, Nature is uniform, and we expect it to be

Inductive strand

1. The Oklo bound. About two billion years ago, natural nuclear fission reactors ran in uranium deposits at Oklo, Gabon. The surviving isotope ratios constrain nuclear physics at that time. Thibault Damour and Freeman Dyson, "The Oklo Bound on the Time Variation of the Fine-Structure Constant Revisited" (Nuclear Physics B 480, 1996), found that the fine-structure constant could have changed since then by no more than about one part in ten million.
2. Distant light. Michael T. Murphy and colleagues, "Fundamental Physics with ESPRESSO: Precise Limit on Variations in the Fine-Structure Constant towards the Bright Quasar HE 0515-4414" (Astronomy and Astrophysics 658, 2022), measured the constant in an absorber whose light has travelled for billions of years and found a value consistent with today's at the level of about one part per million. Earlier hints of variation from quasar spectra were reported by John Webb's group; the newest, laser-calibrated measurements do not confirm them.
3. Infants expect regularity. Renee Baillargeon, Elizabeth Spelke and Stanley Wasserman, "Object Permanence in Five-Month-Old Infants" (Cognition 20, 1985), found that infants looked longer at an event in which a screen appeared to pass through a hidden box, as if surprised. Spelke and Katherine Kinzler, "Core Knowledge" (Developmental Science 10, 2007), summarise evidence that infants expect objects to persist, move continuously and not pass through one another.

Abductive strand

Convergence

Nuclear geochemistry (Oklo), astronomical spectroscopy (quasars) and developmental psychology (infant looking-time studies) reach P3 by unrelated methods, in unrelated laboratories, with no theological interest.

Objections and rebuttals

1. "Evolution explains the infant expectation." Rebuttal. Partly, and the argument allows it. Selection explains why organisms expect the regularities that matter for survival. It does not explain why the world is uniform, which is the condition for any such selection to work.
2. "The constants might change in the future." Rebuttal. Exactly Hume's point, and it is why P2 asks for a ground beyond the record.

P4, The rival grounds

Inductive strand

1. Humean regularity. On a Humean view, laws are just patterns in what happens. Then nothing makes the pattern continue; the future's resemblance to the past is a brute coincidence that has held so far.
2. Pragmatic vindication. Hans Reichenbach (Experience and Prediction, 1938) argued that if any method of prediction works, induction will. That gives a reason to use induction, but no reason to expect it to work.
3. Necessitarian laws. David Armstrong (What Is a Law of Nature?, 1983) holds that laws are relations of necessitation between properties, and that such laws best explain observed regularities, which answers inductive scepticism by inference to the best explanation. This is the strongest naturalistic answer; it is the subject of the defeater gate.

Abductive strand

Convergence

Foster's The Divine Lawmaker (Oxford, 2004) argues the same comparison in technical detail: regularities need explaining, laws are best understood as regularities imposed on the world, and the best ultimate explanation of their imposition is the agency of God.

Objections and rebuttals

1. "We don't need a ground; induction just works." Rebuttal. That is the Humean answer, and it concedes the point: on that view, the working is unexplained.

P5, The theistic ground

Inductive strand

1. The promise. "While the earth remaineth, seedtime and harvest, and cold and heat, and summer and winter, and day and night shall not cease" (Genesis 8:22, ASV).
2. The covenant with day and night. God stakes His covenant with David on the fixity of creation: "If my covenant of day and night stand not, if I have not appointed the ordinances of heaven and earth; then will I also cast away the seed of Jacob" (Jeremiah 33:25-26, ASV). The regularity of nature is presented as a pledge as reliable as God's word.
3. Faithfulness as the ground of order. "Thy faithfulness is unto all generations: Thou hast established the earth, and it abideth. They abide this day according to thine ordinances; For all things are thy servants" (Psalm 119:90-91, ASV).
4. Present sustaining. The Son is "upholding all things by the word of his power" (Hebrews 1:3, ASV), so the uniformity is not a past setting but a present, constant act.
5. A world made to be learned. If God made rational creatures to know and care for His world, He has reason to make it regular enough to learn from, and to give those creatures the expectation of regularity in advance, which is what P3 found.

Abductive strand

On theism, uniformity is grounded in the character of the One who sustains nature: He does not change His mind about how the world works, because His will is constant and He has promised. That ground is not reached by induction from nature; it belongs to what God is. So the circle in P2 is broken. Naturalism has no corresponding ground: nature's regularity is either brute (Hume) or rests on necessitations that are themselves brute (Armstrong).

Convergence

Biblical theology (Genesis, Jeremiah, Psalms, Hebrews), philosophical theology (Foster), and the founders of modern physics converge on P5. Descartes, in Principles of Philosophy (1644), part 2, derived the conservation of motion and his laws of nature from the immutability of God. See Rene Descartes.

Objections and rebuttals

1. "If God can work miracles, nature is not uniform." Rebuttal. The promise concerns the general order ("while the earth remaineth"). A miracle is a rare, purposeful exception that can only be recognised as one against a background of regularity, a point C. S. Lewis develops in Miracles (1947). A God who acts for reasons makes both the order and the rare sign intelligible; a world of brute regularities makes neither. See Miracles.
2. "The theist knows God is faithful only by induction from experience." Rebuttal. The argument concerns what makes induction rational and reliable, not how each believer first comes to trust God. If theism is true, uniformity has a ground whether or not anyone has reasoned it out, just as eyesight is reliable before anyone studies optics.

The history strand

This strand corroborates P5; it is not a premise.

1. Whitehead's thesis. Alfred North Whitehead, Science and the Modern World (1925), argued that the faith in the possibility of science, generated before modern scientific theory developed, was "an unconscious derivative from medieval theology," specifically from belief in the rationality of God.
2. Laws of nature as a theological concept. Francis Oakley, "Christian Theology and the Newtonian Science: The Rise of the Concept of the Laws of Nature" (Church History 30, 1961), traced the early modern idea of laws imposed on nature to the theology of a free divine Legislator.
3. Confidence and humility together. Peter Harrison, The Fall of Man and the Foundations of Science (Cambridge, 2007), argues that early modern beliefs about the Fall's damage to human faculties motivated the experimental method: nature is lawful, but fallen minds must test rather than assume. Harrison's thesis is debated by historians, and the page uses it only as an illustration that the founders' confidence in order came with an awareness of their own fallibility.

The knot

The strands come from independent sources. The philosophical problem (Hume, Russell) was framed by sceptics. The physical measurements (Damour and Dyson; Murphy and colleagues) are empirical and theologically neutral. The infant studies (Baillargeon, Spelke) come from developmental psychology. The theological ground (Genesis 8:22, Jeremiah 33:25) is ancient. Foster supplies the philosophical bridge, and the historians (Whitehead, Oakley, Harrison) supply corroboration. Two overlaps are counted once: the Oklo and quasar measurements are one row ("nature is uniform"), not two; and the history strand is kept out of the ledger's core rows, because it shows what the founders believed, not that their belief was true.

The ledger

Prior-agnostic: the ledger asks only how expected each datum is under each hypothesis. The ratings are reasoned judgments from the sections above, not measurements.

Net. The weight rests on the fourth row: if one thinks the success of induction is not a brute coincidence, theism offers a ground that naturalism lacks. The physical row adds a modest confirmation, and the psychological and historical rows add little on their own. The argument is as strong as one's conviction that the steadiness of the world calls for an explanation.

The defeater gate

The necessitarian reply. The strongest objection is Armstrong's. Laws are not mere patterns; they are relations of necessitation between properties, holding at every place and time because they are relations between universals, which are wholly present wherever they are instantiated. Observed regularities are best explained by such laws, so inductive inference to them is justified by inference to the best explanation, with no God required. Theism adds an Agent behind the laws, but that is an extra posit with no additional explanatory payoff.

The reply has three parts.

1. The question is relocated, not answered. Why do these necessitation relations hold, and not others? Why does the relation between mass and gravitational attraction obtain with this strength at all? Armstrong takes the relations as contingent and basic. That is a brute fact one level down, where theism says the relations hold because a constant will imposes them.
2. Foster's objection. John Foster argues that a necessitation relation between universals cannot by itself make particular things behave in a regular way; something must impose the regularity on the world, and the imposition is best understood as the act of an agent. The necessitarian view, worked out, points toward an imposer.
3. The payoff is not zero. Theism explains three things the necessitarian leaves as brute: why there are laws at all, why they are stable, and why they are learnable by the minds that live under them. One Agent with reasons explains all three.

What the gate concedes: necessitarian realism is a serious non-theistic view, and it is far better than Humean scepticism. The argument does not claim naturalists cannot do science. It claims that theism explains why their science works.

Concessions

• Grant: atheists use induction just as well as anyone. Now collect: the question was never whether they can use it but what makes it reliable; everyone relies on a steadiness only one view explains.
• Grant: the constants might, for all logic says, change tomorrow. Now collect: then the confidence everyone has that they will not is either a brute habit or a trust with a ground. Name the ground.
• Grant: the history of science does not prove theism. Now collect: the founders who built the method held exactly the ground this argument proposes, which at least shows the ground is fit for the purpose.

Cheatsheet

The 30-second version (read aloud). Every science and every plan assumes tomorrow will be like today. Hume showed you cannot prove that from the past without assuming it. Yet it keeps being true: the basic constants of physics have not changed measurably in two billion years, and babies expect the world to be steady before anyone teaches them. Why is the world so faithful? The Bible's answer is that it rests on the faithfulness of God, who promised that seedtime and harvest would not cease. On atheism the steadiness is just there. On Christianity it is a promise being kept.

Fast facts:

1. Hume (1748): "all our experimental conclusions proceed upon the supposition that the future will be conformable to the past."
2. Damour and Dyson (1996): the fine-structure constant changed by less than about one part in ten million since the Oklo reactors ran about two billion years ago.
3. Murphy et al. (2022): quasar light gives no change at about one part per million.
4. Baillargeon, Spelke and Wasserman (1985): five-month-olds are surprised when solid objects seem to pass through each other.
5. Foster (2004): the best explanation of laws of nature is a divine Lawmaker.

The closing line. Seedtime and harvest have not ceased, and the One who promised it has not changed.

Cross-examination

Round 1, The problem

• Q: How do you know the laws of physics will hold tomorrow?
• If they say "they always have": ask whether that answer uses the very assumption in question.

Round 2, The ground

• Q: On your view, is there anything that makes nature uniform, or is it simply uniform so far?
• If they say "simply so far": then the success of science is an unexplained streak.
• If they say "necessary laws": ask why these laws, and what makes them hold at every moment.

Round 3, The fit

• Q: Why would a universe with no purpose be steady enough, for billions of years, for minds to learn its laws?
• If they say "we couldn't exist otherwise": ask whether life needs constants fixed to parts per million across the whole cosmos.

Dodge table

Live-cite kit

Scripture (ASV):

• Genesis 8:22: While the earth remaineth, seedtime and harvest, and cold and heat, and summer and winter, and day and night shall not cease (Genesis 8:22, ASV)
• Jeremiah 33:25: Thus saith Jehovah: If my covenant of day and night stand not, if I have not appointed the ordinances of heaven and earth (Jeremiah 33:25, ASV)
• Psalm 119:89-91: For ever, O Jehovah, Thy word is settled in heaven. Thy faithfulness is unto all generations: Thou hast established the earth, and it abideth. They abide this day according to thine ordinances; For all things are thy servants (Psalm 119:89-91, ASV)
• Lamentations 3:23: They are new every morning; great is thy faithfulness (Lamentations 3:23, ASV)

Scholarly:

• David Hume, An Enquiry Concerning Human Understanding (1748), section IV.
• John Foster, The Divine Lawmaker: Lectures on Induction, Laws of Nature, and the Existence of God (Oxford, 2004).
• Thibault Damour and Freeman Dyson, Nuclear Physics B 480 (1996): the Oklo bound.
• A. N. Whitehead, Science and the Modern World (1925): faith in science as "an unconscious derivative from medieval theology."

Aphorism: "The sun rises on a promise, not on a habit."

Connection to Scripture

• Genesis 8:22, the covenant with the seasons, made with all humanity through Noah.
• Jeremiah 33:25-26, the fixed ordinances of heaven and earth as the measure of God's faithfulness.
• Psalm 119:89-91, creation abiding by His ordinances because of His faithfulness.
• Hebrews 1:3, all things upheld now by His word.
• Lamentations 3:22-23, faithfulness new every morning.

Scholarship

Published academic works this page draws on, each matched to a publisher, index or library catalog record.

• D. M. Armstrong, What is a Law of Nature? (Cambridge University Press, 1983). doi:10.1017/cbo9781139171700
• Renée Baillargeon, Elizabeth S. Spelke and Stanley Wasserman, "Object permanence in five-month-old infants," Cognition 20 (1985). doi:10.1016/0010-0277(85)90008-3 PubMed 4064606
• Thibault Damour and Freeman Dyson, "The Oklo bound on the time variation of the fine-structure constant revisited," Nuclear Physics B 480 (1996). doi:10.1016/s0550-3213(96)00467-1 open access
• John Foster, The Divine Lawmaker: Lectures on Induction, Laws of Nature, and the Existence of God (Oxford University Press, 2004). doi:10.1093/0199250596.001.0001
• Peter Harrison, The Fall of Man and the Foundations of Science (Cambridge University Press, 2007). doi:10.1017/cbo9780511487750
• Michael T. Murphy et al., "Fundamental physics with Espresso: Precise limit on variations in the fine-structure constant towards the bright quasar HE 0515−4414," Astronomy & Astrophysics 658 (2022). doi:10.1051/0004-6361/202142257
• Francis Oakley, "Christian Theology and the Newtonian Science: The Rise of the Concept of the Laws of Nature," Church History 30 (1961). doi:10.2307/3161219
• Hans Reichenbach, Experience and prediction (University of Chicago Press, 1938). Open Library
• Alfred North Whitehead, Science and the modern world: Lowell lectures, 1925 (Macmillan, 1925). Open Library

See also

• Ris3n Arguments, the hub of ris3n's original arguments
• Arguments, the master index
• Argument from Intelligibility, why the world can be understood at all; this page asks why it stays the same
• Hand of God Argument, why truth needs an agent; this page narrows the question to the future's resemblance to the past
• Laws of the Universe as Witness to Design and Mathematical Intelligibility of Nature, the laws whose stability this page examines
• Inductive Reasoning, the form of reasoning at issue
• David Hume and Bertrand Russell, the sceptics who framed the problem
• Rene Descartes, laws of nature from God's immutability
• Argument from the Continuance of Being, the present sustaining of the world
• Miracles, exceptions that presuppose the rule
• Cumulative Case for Christian Theism, where this argument contributes