Evidence for God: Six Arguments and a Coherent God
There is evidence for God, and it comes in the form of arguments, each with its own logical force: a few claim to be demonstrations from premises a skeptic may still dispute, and most are inferences to the best explanation that together make a strong cumulative case. Six are surveyed here: the ontological argument, the argument from design and fine-tuning, the kalam, the argument from contingency, the moral argument and Aquinas' Five Ways. Beneath all six sits a harder question, whether a maximal God is a coherent idea at all, and that is where we make our own case for the classical attributes of God.
Key takeaways
- Evidence for God is not the same as proof. Most arguments for God are inferences to the best explanation; the few demonstrations rest on disputed premises.
- Every argument meets a standard objection. Kant, Darwin, the multiverse, open cosmology, Hume and the Euthyphro dilemma each press on one of the six.
- The deepest objection is incoherence. If a maximal God is "an impossible object like a circular square," no argument for Him can succeed.
- Frame-Indexed Classical Theism (FICT) answers with an argument from coherence. Where Richard Swinburne revises timelessness and foreknowledge and open theism trims omniscience, FICT keeps every maximal attribute and answers each charge, which supports the possibility premise the modal ontological argument needs.
What counts as evidence for God?
Asking for evidence for God rarely means expecting God in a test tube; a being who made the physical world is not one more object inside it. The question is whether there are good reasons to think God exists, reasons that make belief more reasonable than its denial.
Two words need separating. A proof, or demonstration, compels assent from premises already granted. Evidence is weaker and far more common. It makes a conclusion more probable than its rivals, the way fingerprints make one suspect likelier than another without a confession. Most of the arguments for God are evidence in this second sense.
Philosophers also sort the arguments by where they start. A priori arguments begin from the idea of God alone; a posteriori arguments begin from the world, its existence, order and beginning. Thomas Aquinas held that God's existence, though not self-evident to us, "can be demonstrated from those of His effects which are known to us."
Free — the complete book
Get Book 1: Resolving the Free Will Dilemma free when you enter your name and email.
This essay takes one piece of it. Resolving the Free Will Dilemma takes the whole dilemma from first axiom to conclusion — the complete book, free.
No spam. Unsubscribe in one click.
Scripture expects this kind of reasoning. Paul writes that "what may be known about God is plain to them, because God has made it plain to them" (Romans 1:19, NIV), and that God's power and nature have been "clearly seen, being understood from what has been made" (Romans 1:20, NIV). Peter asks the same of every believer:
"Always be prepared to give an answer to everyone who asks you to give the reason for the hope that you have." (1 Peter 3:15, NIV)
The question opens every other question in the doctrine of God.
Volume One builds the classical case from first premises, and you can read it free.
Is there no evidence for God? The objection at its strongest
The skeptic's case comes in two strengths. The first is empirical. No observation, the argument runs, separates a world made by God from a world that runs by its own laws, and every gap theism once filled has been closed by science. William Paley saw design in the eye as clearly as in a watch; then Darwin explained the eye's fit to its task without a designer. If the pattern holds, "God" will turn out to be a name for what we have not yet explained.
The second form is deeper, and the most ambitious atheist arguments take it: could the thing the evidence is for exist at all? The God of the classical arguments is maximal: all-powerful, all-knowing, perfectly good, unchanging, simple and eternal. The Stanford Encyclopedia of Philosophy states the strategy: "one way to argue for the nonexistence of the God of omni-theism … is to argue that such a God is an impossible object like a circular square. Many attempts have been made to construct such arguments." Its own example sets two attributes against each other: a perfectly good being would be "incapable of wrongdoing," while an omnipotent being "would be quite capable of doing things that would be wrong to do."
The charges are familiar. Can God make a stone He cannot lift? If God knows what I will choose, am I free? Can there be a set of all truths for God to know? How can a God who never changes answer prayer? If God is simple, did He have to create? And if He loves us, why is He hidden?
This objection comes first in logical order: if a maximal God is incoherent, then nothing can be evidence of God, any more than a footprint can be evidence of a square circle.
The series takes these charges one volume at a time; see every volume.
What are the arguments for the existence of God?
The arguments for the existence of God fall into a handful of families, and the literature defends them in three broad ways. Each succeeds at something and strains somewhere.
The classical approach, Aquinas's, offers demonstrations: arguments from what we observe to a first cause. Its strength is rigor. Its strain is that its premises, about causation and about why anything exists, are exactly the ones a modern skeptic disputes.
The cumulative approach treats each argument as raising the probability of theism. Richard Swinburne, as the Stanford Encyclopedia puts it, rejects deductive versions of the cosmological argument and "proposes an inductive argument that is part of a larger cumulative case for God's existence." Its strength is modesty. Its strain appears when the coherence question arrives, because Swinburne's defense of a coherent God revises two classical attributes.
The third approach, William Lane Craig's, leads with the kalam and meets modern cosmology on its own ground. Its strength is engagement with physics. Its strain is that one premise leans on a question physicists still hold open.
Six arguments follow, each with its standard objection.
The ontological argument: can God be known from the idea of God?
The ontological argument is the one a priori argument in the set. In the Stanford Encyclopedia's summary, "Anselm claims to derive the existence of that than which no greater can be conceived from the concept of that than which no greater can be conceived." Such a being existing only in the mind would be less great than one existing in reality too, so it must exist. Anselm's wider place in the classical tradition is taken up in Volume One, Resolving the Free Will Dilemma.
Two objections have shadowed it for centuries. Gaunilo, Anselm's contemporary, offered a "perfect island" parody: the same reasoning would conjure the greatest conceivable island into existence. Kant held that the argument treats "existence" as "a real predicate," a property that adds to a thing's greatness.
The modern form, from Alvin Plantinga, sidesteps both. A being is maximally great if it "exists necessarily and is essentially omnipotent, omniscient, and morally perfect." If such a being is so much as possible, it exists in some possible world, and a necessary being that exists in one world exists in all, ours included. The argument rests on that first premise, and the encyclopedia names what it calls perhaps the most important objection: the symmetry problem, which runs the same logic from "Possibly, God doesn't exist" to the opposite conclusion. Whether a maximally great being is possible is where this survey returns.
Volume One is free to download, and the series continues from there.
The teleological argument: does fine-tuning point to a designer?
The teleological argument reasons from order to mind. For Paley, a stone needs no explanation, but a watch, its parts framed for a purpose, makes "the inference we think is inevitable, that the watch must have had a maker." Paley carried that inference to the eye, and Darwin answered him there.
The modern version moves from biology to physics. According to many physicists, as the Stanford Encyclopedia reports, "the fact that the universe is able to support life depends delicately on various of its fundamental characteristics." The cosmological constant is the sharpest case: on theoretical grounds "one would expect it to be larger than its actual value by an immense number of magnitudes." A classic response is "to infer the existence of a cosmic designer." The standard objection is the multiverse: given enough varied universes, "it is only to be expected that there is at least one where they are right for life." Critics, following Ian Hacking, call that reply the inverse gambler's fallacy: inferring many unseen trials from one remarkable outcome.
The series adds two arguments of this family. Eugene Wigner called the fit of mathematics to physics "a wonderful gift which we neither understand nor deserve," and Volume Five, Resolving Potential vs Actual Infinity, reads that gift as one Logos behind world and mind. Volume Six, Resolving Rest and Relativity, finds one kind of limit on every axis of physical striving. Both call this evidence for God what it is: inference to the best explanation.
Volume Six sets out the physical limits in full; the series runs volume by volume.
The kalam: did the universe begin to exist?
The kalam cosmological argument, revived by William Lane Craig, runs in three lines: "Everything that begins to exist has a cause of its existence. The universe began to exist. Therefore, the universe has a cause of its existence." For the second premise, an actually infinite past could never have been traversed to reach today, and defenders cite Big Bang cosmology and a 2003 theorem of Arvind Borde, Alan Guth and Alexander Vilenkin: a universe expanding fast enough "must be incomplete in null and timelike past directions."
The standard objection targets that premise. The same paper concludes only that "inflationary models require physics other than inflation to describe the past boundary," and bounce models and the no-boundary proposal remain live candidates. Russell, following Hume, doubted we can ask for the cause of the whole at all.
The series stands with Craig on both of his great conclusions: no actual infinite exists in creation, and God's infinity is not a number. It differs on the route. Volume Five secures creation's finitude by the difference between God's self-existence and the universe's dependence, with the paradoxes and the cosmology serving as corroboration. The infinity of God and creation from nothing take the thread further.
Volume Five gives the infinite its full treatment; browse the series.
The argument from contingency: why is there anything at all?
The argument from contingency asks why anything exists rather than nothing. Everything in the world could have failed to exist. Leibniz and Samuel Clarke built the argument on the Principle of Sufficient Reason, and Aquinas's Third Way reaches the same terminus: "we cannot but postulate the existence of some being having of itself its own necessity, and not receiving it from another, but rather causing in others their necessity."
The standard objection comes from Hume, who held that "when the parts are explained the whole is explained." Others deny the Principle of Sufficient Reason itself.
The argument from contingency has its own article answering both. The series treats contingency, not the impossibility of an infinite past, as the primary ground for the universe's finitude, so it can hold the kalam's conclusion while holding its physics loosely. The necessary being the argument reaches is the God of aseity, the ground of being, and this is the most secure evidence for God of the six.
The first volume opens the series; download it free.
The moral argument: do moral facts need God?
The moral argument reasons from the reality of moral obligation to its ground. Its theoretical form, as the Stanford Encyclopedia gives the template, begins: "There are objective moral facts. God provides the best explanation of the existence of objective moral facts." Cruelty for its own sake is wrong whether or not anyone agrees, and matter in motion has no evident room for that kind of fact.
Two objections meet it. The first is the Euthyphro dilemma, Plato's question whether the good is good because God commands it, which on one horn seems to "reduce ethics to arbitrariness." The second is evolutionary debunking, pressed by Sharon Street: if survival produced our moral beliefs, their match to moral truth would be luck.
Both have their own treatment in this cluster: the moral argument and the Euthyphro dilemma. It, too, is an inference to the best explanation, and it, too, presupposes that a perfectly good God is possible.
Aquinas' Five Ways: from effects to their cause
"The existence of God can be proved in five ways," Aquinas writes in the Summa Theologiae, and the five run from motion, efficient causation, possibility and necessity, degrees of perfection, and the ordering of natural things to their ends. The first concludes, "Therefore it is necessary to arrive at a first mover, put in motion by no other; and this everyone understands to be God." The fifth concludes, "Therefore some intelligent being exists by whom all natural things are directed to their end; and this being we call God."
The standard objection fastens on those closing phrases: granted a first mover, why is it the God of Christian worship? Aquinas does not settle that in Article 3; the questions that follow derive God's simplicity, perfection and goodness. These arguments, the oldest systematic evidence for God from His effects, have their own article on the Five Ways, as does the unmoved mover, and Thomism traces the school.
Read Volume One free for the classical case argued in full.
How does Frame-Indexed Classical Theism answer? An argument from coherence
Frame-Indexed Classical Theism (FICT) is a research program in analytic theology that resolves classical theological dilemmas by diagnosing them as artifacts of a single error: one term carried across two frames as though it meant the same thing in each. Its contribution to the evidence for God begins with what the six arguments share: each reaches a maximal God, and each fails if a maximal God is impossible.
Theists have answered the charge in three ways. Richard Swinburne, in The Coherence of Theism, defends a coherent God by revising Him: he moved from a timeless God to a temporal one, and he holds, as the Stanford Encyclopedia reports, that omniscience "does not include foreknowledge of future free actions." Open theism trims foreknowledge in the same way, and others retreat from maximal power to "optimally powerful." Each revision buys coherence by spending greatness, so the being shown possible is no longer maximal. The classical theist keeps every attribute and appeals to divine simplicity and to analogy, and then meets Alvin Plantinga's charge that a God identical with His properties "is a property," and modal collapse besides.
FICT keeps every classical attribute, untrimmed; we call its answer an argument from coherence. Each incoherence charge carries one term, "power," "know," "change" or "necessary," across the line between God's frame and ours as if it meant the same in both. Read in the frame of the one it describes, the term yields no contradiction, and a God whose full set of maximal attributes yields none is a coherent God. Volume Five states the principle beneath this: coherence is the creature's test of truth, because "God is simple and non-contradictory," and "no incoherent set of ideas can correspond to what He sustains."
The logical force must be exact. The argument from coherence defeats the claim that a maximal God is an impossible object, and it supports the possibility premise the modal ontological argument needs. It does not, by itself, prove that God exists: consistency is the first requirement of possibility, and a critic may still ask for more. Scripture states the gap between the frames first:
"For my thoughts are not your thoughts, neither are your ways my ways," declares the LORD. "As the heavens are higher than the earth, so are my ways higher than your ways and my thoughts than your thoughts." (Isaiah 55:8-9, NIV)
The statement of FICT sets out the position in one place, and Volume One argues it at length, free.
Is a maximal God coherent? Six charges answered
Yes, and each charge has a full article on this site; here are the conclusions, so the pattern shows.
The omnipotence paradox asks whether God can make a stone He cannot lift. A contradiction names no task, so failing to perform it is no lack of power, and God's power does not crowd out ours once the two are not weighed on one scale. See the omnipotence of God.
The charge from foreknowledge asks whether, if God knows what I will choose, I must choose it. God's knowing is not a past belief that fixes the future, and the only necessity His knowledge carries is the necessity of the consequence. See whether God knows the future and theological fatalism.
On omniscience, Patrick Grim argues from Cantor's theorem that there is no set of all truths, and so no omniscient being. The mathematics holds; the premise that omniscience is a set does not. See whether God is omniscient.
The charge from immutability asks how a changeless God can answer prayer. The texts in which He relents and the texts in which He never changes both stand as written, because the words carry a different sense in His frame and in ours. See whether God changes His mind.
The modal collapse argument asks whether a simple God had to create. Necessity belongs to God's act taken in itself; "the creating of this world" names that act by its effect, and the effect could have been otherwise. See modal collapse.
Divine hiddenness asks why a loving God is hidden. Hiddenness is a holding back of His full, unveiled presence in this age, so that a finite creature has room to respond freely. See divine hiddenness.
Six charges, one signature: a term with a frame carried across the Creator-creature line as though it had none. The distinction that answers them was drawn from the natures of God and creature before any dilemma was raised, so it is no subscript applied after the fact.
The series works through each dilemma in its own volume.
Has the evidence for God survived its critics?
We have run thousands of simulated attacks against Frame-Indexed Classical Theism, drawn from academic and peer-reviewed sources. Five bear directly on the evidence for God, the case that Swinburne's cumulative theism and Craig's kalam each defend in their own way.
Isn't frame-indexing just a double truth?
Here's one of those attacks and how the Many Beings framework would address it.
The attack comes from formal logic, in the name of the law of non-contradiction, and it revives an old charge: "double truth," holding a claim true in theology and false in philosophy. Take two contradictory statements, "the future is potential" and "the future is actual," write a subscript for God's frame beside one and for ours beside the other, and declare them compatible. Has a contradiction been removed, or merely relabeled? If indexing can dissolve any contradiction, it proves nothing.
The defense grants the general point: an index with nothing beneath it is bookkeeping, which files a contradiction rather than removing it. But the two frames are not equal peers. God's frame is the ground of the one reality, and ours is a derived and sustained mode of access to it, bounded and true. "Potential" and "actual" here are not one predicate and its negation but the modes in which one grounded reality is known by two kinds of knower, so the condition every contradiction needs, the same thing affirmed and denied in the same sense, fails. Nor is the index free. The series names in advance which terms index, those involving modality, time, agency and value, and which never do: identity, non-contradiction and arithmetic. A rule like that can be broken, and so it can be checked.
Does the Big Bang really show the universe began?
Here's one of those attacks and how the Many Beings framework would address it.
A physicist in the standard relativistic tradition pressed this against Volume Three's account of the beginning. The Big Bang, he points out, names the hot, dense early state of the universe, and whether it was the first moment of time is another question. The Borde-Guth-Vilenkin theorem shows that an expanding universe's past is incomplete, and calls for "physics other than inflation" at that boundary, not a beginning of time. Bounce cosmologies, the no-boundary proposal and eternal inflation all remain open, so to call a finite beginning universally recognized overstates the physics.
The defense accepts the correction, and Volume Three now reads that way. Cosmology "strongly supports a finite cosmic past, commonly identified with the Big Bang," and whether that boundary is the first moment of time "remains physically underdetermined." No premise was lost, because the argument never rested on the physics. The book's primary ground for a finite past is contingency: only God exists from Himself, and the universe exists only as He sustains it. Cosmology points the same way, "never doing the deciding." Volume Six says the same of the no-boundary proposal: if it proved correct, "this book's argument would not lose a premise."
Do the limits of physics prove a Creator?
Here's one of those attacks and how the Many Beings framework would address it.
A physicist in the same tradition targeted an early draft of Volume Six, which claimed the universe's limits prove it was created: "Only a created thing has edges." Planck-scale "pixelation" is not settled physics, absolute zero is the floor of a scale rather than an edge in space, and the speed of light belongs to the structure of spacetime, not to a fence. Physics describes how the world behaves; on its own it cannot rule on whether the world was made.
The defense accepted the point about force and kept the argument. That every dimension of the creature's reach carries the same hard stop, in separate theories that agree in form, is "best explained" by a created order grounded in what exceeds it. "This is inference to the best explanation. It proves nothing." The book even says why: the same gap between creature and Creator that the limits display keeps an argument from the limits from demonstrating what lies beyond them. The naturalist may keep six brute facts, but what he keeps is "a coincidence multiplying across every field of human inquiry."
Doesn't eternal inflation leave the universe unbounded?
Here's one of those attacks and how the Many Beings framework would address it.
A relativistic cosmologist brought current physics against Volume Six's claim that the universe, "examined at its extremes, is bounded on every side." The Borde-Guth-Vilenkin result marks only where inflation stops describing, not a wall. Alan Guth's summary of eternal inflation states, "Although inflation is generically eternal into the future, it is not eternal into the past," and speaks of "the infinity of pocket universes produced by eternal inflation." The concordance model, on the Planck satellite's 2018 results, runs on without end. If the cosmos may be endless in volume and duration, in what sense is it bounded on every side?
The defense points to the page. The chapter defines its sides at once: six walls, one on each axis along which a creature can strive, speed, smallness, cold, heat, density and structure. Size and age are not on the list. The chapter expressly rejects the reading the attack needs, the assumption "that a bounded universe is bounded the way a fenced yard is bounded, by more of the same beyond." Its own reading "is neither brute nor spatial." Of the multiverse, the book says: "a multiverse is not an escape from the box but a larger floor plan; the walls, being structural rather than geographic, come along to every bubble."
Does the effectiveness of mathematics point to God at all?
Here's one of those attacks and how the Many Beings framework would address it.
This attack comes from constructivist philosophy of mathematics, with Solomon Feferman its strongest voice and Hermann Weyl and Errett Bishop behind him. Volume Five argues that Wigner's gift is best explained by one Logos behind world and mind, and that the fit holds "because what is being read was never sized to its readers." Feferman showed that "surprisingly meager (in the proof-theoretical sense) predicatively justified systems suffice for the direct formalization of almost all, if not all, scientifically applicable mathematics," systems no stronger than ordinary arithmetic. The towers of large infinities have no physical use, so what physics reads is sized after all, and the fit lends no weight to an infinite Ground.
The defense is that the attack measures something else. "Never sized to its readers" means not fitted to the scale and purposes of the creatures reading it: selection tuned us to "fruit, predators, and middle-sized objects at walking speed," yet the mathematics works at the scale of quasars and quarks. Feferman measures the logical strength of the tools physics uses, and the book already separates tool from subject: "The objects are the frame's instruments; the structure is not." It concedes that the large infinities describe no creature, and it argues the Ground's actual infinity on separate premises, never from effectiveness. The claim stands as the book states it: inference to the best explanation, not demonstration.
Each of these defenses is set out at length in its own volume; the series lists them all.
Frequently asked questions
Is there scientific evidence for God?
Science offers evidence that bears on God without proving Him. Fine-tuning, a finite cosmic past and the effectiveness of mathematics are each best explained, on the theist's account, by a Creator. Each meets a standing reply, such as the multiverse or brute fact, so the scientific evidence for God is an inference to the best explanation.
What is the strongest argument for the existence of God?
Philosophers disagree. Classical theists often rank contingency highest, because it asks why anything exists at all and does not depend on open physics. The modal ontological argument is the most debated, since the debate turns on whether a maximal God is possible. Taken together, the arguments for God make a stronger case than any one alone.
Can God's existence be proven?
Aquinas held that it can be demonstrated from God's effects, and the argument from contingency claims deductive form. But every demonstration rests on premises a skeptic can dispute, so in practice the arguments work as evidence of God: reasons that make belief more reasonable than its denial, not a proof that compels every hearer.
Is the concept of God coherent?
Yes. The standard charges, from the stone paradox to modal collapse, each carry one term across the line between God's frame and ours as though it meant the same in both. Read in the right frame, each charge dissolves with no attribute trimmed. That defeats the claim that a maximal God is an impossible object.
What are Aquinas' five ways?
They are five arguments in the Summa Theologiae: from motion, efficient causation, possibility and necessity, degrees of perfection, and the ordering of natural things to an end. Each traces a chain of dependence to a first term, which Aquinas identifies as God.
Where to go from here
The question of evidence for God rests on an earlier one. If a maximal God is impossible, no watch, no beginning and no moral law can count in His favor. If He is possible, the six arguments regain their full force, and the cumulative case is strong. That is why the case starts with coherence, and why the series begins from the natures of God and creature rather than from any single argument. Volume One sets out that foundation from first premises, and it is yours free. The series carries the same method into evil, time, change, infinity and physics.
