Concept

Occam's Razor

Intro

Occam's razor is the rule that when two explanations account for the same evidence equally well, you should prefer the one that assumes less. If your car will not start and the battery is dead, "the battery is dead" is a better explanation than "the battery is dead and also a raccoon cut a wire." Both fit. One assumes less.

The razor is genuinely useful and genuinely reliable. It is also the single most misused principle in debates about God, and the misuse is almost always the same one: the four words that accounts for all the data get dropped.

The razor is not "prefer the simpler story." It is "prefer the simpler adequate story." A story that is simpler because it leaves things out has not won on parsimony; it has changed the subject. "There is no God" is a shorter sentence than "God exists," but sentence length is not what the razor measures. What it measures is how many things you have to assume, weighed against how much you have explained.

So when someone says "atheism is simpler," the reply is a question, not a counter-assertion: simpler at explaining what? Naturalism does not explain why there is a universe, why it runs on elegant mathematical laws, why those laws are finely balanced, why conscious minds exist, or why moral obligations bind. It leaves each of those as a brute fact, an unexplained given. Counting the entities, theism posits one. Counting the unexplained brute facts, naturalism posits many. That is the trade the razor is actually asking about, and the shorter word count does not settle it.

Quick reply line: "Occam's razor is not 'the simplest explanation wins.' It is 'the simplest explanation that accounts for all the data wins.' So before we count entities, let us agree on the data. If your account leaves the existence of the universe, its fine-tuning, consciousness and moral obligation as brute unexplained facts, you have not been economical. You have just stopped explaining earlier than I did."

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

In full

Occam's razor, also called the principle or law of parsimony, is a methodological rule of theory choice: among competing hypotheses with equal explanatory power, prefer the one with the fewest ontological commitments.

Three things about that formulation carry the whole weight, and each is routinely lost in debate:

1. Competing. The razor compares rivals answering the same question. It cannot adjudicate between hypotheses aimed at different explananda.
2. Equal explanatory power. The razor is a tiebreaker, applied only after the hypotheses have been shown to account for the evidence equally. It never licenses discarding evidence in order to reach a tie.
3. Ontological commitments. What gets counted is kinds of thing posited, not words used, not conceptual difficulty, not how many propositions the theory needs to state.

Historical note, and a correction worth having

The principle predates its namesake. Aristotle states a version in the Posterior Analytics: other things being equal, the demonstration deriving from fewer postulates is superior. Ptolemy and the medieval scholastics used it routinely.

William of Ockham (c. 1287-1347), the English Franciscan whose name it carries, never wrote the phrase usually attributed to him. Entia non sunt multiplicanda praeter necessitatem, "entities are not to be multiplied beyond necessity," first appears in the 1639 Scotus commentary of the Irish Franciscan John Punch, three centuries after Ockham's death. Ockham's own formulations are Numquam ponenda est pluralitas sine necessitate ("plurality is never to be posited without necessity") and Frustra fit per plura quod potest fieri per pauciora ("it is futile to do with more what can be done with fewer"). The label "Ockham's razor" itself is a nineteenth-century coinage.

This matters more than trivia. Ockham's use of the principle was theological, not anti-theological. He applied it inside a framework of robust classical theism, to trim superfluous metaphysical machinery such as real distinctions and free-floating universals, never to trim God. Ockham is also the source of the "soft facts" distinction that answers Theological Fatalism, and of the nominalism that shapes the debate over Universals. Anyone who deploys "Ockham's razor" as though the man had been an early skeptic has the history exactly backwards, and saying so in an exchange is worth doing, briefly and without gloating.

Newton restates it as Rule I of the Regulae Philosophandi in the Principia: admit no more causes of natural things than are both true and sufficient to explain the appearances. Note both conditions. True and sufficient. A cause that is economical but insufficient fails Newton's rule as badly as one that is superfluous.

What parsimony actually counts

A distinction from David Lewis governs most disputes here:

Positing a new kind of thing (a soul, a deity, a field, a universe-generating mechanism) is a real cost. Positing more instances of a kind already in your theory (another electron, another star) is nearly free. This is why "theism posits one more entity" is not the objection it sounds like: the question is not whether the entity is one or many but whether the kind is doing explanatory work that nothing already in the inventory can do.

The five standard misuses

1. Simplicity of statement mistaken for simplicity of ontology. "There is no God" is four words; "God exists" is two. Word count measures nothing. The razor counts what a theory must assume exists, not how briefly it can be phrased.

2. Applied across unequal explanatory scope. This is the dominant error. A theory that explains less is not thereby more parsimonious; it is less complete. Comparing on parsimony before comparing on adequacy inverts the order of operations. See Genesis ANE Myth Borrowing Objection Defeater for a worked case, where "the simplest explanation is borrowing" is deployed against a body of textual data that the borrowing thesis does not in fact account for.

3. Treated as evidence rather than as a tiebreaker. The razor never establishes a conclusion; it breaks a tie between conclusions already equally supported. "Occam's razor says naturalism" is not an argument, any more than "the coin toss says heads" is a reason to believe something.

4. Counting only the entities you dislike. A consistent razor must be run over your own commitments too. Anyone deploying it against God while holding an unobservable multiverse, brute necessary moral facts, or an unexplained fine-tuned law-generator is not applying a principle; he is applying a preference. See Special Pleading.

5. Confusing parsimony with familiarity. Novel entities feel expensive and familiar ones feel free, but the razor is indifferent to how used to something you are. "Matter" is not ontologically cheaper than "mind" because you thought of it first.

The honest limits

The razor is a heuristic, not a theorem, and overclaiming for it is a real vulnerability.

• Nature is under no obligation to be simple. Quantum mechanics, biochemistry and the taxonomy of particles are all baroque, and in each case the baroque theory is the true one. The razor tracks a defensible prior, not a guarantee.
• There is no single razor. Elliott Sober (Ockham's Razors, 2015) argues that "simplicity" is not one measurable property and that parsimony reasoning is only ever justified relative to background assumptions, whether those come from a model-selection framework such as AIC or from substantive theory. Granting this is better than resisting it, because it dissolves the atheist's version of the razor at the same time it constrains yours. Nobody gets to invoke "simplicity" as a standalone trump.
• The razor does not favour the position with fewer objects in the universe. It favours the theory with the better ratio of explained data to assumed ontology. Those come apart constantly.

Parsimony as a prior: the Bayesian form

The disciplined version of the razor is probabilistic. Simplicity functions as a criterion on the prior probability of a hypothesis, which then gets updated on the evidence. See Bayesian Probability.

Richard Swinburne (Simplicity as Evidence of Truth, 1997; The Existence of God, 2nd ed. 2004) presses the further claim that theism is not merely permitted by parsimony but favoured by it. His argument is that a hypothesis positing a single person with infinite power, knowledge and freedom is simpler than one positing a person with some particular finite degree of each, because infinite values are non-arbitrary while any finite value invites the question "why that value?" On this reading classical theism is the simplest personal explanation available, and the fine-tuned constants are exactly the arbitrary finite quantities a naturalist must accept as brute.

The argument is contested and should be presented as such. It is a strong prima facie case that simplicity does not automatically run against theism, not a proof that it runs for it.

Where the razor cuts for the Christian case

Run correctly, parsimony is more often a friend than an opponent:

• Against the multiverse. Postulating 10^500 or more unobservable universes, plus a universe-generating mechanism that must itself be finely tuned to produce them, in order to avoid a single designer, is a large ontological bill for a modest explanatory return. See Multiverse and Fine-Tuning Argument.
• In the resurrection case. The resurrection hypothesis explains all the accepted historical facts with one cause; the standard naturalistic alternatives require conjoining several independent hypotheses (hallucination and misidentified tomb and independent conversion of a hostile witness), and conjunction is exactly what the razor penalises. See Minimal Facts Argument.
• In philosophy of mind. Whether the soul is an extravagance depends on whether the physicalist account is adequate. If it is not, positing a further kind of thing is not extravagance; it is the minimum sufficient ontology. See Substance Dualism and Argument from Personal Identity.
• Against conspiracy-shaped and syncretic theses. A thesis that keeps the supernatural furniture while adding new machinery on top of it has multiplied entities, not reduced them. See Jesus Is an Egregore Objection Defeater.
• In the demonic-activity debate. Parsimony favours the simplest adequate explanation, so a naturalistic account that handles most cases does not thereby handle all of them, and the residue is what the argument is about. See Demonic Activity is Just Medical Conditions Defeater.

Where the atheist deploys it, and how to answer

The standard deployment is: naturalism gets by with matter, energy and law; theism adds a further entity; therefore parsimony favours naturalism. Three replies, in order of force.

1. Contest the tie before contesting the count. The razor applies only between hypotheses of equal explanatory power. Establish first that naturalism accounts for the existence of a contingent universe, the intelligibility of the world, the fine-tuning of the constants, the arising of consciousness and the bindingness of moral obligation. If it leaves any of these as brute, there is no tie, and the razor never engages.

2. Count brute facts, not just entities. Every unexplained given is an item on the bill. Naturalism's inventory typically includes the universe itself, the laws, the constants, the initial low-entropy condition, the emergence of consciousness and either the reality or the illusion of moral obligation. Theism posits one entity from which those follow. Whether that inference succeeds is the debate; that it is the more economical shape of explanation is not seriously arguable.

3. Ask whether the razor is being run on both sides. Parsimony that is never turned on one's own commitments is a rhetorical device. See Ad-Hoc Rescue.

Compact form for live use: "You have counted my entities and not your brute facts. Add them up. I have one thing I cannot explain further. How many do you have?"

See also

• Epistemology, the wider theory-of-knowledge frame parsimony sits inside
• Bayesian Probability, the formal setting where simplicity operates as a prior
• Debate Logic Toolkit, the razor in its compact debating form alongside the other standard moves
• Special Pleading, the fallacy committed by a razor applied in one direction only
• Ad-Hoc Rescue, the failure mode parsimony is meant to catch
• Multiverse, the clearest live case of parsimony cutting against the naturalistic option
• Fine-Tuning Argument, where the design-versus-multiverse economy comparison is made
• Minimal Facts Argument, parsimony applied to competing historical hypotheses
• Substance Dualism, the razor as the leading objection and why it does not land
• Naturalism, the position most often defended by appeal to parsimony
• Universals, where Ockham's own nominalist use of the principle operates
• Theological Fatalism, where Ockham's soft-facts distinction does the work
• Fallacies, the wider catalogue of reasoning failures