Possible World¶
Represent a complete way things could be as an alternative point at which propositions can be evaluated.
Core Idea¶
A possible world is a complete alternative way things could be, or a point in a formal interpretation that represents such an alternative and at which statements can be evaluated. Instead of asking only whether a proposition is true here, possible-world reasoning asks at which alternatives it is true. The actual world, if designated, is one evaluation point among alternatives; actuality is not the criterion that makes another point possible.[1]
The term has two connected but distinct uses. Philosophers debate what a complete alternative is and whether worlds are concrete realities, abstract states of affairs, sets of propositions or something else. In modal semantics, a world is a model-theoretic index carrying a local valuation or predicate extension; a separate accessibility or similarity relation determines which worlds matter to an operator. A workable formal interpretation does not by itself settle the metaphysical ontology.[1][2]
A possible world is distinct from possible-world semantics, a framework that combines worlds with valuations, relations and truth clauses. The individual alternative and the framework using alternatives answer different questions.
How would you explain it like I'm…
Another Way Things Could Be
Other Ways Things Could Be
Complete Alternative Worlds
Structural Signature¶
Sig role-phrases:
- Alternative total circumstance — A way matters could stand as a whole, complete relative to a specified space of facts or model vocabulary; a single isolated hypothetical sentence is not a world.[1]
- World-indexed evaluation point — A location
wat whichp, predicates or formulas can be true or false, making truth relative to an alternative explicit.[1][2] - Possibility standard — The logical, metaphysical or model-specific constraints that determine whether this alternative belongs among possible worlds. The standard must be stated before contrasting possible and impossible worlds.[1]
- Local valuation or extension — In an interpreted model, assignments that determine atomic truth or predicate extensions at
w. Such assignments are part of the formal representation, not an agreed metaphysical composition of a world.[1][2] - Optional selection relation — Accessibility for necessity/knowledge, or comparative similarity for counterfactuals, links or orders worlds for a use. It is not an intrinsic ingredient of one world's identity.[1][2][3][4]
Condensed: specified possibility space → complete alternative → world-relative evaluation. Relations and modal operators act over these alternatives but are not themselves the alternatives.
What It Is Not¶
- Not any imagined scenario. “The light is red” alone leaves nearly everything unspecified; a world is a complete alternative, even when a finite formal model describes only what its language can distinguish.
- Not the whole possible-world semantics system. A set of worlds, valuations, accessibility and truth clauses is a model or semantic framework, not one world.[1][2]
- Not an impossible world under the same standard. The live Impossible World node allows selected possibility constraints to fail; merely being nonactual does not make a world impossible.
- Not necessarily a concrete parallel universe. Modal realism is one metaphysical interpretation, not a theorem of semantic usefulness.[1]
- Not an intension or relation. An intension can be a function from worlds to extensions; accessibility connects world points. Neither is an individual possible world.[1]
Scope of Application¶
In modal logic, possible worlds provide points for evaluating necessity and possibility under a specified frame. In epistemic logic they can stand for complete alternatives consistent with an agent's information, while an accessibility relation determines which alternatives count as live for that agent. In counterfactual semantics, a similarity ordering can select relevant antecedent worlds. These are uses of the same world-indexed alternative concept, but the selection rules and truth clauses differ.[1][2][5][3][4]
In metaphysics, questions shift from how to use alternatives in a model to what they are: concrete worlds, abstract representations or other entities. The present entry does not choose one account. A formal frame can also omit a specially designated actual world if its purpose is to study validity at arbitrary points; the metaphysical description and the technical model must not be conflated.[1]
Clarity¶
Worlds explain why a statement's actual truth value does not settle its modal profile. Two statements can both be true here yet differ in whether they remain true across relevant alternatives. The world parameter exposes that difference. The key question becomes which world is under evaluation, followed by which other worlds are relevant to this operator. Those questions are not interchangeable.[1]
The distinction also prevents a false metaphysical inference. A mathematical model can assign a proposition true at w without asserting that w is a causally connected, physically existing universe. Conversely, a philosophical claim that worlds are concrete needs an argument beyond the model's convenient indexing.[1]
Manages Complexity¶
Modal, epistemic and counterfactual discourse contain nested references to alternatives. Treating each complete alternative as a point lets a model store local truths and then reuse relations among points rather than restating every scenario in prose. p can be true at w1 and false at w2; a separate relation determines whether a necessity or knowledge claim at w1 must inspect w2.[1][2][5]
This economy has a representational limit. A finite model's valuation covers only its chosen atoms and rules. Calling its points “worlds” does not mean one has literally listed every fact about reality, nor that alternative ontologies give the same answers. Model completeness is relative to a language and constraints.[1]
Abstract Reasoning¶
First state a possibility standard and the proposition vocabulary. Then define alternative world points and what is true at each. Only after that choose a relation appropriate to a modality. If p is false at the actual point but true at an accessible alternative, p may be possible though not actual. If p holds at every relevant alternative, a necessity or knowledge claim may be true under the chosen truth clause; the formal result remains relative to that model and relation.[1][2][5]
Counterfactual evaluation asks a different question: among worlds where the antecedent holds, which are sufficiently similar or closest, and does the consequent hold there? A similarity ordering adds structure to worlds; it does not redefine each world as the ordering itself. If an alleged alternative violates the very logical constraint the model treats as possibility, it is not a possible world for that model and may require the distinct Impossible World notion.[1][3][4]
Knowledge Transfer¶
The idea of a truth-evaluable alternative transfers literally across alethic, epistemic and counterfactual formal settings. What changes is the relation or ordering and the interpretation of “could.” Information compatibility is not automatically metaphysical possibility, and similarity is not automatically an epistemic accessibility relation.[5][1][3][4]
Other disciplines may use scenario sets analogically, but those are not all possible worlds unless the required completeness and evaluation role is established. No prime status follows merely because alternatives appear in many domains; this named concept remains tied to philosophical and logical semantics.
Examples¶
A contingent modal alternative¶
A model distinguishes an actual world where a contingent proposition p is false from another coherent world w where p is true. Given an accessibility relation that reaches w, a possibility claim about p can hold at the actual point even though p is actually false. The exact claim depends on the model and its relation.[1][2]
Mapped back: alternative = complete relevant arrangement with p true; evaluation point = w; possibility standard = model's permitted alternatives; valuation = truth of p at w; optional relation = access from the designated point.
An agent's information-compatible world¶
An agent cannot distinguish two complete states compatible with their information. In one, a proposition q is true; in the other it is false. The second is a possible world of the epistemic model. If the agent's knowledge operator ranges over both, the agent does not know q, even if q happens to be true at the actual state.[5]
Mapped back: alternative = a complete compatible state; evaluation point = epistemic world u; possibility standard = retained information and model constraints; valuation = q false at u; optional relation = the agent's indistinguishability/accessibility.
Near miss: a contradictory world point¶
A nonclassical semantic framework can represent a world-like point where both r and not-r hold. Such a point violates a stipulated classical consistency standard and cannot be treated as an ordinary point with classical negation. It is an Impossible World in that framework, not merely a possible but nonactual alternative.
Structural Tensions¶
Semantic utility versus ontological commitment. World indices can make modal truth conditions tractable, but using them does not decide whether nonactual worlds are concrete or abstract. Treating every model point as a physical reality overcommits; dismissing the formal method until metaphysics is settled sacrifices analytic clarity. Diagnostic: is the assertion about a model's truth conditions or about independent existence?[1]
Complete alternatives versus tractable models. A philosophical world is maximally inclusive, yet a formal model records only distinctions expressible in its vocabulary. Requiring literal exhaustive description defeats practical modeling; treating a finite valuation as all facts overstates what it represents. Diagnostic: which propositions and constraints can this model distinguish, and which are left unrepresented?[1][2]
Structural–Framed Character¶
Possible World has a structural alternative-and-evaluation skeleton, but its name is philosophical-semantic-framed. The vocabulary travels among modal, epistemic and counterfactual theories, while export to ordinary planning imports the formal or metaphysical world concept rather than automatically recognizing the same entity. Its evaluation depends on human-selected languages, relations and possibility standards; its origin lies in a philosophical and logical research tradition, not in an institution's policy. Evaluative weight is low in the moral sense, though “possible” is relative to a normative choice of logical constraints. A broader representation of alternatives can resemble this pattern, though representing an alternative does not itself decide the metaphysical status of a possible world. Its character: structurally precise within formal use, yet substantially framed by modal semantics and metaphysical interpretation.
Structural Core vs. Domain Accent¶
The core relation is an alternative evaluation point carrying different local truths under a specified possibility standard. The domain accent is modal and philosophical: completeness of a way things could be, world-relative truth and disputed world ontology. A mere scenario list lacks that identity; a complete modal model adds relations and operators beyond one world. The named concept depends on modal possibility, semantic interpretation and philosophical ontology; the bare act of representing an alternative does not settle any of these.
Instantiates / Related Primes¶
Logical Possibility is a related criterion, not a world-object genus. Accessibility Relation connects worlds, rather than subsuming them. Normal Modal Logic is a system using world semantics, not a parent kind of world. Impossible World is a contrasting sibling whose possibility constraint is relaxed. Representation may cover some formal uses but does not necessarily subsume metaphysical possible worlds. These relationships distinguish criteria, relations, logical systems and individual world-like alternatives.
Neighborhood in Abstraction Space¶
Possible World sits in a sparse region of the domain-specific corpus (68th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Formal Models & Logical Foundations (37 abstractions)
Nearest neighbors
- Ranking Theory — 0.86
- Impossible World — 0.85
- Logical possibility — 0.83
- Marginal man theory — 0.83
- Formal Model — 0.83
Computed from structural-signature embeddings · 2026-10-31
Not to Be Confused With¶
Do not treat Possible-World Semantics as an alias for one Possible World. Do not infer that all possible worlds are physically instantiated, that an accessibility relation is built into each world, or that a nonactual world is thereby impossible. An intension is a world-to-extension function, not the world argument itself. Counterfactual similarity is a further ordering on alternatives, not the definition of an alternative.
References¶
[1] Menzel, Christopher, “Possible Worlds”, Stanford Encyclopedia of Philosophy, introduction and §§1–2; directly inspected philosophical and formal exposition. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v
[2] Kripke, Saul A. (1963). “Semantical Considerations on Modal Logic”, Acta Philosophica Fennica 16, 83–94. Original scan discoverable; individual source passages were not directly verified in this review. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j
[3] Starr, Willow (2019). “Counterfactuals”, Stanford Encyclopedia of Philosophy; winter 2025 revision by Starr and Alex Kocurek, §5; directly indexed explanatory account of Lewis–Stalnaker similarity semantics. registry ↩a ↩b ↩c ↩d
[4] Lewis, David (1973). Counterfactuals, Harvard University Press. Original scan discoverable; relevant passages were not directly verified in this review. registry ↩a ↩b ↩c ↩d
[5] Rendsvig, Rasmus, John Symons, and Yanjing Wang, “Epistemic Logic”, Stanford Encyclopedia of Philosophy, summer 2025 archive, §2.4; directly inspected accessible exposition, not an original 1960s epistemic-logic source. registry ↩a ↩b ↩c ↩d ↩e