pdf version of the entry

Possible Worlds
http://plato.stanford.edu/archives/sum2015/entries/possible-worlds/ Possible Worlds
from the Summer 2015 Edition of the First published Fri Oct 18, 2013; substantive revision Mon Oct 21, 2013

Stanford Encyclopedia Anne is working at her desk. While she is directly aware only of her
immediate situation — her being seated in front of her computer, the
of Philosophy music playing in the background, the sound of her husband's voice on the
phone in the next room, and so on — she is quite certain that this situation
is only part of a series of increasingly more inclusive, albeit less
immediate, situations: the situation in her house as a whole, the one in her
neighborhood, the city she lives in, the state, the North American
continent, the Earth, the solar system, the galaxy, and so on. On the face of
Edward N. Zalta Uri Nodelman Colin Allen R. Lanier Anderson
it, anyway, it seems quite reasonable to believe that this series has a limit,
Principal Editor Senior Editor Associate Editor Faculty Sponsor
that is, that there is a maximally inclusive situation encompassing all
Editorial Board
http://plato.stanford.edu/board.html others: things, as a whole or, more succinctly, the actual world.

Library of Congress Catalog Data Most of us also believe that things, as a whole, needn't have been just as
ISSN: 1095-5054
they are. Rather, things might have been different in countless ways, both
Notice: This PDF version was distributed by request to mem- trivial and profound. History, from the very beginning, could have
bers of the Friends of the SEP Society and by courtesy to SEP unfolded quite other than it did in fact: The matter constituting a distant
content contributors. It is solely for their fair use. Unauthorized star might never have organized well enough to give light; species that
distribution is prohibited. To learn how to join the Friends of the survived might just as well have died off; battles won might have been
SEP Society and obtain authorized PDF versions of SEP entries, lost; children born might never have been conceived and children never
please visit https://leibniz.stanford.edu/friends/ . conceived might otherwise have been born. In any case, no matter how
things had gone they would still have been part of a single, maximally
Stanford Encyclopedia of Philosophy
Copyright c 2015 by the publisher inclusive, all-encompassing situation, a single world. Intuitively, then, the
The Metaphysics Research Lab actual world of which Anne's immediate situation is a part is only one
Center for the Study of Language and Information
among many possible worlds.
Stanford University, Stanford, CA 94305
Possible Worlds The idea of possible worlds is evocative and appealing. However, possible
Copyright c 2015 by the author
Christopher Menzel worlds failed to gain any real traction among philosophers until the 1960s
All rights reserved.
when they were invoked to provide the conceptual underpinnings of some
Copyright policy: https://leibniz.stanford.edu/friends/info/copyright/

1
 Possible Worlds Christopher Menzel

powerful developments in modal logic. Only then did questions of their 2.3.5 Analytic and Emergent Modalities; Essential
nature become a matter of the highest philosophical importance. Properties
Accordingly, Part 1 of this article will provide an overview of the role of 2.3.6 Fewer Things and Other Things: Modified
possible worlds in the development of modal logic. Part 2 explores three Combinatorialism
prominent philosophical approaches to the nature of possible worlds.[1] 2.3.7 A Brief Assessment of Combinatorialism
Although many of the finer philosophical points of Part 2 do presuppose Bibliography
the technical background of Part 1, the general philosophical landscape Academic Tools
laid out in Part 2 can be appreciated independently of Part 1. Other Internet Resources
Related Entries
1. Possible Worlds and Modal Logic
1.1 Extensionality Lost
1.2 Extensionality Regained 1. Possible Worlds and Modal Logic
1.3 Two Applications: The Analysis of Intensions and the De Re
/ De Dicto Distinction Although ‘possible world’ has been part of the philosophical lexicon at
2. Three Philosophical Conceptions of Possible Worlds least since Leibniz, the notion became firmly entrenched in contemporary
2.1 Concretism philosophy with the development of possible world semantics for the
2.1.1 Concrete Worlds and Existence Therein languages of modal logic. In addition to the usual Boolean sentence
2.1.2 Actuality operators of classical logic such as ‘and’ (‘∧’), ‘or’ (‘∨’), ‘not’ (‘¬’), and
2.1.3 Modal Reductionism, Counterparts, and the Analysis ‘if...then’ (‘→’), these languages contain operators intended to represent
of Intensions the modal adverbs ‘necessarily’ (‘□’) and ‘possibly’ (‘◇’). Although a
2.1.4 Plenitude and Recombination prominent aspect of logic in both Aristotle's work and the work of many
2.1.5 A Brief Assessment of Concretism medieval philosophers, modal logic was largely ignored from the modern
2.2 Abstractionism period to the mid-20th century. And even though a variety of modal
2.2.1 Abstract Possible Worlds and Existence Therein deductive systems had in fact been rigorously developed in the early 20th
2.2.2 Irreducible Modality and Intensional Entities century, notably by Lewis and Langford (1932), there was for the
2.2.3 Actuality and Actualism languages of those systems nothing comparable to the elegant semantics
2.2.4 A Brief Assessment of Abstractionism that Tarski had provided for the languages of classical first-order predicate
2.3 Combinatorialism logic. Consequently, there was no rigorous account of what it means for a
2.3.1 The Basic Ontology of Combinatorialism sentence in those languages to be true and, hence, no account of the
2.3.2 States of Affairs and Recombination critical semantic notions of validity and logical consequence to underwrite
2.3.3 Structural States of Affairs and Supervenience the corresponding deductive notions of theoremhood and provability. A
2.3.4 Combinatorial Possible Worlds and Existence Therein concomitant philosophical casualty of this void in modal logic was a deep

2 Stanford Encyclopedia of Philosophy Summer 2015 Edition 3

 Possible Worlds Christopher Menzel

skepticism, voiced most prominently by Quine, toward any appeal to substitutivity principle is typically rendered invalid.
modal notions in metaphysics generally, notably, the notion of an essential
property. (See Quine 1953 and 1956, and the appendix to Plantinga 1974.) Extensionality is a well known and generally cherished feature of classical
The purpose of the following two subsections is to provide a simple and propositional and predicate logic. Modal logic, by contrast, is intensional.
largely ahistorical overview of how possible world semantics fills this To illustrate: the substitutivity principle for sentences tells us that
void; the final subsection presents two important applications of the sentences with the same truth value can be substituted for one another
semantics. (Readers familiar with basic possible world semantics can skip salva veritate. So suppose that John's only pets are two dogs, Algol and
to §2 with no significant loss of continuity.) BASIC, say, and consider two simple sentences and their formalizations
(the predicates in question indicating the obvious English counterparts):
1.1 Extensionality Lost
(1) All John's dogs are mammals: ∀x(Dx → Mx).
Since the middle ages at least, philosophers have recognized a semantical
(2) All John's pets are mammals: ∀x(Px → Mx)
distinction between extension and intension. The extension of a denoting
expression, or term, such as a name or a definite description is its referent,
As both sentences are true, they have the same extension. Hence, in
the thing that it refers to; the extension of a predicate is the set of things it
accordance with the classical substitutivity principle for sentences, we can
applies to; and the extension of a sentence is its truth value. By contrast,
replace the occurrence of (1) with (2) in the false sentence
the intension of an expression is something rather less definite — its
sense, or meaning, the semantical aspect of the expression that determines (3) Not all John's dogs are mammals: ¬∀x(Dx → Mx)
its extension. For purposes here, let us say that a logic is a formal language
together with a semantic theory for the language, that is, a theory that and the result is the equally false sentence
provides rigorous definitions of truth, validity, and logical consequence
for the language.[2] A logic is extensional if the truth value of every (4) Not all John's pets are mammals: ¬∀x(Px → Mx).
sentence of the logic is determined entirely by its form and the extensions
of its component sentences, predicates, and terms. An extensional logic However, when we make the same substitution in the true sentence
will thus typically feature a variety of valid substitutivity principles. A
(5) Necessarily, all John's dogs are mammals: □∀x(Dx → Mx),
substitutivity principle says that, if two expressions are coextensional, that
is, if they have the same extension, then (subject perhaps to some the result is the sentence
reasonable conditions) either can be substituted for the other in any
sentence salva veritate, that is, without altering the original sentence's (6) Necessarily, all John's pets are mammals: □∀x(Px → Mx),
truth value. In an intensional logic, the truth values of some sentences are
determined by something over and above their forms and the extensions of which is intuitively false, as John surely could have had a non-mammalian
their components and, as a consequence, at least one classical pet. In a modal logic that accurately represents the logic of the necessity

4 Stanford Encyclopedia of Philosophy Summer 2015 Edition 5

 Possible Worlds Christopher Menzel

operator, therefore, the substitutivity principle for sentences will have to Given this, the failures of the classical substitutivity principles can be
fail. traced to the fact that modal operators, so interpreted, introduce contexts
that require subtler notions of meaning for sentences and their component
The same example illustrates that the substitutivity principle for predicates parts than are provided in classical logic; in particular, a subtler notion (to
will have to fail in modal logic as well. For, according to our example, the be clarified shortly) is required for predicates than that of the set of things
predicates ‘D’ and ‘P’ that are true of John's dogs and of John's pets, they happen to apply to.
respectively, are coextensional, i.e., ∀x(Dx ↔ Px). However, while
substituting the latter predicate for the former in (3) results in a sentence Tarskian Semantics. Standard model theoretic semantics for the
with the same truth value, the same substitution in (5) does not. languages of predicate logic deriving from the work of Tarski (1933,
1944) is the paradigmatic semantic theory for extensional logics. Given a
Modal logic, therefore, is intensional: in general, the truth value of a standard first-order language ℒ and a set D for the quantifiers of ℒ to
sentence is determined by something over and above its form and the range over (typically, some set of things that ℒ has been designed to
extensions of its components. Absent a rigorous semantic theory to describe), a Tarskian interpretation M for ℒ assigns, to each term
identify the source of its intensionality and to systematize intuitions about (constant or variable) τ of ℒ, a referent aτ ∈ D and, to each n-place
modal truth, validity, and logical consequence, there was little hope for the predicate π of ℒ, an appropriate extension Eπ — a truth value (TRUE or
widespread acceptance of modal logic. FALSE) if n = 0, a subset of D if n = 1, and a set of n-tuples of members
of D if n > 1. Given these assignments, sentences are evaluated as true
1.2 Extensionality Regained
under the interpretation M — trueM, for short — according to a more or
The idea of possible worlds raised the prospect of extensional less familiar set of clauses. To facilitate the definition, let M[ν/a] be the
respectability for modal logic, not by rendering modal logic itself interpretation that assigns the individual a to the variable ν and is
extensional, but by endowing it with an extensional semantic theory — otherwise exactly like M. Then we have:
one whose own logical foundation is that of classical predicate logic and,
An atomic sentence ⌈πτ1...τn⌉ (of ℒ) is trueM if and only if
hence, one on which possibility and necessity can ultimately be
n = 0 (i.e., π is a sentence letter) and the extension of π is the
understood along classical Tarskian lines. Specifically, in possible world
truth value TRUE; or
semantics, the modal operators are interpreted as quantifiers over possible
n = 1 and aτ1 is in the extension of π; or
worlds, as expressed informally in the following two general principles:
n > 1 and ⟨aτ1, ..., aτn⟩ is in the extension of π.
Nec A sentence of the form ⌈Necessarily, φ⌉ (⌈◻φ⌉) is true if and A negation ⌈¬ψ⌉ is trueM if and only if ψ is not trueM.
only if φ is true in every possible world.[3] A material conditional⌈ψ → θ⌉ is trueM iff, if ψ is trueM, then θ is
trueM.
Poss A sentence of the form ⌈Possibly, φ⌉ (⌈◇φ⌉) is true if and only A universally quantified sentence ⌈∀νψ⌉ is trueM if and only if, for
if φ is true in some possible world. all individuals a ∈ D, ψ is trueM[ν/a].[4]

6 Stanford Encyclopedia of Philosophy Summer 2015 Edition 7

 Possible Worlds Christopher Menzel

Clauses for the other standard Boolean operators and the existential quantification and the extension of the predicate ‘P’ could, in some sense
quantifier under their usual definitions follow straightaway from these or other, have been different.
clauses. In particular, where
Possible world semantics, of course, uses the concept of a possible world
(7) ∃νφ =def ¬∀ν¬φ to give substance to the idea of alternative extensions and alternative
domains of quantification. (Possible world semantics can be traced most
it follows that: clearly back to the work of Carnap (1947), its basic development
culminating in the work of Hintikka (1957, 1961), Bayart (1958, 1959),
An existentially quantified sentence ⌈∃νψ⌉ is is trueM if and only if,
and Kripke (1959, 1963a, 1963b).[5]) As in Tarskian semantics, a possible
for some individual a ∈ D, ψ is trueM[ν/a].
world interpretation M of a modal language ℒ starts with nonempty set D,
It is easy to verify that, in each of the above cases, replacing one although thought of now as the set of “possible individuals” of M. Also as
coextensional term, predicate, or sentence for another has no effect on the in Tarskian semantics, M assigns each term τ of ℒ a referent aτ in D.[6]
truth values rendered by the above clauses, thus guaranteeing the validity Additionally however, M contains a set W, the set of “possible worlds” of
of the classical substitutivity principles and, hence, the extensionality of M, one of which is designated its “actual world”, and each world w in W
first-order logic with a Tarskian semantics. is assigned its own domain of quantification, d(w) ⊆ D, intuitively, the set
of individuals that exist in w.[7] To capture the idea of both the actual and
From Tarskian to Possible World Semantics. The truth conditional possible extensions of a predicate, I assigns to each n-place predicate π a
clauses for the three logical operators directly reflect the meanings of the function Iπ — the intension of π — that, for each possible world w,
natural language expressions they symbolize: ‘¬’ means not; ‘→’ means returns the extension Iπ(w) of π at w: a truth value, if n = 0; a set of
if...then; ‘∀’ means all. It is easy to see, however, that we cannot expect to individuals, if n = 1; and a set of n-tuples of individuals, if n > 1.[8] We
add an equally simple clause for sentences containing an operator that can thus rigorously define a “possible extension” of a predicate π to be
symbolizes necessity. For a Tarskian interpretation fixes the domain of any of its w-extensions Iπ(w), for any world w.
quantification and the extensions of all the predicates. Pretty clearly,
however, to capture necessity and possibility, one must be able to consider The Tarskian truth conditions above are now generalized by relativizing
alternative “possible” domains of quantification and alternative “possible” them to worlds as follows: for any possible world w (the world of
extensions for predicates as well. For, intuitively, under different evaluation):
circumstances, fewer, more, or other things might have existed and things
An atomic sentence ⌈πτ1...τn⌉ (of ℒ) is trueM at w if and only if:
that actually exist might, in those circumstances, have had very different
n = 0 and the w-extension of π is the truth value TRUE; or
properties. (6), for example, is false because John could have had non-
n = 1 and aτ1 is in the w-extension of π; or
mammalian pets: a canary, say, or a turtle, or, under very different
n > 1 and ⟨aτ1,..., aτn⟩ is in the w-extension of π.
circumstances, a dragon. A bit more formally put: Both the domain of
A negation ⌈¬ψ⌉ is trueM at w if and only ψ is not trueM in w.

8 Stanford Encyclopedia of Philosophy Summer 2015 Edition 9

 Possible Worlds Christopher Menzel

A material conditional⌈ψ→θ⌉ is trueM at w iff, if ψ is trueM at w, the exact form of our informal principle Nec.
then θ is trueM at w.
Call the above basic possible world semantics. Spelling out the truth
A quantified sentence ⌈∀νψ⌉ is trueM at w if and only if, for all
conditions for (6) (relative to the intended interpretation of its language),
individuals a that exist in w, ψ is trueM[ν/a].
basic possible world semantics tells us that (6) is true if and only if
And to these, of course, is added the critical modal case that explicitly
interprets the modal operator to be a quantifier over worlds, as we'd (8) For all possible worlds w, ‘∀x(Px → Mx)’ is true at w.
initially anticipated informally in our principle Nec:
And by unpacking (8) in terms of the quantificational, material
A necessitation ⌈◻ψ⌉ is trueM at w if and only if, for all possible conditional, and atomic clauses above we have that (6) is true if and only
worlds u of M, ψ is trueM at u.[9] if

A sentence φ is falseM at w just in case it is not trueM at w, and φ is said (9) For all possible worlds w, and for all possible individuals a that
to be trueM just in case φ is trueM at the actual world of M. exist in w, if a is in the w-extension of ‘P’ then a is in the w-
extension of ‘M’.
On the assumption that there is a (nonempty) set of all possible worlds and
a set of all possible individuals, we can define “objective” notions of truth Since we are evaluating (6) with regard to the intended interpretation of its
at a world and of truth simpliciter, that is, notions that are not simply language, the w-extension of ‘P’ that is returned by its intension, for any
relative to formal, mathematical interpretations but, rather, correspond to world w, is the (perhaps empty) set of John's pets in w and that of ‘M’ is
objective reality in all its modal glory. Let ℒ be a modal language whose the set of mammals in w. Hence, if w is a world where John has a pet
names and predicates represent those in some fragment of ordinary canary — COBOL, say — COBOL is in the w-extension of ‘P’ but not
language (as in our examples (5) and (6) above). Say that M is the that of ‘M’ , i.e., ‘∀x(Px → Mx)’ is false at w and, hence, by the truth
“intended” interpretation of ℒ if (i) its set W of “possible worlds” is in condition (9), (6) is false at the actual world — that is, (6) is false
fact the set of all possible worlds, (ii) its designated “actual world” is in simpliciter, as it should be.
fact the actual world, (iii) its set D of “possible individuals” is in fact the
set of all possible individuals, and (iv) the referents assigned to the names Note that interpreting modal operators as quantifiers over possible worlds
of ℒ and the intensions assigned to the predicates of ℒ are the ones they in provides a nice theoretical justification for the usual definition of the
fact have. Then, where M is the intended interpretation of ℒ, we can say possibility operator in terms of necessity, specifically:
that a sentence φ of ℒ is true at a possible world w just in case φ is trueM
(10) ⌈◇φ⌉ =def ⌈¬◻¬φ⌉.
at w, and that φ is true just in case it is trueM at the actual world. (Falsity
at w and falsity, simpliciter, are defined accordingly.) Under the
That is, a sentence is possible just in case its negation isn't necessary.
assumption in question, then, the modal clause above takes on pretty much
Since, semantically speaking, the necessity operator is literally a universal

10 Stanford Encyclopedia of Philosophy Summer 2015 Edition 11

 Possible Worlds Christopher Menzel

quantifier, the definition corresponds exactly to the definition (7) of the sort of mismatch between the surface syntax of those operators and their
existential quantifier. For, unpacking the right side of definition (10) semantics: syntactically, they are unary sentence operators like negation;
according to the negation and necessitation clauses above (and invoking but semantically, they are, quite literally, quantifiers. Their syntactic
the definitions of truth and truth at a world simpliciter), we have: similarity to negation suggests that, like negation, the truth values of
⌈□φ⌉ and ⌈◇φ⌉, insofar as they are determinable at all, must be
(11) ⌈◇φ⌉ is true iff it is not the case that, for all possible worlds w, determined by the truth value of φ. That they are not (in general) so
φ is not true at w. determined leads to the distinctive substitutivity failures noted above. The
possible worlds analysis of the modal operators as quantifiers over worlds
Clearly, however, if it is not the case that φ fails to be true at all possible
reveals that the unary syntactic form of the modal operators obscures a
worlds, then it must be true at some world; hence:
semantically relevant parameter. When the modal operators are interpreted
as quantifiers, this parameter becomes explicit and the reason underlying
(12) ⌈◇φ⌉ is true iff, for some possible world w, φ is true at w.
the failure of extensionality in modal logic becomes clear: That the truth
And that corresponds exactly to our intuitive truth condition Poss above. values of ⌈□φ⌉ and ⌈◇φ⌉ are not in general determined by the truth value
Thus, spelling out the negation ‘¬□∀x(Px → Mx)’ of our false sentence (6) of φ at the world of evaluation is, semantically speaking, nothing more
above in accordance with definition (10) (and the standard definition of than the fact that the truth values of ‘∀xFx’ and ‘∃xFx’ are not in general
conjunction ∧), we have: determined by the truth value of ‘Fx’, for any particular value of ‘x’.
Possible world semantics, therefore, explains the intensionality of modal
(13) Possibly, one of John's pets is not a mammal: ◇∃x(Px ∧ ¬Mx), logic by revealing that the syntax of the modal operators prevents an
adequate expression of the meanings of the sentences in which they occur.
for which (12) and the possible world truth conditions for quantified, Spelled out as possible world truth conditions, those meanings can be
Boolean, and atomic sentences yield the correct truth condition: expressed in a wholly extensional fashion. (For a more formal exposition
of this point, see the supplemental article The Extensionality of Possible
(14) There is a possible world w and an individual a existing in w that World Semantics.)
is in the w-extension of ‘P’ but not that of ‘M’,
1.3 Two Applications: The Analysis of Intensions and the De Re
that is, less stuffily, there is a possible world in which, among John's pets, / De Dicto Distinction
at least one is not a mammal.
As noted, the focus of the present article is on the metaphysics of possible
Summary: Intensionality and Possible Worlds. Analyzed in terms of worlds rather than applications. Of course, the semantics of modal
possible world semantics, then, the general failure of classical languages is itself an application, but one that is of singular importance,
substitutivity principles in modal logic is due, not to an irreducibly both for historical reasons and because most applications are in fact
intensional element in the meanings of the modal operators, but rather to a themselves applications of (often extended or modified versions of) the

12 Stanford Encyclopedia of Philosophy Summer 2015 Edition 13

 Possible Worlds Christopher Menzel

semantical apparatus. Two particularly important examples are the not follow that intensions cannot be analyzed in terms of possible worlds,
analysis of intensions and a concomitant explication of the de re/de dicto but only that more subtle constructions might be required. This reply
distinction.[10] appears to side-step the objections from granularity while preserving the
great advantage of the possible worlds analysis of intensions, viz., the
The Analysis of Intensions. As much a barrier to the acceptance of modal rigorous definability of these philosophically significant notions.
logic as intensionality itself was the need to appeal to intensions per se —
properties, relations, propositions, and the like — in semantical The De Re / De Dicto Distinction. A particularly rich application of the
explanations. Intensional entities have of course featured prominently in possible world analysis of intensions concerns the analysis of the
the history of philosophy since Plato and, in particular, have played venerable distinction between de re and de dicto modality.[11] Among the
natural explanatory roles in the analysis of intentional attitudes like belief strongest modal intuitions is that the possession of a property has a modal
and mental content. For all their prominence and importance, however, the character — that things exemplify, or fail to exemplify, some properties
nature of these entities has often been obscure and controversial and, necessarily, or essentially, and others only accidentally. Thus, for
indeed, as a consequence, they were easily dismissed as ill-understood and example, intuitively, John's dog Algol is a pet accidentally; under less
metaphysically suspect “creatures of darkness” (Quine 1956, 180) by the fortunate circumstances, she might have been, say, a stray that no one ever
naturalistically oriented philosophers of the early- to mid-20th century. It adopted. But she is a dog essentially; she couldn't have been a flower, a
is a virtue of possible world semantics that it yields rigorous definitions for musical performance, a crocodile or any other kind of thing.
intensional entities. More specifically, as described above, possible world
semantics assigns to each n-place predicate π a certain function Iπ — π's Spelling out this understanding in terms of worlds and the preceding
intension — that, for each possible world w, returns the extension Iπ(w) of analysis of intensions, we can say that an individual a has a property F
π at w. We can define an intension per se, independent of any language, to essentially if a has F in every world in which it exists, that is, if, for all
be any such function on worlds. More specifically: worlds w in which a exists, a ∈ F(w). Likewise, a has F accidentally if a
has F in the actual world @ but lacks it in some other world, that is, if a ∈
A proposition is any function from worlds to truth values. F(@) but, for some world w in which a exists, a ∉ F(w). Thus, let ‘G’ and
A property is any function from worlds to sets of individuals. ‘T’ symbolize ‘is a dog’ and ‘is someone's pet’, respectively; then, where
An n-place relation (n > 1) is any function from worlds to sets of n- ‘E!x’ is short for ‘∃y(x=y)’ (and, hence, expresses that x exists), we have:
tuples of individuals.
(15) Algol is a dog essentially: □(E!a → Ga)
The adequacy of this analysis is a matter of lively debate that focuses
chiefly upon whether or not intensions, so defined, are too “coarse- (16) Algol is a pet accidentally: Ta ∧ ◇(E!a ∧ ¬Ta)
grained” to serve their intended purposes. (See, e.g., Stalnaker 1987 and
More generally, sentences like (15) and (16) in which properties are
2012 for a strong defense of the analysis.) However, Lewis (1986, §1.5)
ascribed to a specific individual in a modal context — signaled formally
argues that, even if the above analysis fails for certain purposes, it does
by the occurrence of a name or the free occurrence of a variable in the

14 Stanford Encyclopedia of Philosophy Summer 2015 Edition 15

 Possible Worlds Christopher Menzel

scope of a modal operator — are said to exhibit modality de re[12]

(modality of the thing). Modal sentences that do not, like QW What, exactly, is a possible world?

(17) Necessarily, all dogs are mammals: □∀x(Gx → Mx) And, given QW:

are said to exhibit modality de dicto (roughly, modality of the QE What is it for something to exist in a possible world?
proposition). Possible world semantics provides an illuminating analysis
In this section we will concern ourselves with, broadly speaking, the three
of the key difference between the two: The truth conditions for both
most prominent philosophical approaches to these questions.[13]
modalities involve a commitment to possible worlds; however, the truth
conditions for sentences exhibiting modality de re involve in addition a 2.1 Concretism
commitment to the meaningfulness of transworld identity, the thesis that,
necessarily, every individual (typically, at any rate) exists and exemplifies Recall the informal picture that we began with: a world is, so to say, the
(often very different) properties in many different possible worlds. More “limit” of a series of increasingly more inclusive situations. Fleshed out
specifically, basic possible world semantics yields intuitively correct truth philosophical accounts of this informal idea generally spring from rather
values for sentences of the latter sort by (i) permitting world domains to different intuitions about what one takes the “situations” in the informal
overlap and (ii) assigning intensions to predicates, thereby, in effect, picture to be. A particularly powerful intuition is that situations are simply
relativizing predicate extensions to worlds. In this way, one and the same structured collections of physical objects: the immediate situation of our
individual can be in the extension of a given predicate at all worlds in initial example above, for instance, consists of, among other things, the
which they exist, at some such worlds only, or at none at all. (For further objects in Anne's office — notably Anne herself, her desk and her
discussion, see the entry on essential vs. accidental properties.) computer, with her seated at the former and typing on the latter — and at
least some of the things in the next room — notably, her husband and the
2. Three Philosophical Conceptions of Possible phone he is talking on. On this view, for one situation s to include another
Worlds r is simply for r to be a (perhaps rather complex and distributed) physical
part of s. The actual world, then, as the limit of a series of increasingly
The power and appeal of basic possible world semantics is undeniable. In more inclusive situations in this sense, is simply the entire physical
addition to providing a clear, extensional formal semantics for a formerly universe: all the things that are some spatiotemporal distance from the
somewhat opaque, intensional notion, cashing possibility as truth in some objects in some arbitrary initial situation, structured as they in fact are; and
possible world and necessity as truth in every such world seems to tap into other possible worlds are things of exactly the same sort. Call this the
very deep intuitions about the nature of modality and the meaning of our concretist intuition, as possible worlds are understood to be concrete
modal discourse. Unfortunately, the semantics leaves the most interesting physical situations of a special sort.
— and difficult — philosophical questions largely unanswered. Two arise
with particular force: 2.1.1 Concrete Worlds and Existence Therein

16 Stanford Encyclopedia of Philosophy Summer 2015 Edition 17

 Possible Worlds Christopher Menzel

The originator and, by far, the best known proponent of concretism is other,[14] and that a is maximal if none of its parts is spatiotemporally
David Lewis. For Lewis and, as noted, concretists generally, the actual related to anything that is not also one of its parts. Then we have the
world is the concrete physical universe as it is, stretched out in space-time. following concretist answers to our questions:
As he rather poetically expresses it (1986, 1):
AW1 w is a possible world =def w is a maximal connected object.[15]
The world we live in is a very inclusive thing....There is nothing so
far away from us as not to be part of our world. Anything at any And, hence, to exist in a world is simply to be a part of it:
distance is to be included. Likewise the world is inclusive in time.
No long-gone ancient Romans, no long-gone pterodactyls, no long- AE1 Individual a exists in world w =def a is a part of w.
gone primordial clouds of plasma are too far in the past, nor are the
dead dark stars too far in the future, to be part of this same world.... It follows from AW1 (and reasonable assumptions) that distinct worlds do
[N]othing is so alien in kind as not to be part of our world, not overlap, spatiotemporally; that no spatiotemporal part of one world is
provided only that it does exist at some distance and direction from part of another.[16] Moreover, given Lewis's counterfactual analysis of
here, or at some time before or after or simultaneous with now. causation, it follows from this that objects in distinct worlds bear no causal
relations to one another; nothing that occurs in one world has any causal
The actual world provides us with our most salient example of what a impact on anything that occurs in any other world.
possible world is. But, for the concretist, other possible worlds are no
different in kind from the actual world (ibid., 2): 2.1.2 Actuality

There are countless other worlds, other very inclusive things. Our Critically, for Lewis, worlds and their denizens do not differ in the manner
world consists of us and all our surroundings, however, remote in in which they exist. The actual world does not enjoy a kind of privileged
time and space; just as it is one big thing having lesser things as existence that sets it apart from other worlds. Rather, what makes the
parts, so likewise do other worlds have lesser other-worldly things actual world actual is simply that it is our world, the world that we happen
as parts. to inhabit. Other worlds and their inhabitants exist just as robustly as we
do, and in precisely the same sense; all worlds and all of their denizens are
It is clear that spatiotemporal relations play a critical role in Lewis's equally real.[17] A significant semantic corollary of this thesis for Lewis is
conception. However, it is important to note that Lewis understands such that the word ‘actual’ in the phrase ‘the actual world’ does not indicate
relations in a very broad and flexible way so as to allow, in particular, for any special property of the actual world that distinguishes it from all other
the possibility of spirits and other entities that are typically thought of as worlds; likewise, an assertion of the form ‘a is actual’ does not indicate
non-spatial; so long as they are located in time, Lewis writes, “that is good any special property of the individual a that distinguishes it from the
enough” (ibid., 73). So with this caveat, let us say that that an object a is objects existing in other worlds. Rather, ‘actual’ is simply an indexical
connected if any two of its parts bear some spatiotemporal relation to each whose extension is determined by the context of utterance. Thus, the

18 Stanford Encyclopedia of Philosophy Summer 2015 Edition 19

 Possible Worlds Christopher Menzel

referent of ‘the actual world’ in a given utterance is simply the world of explicit quantifiers over worlds.[18] The mature account of Lewis 1986 is
the speaker, just as the referent of an utterance of ‘the present moment’ is much more semantic in orientation: it avoids any talk of translation and
the moment of the utterance; likewise, an utterance of the form ‘a is offers instead a (somewhat informal) account of concretist possible world
actual’ indicates only that a shares the same world as the speaker. The truth conditions for a variety of modal assertions. Nonetheless, it is useful
speaker thereby ascribes no special property to a but, essentially, to express the logical forms of these truth conditions explicitly in terms of
expresses no more than when she utters ‘a is here’, understood in the worlds, existence in a world (in the sense of AE1, of course), and the
broadest possible sense. By the same token, when we speak of non-actual counterpart relation, which will be discussed shortly:
possibilia — Lewis's preferred label for the denizens of possible worlds —
we simply pick out those objects that are not here in the broadest sense. In Wx: x is a world
the mouth of an other-worldly metaphysician, we here are all among the Ixy: x exists in world y
non-actual possibilia of which she speaks in her lectures on de re Cxy: x is a counterpart of y
modality.
For sentences like (17) involving only de dicto modalities, Lewis's truth
2.1.3 Modal Reductionism, Counterparts, and the Analysis of Intensions conditions are similar in form to the truth conditions generated by the
modal clauses of basic possible world semantics; specifically, for (17):
Modal Reductionism and Counterparts. Lewis parted ways
(18) For every world w, every individual x in w that is a dog is a
dramatically with his mentor W. V. O. Quine on modality. Quine (1960,
mammal: ∀w(Ww → ∀x(Ixw → (Gx → Mx))).
§41) stands in a long line of philosophers dating back at least to David
Hume who are skeptical, at best, of the idea that modality is an objective
As in possible world semantics, the modal operators ‘□’ and ‘◇’ “turn
feature of reality and, consequently, who question whether modal
into” quantifiers over worlds in concretist truth conditions (1986, 5). Also
assertions in general can be objectively true or false, or even coherent.
as in possible world semantics, a quantifier (in effect) ranging over
Lewis, by contrast, wholly embraces the objectivity of modality and the
individuals that occurs in the scope of a quantifier (in effect) ranging over
coherence of our modal discourse. What he denies, however, is that
worlds — ‘∀x’ and ‘∀w’, respectively, in (18) — is, for each value w of
modality is a fundamentally irreducible feature of the world. Lewis, that
the bound world variable, restricted to the objects existing in w. However,
is, is a modal reductionist. For Lewis, modal notions are not primitive.
unlike possible world semantics, predicates are not to be thought of as
Rather, truth conditions for modal sentences can be given in terms of
having different extensions at different worlds. Rather, for Lewis, each (n-
worlds and their parts; and worlds themselves, Lewis claims, are defined
place) predicate has a single extension that can contain (n-tuples of)
entirely in non-modal terms. The earliest presentation of Lewis's theory of
objects across many different worlds — intuitively, all of the objects that
modality (Lewis 1968) — reflecting Quine's method of regimentation —
have the property (or n-tuples of objects that stand in the relation)
offers, rather than a possible world semantics, a scheme for translating
expressed by the predicate across all possible worlds. Thus, in particular,
sentences in the language of modal predicate logic into sentences of
the predicate ‘G’ picks out, not just this-worldly dogs but other-worldly
ordinary first-order logic in which the modal operators are replaced by

20 Stanford Encyclopedia of Philosophy Summer 2015 Edition 21

 Possible Worlds Christopher Menzel

canines as well. Likewise, the pet predicate ‘T’ picks out both actual and differ in important ways from its other-wordly counterparts. A typical
other-worldly pets. Such a move is not feasible in basic possible world other-worldly counterpart of Algol, for example, might resemble her very
semantics, which is designed for a metaphysics in which one and the same closely up to some point in her history — a point, say, after which she
individual can exemplify a given property in some worlds in which they continued to live out her life as a stray instead of being brought home by
exist but not others. Hence, a typical predicate will be true of an individual our kindly dog-lover John. Hence, sentences making de re assertions about
with respect to some worlds and false of it with respect to others. But, for what Algol might have done or what she could or could not have been are
Lewis, as we've seen, distinct possible worlds do not overlap and, hence, unpacked, semantically, as sentences about her counterparts in other
objects are worldbound, thereby eliminating the need to relativize possible worlds. Thus, when we analyze (16) accordingly, we have the
predicate extensions to worlds. entirely unproblematic concretist truth condition:

However, this very feature of Lewis's account — worldboundedness — (20) Algol is a pet, but there is a world in which exists a counterpart
might appear to threaten its coherence. For example, since Algol is in fact of hers that is not:
a pet, given worldboundedness and the definition AE1 of existence in a Ta ∧ ∃w(Ww ∧ ∃x(Ixw ∧ Cxa ∧ ¬Tx)).
world w, we have:
Ascriptions of essential properties, as in (15), are likewise unpacked in
(19) There is no world w such that Algol exists in w and fails to be terms of counterparts: to say that Algol is a dog essentially is to say that
someone's pet: ¬∃w(Iaw ∧ ¬Ta),
(21) All of Algol's counterparts in any world are dogs:
But, according to Lewis's analysis, the modal operators ‘□’ and ‘◇’, ∀w(Ww → ∀x((Ixw ∧ Cxa) → Gx)).
semantically, are quantifiers over worlds. Hence, (19) might appear to be
exactly the concretist truth condition for the denial of (the right conjunct The Analysis of Intensions. Lewis's possible world truth conditions are
of) (16), i.e., it might appear that, on Lewis's analysis, Algol is not a pet expressed in classical non-modal logic and, hence, they are to be
accidentally but essentially; likewise, more generally, any individual and interpreted by means of standard Tarskian semantics. Thus, n-place
any intuitively accidental property of that individual. predicates π are assigned extensions Eπ — in particular, for 1-place
predicates, sets of individuals — as their semantic values, as described in
In fact, Lewis whole-heartedly accepts that things have accidental the exposition in §1.2 above. However, given worldboundedness and the
properties and, indeed, would accept that (16) is robustly true. His fact that predicate extensions are drawn not simply from the actual world
explanation involves one of the most interesting and provocative elements but from all possible worlds, these extensions are able to serve as
of his theory: the doctrine of counterparts. Roughly, an object y in a world intensions in Lewis's theory. As in basic possible world semantics,
w2 is a counterpart of an object x in w1 if y resembles x and nothing else intensional entities in general can be defined in terms of the basic ontology
in w2 resembles x more than y.[19] Each object is thus its own (not of the theory independent of the linguistic roles they can play as the
necessarily unique) counterpart in the world it inhabits but will typically intensions of predicates. And because individuals are worldbound, Lewis

22 Stanford Encyclopedia of Philosophy Summer 2015 Edition 23

 Possible Worlds Christopher Menzel

is able to simplify the definitions given in §1.3 by defining intensions as considered successful only if it is complete in this sense.
sets rather than functions:
The chief question Lewis faces in this regard is whether there are enough
A proposition is any set of worlds. worlds to do the job. The truth condition (20) for the intuitively true (16)
A property is any set of individuals. says that there exists a possible world in which a counterpart of Algol is
An n-place relation (n > 1) is any set of n-tuples of individuals.[20] no one's pet. By virtue of what in Lewis's theory does such a world exist?
The ideal answer for Lewis would be that some principle in his theory
Thus, on this analysis, a proposition p is true in a world w just in case w ∈ guarantees a plenitude of worlds, a maximally abundant array of worlds
p and an individual a has a property P just in case a ∈ P. (Note that that leaves “no gaps in logical space; no vacancies where a world might
propositions are thus simply properties of worlds on these definitions.) a have been, but isn't” (Lewis 1986, 86). From this it would follow that the
has P accidentally just in case a ∈ P but b ∉ P for some other-worldly worlds required by the concretist truth condition for any intuitive modal
counterpart of b of a; and a has P essentially if b ∈ P for every truth exist. Toward this end, Lewis initially considers the evocative
counterpart b of a. principle:

In Lewis's theory of modality, then, modal operators are understood

Ways Absolutely every way that a world could be is a way that some
semantically to be quantifiers over concrete worlds, predicates denote
world is.
intensions understood as sets of (n-tuples of) parts of those worlds, and
sentences involving de re modalities are understood in terms of Since, in particular, a world satisfying (20) seems quite obviously to be a
counterparts. To the extent that these notions are free of modality, Lewis way a world could be, by Ways such a world exists. But there is a fatal
has arguably reduced modal notions to non-modal. flaw here: Lewis himself (1973, 84) identifies ways that a world could be
with worlds themselves. So understood, Ways collapses into the triviality
2.1.4 Plenitude and Recombination that every world is identical to some world.[22]

That Lewis's truth conditions for modal statements are themselves free of Lewis finds a replacement for Ways in a principle of recombination
modality and, hence, that his theory counts as a genuine reduction of whereby “patching together parts of different possible worlds yields
modal notions to non-modal is not terribly controversial (albeit not another possible world” (1986, 87–88). The principle has two aspects. The
undisputed — see Lycan 1991, 224–27; Divers and Melia 2002, 22–24). first is the principle that “anything can coexist with anything”. For “if
Significantly more controversial, and perhaps far more critical to the there could be a dragon, and there could be a unicorn,” Lewis writes, “but
project, is whether or not his account is complete, that is, whether or not, there couldn't be a dragon and a unicorn side by side, that would be ... a
for all modal statements φ, (i) if φ is intuitively true, then its Lewisian failure of plenitude” (ibid., 88). Given that individuals are worldbound,
truth condition holds (ii) if φ is intuitively false, then its Lewisian truth however, the principle is expressed more rigorously (and more generally)
condition fails.[21] The challenge to Lewis, then, is that his account can be in terms of other-worldly duplicates:

24 Stanford Encyclopedia of Philosophy Summer 2015 Edition 25

 Possible Worlds Christopher Menzel

R1 For any (finite or infinite) number of objects a1, a2, ..., there is a regarding plenitude is still a matter of some dispute.[24] However, even if
world containing any number of duplicates of each of those it doesn't, it is less than clear whether this counts against the success of
objects in any spatiotemporal arrangement (size and shape Lewis's reductionist project. For, as a realist about worlds, Lewis does not
permitting). seem to be under any obligation to “derive” plenitude from more
fundamental principles. Hence, there is no obvious reason why he cannot
The second aspect of the principle expresses “the Humean denial of respond to charges of incompleteness by saying that it is simply a
necessary connections” (ibid., 87), that is, the idea that anything can fail to presupposition of his theory that logical space has no gaps, that there are
coexist with anything else. For “if there could be a talking head always enough worlds to satisfy the concretist truth condition for any
contiguous to the rest of a living human body, but there couldn't be a intuitive modal truth.[25] So understood, the role of recombination for a
talking head separate from the rest of a human body, that too would be a realist about worlds like Lewis is something like the role of such axioms
failure of plenitude” (ibid). To express this a bit more rigorously, say that as powerset and replacement for a realist about sets: given some sets, these
objects a1, a2, ..., are independent of objects b1, b2, ..., if no sum of any principles provide us with a detailed — but always less than complete —
parts of the former are parts or duplicates of any sum of any parts of the characterization of what further sets there are. Their role, therefore, is to
latter and vice versa; then we have: give us insight into the richness and diversity of set theoretic space, not a
complete mechanism for proving which particular sets do or do not exist.
R2 For any world w any (finite or infinite number of) objects a1, a2, Likewise recombination vis-à-vis worlds and logical space.
..., in w and any objects b1, b2, ..., in w that are independent of
a1, a2, ..., there is a world containing duplicates of a1, a2, ..., and 2.1.5 A Brief Assessment of Concretism
no duplicates of b1, b2, ... .
Lewis's theory is particularly commendable for its striking originality and
Worlds that satisfy the concretist truth conditions for workaday ingenuity and for the simple and straightforward answers AW1 and AE1
possibilities like (16) are easily conceived as consisting of duplicates of that it provides to our two questions QW and QE above. Furthermore,
relevant parts of the actual world — suitably organized to retain their because worlds are (plausibly) defined entirely in nonmodal terms, the
actual properties, or not, as needed. Hence, the existence of such worlds truth conditions provided by Lewis's translation scheme themselves appear
does indeed appear to follow from the existence of the actual world by to be free of any implicit modality. Hence, unlike many other popular
recombination. Worlds containing talking donkeys, exotic species accounts of possible worlds (notably, the abstractionist accounts discussed
resulting from a wholly different evolutionary history, worlds with silicon- in the following section), Lewis's promises to provide a genuine analysis
based life forms, and so on present a bigger challenge to the view. of the modal operators.
Nonetheless, it is not entirely implausible to think such worlds exist given
suitable duplication and reorganization of microphysical objects.[23] Perhaps the biggest — if not the most philosophically sophisticated —
challenge to Lewis's theory is “the incredulous stare”, i.e., less colorfully
Whether recombination completely captures our modal intuitions put, the fact that its ontology is wildly at variance with common sense.

26 Stanford Encyclopedia of Philosophy Summer 2015 Edition 27

 Possible Worlds Christopher Menzel

Lewis faces this objection head on: His theory of worlds, he To give the notion of a state, or condition, of the world a little more
acknowledges, “does disagree, to an extreme extent, with firm common metaphysical substance, abstractionists typically appeal to more traditional
sense opinion about what there is” (1986, 133). However, Lewis argues ontological categories. Thus, for example, that things could be in the
that no other theory explains so much so economically. With worlds in simple state described above might be spelled out in one of the following
one's philosophical toolkit, one is able to provide elegant explanations of a ways:
wide variety of metaphysical, semantical, and intentional phenomena. As
high as the intuitive cost is, Lewis (135) concludes, the existence of The proposition that Anne is in her office and at her desk is
worlds “ought to be accepted as true. The theoretical benefits are worth possibly true.
it.” The set of propositions {that Anne is in her office, that Anne is at
her desk} is such that, possibly, all of its members are true.
Additional discussion of, and objections to, concretism can be found in the The property being such that Anne is in her office and at her desk
supplemental document Further Problems for Concretism. is possibly exemplified (by “things as a whole”).

2.2 Abstractionism Possible worlds are then defined as special cases of the type of entity in
question that are in some relevant sense total. Adams (1974), for example,
A rather different set of intuitions about situations is that they are abstract defines possible worlds to be consistent sets of propositions that are total
entities of a certain sort: They are states or conditions, of varying detail in the sense of containing, for every proposition p, either p or its negation;
and complexity, that a concrete world could be in — they are ways that Fine (1977), fleshing out ideas of Prior, defines a possible world to be a
things, as a whole, could be.[26] Thus, returning to our original example, consistent proposition w that is total in the sense that, for every
one very simple way things could be is for our philosopher Anne to be in proposition p, w entails either p or its negation. For purposes here,
her office. We can now imagine, as in our example, further detail being however, we will sketch the fundamentals of the abstractionist view in
successively added to that description to yield more complex ways things terms of states of affairs, following the basic features of the account
could be: Anne working at her desk in her office; music being in the developed by Plantinga (1974, 1976), an account that, in the literature,
background; her husband being on the phone in the next room; her frequently serves as a particularly trenchant abstractionist counterpoint to
neighbor mowing the lawn next door; and so on. Roughly speaking, then, Lewis's concretism.[27]
a possible world for an abstractionist is the limit of such a “process” of
consistently extending and adding detail to some initial state of the world; States of affairs (SOAs) are abstract, intensional entities typically signified
it is a total way things could be, a consistent state of the world that settles by sentential gerundives like “Algol's being John's pet” and “There being
every possibility; a consistent state to which no further detail could be more than ten solar planets”. Importantly, SOAs constitute a primitive
added without rendering it inconsistent. ontological category for the abstractionist; they are not defined in terms of
possible worlds in the manner that propositions are in §1.3. Just as some
2.2.1 Abstract Possible Worlds and Existence Therein propositions are true and others are not, some SOAs are actual and others

28 Stanford Encyclopedia of Philosophy Summer 2015 Edition 29

 Possible Worlds Christopher Menzel

are not.[28] Note, then, that to say an SOA is non-actual is not to say that it And non-actual worlds are simply those total possible SOAs that do not.
does not actually exist. It is simply to say that it is not, in fact, a condition,
What of existence in such worlds? As we've seen, on Lewis's account, to
or state, that the concrete world is actually in. However, because ‘____ is
exist in a concrete world w is literally to exist in w, that is, within the
actual’ is often used simply to mean ‘____ exists’, there is considerable
spatiotemporal boundaries of w. Clearly, because SOAs are abstract,
potential for confusion here. So, henceforth, to express that an SOA is
individuals cannot exist in abstractionist worlds in anything like the same
actual we will usually say that it obtains.
literal, mereological sense. Accordingly, the abstractionist defines
An SOA is said to be possible (necessary, impossible) insofar as it is existence in a world simply to be a special case of the inclusion relation:
possible (necessary, impossible) that it obtain. One SOA s is said to
include another t if, necessarily, s obtains only if t does; s precludes t if, AE2 Individual a exists in possible world w =def w includes a's
necessarily, s obtains only if t doesn't. So, for example, Algol's being existing.
John's pet includes Algol's being someone's pet and precludes there
Unlike concretism, then, abstractionism does not entail that individuals are
being no pets. Thus, on the abstractionist's understanding of a situation as
worldbound; there is no inconsistency whatever in the idea that many
a state or condition of the physical world rather than a concrete, structured
distinct worlds can include the existence of one and the same individual.
piece of it, the inclusion of one situation in another is a purely logical
Indeed, typically, abstractionists are staunchly committed to transworld
relation, not a mereological one. Finally, say that an SOA s is total if, for
identity and hold that most any given individual exists in many possible
every SOA t, s either includes or precludes t. (Abstractionists often use
worlds and, moreover, that contingent individuals, at least, can exemplify
‘maximal’ instead of ‘total’, but we have already introduced this term in
very different properties from world to world. The abstractionist,
the context of concretism.) Abstractionist possible worlds are now
therefore, has no need to appeal to counterparts to understand de re
definable straightaway:
modalities and can therefore accept the truth conditions for such
AW2 w is a possible world =def w is an SOA that is both possible and modalities given by basic possible world semantics (spelled out, of course,
total.[29] in terms of his definitions AW2 and AE2). In particular, he can take the
standard possible world truth condition for, e.g., the right conjunct of (16)
It is easy to see that this definition covers the more intuitive at face value: ‘◇(E!a ∧ ¬Ta)’ is true on the abstractionist's approach if and
characterizations of abstract possible worlds above: they are consistent — only if there is is a world in which Algol herself, rather than some
i.e., possible — states of the world that settle every possibility, consistent counterpart of hers, exists but fails to be anyone's pet.
states to which no further detail could be added without rendering them
inconsistent. Note also that, for the abstractionist, as with the concretist, 2.2.2 Irreducible Modality and Intensional Entities
the actual world is no different in kind from any other possible world; all
It is important to note that the possible worlds of abstractionism do not
possible worlds exist, and in precisely the same sense as the actual world.
yield a reductive analysis of modality. The reason for this is clear: abstract
The actual world is simply the total possible SOA that, in fact, obtains.

30 Stanford Encyclopedia of Philosophy Summer 2015 Edition 31

 Possible Worlds Christopher Menzel

possible worlds are defined in irreducibly modal terms — a possible world However, it is not entirely clear that this observation constitutes an
is an SOA that (among other things) possibly obtains; or a set of objection to abstractionism. For the abstractionist can argue that the goal
propositions such that it is possible that all of its members are true; or a of his analysis is the converse of the reductionist's goal: The reductionist
property that is possibly exemplified; and so on. Hence, unpacked in terms wants to understand modality in terms of worlds; the abstractionist, by
of the abstractionist's definitions, the possible world truth conditions for contrast, wants to understand worlds in terms of modality. That is, the
modal propositions are themselves irreducibly modal. For example, when abstractionist can argue that we begin with a primitive notion of modality
we unpack Plantinga's definition of a possible world in the semantic clause and, typically upon a certain amount of philosophical reflection, we
for sentences of the form ⌈◻ψ⌉ in order to derive the truth condition for subsequently discover an intimate connection to the notion of a possible
(17), ‘□∀x(Gx → Mx)’, we end up with this: world, as revealed in the principles Nec and Poss. The analysis that the
abstractionist provides is designed to make this connection explicit,
(22) For all SOAs w, if (i) possibly, w obtains and (ii) for all SOAs s, ideally, in such a way that Nec and Poss fall out as theorems of his theory
either (a) necessarily, w obtains only if s does or (b) necessarily, (see, e.g., Plantinga 1985 and Menzel and Zalta 2014).
w obtains only if s doesn't, then, ‘∀x(Gx → Mx)’ is true at w.
Hand in glove with the irreducible nature of modality is the nature of
If we now unpack the modal operators in (22) using the corresponding intensional entities. Concretists define intensional entities in terms of
truth conditional clauses of standard possible world semantics, the result worlds, as described in §2.1.3. Abstractionists, by contrast, define worlds
will contain further world quantifiers. And spelling out those world in terms of intensional entities. This divergence in their choice of
quantifiers in turn using Plantinga’s definition will re-introduce those ontological primitives reflects, not only their differing stances toward
same modal operators yet again. modality, but also an important methodological difference with regard to
metaphysical inquiry. The concretist is far more pragmatic; notions of
More generally, and a bit more exactly, put: As noted above, the logical property, relation, proposition, and the like play certain roles in our
framework of basic possible world semantics is classical predicate logic. theorizing and are subject to a “jumble of conflicting desiderata” (Lewis
The logical framework of abstractionism is modal predicate logic. Hence, 1986, 54). Within a given theory, any entities that can play those roles
if possible world semantics is supplemented with abstractionist definitions fruitfully for the purposes at hand are justifiably identified with those
of possible worlds, then the logical framework of possible world notions — regardless of how well they comport with pre-theoretic
semantics becomes modal predicate logic as well and, as a consequence, intuitions. Thus, Lewis finds it to be a strength of his position that he is
the extensionality of the semantics is lost once again. (This point is able to adopt the set theoretic definitions in §2.1.3. By contrast, at least
expressed somewhat more formally in the supplemental document The some abstractionists — Plantinga (1987) perhaps most notably — believe
Intensionality of Abstractionist Possible World Semantics.) Since, as that we have intuitive, pre-theoretic knowledge of intensional entities that
noted above, the central motivation for possible world semantics was to precludes their being identified with set theoretic constructions of any
deliver an extensional semantics for modal languages, any motivation for sort.[31] (See Stalnaker 1976 for a particularly illuminating discussion of
abstractionism as a semantic theory is arguably undermined.[30] the contrast between concretism and abstractionism with respect to the

32 Stanford Encyclopedia of Philosophy Summer 2015 Edition 33

 Possible Worlds Christopher Menzel

treatment intensional entities.) possibilia. However, we have just seen that the abstractionist can avoid
this apparent commitment to possibilism by defining possible worlds to be
2.2.3 Actuality and Actualism SOAs of a certain sort. So defined, non-actual worlds, i.e., worlds that fail
to obtain, can still actually exist. Hence, the commitment of basic possible
As was noted in §2.1.2, for the concretist, there is no special property of world semantics to non-actual worlds does not in itself threaten the
the actual world — actuality — that distinguishes it, in any absolute sense, actualist's ontological scruples.
from all of the others; it is simply the world that we inhabit. For
abstractionists, however, actuality is a special property that distinguishes However, the specter of possibilism is not so easily exorcised. For non-
exactly one possible world from all others — the actual world is the only actual worlds are not the only, or even the most compelling, examples of
world that happens to obtain; it is the one and only way things could be mere possibilia that seem to emerge out of basic possible world semantics.
that is the way things as a whole, in fact, are. However, for most For instance, it is quite reasonable to think that evolution could have taken
abstractionists, the distinctiveness of the actual world does not lie simply a very different course (or, if you like, that God could have made very
in its actuality but in its ontological comprehensiveness: the actual world different creative choices) and that there could have been individuals —
encompasses all that there is. In a word: most abstractionists are actualists. call them Exotics — that are biologically very different from all actually
existing individuals; so different, in fact, that no actually existing thing
Actualism is the thesis that everything that there is, everything that has could possibly have been an Exotic. According to basic possible world
being in any sense, is actual. In terms of possible worlds: Everything that semantics, the sentence ‘There could have been Exotics’ or, more
exists in any world exists in the actual world.[32] Possibilism, by contrast, formally,
is the denial of actualism; it is the thesis that there are mere possibilia, i.e.,
things that are not actual, things that exist in other possible worlds but fail (23) ◇∃xEx
to exist in the actual world. Concretists are obviously not actualists (on
their understanding of ‘actual’, at any rate).[33] Indeed, for the concretist, is true just in case there is a world in which ‘∃xEx’ is true, i.e., when all is
since individuals are worldbound, everything that exists in any nonactual said and done, just in case:
possible world is distinct from everything in the actual world. However,
although possibilism and abstractionism are entirely compatible — Zalta (24) There is a possible world w and an individual a in w such that a
(1983), for example, embraces both positions — abstractionists tend to be is an Exotic in w,
actualists. The reason for this is clear: Basic possible world semantics
which, a bit less formally, is simply to say that
appears to be committed to possibilism and abstractionism promises a way
of avoiding that commitment.
(25) Some individual is an Exotic in some possible world.
The specter of possibilism first arises with regard to non-actual possible
However, since no actually existing thing could have been an Exotic,
worlds, which would seem by definition to be prime examples of mere
anything that is an Exotic in some possible world cannot be among the

34 Stanford Encyclopedia of Philosophy Summer 2015 Edition 35

 Possible Worlds Christopher Menzel

things that exist in the actual world. Thus, the truth conditions that basic — haecceities — to similar ends. The haecceity of an individual a is the
possible world semantics assigns to some of our intuitive modal beliefs property of being that very individual, the property being a. A property is
appear to entail that there are non-actual individuals as well as non-actual a haecceity, then, just in case it is possible that it is the haecceity of some
possible worlds. Defining possible worlds as SOAs provided a way for the individual.[36] It is a necessary truth that everything has a haecceity. More
actualist to embrace non-actual worlds without compromising her importantly, for haecceitists, haecceities are necessary beings. Thus, not
actualism. But how is the actualist to understand the apparent commitment only is it the case that, had any particular individual a failed to exist, its
to non-actual individuals in such truth conditions as (25)? haecceity ha would still have existed, it is also the case that, for any
“merely possible” individual a, there is an actually existing haecceity that
Answers that have been given to this question represent a rather deep would have been a's haecceity had a existed. More generally (and more
divide between actualist abstractionists. On the one hand, “trace” carefully) put: Necessarily, for any individual a, (i) a has a haecceity h
actualists introduce actually existing entities into their ontologies that can and (ii) necessarily, h exists.
play the role of mere possibilia in (25) and its like. Trace actualists come
in two varieties: new actualists and haecceitists. New actualists like Like the new actualists, then, the haecceitist's metaphysics enables him to
Linsky and Zalta (1996) and Williamson (1998, 2000, 2013) argue that, in systematically reinterpret possible world semantics in such a way that the
fact, all individuals are actually existing, necessary beings but not all of truth conditions of modal discourse are expressed solely in term of
them are necessarily concrete. Some concrete individuals — those actually existing entities of some sort rather than actual and non-actual
traditionally (mis-)categorized as contingent beings — are only individuals. More specifically, for the haecceitist, the domain d(w) of a
contingently concrete. Likewise, some non-concrete individuals — those, world w is taken to be the set of haecceities that are exemplified in w, that
like possible Exotics, traditionally (mis-)categorized as contingently non- is, the set of haecceities h such that w includes h's being exemplified.
actual mere possibilia — are only contingently non-concrete.[34] Likewise, the w-extension of a (1-place) predicate π is taken to be a set of
haecceities — intuitively, those haecceities that are coexemplified in w
This novel take on modal metaphysics allows the new actualist to with the property expressed by π. So reinterpreted, the truth condition for
reinterpret possible world semantics so as to avoid possibilism. Notably, (23) is:
the domain d(w) of a world w is understood not as the set of things that
exist in w — for all individuals exist in all worlds — but the set of things (27) There is a possible world w and a haecceity h that is (i)
that are concrete in w.[35] Hence, for the new actualist, the correct truth exemplified in w and (ii) coexemplified with the property being
condition for (23) is: an Exotic in w.

(26) There is a possible world w and an individual a that is (i) By contrast, “no-trace”, or strict, actualists like Prior (1957), Adams
concrete in w and (ii) an Exotic in w. (1981), and Fitch (1996) hew closely to the intuition that, had a contingent
individual a failed to exist, there would have been absolutely no trace, no
On the other hand, haecceitists like Plantinga introduce special properties metaphysical vestige, of a — neither a itself in some non-concrete state

36 Stanford Encyclopedia of Philosophy Summer 2015 Edition 37

 Possible Worlds Christopher Menzel

nor any abstract proxy for a. Hence, unlike trace actualism, there are no analyzing and developing solutions to philosophical problems. In this
such vestiges in the actual world of objects that are not actual but only regard, particularly noteworthy are: Plantinga's (1974) influential work on
could have been. the ontological argument and the free will defense against the problem of
evil; Adams' (1974, 1981) work on actualism and actuality; and
The logical consequences for no-trace actualists, however, appear to be Stalnaker's (1968, 1987) work on counterfactual conditionals and mental
severe; at the least they cannot provide a standard compositional semantics content.
for modal languages, according to which (roughly) the meaning of a
sentence is determined by its logical form and the meanings of its A number of important objections have been voiced in regard to
semantically significant constituents. In particular, if there is nothing to abstractionism. Some of these are addressed in the document Problems
play the role of a “possible Exotic”, nothing that is, or represents, an with Abstractionism.
Exotic in some other possible world — a mere possibile, a contingently
non-concrete individual, an unexemplified haecceity — then the strict 2.3 Combinatorialism
actualist cannot provide standard, compositional truth conditions for
As its name might suggest, our third approach — combinatorialism —
quantified propositions like (23) that yield the intuitively correct truth
takes possible worlds to be recombinations, or rearrangements, of certain
value. For, understood compositionally, (23) is true if and only if ‘∃xEx’ is
metaphysical simples. Both the nature of simples and the nature of
true at some world w. And that, in turn, is true at w if and only if ‘Ex’ is
recombination vary from theory to theory. Quine (1968) and Cresswell
true at w for some value of ‘x’. But, as just noted, for the strict actualist,
(1972), for example, suggest taking simples to be space-time points
there is no such value of ‘x’. Hence, for the strict actualist, ‘Ex’ is false at
(modeled, perhaps, as triples of real numbers) and worlds themselves to be
w for all values of ‘x’ and, hence, (23) is false as well. (These issues are
arbitrary sets of such points, each set thought of intuitively as a way that
explored in much greater detail in §4 of the entry Actualism.)
matter could be distributed throughout space-time. (A world w, so
2.2.4 A Brief Assessment of Abstractionism construed, then, is actual just in case a space-time point p is a member of
w if and only if p is occupied by matter.) Alternatively, some philosophers
Like concretism, abstractionism provides a reasonably clear and intuitive define states a world could be in, and possible worlds themselves, simply
account of what worlds are and what it is to exist in them, albeit from a to be maximally consistent sets of sentences[37] in an expressively rich
decidedly different perspective. Although, as noted in §2.2.2, the fact that language — “recombinations”, certainly, of the sentences of the language.
modality is a primitive in abstractionist definitions of possible worlds (Lewis refers to this view as linguistic ersatzism.[38]) However, the
arguably compromises its ability to provide semantically illuminating truth predominant version of combinatorialism finds its origins in Russell's
conditions for the modal operators, those definitions can be taken to (1918/1919) theory of logical atomism and Wittgenstein's (1921, 1922,
illuminate the connection between our basic modality concepts and the 1974) short but enormously influential Tractatus Logico-Philosophicus. A
evocative notion of a possible world that serves as such a powerful suggestive paper by Skyrms (1981) spelling out some of the ideas in the
conceptual tool for constructing philosophical arguments and for Tractatus, in turn, inspired a rich and sophisticated account that is

38 Stanford Encyclopedia of Philosophy Summer 2015 Edition 39

 Possible Worlds Christopher Menzel

developed and defended in great detail in an important series of books and molecular.) One fact f includes another g if every conjunct of g is a
articles by D. M. Armstrong (1978a, 1978b, 1986a, 1989, 1997, 2004b, conjunct of f. (Note, importantly, that inclusion, so defined, is quite
2004c). In this section, we present a somewhat simplified version of different from the homonymous notion defined in the discussion of
combinatorialism that draws primarily upon Armstrong's work. Unless abstractionism above — most notably, combinatorial inclusion is not a
otherwise noted, this is what we shall mean by ‘combinatorialism’ for the modal notion.) For purposes below, say that an object a is a bare
remainder. particular in a molecular fact f if there is no monadic conjunct of f of
which a is the constituent object, no conjunct of the form a exemplifies F,
2.3.1 The Basic Ontology of Combinatorialism for some property F. a is a bare particular if it is bare in every molecular
fact. Intuitively, of course, a bare particular is an unpropertied object.
Wittgenstein famously asserted that the world is the totality of facts, not of
things (ibid., §1.1). The combinatorialist spells out Wittgenstein's There is no upper bound on the “size” of a molecular fact and no
aphorism explicitly in terms of an ontology of objects (a.k.a., particulars), restriction on which atomic facts can form a conjunction; for any atomic
universals (a.k.a., properties and relations), and facts. Facts are either facts at all, there is a molecular fact whose conjuncts are exactly those
atomic or molecular. Every atomic fact — Sachverhalt, in the language of facts. As a first cut, then, we can spell out Wittgenstein's characterization
the Tractatus — is “constituted” by an n-place relation (= property, for of the (actual) world as the totality of facts by defining the world to be the
n=1) and n objects that stand in, or exemplify, that relation. Thus, for largest molecular fact, the molecular fact that includes all of the atomic
example, suppose that John is 1.8 meters tall. Then, in addition to John facts.[39]
and the property being 1.8 meters tall, there is for the combinatorialist the
atomic fact of John's exemplifying that property. More generally, atomic Although objects and universals are typically included along with facts in
facts exist according to the following principle: the basic ontology of combinatorialism, facts are typically considered
more fundamental. Indeed, taking his queue from the Tractarian thesis that
AF Objects a1, ..., an exemplify n-place relation R iff there is the fact the world consists of facts, not things, Armstrong (1986a, 577) argues that
a1, ..., an's exemplifying R ([R,a1,...,an], for short). facts alone are ontologically basic and that objects and universals are
simply “aspects of, abstractions from” facts. Accordingly, he calls the
Say that the ai are the constituent objects of the fact in question and R its object constituent of a fact of the form [P,a] a “thin” particular, an object
constituent universal, and that R and the ai all exist in [R,a1,...,an]. “considered in abstraction from all its [intrinsic] properties” (1993, 433);
and where N is the conjunction of “all the non-relational properties of that
A fact is monadic if its constituent universal is a property. A molecular particular (which would presumably include P), the atomic fact a's
fact f is a conjunction of atomic facts. Its constituent objects and exemplifying N itself is the corresponding “thick” particular ” (ibid., 434
universals are exactly those of its conjuncts and an entity exists in f just in — we will occasionally use italics to distinguish a thin particular a from
case it exists in one of its conjuncts. (For simplicity, we stipulate that an the corresponding thick particular a). Though not all combinatorialists of
atomic fact has (only) itself as a conjunct and, hence, is “trivially” every stripe buy into Armstrong's “factualist” metaphysics (Bricker 2006),

40 Stanford Encyclopedia of Philosophy Summer 2015 Edition 41

 Possible Worlds Christopher Menzel

they do generally agree that facts are more fundamental, at least to the h1 and h2. For the combinatorialist, “thick” particulars like the molecule
extent that both the notion of a bare particular, i.e., an object exemplifying itself as well as its constituent atoms are themselves facts: o is the fact
no properties, and that of an unexemplified property are considered [O,o] in which the universal oxygen (O) is exemplified by a thin
incoherent; insofar as they exist at all, the existence of both particulars and particular o;[43] likewise h1 and h2. W in turn comprises those monadic
universals depends on their “occurring” in some fact or other. Whatever facts and the relational facts [B,o,h1], [B,o,h2] wherein the covalent
their exact ontological status, it is an important combinatorialist thesis that bonding relation B holds between the oxygen atom and the two hydrogen
exactly what objects and universals exist is ultimately a matter for natural atoms. The structural universal Water itself, then, shares this structure —
science, not metaphysics, to decide. it is, so to say, an isomorph consisting of the monadic universals O and H
and the binary relation B, structured as indicated in the right-hand diagram
Objects can be either simple or complex. An object is simple if it has no of Figure 1.[44]
proper parts, and complex otherwise. Like objects, universals too divide
into simple and complex. A universal is simple if it has no other universal
as a constituent, and complex otherwise. Complex universals accordingly
come in two varieties: conjunctive — the constituents of which are simply
its conjuncts — and structural. A structural universal U is one that is
exemplified by a complex object O, and its constituents are universals
(distinct from U) exemplified by simple parts of O that are relevant to O's
being an instance of U.[40] It is important to note that, for Armstrong, the
Figure 1: A Water Molecule W and the Structural Universal Water
constituency relation is not the mereological parthood relation. Rather,
complex universals (hence also complex facts of which they are
2.3.2 States of Affairs and Recombination
constituents) enjoy a “non-mereological mode of composition” (1997,
119–123) that, in particular, allows for a richer conception of their It should be clear from principle AF that all atomic facts hold; that is, all
structure.[41] (An assumption of our simplified account here will be that of them reflect actual exemplification relations. Obviously, however,
both the proper part of relation and the constituency relation are well- possibility encompasses more than what is actual, that is, there are
founded. It follows that (i) there is no gunk, i.e., that every complex object possible facts as well as actual facts; the world's universals might have
is composed, ultimately, entirely of simples and (ii) complex universals — been exemplified by its objects very differently. If they had — if the
hence the complex facts in which they are exemplified — are ultimately world's objects and universals had combined in a very different way —
“grounded” in simple facts, i.e., that they cannot be infinitely decomposed there would have been a very different set of atomic facts and, hence, a
into further complex universals/facts.[42]) very different world.
To illustrate the basic idea: in Figure 1, the left-hand diagram depicts a To spell out the idea of a possible fact, the combinatorialist introduces the
water molecule W comprising an oxygen atom o and two hydrogen atoms

42 Stanford Encyclopedia of Philosophy Summer 2015 Edition 43

 Possible Worlds Christopher Menzel

more general notion of an atomic (combinatorial) state of affairs, that is, Constituency for states of affairs is understood as for facts. Additionally,
an entity that simply has the form of an atomic fact — n objects analogous to molecular facts, there are molecular states of affairs —
exemplifying an n-place relation — but without any requirement that the conjunctions of atomic states of affairs. Inclusion between states of affairs
exemplification relation in question actually holds between them. More is understood exactly as it is between facts and being a bare particular in a
exactly: molecular state of affairs s is understood as for facts: a is a bare particular
in s if there is no monadic conjunct of s of the form a exemplifies F. The
AS For any objects a1, ..., an and any n-place relation R, there is an notion of recombination is now definable straightaway:
atomic (combinatorial) state of affairs a1, ..., an's exemplifying R
(again, [R,a1,...,an], for short). (28) s is a recombination of a molecular state of affairs f =def s is a
molecular state of affairs whose constituent objects and
Thus, even if the two hydrogen atoms h1 and h2 in a water molecule do constituent universals are exactly those of f. s is a non-trivial
not in fact stand in the covalent bonding relation B, there is nonetheless recombination of f if it does not include the same states of affairs
the (non-factual) state of affairs [B,h1,h2]. as f.

Combinatorialism takes facts to be literal, structured parts of the physical Very roughly then, a possible world will be a certain sort of recombination
world. This suggests that a non-factual state of affairs — a merely possible of (some portion of) the actual world, the molecular fact that includes all
fact — must be part of a merely possible physical world. This idea is at of the atomic facts. This idea will be refined in the following sections.
odds with the strong, scientifically-grounded form of actualism that
typically motivates combinatorialism. Two options are available: The 2.3.3 Structural States of Affairs and Supervenience
combinatorialist can follow the (actualist) abstractionists and define states
of affairs to be philosophical or mathematical constructs consisting only of Say that a state of affairs is structural if it is atomic and its constituent
actual objects, properties, relations, and facts. For example, the state of universal is structural or it is molecular and includes a structural state of
affairs [R,a1,...,an] can simply be identified with the ordered n-tuple affairs; and say that it is simple otherwise. The difference between
⟨R,a1,...,an⟩. So long as the combinatorialist is willing to adopt the structural and simple universals and states of affairs is particularly
additional metaphysical or set theoretic machinery, this sort of approach significant with regard to the important concept of supervenience
offers a way of introducing non-factual states of affairs that does not (Armstrong 1989, Ch 8).[45] Entity or entities S supervene on entity or
involve any untoward ontological commitments to merely possible entities R if and only if the existence of R necessitates that of S (ibid.,
entities. Alternatively, following Armstrong (1989, 46–51; 1997, 172–4), 103). (Necessitation here is, of course, ultimately to be spelled out in
the combinatorialist can refuse to grant non-factual states of affairs any terms of combinatorial possible worlds.) Non-structural states of affairs
genuine ontological status and adopt a form of modal fictionalism that supervene directly on their atomic conjuncts.[46] However, things are not
nonetheless permits one to speak as if such states of affairs exist. The in general quite so straightforward for structural states of affairs. For,
exposition to follow will remain largely neutral between these options. although structural states of affairs are ultimately constituted entirely by

44 Stanford Encyclopedia of Philosophy Summer 2015 Edition 45

 Possible Worlds Christopher Menzel

simple states of affairs, unlike non-structural states of affairs, structural pairs of states of affairs, while individually possible, are not compossible
states of affairs typically supervene on more than the totality of their — the states of affairs [having 1kg mass, a] and [having 2kg mass, a] for
constituents. For, in many cases, whether or not a structural fact exists a given object a, or, for a given mereological sum m of simples, the states
depends not only on the existence of its constituent facts but also on the of affairs [being a baboon, m] and [being a hoolock, m]. But nothing that
absence of certain others (Armstrong 1997, 34ff). For example, as noted in has been said rules out the existence of recombinations of the actual world
our example above, our water molecule W comprises two further facts in — rearrangements of its objects and universals — that include such states
which two hydrogen atoms h1 and h2 both stand in the covalent binding of affairs. Obviously, however, such recombinations cannot be thought to
relation with an oxygen atom o. However, if o were to bind with a further represent genuinely possible worlds. Of course, like the abstractionist, the
hydrogen atom h3, then, despite the fact that the constituent facts of W combinatorialist could simply stipulate as part of the definition that all
would still hold, W would not be water; there would be no such fact as legitimate recombinations must be genuinely possible states of affairs of a
W's being water.[47] Rather, W would exist only as a complex part of a certain sort, genuinely possible recombinations. But this will not do. For,
hydronium ion; the new binding [B,o,h3] would, so to say, “spoil” the like concretism, combinatorialism purports to be a reductive account of
instantiation of Water. Thus, more generally, whether or not a structural modality, an account of possible worlds that does not depend ultimately on
state of affairs S exists in a possible world typically requires something modal notions (see Armstrong 1989, 33).[51]
over and above its constituent states of affairs being “welded together” in
the right sort of way (Armstrong, 1997, 36); it requires also that there be Here the distinction between simple and structural states of affairs together
no relevant “spoilers” for S.[48] Armstrong draws directly on the initial with the combinatorialist's strong notion of supervenience come to the
passages of the Tractatus[49] for the necessary apparatus: a structural state fore. For, given that structural facts supervene on simple facts and the
of affairs S in any possible world w, supervenes, not simply on its actual totality fact T@, the actual world can be defined more
constituent atomic states of affairs but on a certain higher-order state of parsimoniously as the molecular fact that includes all the simple atomic
affairs Tw, namely, the state of affairs that the (first-order) atomic states of facts and the totality fact T@. And at the level of simples, there are no
affairs of w are all the (first-order) atomic states of affairs and, hence, that limitations whatever on recombination (Wittgenstein 1921, 2.062-2.063);
w includes no spoilers for S. Armstrong (ibid., 35, 134–5, 196–201) calls hence, any recombination of simple objects and universals is by definition
Tw the totality state of affairs for the atomic states of affairs of w.[50] considered possible. Thus Armstrong (1986a, 579):

The simple individuals, properties, and relations may be combined

2.3.4 Combinatorial Possible Worlds and Existence Therein
in all ways to yield possible [simple] atomic states of affairs,
The idea of possibility being rooted in arbitrary recombinations of the provided only that the form of atomic facts is respected. That is the
actual world, rearrangements of its objects and universals, is intuitively combinatorial idea.
appealing. Clearly, however, not just any such recombination can count as
Worlds, in particular, can be defined as special cases of such
a possible world. Some states of affairs are intuitively impossible —
recombinations, together with appropriate totality facts. To state this, we
[being an elephant, e], where e is an individual electron, say — and some

46 Stanford Encyclopedia of Philosophy Summer 2015 Edition 47

 Possible Worlds Christopher Menzel

need a condition that ensures the existence of a unique actual world: definable as follows:

(29) States of affairs s and t are identical iff they include exactly the AE3 Entity a exists in (combinatorial) possible world w =def either (i) a
same states of affairs. exists fundamentally in w or (ii) a supervenes on entities that
exist in w.
Given this, we have:
Semantics receives rather short shrift in Armstrong's version of
(30) The (combinatorial) actual world =def the fact @ that includes combinatorialism — at least, semantics in the model theoretic sense of
exactly all the simple atomic facts and the totality state of affairs §1.2 — but, as it has played an important role in our discussion of
T@ for the conjunction of those facts. concretism and abstractionism, we note briefly how the ontology of
combinatorialism might be taken to populate a possible world
AW3 w is a (combinatorial) possible world =def w is a recombination of
interpretation of the language of modal predicate logic. Specifically, we
the simple atomic facts of the actual world conjoined with the
can take the range of the modal operators — understood, semantically, as
totality fact Tw for that recombination.[52]
quantifiers — to be all of the combinatorial possible worlds in the sense of
Armstrong's ontological commitments are notoriously rather slippery but, AW3. The domain d(w) of each world w is the set of all simple and
given AW3, a reasonably complete notion of existence in a world is complex objects that exist in w according to AE3 and the w-extension
forthcoming. First, let us note that, for Armstrong, the “combinatorial Iπ(w) of a predicate π expressing a simple or complex universal R is the
idea” yields a substantial metaphysical thesis, as well, viz., the ontological set of all n-tuples, ⟨a1, ..., an⟩ such that the atomic fact [R,a1,...,an] exists
free lunch (1986, 12ff), i.e., the thesis that “[w]hat supervenes is no in w.
addition of being”; that “whatever supervenes ... is not something
2.3.5 Analytic and Emergent Modalities; Essential Properties
ontologically additional to the subvenient entity or entities.” Hence, for
Armstrong, it appears that simple states of affairs and their constituents
There are, then, for the combinatorialist no intrinsically modal
exist most fundamentally and that the existence of more complex entities
phenomena; there are just all of the various worlds that exist on
is in a certain sense derivative. Thus:
unrestricted combinatorial grounds alone. Ultimately, all genuine
(31) Entity a exists fundamentally in (combinatorial) possible world w possibilities, simple or not, are just states of affairs that exist in these
=def (i) a is a simple state of affairs that w includes or (ii) a is a combinatorial worlds in the sense of AE3. However, it is not immediately
constituent or conjunct of an entity that exists fundamentally in as clear how to understand many intuitive necessities/impossibilities
w. involving complex structural universals, for example, the impossibilities
noted in the previous section, viz., that something simultaneously have a
Given this, existence in a world generally — both fundamental and mass of both 1kg and 2kg or simultaneously be both a baboon and a
derivative — both for simples and (first-order[53]) non-simples alike, is hoolock. Likewise, it is not entirely clear how combinatorialism accounts

48 Stanford Encyclopedia of Philosophy Summer 2015 Edition 49

 Possible Worlds Christopher Menzel

for intuitive facts about essential properties, such as that our water constitute a hydronium ion I. Conversely, however, given the unrestricted
molecule W is essentially water or that Algol is essentially a dog. nature of recombination, there is a world w2 that includes W structured as
Combinatorialists argue that such modal facts can nevertheless be it actually is in w1 but which also includes the spoiler [B,o,h3] — where o
explained in terms that require no appeal to primitive modal features of the and h3 bond — and, hence, the structural state of affairs I's being
world (Armstrong 2004b, 15). hydronium. Thus, the absence of [B,o,h3] in w1 enables the emergence of
W's being water and precludes I's being hydronium whilst its presence
Analytic Modalities. Armstrong argues that many intuitive modal facts in w2 enables the emergence of the latter but precludes the former. As a
— notably, the impossibility of an object exemplifying more than one consequence, it is impossible that the states of affairs W's being water
determinate of the same determinable — can be understood ultimately as and I's being hydronium coexist.[56]
logical, or analytic, modalities that are grounded in meaning rather than
any primitive modal features of reality. For example, intuitively it is
impossible that an object simultaneously exemplify the structural
properties having 2kg mass and having 1kg mass. The combinatorial
reason for this (cf. Armstrong 1989, 79) is that, for an object a to
exemplify the former property is simply for there to be a division of a into
two wholly distinct parts, both of which exemplify the latter property.
Moreover, this division into parts is entirely arbitrary, that is, for any part
a1 of a exemplifying having 1kg mass, there is a (unique) part a2 of a Figure 2: W's being water and (given a bond between o and h3) I's being
wholly distinct from a1 that also exemplifies that property. It follows that, hydronium
if our 2kg object a itself also exemplifies having 1kg mass, then, as a is a
part of itself, there must be a 1kg part of a that is wholly distinct from a. Although more dramatic, large-scale examples of incompatible states of
And that is analytically false, false “solely by virtue of the meaning we affairs — such as a thing's being simultaneously both a baboon and a
attach to” the word ‘part’ (ibid., 80).[54] hoolock — might be vastly more complex, there is no obvious reason why
their impossibility could not have the same sort of combinatorial
Emergent Modalities. Combinatorialism purports to explain a further explanation.
class of intuitive modal facts as features that simply “emerge” from facts
about structural properties.[55] The discussion of structural states of affairs Essential Properties. It follows from the unrestricted nature of
and supervenience above provides an example. Let us suppose the actual recombination that, for any simple object a and simple universal P, a
world w1 includes our water molecule W from Figure 1 plus a further recombines with P in some worlds and fails to recombine with P in others.
hydrogen atom h3. In this world, only h1 and h2 bind to o. Hence, this Generalizing from this fact, it follows that no simple or sum of simples has
world includes the state of affairs W's being water but not the state of any simple universal or conjunction of simple universals essentially. It
affairs I's being hydronium in which o, h1, h2, and h3 are so bonded as to also follows that no such object has any structural property essentially. For

50 Stanford Encyclopedia of Philosophy Summer 2015 Edition 51

 Possible Worlds Christopher Menzel

assume o is such an object and that it exemplifies a structural property P. it exists (Armstrong 1997, 99–103, 169).[57]
Since P is structural, it supervenes on some set of simple states of affairs.
But by the nature of recombination, there are combinatorial worlds in 2.3.6 Fewer Things and Other Things: Modified Combinatorialism
which those states of affairs do not exist and, hence, in which P doesn't
but o — being a simple or a sum of simples — does. Since a possible world is a recombination of the actual world and every
recombination includes states of affairs involving every simple individual
Thick particulars like our water molecule W don't fare much better and every simple universal, by AE3, every simple entity exists in every
because of the possibility of spoilers. For Armstrong (1997, 35), W is world. Hence, there could not have been fewer of them; nor could there
simply the conjunction of its constituent states of affairs. As we've just have been simples other than the ones there actually are. In this section,
seen, however, in the presence of spoilers, that conjunction would exist — we address this issue and the issue of contingent existence generally in
hence, W would exist — without being Water. Hence, it would seem that combinatorialism.
at least some properties that, intuitively, are essential to their bearers turn
out not to be for the combinatorialist. The problem is compounded by the Fewer things. Combinatorialism as it stands has no problem accounting
fact that some intuitively non-essential properties of some thick particulars for the general intuition that there could have been fewer things. We have
are arguably essential for the combinatorialist. The shape properties of a already noted in §2.3.3 and again in §2.3.5 how our water molecule W, as
thick particular A, for example, would seem to be a function of its such, might not have existed. More generally, given the unrestricted nature
constituent states of affairs. Moreover, the exemplification of such of recombination, for any a involving a structural fact S, there are
properties are not obviously subject to spoilers the way that natural kind recombinations of the actual world wherein either (a) some of the relations
properties like Water are. Hence, as A is identical to the conjunction of its among a's constituents that are critical to S's structure fail to be
constituent states of affairs, it would seem that it will have the same shape exemplified by those constituents, or (b) there are further states of affairs
in any world in which it exists, i.e., it will have that shape essentially. included by those recombinations that act as spoilers for S. Consequently,
the combinatorialist seems to have no difficulty explaining how there
That said, combinatorialism can arguably provide a reasonably robust might have been fewer water molecules, humans, etc.
analysis of intuitions about the essential properties of ordinary thick
particulars like dogs or persons. Such objects can be taken to be temporal Intuitively, however, there isn't anything in the idea of a simple that
successions of sums of simples and each sum in the succession as its suggests that simples are necessary beings — especially if, as
temporal parts. Sums in the same rough temporal neighborhood are combinatorialists generally agree, simples are physical things of some sort
composed of roughly the same simples and are structured in roughly the and simple universals are properties of, and relations among, those things.
same way. Similarities between such objects across worlds in turn For there is nothing in the nature of a simple object to suggest that any
determine counterpart relations. Following Lewis, the essential properties given simple had to have existed. Likewise, there is nothing in the nature
of such objects can then be identified with those properties exemplified by of a simple universal to suggest it had to have been exemplified and,
(all of the temporal parts of) all of its counterparts in every world in which hence, on the combinatorialist's own conception of universals, that it had

52 Stanford Encyclopedia of Philosophy Summer 2015 Edition 53

 Possible Worlds Christopher Menzel

to exist. Otherwise put, as simples exist only insofar as they are account for the possibility of conjunctive and structural universals that are
constituents of facts, there seems no reason why there couldn't have been a simply rearrangements of actual simples. It is not implausible to think that
very small number of facts, indeed, just a single simple, atomic, monadic such recombinations can give rise to, say, exotic biological kinds that have
fact and, hence, a lone simple object and a lone simple universal. no actual instances (Armstrong 1989, 55–56). Thus, in particular,
combinatorialism seems quite able to provide the truth condition (24) for
In fact, however, AW3 can be easily modified to accomodate these (23) and, hence, can account for some possibilities involving “missing”
intuitions without doing any serious violence to combinatorialist universals that, intuitively, ought to be possible.
intuitions. Specifically, the combinatorialist can admit “contracted” worlds
in which fewer simples exist by allowing any recombination of any simple However, it is far from clear that such possibilities exhaust the modal
fact — that is, equivalently, by allowing any state of affairs — to count as intuition that other things could have existed. Notably, intuitively, there
a possible world: could have been different simple universals distinct from any that actually
exist — different fundamental properties of simples, for example.
AW3ʹ′ʹ′ w is a (combinatorial) possible world =def w is a recombination of Likewise for simple objects. Either way, there seems to be nothing in the
some simple fact f conjoined with the totality state of affairs Tw idea of a simple object or simple universal that suggests there couldn't
for that recombination. have been simples other than, or in addition to, the simples there are in
fact. But AW3ʹ′ does not allow for this; the simples of every possible
AE3 requires no modification, as it was defined with sufficient generality world are a subset of the actual simples and there is no obvious way of
above. Under AW3ʹ′ʹ′, however, AE3 entails that all entities alike — modifying the principle to accommodate the intuition. Nor is there any
objects and universals, simple and structural — are contingent and, obvious way of modifying the principle to accomodate the intuition in
indeed, that every simple object is the sole constituent of some question.[59]
combinatorial possible world.
The combinatorialist could of course abandon actualism and include
Other things. Intuitively, not only could there have been fewer things, merely possible simples into her ontology. Again, she could follow the
there could have been more things or, more generally, things other than new actualists and draw a division between actually concrete and non-
those that actually exist. As above, combinatorialism as it stands seems actual, possibly concrete simples; or she could introduce Plantinga-style
able to account for many instances of this intuition: Figure 2 illustrates haecceities to go proxy for merely possible simples. But all of these
how a non-actual hydronium ion I might exist in another world. Likewise, options would be badly out of step with the strong, naturalist motivations
there seems no reason to deny, e.g., that there are rearrangements w of the for combinatorialism: There is but the one physical world comprising all
actual world's simples wherein exist all of the human beings that actually of the facts; recombinations of (at least some of) those facts — arbitrary
exist (at, say, 0000GMT 1 January 2013) and more besides that are rearrangements of their simple objects and universals — determine the
composed of simples that, in fact, constitute things other than human possible worlds. Mere possibilia, merely possible non-concretia, and non-
beings (Armstrong 1997, 165).[58] Combinatorialism also seems able to qualitative haecceities have no real place in that picture.

54 Stanford Encyclopedia of Philosophy Summer 2015 Edition 55

 Possible Worlds Christopher Menzel

The “purest” option for the combinatorialist is simply to brazen it out and statements, ultimately, are true or false in virtue of how things stand with
argue that the actual simples are, in fact, all the simples there could be respect to worlds that are themselves defined in non-modal terms.
(Armstrong 1989, 54ff; Driggers 2011, 56–61). A more robust option
suggested by Skyrms (1981) makes some headway against the problem by Combinatorialism's ontological modesty, however, is also a weakness.
introducing an “outer”, or “second-grade” realm of possibility, but at the For, unlike, the two competing approaches, there are modal intuitions that
cost of moving beyond the basic intuitions of combinatorialism the combinatorialist is not easily able to account for, notably, the intuition
(Armstrong 1989, 60; 1997, 165–167). Finally, Sider (2005, 681) suggests that there could have been other things. Additional difficulties are
that combinatorialists who (like Armstrong) are modal fictionalists can discussed in the supplemental document Further Problems for
deal with the problem of missing entities simply by appealing to yet more Combinatorialism.
fictionalism: As the combinatorialist fiction already includes non-actual
states of affairs with actually existing constituents, there seems no reason Bibliography
not to extend the fiction to include non-actual states of affairs whose
Adams, R., 1974. ‘Theories of Actuality’, Noûs, 8: 211–31; reprinted in
constituents include non-actual particulars and universals. Fictionalism
Loux (1979): 190–209
itself, however, leaves the combinatorialist with the deep problems
–––, 1981. ‘Actualism and Thisness’, Synthese, 49: 3–41
detailed by Kim (1986), Lycan (1993), and Rosen (1993).[60]
Armstrong, D. M., 1978a, Universals and Scientific Realism, Volume I:
2.3.7 A Brief Assessment of Combinatorialism Nominalism and Realism, Cambridge: Cambridge University Press.
–––, 1978b, Universals and Scientific Realism, Volume II: A Theory of
As with concretism and abstractionism, combinatorialism provides Universals, Cambridge: Cambridge University Press.
reasonably clear definitions of possible worlds and existence in a world –––, 1986a, ‘The Nature of Possibility’, The Canadian Journal of
and is noteworthy for its attempt to avoid what might be thought of as the Philosophy, 16(4): 575–594.
metaphysical excesses of the two competing views. In contrast to –––, 1986b. ‘In Defense of Structural Universals’, Australasian Journal of
concretism, combinatorialism is staunchly actualist: instead of an infinity Philosophy, 64(1): 85–88.
of alternative physical universes, each with its own unique inhabitants –––, 1989. A Combinatorial Theory of Possibility, New York: Cambridge
existing as robustly as the inhabitants of the actual world, the worlds of University Press.
combinatorialism are simply rearrangements of the universals and –––, 1993. ‘A World of States of Affairs’, in J. Tomberlin (ed.),
particulars of the actual world; and commitment even to them might be Philosophical Perspectives, 7: 429–440.
avoided if some version of fictionalism is tenable. Likewise, in contrast to –––, 1997. A World of States of Affairs, New York: Cambridge University
abstractionism's rather rich and unrestrained ontology of SOAs, Press.
combinatorialism's states of affairs are comparatively modest. Moreover, –––, 2004a. ‘Theorie Combinatoire Revue et Corrigée’, in J-M. Monnoyer
in contrast to nearly all versions of abstractionism, combinatorialism (Ed.) La Structure du Mond: Objets, Propriétés, États et Choses,
shares with concretism the virtue of a reductive theory of modality: Modal Paris: Vrin, 185–198.

56 Stanford Encyclopedia of Philosophy Summer 2015 Edition 57

 Possible Worlds Christopher Menzel

–––, 2004b. ‘Combinatorialism Revisited’, translation (with minor –––, 2008. ‘Concrete Possible Worlds’, in T. Sider, J. Hawthorne, and D.
corrections) of Armstrong 2004a, URL = Zimmerman (eds.), Contemporary Debates in Metaphysics Blackwell
http://eprints.nottingham.ac.uk/716/, ISBN: 2-7116-1627-4. Publishing, 111–134.
–––, 2004c. Truth and Truthmakers, Cambridge: Cambridge University Bringsjord, S., 1985. ‘Are There Set Theoretic Possible Worlds?’,
Press. Analysis, 45(1): 64.
–––, 2009. ‘Reply to Keller,’, in Monnoyer 2007, 157–162. Bueno, O., C. Menzel, and E. Zalta, 2014. ‘Worlds and Propositions Set
Barcan, R., 1946. ‘A Functional Calculus of First Order Based on Strict Free’, Erkenntnis, 79: 797–820.
Implication’, Journal of Symbolic Logic, 11: 1–16. Carnap, R., 1947. Meaning and Necessity. Chicago: The University of
Bayart, A., 1958, ‘La correction de la logique modale du premier et Chicago Press.
second ordre S5’, Logique et Analyse, 1: 28–44. Translated in Chihara, C., 1998. The Worlds of Possibility, Oxford: Clarendon Press
Cresswell forthcoming. Copeland, B. J., 2002. ‘The Genesis of Possible Worlds Semantics’,
–––, 1959, ‘Quasi-adéquation de la logique modale de second ordre S5 et Journal of Philosophical Logic, 31: 99–137.
adéquation de la logique modale de premier ordre S5’, Logique et –––, 2006. ‘Meredith, Prior, and the History of Possible Worlds
Analyse, 2: 99–121. Semantics’, Synthese, 150(3): 373–397.
Beall, J., 2000. ‘A Neglected Response to the Grim result’, Analysis, Cresswell, M. J., 1972. ‘The World is Everything that is the Case’,
60(1): 38–41. Journal of Philosophy, 50: 1–13. Reprinted in Loux (1979), 129–145.
Bigelow, J. and R. Pargetter, 1987. ‘Beyond the Blank Stare’, Theoria, 53: –––, 1973. Logics and Languages, London, Methuen.
97–114. –––, 1985a. Structured Meanings: The Semantics of Propositional
Bradley, R., 1989. ‘Possibility and Combinatorialism: Wittgenstein Versus Attitudes, Bradford Books/MIT Press.
Armstrong’, Canadian Journal of Philosophy, 19(1): 15–41. –––, 1985b. Adverbial Modification. Dordrecht, Reidel.
Bricker, P., 1980. ‘Prudence’, Journal of Philosophy, 77(7): 381–401. –––, 1988. Semantical Essays: PossibleWorlds and Their Rivals.
–––, 1987. ‘Reducing Possible Worlds to Language’, Philosophical Dordrecht, Kluwer Academic Publishers.
Studies, 52(3): 331–355. –––, 1990. Entities and Indices. Dordrecht, Kluwer.
–––, 1996. ‘Isolation and Unification: The Realist Analysis of Possible –––, 1994. Language in the World. Cambridge: Cambridge University
Worlds’, Philosophical Studies, 84(2/3): 225–238. Press.
–––, 2001. ‘Island Universes and the Analysis of Modality’, in G. Preyer, –––, 1996. Semantic Indexicality. Dordrecht: Kluwer.
F. Siebelt (eds.), Reality and Humean Supervenience: Essays on the –––, 2004. ‘Adequacy Conditions for Counterpart Theory’, Australasian
Philosophy of David Lewis, Rowman and Littlefield. Journal of Philosophy, 82(1): 28–41.
–––, 2006. ‘The Relation Between General and Particular: Supervenience –––, forthcoming. ‘Arnould Bayart’s Modal Completeness Theorems’,
vs. Entailment’, in D. Zimmerman (ed.), Oxford Papers in Logique et Analyse.
Metaphysics, vol. 3, Oxford: Oxford University Press, 251–287. Darby G., and D. Watson, 2010. ‘Lewis's Principle of Recombination:

58 Stanford Encyclopedia of Philosophy Summer 2015 Edition 59

 Possible Worlds Christopher Menzel

Reply to Efird and Stoneham’, Dialectica, 64(3): 435–445. Fitch, G. W., 1996, ‘In Defense of Aristotelian Actualism’, Philosophical
Davies, M., 1981. Meaning, Quantification, Necessity: Themes in Perspectives, 10: 53–71.
Philosophical Logic, London: Routledge and Kegan Paul. Forrest, P. and D. M. Armstrong, 1984. ‘An Argument against David
DeRosset, L., 2009a. ‘Possible Worlds I: Modal Realism’, Philosophy Lewis's Theory of Possible Worlds’, Australasian Journal of
Compass, 4(6): 998–1008. Philosophy, 62: 164–168.
–––, 2009b. ‘Possible Worlds II: Non-reductive Theories of Possible Garson, J., 2006. Modal Logic for Philosophers. New York: Cambridge
Worlds’, Philosophy Compass, 4(6): 1009–1021. University Press.
Divers, J., 2002. Possible Worlds, London: Routledge. Goldblatt, R., 2003. “Mathematical Modal Logic: A View of its
Divers, J. and J. Melia, 2002. ‘The Analytic Limit of Genuine Modal Evolution”, in D. M. Gabbay and J. Woods (eds.), Handbook of the
Realism’, Mind, 111(441): 15–36. History of Logic, Vol. 7: Logic and the Modalities in the Twentieth
Driggers, R. K., 2011. ‘M-Combinatorialism and the Semantics of Century. Amsterdam: Elsevier, pp. 1–98.
SQML’, M.A. Thesis, Texas A&M University. URL = Grim, P., 1986. ‘On Sets and Worlds: A Reply to Menzel’, Analysis,
http://hdl.handle.net/1969.1/ETD-TAMU-2011-05-9484 46(4): 186–191.
Eddon, M., 2007. ‘Armstrong on Quantities and Resemblance’, Hazen, A., 1979. ‘Counterpart Theoretic Semantic for Modal logic’,
Philosophical Studies, 136(3): 385–404. Journal of Philosophy, 76: 319–338.
Efird, D. and T. Stoneham, 2008. ‘What is the Principle of Hunter, G. and W. Seager, 1981. ‘The Discreet Charm of Counterpart
Recombination?’, Dialectica, 62(4): 483–494. Theory’, Analysis, 41(2): 73–76.
Egan, A., 2004. ‘Second-order predication and the Metaphysics of Heil, J., 2007. ‘The Legacy of Linguisticism’, Australasian Journal of
Properties’, Australasian Journal of Philosophy, 82(1): 48–66. Philosophy, 84(2): 233–244.
Etchemendy, J., 1990. The Concept of Logical Consequence, Cambridge, Hintikka, J., 1957. ‘Modality as referential multiplicity’, Ajatus, 20: 49–
MA: Harvard University Press. (Reissued by Stanford: CSLI 64.
Publications, 1999.) –––, 1961. ‘Modality and Quantification’, Theoria, 27: 119–28.
Feldman, F., 1971. ‘Counterparts’, Journal of Philosophy, 68: 406–9. Hughes, G. and Cresswell, M., 1968. An Introduction to Modal Logic,
Fara, D. G., 2009. ‘Dear Haecceitism’, Erkenntnis, 70: 285–297. London: Methuen.
Fara, M. and T. Williamson, 2005. ‘Counterparts and Actuality’, Mind, Jager, T., 1982. ‘An Actualist Semantics for Quantified Modal Logic’,
114(453): 1–30. Notre Dame Journal of Formal Logic, 23(3) (July): 335–49.
Fine, K., 1977. ‘Postscript’, in Prior 1977, 116–161. Kanger, S., 1957. Provability in Logic, Stockholm: Almqvist and Wiksell.
–––, 1978. ‘Model Theory for Modal Logics: Part I — The De Re/De Kaplan, D., 1975. ‘How to Russell a Frege-Church’, Journal of
Dicto Distinction’, Journal of Philosophical Logic, 7: 125–56. Philosophy, 72: 716–29; reprinted in Loux (1979), pp. 210–24.
Forbes, G., 1985. The Metaphysics of Modality, Oxford: Clarendon Press. Kaplan, D., 1979. ‘Transworld Heir Lines’, in Loux (1979), pp. 88–109.
–––, 1982. ‘Canonical Counterpart Theory’, Analysis, 42(1): 33–37. –––, 1995. ‘A Problem in Possible World Semantics’, in W. Sinnott-

60 Stanford Encyclopedia of Philosophy Summer 2015 Edition 61

 Possible Worlds Christopher Menzel

Armstrong, D. Raffman, and N. Asher (eds.), Modality, Morality and Possibility’, Australasian Journal of Philosophy, 70(2): 211–224.
Belief: Essays in Honor of Ruth Barcan Marcus, Cambridge: Lewis, C. I. and C. H. Langford, 1932. Symbolic Logic, New York: The
Cambridge University Press, 41–52. Appleton-Century Company; reprinted in paperback by New York:
Keller, P., 2007. ‘A World of Truthmakers’, in Monnoyer 2007, 105–156. Dover Publications, 1951.
–––, 2009. ‘Review of D. M. Armstrong, Truth and Truthmakers’, Mind, Linsky, B. and E. Zalta, 1994. ‘In Defense of the Simplest Quantified
118(472): 1101–1105. Modal Logic’, in J. Tomberlin (ed.), Philosophical Perspectives, 8:
Kemeny, J. G., 1956a. ‘A New Approach to Semantics — Part I’, The 431–458.
Journal of Symbolic Logic, 21(1): 1–27. –––, 1996. ‘In Defense of the Contingently Concrete’, Philosophical
–––, 1956b. ‘A New Approach to Semantics — Part II’, The Journal of Studies, 84: 283–294.
Symbolic Logic, 21(2): 149–161. Loux, M. (ed.), 1979. The Possible and the Actual, Ithaca: Cornell.
Kim, J., 1986. ‘Possible Worlds and Armstrong’s Combinatorialism’, Lycan, W., 1988. ‘On the Plurality of Worlds by David Lewis’, The
Canadian Journal of Philosophy, 16(4): 595–612. Journal of Philosophy, 85(1): 42–47.
Kripke, S., 1959. ‘A Completeness Theorem in Modal Logic’, Journal of –––, 1991. ‘Two — No, Three — Concepts of Possible Worlds’,
Symbolic Logic, 24(1): 1–14. Proceedings of the Aristotelian Society, New Series, 91: 215–227.
–––, 1963a. ‘Semantical Analysis of Modal Logic I: Normal Modal –––, 1993. ‘Armstrong's New Combinatorial Theory of Modality’, in J.
Propositional Calculi’, Zeitschrift für Mathematische Logik und Bacon, K. Campbell, and L. Reinhardt (eds.), Ontology, Causality,
Grundlagen der Mathematik, 9: 67–96. and Mind: Essays in Honour of D. M. Armstrong, Cambridge:
–––, 1963b. ‘Semantical Considerations on Modal Logic’, Acta Cambridge University Press, 3–22.
Philosophica Fennica, 16: 83–94. Maddy, P., 1980. ‘Perception and Mathematical Intuition’, The
–––, 1972. ‘Naming and Necessity’, Cambridge, Massachusetts: Harvard, Philosophical Review, 89(2): 163–196.
1980. Mates, B., 1968. ‘Leibniz on Possible Worlds’, in B. van Rootselaar and J.
Lewis, D., 1968. ‘Counterpart Theory and Quantified Modal Logic’, F. Staal (eds.), Logic, Methodology and Philosophy of Science, vol. 3,
Journal of Philosophy, 65: 113–126. Amsterdam: North-Holland Publishing Company, 507–29; reprinted
–––, 1970. ‘General Semantics’, Synthese, 22: 18–67. in H. Frankfurt (ed.), Leibniz, Notre Dame: University of Notre Dame
–––, 1973. Counterfactuals, Cambridge, Massachussetts: Harvard Press (1972), 335–364.
University Press. Marcus, R., 1986. ‘Possibilia and Possible Worlds’, Grazer
–––, 1986. On The Plurality of Worlds, Oxford: Blackwell. Philosophische Studien, 25/26 (1985/1986): 107–33.
–––, 1986a. ‘Against Structural Universals’, Australasian Journal of McDaniel, K., 2006. ‘Modal Realisms”, Philosophical Perspectives, 20:
Philosophy, 64(1): 25–46. 303–331.
–––, 1991. Parts of Classes, Oxford: Basil Blackwell Ltd. McNamara, P., 1993. ‘Does the Actual World Actually Exist?’,
–––, 1992, ‘Critical Notice: Armstrong, D. M., A Combinatorial Theory of Philosophical Studies, 69: 59–81.

62 Stanford Encyclopedia of Philosophy Summer 2015 Edition 63

 Possible Worlds Christopher Menzel

Menzel, C., 1986. ‘On Set Theoretic Possible Worlds’, Analysis, 46(2): Logic, 23: 253–265.
68–72. Novakofski, J. M. S. Brewer, N. Mateus-Pinilla, J. Killefer, and R. H.
–––, 1989. ‘On an Unsound Proof of the Existence of Possible Worlds’, McCusker, 2005. ‘Prion Biology Relevant to Bovine Spongiform
Notre Dame Journal of Formal Logic, 30(4): 598–603. Encephalopathy’, Journal of Animal Science, 83(6): 1455–1476
–––, 1990. ‘Actualism, Ontological Commitment, and Possible Worlds [available online].
Semantics’, Synthese, 85: 355–89. Pickavance, T., 2014. ‘Bare Particulars and Exemplification’, American
–––, 1991. ‘The True Modal Logic’, Journal of Philosophical Logic, 20: Philosophical Quarterly, 51(2): 95–108.
331–374. Plantinga, A., 1974. The Nature of Necessity, Oxford: Oxford University
–––, 2012. ‘Sets and Worlds Again’, Analysis, 72(2): 304–309. Press.
Menzel, C., and E. Zalta, 2014, ‘The Fundamental Theorem of World –––, 1976. ‘Actualism and Possible Worlds’, Theoria, 42: 139–60;
Theory’, Journal of Philosophical Logic, 43(2): 333–363. reprinted in Loux (1979): 253–73.
Meredith, C. and A. Prior, 1956. ‘Interpretations of Different Modal –––, 1985. ‘Replies’, in Alvin Plantinga, J. Tomberlin and P. van Inwagen
Logics in the “Property Calculus”’, in Copeland, B. (ed.) 1996, Logic (eds.), Dordrecht: D. Reidel: 313–96.
and Reality: Essays on the Legacy of Arthur Prior, Oxford: –––, 1987. ‘Two Concepts of Modality: Realism and Modal
Clarendon Press. Reductionism’, in J. Tomberlin (ed.), Philosophical Perspectives, 1:
Merricks, T., 2003. ‘The End of Counterpart Theory ’, The Journal of 189–231.
Philosophy, 100(10): 521–549. Pollock, J., 1985. ‘Plantinga on Possible Worlds’, in Alvin Plantinga, J.
Merrill, G., 1978. ‘Formalization, Possible Worlds and the Foundations of Tomberlin and P. van Inwagen (eds.), Dordrecht: D. Reidel, pp. 121–
Modal Logic’, Erkenntnis, 12: 305–327. 44.
Monnoyer, J. (ed.), 2007. Metaphysics and Truthmakers, Frankfurt am Prior, A. N., 1952. ‘Modality de dicto and modality de re, Theoria, 18(3):
Main: Ontos Verlag. 174–180.
Montague, R., 1974. Formal Philosophy, New Haven, CT: Yale –––, 1977. Worlds, Times, and Selves, Amherst: University of
University Press Massachusetts Press.
Moreland, J. P., 2011. ‘Exemplification and Constituent Realism: A –––, 1956. ‘Modality and Quantification in S5’, Journal of Symbolic
Clarification and Modest Defense’, Axiomathes DOI Logic, 21: 60–2.
10.1007/s10516-011-9148-x Pruss, A., 2001. ‘The Cardinality Objection to David Lewis's Modal
Nolan, D., 1996. ‘Recombination Unbound’, Philosophical Studies, Realism’, Philosophical Studies, 104: 169–178.
84(2/3): 239–262. Quine, W. V. O., 1948. ‘On What There Is’, in From a Logical Point of
Nortmann, U., 2002. ‘The Logic of Necessity in Aristotle: An Outline of View, New York: Harper, 1953, 1–19.
Approaches to the Modal Syllogistic, Together with a General –––, 1953. ‘Three Grades of Modal Involvement’, Proceedings of the XIth
Account of de dicto- and de re-Necessity’, History and Philosophy of International Congress of Philosophy, 14: 65–81.

64 Stanford Encyclopedia of Philosophy Summer 2015 Edition 65

 Possible Worlds Christopher Menzel

–––, 1956. ‘Quantifiers and Propositional Attitudes’, The Journal of Oxford: Blackwell, pp. 98–112.
Philosophy, 53: 177–187. –––, 1976. ‘Possible Worlds’, Noûs, 10(1): 65–75.
–––, 1960. Word and Object, Cambridge, MA: MIT Press. –––, 1987. Inquiry, Boston: Bradford Books, MIT Press.
–––, 1968. ‘Propositional Objects’, Critica: Revista Hispanoamericana de –––, 2012. Mere Possibilities: Metaphysical Foundations of Modal
Filosofia, 2(5): 3–29. Reprinted in W. V. Quine, 1977, Ontological Semantics, Princeton: Princeton University Press.
Relativity, Columbia University Press: 139–160. Tarski, A., 1933. ‘The concept of truth in the languages of the deductive
Rosen, G., 1993. ‘A Problem for Fictionalism about Possible Worlds’, sciences’ (Polish), Prace Towarzystwa Naukowego Warszawskiego,
Analysis, 53(2): 71–81. Wydzial III Nauk Matematyczno-Fizycznych, 34, Warsaw; expanded
Roy, T., 1995. ‘In Defense of Linguistic Ersatzism’, Philosophical English translation in Tarski (1983), 152–278.
Studies, 80(3): 217–242. –––, 1944. ‘The semantic conception of truth’, Philosophy and
Russell, B., 1918/1919. ‘The Philosophy of Logical Atomism’, The Phenomenological Research, 4: 13–47.
Monist, 28: 495–527; 29: 32–63, 190–222, 345–380. Reprinted in –––, 1983. Logic, Semantics, Metamathematics, papers from 1923 to
Russell 1956, 177–281 and published in book form as Russell 1985. 1938, edited by John Corcoran. Indianapolis: Hackett Publishing
–––, 1956. Logic and Knowledge, London: Allen and Unwin. Company.
–––, 1985. The Philosophy of Logical Atomism, with an introduction by D. Tomberlin, J., and van Inwagen, P. (eds.), 1985. Profiles: Alvin Plantinga,
Pears, London: Routledge. Dordrecht: D. Reidel, pp. 121–44.
Salmon, N., 1988. ‘Review of On the Plurality of Worlds, Philosophical van Inwagen, P., 1986. ‘Two Concepts of Possible Worlds’, in Midwest
Review, 97(2): 237–244. Studies in Philosophy, XI, P. French, T. Uehling, and H. Wettstein
Shalkowski, S., 1994. ‘The Ontological Ground of the Alethic Modality’, (eds.), Minneapolis: University of Minnesota Press, 185–213.
The Philosophical Review, 103(4): 669–688. –––, 2008. ‘McGinn on Existence’, The Philosophical Quarterly, 58(20):
Schneider, S., 2001. ‘Alien Individuals, Alien Universals, and Armstrong's 36–58.
Combinatorial Theory of Possibility’, The Southern Journal of Washington, C., 1998. ‘Use/Mention Distinction and Quotation’, in
Philosophy, 39: 575–593. Routledge Encyclopedia of Philosophy, vol. 9, W. Craig (ed.),
Sider, T., 2002. ‘The Ersatz Pluriverse’, Journal of Philosophy, 99(6): London and New York: Routledge.
279–315. Williamson, T., 1998. ‘Bare Possibilia’, Erkenntnis, 48: 257–273.
–––, 2005. ‘Another Look at Armstrong's Combinatorialism’, Noûs, 39(4): –––, 2000. ‘The Necessary Framework of Objects’, Topoi, 19: 201–208.
679–695. –––, 2013. Modal Logic as Metaphysics. Oxford: Oxford University Press.
Skyrms, B., 1981. ‘Tractarian Nominalism’, Philosophical Studies, 40(2): Wittgenstein, L., 1921. ‘Logisch-Philosophische Abhandlung’, with a
199–206. forward by Bertrand Russell, Annalen der Naturphilosophie, 14,
Stalnaker, R., 1968. ‘A Theory of Conditionals’ in Studies in Logical published by Wilhelm Ostwald, Leipzig: Verlag Unesma: 185–262.
Theory, American Philosophical Quarterly Monograph Series, 2. Also available online in HTML, PDF, and ePub formats in a side-by-

66 Stanford Encyclopedia of Philosophy Summer 2015 Edition 67

 Possible Worlds Christopher Menzel

side presentation with the translations Wittgenstein (1922) and modal ersatzism (roughly what is referred to as abstractionism in
Wittgenstein (1974) at http://people.umass.edu/klement/tlp/. this entry)
–––, 1922. Tractatus Logico-Philosophicus, trans. of Wittgenstein (1921) modal fictionalism
by C. K. Ogden, Routledge & Kegan Paul. modal realism (referred to as "concretism" in this entry)
–––, 1974. Tractatus Logico-Philosophicus, revised edition, trans. of possible worlds
Wittgenstein (1921) by D. F. Pears and B. F. McGuinness, New York possible world semantics
and London: Routledge & Kegan Paul. Theories of modality
Yagisawa, T., 2010. Worlds and Individuals, Possible and Otherwise, modal primitivism
Oxford: Oxford University Press. universals
Zalta, E., 1983. Abstract Objects: An Introduction to Axiomatic
Metaphysics, Dordrecht: D. Reidel. Related Entries
–––, 1993. ‘Twenty-Five Basic Theorems in Situation and World Theory’,
Journal of Philosophical Logic, 22(4): 385–428. actualism | essential vs. accidental properties | identity: transworld | Lewis,
David: metaphysics | logic, history of: modal logic | logic: classical | logic:
Academic Tools intensional | logic: modal | modality: epistemology of | natural kinds |
possible objects | Prior, Arthur | rigid designators | states of affairs
How to cite this entry.
Acknowledgments
Preview the PDF version of this entry at the Friends of the SEP
Society. The author wishes to express his deep gratitude to Phillip Bricker and Max
Look up this entry topic at the Indiana Philosophy Ontology Cresswell for extensive comments on several drafts of this entry and for
Project (InPhO). numerous illuminating discussions of its content and related topics. The
Enhanced bibliography for this entry at PhilPapers, with links entry is vastly better for their generous input. Errors and other infelicities
to its database. that remain are of course the sole responsibility of the author. A great deal
of this entry was written with the support of the Alexander von Humboldt
Other Internet Resources Foundation while the author was a Visiting Fellow at the Munich Center
for Mathematical Philosophy in 2011–12. Thanks are due to the Center's
Browse papers by related topics at Phil Papers: director, Professor Hannes Leitgeb, for making the author's stay at this
actualism remarkable venue possible. Finally, the author would like to express his
counterpart theory thanks to the SEP Editors for their extraordinary patience in dealing with
essentialism and de re modality the very tardy author of a badly-needed entry.
essentialism and quantified modal logic

68 Stanford Encyclopedia of Philosophy Summer 2015 Edition 69

 Possible Worlds Christopher Menzel

The Extensionality of Possible World Semantics extensional first-order logic in a dedicated language. It is in this clear
sense that basic possible world semantics is an extensional semantic
As noted, possible world semantics does not make modal logic itself theory for modal languages. It can therefore be said that modal logic with
extensional; the substitutivity principles all remain invalid for modal a basic possible world semantics is itself extensional in a derivative sense:
languages under (basic) possible worlds semantics. Rather, it is the the logic that gives full expression to the meanings of modal sentences is
semantic theory itself — more exactly, the logic in which the theory is extensional in the primary sense that, in that logic, all substitutivity
expressed — that is extensional. Specifically, basic possible world principles are valid.
semantics for a given modal language ℒ, when formalized, is expressed in
a (non-modal) first-order language ℒPWS that contains the set membership Return to the article “Possible Worlds”
predicate ‘∈’ supplemented with dedicated predicates, function symbols,
and constants, as well as mechanisms for talking about the expressions of Further Problems for Concretism
ℒ and their possible world interpretations, notably:
The Charge of Circularity
World(w): w is a world
Lycan (1988, 1991) and others have disputed the claim that Lewis's
T(φ,w): (formula) φ is true at (world) w
account is free of any modalities and hence have questioned whether it
dom(w): the domain of world w
really provides an analysis of our modal intuitive notions. The usual
ext(π,w): the extension of (n-place predicate) π at world w
charge is that 'world' for Lewis can only mean 'possible world', lest, e.g.,
den(τ): the denotation of (constant or variable) τ
there be Lewisian worlds containing impossible objects. It is not clear that
@: the actual world
this charge has any merit. If an astrophysicist were to conjecture for
Thus, for example, by formalizing the definition of truth simpliciter as theoretical reasons the existence of a distant galaxy beyond the reach of
truth in the actual world: any telescope, we would not say that by 'galaxy' the astrophysicist can
only mean 'possible galaxy'. We would say simply that the astrophysicist
True(φ) =def T(φ,@) is conjecturing that there is yet another thing of an already familiar sort,
viz., galaxies. Modality need not enter the picture at all. It seems that an
the complete statement of the truth conditions for (6), expressed more
analogous response is open to Lewis. He is simply conjecturing for
formally in ℒPWS, take on the following form:
theoretical reasons that there is yet another thing — indeed many other
True(‘◻∀x(Px → Mx)’) ↔ ∀w∀x((World(w) ∧ x ∈ dom(w)) → (x ∉ things — of an already familiar sort, viz., worlds and, on the face of it
ext(‘ P’,w) ∨ x ∈ ext(‘M’,w))). anyway, those entities are themselves analyzed in non-modal terms, viz.,
spatiotemporal relations and mereology. Granted, the picture is a bit
The semantic theory for this language ℒPWS, of course, is just our Tarskian muddier in Lewis's case insofar as he wants to claim that the intuitive
semantics above. Thus, the logic of possible world semantics is simply an possibilities include universes in which the laws of nature are quite a lot

70 Stanford Encyclopedia of Philosophy Summer 2015 Edition 71

 Possible Worlds Christopher Menzel

different from those in our universe and, hence, that there are worlds in our semantical intuitions tell us otherwise. Other things being equal,
which these possibilities are realized. But it is not clear how Lewis's dissonance with strong intuitions can be decisive in a philosophical
commitment even to such remote worlds requires him to acknowledge a dispute. However, while consanance with intuitions is desirable,
primitive modality. Likewise, we may have strong intuitions that, say, it is semantical intuitions can be trumped by a theory's explanatory breadth.
impossible that there be stones capable of thought or that 5 plus 7 be Thus, as he does in response to the incredulous stare, Lewis would likely
something other than 12, and hence that there are no worlds in which those argue that we are entirely warranted in ignoring the intuitions in question
propositions are true. But there is no obvious point at which Lewis is so long as there is no theory that accommodates them whose explanatory
forced to say that there cannot be any such worlds. Rather, all he needs to power is at least roughly equivalent to his theory.
say is that there is no reason to think that the existence of such worlds is in
any way suggested by the recombination principles R1 and R2. (For more Coarse-grained Intensions
in defense of Lewis on this point, see Divers and Melia 2002.)
Lewis inherits a general problem for the possible world analysis of
The Charge of Irrelevance intensional entities, namely, that they are too "coarse-grained". A
proposition p is necessary just in case it is true in all possible worlds. On
A frequent objection raised against Lewis — presented forcefully by, e.g., Lewis's analysis, p is true in w just in case w ∈ p. It follows that p is
Salmon (1988) — is that, in general, the truth value of a de re modal necessary just in case, for all worlds w, w ∈ p, i.e., just in case p is the set
statement is obviously not determined in any way by how things stand of all possible worlds. As sets with the same members are identical, it
among entities that bear absolutely no spatial or temporal connection to follows that, on the concretist analysis, there is only one necessary truth.
the actual world, among entities that bear no intrinsic connection to us. In This is intuitively problematic. The propositions that all red things have
a word, what goes on in Lewis's worlds is largely irrelevant to the modal a color and that there are infinitely many prime numbers are,
facts of our world. Intuitively, that Socrates could have been a politician intuitively, quite distinct; one is about a color, the other about a certain
has something intrinsically to do with Socrates himself. On Lewis's telling, type of natural number. Both, however, are necessary and are identical
however, whether Socrates could have been a politician is determined by according to the analysis above, intuitions to the contrary notwithstanding.
what happens to someone else, somewhere else. Intuitively, however, what Similar problems arise for properties and relations.
happens to someone other than Socrates in some other world — even a
spitting image of him in a world that is a spitting image of ours — has no Lewis (1986, 55–59) is of course well-aware of this objection and mounts
more to do with what could have happened to him than does what happens a sophisticated response. His basic line is that there are numerous
to someone who resembles him in, say, New York City. conceptions of propositions, properties, and relations, only one of which,
admittedly coarse-grained, is captured by the definitions in §2.1.4. More
For all its intuitive force, the objection is arguably question-begging: complex set theoretic constructions however can be defined that play the
Lewis's claim is precisely that the allegedly irrelevant connections are roles of finer-grained intensions as needed. (See §2.2.2 for further
what determine modal truth values. The objection, at root, is simply that discussion.)

72 Stanford Encyclopedia of Philosophy Summer 2015 Edition 73

 Possible Worlds Christopher Menzel

Challenges to the Completeness of Lewis's Program concretist unable to countenance the intuitive possibility that the actual
world comprises multiple island universes without abandoning the
Island Universes indexical account of ‘actual’ and its cognates and invoking instead an
absolute metaphysical distinction between those worlds that are actual and
A surprisingly simple challenge to completeness is the problem of island those that are merely possible, a move that Lewis would in no wise accept.
universes. Intuitively, one way things could be is that there are multiple
physical universes each of which is spatio-temporally unrelated to all of Alien Properties
the others. Why could it not be the case, for example, that the actual world
consists of exactly two, symmetrical, physical universes, each the mirror Another challenge to the concretist concerns so-called “alien properties”
image of the other, as described in Adams 1981, albeit, in addition, spatio- (Lewis 1986, 91–92). Intuitively, there could have been things of a very
temporally unconnected from each other? For the concretist, such different kind from very different from anything in the actual world, so
purported possibilities appear to be logically incoherent. For a physical different, in fact that they are not reasonably thought of as the result of
universe is a maximal object in the concretist sense — an object such that reorganizing duplicates of parts of the actual world — a very different sort
anything spatio-temporally related to it is part of it. Hence, the translation of fundamental particle, say. But if so, the challenge goes, recombination
of "Possibly, there are multiple spatio-temporally unrelated physical does not suffice to account for the existence of the worlds needed to
universes" into the concretist framework is "There is a world in which provide truth conditions for this intuitive possibility and, hence, needs
there are multiple worlds". Since all of the objects within a single world supplementation. Divers and Melia (2002) propose a strong principle
are spatio-temporally related, the translation entails that there is an object OAN that entails the existence of infinitely many alien properties but go
— a world — all of whose parts are spatio-temporally related such that on to show that, even given the principle, Lewis's translation scheme is
some of those parts are not spatio-temporally related. still incomplete: Given OAN, there will still be intuitive possibilities that
do not translate into truths of concretism (ibid., 31–35).
Lewis (1986, 66ff) rejects the possibility of island universes and attempts
to explain away the initial intuition. Bricker (2001) argues that Lewis's However, the challenge of alien properties itself is arguably the product of
response does not adequately explain away the intuition and, indeed, the same misconception noted in §2.1.5. Lewis agrees that recombinations
argues that they follow from the Humean denial of necessary connections of this-world duplicates do not suffice to “generate” alien possibilities. But
that underlies Lewis's recombination principle R2. He then argues (§3.2) this it is a confusion to think that this is the purpose of the principle of
that concretism can accommodate island universes simply by amending Recombination. Recombination is a principle designed, vividly if perhaps
the standard analysis of possibility such that a proposition is possible if only partially, to express the plenitude of worlds; it is not meant as a sort
and only if it is true at some nonempty class of worlds. This allows island of axiom from which the existence of arbitrary worlds can in some sense
universes to be possible (as they will just comprise multiple concrete be derived. Rather, given any intuitive possibility, it simply follows from
worlds) without there being a world at which any island universes exist. the stipulated completeness of Lewis's account that the concrete worlds
However, even still, Bricker (ibid., §3.3) points out, this move leaves the required for its truth condition exist; Recombination, in turn, expresses the

74 Stanford Encyclopedia of Philosophy Summer 2015 Edition 75

 Possible Worlds Christopher Menzel

richness of the worlds whose existence follows from those concrete worlds propositions, properties, and relations do not exist. Notably, there are no
in virtue of the plenitude of logical space. necessary propositions. For, on Lewis's account, a proposition p is true in
w just in case w ∈ p and p is necessary just in case it is true in all worlds,
The Threat of Paradox that is, just in case p is exactly the set of all worlds. Likewise, if there is
no cardinal limit on the number of counterparts any object can have (and it
Lewis's liberal use of sets in his theory raises the specter of paradox, one
is difficult to see why there should be), then there is no such thing as the
concerning the principle of Recombination and another that stems from his
counterpart relation, as the number of pairs ⟨x,y⟩ such that y is a
adoption of the possible worlds conception of propositions.
counterpart of x is simply too large to be a set. Lewis, therefore, finds
himself forced to place restrictions on R that are arguably ad hoc —
Paradox and Recombination
notably, that there is a cardinal upper bound on the number of objects that
As it stands, the principle R of recombination leads directly to paradox. can fit into any possible physical spacetime (1986, 89–90, 101–4).
Forrest and Armstrong 1984 suggest a particularly simple one: The worlds Moreover, these restrictions threaten the completeness of Lewis's
themselves are objects. Moreover, according to Lewis, the worlds form a program, as certain intuitive modal truths are false under translation into
set W and, hence, there is some number of them. Hence, by R, there is a Lewis's metaphysics, notably, the proposition that, for any number n, it is
distinct world w* that itself contains duplicates of all the worlds. Now, let possible that there are n things. (For further discussion, see Forrest and
E be the set of electrons in w* and suppose the cardinality of E is κ, which Amstrong 1984, Lewis 1986, 89–92 and 101–104, Nolan 1986, Pruss
we can safely assume to be infinite. By R again, for every subset Eʹ′ of E 2001, and Sider 2009.)
there is a distinct world wE containing duplicates of exactly those
Paradox and Propositions
electrons and no others. Hence, there are exactly 2κ such worlds. By
hypothesis, there are distinct duplicates of each of those worlds in w* and,
A second paradox deriving from Kaplan (1995) threatens any account —
hence, w* must contain more electrons than, by assumption, it contains.
Lewis's in particular — that follows possible world semantics in defining
A way out of this paradox is to deny that worlds and, hence, objects propositions to be arbitrary sets of worlds (see Davies 1981, 262; Lewis
generally, form sets. Indeed, that they do not seems to follow from R. For 1986, 104–108; Bueno et al. 2014). Let t be a given time and consider the
R entails that, for any cardinal number κ, there is a world containing at following thesis:
least κ objects. It follows that, for any cardinal κ, there are more than κ
E For every proposition p, it is possible that p is the only
objects. Hence, by some basic set theory, there is no set containing all the
proposition entertained (by anyone) at t.
objects. Assuming furthermore that there is a definite cardinal number of
objects in any world, it follows that there is no set of all worlds. But this
It follows from E that, for every proposition p, there is at least one
appears to lead to theoretical disaster as well. For, given Lewis's
possible world unique to p, a world in which that proposition alone is
definitions of intensional entities in §2.1.3, it follows that certain intuitive
entertained at t. Hence, there have to be at least as many worlds as

76 Stanford Encyclopedia of Philosophy Summer 2015 Edition 77

 Possible Worlds Christopher Menzel

propositions. But if every set of worlds is a proposition, then, by Cantor's We assume there are appropriate abstractionist axioms for these
Theorem, there are more propositions than worlds. Contradiction. primitives, e.g., that only SOAs can obtain — ◻∀x(Ob(x) → SOA(x)) —
that anything that could be an SOA is an SOA, ◻∀x (◇SOA(x) →
Lewis's response is, not implausibly, to deny the thesis E that every SOA(x)) — and so on. To formalize the abstractionist notion of a possible
proposition — every arbitrary subset of an infinite plenitude of worlds — world and to pick out the actual world, four auxiliary definitions are
is possibly the content of some agent's thought (ibid., 105). A consequence needed:
of this move is that it raises doubts about the identification of propositions
with arbitrary sets of worlds. As noted already in §2.2.2, however, Lewis • Poss(s) =def ◇Ob(s)
adopts a very flexible, pragmatic attitude toward the nature of
propositions, and, hence, would not find this consequence particularly • Inc(s,t) =def ◻(Ob(s) → Ob(t))
worrisome. • Pre(s,t) =def ◻(Ob(s) → ~Ob(t))

The Intensionality of Abstractionist Possible World • Max(s) =def ∀t(Inc(s,t) ∨ Pre(s,t))

Semantics Given these we have:
The intensionality of possible world semantics under an abstractionist
• World(s) =def Poss(s) ∧ Max(s)
ontology is vividly illustrated by extending the formal analysis provided in
the supplementary document The Extensionality of Possible World • @ =def the s(World(s) ∧ Ob(s))
Semantics. Recall that the formalized first-order metalanguage in that
document included the set membership predicate ∈ as well as some As before, truth can be defined as truth in the actual world:
additional metalinguistic primitives. However, we will not need the world
True(φ) =def T(φ,@).
predicate ‘World’ or the constant ‘@’ for the actual world, as these are
definable for the abstractionist. Abstractionist semantics requires the Now, recall that the formalized definition of truth in basic possible world
following: semantics yielded the following clause for (6):

SOA(x): x is a state of affairs True(‘◻∀x(Px → Mx)’) ↔ ∀w∀x((World(w) ∧ x ∈ dom(w)) → (x ∉

Ob(s): (state of affairs) s obtains ext(‘P’,w) ∨ x ∈ ext(‘M’,w))).
T(φ,w): (formula) φ is true at (world) w
dom(w): the domain of world w Spelling out the 'World' predicate according to our definition, then, we
ext(π,w): the extension of (n-place predicate) π at world w have:
den(τ): the denotation of (constant or variable) τ
True(‘◻∀x(Px → Mx)’) ↔ ∀w∀x((Poss(w) ∧ Max(w) ∧ x ∈

78 Stanford Encyclopedia of Philosophy Summer 2015 Edition 79

 Possible Worlds Christopher Menzel

dom(w)) → (x ∉ ext(‘P’,w) ∨ x ∈ ext(‘M’,w))). The proof of the left-to-right direction, however — which is, in fact,
offered as a general proof that possible worlds exist — depends upon a
And unpacking the definitions of 'Poss and 'Max' in turn yields: metaphysical analog of the compactness theorem for first-order logic that
is demonstrably false in the context of Plantinga's rich ontology of states
True(‘◻∀x(Px → Mx)’) ↔ ∀w∀x((◇Ob(w) ∧ (∀s◻(Ob(s) → Ob(t))
of affairs. (See Menzel 1989 for details.)
∨ ∀s◻(Ob(s) → ~Ob(t))) ∧ x ∈ dom(w)) → (x ∉ ext(‘P’,w) ∨ x ∈
ext(‘M’,w))), And again: Plantinga (1974, 45) offers up a proof that there is a unique
actual world. However, as McNamara (1993) demonstrates, the argument
and the irreducibly intensional character of the truth condition becomes
rests essentially on the principle that necessarily equivalent SOAs are
evident.
identical, a principle clearly at odds with the fine-grainedness of SOAs
Return to the article “Possible Worlds” that Plantinga assumes elsewhere (and often) in various applications of his
theory.
Problems with Abstractionism
The Threat of Paradox
Lack of Formal Rigor
A more pernicious threat arising from the lack of adequate formal rigor is
With the notable exception of Zalta (1983, 1993), a pervasive problem for the threat of paradox. Adams (1974), for example, defines a possible
nearly all abstractionist accounts is that they lack sufficient formal rigor. world (or "world story") to be maximal consistent set of propositions,
More specifically, abstractionists rely heavily upon an ontology of abstract where a set S of propositions is maximal just in case, for every proposition
entities but typically do not provide adequate formal underpinnings for p, S contains either p or its negation ¬p, and consistent just in case it is
their theories, notably, a formal language, model theory, logic, and a set of possible that all of S's members be jointly true. Suppose now the following
axioms. (Sider's (2002) linguistic ersatzism is not lacking in rigor but he three propositions (all of which appear to be embraced by Adams):
does not axiomatize his theory and its heavily model theoretic approach is
(A) For every set S of propositions there is a distinct proposition pS
subject to well known cardinality problems to which Zalta's theory
(e.g., the proposition that S is nonempty).
appears to be immune.)
(B) Every proposition p has a negation ¬p.
To illustrate: Plantinga (1985) offers up a proof of the extensional
adequacy of his notion of a possible world, that is, a proof of: (C) If S and Sʹ′ are distinct sets of propositions, then pS, pSʹ′, and their
negations are all pairwise distinct.
EA For every SOA s, s is possible iff there is a possible world that
includes s. Then a paradox follows. For let w be a world story. By (A), (B), and the
maximality of w, for every subset S of w either pS ∈ w or ¬pS ∈ w and, by

80 Stanford Encyclopedia of Philosophy Summer 2015 Edition 81

 Possible Worlds Christopher Menzel

(C) and the consistency of w, this proposition is correlated uniquely with that includes both [P,o] and [Q,o]. On a priori grounds alone, this seems
S. It follows that function f(S) = pS maps the set ℘(w) of all subsets of w implausible; why should it not turn out that some of the simple properties
one-to-one into w. Hence ℘(w) is no larger than w, in violation of of the simple are simply, primitively incompatible? Likewise, for any
Cantor's Theorem. (For more on this paradox, see Bringsjord 1985, simple binary relation R, unrestricted recombination requires that, for any
Menzel 1986, Grim 1986, Beall 2000, and Menzel 2012.) Because distinct simple objects a and b, there is a world that includes both [R,a,b]
Plantinga (1974, 44–5) accepts (A)–(C) and postulates, for each of his and [R,b,a]; there can therefore be no simple, necessarily asymmetric
worlds w, the existence of a corresponding world story (which he calls the relations. For the same reason there must be a world that includes [R,a,a]
book on w), this paradox arises for him as well. However, making and, hence, there can be no simple, necessarily irreflexive relations.
reasonable assumptions about the nature of Plantinga's underlying Armstrong (1997, 146–7) acknowledges that unrestricted recombination,
metaphysics, Chihara (1998, 126–7) reconstructs a similar paradox or independence, is ultimately an empirical matter but argues that, so long
directly for Plantinga's notion of a possible world. as it is not empirically refuted, the doctrine is warranted by the simplicity
it brings to the theory of modality.
Divers (2002, Ch. 15) presents a detailed and illuminating overview of the
problem of paradox for abstractionism. Further difficulties for Adams' The Prospect of Infinite Decomposition
conception of a world story and for Plantinga’s notion of an haecceity, as
well as for other abstractionist approaches can be found in Problems with The combinatorialist definition of a possible world appears to hinge
Actualist Accounts, a supplementary document to the companion article critically on the assumption that, ultimately, there are simple facts on
Actualism. A discussion of how Zalta's world theory avoids cardinality which all other (first-order) facts supervene. A challenging possibility for
problems and paradox can be found in Bueno et al. (2014). the combinatorialist is that this assumption is false. In particular, it seems
conceptually possible that there are no simples at all, that every particular
Return to Possible Worlds comprises yet simpler particular and that every universal is a conjunction
of yet simpler universals. Should this be the case, then there are simply no
Further Problems for Combinatorialism possible worlds the sense of AW3 or AW3ʹ′ʹ′.

Problems with Unrestricted Recombination Armstrong (1986a, ch. 5) takes up this challenge by introducing the notion
of relative atoms. The idea is this: Let s be any fact, simple or not, and call
It is critical to combinatorialism that recombination among simples is its constituent objects and universals its atoms (Armstrong 1986, 69):
unrestricted — any simple object can exemplify any number of simple
universals. Any modal qualifications of this principle would undermine The totality of recombinations of these 'atoms' yields a set of
combinatorialism's modal reductionism (cf. Bradley 1989, 35–38). Thus, possible worlds….If the atoms are genuine atoms [i.e., simples],
for the combinatorialist, all simple universals must be compatible; for any then no more remains to be done. But if the atoms are not
simple object o and any simple univesals P and Q, there must be a world genuinely atomic, then this set is a mere subset of the worlds

82 Stanford Encyclopedia of Philosophy Summer 2015 Edition 83

 Possible Worlds Christopher Menzel

which can be formed. With one or more of the 'atoms' broken up, the recombination of the relative atoms of a world. Say that universals
we can go on to an enlarged set of worlds. If the breaking-up goes overlap if they share a common constituent and that particulars overlap if
on forever…, then at each point in the break-up new worlds they share a common part. Then the proposed definition of a possible
emerge. world is the following

Clearly, however, a condition has to be placed on the initial state of affairs AW3R w is a (combinatorial) possible world =def w is the conjunction of
s. For let W be our water molecule from Figure 1 and let I be an (actual) (i) a recombination of some simple fact f such that none of f's
hydronium ion, hence one with entirely distinct atomic constituents, as constituent objects, and none of its constituent universals,
indicated (along with the correpsonding structural universal) in Figure 3. overlap and (ii) the totality state of affairs Tw for that
recombination.

As simples by definition do not overlap, AW3R obviously generalizes

AW3ʹ′ʹ′. At the same time, because Water and Hydronium overlap —
Oxygen and Hydrogen are constituents of both — [[Water,t] &
[Hydronium,s] & T] does not represent a genuine possible world.

Sider (2002) points out that serious problems remain for Armstrong's idea.
Figure 3: Hydronium Ion I and the Structural Universal Hydronium (H) By AW3R, [Water,s] (together with its totality state of affairs) itself
constitutes a (very sparse) possible world w. But, as Sider notes, as none
Now, W and I, according to Armstrong, are both facts, presumably of the of the constituents of Water exist in w, “its use in representation of a
form [Water,s] and [Hydronium,t], respectively, and s and t are, possibility has little to do with its actual nature”; similarly for the
respectively, the sums of the constituent atoms of W and I. Since both W (actually) complex object s. The world w thus does not represent “much of
and I exist, there is the conjunctive state of affairs [W & I]. If, however, anything, beyond that there exist[s something]…which instantiates a
we consider any recombination of the “atomic” universals and particulars single universal” (ibid., 683). Sider then offers up a plausible
constituting W and I — viz., Water, Hydronium, s and t — (together reconstruction of Armstrong's basic idea of relative atoms that avoids this
with its totality state of affairs T) to count as a (contracted) possible world, objection, but this effort leads to an array of further objections to the view.
then we have in particular the world [[Water,t] & [Hydronium,s] & T]
(hence a world where t exemplifies Water and s exemplifies Return to “Possible Worlds”.
Hydronium), which is clearly impossible for obvious structural reasons
(notably, in reality, W consists of only two hydrogen atoms and I of Notes to Possible Worlds
three).
1. For important applications of possible worlds, see the SEP entries
To avoid this problem, Armstrong places a “non-ovalapping” condition on

84 Stanford Encyclopedia of Philosophy Summer 2015 Edition 85

 Possible Worlds Christopher Menzel

Supervenience, Rigid Designators, Two-dimensional Semantics, 4. Tarski's own early work on the semantics of predicate logic was much
Conditionals, Logic of Belief Revision, Common Knowledge, and Belief. more informal than the account here, which for the most part reflects the
A particularly illustrative possible worlds analysis of the concept of development found in modern expositions that came to full flower in,
prudence is found in Bricker 1980. notably, Kemeny 1956a and 1956b. The account here differs from both
Tarski's approach and its typical modern incarnations in that
2. A logic is often (quite legitimately) defined more generally to be a interpretations assign values to variables directly, much as if they were
language together with either a semantic theory or some sort of deductive names. On the more typical approach, interpretations do not assign values
apparatus, as in the definition found in the Encyclopedia entry Classical to variables; rather, variables can be assigned different values within one
Logic. However, as the distinction between extensional and intensional and the same interpretation. (See §4 of the entry Classical Logic for
logics drawn here is a purely semantic one, it is convenient for our details.) The chief advantage of the approach adopted here (which does
purposes to use the more limited, semantically-oriented definition. not alter the critical notions of validity and logical consequence in any
significant way) is that it enables one to define the notion of truth under an
3. A word about the corner quotes, or “quasi-quotes”, around the 2-
interpretation directly (and rather more simply). For Tarski's own
element expressions ‘◇φ’ and ‘□φ’ here. Poss and Nec are metalinguistic
approach, see §2 of the entry on Tarski and §2.1 of the entry on Tarski's
statements, statements of a language — in this case, English enhanced
Truth Definitions. See also the seminal work of Etchemendy (1990) for a
with a bit of logical apparatus — made about the statements in (typically)
more philosophical examination of Tarski's work.
some other language — in this case, the statements of some (unspecified)
modal language ℒ. We need some type of quotation marks around ‘◇φ’ 5. Bayart 1958 and 1959 are surprisingly unfamiliar to contemporary
and ‘□φ’ in Poss and Nec because ‘◇’ is not itself a part of our English philosophers and logicians. In these papers, Bayart — working largely
metalanguage; rather, we are referring to certain statements of ℒ that independently, it appears — developed a possible world semantics for
contain it, viz., those that begin with it. Ordinary single quotation marks first-order modal languages and proved the soundness and completeness
will not do, as they create names for the very expressions they enclose and of first-order S5. The articles are translated and given a very informative
our intent here is to refer, not to the strings ‘□φ’ and ‘◇φ’ themselves but, introduction by Cresswell (forthcoming), whom the author thanks for
rather, in each case, to sentences of ℒ that have the forms those introducing him to Bayart's work. See the entry Modern Origins of Modal
expressions indicate. This is the purpose of the quasi-quotes. Thus, Nec Logic as well as Copeland 2002 and Goldblatt 2003 for a comprehensive
can be read as follows: For any sentence φ of ℒ, the sentence that results overview of the historical development of possible world semantics in the
from prefixing the symbol ‘□’ to φ is true if and only if φ is true in every 20th century.
possible world. Similarly for Poss. (Readers familiar with the
programming language Lisp will recognize the similarity between quasi- 6. We follow Kripke (1963) in taking names to be “rigid”, i.e., to have a
quotes and the Lisp backquote operator.) For more on this topic, see fixed denotation that is independent of worlds. A more general (but, for
Washington 1998 as well as the entry Quotation. purposes here, needlessly more complicated) approach, following Carnap
1947, is to assign intensions, not only to predicates, but to names as well.

86 Stanford Encyclopedia of Philosophy Summer 2015 Edition 87

 Possible Worlds Christopher Menzel

See, e.g., ch. 13 of Garson 2006, esp. §13.2, as well as §3 of the entry of this approach where all worlds accessible to each other.
Intensional Logic, especially §§3.3–3.5. See §3.6 of the entry for a
discussion of problems arising from the assumption that names are rigid. 10. For examples of the use of intensional entities in more sophisticated
developments of possible world semantics, particularly with regard to the
7. We again follow Kripke (1963) here and adopt a variable domain semantics of natural language, see, e.g., Lewis 1970, Montague 1974,
semantics for quantified modal logic to capture the intuition that, “under Kaplan 1979, and Cresswell 1973, 1985a/b, 1988, 1990, 1994, 1996 as
different circumstances, fewer, more, or other things might have existed”. well as the Encyclopedia entry Intensional Logic.
A common alternative is simply to let the quantifiers range over the single
fixed domain D of all “possible objects” and, instead, introduce a primitive 11. The de re/de dicto distinction traces back to Aristotle (see Nortmann
existence predicate ‘E!’ whose extension varies from world to world. 2002) and was a matter of robust discussion in the medieval period (see
Aside from the fact that Kripke's treatment of the quantifiers arguably the Encyclopedia entry Medieval Theories of Modality).
yields an intuitively more correct semantics of quantification in modal
12. More exactly, φ exhibits modality de re if there is a subformula of φ
contexts, formally, unlike a fixed domain semantics, it prevents certain
of the form ⌈◻ψ⌉ or ⌈◇ψ⌉ containing either an occurrence of a name or
controversial modal principles — most notably, the so-called Barcan
a free occurrence of a variable. Thus, for example, ‘◇∃xFx → ∃x◇Fx’ —
Formula ⌈◇∃νφ → ∃ν◇φ⌉ (Barcan 1946) — from being logical truths.
an instance of the controversial Barcan Formula — is de re because ‘x’
See the articles on actualism (especially §2 and §3) and modal logic
occurs free in the subformula ‘◇Fx’. This definition corresponds to Fine's
(especially §13) for detailed discussion.
notion of modality de re in the strict sense (1978, 143). Sentences are de
8. Some possible world semantic theories (e.g., Menzel 1991 and (in its re in the loose sense if they simply contain a free variable in a modal
own way) Jager 1982) impose the additional condition that Iπ(w) contain context. As he notes, the strict sense is appropriate if names are
only (n-tuples of) things that exist in w. This condition reflects the thesis understood semantically (as they are in this exposition) to be rigid
of serious actualism that exemplification entails existence, i.e., that it is designators. (Fine defines strictness and looseness for modality de dicto,
not possible to exemplify a property at a possible world without existing in but it amounts to the same thing.)
that world. For more on serious actualism, see the entry Actualism,
13. This is of course not to deny that there are robust alternatives to these
especially, §2.2.3 and note 6.
three. Two notable recent examples are found in McDaniel 2006 and
9. In a more comprehensive exposition of possible world semantics, the Yagisawa 2010. Both develop detailed theories of worlds that bear
definition of a possible world interpretation would include a binary important similarities to concretism but depart from it sharply on several
accessibility relation R on the set W of worlds and the modal clause counts, in particular in their answers to QW.
would say that ⌈◻ψ⌉ is true at w if and only if ψ is true at all worlds u
14. Note that ‘part’ here does not mean ‘proper part’, nor does “any two
that are accessible from w i.e., all worlds u such that Rwu. (See §§7–8 of
parts” mean “any two distinct parts”. Even if there are (as Lewis seemed
the entry Modal Logic.) Our semantics here is essentially the special case
to hold) simple objects like spatiotemporal points that have no proper

88 Stanford Encyclopedia of Philosophy Summer 2015 Edition 89

 Possible Worlds Christopher Menzel

parts, they are still parts of themselves. Moreover, such objects bear 136) labels such views “ersatz” modal realism; see notes 23 and 34 for
spatio-temporal relations to themselves — coexisting at some time, for more on ersatzism.) Second, the rubric is arguably misleading. Some uses
example — and hence are connected by the given definition. of ‘realism’ are intended to indicate that the fundamental locutions in a
relevant body of discourse are semantically primitive and, hence, cannot
15. It is worth commenting on the fact that, the rubric “concretism” be reduced to more fundamental notions. Some forms of scientific realism,
notwithstanding, the notion of concreteness is not actually part of the for example, hold that discourse about the theoretical entities of physics is
definition of a world. Lewis (1986, §1.7) himself acknowledges that his not simply shorthand for discourse about, say, meter readings and patterns
worlds are concrete according to several ways of understanding the notion, on cloud chamber photographs. As discussed in §2.1.3, however, Lewis is
but is skeptical about whether the abstract/concrete distinction could be a reductionist about modal discourse — the modal operators, in particular,
clearly made. Concrete objects, for example, are often defined to be are not semantically primitive but, rather, are to be unpacked semantically
objects whose parts all bear spatiotemporal relations to other parts. But in non-modal terms. See Plantinga 1987 for more on this point.
there are conceptions of certain abstract objects (notably, sets of
spatiotemporally located objects (Maddy 1980) and physical universals 18. In his well known 1968 paper, Lewis axiomatizes the three predicates
(Armstrong 1986b)) that arguably satisfy this description as well. below (and a fourth) that express the primitives of his translation scheme
Nonetheless, we are sticking with the “concretism” label for Lewis's view explicitly, and calls the result counterpart theory. Hazen (1979) raises
— albeit “advisedly” (Bricker 2008) — because it still has a good bit of several challenges to the theory against which Hunter and Seager (1981)
traction in the philosophical literature (due in large measure to the mount a defense that is tightened up formally by Forbes (1982). Notable
influential van Inwagen 1986). more recent critiques of counterpart theory are found in Merricks 2003,
Cresswell 2004, Fara and Williamson 2005, and Fara 2009.
16. Let Rxy mean that x and y are spatiotemporally related and suppose
w1 and w2 overlap, i.e., that they share a common part c. Let a be a part of 19. It should be noted that, in Lewis's mature theory of 1986, resemblance
w1 and b a part of w2. Then, as w1 and w2 are both connected, we have relations are very fluid and contextually-dependent, and such relations can
Rca and Rcb. Hence, on the reasonable assumption that R is Euclidean, actually vary even within the semantics of a single sentence. See, for
Rab. Hence, by maximality, b is a part of w1. A parallel argument shows example, Lewis's response (ibid., 254–5) to the “there but for the grace of
that a must be a part of w2. Hence, our two worlds have the same parts God” examples of Feldman (1971) and others.
and, hence, by basic principles of mereology, they are identical.
20. These are of course first-order intensions. Lewis does in fact allow for
17. For this reason concretism is often referred to as modal realism. That higher-order intensions as well, e.g., properties of properties of individuals
rubric is avoided here for two reasons. First, it is arguably tendentious, like being a property possessed by every great general. Egan (2004)
insofar as it suggests that abstractionist (§2.2 below) and (some) argues, however, that Lewis cannot generalize the definitions here and
combinatorial (§2.3 below) alternatives to concretism are somehow less define, e.g., a property to be any set but, instead, must take intensions
than robust forms of realism about possible worlds. (Indeed, Lewis (1986, generally to be functions from worlds to extensions, much like the original

90 Stanford Encyclopedia of Philosophy Summer 2015 Edition 91

 Possible Worlds Christopher Menzel

definition above. 27. See van Inwagen 1986 for an especially clear and illuminating
comparison of the two approaches and Plantinga 1987 for his own rather
21. We use only the conditional rather than the biconditional here because more tendentious comparison. See Lewis 1986 (§3.4) for an extended
there is no reason to think that every sentence φ* that is the translation of critique of abstractionism. Lewis (no less tendentiously) refers to
a modal sentence and happens to be true in Lewis's metaphysics should be abstractionism as magical ersatzism — “ersatzism” because, in Lewis's
the translation of an intuitive modal truth. Certain specifics of Lewis's view, concrete worlds constitute the most natural ontology for modality
metaphysics might well decide the truth values of sentences that are and abstractionists seek to replace them with (inferior) abstract substitutes;
undecided by ordinary modal intuitions. “magical” because it is mysterious exactly how these abstract entities, as
understood by the abstractionist, manage to represent possible worlds at
22. Lewis (1986, 86) himself makes this point; van Inwagen (1986, 197),
all.
to whom Lewis credits the observation, makes the point as well.
28. Indeed, Plantinga acknowledges that states of affairs might just be
23. DeRosset (2009a, 1003) argues that “pushing down” to the
propositions, but finds it more natural to distinguish them. See Plantinga
microphysical level to account for possibilities that do not seem to be the
1974, 45.
product of macrophysical recombination still leaves possibilities at the
micro-level that seem unaccounted for by R. For more on the question of 29. Note that the qualification that a world be a possible SOA is required,
the success of Lewis's reductionism, see e.g., Shalkowski 1994 and as every impossible SOA satisfies the definition of totality. For suppose s
Bricker 2008. is impossible and let t be any SOA. Then it is not possible that s obtain
and, hence, trivially, it is not possible both that s obtain and t fail to
24. See Efird and Stoneham 2008 and Darby and Watson 2010 for a
obtain. Hence, s includes t and, hence, trivially, s either includes or
particularly interesting and sophisticated exchange.
precludes t, i.e., s is total.
25. I am indebted to Phillip Bricker for suggesting this line of response.
30. Merrill (1978) argues quite forcefully that the strict circularity of such
26. As noted already, Lewis (1973, 86) originally identified his concrete analyses does not preclude their being semantically enlightening.
worlds with “ways things could have been” and realized later that the
31. Thus Plantinga (1987, 207):
identification was vacuous. Even earlier, however, Stalnaker (1976, 68)
had called him out about the use of ‘the way things are’ to indicate the Clearly [Lewis] is partly right: there is much about the nature of
concrete actual world: “If possible worlds are ways things might have propositions we don't pretheoretically know....But we do know
been, then the actual world ought to be the way things are rather than I something about the nature of propositions, prior to theory.
and all my surroundings. The way things are is a property or a state of the Conceivably they could turn out to be idealized sentences or divine
world, not the world itself.” thoughts; but they couldn't turn out to be just anything—donkeys,
or fleas, or tables, for example. We know that no propositions are

92 Stanford Encyclopedia of Philosophy Summer 2015 Edition 93

 Possible Worlds Christopher Menzel

donkeys, and we know that none are fleas. the actualist denies is that there is a definite (non-actual) individual a such
that a could have been Bergoglio's child; that there is a possible world in
The concretist would likely agree. But Plantinga (ibid., 208) continues: which it is true that a is Bergoglio's child.

Even as we can see that a proposition can't be a donkey or a flea, 33. This is an important qualification, as there is a clear sense in which
so we can see that a proposition can't be the unit set of a flea, or concretists are not possibilists at all. One way to formulate actualism is
any other set of fleas or donkeys,....or a set of concrete objects of say that there is no more general kind of being than that enjoyed by
any sort. The problem, fundamentally, is that sets, like donkeys, actually existing things. On this formulation, concretists are actualists.
obviously lack the relevant intentional properties — the intentional For, according to the concretist, all individuals in non-actual possible
properties propositions have. worlds exist in precisely the same way that individuals in the actual world
do. They are simply not here (in the broadest possible sense). van Inwagen
Here the concretist would balk. The reason it seems obvious that donkeys
(2008, 40–41) makes essentially this point.
or, for that matter, individual concrete worlds, are not propositions, or any
other type of intensional entity, is that it is difficult imagine a coherent, let 34. For further details of the new actualist understanding of modal truth
alone natural, theory in which concrete individuals play the sorts of conditions, see §4.1.1 of the entry Actualism.
philosophical and logical roles that propositions are often invoked to play
— exhibiting closure under boolean operations, for example. By contrast, 35. In fact, new actualists generally tend to opt for fixed domain semantics
sets of worlds (and perhaps more elaborately structured sets built up from for modal logic (as discussed briefly in note 6 above) and a modal
individuals generally) are able to play those roles. As to the charge that language that contains a dedicated predicate ‘C!’ to express the property
propositions so construed lack the relevant intentional properties, the of concreteness. Thus, the intuition that there could have been something
concretist would likely argue that the charge is question-begging, insofar that does not actually exist is understood as the claim that there are non-
as (a) the properties in question are not available within the concretist concrete things that could have been concrete: ∃x(¬C!x ∧ ◇C!x).
framework and (b) are not obviously doing any genuine theoretical work.
(Cf. the responses of Hazen (1979, 322ff) to similar arguments of 36. More exactly, a haecceity is a type of individual essence, i.e., a
Plantinga's against Lewis's analysis of de re modality.) property P such that (i) P is possibly exemplified, (ii) necessarily, if an
individual s exemplifies P, then s exemplifies P essentially and, moreover,
32. Note, importantly, that actualism is not the thesis that there couldn't (iii) necessarily, nothing other than s could exemplify P. (s exemplifies a
have been anything other than the things that are, in fact, actual. Most property P essentially if it is not possible that s exist and fail to exemplify
every actualist believes there could have been. For instance, assuming his P.) That haecceities are considered individual essences is enough for them
lifelong chastity, the Pope (as of April 2013), Jorge Bergoglio, has no to play the role Plantinga intends for them but, intuitively, haecceities can
children. However, most actualists would agree that he could have had be thought of as purely non-qualitative individual essences.
children, i.e., that there are possible worlds in which that is the case. What
37. A set Γ of sentences in the language ℒ of a logic is (semantically)

94 Stanford Encyclopedia of Philosophy Summer 2015 Edition 95

 Possible Worlds Christopher Menzel

consistent if it has a model, i.e., if there is an interpretation of ℒ under objections from Lewis (1986a, §2.3.1) in regard to the notion of a
which every member of Γ is true. Γ is maximally consistent if it is structural universal, understood “pictorially” as isomorphs of their
consistent and no proper superset of Γ is consistent. Consistency is also instances in the manner laid out in §2.3.1. Lewis's strongest objection
often defined proof theoretically — Γ is (proof theoretically) consistent if suggests that the notion is ultimately inconsistent. Consider the structural
there is no proof of a contradiction from Γ — but the definition above universal Water in Figure 1. As can be seen there, the Water involves the
comports with the definition of a logic used in §1.1 simply as a language universal hydrogen two times over; likewise for the bonded relation B.
together with a semantics in which truth, validity, etc are definable. Lewis then asks:

38. Lewis 1986, 142–165. See note 23 above for the explanation of But what can it mean for something to have a part [two] times
‘ersatzism’. Carnap's (1947) notion of a state description can be seen as an over? What are there [two] of? There are not [two] of the universal
early version of linguistic ersatzism (see esp. §41). See Roy 1995 and hydrogen, or the universal bonded; there is only one. The pictorial
Sider 2002 for particularly sophisticated developments of this approach. conception...has many virtues, but consistency is not one of them.
Bricker 1987 provides an illuminating account of linguistic ersatzism and
a cogent assessment of its prospects. Armstrong (1997, 119–123) argues in response that the objection
presupposes that metaphysical complexity can only be mereological and,
39. The use of “the complex fact” here assumes that facts with the same hence, that structural universals are simply sums of their constituent
conjuncts are identical; nothing terribly important hangs on this principle, universals. To the contrary, as noted, structural universals enjoy a richer
but it is a natural one for the combinatorialist and, hence, it will be “non-mereological mode of composition”. Armstrong illustrates the idea
introduced explicitly below (proposition (29). chiefly by pointing to cases of distinct states of affairs with identical
constituents. For example, there is obviously more to the state of affairs
40. See Armstrong (1978b, 69ff). Note that, on this definition, a's loving b than the sum of a, b, and the universal loving, lest it be
“[s]tructural properties may or may not involve certain relations among identical to the state of affairs b's loving a. The states of affairs must
the part of the particulars having the properties.” Thus, on this definition, therefore in general exhibit a richer sort of composition than the mere
properties of, e.g., the form being exactly two Fs appear to count as mereological composition of its constituents to explain this difference. As
structural, even for simple universals F. Armstrong gives the example structural universals are simply abstractions from states of affairs, the
being exactly two electrons, although of course he does not claim that argument presumably concludes, it is reasonable to appeal to this richer
being an electron is a simple). However, in a later incarnation, Armstrong sort of composition to explain their structural complexity. The response is
(1997, 32) suggests that, for a property U to count as structural, at least promising but, unfortunately, Armstrong does not develop the concept of
one of the universals constituting U must be a relation that at least two of non-mereological composition any further.
the simpler parts of any instance of U bear to one another.
42. This sense of ‘gunk’ originates in Lewis 1991, 19–21, 133–136. For
41. Armstrong developed this notion of constituency in response to serious reasons discussed in §2.3.3, structural facts cannot be identified with the

96 Stanford Encyclopedia of Philosophy Summer 2015 Edition 97

 Possible Worlds Christopher Menzel

sum or conjunction of the simple facts they are grounded in. Problems that supervenient entity is still an entity, but it is altogether superfluous as a
arise when the assumption that the constituency relation is well-founded truth-maker.”
(and hence that structural facts cannot be infinitely decomposed) is lifted
are discussed in the supplemental document Further Problems with 46. This claim is a bit ambiguous. Non-structural states of affairs do not
Combinatorialism. supervene on their atomic conjuncts individually, but jointly, a claim that,
it appears, must be understood via plural quantification: for any non-
43. That is, assuming O is the thin particular o's only intrinsic property, structural state of affairs s, there are some states of affairs such that (a) s
the atom/fact o is a “thick“ particular, i.e., the thin particular o “considered supervenes on them and (b) something is one of them if and only if it is
along with all of its intrinsic...properties” (ibid., 434). atomic and s includes it.

44. The example illustrates the sort of thing Armstrong has in mind in 47. The astute reader will have noticed that we have avoided displaying
suggesting that complex universals enjoy a “non-mereological mode of the exact form of the state of affairs W's being water. The grammatical
composition”. If the complexity of the structural universal Water were form suggests that it has the form [Water,W], but that would not be
simple mereological complexity, there would only be just the one correct, in light of the fact noted in §2.3.1 that the constituent object in
universal, hydrogen (H) and the one bonding relation (B) and, hence, the states of affairs of this form are thin particulars, and W is, by assumption,
structure of the universal Water could not be isomorphic to that of our thick. Armstrong himself is not terribly clear on this point but,
molecule M, in which H and B are exemplified twice. The “non- extrapolating from Armstrong 1993 (433–434) and 1997 (34–35) (which
mereological” composition of the structural universal, by contrast, permits are not obviously consistent), the most plausible form of W's being water
there to be, as it were, two “copies” of both H and B in Water so as to is [Water,w], where w is the mereological sum o+h1+h2 of the thin
reflect accurately the structure of M. particulars underlying the oxygen and hydrogen molecules o, h1, and h2.
In fact, the notion of exemplification in constituent ontologies like
45. For Armstrong, supervenience has significant ontological heft: “[w]hat Armstrong's — or, at least, like our simplified reconstruction here — is a
supervenes is no addition of being” (Armstrong 1997, 12). That is, in rather tricky business. See, e.g., Moreland 2011 and Pickavance 2014.
particular, complex states of affairs — whether mere conjunctions of
simple, atomic states of affairs or structural states of affairs like our water 48. It is natural to understand spoilers in primitive modal terms, e.g., a is a
molecule W — are nothing, ontologically, over and above the states of spoiler for S just in case it is not possible that both a and S exist. But there
affairs on which they supervene (see also TLP 2.032; Armstrong 1989, does not seem to be any reason the combinatorialist cannot simply deny
113; Armstrong 1997, 44ff). Lewis (1992, 216) finds the claim difficult to the primitiveness of the modality here and assert instead that it is just a
understand in terms of a standard notion of ontology and, hence, proposes fact about recombination that a and S do not co-exist in any combinatorial
that “[Armstrong's] question is not: what is there? But rather: what does it possible world, that a world in which a exists is not one where S does and
take to provide truth-makers for all the truths. That way, it makes perfect that this non-modal fact is the truthmaker for the modal proposition above.
sense to say that supervenient entities add nothing to our ontology. A Thus, a world in which a third hydrogen atom bonds to our the oxygen

98 Stanford Encyclopedia of Philosophy Summer 2015 Edition 99

 Possible Worlds Christopher Menzel

atom o in Figure 1 is simply not a world in which W exists; it is not a not seem essential to combinatorialism per se.
world in which o, h1, and h2, even bonded as they are in W, jointly
constitute an instance of Water. 53. Totality states of affairs and so-called law states of affairs, among
others, are “higher-order” for Armstrong, insofar as they involve
49. Notably, §§1.11–1.12: “The world is determined by the facts, and by properties of, or relations among, properties. Law states of affairs in
their being all the facts. For the totality of facts determines what is the particular are causal relations between universals (Armstrong 1986,
case, and also whatever is not the case.” §15.2).

50. A bit more specifically, Armstrong (2004, 74) takes a totality state of 54. As Armstrong (1997, 169) notes, on the combinatorialist picture,
affairs to be a higher-order atomic state of affairs of the form [Tot,T,being animal bodies and other large, complex, highly structured physical objects
a state of affairs], where T is a sum of states of affairs — intuitively, the can only be “loosely” identified with similar objects in other possible
state of affairs that the sum in question is a totality with regard to the worlds (or, for that matter, at other times (ibid., 104–7)), as any difference
property in question. Heil (2007) argues cogently that the perceived need in the internal structure of complex objects o and oʹ′ entails that o ≠ oʹ′. Cf.
for totality states of affairs is a confusion based upon a “shady the discussion of loose and strict identity (ibid., 14–17).
‘linguisticizing’ tendency to conflate features of descriptions and features
of what is described“ (ibid., 233). Keller (2007; 2009, 125–126) argues, 55. Armstrong himself argues from more or less these premises to the
more ominously, that the idea of a totality state of affairs is paradoxical conclusion that O must be identical to one of its proper parts but it is not
and, hence, that such states of affairs simply cannot exist. However, the entirely clear how to reconstruct the argument. Lewis (1992, 212)
argument involves the assumption that if there are totality states of affairs, questions the reductive adequacy of Armstrong's account of analytic
then they, too, form a totality. But, as Armstrong (2007) points out in a necessities and impossibilities, as the claim of analyticity seems to rest on
reply to Keller, it seems quite consistent to accept the existence of totality the modal assumption “that a whole cannot share a universal with its part.”
states of affairs and deny that they jointly constitute a further totality, But, Lewis asks, why not? “If something has a mass of [two] kilograms by
much in the way that ZF set theory denies that the sets jointly constitute a consisting of [two] one-kilogram parts, how does that prevent it from also
further set. having a mass of one kilogram?” See also Eddon 2007 for a detailed and
incisive critique of Armstrong's analysis of quantitative properties.
51. That said, Armstrong goes a bit soft on his commitment to modal
reductionism in light of, especially, apparently irreducible modal 56. Somewhat surprisingly, Armstrong himself does not seem to avail
properties of certain relations. See 1997, Ch. 9, esp. 147. himself of explanations of exactly the sort given in the paragraph
containing the reference to this footnote. He does, however, explicitly
52. Armstrong (1989, 47) adds a further condition on possible worlds: appeal to the notion of emergence. (Indeed, he defines the notion of an
they must not include any (simple) bare particulars, but the requirement emergent property as a non-structural “anomoeomerous” property, i.e., a
stems more from Armstrong's more global metaphysical views and does non-structural property such that some particular that has it has a part that

100 Stanford Encyclopedia of Philosophy Summer 2015 Edition 101

 Possible Worlds Christopher Menzel

does not have it; see Armstrong 1978b, 69–71, 171.) In his 1997 (152–3), part of the structure that defines the other.
he uses the notion of emergence with regard to the prospect of a simple
property Q of a complex individual a+b that arises nomically from a and 58. As with AW3, Armstrong himself (1989, 61–5) adds a proscription
b having certain specific properties other than Q and standing in specific against bare particulars. The continuation of the quote above in Armstrong
simple relations. (It is unclear exactly how such a Q is to be distinguished 1989 (579) reflects the modifications of AW3ʹ′ that allow for contracted
from a structural property whose constituents are those simple properties worlds: “Such possible atomic states of affairs may then be combined in
and relations, much like the properties B and H of Figure 1.) And in 2004b all ways to yield possible molecular states of affairs. If such a possible
(14–15) Armstrong again appeals to the notion with regard to complex molecular situation is thought of as the totality of being, then it is a
relations that are, e.g., asymmetric, and, hence, give rise to such possible world.”
impossibilities as: If R is necessarily asymmetric, there can be no state of
59. This is of course the problem of alien universals and alien particulars
affairs that includes both [R,a,b] and [R,b,a]. (Recall that, due to the
(Lewis 1986), which is discussed in the context of concretism in the
unrestricted nature of recombination, no simple relation can be
supplemental document Further Problems with Concretism. In a detailed
intrinsically asymmetric.) For more on emergence, see the Encyclopedia
discussion, Schneider (2001) argues that Armstrong (1997) plausibly
entry Emergent Properties, especially §3.
meets the challenge of alien particulars but fails with regard to alien
57. More complex real-world examples of incompatible states of affairs universals.
can be found in molecular biology. The misfolded prions responsible for
60. The problem with possible world fictionalism identified in Rosen 1993
the class of TSE diseases (Novakofski et al. 2005) constitute a particularly
in particular led to the revised account in Armstrong 1997.
dramatic example of the impossibility of a given complex object
simultaneously exemplifying structural properties that it can exemplify Copyright © 2015 by the author
serially. Instances p of the normal prion protein, PrP, are routine on the Christopher Menzel
surface of mammalian neurons and consist of a polypeptide sequence s
folded into a certain structure. Thus, both the component amino acids of s
and the structural relations they bear to one another are constitutive of that
instance of the structural property PrP. If that structure is altered in a
certain way, s ceases to be such a protein and transforms instead into a
pathogenic PrPSc protein pʹ′ʹ′ with very different causal properties. (The
“Sc” suffix here derives from the TSE disease scrapie that affects sheep
and goats.) The conjunction [p & pʹ′ʹ′] of the two component structural
states of affairs, therefore, is impossible simply in virtue of the fact that
the structure that defines either (and endows it with its particular causal
properties) involves relations between the constituents of s that are not

102 Stanford Encyclopedia of Philosophy Summer 2015 Edition 103