Organon F 21 (4) 2014: 471-491

Tichý’s Possible Worlds

JIŘÍ RACLAVSKÝ
Department of Philosophy. Faculty of Arts. Masaryk University
Arne Nováka 1. 602 00 Brno. Czech Republic
raclavsky@phil.muni.cz

RECEIVED: 30-04-2014  ACCEPTED: 11-08-2014

ABSTRACT: Pavel Tichý originally published his interesting conception of possible

worlds in 1968. Even though he modified it over the following twenty five years, its core
remained unchanged. None of his thirty journal papers or books containing the notion
of possible worlds was a study in metaphysics. Tichý (and most of his followers) always
introduced the notion in the context of other investigations where he applied his
Transparent intensional logic either to the semantic analysis of natural language or to
the explications of other notions. Tichý presented his conceptions using rather short
descriptions occurring on a number of places; his proposal appears not only fragmentary
but also somehow incoherent. The main contribution of this paper is thus not only
a complete survey of Tichý’s development of his conception but also a certain comple-
tion of the very proposal.
KEYWORDS: Actualism – combinatorialism – possible worlds – possible world semantics
– Transparent intensional logic.

1. Introduction

It is well known that the modern story of possible worlds began with
Wittgenstein’s states-of-affairs (see Wittgenstein 1921/1922). Carnap
(1947) introduced a linguistic mirror of this conception (viz. state-des-
criptions) to formal semantics. Through a development best summarized
elsewhere (e.g. Copeland 2006), Kripke (1963) and others supplemented
formulas of modal logic with model-theoretic semantics based on the no-

© 2014 The Author. Journal compilation © 2014 Institute of Philosophy SAS

472 JIŘÍ RACLAVSKÝ

tion of possible worlds. During the 1970s, Lewis proposed basics of his
theory (1973); Plantinga (e.g. 1970), Adams (1974), Stalnaker (1976) and
others significantly contributed to the debate. As the reader will be aware,
the contemporary debate revolves around Lewis’ landmark monograph
(1986); further notable books include Armstrong (1989), Divers (2002),
Nolan (2002) and Yagisawa (2010).
The present paper deals with a remarkable conception of possible
worlds developed by the post-Prague Spring refugee Pavel Tichý (1936
Brno – 1994 Dunedin) who moved to New Zealand where he became
a professor of philosophy. Tichý never wrote a systematic paper on possible
worlds, nor a book, which is one of the reasons why his conception is not
well known. Another reason might be simply that he exposed the notion
outside metaphysical debate, in the context of different investigations –
where he applied his system of intensional logic. 1
My main objective is to present Tichý’s proposal, while focusing on its
recent version. Though his conception of possible worlds is shared and
sometimes discussed by his sympathisers (cf. below), the present paper pro-
vides its first complex survey.
Tichý developed his views in four stages:

1) around 1969 (the papers published in 1968, 1969 and 1971)

2) 1971–79 (especially the unpublished book 1976 and a number of re-
lated papers mainly in the late 1970s)
3) between 1980 and 1988 (partly in cooperation with G. Oddie)
4) 1988 and after (his supreme proposal published mainly in 1988).

However, such division is somehow artificial because the stages are overlap-
ping and a number of ideas persisted throughout. One of my main goals
then is to show a unity of Tichý’s conception – even if there is an apparent
incompatibility between its historical variants.
None of Tichý’s expositions of the notion is extensive: it is usually less
than half of one page. Tichý’s most comprehensive treatise can be found in
his excellent monograph The Foundations of Frege’s Logic (see Tichý 1988),
but the core part of its section ‘36. Possible Worlds’ covers only three pag-
es. If we compare Tichý’s presentation with rivalling proposals of 1970s and
early 1980s, we may say that it is likewise brief. After Lewis’ monograph

1
As remarked by one of the reviewers, there is also a reason that Tichý took possible
worlds as primitive; however, this is not so simple (cf. below).
 TICHÝ’S POSSIBLE WORLDS 473

(1986) and subsequent philosophical debates, however, such proposals seem

sketchy. Though I fill some gaps in presentation of Tichý’s position, limits
of the paper format force me to omit an extensive discussion of his concep-
tion. Recall also that Tichý as well as his followers are more interested in
logical matters than in philosophy, which is why there has been no philo-
sophical elaboration of Tichý’s conception so far.
I will, of course, add explanatory remarks that should help the reader to
compare Tichý’s conception with other proposals. But it should be kept in
mind that the contemporary understanding of possible worlds and the clas-
sification of their conceptions was proposed later (cf., e.g., Haack 1978,
Menzel 2014) than Tichý developed his views. Since Tichý is a distinctive
thinker, his conception is sometimes difficult to subsume into one particu-
lar category.
In the following Section (2), I explain Tichý’s first conception followed
by an immediate assessment (Section 3). In Section 4, I introduce its
second and third stage. Then, in Section 5, I describe Tichý’s supreme
conception. In Section 6, I briefly explain his intensional logic and seman-
tics and provide a conclusion (Section 7).

2. The first stage: procedures and intensional basis

During the late 1960s, Tichý faced limits of analysis of our conceptual
scheme by means of classical extensional logic. Being well acknowledged
with algorithms and related problems, he proposed in his textbook intro-
ducing logic as a framework of science (1968a), to enrich predicate logic by
means of higher-order logic developed by Church (e.g., Church 1940).
From a debate between Tichý (e.g., Tichý 1966) and Czech extensional lo-
gicians it follows that he knew Tractarian/Carnapian ideas concerning the
content of empirical/non-empirical sentences, whereas the latter ones are
valid in all circumstances.
Tichý’s paper “Intensions in Terms of Turing Machines” published in
Studia Logica (see Tichý 1969) benefited from his previous thoughts;
a shorter version already appeared in a Czech philosophy journal (see Tichý
1968; for its English translation, see Tichý 2004). Tichý presented there
his first compact conception of possible worlds.
As Tichý (1969, 7–9) explained, the aim of empirical investigation is to
provide results of tests or procedures examining which attributes (i.e. proper-
474 JIŘÍ RACLAVSKÝ

ties and relations) are possessed by investigated objects. The sentence such
as “The object A is heavier than B” is a record of this: it amounts to saying
that A was positively tested on having that attribute.
To verify the sentence, one executes the procedure consisting of testing
A on being such and such. The meaning of the sentence is identified with
a complex concept – the procedure. Procedure is an older term for algo-
rithm. Procedures split into two kinds, empirical and non-empirical. For an
execution of any empirical procedure, one has to introduce state of the ex-
ternal world.
According to Tichý (1969, 9), language is based on an empirical system
which consists of a particular fixed finite set of individuals and a fixed
register of elementary tests, such as BE HEAVIER THAN B, called intension-
al basis.
Tichý always repeated this idea, only deleting “finite” and replacing
“empirical system” with “epistemic framework” and “set of individuals” with
“universe of discourse”. He admitted partial attributes to be in intensional ba-
sis (cf., e.g., Oddie – Tichý 1982, 234, 3.1.a). Tichý also dismissed the idea
of intensional independence of tests employed in Tichý (1969) because he
soon realized (though he did not mention) the force of Kemeny’s (1951)
objection to Carnap (and Wittgenstein) according to which some intuitive
attributes depend on others; in other words, that atomic facts are not nec-
essarily independent.
Possible worlds are systems of possible outcomes of applications of proce-
dures collected in intensional basis:

each combinatorial possibility as to the outcome of applications of all

the tests in the intensional basis to all individuals (or to couples of indi-
viduals etc.) must be regarded as a conceivable state of the external
world. Let us call these possibilities briefly possible worlds with respect to
the empirical system. (Tichý 1969, 9)

Tichý also counted the number of possible worlds, which reminds us of

Wittgenstein’s counting in Wittgenstein (1921/1922, 4.42). Tichý added
that pure semantics cannot decide or assume which world is the actual
one.
The rest of Tichý’s paper focuses on defining intensions as classes of
equivalent procedures (procedures are equivalent if they yield the same out-
puts for the same systems of inputs and states of external world). Inten-
 TICHÝ’S POSSIBLE WORLDS 475

sions were thus not functions from possible worlds (Ws) as we know it
from possible world semantics and Tichý’s later writings. 2
In the subsequent paper “An Approach to Intensional Analysis” pub-
lished in Noûs as Tichý (1971), he reminded the reader of his definition of
intensions in Tichý (1969) and proposed his novel method of using Church
(1940)’s typed λ-calculus for analysis of natural language (cf. Section 6).

3. Brief characterization of Tichý’s conception

Since Tichý never made a substantial change to the above picture, we

may already briefly compare it with rivalling proposals.
Despite being inspired by Tractarian/Carnapian combinatorialism, 3
Tichý does not subscribe to its contemporary Armstrong’s (1989) version
(the differences between the two will be more evident from the next two
sections). Since Tichý’s worlds are external to language (though they are
linked to epistemic base of a language), his conception of possible worlds is
not linguistic ‘ersatzism’ 4 as, e.g., Hintikka’s (1969) conception.
Though Tichý’s views seem to contain some ideas of abstractionism, 5
Tichý maintained that the mathematical nature of the combinations leads
to the conclusion that logical space, i.e., the set of possible worlds, has to be
homogenous (see Tichý 1988, 179). He thus rejected Stalnaker’s idea (1976)
to set apart our actual world from the other possible worlds, which are only
admitted as (useful) fictions. Modal fictionalism was dismissed by Tichý be-
fore it came into existence:

2
In intensional semantics and logic, intensions are functions from possible worlds
(Ws). Intensions include propositions whose values are truth-values, properties whose val-
ues are classes of objects, relations whose values are classes of n-tuples of objects, etc.
That an object instantiates (exemplifies, possesses) a property in (or: at) a given possible
world W means that the object occurs in the extension of that property in (at) W.
3
The view that (alternative) possible worlds are results of recombination of meta-
physical elements.
4
The view that (alternative) possible worlds are mere maximal consistent sets of sen-
tences.
5
The view that alternative possible worlds are mere abstractions from the actual
(real) world.
476 JIŘÍ RACLAVSKÝ

If unrealized determination systems [i.e. ‘possible worlds’] are mere fic-

tions, then so is the realized one. (Tichý 1988, 179)

Obviously, Tichý is an actualist about possible worlds: all possible worlds

exist (yet only one of them is actualized).
Tichý’s worlds are not concrete entities, they are classes of tests and
each test is distinguished from its empirical execution (compare it with
a computer program as such and its concrete execution). Tichý’s concep-
tion is thus not (Lewisian) concretism. Tichý sharply rejected Lewis’ con-
ception as an absurdity (cf. Tichý 1988, 177-180). 6
Lewis proposal is the only rival conception of possible worlds Tichý
discussed and explicitly referred to. 7 In Tichý (1975, 91-92), he objected
also to Lewis’ construal of actuality. According to Tichý, the word “actual”
stands for the identity function on worlds; this explains why an addition of
the word to a descriptive phrase is redundant. He maintained that we, ig-
norant of many facts, are not omniscient, thus we are hardly capable to
identify the actual possible world. Tichý repeated this idea in a number of
places.
In Tichý (1971, 274-277) and also in Tichý (1988, 180-183), he passio-
nately argued against varying domains and possible individuals. Without
a demarcation of a domain, quantification over x cannot be logically satis-
factory. The alleged examples of possible individuals (Kripke’s Pegasus, 1963)
are only examples of this or that individual concept – Tichý instead used
his terms individual role or individual office (an entity an individual can oc-
cupy). Tichý explicated individual offices as intensions having individuals as
values. He elaborated and generalized the theory of offices to interesting
heights. 8
Tichý was a strong antiessentialist and ‘haecceitist’ (cf. Raclavský 2008;
2011). Though individuals do not have a genuine essence (but see Cmorej

6
According to Tichý (1988), Lewis’ worlds are not in space, since they are causally
unrelated to our world. Thus, they are simply nowhere because space is by definition
all-embracing. The ‘newspeak’ Lewis proposed does not help explain modality as ap-
pearing in everyday sentences.
7
The other metaphysical conception Tichý provably knew was Plantinga’s early
theory (1970), but he criticized Plantinga (1972)’s essentialism.
8
See Tichý (1987), which is an adaptation of his introduction to Tichý (1976). For
its application see Tichý (1978d); for his analysis of Anselm’s ontological proof see
Tichý (1979).
 TICHÝ’S POSSIBLE WORLDS 477

1996; 2001), individual offices do have essences (an essence is a unique sum
of defining properties called “requisites”). Tichý quantified over all individ-
uals there are and, with the help of another variable, over all individual of-
fices. Tichý is thus an actualist, though not a paradigmatic one: alleged
possible individuals, excluded by a paradigmatic actualist from quantifica-
tion, are treated by Tichý as individual offices and these are in the domain
of quantification.
According to Tichý, the kind of existence which is applicable to individ-
uals is trivially applicable to them. But he introduced (e.g., in Tichý 1979;
or Tichý 1988; 1976) a kind of existence nontrivially applicable to offices;
to say that the Pope exists in a given world W amounts to saying that the
Pope-office has a holder in W. Nontrivial existence ascribable to properties
(relations) says that the property has an instance in W (cf. e.g. Raclavský
2010 for more).

4. The second and third stage: intensions, primary / derived

attributes, nexuses

The second stage of Tichý’s considerations on possible worlds begins

after the 1971-paper. At that time, Tichý (1974) disproved Popper’s defini-
tion of verisimilitude (likeness of theories of truth) and worked on a positive
proposal within the framework of his intensional logic, which he published
in Tichý (1976a). This was followed by its slightly modified version in
Tichý (1978d). 9 Recall that truthlikeness measured within the framework
of intensional logic (Tichý, Oddie, Niiniluoto and others) is closely related
to the topic because it deploys possible worlds underlying scientific theo-
ries.
Especially in Tichý (1978d), he focused more on entities in intensional
basis. He emphasized the difference between primary attributes which are
collected in the intensional basis and derived attributes. The idea is ex-
plained further in his unpublished book Tichý (1976) where he remarked
that to find out whether Xantippa is a widow one has to ascertain some
more basic facts, e.g. that Socrates is dead. The idea goes back to the

9
Tichý’s former pupil Graham Oddie wrote a whole book (see Oddie 1986) elaborat-
ing on and defending Tichý’s approach to truthlikeness. In Oddie (1987), he indirectly
supports Tractarian inspiration for Tichýan worlds.
478 JIŘÍ RACLAVSKÝ

aforementioned Kemeny’s criticism of Carnap and it seems to be a prede-

cessor of Kim’s idea of supervenient properties. 10
The second feature distinguishing the second period from the first is
that intensions are not equivalent classes of procedures, but total or partial
functions (as mappings) from possible worlds (cf. footnote 3).
A significant Tichý’s work of this period is an extensive unpublished
monograph “Introduction to Intensional Logic” (see Tichý 1976) written
between 1973 and 1976. The book contains the first main, atemporal, ver-
sion of his logical system (see Section 6) and its applications to natural lan-
guage. Its important part is a large chapter on subjunctive conditionals and
related phenomena, which makes a third distinctive feature of this stage. 11
Most of Tichý’s papers published between 1975 and 1979 are nothing
but selections from this book. Needless to say that his Czech and Slovak
namesakes Pavel Materna (known for his later popularization of Tichý’s
logic) and Pavel Cmorej (who also utilizes his system) persuaded Tichý to
adopt temporal parameter and most of his papers published at that time al-
ready contained its adoption in a rudimentary form. A typical example of
a brief exposition of possible worlds in such a paper can be found in “New
Theory of Subjunctive Conditionals” (see Tichý 1978). Note the similarity
with his first stage proposal:

The aim of the investigation delimited by an epistemic framework is to

determine exactly how the attributes from the intensional base are dis-
tributed through the universe of discourse at various moments of time.
Before the investigation gets off the ground, the investigator faces
a range of possibilities on that score. These possibilities are usually
called, somewhat dramatically, possible worlds, and the totality of possible
worlds is known as the logical space of the framework. (Tichý 1978, 435)

A feature more characterizing the third stage is Tichý’s emphasis on

nexuses (connections), i.e. higher-order relations such as cause-effect relation

10
Later, Tichý wrote a paper (co-authored by Oddie, 1990) on resplicing properties
over a supervenience base.
11
Tichý published three papers on subjunctive conditionals. The first one (Tichý
1976b) disproves Lewis-Stalnaker’s theory. The last one (Tichý 1984) disproves, among
others, Tichý’s own theory from (1978). In Tichý (1984), the truth of subjunctive con-
ditionals depends, inter alia, on a certain unarticulated, tacitly assumed parameter (such
as usual laws of nature).
 TICHÝ’S POSSIBLE WORLDS 479

(or even attitudes), as discriminating between possible worlds. He puts

them in an intensional basis:

We thus see that a possible world is not fully described in terms of ob-
servable events such like X’s being dry, X’s being stuck, and X’s burn-
ing, taking place at definite times. What makes a world the world it is
are also connections which hold between such events. Worlds differ from
one another not just in what observable events take place in them, but
also which events have the power bring others about. (Tichý – Oddie
1983, 136)

However, we may find that Tichý already considered nexuses to be among

the main characteristics of worlds in his (1978, 435).
Thus, the only decisive feature differentiating the third stage from the
second one is Tichý’s stress on temporality. In the third stage, he signifi-
cantly employed verb tenses and also events/episodes for characterizing
possible worlds. 12 Here is an illustration from the introductory parts of the
papers which utilize temporality in explication of ability and freedom:

worlds must be allowed to branch: there must be worlds which are po-
pulated by the same individuals and whose histories are the same up to
a certain time, and different after that time. (Tichý – Oddie 1983, 135)

As the possession of an attribute by an object is a time-dependent affair,

the possibilities are, more particularly, possible histories of the distribu-
tion. (Oddie – Tichý 1982, 228)

5. Tichý’s late conception: determination systems

Tichý’s late conception is clearly recognizable from its stress on the idea
of determination systems. These are discussed only in his masterpiece The
Foundations of Frege’s Logic (see Tichý 1988) and in Tichý (1994), which is
a posthumously published introduction to his just written, but unfinished,
book Meaning Driven Grammar which aimed to provide a highly ambitious
semantic analysis of natural language. In his explanations, Tichý presup-

12
See mainly his excellent analyses of verb tenses and temporal adverbials in (1980)
and analyses of episodic verbs and verb aspects, as well as events and episodes, in (1980a).
See also his philosophical defence of temporal dependence of truth (1980b).
480 JIŘÍ RACLAVSKÝ

posed a reader who has only a cursory knowledge of possible worlds; he

even cited no his earlier relevant paper.
To introduce determination systems, Tichý deploys the notion of
fact:

What are facts? The notion of fact is correlative with that of determin-
er. To each fact there corresponds a determiner in such a way that the
fact consist either in determiner’s singling out a definite object or in its
failing to single out anything at all. The fact that Scott is the author of
Waverley, for instance, consists in determiner A’s [= THE AUTHOR OF
WAVERLEY’s] singling out Scott. The fact that the author of Waverley
is a poet consists in the proposition P’s [= THAT THE AUTHOR OF
WAVERLEY IS A POET’s] singling out [the truth-value] T. (Tichý 1988
178)

Obviously, this is nothing but Tichý’s ‘tests-on-individuals’ story. In-

stead of “attribute” or “office”, Tichý uses the term “determiner”; determin-
ers are explicated (in Tichý 1988, 198) as possible world intensions. Instead
of “test with a positive/negative outcome”, Tichý says fact; facts are actual
(i.e. obtaining) or possible. Tichý explicated the notion of fact as proposi-
tion because he considered A IS GREATER THAN B and B IS SMALLER
THAN A being one and the same intuitive fact. Tichý’s most elaborated
view on the notion of fact appears in his re-examination of Wittgenstein’s
Tractatus (cf. Tichý 1994a).
Intuitive possible worlds are total (i.e. maximal) collections of facts:

Thus the determiner’s [= THE AUTHOR OF WAVERLEY] picking out

George IV is a possible fact and so is its picking out nothing at all. Hence
a possible world, if conceived as a totality of possible facts, will be fully
characterized by an assignment of objects (of appropriate kinds) to some
determiners. Let us call such assignment a determination system. A de-
termination system is thus any many-to-one correspondence associating
(some) individual determiners with individuals, (some) truth-value de-
terminers with truth-values, etc. (Tichý 1988, 178)

Briefly, a determination system specifies one combinatorial possibility as

to what objects are determined (or singled out) by what intensions at
what times. (Tichý 1988, 199)

The determiners involved in intensional basis can be of various types:

 TICHÝ’S POSSIBLE WORLDS 481

[determiners] may come classified into categories, like colours, heights,

propositional attitudes, and the like. (Tichý 1988, 199)

Realize that every combinatorially possible association of determiners with

determinees is a determination system only relative to a given epistemic
framework.
Let us assume an illustrative epistemic framework with a universe only
containing A(lan) and B(arbra) and intensional basis only containing
attributes (determiners) MAN and WOMAN. When ignoring partiality, there
are exactly 16 mappings associating the determiners with their possible val-
ues (cf. Tichý 1988, 178), i.e. 16 determination systems:

MAN WOMAN MAN WOMAN

1. {A, B} {A, B} 9. {B} {A, B}
2. {A, B} {A} 10. {B} {A}
3. {A, B} {B} 11. {B} {B}
4. {A, B} ∅ 12. {B} ∅
5. {A} {A, B} 13. ∅ {A, B}
6. {A} {A} 14. ∅ {A}
7. {A} {B} 15. ∅ {B}
8. {A} ∅ 16. ∅ ∅

As Tichý explains in (1988, 178-179), since determiners are not always

mutually independent, not every combinatorically possible determination
system is realizable. The two determiners MAN and WOMAN cannot pick
out overlapping classes of individuals. Thus, not every determination sys-
tem is a possible world. On the other hand, every possible world is
a determination system. Given our intensional basis and universe of dis-
course, only the determination systems whose numerals are written in bold
(4, 7-8, 10, 12-16) count as (intuitive) possible worlds; we therefore have 9
possible worlds altogether. 13

13
In Tichý (1988, 179), Tichý speaks about 8 worlds but, when discussing an entirely
analogous example in his (1994, 60), he mentioned the number 9, which is, I maintain,
the correct one (my reason occurs in the next paragraph).
482 JIŘÍ RACLAVSKÝ

Before proceeding further, I should state a few remarks explaining

Tichý’s position. The possible world no. 16 seems to be a suspicious empty
world. However, this idea is mistaken. Firstly realize that Tichý’s worlds are
not individuated by a certain population of individuals. 14 The possible
world no. 16 is simply a world in which A and B are tested on instantiation
of the properties MAN and WOMAN with a negative result. The individuals
have different properties than the two primary ones contained in our mi-
niature intensional basis, e.g. the property BE NOT A MAN. 15
Unrealizable determination systems in some ways resemble impossible
possible worlds which are discussed so much today (cf., e.g., Berto 2013; Va-
cek 2013). Impossible worlds are proposed for a variety of reasons, none of
them are important for Tichý. For instance, he prevented the problem of
logical omniscience by employing hyperintensional entities (cf. Section 6).
Recall also that Tichý proposed possible worlds in connection with the
idea of subject investigating the external, empirical reality. Both empty and
impossible worlds are beyond such considerations. In other words, they are
excluded on the basis of Tichý’s pre-theoretic motivation.
Surprisingly, Tichý (1988) did not employ and develop an idea which
suggests itself, viz. that a determination system is only a single ‘slice’ (or
‘time point’, Kuchyňka – Raclavský 2014) of a possible world, which is thus
a sequence of such slices. However, an anticipation of this idea can be
found in (Oddie – Tichý 1982, 228, cf. the quotation above), its full ex-
pression is this:

we must think of it [i.e. possible world] as a world history, a whole

course of events unfolding in time. In other words, a world must be
conceived of not as a single distribution of the traits from the inten-
sional base through the universe of discourse, but as a series of such dis-
tributions, one for each moment of time. (Tichý 1994, 62)

14
Vladimír Svoboda (2001) imagined that individuals may escape ‘our world’, moving
thus to the world no. 16, by losing any (remarkable) property and reaching a kind of
‘limbo’. A closer examination of this idea (e.g. Raclavský 2008b) leads to the conclusion
that such proposal contradicts Tichý.
15
Some metaphysicians reject ‘negative’ properties but Tichý belongs to the opposite
camp: BE A NON-FERROUS, for instance, is a good property the possession of which can
be sensibly empirically tested. When properties are explicated as possible world inten-
sions (cf. Raclavský 2011, I.2) any trace of negativity evaporates, because no intension
displays any mark of negation.
 TICHÝ’S POSSIBLE WORLDS 483

So far, we have an intuitive notion of possible world as an entity which

contains facts whereas a fact consists in that a determiner picks out a de-
terminee. What then are the logically simple entities W1, W2,…, Wn which
are the arguments of Tichý’s intensions?
The question is very important because Tichý’s construal appears to be
incoherent: Tichý often speaks about possible worlds in the spirit of com-
binatorialism but then he says that in his type-theoretic logical framework
possible worlds are members of the type ω, which is an atomic type thus its
members are primitive – this seems to mean that they do not possess com-
binatorialistic features. 16
Oddie (1986, 125) first reacted to the problem and suggested to call
these Ws proto-worlds and assumed their correlation with the ‘thick’ possi-
ble worlds. Tichý in (1988, 194-200) enlightened us of the link between
the two kinds of worlds as follows. (The following explanation appeared
firstly in Raclavský 2009, 9-11.)
Firstly recall Carnap’s notion of explication: it consists in the replacing
of an intuitive notion – here the notion of possible world as a collection of
facts – by its rigorous mate, its explicatum. Now there is a small complica-
tion. Tichý explicated facts as possible world propositions, which are classes
of world/time couples. As noted already by Stalnaker (1976) and even
Adams (1974), neither mentioned by Tichý, there is a question how can
possible worlds be classes of propositions when propositions are classes of
worlds (or world/time couples). 17
Tichý realized this circularity problem (cf. Tichý 1988, 194) and solved it
by carefully distinguishing between logically primitive and complex entities
and their role in explication. 18 For Tichý, propositions are the primary goal
of his investigation, thus possible worlds must be taken as logically primi-
tive, while propositions will then be defined in terms of worlds.

16
As Pavel Cmorej reminded me in personal communication, these possible worlds
are not quite pure entities because a system of functions based on such worlds encodes
the properties of intuitive determination systems. Thus, they are rather surrogates of
‘full blooded’ worlds as we will show in our argument.
17
Oddie (1986, 125) considered a slightly different puzzle: possible worlds have to be
specified even by some functions defined on worlds (e.g. nexuses between propositions);
but how then can one specify a single world? Oddie’s solution anticipated Tichý’s.
18
A logically primitive entity differs from another entity of the same sort only by its
different numeric identity.
484 JIŘÍ RACLAVSKÝ

But to explain anything, propositions must be tied into the system of

explication to the intuitive notions because it does matter whether a par-
ticular class of world/time couples explicates the intuitive proposition
ALAN IS A MAN or rather ALAN IS NOT A MAN. Such a link of a possible
world proposition to its intuitive correlate is guaranteed by an interpretation
of the members of ω and the two truth-values. Interpretation is obviously
a reverse of explication:

Now to interpret the basic category ω is to assign to each of its members

a unique determination system. … The determination systems which are
assigned, within the [given epistemic] framework, to ω-objects are called
the possible worlds of the framework. (Tichý 1988, 199-200)

Let us briefly reflect on Tichý’s position. Tichý’s contribution to meta-

physics consists in explaining possible worlds as determinations systems.
This is a refinement of the idea that possible worlds are total classes of
compatible facts. 19 For logical investigations, however, Tichý decided to
use only surrogates of these worlds, namely the logically primitive entities
W1, W2,…, Wn, which are interpreted by the possible worlds as determina-
tion systems (or rather their chronologies).
A metaphysician can be perhaps a bit uncomfortable with the fact that
Tichý’s framework, used by him and others for a number of useful explica-
tions of various notions, does not provide a ‘full blooded’ explication of the
notion of possible world. Yet there is a way out of this suggested by Rac-
lavský (2009a) who utilized ideas and apparatus from Raclavský (2008a) and
Raclavský – Kuchyňka (2011). 20 We can retain Tichý’s framework as it is,
with all those explications, but we must rethink the role of Ws. I suggest
understanding Ws as a mere modal factor (I suppress some questions related
to interaction with temporality). A proposition, for instance, is clearly
a function whose values depend on logical modality (modal factor). Possible
worlds are then explicated as some other, complex entities which reflect the
intuitive features of classes of facts, whereas facts are some structured entities,

19
Which goes back to C. I. Lewis (see Lewis 1923).
20
The original motivation is Raclavský’s solution (2007; and 2014 with Kuchyňka) to
the puzzle studied already by Petr Kolář in his (e.g. 2002): Tichý rejected the idea of
facts as structured entities by evoking the famous aRb fact; but then, there is the prob-
lem of how to preserve correspondence theory of truth which is based on some sort of
isomorphism between sentences and their significance (facts).
 TICHÝ’S POSSIBLE WORLDS 485

not possible world propositions. It is interesting that such possible worlds

would be very close to the worlds of Tichý’s first stage conception (recall that
those worlds were classes of structured procedures expressed by sentences).

6. Possible worlds in Tichý’s Transparent intensional logic

In this section, I am going to briefly explain the role of possible worlds

(as Ws) in Tichý’s semantics and logic. It is important to stress that Tichý
adopted the notion of possible world for utilization in philosophical logic,
not as an object of metaphysical considerations; this is somehow obscured
by the fact that in Tichý’s work logic and philosophy go hand in hand.
An admissible introduction to Tichý’s semantics likens it to Montague’s
semantics which was published mainly around 1970, cf. the posthumous
collection (see Montague 1974). Tichý referred to Montague’s work in
(1971) and it seems that he became acknowledged with Montague’s results
after developing his own system. Unlike Montague, Tichý made
a straightforward adaptation of Church’s typed λ-calculus because he
enriched Church’s type basis containing ο (two truth-values) and ι (indi-
viduals) by the type ω. Over such basis, there are numerous functions of
composite types; for instance, the type of propositions is (οω) (in another
notation: ω→ο). After his (1971), Tichý adopted partial functions, thus
some intensions can be without a value (gappy). In the late 1970s, he ad-
mitted the temporal parameter τ; intensions (e.g. propositions, properties,
…) are then functions from world/time couples (equivalently speaking, they
are functions from worlds to ‘chronologies’ of objects; for the sake of brevi-
ty, I will omit temporal parameter).
Tichý’s technical treatment of intensions is distinct from Montague’s
(cf. e.g. Tichý 1978b). Tichý’s terms standing for intensions are
λ-abstractions over possible worlds; but such λ-abstraction λw[…w…] can
be combined with variable w ranging over possible worlds in order to ex-
press a recourse to a value of that intension in the value of w. This way
Tichý could sensibly assign intensions to expressions having modally con-
ditioned reference (common sentences, descriptions, predicates, etc.) in
every context, including the transparent one, 21 which is the main difference
from Montague’s approach.

21
This is why he called his system Transparent intensional logic, TIL.
486 JIŘÍ RACLAVSKÝ

λ is a variable binding operator and λ-formalism is capable of expressing

subtle differences in scope. This enables a logically lucid treatment of de
re/de dicto propositional attitudes as well as scopes of modal operators. Let
us add that modal operators are classes of propositions, being of type
(ο(οω)), which means that they are ‘quantifiers’ of possible worlds. Famous
puzzles concerning modality are quickly solvable in TIL (cf. Tichý 2004;
Raclavský 2009).
Already in (1970), Lewis published a criticism of intensional (or possible
world) semantics. His main argument roughly says that possible world inten-
sions are too coarse-grained to be meanings of expressions. In other words,
intuitive meanings have a more fine-grained structure. We thus need hyperin-
tensional entities which would stand in semantic scheme between expressions
and extensions/intensions signified by them. For another argument, when
X believes that 1+1=2, we can hardly entail that X believes Fermat’s Last
theorem, despite that “1+1=2” and the famous theorem stand for one and
the same proposition which is true in all possible worlds.
After Tichý (1971), he found a solution to the problem. He realized
that his λ-terms can be read in an extensional way as representing usual
set-theoretic objects, e.g. possible world intensions, or they can be read in
‘intensional’ way as representing entities which determine the set-theoretic
objects. Tichý called those ‘intensional’ entities “constructions” and he later
explained (in his 1986) that he borrowed the term from geometry where
a point or circle can be constructed one way or another. Constructions are
structured, abstract, extra-linguistic entities which are akin to algorithmic
computations (Tichý 1986, 526). For a defence of constructions and their
careful description see esp. Tichý (1988).
Constructions can be used for capturing features of intuitive entities
that cannot be explicated by means of mere possible world intensions.
Tichý stressed (already in his 1976) mainly their usefulness as the explicata
of meanings. His semantic scheme is thus entirely Churchian (‘Fregean’): an
expression E expresses (in L) a construction C (= the meaning of E) whe-
reas C constructs an intension or extension (= the significance of E). For
instance, the meaning of the sentence “Fido is a dog” is the propositional
construction λw[Dogw Fido], which constructs the proposition FIDO IS A
DOG. Propositional attitudes are explicated in Tichý (1988) as attitudes to-
wards propositional constructions, not towards mere propositions; this
proposal blocks undesirable consequences embraced by intensional seman-
tics and solves the problem of logical omniscience.
 TICHÝ’S POSSIBLE WORLDS 487

The type of constructions is split. Tichý’s late system is thus a special

ramified theory of types. Consequently, it is easily capable of avoiding a num-
ber of paradoxes concerning ‘propositions’ (propositions not in the strict sense
of possible world semantics). One of the best known paradoxes concerning
possible worlds and ‘propositions’ is Kaplan’s (1994). In Tichý (1988, sec.
42, 218 and passim), he reminded the reader of the fact proven by Cantor
that there are more mappings from (nonempty) set S to (nonempty) set S’
than there are members of S. Thus, there are more propositions than
worlds: no mapping can associate, in 1-1 fashion, every proposition with
a certain world (or world/time couple). Consequently, no matter how one
explicates relations such as belief, assertion,… between individuals and
propositions, “there will always be a proposition such that at no world/time
couple is it the only proposition believed, or asserted, by George IV.”
(Tichý 1988, 219). Tichý demonstrated the situation and immediately
proved Theorem 42.1, the corollaries of which provide his solution to the
Liar paradox (which bears resemblance to his solution in his 1976).

7. Conclusions

To repeat the most essential features of Tichý’s conception of possible

worlds, Tichý proposed it in close connection with his sophisticated and
extensive logical system. Similarly as other intensional logicians/seman-
ticists, he used possible worlds for semantic analysis of natural language and
also further explication of various notions of our conceptual scheme such as
fact, causality, event, ability, freedom, propositional attitudes, subjunctive
conditionals and modalities. Though it is problematic to subsume his con-
ception under the known ones, “combinatorialism” and “actualism” seem to
be the most appropriate labels.
Tichý’s first proposal was published in 1968. His conception underwent
various modifications, but its main character remained unchanged. In the
first stage, possible worlds were classes of certain procedures (algorithms)
consisting mainly of tests on individuals. In the second and subsequent
stages, they were intuitively total classes of facts but Tichý technically
treated them as arguments for his intensions, i.e. as logically primitive enti-
ties. The combinatorial character was underlined by his late conception in
which the intuitive possible worlds are explicitly specified as some determi-
nation systems; determiners are then explicated as intensions.
488 JIŘÍ RACLAVSKÝ

A tension in textual evidence is caused by the fact that Tichý always de-
scribed possible worlds in the spirit of combinatorialism but, at the same
time, he treated them as logically primitive entities of his logical frame-
work. Above, I have reconciled the two views utilizing Tichý’s late sugges-
tion that the former worlds are only intuitive, pre-theoretic possible
worlds, while the latter worlds are their rigorous explicata.
Tichý usually stressed intensional dependence of attributes/determiners
and also the idea that worlds are individuated even by the nexuses and atti-
tudes which are realized in them. In the late 1970s, he added temporality.
Tichý always preferred fixed domain; he modelled ‘possibilia’ as individual
offices, i.e. some intensions. Since his recent type-theoretic framework is
explicitly ramified, his approach can avoid various paradoxes concerning
possible worlds and ‘propositions’.
Some questions not answered by Tichý, but posed and answered by
some his followers, concern the structured nature of facts, the nature of
correspondence, and the possibility to construct possible worlds from struc-
tured facts. Another important task is to elaborate consequences of tempo-
rality involved in Tichý’s conception of possible worlds. 22

References
ADAMS, R.M. (1974): Theories of Actuality. Noûs 8, No. 3, 211-231.
ARMSTRONG, D.M. (1989): A Combinatorial Theory of Possibility. Cambridge University
Press.
BERTO, F. (2013): Impossible Worlds. In: Zalta, E. N. (ed.): The Stanford Encyclopedia
of Philosophy. (Winter 2013 Edition). Available at: http://plato.stanford.edu/archives
/win2013/entries/impossible-worlds/.
CARNAP, R. (1947): Meaning and Necessity. The University of Chicago Press.
CHURCH, A. (1940): A Formulation of the Simple Theory of Types. The Journal of
Symbolic Logic 5, No. 2, 56-68.
CMOREJ, P. (1996): Empirické esenciálne vlastnosti. Organon F 3, No. 3, 239-261.
CMOREJ, P. (2001): Esencializmus verzus antiesencializmus. In: Na pomedzí logiky a filo-
zofie. Bratislava: Veda, 91-113.
COPELAND, B.J. (2006): Meredith, Prior, and the History of Possible Worlds Semantics.
Synthèse 150, No. 3, 373-397.
DIVERS, J. (2002): Possible Worlds. Routledge.

22
Thanks to Pavel Cmorej and both anonymous reviewers for their valuable sugges-
tions.
 TICHÝ’S POSSIBLE WORLDS 489

HAACK, S. (1978): Philosophy of Logics. Cambridge University Press.

HINTIKKA, J. (1969): Models for Modalities. Dordrecht: Reidel.
KAPLAN, D. (1994): A Problem in Possible-World Semantics. In: Sinnot-Armstrong,
W. – Raffman, D. – Asher, N. (eds.): Modality, Morality and Belief: Essays in Honor
of Ruth Barcan Marcus. Cambridge University Press, 41-52.
KOLÁŘ, P. (2003): Pravda a fakt. Praha: Filosofia.
KEMENY, J.G. (1951): Review of R. Carnap, Logical Foundations of Probability. The
Journal of Symbolic Logic 16, 205-207.
KRIPKE, S.A. (1963): Semantic Considerations on Modal Logic. Acta Philosophica Fenni-
ca 16, 83-94.
KUCHYŇKA, P. – RACLAVSKÝ, J. (2014): Pojmy a vědecké teorie [Concepts and Scientific
Theories]. Brno: Masarykova univerzita.
LEWIS, C.I. (1923): Facts, Systems, and the Unity of the World. The Journal of Philoso-
phy 20, No. 6, 141-151.
LEWIS, D. (1970): General Semantics. Synthèse 22, 18-67.
LEWIS, D. (1973): Counterfactuals. Cambridge (Mass.): Harvard University Press.
LEWIS, D. (1986): On the Plurality of Worlds. Blackwell.
MENZEL, C. (2014): Possible Worlds. In: Zalta, E. N. (ed.): The Stanford Encyclopedia of
Philosophy. (Summer 2014 Edition). Available at: http://plato.stanford.edu/archives/
sum2014/entries/possible-worlds/.
MONTAGUE, R. (1974): Formal Philosophy. New Haven – London: Yale University
Press.
NOLAN, D. (2002): Topics in the Philosophy of Possible Worlds. Routledge.
ODDIE, G. (1986): Likeness to Truth. Dordrecht: Reidel.
ODDIE, G. (1987): The Picture Theory of Truthlikeness. In: Kuipers, T.A.F. (ed.):
What is Closer-to-the-Truth? Amsterdam: Rodopi, 25-45.
ODDIE, G. – TICHÝ, P. (1982): The Logic of Ability, Freedom, and Responsibility.
Studia Logica 41, 2-3, 227-248.
ODDIE, G. – TICHÝ, P. (1990): Resplicing Properties in the Supervenience Base. Philo-
sophical Studies 58, No. 3, 259-269.
PLANTINGA, A. (1970): World and Essence. Philosophical Review 79, 461-492.
RACLAVSKÝ, J. (2008): Reformulating Tichý’s Conception of Bare Individuals. Organon
F 15, No. 2, 143-167.
RACLAVSKÝ, J. (2008a): Conceptual Dependence of Verisimilitude Vindicated. A Fare-
well to Miller’s Argument. Organon F 15, No. 3, 369-382.
RACLAVSKÝ, J. (2008b): Holá individua nejsou bez vlastností [Bare Individuals are not
without Properties]. In: Zouhar, M. (ed.): Jednotliviny, všeobecniny, významy. Bratis-
lava: Filozofický ústav SAV, 158-172.
RACLAVSKÝ, J. (2009): Jména a deskripce: logicko-sémantická zkoumání [Names and De-
scriptions: Logico-Semantical Investigation]. Olomouc: Nakladatelství Olomouc.
RACLAVSKÝ, J. (2009a): Structured Language Meanings and Structured Possible
Worlds. In: Munz, V. A. – Puhl, K. – Wang, J. (eds.): Language and the World.
Wien: Austrian Ludwig Wittgenstein Society, 349-350.
490 JIŘÍ RACLAVSKÝ

RACLAVSKÝ, J. (2010): Co obnáší kontingentní existence individuí [What is Contingent

Existence of Individuals?]. Organon F 17, No. 3, 374-387.
RACLAVSKÝ, J. (2011): Individua a jejich vlastnosti: studie z intenzionální metafyziky [Indi-
viduals and Their Properties: Studies in Intensional Metaphysics]. Olomouc: Naklada-
telství Olomouc.
RACLAVSKÝ, J. – KUCHYŇKA, P. (2011): Conceptual and Derivation Systems. Logic and
Logical Philosophy 20, 1-2, 159-174.
STALNAKER, R.C. (1976): Possible Worlds. Noûs 10, No. 1, 65-75.
TICHÝ, P. (1966): K explikaci pojmu obsah věty [On Explication of the Notion of Con-
tent of a Sentence]. Filosofický časopis 14, 364-372. 23
TICHÝ, P. (1968): Smysl a procedura [Sense and Procedure]. Filosofický časopis 16, 222-
232.
TICHÝ, P. (1968a): Logická stavba vědeckého jazyka [The Logical Structure of Scientific
Language]. Praha: Státní pedagogické nakladatelství.
TICHÝ, P. (1969): Intensions in Terms of Turing Machines. Studia Logica 26, 7-52.
TICHÝ, P. (1971): An Approach to Intensional Analysis. Noûs 5, No. 3, 273-297.
TICHÝ, P. (1972): Plantinga on Essence: A Few Questions. The Philosophical Review 81,
No. 1, 82-93.
TICHÝ, P. (1974): On Popper’s Definitions of Verisimilitude. British Journal for the Phi-
losophy of Science 25, No. 2, 155-188.
TICHÝ, P. (1975): What Do We Talk About? Philosophy of Science 42, No. 1, 80-93.
TICHÝ, P. (1976): Introduction to Intensional Logic. Unpublished book manuscript.
TICHÝ, P. (1976a): Verisimilitude Redefined. The British Journal for the Philosophy of
Science 27, No. 1, 25-42.
TICHÝ, P. (1976b): A Counterexample to the Stalnaker-Lewis Analysis of Counterfac-
tuals. Philosophical Studies 29, No. 4, 271-273.
TICHÝ, P. (1978): A New Theory of Subjunctive Conditionals. Synthèse 37, No. 3, 433-
457.
TICHÝ, P. (1978a): Verisimilitude Revisited. Synthèse 38, No. 2, 175-196.
TICHÝ, P. (1978b): Two Kinds of Intensional Logic. Epistemologia 1, 143-164.
TICHÝ, P. (1978c): De Dicto and De Re. Philosophia 8, No. 1, 1-16.
TICHÝ, P. (1978d): Verisimilitude Revisited. Synthèse 38, No. 2, 175-196.
TICHÝ, P. (1979): Existence and God. The Journal of Philosophy 76, No. 8, 403-420.
TICHÝ, P. (1980): The Logic of Temporal Discourse. Linguistics and Philosophy 3, No.
3, 343-369.
TICHÝ, P. (1980a): The Semantics of Episodic Verbs. Theoretical Linguistics 7, No. 3,
264-296.
TICHÝ, P. (1980b): The Transiency of Truth. Theoria 46, 2-3, 165-182.
TICHÝ, P. (1984): Subjunctive Conditionals: Two Parameters vs. Three. Philosophical
Studies 45, No. 2, 1-13.
TICHÝ, P. (1986): Constructions. Philosophy of Science 53, No. 4, 514-534.

23
Tichý’s papers are reprinted in Tichý (2004).
 TICHÝ’S POSSIBLE WORLDS 491

TICHÝ, P. (1994): The Analysis of Natural Language. From the Logical Point of View 3,
No. 2, 42-80.
TICHÝ, P. (1994a): The Tractatus in the Light of Intensional Logic. From the Logical
Point of View 3, No. 2, 32-41.
TICHÝ, P. (1987): Einzeldinge als Amtsinhaber. Zeitschrift für Semiotik 9, 13-50. Transl.
to Slovak as Tichý, P. (1994): Jednotliviny a ich roly [Particulars and their Roles].
Organon F 1, No. 1, 29-42; No. 2, 123-132; No. 3, 208-224; No. 4, 328-333.
TICHÝ, P. (2004): Pavel Tichý’s Collected Papers in Logic and Philosophy. Svoboda, V. –
Jespersen, B. – Cheyne, C. (eds.). Dunedin: University of Otago Press; Praha: Filo-
sofia.
TICHÝ, P. (1988): The Foundations of Frege’s Logic. Walter de Gruyter.
TICHÝ, P. – ODDIE, G. (1983): Ability and Freedom. American Philosophical Quarterly
20, No. 2, 135-147.
SVOBODA, V. (2000): Where Do All the Individuals Go? In: Childers, T. – Palomäki, S.
(eds.): Between Words and Worlds (A Festschrift for Pavel Materna). Praha: Filosofia,
21-33.
VACEK, M. (2013): Impossibilists’ Paradise on the Cheap? Organon F 20, No. 3, 283-
301.
WITTGENSTEIN, L. (1921/1922): Tractatus logico-philosophicus. Routledge & Kegan
Paul.
YAGISAWA T. (2010): Worlds and Individuals, Possible and Otherwise. Oxford University
Press.