|
|
|
|
|
|
ing to Plantinga (1974), "any possible world has its own book: for any possible world W the book on W is the set S of propositions such that p is a member of S if W entails p. . . . Each maximal possible set of propositions is the book on some world" (p. 46). |
|
|
|
|
|
|
|
|
To say, however, that setting up a possible world is like setting up a text does not imply that setting up a text necessarily means setting up a possible world. If I am telling someone how Columbus discovered America, I am undoubtedly producing a text which refers to what is commonly believed to be the 'real' world. By describing a portion of it, I am taking for granted all the rest of its individuals along with their properties and all the rest of the propositions holding in it. |
|
|
|
|
|
|
|
|
But something different happens when I set up a fictional possible world such as a fairy tale. When telling the story of Little Red Riding Hood, I furnish a world with a limited number of individuals (mother, girl, grandmother, wolf, hunter, a wood, two houses, a gun) endowed with a limited number of properties holding only for that world; for instance, in this story wolves can speak and human beings have the property of not dying when devoured by wolves.
6 |
|
|
|
|
|
|
|
|
Within this fictional world (a possible one constructed by the author), human individuals assume propositional attitudes; for instance, Little Red Riding Hood believes that the wolf is trustworthy. This world is a doxastic construct of the character (it is immaterial by now how much it overlaps the world of the story). As a doxastic construct it is presented by the author as one of the events of the story. Thus we have two partially different constructs: in the former wolves are not trustworthy, whereas in the latter they are. Since the final state of the story disproves the doxastic world of the character, one must ask to what extent those constructs are mutually comparable and accessible. To answer this question let me make some theoretical assumptions. |
|
|
|
|
|
|
|
|
8.4.3. Possible Worlds as Cultural Constructs |
|
|
|
|
|
|
|
|
First, let me assume that a possible world (hereafter W) is an ens rationis, or a rational construct. Within its frame the difference between individuals and properties should disappear, individuals being singled out as bundles of properties. Nevertheless, the distinction must be maintained for practical purposes, since no possible world sets up ex nihilo all its elements. Hintikka (1973) has shown how one can construct different possible worlds by differently combining four different properties. Given the properties |
|
|
|
|
|
|
|
|
shared by four individuals, as shown in Figure 8.3, there can be a W1 in which x1 and x3 exist and in which x2 and x4 do not. And there can be |
|
|
|
|
|