This was the case I was trying to distinguish:
(1) In our world, at some time prior to 1887, at 1 Bush Villas, Elm Grove, SouthSea, Portsmouth, England, Arthur Conan Doyle conceived of the fictional character "Sherlock Holmes".
MW: Arthur Conan Doyle identified a possible world in which these things are true.
EJB: Per Plantinga, Doyle constructed this “possible world”. But I would argue that he did that when he fleshed out that world, or part of it, in his first story (A Study in Scarlet, I think). I’m not sure that developing that story is what Simon meant by “conceived of the fictional character”. I don’t believe that it is necessary to have a “possible world” in order to have an intensional characterization, or a design, that may not have an instance.
(2) Before 1887 the character "Sherlock Holmes" did not exist as something to which the conception of could be attributed.
MW: Before 1887 there was no representation of this possible world in our world.
EJB: Simon says that before 1887, the classifier “Sherlock Holmes” itself did not exist – the term itself was meaningless. Matthew says only that Doyle’s constructed world, in which that concept is populated by an individual, did not exist at that time. Simon’s phrasing argues that the possible world somehow includes the intensions, the concepts, not just the population. So I can’t match this higher-order “possible world” with my (admittedly limited) understanding of Plantinga’s “possible world”. While I recognize the associated social phenomena, I find it very difficult philosophically to think of concepts themselves as having temporal parts.
EJB: I think it is more useful to model Sherlock Holmes as something more like a ProductDesign, which is an individual but behaves somewhat like a classifier in that it has an instantiation relationship to other individuals. We don’t have to ask whether a given ProductDesign exists as a concept outside of time and actualities. It is not a pure abstraction; it is a thing, and it came into existence when it was written down.
(3) In our world, the fictional character of Holmes was partially inspired by the actual person Joseph Bell, who was a Doctor.
MW: There is a similarity link between the Joseph Bell of this world and Sherlock Holmes in his possible world.
EJB: I think Simon’s point is: In what world does that relation have a corresponding actuality? Since Doyle’s world is constructed in our historical world, it exists in our world as a possible world, and the link exists in our world.
(4) In our world, the BBC commissioned a number of films featuring modern reinterpretations of the Conan Doyle stories, featuring a Holmes who in the worlds of those movies differed in some respects from the original source. [Written by Stephen Moffat, who isn't a Doctor, but writes one on TV]
MW: The films are representations of possible worlds that are similar to varying degrees to the possible world of Conan Doyle.
EJB: Here I will argue that Holmes as the licensed ProductDesign can have multiple implementations by different producers, all of which are distinguishable in ways that are irrelevant to satisfying the one common ProductDesign. The Sherlock Holmes design object exists in our world as a fictitious character, and it has instances in each of the films and stage plays, as well as an instance in Doyle’s work. But unlike the others, Doyle’s work created the design object itself. Other playwrights and screenplay authors may have constructed derivative designs, which does not necessarily make them inconsistent with the original design. But all the distinguishable instantiations are different precisely because they are distinguishable individuals.
(5) In the fictional world in which our Conan Doyle conceived, Holmes was a detective, who resided at 221B Baker St.
MW: Just some facts that are true in Conan Doyle’s (and possibly some of the other Sherlock Holmes variations) worlds.
EJB: Exactly. This is just a state that is included in Doyle’s possible world and others.
(6) In a possible real world, that Conan Doyle may have conceived of a fictional world in which Holmes was a baker, who lived at 221B Detective St.
MW: Just another possible world. Your possible real world is as fictional as Conan Doyle’s world of Sherlock Holmes.
EJB: And this possible world associates the term “Sherlock Holmes” with a significantly different ProductDesign. It is a homonym.
(7) In yet another possible world, Bell may have been partially inspired by Conan Doyle to conceive of a Holmes and write fictional stories featuring that character word-for-word identical with the ones in our Conan Doyle wrote.
MW: That is interesting, but no reason why people in two different possible worlds should not create a representation of another possible world.
EJB: And here Simon returns very clearly to the question of identity of a possible world. If Bell’s world is word-for-word identical in description to Doyle’s, how can it be different? Matthew says later that it is the same possible world, even if it is constructed by someone else. I agree. That is, I think we agree that Holmes-the-design is what is written, not the provenance metadata.
Is the fictional character in (1) the same fictional character as in (4)?
MW: They might share some temporal parts (if you take possible worlds to be branching) but they are not the same for the whole of their lives.
EJB: The underlying question here is what the term “Sherlock Holmes” refers to. Is it a collection of properties or a specific thing? If the latter, then it can only be the thing in Doyle’s possible world(s). If the former, it is a collection of properties exhibited by that thing, and the term then refers to a classifier which may have technically different extensions in different possible worlds. It seems to me that the phenomena of stage plays and films made from a book create different possible worlds in which the SAME “fictional character” has an extension. If we take Matthew’s view that these extensions are different things, then the term “Sherlock Holmes” must refer to a classifier, not an individual.
EJB: This is why I prefer the ProductDesign idea over the “classifier” idea. The design object is an individual thing (with one or more temporal parts), and it is related to other individual things that are its instantiations in various constructed worlds. That relationship might look like “Predicate maps thing to true/false”, but it isn’t – it is just a binary relation.
Is the fictional character in (5) the same fictional character as in (6)?
MW: No.
Is the fictional character in (1) the same fictional character as in (7)?
MW: Yes.
EJB: Because the worlds are identical in Matthew’s philosophy.
In our world could two people in the year 1700 discuss the character Sherlock Holmes?
MW: Only if they made a representation of him first (no reason why they should not, just like Bell in the other possible world, just incredibly unlikely).
EJB: It is not just incredibly unlikely. If we take (1) to be true “in our world”, then the two people in 1700 would have been discussing a meaningless term. And if they described the same classifier in terms of a sufficiently narrow set of properties, then Doyle did not invent the classifier, just a world in which that classifier had a non-empty extension. But this is just the kind of question that arises from deciding that a concept itself can be temporal, as distinct from having different possible and temporal extensions.
In our world, could two people in 2013 discuss real properties of the character Sherlock Holmes (e.g. the street address where the character was conceived).
MW: Of course. This is the place where the first (as far as we know) representation of that possible world was written down.
EJB: Agree. This part is at least consistent with Plantinga – the constructed possible world exists in the “real world” as a constructed world.
Regards
Matthew West
Information Junction
Tel: +44 1489 880185
Mobile: +44 750 3385279
Skype: dr.matthew.west
matthew.west@xxxxxxxxxxxxxxxxxxxxxxxxx
http://www.informationjunction.co.uk/
http://www.matthew-west.org.uk/
This email originates from Information Junction Ltd. Registered in England and Wales No. 6632177.
Registered office: 2 Brookside, Meadow Way, Letchworth Garden City, Hertfordshire, SG6 3JE.