The problem of fictional discourse -- do statements involving fictional objects have truth-values or are they truth-valueless; if the former, how are these truth values determined? -- is one that goes back at least to Frege (his views on the compositionality of language and what the referent of a sentence is entail that sentences which have non-referring parts have no truth-value) and for which no adequate solution has yet been found, given, e.g., the publication of books such as Tim Crane's The Object of Thought in 2013. I'm not going to survey all the issues arising from these problems, but instead present a problem for some accounts of the truth values of fictional statements which is, so far as I am aware, not often considered.
In Part 1 last week, one of my example languages was Adûnaic, one of the languages developed by J.R.R. Tolkien. I picked Adûnaic because of the context in which it appears in its most full form, Lowdham's report to the members of the Notion Club (The Notion Club Papers are themselves full of extremely interesting ideas about the nature of language and how linguistic meaning can be acquired, and will be the focus of a future Lesson), but any of his languages -- Quenya, Sindarian -- or indeed any constructed language of sufficient sophistication -- e.g., Klingon -- will serve my purposes.
One thing that is remarkable about constructed fictional languages such as Klingon, Quenya, etc., is the academic scholarship that they have generated. There is the The Elvish Linguistic Fellowship (a special interest group of the Mythopoeic Society, "devoted to the scholarly study of the invented languages of J.R.R. Tolkien"), which publishes two print journals, Vinyar Tengwar and Parma Eldalamberon, and an online journal, Tengwestië; Helge Kåre Fauskanger's Ardalambion, an extensive collection of links regarding Tolkienian languages; the Klingon Language Institute; and even the journal Tolkien Studies, which will not specifically devoted to his constructed languages publishes articles on them, such as Christopher Gilson's "Essence of Elvish: The Basic Vocabulary of Quenya". One conclusion that can be drawn from this extensive academic activity is that these fictional languages are taken as worthy objects of study, When it comes to languages, one of the basic requirements that it has to meet in order for it to be something worth studying is that it be meaningful. (While people can, and do, create "nonsense" languages, which are not meaningful, these do not have a history of generating the sort of scholarship that fictional languages like Klingon and Sindarin have.)
There are two types of contexts in which these languages can be found: In the original context of construction (e.g., in Tolkien's writings for the Elvish languages; in Star Trek TV episodes and movies for Klingon), and in external contexts (e.g., a wholly new composition of poetry in Quenya, or translation of the Bible into Klingon). Both of these contexts cause problems for straightforward explanations of how these languages are meaningful.
Adûnaic is set apart from other fictional languages by the paucity of examples that we have in it. As Fauskanger notes, "There are no coherent Adûnaic texts. Except single words scattered around in Lowdham's Report, most of the corpus consists of a number of fragmentary sentences given in SD:247, with Lowdham's interlinear translation" . Fauskanger supplements the translation that is found in Sauron Defeated with words whose meaning is known from other contexts, and gives it as follows:
Kadô Zigûrun zabathân unakkha... "And so / [the] Wizard / humbled / he came..." ...Êruhînim dubdam Ugru-dalad... "...[the] Eruhíni [Children of Eru] / fell / under [the] Shadow..." ...Ar-Pharazônun azaggara Avalôiyada... "...Ar-Pharazôn / was warring / against [the] Valar..." ...Bârim an-Adûn yurahtam dâira sâibêth-mâ Êruvô "...[the] Lords of [the] West / broke / the Earth / with [the] assent / of Eru..." ...azrîya du-phursâ akhâsada "...seas /so as to gush/ into [the] chasm..." ...Anadûnê zîrân hikallaba... "...Númenor / [the] beloved / she fell down..." ...bawîba dulgî... "...[the] winds [were] black..." (lit. simply "winds / black") ...balîk hazad an-Nimruzîr azûlada... "...ships / seven / of Elendil / eastward..." Agannâlô burôda nênud... "Death-shadow / heavy /on us..." ...zâira nênud... "...longing [is] / on us..." ...adûn izindi batân tâidô ayadda: îdô kâtha batîna lôkhî... "...west / [a] straight / road / once / went / now / all / roads / [are] crooked..." Êphalak îdôn Yôzâyan "Far away / now [is] / [the] Land of Gift..." Êphal êphalak îdôn hi-Akallabêth "Far / far away / now [is] / She-that-hath-fallen"
Though this is fragmentary, it is clear that what is being expressed is in the context of the legends of Middle Earth, that is, Adûnaic is being used to write what is from our point of view fiction. This fragmentary poem is not intended to be read literally as expressing statements about the actual world. This feature -- the use of the language to say something within the context of a fiction -- is to some extent shared by "external" uses of fictional languages, which by and large are also used for non-literal or non-factive writing or speech: Almost no one writes lesson plans in Quenya, composes a shopping list in Klingon, or sends submits meeting minutes in Sindarin.
The question then, is: If these languages are meaningful, how are they meaningful? What do we mean when we say that these fictional languages are meaningful, or at least intended to have meaning? If we mean that the sentences that are expressed by these languages have truth-conditions (e.g., a truth-conditional account of meaning has been adopted), then we are stymied by the fact that there is no good account of the truth-conditions of fictional language. One might say that there are truth conditions, we simply do not know them, and hence (via an epistemic rather than ontological truth-conditional account of meaning) we do not know the meaning of these sentences, even though they are meaningful. This sceptical position, however, does not seem to take account of our behaviour regarding sentences in these languages: We evaluate translations from, e.g., Quenya into English as correct or incorrect, we write literary criticism about the propositions expressed, and in general, we sure act as if we know what these sentences mean, at least in part, even though by assumption we don't know the relevant truth conditions.
Of course, it always possible to stick to one's philosophical guns, and simply reply that we are mistaken when we think we know what the sentences mean: But that seems to be a pretty tough horn to impale oneself on, because it doesn't provide any account for our behaviour with respect to these sentences. There is clearly some factor that guides our behaviour: There must be something in which referees evaluating prospective articles for Tengwestië ground their reports. If this is not the meaning of the sentences, then whatever it is, it certainly seems to be something which is functionally equivalent to meaning.
Lesson: We treat statements in constructed languages as meaningful. But if these languages are only ever used in fictional contexts, then any truth-conditional account of meaning which denies truth values to fictional discourse will have difficulty accounting for the fact that we do treat these languages in that way (that is, as meaningful).
© 2015, Sara L. Uckelman