J.R.R. Tolkien concluded his 1931 essay on the "Secret Vice" of language invention with some specimens of his own efforts in what he called
the one language which has been expressly designed to give play to my own most normal phonetic taste . . . and which has had a long enough history of development to allow of this final fruition: verse. It expresses, and at the same time has fixed, my personal taste. Just as the construction of a mythology expresses at first one's taste, and later conditions one's imagination, and becomes inescapable, so with this language. I can conceive, even sketch, other radically different forms, but always insensibly and inevitably now come back to this one, which must therefore be or have become peculiarly mine.(MC 212–13).
The language to which Tolkien was referring is represented in three of the specimen poems, "Oilima Markirya" ("The Last Ark"), "Nieninque" and "Earendel" (MC 213–16). It can now be traced back in all of its details, insofar as they were recorded by Tolkien in surviving documents, to its emergence in the Qenyaqetsa or "Qenya Lexicon" compiled about fifteen years earlier. It was in this dictionary with two and a half thousand entries that some of the earliest of Tolkien's mythological or legendary names were first recorded, or those that appeared in his contemporary poems were given linguistic explanations. In 1917 Tolkien compiled the lexicon of another invented language called Goldogrin or "Gnomish," many of whose words he designed to have etymological connections with Qenya, and words from the earlier language were cited in the "Gnomish Lexicon" to elucidate these connections. In The Book of Lost Tales, composed over the next few years, Tolkien elaborated and consolidated the mythological conceptions that appear in the early poetry and in some of the entries of the lexicons. The tales included a fictional history of the Elves and Gnomes for whom Tolkien imagined that his invented languages were their native speeches. In the early 1920s Tolkien wrote a grammar of Qenya which shows numerous conceptual associations with the Qenya Lexicon, but also expanded and revised some features of the inflexional patterns of the language that could be observed in specimens incorporated into earlier works. In the later 1920s [End Page 213] Tolkien continued this revision, recorded in a sequence of successively more elaborate paradigms, and he also compiled various Qenya word-lists. During the 1920s Tolkien also worked on Gnomish, writing a grammar of that language and various word-lists in which a prominent feature continued to be the etymological relation of the two languages, which underwent some theoretical changes as well.
This is the history of development to which Tolkien was referring in the essay, and the "one language" that he concluded had become peculiarly his is Qenya, although he does not give its name in the essay. The fourth specimen poem in the essay was composed in Gnomish, which is simply identified as "a totally different if related language" (MC 217). A study of the vocabulary of the poems included in the essay, as well as the draft versions that preceded them, has revealed that Tolkien used some words that go back to the Qenya Lexicon, and others that emerged in the later revisions to Qenya during the 1920s. It is apparently in this sense that Tolkien meant that the history of development of Qenya would "allow this final fruition: verse," insofar as it provided a sufficient accumulation of vocabulary for the variety of expression needed in poetry. And it is in this regard that the earliest invented words of Qenya would still have been part of the language unless Tolkien had consciously rejected them or intentionally replaced them with something else.
In 1926 Tolkien had written a "Sketch of the Mythology" (revised around 1930), in which he summarized the features of the Lost Tales necessary to an understanding of the long poems based on the "Tale of Turambar" and the "Tale of Tinúviel" that he was composing around that time (Shaping 11). Successive expansions and revisions of the sketch would produce the various annals and historical...