Time for illustrating those languages, yes? Yes.
At first I wanted to provide you guys with a propćdeutic to the ideas presented by Borges in this story. But then I realise that would easily become one of the most boring posts in the history of my posting, and that no-one wants a propćdeutic, even without knowing that word means. So I decided a good post to help ideas flow would be an application
of "Tlön, Uqbar etc". Because the Tlönian languages --- believe it or not --- do exist
in one way or another in the world. There are two flavours of Tlönian languages in the story: strictly verbal, and strictly adjective.
Strictly verbal languages
are funny birds. While, theoretically, they would be the most probable kind of minimalistic language to exist, there are no natural languages of such kind. I say they're probable based on the idea that all gramatical categories can be simulated by verbs.
An adjective, like "orange", can be substituted by the verb "to be orange". Any adverb can be converted into an auxiliary verb: "quickly" would be substituted with some broader distant cousin of "to rush". And so on: "upward behind the onstreaming it mooned."
Still. Perhaps the closest parallel to a strictly verbal language we have, as far as I know, is mathematics. Maths are an interesting kind of language. It presents broad concepts and their transformations: its main objective is to describe transformations, therefore it is essentially verbal. In the middle of the 20th century, two mathematicians came up with minimalistic mathematical systems which are indeed strictly verbal. The first of them is Haskell Curry's Combinatory Logic,
and the second is Alonzo Church's Lambda Calculus.
The two systems are fascinating and extremely insightful. For those interested, here's a primer to both languages by means of colourful alligator eggs.
But there's a less evident example of verbal languages. Visual languages as a whole (like paintings and traffic signs) tend to be adjective or nominal. One fascinating exception comes to mind: it's impressionistic paintings. Each trace in an impressionist painting is a verb. Some of them are "upwards it oranged", some of them are "it greened sideways, boldly", some of them are "behind the onstreaming it happened". No trace on its own ("it reddened downwards") makes sense, and it takes tens of them to make a unit of sense, and thousands of them to portray the sublime aspect of a situation in detail. They might not be sentences as simple as "upward behind the onstreaming it mooned", but they sure base themselves on the same principle.
Now, concerning adjective languages
... believe it or not, but they do exist as natural languages. Not purely adjective, I'm afraid, as no language is pure. But based on an adjective paradigm, which is to say they're nothing like anything you've seen. I take it as my example the only one of those languages I'm actually acquainted with: Old Tupi (or, as it refers to itself: nheʿengatu
, "the good language" --- literally, "good spoken").
Old Tupi is a dead language. It was spoken by the tribespeople from the coast of Brazil by the time of the discovery. Its fundamental building block is the adjective, for instance, porang
= "beautiful". Most adjectives can be appended a nominating suffix, -a, which turns them into nouns: hence poranga
= "beauty". Some nominalised adjectives can be possessed: xe poranga
= "my beauty". Therefore, their possesed adjective forms can be taken to work as verbs: xe porang
= "I'm beautiful" (more literally, "I have beauty", or even: "my being beautiful").
Some words cannot be posessed, like ybyrá
= tree. This must be taken to mean two things: there is no adjective or posessive relation between people and trees --- you cannot "have" a tree, and you cannot "be" a tree. Those words are taken to work as nouns. The rest of the language is strictly adjective made by either adjectives or modifiers, so what you have to understand is that the Old Tupi people probably thought of nouns as "funny adjectives" rather than as something else in its own right. This difference in the way of facing things leads to extreme consequences, which I invite the reader to think about.
More importantly though is that there are many classes of "pronominal modifiers", that is to say, modifiers that include a relation between a person and an adjective in a given context. For instance, for nheʿeng
= "spoken", if xe nheʿeng
= "I have speech", then it is possible that anheʿeng
= "I speak" (or, more literally, "I manifest my speech", or even more literally: "manifestly spoken by me"). So in Old Tupi, adjectives can be adequated into verbal functions, but they're still the building block: the adjective quality is still the basic unit of sense, which is declined into a "manifesting", "by me" form.
In fact, this is a very important feature of Old Tupi, because it makes the language non-redundant. There is only one word for "speech", "talk", "conversation", "language" etc., and it can only be possessed by one person at a time: which means I never "speak to you", and I never "say something" in Tupi: but rather "I say". What I said, and to whom I said it will depend on more adjectives and modifiers, but the farthest "spoken" can be bended into meaning is "I speak".
Actually, an adjective in Old Tupi can be transitive: it can by itself posess something else. That is the case of kuab
= "conscient", "aware". When I say xe aĭkuab i
= "I know it", I'm actually concatenating three adjectives. "It" is posessed by "knowledge" ( = "knowledge of it") which is posessed by me: "'knowledge of it' of mine" = "I know it". Headache much? Think of the possibilities.