In Kokanu's context system, things that don't have a dedicated lexicalization are described using existing words with detail sufficient to make the reader realize what is being referred to if they are already familiar with it. In many cases, such a description needs to be very long, and Kokanu users are encouraged to employ a sufficiently lengthy description at least once, and then refer back to it in abbreviated terms if the concept needs to recur.
Two issues arise in this system:
Kokanu makes no distinction between cases where modifiers and genitive phrases are used as descriptors to point the audience towards a certain concept, and when they are used as descriptors for a unique instance of something. This can lead to slow comprehension and confusion when the audience does not know whether they are supposed to be guessing at the meaning of a compound or just receiving a description. This problem can be fixed through the use of a compound-marking particle: a particle that, when used, specifically indicates that a noun phrase with descriptors is to be interpreted as a compound for a specific concept that the audience should be familiar with rather than a set of descriptions in a unique circumstance. This particle could couple to a previous phrase marking particle, like no, but a better form may be as an alternative to je, so that Kokanu would have both a descriptive genitive marker and a compound-forming genitive marker.
Example:
In a story translation, I translated camel as neje wiki je osa niku pice takilo. If I just wrote neje wiki je pice takilo, it could be pretty difficult to realize I'm talking about a camel and not a fast horse with some kind of cone shaped object on its back. A word that marks that this is supposed to be a compound for an animal with a specific notable quality could make the comprehension of what I'm trying to say a lot faster, and could even allow me to use a shorter compound without sacrificing read comprehension.
The strategy of fully describing a necessary concept once so that the audience locks on to what is being indicated, and then referring back to the concept in abbreviated terms, works well in simpler situations, but can become limiting in more complex ones. In communications of high complexity, like scientific or political articles, multiple forms of a concept may need to be referred to, and various highly specific concepts may be difficult to distinguish as they employ the same Kokanu descriptors. In these cases, a way to distinguish between a phrase that refers back to a previous description in abbreviated terms, and a phrase that indicates a seperate concept, becomes necessary. This could be accomplished with an anaphoric pronoun, which would be a demonstrative pronoun like na and ne, but instead of pointing to something near or far, it would point to an idea previously mentioned. This kind of distinction exists in natural languages, and is similar to some uses of the English definite article, distinguishing something that should already be known of in the conversation from a new idea.
The solutions to these two issues are similar in some ways, and it is possible that only one of the two may be necessary.
In Kokanu's context system, things that don't have a dedicated lexicalization are described using existing words with detail sufficient to make the reader realize what is being referred to if they are already familiar with it. In many cases, such a description needs to be very long, and Kokanu users are encouraged to employ a sufficiently lengthy description at least once, and then refer back to it in abbreviated terms if the concept needs to recur.
Two issues arise in this system:
no, but a better form may be as an alternative toje, so that Kokanu would have both a descriptive genitive marker and a compound-forming genitive marker.Example:
naandne, but instead of pointing to something near or far, it would point to an idea previously mentioned. This kind of distinction exists in natural languages, and is similar to some uses of the English definite article, distinguishing something that should already be known of in the conversation from a new idea.The solutions to these two issues are similar in some ways, and it is possible that only one of the two may be necessary.