Nalnuàntir

Orthography
[...]

Morphology
~slightly inflectional, mostly isolating

Personal Pronouns
Personal pronouns are inflected depending on their role in a sentence.

Pragmatics
How sophisticated someone's Nalnuàntir is, is largely related to how well they handle "Lingering" and how skillfully they manipulate the stack.

Lingering
Lingering is the process of..

Using Stack Manipulators
use stack manipulators whenever you can. if a sentence can be said both with and without the use of stack manipulators, they version *with* stack manipulators is always preferred. avoiding them comparable to the way young children speak English.

construction site
Marking for Θ-roles instead of case — using Θ-roles to define place structures Θ-roles (theta) in linguistics refers to things like agent, patient, focus etc. which are semantical funtions rather than grammatical ones. While Fith apprarently used a "focus" place for some verbs, it wasn't clear to me how to use it or what it was for. Today I realized how cool it would be to mark nouns for Θ-role instead of case, and here I'll show this for the interrogative pronouns. (A table of them is on sheet 1) Instead of defining a verb for "to give" as "x1 gives x2 to x3" or "Subject Direct-Object Indirect-Object", we use Θ-roles. I'd define them as follows: Agent (abbreviated as A): classical subject role, the doer of an action, one who brings something about Patient (abbreviated as P): the "victim" of an action, the thing directly affected by an action, experiencer role And lastly a Θ-role that combines several ones, (we can call it Focus), which includes focus, beneficiary, theme, goal. Let's abbreviate it as F. Doing this, we can make do with just three different interrogative pronouns. A: na, P: nara, F: nawe. "Using the verb pāra (A goes to F), we can ask the question: ""nawe pārà u?"" — ""where are you going?"" " We can use the *same* question word "nawe" in a sentence that uses "to give" (A gives P to F): (let's say just for this example that this verb is "hilo") "nawe hilo u?" — "Who do you give it to?" And probably in many more words like "to think" if it's A thinks about F, we can say "la nawe pensi e" — "what am i thinking."

Making case/Θ-role marking optional/situational It seems to me that if case/Θ-role marking is mandatory in every context, it introduces a lot of redundancy, but more importantly, it makes the swapping mechanisms much less useful. the SWAP was useful in Fith for allowing the object to be uttered before the subject and then switching them and then saying the verb. (kinda like SE, imo). if case is mandatory, this swapping is totally useless. i think it stands to reason to make case marking optional/situational. that is, force it in relative clauses as well as in sentences with elided arguments if the elided arguments are not the trailing nouns right before the verb. what i mean is: noun noun noun verb/3 ~ using case is not necessary and probably should be forbidden, dunno for sure noun noun [] verb/3 ~ again, case is not necessary because the two nouns can be unambigously assigned a sumti place. (namely, x1 and x2) noun [] noun verb/3 ~ here, a middle argument was omiited. the second noun can't be assigned the right sumti place anymore, therefore, make case marking on that noun mandatory. as seen in the above article, this allows for short and efficient sentences when all but one argument are elided e.g. "nawe parà u". nawe can be unambiguously assigned to the focus place of to go here. if it hadn't been marked however, it would fall into the agent place. btw, i would like to mention that even though this *might* seem weird, it's not that unnatural. korean has a similar curiosity, where marking certain verbs for person is ungrammatical when the subject is explicitly mentioned. oO Weird, but kinda awesome.

Relative Clauses (sel's ideas) "the man who sees the dog" using some of the old vocab mixed with some new words, we get the parts: the man = hong ke to see = vil (new) the dog = bagèr ke (new) The basic idea is that relative clauses work in a way similar to adjectives. They are atrributes that come after the noun they specify. However, for sentences to be used as "adjectives", we must prevent the following: "hong ke bagèr ke vil e" - this would make a complete sentence already, and wouldn't specify the man at all. Firstly, I think it might be a good idea to say that the article ke closes a noun phrase, so that no further attributes can be added (except maybe with some weird cmavo). So the syntax instead is: NOUN - ADJ - ART / NOUN - relative clause - ART Okay, next, before continuing, I think this syntax makes tanru very easy. (if articles are mandatory) There is an obvious difference between "hong ke bagèr ke" (the man the dog), and "hong bagèr ke" (the dog type-of man). Anyway, now to get "bagèr ke vil e" to be an attribute of hong, we must seperate it from hong, or else hong will become part of it as above. Let's say this seperator is "ral". We then get: hong ral ___________ e ke where the __ part is the attributive phrase/relative clause so for example: "hong ral bagèr ke vil è ke" - the man who sees the dog. once the sentence gets closed by "è", it goes back to the last occurence of "ral" and makes the whole thing into an attribute. Now, the above sentence is clearly ambiguous, how do we know whether it means "the man who sees the dog" or "the man who is seen by the dog", since the relative clause only has one argument in it. There are several ways to handle this, 3 come to mind. Method 1: Always assume that the first argument fills the first argument place, while the empty argument place is understood be filled by the head of the relative clause. This would mean that the above sentence unambiguously means "The man who is seen by the dog" Method 2: Use a ke'a-like word. This is a cheap solution and seems like plagiarizing lojban. Method 3 is what i came up with and I really like it. (i hope you do too) Method 3: Introduce two different cases. One for the direct object, and one for everything else. This makes the handling of relative clauses extremely nice: let the direct object form of "bagèr" be "bagrà", then we get the following two possible relative clauses: (1) "hong ral bagrà ke vil è ke" - the man who sees the dog (2) "hong ral bagèr ke vil è ke" - the man who is seen by the dog Here's how it works: Every relative clause must have an empty slot. The empty slot is then understood to be the same as the head of the relative clause. This is similar to NOI and ke'a. Next, if a relative clause contains a direct object, then the head of the clause must be the subject/agent, whereas if the relative clause contains no direct object, the head of the relative clause must be the direct object. I also played around with some sentences and found that it sounds better when the article "ke" becomes "ge" between vowels. Also, if the direct object ends in "a" like "bagrà", then "ke" sounds nicer as "ga". This makes it sound more naturalistic. Applying these sound changes we get: "hong ral bagrà ga vil è ge" "hong ral bagèr ke vil è ge"

Here's one more thing we get for free if we decide to do relative clauses this way: For verbs with a valence of 3 (e.g. dunda), we get a third type of relative clause for free by only having two grammatical cases and if we use a postposition for indirect objects (e.g. "to") Using the words "book", "man", "I" and "give" we this third type of relative clauses: "man ral I book give è ge" ~ "the man I give the book to", because both the direct object and the subject are in the relative clause, the head can only be the indirect object! Seeing useful these two cases are (and that's not too much I think, it's just TWO cases and only one of them gets an ending -a), I think me might keep them. it also makes it easier to understand the arguments roles in the sentence when things get complicated *and* it allows us to free up the word order.

Should articles be optional? if articles are optional, we can do things like the following, but we'd lose lojban-type tanru: hong bagèr ke praj ma pâra è mile ke man dog the tree a go VAU beautiful the "man - the dog goes to the tree. beautiful the" - the dog goes to the tree. the beautiful man