Mostly more old stuff.
“do” is long deprecated
do used to be a word before the current binding system took effect. It is no longer needed and is free to be assigned to something useful.
The “anonymous” variable is raı.
“ju” takes a full sentence
ju was previously only able to take a subordinate clause and could only be terminated with na. Now it takes a full sentence and is terminated by that sentence’s illocution marker:
Shủe nỉe déo, ju tẻa hó séoqtue da, búe da.
“The child, who is scared of the weather, stays inside the home.”
Gúobe, ju mả kảqgāı súq hó moq, bı jỏaı hó fú da.
“The cow -do you see it?- it’s looking for [its] children.”
No more na na na na Batman.
What about old “la”?
Old la was probably one of the most unstable words, being the testing ground for all sorts of experimental grammar. Its original purpose was to make meta-statements, that is, statements about the containing text. Since the spelling la is now taken by a different word, a new spelling is needed for “meta-la” (at least if it is to remain in the language). I don’t have a form in mind, though I would prefer if it weren’t a Cy.
A more interesting question is how to best define this word. There are mainly two possible design-choices:
- Make it identical to ju in that it takes a fully satiated sentence. The difference would be that the sentence has to be about the containing text. The meta-la word would mostly act as a signal to the listener that the speaker intends to make a meta-statement.
- Make it take a full sentence, but require a hoa to appear in it. This hoa would refer to the containing text (sentence). This would guarantee that the meta-phrase is always actually meta.
With option 1, it is unclear how to deal with something like this (where X stands for the meta-la replacement):
Je, X mả chủqkuaı súq moq, fả jí jéarıaq da.
“So, are you hungry?, I’m going to the market.”
This is clearly not a meta-statement, but it would be a valid way to use the grammatically equivalent ju. If this kind of use of meta-la is acceptable, then there is no clear difference between it and ju, and the two should be merged. With option 2, the difference would be a lot clearer, since it would be impossible to make a statement that is not about the text due to the mandatory presence of hoa:
Je, X mủa hóa da, kảı jí ní da.
“So, ~this is as example~ I am writing this.”
I dunno.
Terminators for CPs and “lu”
CPs, that is, 
clauses and 
clauses, are terminated by cy:
Shủe tỉ tu düı nỉo cy búı da.
“Everyone who was too young stayed outside.”
LU, which was previously a pseudo-class of related but distinct function words is now an actual class whose sole member is lu. Its terminator is ky:
Lủ mảı hóa hóa ky súq
“You are such that you love yourself.”
Lủ zảı hóa tâı hóa ky jí ru súq
“Me and you (each) are X such that X hopes that X succeeds.”
LU benefits from having its own terminator, because usually lu is only used in complicated cases, and the terminator lets you close off the structure no matter the amount of / nesting within.
Tenses are no longer weird
This paragraph isn’t a change as much as it’s just a clarification of another change: One consequence (and purpose) of the introduction of predicate classes was to make the interaction between tenses and adverbials less awkward. The syntax described in that one Tense article was a more or less necessary workaround in the previous system, because Tense and rao wouldn’t combine in the right way due to the old adverb behavior. This is no longer the case, since Tense now correctly scopes over the rào:
| Old (deprecated) | New |
| Pù kíachāq bı chủq jí sa jảq nỏqgī da. “On Monday, I ate something very tasty.” | Pủ chủq rào kíachaq jí sa jảq nỏqgı da. “On Monday, I ate something very tasty.” |
Tense predicates no longer have a 2nd argument place.
An innocent question
As explained in the Auto-hoa Suite, and as anyone involved in the language is aware, it remains an open question which of the many possible auto-hoa proposals should be implemented. As this has gone on for so long, one specific question needs asking: If none of the proposals are clearly better than all the others (i.e., they all have flaws), could it be that auto-hoa on the whole is just not sufficiently much better than no auto-hoa to be worth it?
Toaq’s pronominal system
… will be explained in a separate post for easier reference.
The Toaq Blog 
My thoughts:
— “ju” takes a full sentences —
Nice, I’ve been using it that way ever since I noted the other day that I thought it would be a good idea. The only downside I’ve encountered so far is the “double da” you can need if it comes at the end of a sentence. I have simply been using “je” instead in these cases.
— What about old “la”? —
The grammatical equivalence of lu and old-la (I’ll call it “luy” here) doesn’t bother me much. One could simply say that luy-statements that aren’t really about the clause are just uncooperative speech.
But, that’s from a purely ivory-tower prescriptivist perspective — maybe you mean that if non-meta luy clauses are grammatically possible then people will slide into using them no matter what we say. That’s a topic I was wondering about for an unrelated Toaq thing the other day. I forget what it was, though.
I’ve gotten quite far while basically not using luy at all. Maybe kio-ki parentheticals could absorb this use-case?
— Terminators for CPs and “lu” —
Thank you for the clarification! I agree about the usefulness of lu having its own terminator.
— Tenses are no longer weird —
Nice! This way feels cleaner.
— An innocent question —
I have been using proposal #3 (lazy initial auto-hoa) all along, and my gut feeling is that it’s better than no auto-hoa. But that’s mostly coming from a writer’s perspective — I’m not a speaker yet, and I have yet to be an audience to anyone who uses this auto-hoa in a confusing way to create garden-path sentences, as Ilmen fears.
LikeLike
((do you mean “ju”, instead of “lu”? It’s those two which were in the same class, not “la” and “lu”))
What’s interesting is that every person seems to favor a different auto-hoa proposal. If everyone could agree on one as being the best, it would have a bit more weight, if that makes sense. In pure number terms, any auto-hoa is better than no auto-hoa, but that doesn’t say much. Maybe, instead of picking the best one out of 10, it would be easier to eliminate the worst over and over until only one remains (I could imagine that #3 and #7 would get pretty far, while #6 should just abstain from the competition).
LikeLiked by 1 person
((Oops, yes. Must’ve gotten my wires crossed.))
That does make sense, yeah. Repeatedly removing the worst options sounds like a good idea, at the least it would make the choice feel less overwhelming once it’s pruned down a bit.
LikeLiked by 1 person
Thanks for the update!
Does “li” still exist? Does “na” still exist?
I mostly like old-la for replacing markers like Lojban’s “zo’o” (la buoq — this sentence is non-serious). Maybe old-la can be replaced with ju and a pronoun for “this sentence”?
I really like option #2 in the auto-hoa suite (which seems to align with what’s been on toaq.org all along). I don’t totally relate to the desire to drop the resumptive pronoun. I suppose that a null resumptive pronoun is really common in natural languages, but I like Toaq tending toward having it be explicit — it just feels like it fits aesthetically, somehow. It’s also really really simple.
LikeLiked by 1 person
It’s not clear whether “lı” is worth keeping at this point. People tend to only use it to be able to omit “ja” (as in “jỏe súq lî sủaq”), but spending one syllable (and in forethought, too), to omit one syllable doesn’t immediately seem like enough of a reason to keep this word. On the other hand, using “lı” for additional redundancy could be helpful for communication in certain cases. It’s up for debate.
I think “na” has to exist for “tỉ jí búe na ru lẻo nủo jí” to continue to work. Elsewhere it might not be needed anymore, I’m not sure.
LikeLiked by 1 person
“li” could be kept around for making drop-in predicates of any arity. It could get the same treatment as “lu” just did — take “lỉ” as the base cases and make the toned versions by adding tones. I used it like this the other day:
“Lỉ dủa jí ja pỏ hóq na, tou jâq dảqbūaı hóq da”
Pardon the now-improper use of hóq!
LikeLiked by 1 person