From The Toaq Wiki

is a proposal that establishes a new variant of ꝡä with slightly different rules regarding adverbial placement. This solves a long-standing issue with adverbial use.

The problem

Given an adverbial in a complex sentence like Ruaq nháo ꝡä ruqshua râo níchaq, should we treat it as belonging to the subclause or the root clause?

Ruaq nháo, ꝡä ruqshua râo níchaq
‘They stated that it rained today.’ (Behavior Ⓐ)

Ruaq nháo, ꝡä ruqshua, râo níchaq
They stated today that it rained.’ (Behavior Ⓑ)

Behavior Ⓑ is useful in that one may always add details to the outer sentence in afterthought – details such as when somebody said something, as in the example – whereas behavior Ⓐ is more consistent, if we think of it as wrapping the simpler sentence Ruqshua râo níchaq.

Hoemaı has also stated[1] that “[a] core syntactic property of Toaq is that you can take any clause and embed it”, also explicitly dispreferring behavior Ⓑ for being “gardenpathy and [in need of being] rephrased”.

The solution: a new kind of ꝡä

Toaq has a tendency to branch heavily to the right, so why not allow the rightmost branch special rights? Enter . Under the proposal, ꝡä would not capture clause-final adverbials, whereas would, but can only appear as the final element in a sentence (save for the speech act particle).

In other words, ꝡä makes "small" clauses that can't have trailing adverbials, whereas opens a "big" clause that can never be closed.

‘that’ ‘if’ ‘how much’
Big clause Ⓐ mö (ꝡa) mö ma mö tıo
Small clause Ⓑ ꝡä tïo

may be used for object incorporation – in which case it turns into – so long as it still appears as the final constituent in a sentence. For instance:

He mıaqgaı jí báq moaq, mô dıabete jí pêaboı báq nıaq saq.
I appreciate those who remember that I’ve had diabetes for the last three years.

(ꝡâ would result in something that parses as I’ve appreciated those who remember that I have diabetes for three years.)

Both particles still allow for fronted adverbials or pre-subject adverbials, as in ꝡä/mö râo níchaq nä ruqshua or ꝡä/mö tao râo níchaq jí ní.

In this model, acts like indirect speech: any clause of any type, so long as it’s placed as the final phrase in the entire sentence, may be wrapped in a and should behave as expected.

References

  1. https://discord.com/channels/311223912044167168/646607726817968138/1088465798047207496

    A core syntactic property of Toaq is that you can take any clause and embed it, as is, inside ꝡä. I would be sad to give up this property.
    If A says Ruqshua râo níchaq, and B can no longer say Dua jí, ꝡä ruqshua râo níchaq but has to reshuffle the clause, I wouldn't like that.
    […]
    Also my intuition is that trying to ban clause-final AdjunctPs in subordinate clauses is trying to fix the wrong thing by basically saying that V [CP long embed] AdjunctP is more often useful than V [CP AdjunctP], while I would think that the former is gardenpathy and should be rephrased.