Em segunda-feira, 26 de setembro de 2016 18:58:35 UTC+3, Jacob Errington escreveu:
> On Sep 26, 2016, at 11:31 AM, selpahi <sel...@gmx.de> wrote:
>
> On 26.09.2016 17:22, Jacob Errington wrote:
>> In the official grammar, it can be observed that {cu} is a terminator
>> by noticing that it is elided only when one or more terms precede the
>> bridi-tail: contrast {ko'a [cu] broda [vau]} with {broda [vau]}. The
>> proposed connective reform does more than reform connectives. It
>> alters the operation of {cu}, so that rather act as a bridi-head
>> elidible terminator, it acts as a bridi-tail elidible *initiator*.
>
> {cu} is not really a terminator. The fact that you can't have {cu} without a bridi-tail is evidence of that. If {cu} were really a bridi-head terminator, {ko'a cu} should be a valid fragment. Note also that people think that things like {sei broda cu brode} should be legal. This shows that psychologically {cu} belongs to the tail, rather than the head.
{ko'a cu} is not a bridi-head, since a bridi-head can appear only before a bridi-tail. (I gave an informal characterization of bridi-heads and bridi-tails in my last message. Notice that my characterization rules out {ko'a cu} as a bridi-head, as it does not precede a bridi-tail.)
As for people wanting to say such things as {sei broda cu brode} as the basis of a psychological argument in favour of {cu} belonging to the tail, I argue that people think this way because the most common way of explaining {cu} to people is a lie: "{cu} is a cmavo that can be placed before the main selbri in order to force the insertion of all elidible terminators necessary to make the appearance of the main selbri grammatical". Sure, according to this explanation, {cu} belongs to the tail, but as I said, this explanation is a lie. According to this explanation, {cu broda} would be legal, since it forces the inclusion of all (zero) terminators necessary to make the appearance of the main selbri {broda} grammatical. However, {cu broda} is not allowed in the official grammar.
Therefore, as I understand it, your psychological argument boils down to this: we should change the operation of {cu} because of a convenient lie we've told to too many people. (I'm not at all claiming that I'm not guilty of perpetuating this lie too; I've explained {cu} this way countless times.)
As for people who want things like {sei broda cu brode} to be legal, according to my (rather straightforward) characterization of bridi as a bridi-tail plus an optional head, {sei broda cu brode} is ruled out, as {sei broda} is not a term. Hence, there is no bridi-head. ("If a bridi-tail is present, it may be preceded with a bridi-head, which begins after {zo'u}, ends with {cu}, and encloses *one or more* terms." Added emphasis.)
Perhaps we can just teach people this characterization instead of fundamentally changing the operation of {cu}. In fact, this is already how I explain {cu} to people now. Rather than change the grammar so that we can have been right all along, let's just admit that our explanation was wrong and move on.
>
> Also, {cu} and {vau} don't really form a pair. {vau} appears even in phrases that contain zero {cu}, and the number of {vau} in sentences containing bridi-tails exceeds the number of {cu}. If they were a real pair, they should always appear equally often.
>
There's no argument here. I agree with you entirely that {cu} and {vau} do not form a pair. This contradicts nothing of what I said though; I never said that they formed a pair. I merely said that there was a kind of elegant symmetry between bridi-heads and bridi-tails, since each has a terminator.
>> Reforming the overly complex connective system is an excellent goal,
>> but I am against the means to that end employed in this proposal,
>> namely the alteration of {cu} to become an elidible initiator (which
>> would also make it the first of its kind, I think).
>
> I like to compare {cu} to FA. A FA is an optional marker for a particular sumti, while {cu} is an optional marker for the selbri. Both FA and CU mark particular slots in the bridi.
>
And I like to compare {cu} to {vau}! {vau} terminates a bridi-tail whereas {cu} terminates a bridi-head. Both terminate important parts of the bridi.
Of course, {cu} cannot appear without {vau}, since a bridi-head cannot be present without an accompanying tail. But then again, this is no different than how CU differs from FA in your proposal, since {fa mi} is not a bridi whereas {cu broda} is.
.i mi'e la tsani mu'o
I have to agree with la tsani's arguments. Besides, feedback from users showed to me that variations of mad proposals lead to no gain.