Sometimes real-world terminology gets as mucked up as Lojban's. Here is a real case, which not only messes together logic and grammar but grammars of several different languages and tht two different time related systems of tenses and aspects (and, hence, of various event types as well). And, of course, different theories of tense as well. Both "perfect" and "perfective" get involved in all of these and Lojban, as usual, tries to deal with all of them, more or less. In terms of logical tense, an event occupies a point or area of time and reference to that event is given in terms of of the temporal relation between that event and the time of reference: prior, concurrent, or posterior. In abstract grammar, an event is placed at or with reference to a "present" and that present is placed with
reference to further moments involved in the conceptualization of that event as it is spoken of, giving a fourfold scheme, four times repeated, of points and vectors. In the logical system, neither "perfect" nor "perfective" have a place: whether an event is considered as a whole or as having parts does not enter in, nor does the matter of its present relevance. In the grammatical system, the difference between current relevance of a past event and simple pastness is covered by the difference between a minus vector on a present axis and a past axis Lojban uses only the logical system at this level. Logically, of course, the present relevance is not a tense feature (one might say), but, like other features of grammatical tense, a psychological projection. Thus,
it should be part of a psychological system like aspect, which involves the speaker's expectations, etc. as well as the temporal order. But here we get a terminological problem, since the "usual" terms for the aspects ("inchoative, initiative, continuative, terminative, completive, superfective") all end in -ive", we expect the same to apply to the converse of "inchoative" and thus get "perfective". But that term has already been taken over in yet a third system of terminology, essentially a counter to the whole aspect system, which takes events as extended (so applies only to states, activities, and processes, not achievements). Since this third system is not relevant here (whatever may be the case in the grammars of particular languages -- where the supposed prefective-imperfective distinction actually usually turns out to be several other things as well -- or instead), we chose to ignore it and use the term that fit our patterns
best (hardly the worst violation of usage in Lojban).
Now, as far as changing labels is concerned, I suppose, if this label actually confused someone, even after reading the explanation, the change should probably be made (and maybe the other aspect terms modified to coincide). But I doubt that is really necessary. As for using "retrospective" instead, this is a term used, with obvious problems, sometimes for minus vectors, sometimes for past axes in grammatical tense systems, so only occasionally relevant to {ba'o} and even then not correct since not aspect. The use of "perfective" for {co'i} is even worse, since it is used, as pointed out, for complete events, not the event of completion, and, again, is not aspectual.
From: Robert LeChevalier <lojbab@lojban.org>
To: bpfk-list@googlegroups.com; John E Clifford <kali9putra@yahoo.com>
Sent: Friday, August 30, 2013 8:38 AM
Subject: Re: [bpfk] Changing the definitions of {ba'o} and {co'i}
la gleki wrote:
> 1.
>
> ba'o is currently defined (according to CLL
> <
http://dag.github.io/cll/10/10/>) as
>
> ba'o ZAhO perfective
>
>
> If we are to use normal linguistic terminology then this is wrong.
>
> {ba'o} is perfect, not perfective - those are completely different things.
> According to Wikipedia <http://en.wikipedia.org/wiki/Perfect_tense> it's
> better to avoid using the term "perfect" and change to "retrospective".
>
> So I propose changing the definition of ba'o to
>
> ba'o ZAhO retrospective
>
>
>
> 2.
>
> As for "perfective" it looks like it's expressed
using {co'i}.
> Another independent proposal is that co'i should be defined as
>
> co'i ZAhO perfective/achievative
>
>
> "perfective" is used quite extensively when describing Chinese and
> Russian grammar so normalising terminology is a must pe'i.
>
> Neither proposal changes anything in Lojban itself, only in translation.
I don't necessarily have a problem with such a change (especially since
I've been an incompetent student of Russian for 20+ years now), but
would like pc's input. IIRC, the terminology came from his exposition
to me of tense logic's terminology used for Aristotelian events, and pc
was at the time a specialist in tense logic.
The "perfective" term, IIRC, was consistent with the "superfective" term
(za'o itself), for which I don't know any other
linguistic equivalent.
So the choice may be between using linguistic terminology or tense logic
terminology.
I've cc'd this message to pc to make sure that he sees it.
lojbab