Was: RE: cha'DIch KLBC rI' BG

[willm@cstone.net]

This seems like it has evolved beyond KLBC, so I feel okay commenting on it.

> > > >I think he should have left of the -pu'.
> >
> > > Now that I look at it again, I think you're
> > > right. Since all three sentences have it, probably none need it (or else
> >only the > first needs it, and the others can inherit its time context).
> >
> >Aaah, no, there is no time context implied by -pu'.  It's RELATIVE
> >COMPLETION.  The time context is provided by time words.
> 
> Relative completion implies a relative time context; if it is completed at 
> the time of the sentence, then it must have been performed sometime prior to 
> that.  You are correct that there is no absolute time context involved in 
> {-pu'}, but there very definitely *is* a relative one.

The main issue that doesn't seem to be getting much light here is that in a 
language that lacks tense, time stamps are significant. If you want to focus on 
aspect while discussing multiple sentences, be sure to provide a time stamp of 
some sort. It can be a {-DI'} marked dependent clause or a timestamp word like 
{wa'Hu'}, or a clock oriented timestamp or whatever, but do SOMETHING to set 
the anchor on the time and then choose aspect appropriate for that time anchor 
until you set down a new time stamp.

> What I was trying to say is that either no sentence needed the {-pu'}, or 
> else the fact that the first sentence already provided the information that 
> the action was already completed, the others could inherit that.  If there 
> were absolutely zero time context, then all three sentences would need 
> {-pu'}, as all three actions had been completed.

You are the first person I've ever heard suggest that such a mechanism exists, 
such that you can essentially use "present perfect" in one sentence in order to 
set a time stamp, and one would assume past tense for everything else. I've 
never seen any canon supporting this and I've never heard anyone or read anyone 
exibit this behavior. It doesn't feel natural for me. Instead, it feels like a 
new path to an old goal. Beginners want to use perfective to mean past tense. 
You are not a beginner, but you are exhibiting this small, familiar trait of 
inexperience with the language, though you are exhibiting it in a refreshingly 
new way.

If I saw three sentences and the first one had a verb marked {-pu'} and the 
next two didn't, I'd assume that there was an unstated time stamp that all 
three sentences shared, and the action of the first sentence was complete at 
that time, while the action of the verbs of the other two sentences happened at 
the time of that unstated time stamp.

Meanwhile, if you put {-pu'} on all three sentences, it would be less "wrong" 
than just somewhat inept. Why use one time stamp that fails to relate to all 
three sentences? Why not just give the REAL time context for the action of the 
verbs and just use the perfective to mark exceptions to that time stamp?
 
> At the time of utterance, the actions of all three statements were completed 

In other words, you are trying to use the perfective to mean the past tense. 
Klingons don't care about the time of utterance. They care about the time 
setting for the action of the sentences. Instead of a vague tense that only 
gives you the three state information of "before now", "now" or "after now", 
Klingon replaces tense with far more versatile and informative time stamps. It 
then allows verbs to efficiently explain their state of intent and completion 
with Type 7 verb suffixes, measuring each verb in time against the specific 
time context set by the time stamp.

When I say in English, "I ate spaghetti," you know the action is past tense. 
So? Did I mean "this morning", "yesterday", or "last week". You can't tell. 
This is vague. In Klingon, I'd say, {povam qagh vISop} or {wa'Hu' qagh vISop} 
or {Hogh ret qagh vISop.} If I wanted to be as vague as English, I might say 
{qen qagh vISop.}

The one thing you do NOT want to do in Klingon is try to use the perfective to 
mean past tense. You are making an assumption that an unstated timestamp 
implies the time of utterance and that is simply not the case in Klingon. An 
unstated timestamp is either the timestamp of the ongoing context, or it is the 
mark of an imprecise speaker. If there is no time context, give a time stamp of 
some sort. Just do it. Don't assume the time of utterance has any significance. 
That's a faulty assumption in Klingon, based on the English bias toward tense.

> (while some of the actions may have been done again later, the instances I 
> was referring to were completed).  So should they all have {-pu'}?  That was 
> my first instinct.  But upon hearing DloraH's advice, I reconsidered.

Good. Keep reconsidering.

> If what you are saying is that the relative time context implied by {-pu'} 
> does not extend to other statements (unlike absolute time contexts supplied 
> by time words), then my first instinct was correct.  On the other hand, we 
> already know that in at least one case such contexts are inherited; a 
> sentence for which {'e'} is the object inherits the aspect (and thus the 
> relative time context) of the sentence which {'e'} represents (TKD, p 66).
> 
> >choQaHpu' 'e' DanIDba'.  tlhoy jIDoy'ba'pu' 'ej pe'vIl jIjangpu'.

Realize that the relationship between {'e'} and Type 7 verb suffixes can be 
historically explained. During one of Okrand's famous retrofits (fixing a 
botched movie line with a subtitle), he was forced to put the perfective marker 
on the first verb of SAO when he really wanted it to go on the second verb. 
This happened while he was developing TKD, so he decided to make this an 
arbitrary rule. He explicitly explained this rule in TKD.

Later on, he has repeatedly broken this rule. Likely, he did so by mistake, 
since the rule is arbitrary and contrary to what one is naturally drawn to do 
while actually writing things using SAO. Possibly, he has instead decided that 
he doesn't like the rule and consciously breaks it. He evades explaining this, 
though if we think to ask point blank, maybe we'll get an answer someday. Maybe.

Now comes the interesting part: During my interview with Okrand all those years 
ago, for HolQeD, he looked me straight in the eye and made very clear emphasis 
by gesture and speech that I should pay a lot of attention to him while he told 
me a golden truth: "Learn proper Klingon by studying usage, not rules." He 
strongly and repeatedly indicated that we need to observe his usage of the 
language and consider that a higher priority than the rules.

Given that, he has broken the "Don't use Type 7 on the second verb of SAO" rule 
enough times that I personally think it's okay to consider it to be an obsolete 
rule. I break it from time to time and don't feel bad about it. I always 
consider following it and often discover that it doesn't make any difference, 
so if it doesn't, I'll favor following the rule, but if it really seems more 
meaningful to break the rule, I break it.

I would definitely not lean heavily on a potentially obsolete rule as a 
precedent used to justify using perfective as past tense, which is what you are 
doing, whether you know it or not. Your attachment to the moment of utterance 
is the way you are mistakenly associating the perfective with the past tense.

> As you can see from my original sentence, in the SAO which I used, {-pu'} 
> does not appear on the second verb of the SAO, although it appears on all 
> three of the other verbs.
> 
> So I guess my question now is, is the aspect (and thus the relative time 
> context) inherited by any sentence other than the second sentence of an SAO 
> contruction?

Change your model of time in Klingon. It's more like American Sign Language. 
You give some sort of time stamp. That is now your time anchor. All sentences 
you express from that point forward relate to that time context unless you 
express a new time stamp. The current time context is always the most recently 
expressed time stamp. You don't use the perfective to establish a time stamp. 
You use it only to mark an exception to the time context; a brief, one-verb 
reference to an action that was complete at the time of the current time 
context set by the most recent time stamp.

No other verb "inherits" the perfective. All verbs "inherit" the time context 
stated in the most recent time stamp.

Does this help?

Will