[Tlhingan-hol] Interactions between verb suffixes

[De'vID]

De'vID:
>> There isn't a one-to-one mapping between Klingon and English
>> sentences. This is a consequence of the fact that Klingon has aspect
>> but not tense, and English is the other way around.

QeS:
> I admit I'm somewhat lost as to why you raise the issue of aspect here. My
> problem with the interpretation would be the same if we were talking about
> Type 6 suffixes rather than Type 7s.

Because we're talking about type 6 suffixes. If we were talking about
type 7 suffixes, we'd apply the criteria for evaluating the meaning of
those suffixes. You seem to be saying that there is one set of rules
we can use to evaluate all suffixes (like a mathematical function).
What I'm saying is that each type of suffix actually behaves
differently, and some interact while others don't. (For all I know,
each suffix within a type can behave differently, but I don't think
so.)

For example: {luyu'nISbej} "I am certain that they need to interrogate him"

{-bej} "certainly, undoubtedly"; "these suffixes show how sure the
speaker is about what is being said"

There is no confusion about whether the Klingon sentence means "they
are certain they need to interrogate him" or "they need to be certain
to interrogate him". It means neither of those things. {-nIS} applies
to the subject of the verb. {-bej} applies to the speaker (writer) of
the sentence.

De'vID:
>> I agree that "they needed to interrogate him" is one possible
>> translation of that sentence, but for a different reason than your
>> explanation.
>>
>> The suffix {-pu'} means exactly that the action of the interrogation
>> is a completed action.

QeS:
> No, it could well mean that the action of *needing to interrogate* is a
> completed action. That's my point, and doesn't have anything to do with what
> the aspect is.

I understand your point. I just don't think that there's any reason to
believe that the volition or predisposition of someone to perform an
action is itself an action, or that the meaning of other suffixes
somehow become attached to the latter (the "action" of needing to
interrogate) rather than the former (the action of interrogation). The
example {De' vItlhapnISpu'} makes perfect sense without needing to
postulate that the two suffixes somehow interact (and I think doesn't
make sense otherwise).

Some suffixes affect "the action" itself: {-choH}, {-qa'}, {-moH},
{-laH}, {-lu'}. {-moH} is a bit weird because the subject of the verb
with this suffix is not the subject of the verb without it. {-lu'} is
even weirder because it changes the roles of the subject and object.
Some other suffixes change what/who the action is being done to:
{-'egh}, {-chuq}. Some express something about the speaker or the
listener: {-chu'}, {-bej'}, {-law'}, {-ba'}, {-neS}.

There is some ambiguity as to what meaning is applied to what
primarily when {-moH} and {-lu'} are involved with another suffix.
Some other pairs can also cause ambiguity in combination, like {-choH}
and {-laH}. (Does {QongchoHlaH} mean "begin to be able to sleep" or
"able to begin to sleep"?)

As the type 2 and type 6 suffixes are described in TKD, {luyu'nISpu'}
cannot mean that the need to interrogate is a completed action. Type 2
suffixes don't change the action described by the verb, they merely
describe the amount of volition the subject has in carrying it out.

QeS:
> Again, to cast the example with a Type 6 suffix rather than a
> Type 7, {luyu'nISbej} can mean that the action of *needing to interrogate*
> is certain. It doesn't necessarily imply that the action of interrogation
> per se is certain, and that's what I'm trying to argue for here.

Indeed, it means that the *speaker* is certain that "they need to
interrogate him". It cannot mean anything else (going by its
definition in TKD).

De'vID:
>> {luyu'nISpu'} indicates that: (1) they interrogate him (neutral as to
>> time: this could mean they interrogated him, they are interrogating
>> him, or they will interrogate him); (2) they have a need to do so; (3)
>> the action of interrogating him is completed (was completed, is being
>> completed, will be completed).

QeS:
> And what is the status of completion of the need to do so? How does {-pu'}
> affect, or not affect, {-nIS}? That's the part that, with respect, I think
> you're missing from my question.

What does it mean to "complete" a volition? The amount of volition
someone has in carrying out an action is not an action itself. To me,
asking whether {-pu'} affects the completion of {-nIS} is like asking
whether it affects the completion of {-bej} or {-neS} or {-chugh}. I
guess if you want a "yes" or "no" answer then I would answer "no".

De'vID:
>> I don't agree that the sentence indicates that they have completed the
>> needing to interrogate him, because if that were true, the meaning of
>> the sentence would be that they don't need to interrogate him.

QeS:
> Yes, but there are easily-imaginable contexts in which such a meaning would
> be called for.

I'm not saying that the sentence would be ambiguous in that case. I'm
saying that, under your hypothesis (that a volition suffix changes the
action of the sentence into the volition itself), {luyu'nISpu'} would
mean "they no longer need to interrogate him" and cannot mean "they
needed to interrogate him". Similarly, {De' vItlhapnISpu'} would have
to mean something like "I no longer need the information".

De'vID:
>> It also doesn't follow from the grammar given in TKD: {-nIS} indicates
>> the subject needs to carry out the action ({yu'}), and {-pu'} indicates
>> perfective aspect (that an action is completed).

QeS:
> That begs the question. How narrowly do we construe "action" here? Does it
> refer only to the verb root? Or does it refer to the verb plus some
> suffixes? That was and continues to be my question.

It applies to the root verb plus those suffixes whose descriptions say
they affect the action. And I'm not begging the question: the
description of type 2 suffixes say that they express the amount of
volition the subject has. It says nothing about changing the action of
the verb (unlike, say type 3 suffixes).

I agree that there is ambiguity when certain pairs of suffixes are
used together, but I don't think there is any ambiguity when {-nIS}
(volition) and {-pu'} (aspect) are used together. I think you are
reading something into the behaviour of type 2 suffixes which isn't
actually stated anywhere in TKD.

QeS:
> The given meaning of the
> sentence {HeghqangmoHlu'pu'} "it causes him to be willing to die" shows that
> for the purposes of {-moH}, the action is {Heghqang}, not {Hegh}.

{-moH} "Adding this suffix to a verb indicates that the subject is
causing a change of condition or causing a new condition to come into
existence."

Like I wrote, {-moH} is weird, and {-lu'} is weirder. Together,
they're even worse. But I have no problem with {-moH} indicating a
change in volition (from non-willing to willing) here. That doesn't
mean that the "action" here is {Heghqang}. The action of the verb
isn't mentioned at all in the definition of {-moH}.

Compare the difference between the definitions of {-moH} and {-choH}:

{-choH} "change in state, change in direction"; "Suffixes of this type
indicate that the action described by the verb involves a change of
some kind from the state of affairs that existed before the action
took place."

{-moH} changes the role of the subject, whereas {-choH} changes the action.

QeS:
> Similarly for {SuvqangmoHbej} and {SuvqangmoHchu'} from the paq'batlh.

So this is an example which shows that {-qang} (type 2) and {-moH}
(type 4) together can result in ambiguity.

Going strictly by what's in TKD, {-qang} should express willingness on
the part of the subject. We saw with {-lu'} that this isn't
necessarily the case (because it changes the role of the subject).
Here, it seems {-moH} with {-qang} can also result in ambiguity.

Here is the relevant canon (any other examples?):
{jIbwIj vISay'nISmoH} "I need to make-clean my hair" (and not "I make
my hair need to be clean")
{[qeylIS] SuvqangmoHchu' molor} "[Molor] fueled [Kahless]' will to
fight" (and not "Molor is willing to make Kahless fight")

One of three things must be true:
1) a type 2 suffix with {-moH} results in ambiguity
2) {-nIS} and {-qang} behave differently with respect to {-moH}
3) the paq'batlh uses weird grammar

Absent other evidence, it seems (1) should be assumed to be true.

(Again, we know almost nothing about stress in Klingon. Perhaps we can
distinguish "I need to make-clean my hair" from "I make my hair need
to be clean" by stressing either {-nIS} or {-moH}. That's how one
would disambiguate something like this in a natural language.)

QeS:
> By "scope", all I'm talking about is the limits (or absence of limits) of
> the ability of one suffix to affect meaning of either the verb root, other
> suffixes, or some combination of the two.

No, I mean, I understand what scope is. I don't understand why you
think {-pu'} applies to "everything preceding it", when the
description given in TKD specifies what it applies to: the action of
the verb (which some suffixes simply don't have anything to do with).
I agree that the description in TKD is underspecified and leads to
some ambiguity, but it's not so underspecified that, for example, that
you can say using {-bej} together with {-pu'} allows to you express
that the "certainty" is completed.

QeS:
> Taking away {-lu'}, then, what do you make of {HeghqangmoHpu'}? We know for
> a fact that in such constructions, it's not necessarily the surface subject
> - the agent of causation - that exhibits the willingness, even though that's
> what your reading of TKD would dictate. We have {SuvqangmoHbej} and
> {SuvqangmoHchu'} from the paq'batlh (paq'raD 7.18, 20) where the meaning has
> the surface *object* who's exhibiting the willingness - this indicates that
> {Suvqang} is acting more derivationally than inflectionally.

If I saw {HeghqangmoHpu'} without any context, I'd be inclined to
interpret it as "he/she was willing to cause him/her to die".

The paq'batlh examples open up the possibility that this means "he/she
caused him/her to be willing to die".

But I'd consider any interpretation involving "completion of the
willingness to die" to be out of bounds.

De'vID:
>> {loD HeghqangmoHlu'pu'} contains the idea of {Heghpu' loD}, as well as,
>> say {Heghqang loD}. I think you are confusing tense and aspect: just
>> because an action is a completed one, it doesn't mean it has actually
>> happened or already happened. {loD HeghqangmoHlu'pu'} doesn't imply
>> that the man has died, but it does "imply" {Heghpu' loD}.

QeS:
> No, it only implies {Heghqangpu' loD}. A willingness to engage in an act
> does not imply that the act will take place, regardless of what the aspect
> is.

Indeed, I'm not claiming that the act has taken or will take place.

I believe that you are mistaken in believing that {-qang} somehow
falls into the "scope" of {-pu'}. They act on different things about
the verb. {Heghqangpu' loD} means exactly that the man is willing that
{Heghpu' loD} happens. His willingness does not mean that the act will
take place, but it does necessarily mean that the act to which he is
willing is a completed one.

De'vID:
>> Also, why do you think "it made him/her willing to die" means that the
>> person described is still alive?

QeS:
> I don't remember saying anything of the kind. All I'm saying is that the
> person described is not necessarily dead and will not necessarily become so.

Agreed. And that's not what {Heghpu'} means.

QeS:
>> For that reason, I also don't agree that {luyu'nISpu'} implies {luyu'pu'}.

De'vID:
>> Again, I just don't understand what you mean. {luyu'nISpu'} expresses
>> a need for the subject ("they") to carry out the action described by
>> {luyu'pu'}. I think that, in this sense, it does imply it.

QeS:
> This also begs the question. The need does not imply the action: that is to
> say, I don't think it's a tautology to say {luyu'nISmo' luyu'} "they
> interrogate him because they need to interrogate him".

Agreed. But I don't understand why you think that would be a rebuttal
to what I wrote.

I'm not saying that {luyu'nISpu'} means that {luyu'pu'} has actually
happened. I am saying that {luyu'pu'} (as opposed to merely {luyu'},
or {luyu'taH}) is the action that "they" (the subject) need to happen.

Compare the following:
{qa'ma' luHoHpa' luyu'nIS} "they need to interrogate a prisoner before
killing him" (as a general rule)
{qa'ma' luHoHpa' luyu'nISpu'} "they needed to interrogate the prisoner
before killing him" (referring to a completed event, which hasn't
necessarily happened)

De'vID:
>> TKD says "neutral as to time", and I believe it actually means
>> "neutral as to time" (i.e., it is talking very clearly about tense and
>> not aspect when it does so).

QeS:
> Why single it out in this instance, then?

Three specific instances.

QeS:
> When *all* verbs are neutral as to
> time, a fact which is clearly discussed in TKD already (p. 40), surely it's
> superfluous to make mention of it so specifically here,

Because, while it is generally true that a Klingon sentence can be
translated in past, present, or future tense, in the three specific
instances noted otherwise, they can't. It's not superfluous in these
three instances, because they illustrate exceptions to the general
rule.

{qaHoH bIjatlhHa'chugh} "If you say the wrong thing, I will kill you."

As TKD explains, while generally {qaHoH} can mean "I killed you" or "I
am killing you" or "I will kill you", here, it can only mean "I will
kill you".

{yaS qIppu' 'e' vIlegh} "I saw him/her hit the officers."

Here, {yaS qIppu'} describes an event that has actually happened, and
which I actually saw. Thus, {vIlegh} here is "I saw", not "I am
seeing" or "I will see". (I suppose that you could
be watching a video of someone hitting the officer, and in that
context, {vIlegh} would be "I see". But that's not the scenario TKD is
describing here.)

{qama'pu' vIjonta' vIneH} "I wanted to capture prisoners."

Here, the speaker, Kruge, wanted to capture prisoners. He's not
expressing a general desire to capture prisoners, but for the act to
be a completed one ({vIjonta'} "I have deliberately captured them"),
i.e., specifically before the enemy ship was destroyed. Note that
{-ta'} here is attached to a verb describing something which didn't
happen. (This is what I'm trying to explain above: that {Heghpu' loD}
is not the same as saying that the man is dead or will die.)

QeS:
> especially in
> phrasing such as this: "...the aspect marker (in this case, {-ta'}
> "accomplished") goes with the first verb only; the second verb, {vIneH} "I
> want it", is neutral as to time." (TKD p.67). Surely *both* verbs are
> neutral as to time, and it's only aspect that is characteristic of the first
> sentence?

The next sentence makes clear why:
"The past tense of the translation ("I wanted...") comes from the
aspect marker on the first verb."

The explanation is saying that the reason the English translation of
this sentence is "I wanted to capture prisoners" and not "I want..."
or "I will want..." is due to the aspect marker on the first verb
(since it refers to an event which should have happened - but didn't -
in the past).

Aspect with {'e'} (and {neH}) is weird. Forbidding aspect on the
second sentence reduces the expressiveness of {'e'}, but also limits
the possible ambiguity. The interpretation of the second sentence is
forced to depend somehow on the meaning of the first sentence. In a
way, that makes the verb of the second sentence act like a verbal
suffix: {vIjonta' vIneH} is similar to {vIjonnISta'} in that {neH}
expresses something like a kind of volition, and relates to the aspect
suffix {-ta'} on the verb {jon} in the same way that {-nIS} would. The
second verb isn't allowed to take an aspect suffix, because it's
already governed by the aspect marker on the first verb.

-- 
De'vID