tlhIngan-Hol Archive: Wed Mar 10 15:30:26 1999

Back to archive top level

To this year's listing

[Date Prev][Date Next][Thread Prev][Thread Next]

Re: Deep structure

At 02:37 PM 3/10/99 -0800, charghwI' wrote:
>I just need to comment that I'm impressed with ter'eS. This 
>description is more internally consistent and better thought 
>through than my own. This is what I felt, but it was not so 
>clearly thought or expressed.
>
>I respect both the analysis and the tone. I hope to take 
>inspiration from both.
>

choquvneS

>> 
>> Basically, I see /-moH/, besides giving the verb a causative meaning,
>> as opening up a new noun role to associate with the verb.  I also see 
>> noun roles as hierarchical, with subject most common, followed by object.
>> An intransitive verb with no suffix has only a subject role. When you 
>> add /-moH/ to it, the object role is now available. 
>> 
>> /bel yaS/ The officer is pleased.
>> /yaS belmoH Duj/ The ship pleases the officer.  
>> 
>> It's true that a particular noun may move from the subject to the
>> object role, as /yaS/ did, but the role of subject in general stays
>> the same: the one which performs the action of the verb.
>
>While I see this slightly differently, I see that this still 
>works by my perspective or yours. I'd say the subject is the one 
>doing the action before {-moH} is added and the one causing the 
>action after {-moH} is added, and this implies a change in role 
>for the nouns, but grammatical analysis is an example of 
>modeling. One model is not better than another model until one 
>model fails to behave like that which is it modeling while the 
>other one succeeds. I don't see either of our models as failing.
>

I agree.  I think I took a more "structuralist" approach, while you are
taking a more "semantic" approach.  That is, I focused on the presence
or absence of additional noun "slots" related to the verb when /-moH/ is
added, and you have focused on the changes in meaning that those "slots"
undergo.

[...]
>> We were pretty stumped for a while, because we didn't seem to
>> have any tools to distinguish the direct object from the ergative
>> object. Finally, MO gave us the suffix /-vaD/ to mark this, 
>> in the /ghaHvaD quHDaj qawmoH/ example. I don't think the two 
>> uses of /-vaD/ are that closely related, so it wasn't obvious until MO 
>> pointed it out. Maybe in /no' Hol/ there were two different suffixes, 
>> whose sounds converged over time.
>
>As I see it, Okrand first explained {-vaD} as benefactor in the 
>original grammar section of TKD. He expanded on that to use it 
>for "indirect object" in the Addendum grammar section. I see 
>this use with {-moH} as an example of {-vaD} marking the 
>indirect object, and I see the roles of direct and indirect 
>object as being somewhat arbitrary depending upon the finer 
>points of meaning of a given verb. Meanwhile, adding {-moH} to a 
>verb already with a direct object lends one to consider the 
>"ergative" object to be the indirect object, hence appropriate 
>for {-vaD}.
>

I'd have no trouble equating the "ergative object" to an indirect
object, except that I thought there was some confusion in
a previous post due to focusing too closely on the benefactive
meaning of /-vaD/.  MO could have chosen some other means to mark
this object, but position in the sentence was already taken by
the subject and object, and he seems to avoid making new suffixes,
so I think he pressed /-vaD/ into service.  As you say, there is a
sort of connection between the two meanings, but I don't think the use
with /-moH/ is directly derivable from its use as a simple indirect
object.

>> Adding /-moH/ to a verb doesn't change the roles that already exist
>> for the unsuffixed verb, it just adds a new one. 
>
>While that is true, it does change which noun is place in each 
>role. THAT is what I'm talking about when I talk about changed 
>roles. I'm not saying that the nature of the roles change. I'm 
>saying something very similar to you, with the adjustment that 
>the nouns get moved into the different grammatical positions to 
>fit the grammatical function called for by the idea that is 
>being expressed.
>

You are looking at what happens to particular nouns in the various
transformations of a particular sentence, while I was speaking more
generally.


>I doubt I'm saying this as clearly as you, but I'm trying to 
>describe whatever difference I can see between our models of 
>this grammar.
>

qayajchu'.

>> With an intransitive
>> verb, the subject remains the actor of the verb, but the object role
>> is added.  With a transitive verb, the subject and direct object do
>> not change roles, but the ergative object role is added.  If the direct
>> object is not present, it does not change this relationship. 
>
>It does change which noun would be placed into these roles, 
>however. I suspect that we are talking about slightly different 
>things when we use the word "roles". You seem to be talking 
>about the slots in the grammar. I seem to be talking about the 
>things nouns do when they are in those slots.
>

Exactly. 

>When {-moH} is added to a stative verb, there is indeed a new 
>role and a new function. {bIr HIq} -> {HIq vIbIrmoH}. The 
>noun {HIq} was moved (from subject) to a different slot (direct 
>object) in order to maintain the same function it originally 
>had (being cold). The slot it previously held is not filled by 
>an implied {jIH}, which causes it to be cold.
>
>Qa'Hom Sop targh -> targhvaD Qa'Hom vISopmoH.
>
>Now a new slot in the grammar is opened up and all the nouns 
>shift. In order to introduce the implied {jIH}, {targh} is moved 
>into the indirect object slot.
>
>In the first example, {HIq} moved into a slot which was not just 
>casually vacant. With a stative verb, it was a slot that 
>otherwise could not be filled. With a verb that COULD fill that 
>slot, however, like {Sop}, {targh} moves to a less important 
>slot than that of direct object.
>
>The idea here is to minimize the number of nouns that will need 
>to be shifted around. The relationship between a verb and its 
>direct object is an essential one and to break that relationship 
>in order to add a somewhat different kind of object -- an 
>ergative object as you put it -- hardly seems best. So, what we 
>get is essentially two slightly differing grammars depending on 
>the nature of the verb.
>
>If the verb is stative and would not under normal circumstances 
>have a direct object at all, if {-moH} is added, it is clearly 
>understood that the noun in the direct object position is the 
>one doing the action caused by the subject.
>
>If the verb quite normally has an object, it is less confusing 
>to put that "ergative object" in the indirect object position 
>than the direct object position.
>
>If we hold to these two slightly differing grammars depending 
>upon the nature of the verb, then there is no need for 
>confusion. Things can be stated clearly.
>

Yes, indeed.  From my structuralist viewpoint, I tried to cover
both transitive and intransitive verbs under the same model, by
positing a hierarchy of noun roles/slots associated with the
verb and proposing that /-moH/ adds a new slot to the verb it
is applied to.  From the semantic viewpoint, /-moH/ does very
different things to intransitive and transitive verbs.

>> I can't make any sense out of using /-vaD/ with a causative intransitive.
>> It seems unnecessary.  The verb had only a subject, and when /-moH/
>> was added and a new noun role opened up, that noun was able to act
>> as the object.  There is no other role associated with the verb for
>> /-vaD/ to be applied to.  If a noun with /-vaD/ is present, it must
>> be understood in it's old benefactive role: 
>> 
>> /nguv Duj/ The ship is painted.
>> /Duj nguvmoH jan/ The device paints the ship.
>> /vavwI'vaD Duj nguvmoH jan/ The device paints the ship for my father.
>> /vavwI'vaD nguvmoH jan/ The device paints for my father.
>> /nguvmoH jan/ The device paints.
>
>Again, {-vaD} is defined both as receiving benefit and as being 
>an indirect object, and while the line between these concepts is 
>a fuzzy one, they are not quite the same thing.
>

Right.
 
>> The interpretation of the roles of the nouns with the verb should
>> not change depending on the definition of the verb or on which roles are 
>> filled and which aren't. There's no reason to introduce the sort of
relativity 
>> into the use of /-moH/ that some have been suggesting.
>
>Yes, that is my point, though one phrase doesn't fit what I 
>believe I understand. I think that depending upon what I call 
>transitivity of the verb root, I do see the nature of these 
>positions in grammar as being different. For an intransitive 
>verb, the agent of the verb root is in the direct object 
>position while the agent of causation is in the subject 
>position. With what I call a transitive verb root, the direct 
>object function is served by the direct object position, while 
>the agent of the root verb is moved to the indirect object 
>position and again, the subject is the agent of change.

jIQochbe'bej.

-- ter'eS

Back to archive top level