Difference between revisions of "Talk:Annotating Tumbuka"

Latest revision as of 21:38, 7 March 2010

This page has developed nicely --Dorothee 13:51, 11 February 2010 (UTC)

The use of kwa and na and their annotation as of is interesting. Abdulaziz Lodhi (personal conversation) pointed out to me that kwa is a "master preposition" in many Bantu languages. I interpret that to mean that it is less specific, and it will be interesting to see more data showing the distribution of these two prepositions in Tumbuka. --Dorothee 21:37, 7 March 2010 (UTC)

I am interested in the annotation of the last three words: nchunga kwa Mary translated as beans for Mary.

is there no plural noun class prefix on nchunga?
kwa looks like a complex word composed of kw-a, with kw as the noun class agreement marker and a as a possessive marker. ??

Could you help Jean :)

--Dorothee 10:37, 16 February 2010 (UTC)

Nchunga is both singular as CL9 and plural as CL10. There are no affixes marking plurality. May be I could just add that CL10 on the glosses under nchunga 'beans'?

yes that would be good --Dorothee 20:22, 20 February 2010 (UTC)

Kwa is simply a preposition and its not agreeing with any of the nouns. If it was used as an agreement marker, it would have been ku-a = kwa under class 16 which is a locative class.

I see.

I like to understand the relation between the prepositional phrase kwa Mary and the causative. Mary is the causee, and I was expecting Mary to occur as a direct object of the verb. So I am surprised to see it occur as a prepositional object. --Dorothee 20:22, 20 February 2010 (UTC)

CiTumbuka prohibits occurence of more than two postverbal NPs such that in a sentence where there are are more than two one is sent into the oblique position. In the causative construction where Mary is occurring,the base verb 'to cook' is transitive. Introduction of causative suffix introduces argument number and so does introduction of the applicative suffix. Instead of having the three postverbal NPs in a row, one them has been sent to the oblique position as a strategy to avoid them disobeying the constraint.

Let's compare the example above with the following example:

Here we have a passivized Causee, namely Temwa. The Causer, Temwani, occurs as an oblique object, that is as a prepositional phrase headed by na. Comparing the two phrase we see that kwa and na both are given as corresponding to the English prepositon by.

Yes, they both save as 'by' in different contexts. 'na' is basically the one used to introduce the passive oblique object. But we can equally say "Nchunga zi-ka-phik-isk-ik-a kwa Maria" 'The beans were cooked by Maria' or 'Maria had the beans cooked by someone'

Hm... --Dorothee 20:38, 20 February 2010 (UTC)

@@ Line 1: / Line 1: @@
-We have annotated the following sentence;
+This page has developed nicely --[[User:Dorothee Beermann|Dorothee]] 13:51, 11 February 2010 (UTC)
-<Phrase>2797</Phrase>
+The use of ''kwa'' and ''na'' and their annotation as ''of'' is interesting. Abdulaziz Lodhi (personal conversation) pointed out to me that ''kwa'' is a "master preposition" in many Bantu languages. I interpret that to mean that it is less specific, and it will be interesting to see more data showing the distribution of these two prepositions in Tumbuka.
-Here little stuff: it seems that the meaning of 'ka' in ''akamayeza''is still not reflected in the annotation; perhaps that could be added; also the POS tag for the possessive is still missing.
+--[[User:Dorothee Beermann|Dorothee]] 21:37, 7 March 2010 (UTC)
-Here a questions potentially more important; I am wondering about the the annotation of noun classes If noun classes are not indicated we cannot identify in the glosses which elements the nouns actually agree with.I guess we need class information both on the nouns and the verbs that they agree with.                carefully it also seems that the imported nouns:
-''mavayilasi'', ''maparasayiti'' etc. carry a class marker. It probably would  be a good idea to annotate those.
-<Phrase>2775</Phrase>
+<Phrase>6353</Phrase>
-In the next sentence:
+I am interested in the annotation of the last three words: ''nchunga kwa Mary'' translated as ''beans for Mary''.
-<Phrase>2777</Phrase>
-is ''idwa'' really the passive marker? The sentence does not seem to have a passive reading.
-The following phrase poses a problem:
+* is there no plural noun class prefix on ''nchunga''?
-<Phrase>2805</Phrase>
+* ''kwa'' looks like a complex word composed of ''kw-a'', with ''kw'' as the noun class agreement marker and ''a'' as a possessive marker. ??
+Could you help Jean :)
+--[[User:Dorothee Beermann|Dorothee]] 10:37, 16 February 2010 (UTC)
+Nchunga is both singular as CL9 and plural as CL10. There are no affixes marking plurality. May be I could just add that CL10 on the glosses under nchunga 'beans'?
+yes that would be good --[[User:Dorothee Beermann|Dorothee]] 20:22, 20 February 2010 (UTC)
+Kwa is simply a preposition and its not agreeing with any of the nouns. If it was used as an agreement marker, it would have been ku-a = kwa under class 16 which is a locative class.
+I see.
+I like to understand the relation between the prepositional phrase ''kwa Mary'' and the causative. Mary is the causee, and I was expecting ''Mary'' to occur as a direct object of the verb. So I am surprised to see it occur as a prepositional object. --[[User:Dorothee Beermann|Dorothee]] 20:22, 20 February 2010 (UTC)
+CiTumbuka prohibits occurence of more than two postverbal NPs such that in a sentence where there are are more than two one is sent into the oblique position. In the causative construction where Mary is occurring,the base verb 'to cook' is transitive. Introduction of causative suffix introduces argument number and so does introduction of the applicative suffix. Instead of having the three postverbal NPs in a row, one them has been sent to the oblique position as a strategy to avoid them disobeying the constraint.
+Let's compare the example above with the following example:
+<Phrase>6352</Phrase>
+Here we have a passivized Causee, namely Temwa. The Causer, Temwani, occurs as an oblique object, that is as a prepositional phrase headed by ''na''. Comparing the two phrase we see that ''kwa'' and ''na'' both are given as corresponding to the English prepositon ''by''.
+Yes, they both save as 'by' in different contexts. 'na' is basically the one used to introduce the passive oblique object. But we can equally say "Nchunga zi-ka-phik-isk-ik-a kwa Maria" 'The beans were cooked by Maria' or 'Maria had the beans cooked by someone'
+Hm...
+--[[User:Dorothee Beermann|Dorothee]] 20:38, 20 February 2010 (UTC)