User talk:Ota
Does typecraft need a layer for word categories? In descriptions in languages like Èdó, morpological and phonological processes interact in word formation. This implies that forms at the morpheme level may be different from forms at the word level as in the following example:
| òwá!bé |
| òwá!bé |
| SchoolLH!H |
| CN |
.
The example above illustrates how morphological and phonological processes influence tone derivation in Èdó with focus on a derived tone:the downstepped high tone. It is derived through the processes of vowel assimilation/elision and contraction, tone spreading and tone simplification, when a word ending with a high tone bearing syllable combines with a noun with an initial low tone bearing syllable across word boundaries.
To accurately describe this process, I have had to use the free translation slot to represent the morpheme level. This is necessary because the tonal realization are different for both levels.
A similar observation applies to some syntactic processes such as the formation of some pronouns as in the following example:
| Ọ̣̣̣̣̀ |
| Ọ̣̣̣̣̀ |
| 3SGSBJNOMAGT |
| PN |
| rhié |
| rhié |
| takePASTH |
| Vdtr |
| èbé |
| èbé |
| book.AFFDO |
| CN |
| nẹ̣̣̣̣́rẹ̣̣̣̣̀n | |
| ̣nè | ẹ̣̣̣̣́rẹ̣̣̣̣̀n |
| prep | 3SGOBJindDATBEN |
| PN | |
.
Here the final vowel of pronoun nè is elided and no tone simplification process takes place. Also, the part of speech information for the initial morpheme of the pronoun nè (prep)+érèn (3SG.OBL) is different from the output word nérèn which is of type PRON.I have had to use the meaning slot to represent part of speech information of the morpheme that is different from the part of speech of the resultant word.
Comments from Ota.