Special Adjectives

There are a handful of adjectives including mostly color adjectives and a few physical attributes that have at least two forms that have different syntactic distributional patterns. The adjectives in the table below are the ones collected so far that seem to have at least two forms:

'white'	'black'	'red'	'strong'
ran, tan, ndan	baal, paal	*yax	*saƭ
ranig, ndanig	balig	yaxig	saƭik
ndangu, tangu	balgu	yaxgu	saƭku

Error creating thumbnail: Unable to save thumbnail to destination

The forms in lines 1, 2, and 3 above arise in different grammatical contexts. The derived forms on line 2 can only be used with the na relative constructions; the root form in 1 and the derived form in 3 are both used in either verbal predication constructions noun-adjacent modification, while the root form in 1 is a bare form that can appear only noun-adjacent:

Error creating thumbnail: Unable to save thumbnail to destination

(Special note: it can also be muus ɓalgu magnu ne and muus magnu ɓalgu ne, so these forms do not rule out the possibility of the adjectives being all -u adjectives, rather than bare ones.)

The forms in group 2 are more productive since they can also occur with copular predication:

 ke moʄ na may na akong ake kaate yaxig	
 ‘most of the gorillas are red’

They may or may not occur with a final -u when they are modifying the noun directly, a fact that holds true for group 1 as well, but for group 2 usually occurrences with -u are more rare than they are for the group 1:

 naamam doxang naxar ne yaxigu na
 ‘I eat under the tree that is red’

They also introduce a phonological change in the root form by turning long root vowels to short vowels in the resulting form: baal -- > balig. In terms of phonological change, it stands to reason that the forms in group 2 are derived from those in group 1 with a suffix -ig, -ik, -id, or -in, and in turn those in group 3 are reduced forms of those in 2. In fact, the consultant accepts alternative forms such as the following:

 omaag omagnu le		
 omaag omaginu le
 ‘the ocean (lit. big bigness)’

The suffix -in is commonly a causative suffix in verbal morphology (e.g., yok ‘get long’ yokin ‘lengthen’), and this may signal a grammaticized adjectival form originally derived from a causative verb. Similarly, -id is another causal verbal suffix: fil ‘fly off’ filid ‘brush aside’. The suffix -ik is an deictic venitive derivational morpheme that indicates motion toward, and may in this case add a metaphorical meaning of turning into or coming to be in a certain state denoted by the adjective. Possibly, -ik and -ig are allomorphs. This seems to be the full set of suffixes responsible for the distinction between group 1 and groups 2 and 3.

There is an additional form of reduplicative modification, which seems to be possible with this set of adjectives. Syntactically it consists of a verbal predication with the form of in group 3 followed by the same word repeated, but in the form from group 2:

 oboxole abalga balig
 oboxole abaliga balig
 ‘the dog is black’

This reduplication is common with verbs in general, both regular and stative, and is highly productive (see Reduplication for other similar patterns):

 areta ret			‘he/she/it went’
 agara gar			‘he/she/it came’
 japil fe adura dur		‘the knife is dull’
 japil fe awela wel		‘the knife is sharp’
 obii ole anewa new		‘the fruit is small’
 axaɗa xaɗ			‘he/she/it is bitter’
 agooka gook			‘he/she/it is too big for his/her britches’

A hypothesis is that, given that the second word is optional, the word boundary is between the full inflected verb, e.g. aɓalga, and a verbally uninflected bare adjective or deverbal noun, e.g. ɓalig. It is uncertain whether this second element is a noun or an adjective, since we have aɓalga ɓalig (the second clearly being a derived adjective) as well as aɓalga ɓaal, in which ɓaal may be a noun 'blackness' (the latter expression is mostly used to refer to a person as having dark skin). (See the discussion on Modifying with Nouns).

There is an additional form with o: aɓalga o ɓaal. However, this has not be explored thoroughly yet, and it remains to be determined what the function of 'o' is and what can follow it.

Based on usage patterns, the semantics of each version, at least for some of these adjectives, show subtle difference. An example of subtle semantic distinction is ciig ‘long’. The consultant defines ciig primarily as ‘long’ (sometimes ‘long’ or ‘tall’) and jigid primarily as ‘tall’.

Go back to: Modification Strategies

Oana 01:07, 8 December 2012 (UTC)