|
qhsAn-diz
good-aDvz
mAni-jAr
song-pl
luhu-zwA.
say-ipfv
‘Jusuf sings better than anyone.’
jak
meat
čig-diz
raw-aDvz
t’ü-n
eat-msd
‘to eat meat raw’
In Lezgian, Adverbs can be derived from Adjectives, primarily by two suffixes: -dakaz and
–diz/-z (Haspelmath 1993: 113). Example (3.19a) illustrates the typical use of manner adverbs. But as illustrated in (3.19b), derived Adverbs may also function as depictives. Despite such encoding patterns, Loeb-Diehl argues that depictives primarily share char- acteristics with resultatives, in that they are both participant-oriented. She also shows that it is not particularly common that encoding is shared between manner predications and resultatives in her language sample (2005: 14).
Also Verkerk (2009) places resultatives, depictives, and manner predications within the domain of secondary predication, based on the assumption that they “share the property of containing two predicative constituents”, one expressing an event and the other a state or a property (2009: 115). Here too, event orientation is argued to be a characteristic of manner predications, as opposed to participant orientation in the case of resultatives and depictives. Like Loeb-Diehl (2005), Verkerk (2009) argues that manner predications nonetheless share with depictives the property of being simultaneous with the main event. As already mentioned in section 2.5.2, Verkerk introduces another characteristic, which manner predications and resultatives share, namely that “their controller is not the subject introduced by the primary predicate” (2009: 119). This is straightforward in the case of resultatives, as they most commonly predicate something of the object of a transitive sentence (e.g. the barn in example 3.15 above). But depictives, too, may have a non- subject (or object) controller, as Verkerk herself shows, e.g. in Mary drinks her coffee black, where the object her coffee is depicted as black, as opposed to cases like Carla went to work drunk, where the subject Carla is depicted as drunk (2009: 116). In fact, there are also resultatives that have a subject controller, e.g. The river froze solid, but these are classified as marginal by Verkerk (2009: 117-118). Verkerk argues that, instead of the subject, manner predications “refer to the event encoded by the primary clause as
a whole” (2009: 117). With typical manner adverbs such as Mary ran fast, it is unclear whether these really relate to ‘the event encoded by the primary clause as a whole’ (e.g. Mary’s running), or simply ‘the event encoded by the main predicate’ (e.g. the running). In other words, it is not always clear whether manner adverbs that modify an event also modify the participants of that event. If they do relate to the whole clause, this is a very different scope of modification compared to the participant of a clause found in the subject or object of the main predicate, in the case of resultatives and depictives. The shared characteristic of resultatives and manner predications can also be questioned since this is based on their lack of a property, rather than an actual shared feature. But Verkerk argues for the relevance of this assumption when using her three-way division to motivate cross-linguistic encoding patterns. The three constructions are all encoded in the same way in some languages, and pattern in pairs in either combination in other languages. This is argued to be due to their shared properties. It may appear to be both practical and plausible to treat resultatives, depictives, and manner predications as instances of secondary predication, but whether typological findings provide enough basis for this conclusion can nonetheless be discussed.
Contrary to the accounts described so far, Himmelmann & Schultze-Berndt (2005b) do not classify manner adverbs as secondary predicates. They define secondary predicates as adjuncts that predicate, meaning that “the state encoded by the secondary predicate is interpreted as holding for one of the participants of the main predicate” (2005b: 4). Depic- tive and resultative constructions can both be classified as secondary predicates. Manner adverbs, on the other hand, do not encode a state that can be interpreted as holding for a participant. Thus, it is the difference between participant and event orientation that appears to be the reason why Himmelmann & Schultze-Berndt exclude manner adverbs from the domain of secondary predication. This does not imply that they expect the construction types to always be neatly distinguished: Himmelmann & Schultze-Berndt (2005b), too, state that overlaps between constructions encoding participant and event orientation are common across languages. Both Verkerk (2009) and Himmelmann & Schultze-Berndt (2005b) are studies of secondary predicates specifically (depictives in the case of the latter), but with opposing views on whether manner adverbs are secondary predicates. Contrary to this, Loeb-Diehl (2005) focuses on manner expressions, but de- fines them as secondary predicates. In the more general definition found in the entry on predication by Rothstein (2006: 73–76) in Encyclopedia of Language and Linguistics, sec- ondary predicates are said to comprise resultatives and depictives only, whereas adverbs are not even mentioned.
Do'stlaringiz bilan baham: |
