An entity classifier represents a whole entity, which may be animate or inanimate (but note that bodypart classifiers, which represent a part of an entity, are described separately in the next section). The SignGram Blueprint distinguishes between entity classifiers and size-and-shape specifiers (SASS), the latter coming in two types (static and tracing SASS), but I follow one study in treating static SASS as a type of entity classifiers, while tracing SASS are described separately in MORPHOLOGY 5.2. The only difference between ‘regular’ entity classifiers and static SASS is that the former represent the entity directly by visualizing a ‘filled form’, while the latter represent the entity more indirectly by visualizing (part of) the outline. Apart from this, the two types function alike, as they both combine with (intransitive and unaccusative) verbs of motion and location. Examples and more morphosyntactic information will follow after addressing the set of classifier handshapes.
This same study found 15 handshapes that can act as entity classifier handshapes. Table X shows these handshapes in the first column, together with a verbal description of the form of represented entities in the second column, and examples of these entities in the third – the table is mainly based on the mentioned study, but slightly adapted and extended after discussions with a native signer (see also Info on Data and Consultants). Classifiers that (also) function as static SASS are positioned in the lower half of the table and indicated by (SASS).
|
Handshape |
Description of represented entities |
Examples of represented entities |
|
|
Flat and wide |
Books, sheets of paper, walls, tables, vehicles, flags |
|
|
Tiny |
Small insects, contact lenses, drops of water |
|
Entities of unspecified shape or shape that is difficult to represent by any other classifier. |
|
|
|
( |
Long and narrow |
Poles, pens, knives, toothbrushes, trees (the handshape indicated by * may refer to multiple of these entities). |
|
Animate |
Humans, animals (the handshapes indicated by # and * may refer to multiple of these entities). |
|
|
|
Airplanes |
|
|
|
Trees |
|
|
|
3D |
Circles |
|
|
Small 2D round |
Coins, buttons |
|
|
Flat rectangular |
Paintings, mirrors |
|
|
Large 2D round |
Biscuits, glasses, discs |
|
|
3D round/cylindrical |
Mugs, apples, balls, poles, trees |
|
|
3D round |
Apples, stones |
|
Entities with many long and thin extensions |
Spiders, grabbers |
|
|
Entities of undetermined shape/abstract entities |
Village center
|
Table X. Handshapes that can function as entity classifiers, with a description of the form of referents and examples of referents (partly based on on Zwitserlood 2003: 138-140)
These handshapes can combine with stems of motion and location – the mentioned work further distinguishes between location and existence, I follow others by collapsing them. Four examples of clauses with a classifier predicate of location are visualized below. From Video X.a and X.b, it is clear that the same entity can be represented by different entity classifiers, caused by a difference in perspective of the signer. In X.a, the tree is rather large and close, and depicted by the underarm and hand. In X.b, the tree is far away and fairly small, and therefore visualized by the index-finger. In Video X.c and X.d, it is shown that the same classifier handshape can be used for two different objects: in X.c, the ]-hand is used for a car, while in X.d, it represents a book.
|
Video clips |
Video clips |
|
X.a Tree CL( |
X.b Tree CL(B): ‘be_locatedb’ |
|
Video clips |
Video clips |
|
X.c Car CL(]):‘be_locateda’ |
X.d Book CL(]): ‘be_locatedb’ |
The following examples depict classifier predicates of movement. In Video X.a and X.b, the entities move from left to right; in X.c the entity falls.
|
Video |
Video |
video |
|
X.a Car CL(]):‘move_from_left_to_right’ |
X.b Person CL(B): ‘move_from_left_to_right’ |
X.c Apple CL(<):‘fall_down’ |
Without going into too much detail, two aspects of the (morpho)syntactic behavior of these classifier predicates are worth mentioning. First, note that the classifier predicate follows the noun in all six examples above. According to one Dutch researcher, this is not obligatory for NGT: the noun can be omitted when the referent is clear from the context, or the noun can follow the classifier predicate. Second, there is a relationship between classifier type and argument structure: entity classifiers only combine with verbs that are intransitive and unaccusative. Thus, the verbs displayed in the examples above (be_located, move and fall) only need one argument, and express an activity which the subject simply ‘undergoes’, instead of being actively involved in it (i.e., as an agent).
Often, signers will not reduplicate the noun itself to indicate its spatial distribution, but they will use classifiers instead. When (verbal) classifier constructions are used to indicate distribution of identical entities in a row, there is a third option in which the dominant hand is used as an anchor point while the other hand articulates the iterations.