Syntax-discourse interface

Simon uses the linguistic space to mark grammatical relations at the local level of individual narrative episodes. Yet failure to secure unambiguous referential identity over longer stretches of narrative discourse indicates that he does not yet fully master the use of the relevant linguistic devices to create cohesion at the global narrative level.

Referential establishment and maintenance. In file 1, Simon uses several linguistic means for referential establishment and maintenance. The sequence provided in (153) illustrates how Simon first establishes the loci for the frog group in a semicircle in front of him, recounting afterwards that the dog picks up one of the frogs from this group. In (154) the verbs thank and wave pick up the locus established previously for the frog group (non-manual means, that is, body-shift and eye gaze direction to the right, are used to mark the shifted reference).

Although Simon uses determiners and pronouns only occasionally in this narrative, he uses them appropriately to establish or maintain reference. Example (148), discussed above in relation to complex constructions, documents the use of a personal pronoun to refer to the boy, introduced earlier in the narrative, when the locus for this character was established to the right of the signer. In (155) the determiner det_self appears in combination with the NP boy, at a point in the narrative when the boy is reintroduced as a character (before Simon retells an event with the dog as a protagonist). The loci of the determiners referring to the boy coincide. Further, in example (156) det_loc establishes the locus for the location the boy has fallen into, causing his clothes to be wet.

In this narrative, Simon uses det_exist only once (cf. (157)), when he narrates that the frog is not where the boy expected him to be. It must be noted, however, that the absent subject (the frog) is not referred to explicitly. Incidentally, the sequence in (157) is also illustrative of Simon’s use of h2-classifiers as discourse buoys. In this case, the classifier used to designate the location, though not specified further, is retained in the sign space during the recount of the boy’s search and realisation that the frog searched is not there.

Referential identity in Simon’s recount of the narrative episode involving the frightened dog followed by the bees is only expressed through non-manual means (cf. (158)). Simon uses an NP to reintroduce the dog as a protagonist, without, however assigning him a locus. He then goes on to recount that the dog runs fast. The SRF used for this purpose is marked through head movement and eye gaze direction to the right. Notice that a locus to the right is later picked up in the sequence with the agreement verb sting. So eye gaze direction and head orientation are the only means that are used in this case to maintain reference. However, this type of non-manual marking only represents an optional agreement marker in DGS. Consequently, whether or not the object associated with a locus by sting and the subject of the previous narrative passage (the dog) are identical cannot be established unambiguously.

Another complex sequence involving a rapid change of perspectives is provided in (159), which follows the description of the boy falling and then sitting like a horse rider on something that is not specified any further (we will discuss that sequence [cf. (161) below] in the context of the expression of spatial relations). That description, expressed from a narrator perspective (FRF), is followed by a shift in perspective in (159a), signalled non-manually via eye gaze to the right. The problem with this SRF in (159a) is that the identity of the subject is difficult to establish because the description of somebody supporting himself on something with surprise could represent either the boy’s reaction -after falling on the deer- or the deer’s -realising that something has fallen on his back. As the deer is introduced only after this sequence (that is, in (159b)) the former interpretation seems more likely. In this sequence (159b), the referential framework is shifted to express the deer’s fright. Reference is maintained until (159g) in which Simon switches back to the narrator perspective to describe that the boy falls down from the deer’s head. Note though that the boy is not referred to overtly. That it is the boy falling down can be inferred from the story context and Simon’s choice of a verb form of fall with a classifier element for human beings in (159g).

The analysis makes apparent that POVs are signalled through a lexically overt expression in those contexts in which the signer adopts a character’s perspective other than the boy’s and that character is introduced as a protagonist. It is important to note, in addition, that although non-manual means signal and mark POVs, referential loci are established contrastively only in a few instances. In other words, Simon marks POVs involving the perspectives of different characters via a change of body orientation and eye gaze direction to the right.

Where non-manual marking of POVs is ambiguous and no overt reference forms are used to signal referential shift, an unambiguous interpretation of the utterances is not possible. Consider the sequence in (160), the only sequence containing a performative verb (that is, say). In this sequence, referential loci are not marked contrastively. In (160a) the boy asks for permission to go to sleep (the sequence contains no overt subject, but is part of the initial part of the narrative in which the boy is the protagonist). Neither is the addressee expressed through a lexically overt expression or through non-manual means (eye gaze is directed to a neutral location in the sign space). The same holds of (160b), in which eye gaze is directed toward the audience during the production of the expression OK.

Reference forms and functions. All referents in file 1 are introduced via NPs. It is interesting to note in this context that, compared with other narratives, the dog is only introduced as a protagonist relatively late in the course of the story, that is, after the boy’s realisation of the frog’s escape. Further, the analysis reveals that reference to characters that are reintroduced as a protagonist remains unexpressed in many cases (60% out of 23.8% of reference forms serving this function) (cf. Table 3.21 and Figure 3.3). Against this backdrop, it comes as no surprise that it is difficult, at times, to establish who is the agent of the activities described. This holds equally of those narrative passages that involve the boy as a protagonist. Recall that we repeatedly commented on the lack of an overtly expressed reference to the boy in reintroductory contexts, with the effect that some narrative passages remain ambiguous even if we attributed the main thematic perspective to the boy. This is the case of example (154) above, in which the boy is reintroduced as a protagonist (waving to the frogs) after a sequence in which the dog is reported to take one of the frogs. Notice that the effect of ambiguity is reinforced by the circumstance that referential loci picked up to mark the subject of POVs are not distributed contrastively, as we remarked upon previously.

Table 3.21: Reference forms and functions in Simon’s file 1.*

* Expressed as a percentage of the total number of reference forms (proportions of forms used for respective function in brackets). Absolute numbers are provided in the Appendix Table C-3.

Figure 3.3: Proportion of reference forms and functions in Simon’s file 1.

Expression of spatial relations. In file 1, Simon seldom provides information on ground entities in descriptions of activities that would rather require prior information about them. Consider, for example, Simon’s recount of the boy’s falling on the deer (cf. (161). The boy is reported to fall, ending up in a position like a horse rider on a horse. However, the ground (= the deer) is not specified. Coincidentally, this information gap is consistent with the plot of the narrative as it reflects the misperception of the boy (who thought that he was clinging to the branches of a tree before he eventually finds himself falling on a deer). The misperception as such, however, is not addressed by Simon.

Another example of a missing specification of the ground is provided in example (162), in which Simon describes the frog’s climbing out of a container. The sequence is target-like from a grammatical perspective. However, from a narrative perspective it remains unclear from where the frog escapes, as neither the ground (the jar) has been introduced previously nor has the circumstance that the frog is sitting in the jar been narrated. Moreover, the audience is left to infer that it is actually the frog climbing out of the jar (the two signs preceding (162) are unclear). Another example lacking specification about the ground was discussed above (cf. (157)). Recall, that in (157) Simon reports that the boy is searching the frog, leaning over and looking into a location that has the shape of a hole, backgrounded via an h2-classifier, without any prior specification of where the location might be.

Another factor that contributes to remaining ambiguities in the interpretation of some narrative episodes is the use of generic cl:form signs in the place of conventional signs. This is the case in example (147) above, in which Simon reports on the falling down of an object (the beehive) that is, however, not specified any further. Note that in example (163), too, it is only through the narrative context that the audience might guess who is being reintroduced (the frog, into the jar).

Table 3.22 provides a summary of the linguistic means used to express figure-ground relations in file 1. We can see that out of five spatial configurations, three are narrated without a prior specification of the ground (instead, the signer uses a default h2-classifier). In one case, the ground is overtly referred to via a conventional sign, and in another case through a generic cl:form sign.

Table 3.22: Expression of figure-ground relations in Simon’s file 1.