Types of representations

Signers do not use all the space around their body that potentially could be used. The space below the waist and the space behind the signer, for instance, are generally not used. The actual signing space extends from the top of the head to the waist vertically, and horizontally from slightly past the shoulders forward to about an arm’s reach (Perniss et al., 2007); signs are articulated within this signing space (Figure 2.7).²⁸

Figure 2.7. The signing space: front view, side view, and top view (photo: Annette Jansen, ©RCSI; inspired by a graphical representation of the signing space in Koenen

& Bloem (1992), created by Ruud Janssen).

The signing space is not only used as a location for the articulation of lexical signs, it also plays an integral role in the grammar. Signs can be arranged in space such that the signer, by manipulating the location or the direction of the sign, can signal syntactic relations, spatial layouts, or a combination of both. At discourse level, spatial modification can serve to structure and organize information and to track referents. For now, we will use the term functions to refer to these different uses of the signing space.

In order to achieve these functions, signers need to associate referents with locations in space, i.e., the signer needs to establish location-referent associations (‘localization’, i.e., establishment of a locus or loci). This is achieved by (i) articulating signs from the core lexicon at specific location(s) that deviate from the place of articulation specified for the citation form, (ii) using a pointing sign (glossed as INDEX or IX) from the non-core lexicon, or (iii)

28 For some shared sign languages, such as Adamorobe Sign Language, Kata Kolok and the Sao Tome and Principe Sign Language, a larger signing space has been found (Nyst, 2012; De Vos, 2012; Mineiro et al. 2017).

using classifier signs from the non-core lexicon. These localization devices will be discussed in Section 2.4.3.1.

Location-referent associations can be either abstract or spatially motivated. Abstract loci do not correspond to locations in the real world, while spatially motivated loci imply a correspondence between the location of the hands in signing space and the position of objects in the real (or an imagined) world (Emmorey, 2001). Depending on the information the signer wishes to convey, different 'representations' can be chosen²⁹:

1. Spatial representation 2. Abstract representation

2a. Genuinely arbitrary loci

2b. Abstract loci motivated by semantic-pragmatic considerations or considerations regarding discourse organization

These types of representations suit different functions, for which different devices are recruited. The next three subsections will elaborate on both types of representations, the spatial and the abstract one, as well as a third type of use of space, namely use of the immediate environment.

2.4.2.1 Spatial representations (motivated space)

When the signing space is recruited for spatial representation, the loci in signing space associated with entities reflect the locations of entities in the physical world (which can be a real or an imagined world). Such a spatial description can be presented using two types of ‘spatial formats’ (Emmorey

& Falgier, 1999; Emmorey, 2001) or perspectives (Perniss, 2007, 2012), in which the physical environment (such as the scene depicted in Figure 2.8a) is projected onto the signing space either as a 2D map or a small-scale model seen from a fixed vantage point (‘diagrammatic space’ or ‘observer perspective’, Figure 2.8b), or as seen from a character mapped onto the

29 This proposal diverges from the traditional distinction between 'syntactic use of space' and 'topographic use of space' (summarized in Perniss, 2012), as it adds a distinction between 'genuinely arbitrary locations' and 'motivated abstract locations'. As such, the ‘taxonomy’ proposed here acknowledges the use of space for discourse organization.

signer’s body (‘viewer space’ or ‘character perspective’, Figure 2.8c). In the latter case, the signing space “reflects an individual’s view of the environment at a particular point in time and space” (Emmorey, 2001, p.

154), and consequently, the signing space is conceptualized as 3D, thus encompassing and surrounding the signer’s body (Perniss, 2012).³⁰ In the character perspective, the conceptualized entities in the environment (termed ‘surrogates’ by Liddell (2003a)) are real-life scaled (Emmorey &

Falgier, 1999), and the signer is internal to the event. In the observer perspective, in contrast, the signer describes the event from an external vantage point (Perniss, 2012). The character perspective will be further discussed in the context of constructed action in Section 2.4.5.

Figure 2.8. (a) Real-life scene; (b) Schematic representation of the diagrammatic space/observer perspective (the placement of the persons, fence and bike are schematically represented by lines and dots); (c) the viewer space/character perspective (the conceptualization of the entities is represented by dotted ‘real-life-spaced’ persons and objects) (photo: ©Eveline Boers-Visker).

30 Importantly, the signing space is conceptualized as sourrounding the signer’s body.

This does not mean that the whole space around the body is actually employed to articulate signs. The space behind the signer’s back is not commonly used to articulate signs – with the exception of a few signs that refer to locations on the signer’s body, such as KIDNEY or BACK.

Obviously, the function of the spatial representation is to depict how entities are located in relation to each other. Yet, a spatial locus can also be used to signal syntactic relations. A signer can, for example, localize a person at a locus that corresponds to a real-world location (e.g., seated at a desk) and subsequently direct an agreement verb (signaling syntactic relations) towards this locus (example from Perniss, 2012: 416).

2.4.2.2 Abstract representations

In contrast to spatial representations, abstract representations do not represent actual locations in space. Instead, they convey abstract relations.³¹ Abstract use of space can serve two functions, namely (i) signal syntactic relations, and (ii) organize and structure discourse.

First, use of space can have a syntactic function, in that loci can be utilized to identify a verb’s arguments, or for pronominal reference. A signer may, for example, associate a locus in signing space with a referent ‘teacher’

and another locus with a referent ‘student’, and subsequently modify the agreement verb HELP such that the verb’s movement starts at the locus associated with the teacher and ends at the locus associated with the student (Figure 2.9), resulting in the meaning “The teacher helps the student”. Once established, such loci can be utilized for pronominal reference as well.

Figure 2.9. Abstract use of space to indicate syntactic relations.

31 Please note that the abstract use of space is not attested in some shared sign languages, such as Kata Kolok (De Vos, 2012).

Secondly, space can be used to organize and structure discourse, by relating areas in space to certain topics, and articulating subsequent segments of discourse that are related to these particular topics in these same areas (Winston, 1991; Engberg-Pedersen, 1993). As such, a signer can (i) introduce and elaborate on different topics, (ii) compare two or more topics/ideas/entities, (iii) indicate segments of time and events occurring at various points within these segments (‘time-mapping’, Winston, 1991). An illustrative example of (ii) is provided in Thumann (2013), who describes a signer who has placed Europe at the right side vertical plane in front of him and America at the left side. Subsequently, he provides information about language use on both continents, by directing (some) signs related to language use in Europe towards the locus associated with Europe, and signs related to language use in America towards the locus associated with America. In all these instances, the location of individual signs (e.g., the signs

LANGUAGE, SAME, DIFFERENT, AREA) is modified as compared to their citation form.

In contrast to the spatially motivated loci in spatial representation, the loci in abstract representations are arbitrary, that is, they do not represent actual physical relations. One can, however, distinguish loci that are genuinely arbitrary (that is, neither motivated by physical locations nor by any other considerations) versus loci that are motivated semantic-pragmatic or discourse-organizational considerations. Engberg-Pedersen (1993) has demonstrated that in Danish Sign Language (Dansk Tegnsprog, DTS), the choice of loci can be motivated by any of the following factors: (i) semantic affinity between referents, i.e., referents with semantic affinity to each other are localized at the same locus or area; (ii) canonical locations, where the locus is influenced by the real-world location at which a person or object is normally found; (iii) authority or perceived authority, e.g., when establishing a locus for King Willem-Alexander³² higher in the signing space; (iv) dislike of or affinity with something or someone, e.g., establishing a locus for a person for whom one feels less affinity further away from the body; (v) discourse-organizational considerations, e.g., when juxtaposing loci in order to compare ideas, as described above (Engberg-Pedersen, 1993).

32 King of the Netherlands (2013 – present).

2.4.2.3 Real space

The previous paragraphs have demonstrated that signers can use the space in front of their body for abstract and spatial representation. In both instances, concepts and entities are mapped onto the signing space.

However, when signers are talking about objects and persons present in the immediate environment, they can direct deictic pointing signs and agreement verbs towards these entities.