Previous studies on children’s acquisition of depicting verbs in signed languages have chiefly studied the use of classifiers in verbs of motion and location, particularly the order in which the different classes of handshape are acquired. The age of the children in these studies have ranged from age three to thirteen, and an important finding has been that classifier constructions are not fully acquired until early adolescence. Most of these studies have used an elicitation tool to investigate the production and comprehension of classifiers, but have not provided any adult target norms of the test items when scoring children’s achievement. The present dissertation provides a detailed description of both adults’ and children’s verb constructions in descriptions of cutting and breaking events in Swedish Sign Language (SSL), specifically focusing on the number of hands used in signing, handshape category and hand activity, which has not been previously described for any sign language. As part of this study, 14 deaf adults (ages 20–72) and 11 deaf children (2;1–6;6) of deaf parents, all native-users of SSL, performed a task that involved describing 53 video clips of cutting and breaking events. The clips show an event in which an actor separates material, either with the aid of a tool or without. Additionally, some clips show an entity separating by itself without an actor being involved. The adults described the events with depicting verb constructions that are produced with two hands. The analysis of the handshapes produced three categories: substitutor, manipulator and descriptor. The most frequent construction in the description of events without a tool was two acting manipulators (depicting a hand handling an object), whereas in descriptions of events with a tool the combinations were acting substitutor or manipulator with a non-acting manipulator. The acting hand referred to the tool and the non-acting manipulator to the affected entity. In descriptions of events without an actor, either two substitutors or two manipulators were used. In addition to depicting verb constructions, the descriptions also contained resultative complements, i.e. signs carrying information about the result of the activity being carried out. The complements were either lexical signs or some form of depicting verb construction. Similar observations have not been noted for any other signed language. In the manner of the adults, the children used depicting verb constructions in descriptions of cutting and breaking events (681 tokens), but they also used pointing and lexical signs (64 tokens). Nearly half of the verb constructions that were used by the children corresponded to the adult target forms. The majority of the constructions describing events without a tool corresponded to the adult target forms using two acting manipulators, even among the youngest informants. In events with a tool, only a third of the constructions corresponded to the adult target forms (emerging at 4;8 – 5;0); the remaining two-thirds were deviating constructions in terms of number of hands, handshape category and hand activity. Resultative complement are sparsely used by children (57 tokens), the most chosen type of complement being lexical signs. Pervasive features of children’s constructions were the addition of contact between the hands and a preference for substitutors, something not found in adults’ constructions. These features were elucidated within the framework of Real Space blending theory, with the study showing that children first use visible blended entities and that invisible blended entities do not emerge until 4;8–5;0. / <p>Disputationen teckenspråkstolkas</p>

The ability to recognize handshapes in signing video is essential in algorithms for sign recognition and retrieval. Handshape recognition from isolated images is, however, an insufficiently constrained problem. Many handshapes share similar 3D configurations and are indistinguishable for some hand orientations in 2D image projections. Additionally, significant differences in handshape appearance are induced by the articulated structure of the hand and variants produced by different signers. Linguistic rules involved in the production of signs impose strong constraints on the articulations of the hands, yet, little attention has been paid towards exploiting these constraints in previous works on sign recognition. Among the different classes of signs in any signed language, lexical signs constitute the prevalent class. Morphemes (or, meaningful units) for signs in this class involve a combination of particular handshapes, palm orientations, locations for articulation, and movement type. These are thus analyzed by many sign linguists as analogues of phonemes in spoken languages. Phonological constraints govern the ways in which phonemes combine in American Sign Language (ASL), as in other signed and spoken languages; utilizing these constraints for handshape recognition in ASL is the focus of the proposed thesis. Handshapes in monomorphemic lexical signs are specified at the start and end of the sign. The handshape transition within a sign are constrained to involve either closing or opening of the hand (i.e., constrained to exclusively use either folding or unfolding of the palm and one or more fingers). Furthermore, akin to allophonic variations in spoken languages, both inter- and intra- signer variations in the production of specific handshapes are observed. We propose a Bayesian network formulation to exploit handshape co-occurrence constraints also utilizing information about allophonic variations to aid in handshape recognition. We propose a fast non-rigid image alignment method to gain improved robustness to handshape appearance variations during computation of observation likelihoods in the Bayesian network. We evaluate our handshape recognition approach on a large dataset of monomorphemic lexical signs. We demonstrate that leveraging linguistic constraints on handshapes results in improved handshape recognition accuracy. As part of the overall project, we are collecting and preparing for dissemination a large corpus (three thousand signs from three native signers) of ASL video annotated with linguistic information such as glosses, morphological properties and variations, and start/end handshapes associated with each ASL sign.

In this dissertation, I investigate various aspects of object marking and how these manifest themselves in the signed modality. The main focus is on Swedish Sign Language (SSL), the national sign language of Sweden, which is the topic of investigation in all five studies. Two of the studies adopt a comparative perspective, including other sign languages as well. The studies comprise a range of data, including corpus data, elicited production, and acceptability judgments, and combine quantitative and qualitative methods in the analyses. The dissertation begins with an overview of the topics of valency, argument structure, and object marking, primarily from a spoken language perspective. Here, the interactions between semantics and morphosyntax are presented from a typological perspective, introducing differential object marking as a key concept. With regard to signed language, object marking is discussed in terms of both verbal and nominal strategies. Verbal strategies of object marking among sign languages include directional verbs, object handshape classifiers, and embodied perspective in signing. The first study investigates the use of directionality and object handshapes as object marking strategies in Al-Sayyid Bedouin Sign Language (ABSL), Israeli Sign Language (ISL), and SSL. It is shown that the strategies generally display different alignments in terms of the types of objects targeted, which is uniform across languages, but that directionality is much more marginal in ABSL than in the other two languages. Also, we see that there is a connection between object marking strategies and the animacy of the object, and that the strategies, object animacy, and word order preferences interact. In the second and third studies, SSL is investigated with regard to the transitive–reflexive distinction. Here, we see that there are interactional effects between object handshapes and the perspective taken by the signer. This points to intricate iconic motivations of combining and structuring complex verb sequences, such as giving preference to agent focusing structures (e.g., agent perspective and handling handshapes). Furthermore, the use of space is identified as a crucial strategy for reference tracking, especially when expressing semantically transitive events. Nominal strategies include object pronouns and derivations of the sign PERSON. The fourth study provides a detailed account of the object pronoun OBJPRO in SSL, which is the first in-depth description of this sign. It is found that the sign is in widespread use in SSL, often corresponds closely to object pronouns of spoken Swedish, and is argued to be grammaticalized from the lexical sign PERSON. In the final study, the possible existence of object pronouns in other sign languages is investigated by using a sample of 24 languages. This analysis reveals that the feature is found mostly in the Nordic countries, suggesting areal contact phenomena. However, the study also shows that there are a number of derivations of PERSON, such as reflexive pronouns, agreement auxiliaries, and case markers. The use of PERSON as a source of grammaticalization for these functions is attributed to both semantic and phonological properties of the sign. This dissertation is unique in that it is dedicated to the topic of object marking in the signed modality. It brings a variety of perspectives and methods together in order to investigate the domain of object marking, cross-linguistically and cross-modally.