Tommi Jantunen

Ellipsis in Finnish Sign Language

This paper deals with syntactic ellipsis in clauses in Finnish Sign Language (FinSL). The point of departure for the paper is the observation, confirmed by several studies, that clauses in FinSL are often syntactically incomplete. Building on this, the paper first describes how all core-internal clausal material may be elided in FinSL: core arguments in clauses with a verbal nucleus, core-internal NPs in clauses with a nominal nucleus, and even nuclei themselves. The paper then discusses several grammatical contexts which especially favor ellipsis in FinSL. These are questionanswer pairs, two-clause coordinated structures, topic-comment structures, blend structures, and structures containing gesturally indicating Type 2 verbals. Finally, the paper argues that FinSL conforms to the main characteristics of a discourse-oriented language, and that FinSL clauses are not highly governed units syntactically.

Signs and Transitions: Do They Differ Phonetically and Does It Matter?

The point of departure of this article is the cluster of three pre-theoretical presuppositions (P) governing modern research on sign languages: (1) that a stream of signing consists of signs (S) and transitions (T), (2) that only Ss are linguistically relevant units, and (3) that there is a qualitative (e.g., phonetic) difference between Ss and Ts. Of these, the article focuses on the relatively untested P3, which is used to back up P1 and P2, and investigates the velocity and acceleration properties of Ss and Ts on the basis of continuous motion-capture data from Finnish Sign Language. The main finding of the study is that the speed of Ss is slower (and varies less) than that of Ts but that Ss still involve more accelerating (and varying) motion. The purely linguistic interpretation of the results is that the slower speed of Ss enables the addressee to perceive more accurately the fine S-internal linguistic details, manifested by high acceleration. The more theoretical conclusion is that the results support all three Ps but that only P3 is supported directly. As previous key research findings concerning Ts partly contradict P1 and P2, further research on Ts is called for.

Head Pose Estimation for Sign Language Video

We address the problem of estimating three head pose angles in sign language video using the Pointing04 data set as training data. The proposed model employs facial landmark points and Support Vector Regression learned from the training set to identify yaw and pitch angles independently. A simple geometric approach is used for the roll angle. As a novel development, we propose to use the detected skin tone areas within the face bounding box as additional features for head pose estimation. The accuracy level of the estimators we obtain compares favorably with published results on the same data, but the smaller number of pose angles in our setup may explain some of the observed advantage.

Detecting Hand-Head Occlusions in Sign Language Video

A large body of current linguistic research on sign language is based on analyzing large corpora of video recordings. This requires either manual or automatic annotation of the videos. In this paper we introduce methods for automatically detecting and classifying hand-head occlusions in sign language videos. Linguistically, hand-head occlusions are an important and interesting subject of study as the head is a structural place of articulation in many signs. Our method combines easily calculable local video properties with more global hand tracking. The experiments carried out with videos of the Suvi on-line dictionary of Finnish Sign Language show that the sensitivity of the proposed local method in detecting occlusion events is 92.6%. When global hand tracking is combined in the method, the specificity can reach the level of 93.7% while still maintaining the detection sensitivity above 90%.

Constructed Action, the Clause and the Nature of Syntax in Finnish Sign Language

Abstract This paper investigates the interplay of constructed action and the clause in Finnish Sign Language (FinSL). Constructed action is a form of gestural enactment in which the signers use their hands, face and other parts of the body to represent the actions, thoughts or feelings of someone they are referring to in the discourse. With the help of frequencies calculated from corpus data, this article shows firstly that when FinSL signers are narrating a story, there are differences in how they use constructed action. Then the paper argues that there are differences also in the prototypical structure, linkage type and non-manual activity of clauses, depending on the presence or non-presence of constructed action. Finally, taking the view that gesturality is an integral part of language, the paper discusses the nature of syntax in sign languages and proposes a conceptualization in which syntax is seen as a set of norms distributed on a continuum between a categorial-conventional end and a gradient-unconventional end.

Acceleration peaks and sonority in Finnish Sign Language syllables

This paper deals with the phonetic syllable and sonority in signed language. Specifically, the paper investigates, on the basis of motion capture data from Finnish Sign Language, the Acceleration Peak Hypothesis, according to which the sonority peak that defines the empirical syllable corresponds physically to an acceleration peak. The prerequisite for the hypothesis to be supported was that acceleration peaks should be found only in syllables and that each syllable should contain only one such peak. Results based on measuring the acceleration of the movement of the tip of the dominant hand index finger show that the number of acceleration peaks in syllables can vary from zero to three and that acceleration peaks can also be found in transitions preceding and following the syllable. Furthermore, the highest acceleration peaks associated equally with syllables and transitions. Consequently, the Acceleration Peak Hypothesis was not supported by the study. Alternatives for the physical basis of sonority as well as more general implications for research into the signed syllable are discussed.