5.2 The structure of signs

Considerable attention has been paid in sign language linguistics to the internal structure of signs, and research in this field has established the existence of sign language phonology.³ However, two fundamentally different ways of analysing the sign have emerged: one emphasising the simultaneity of the sign and the other the sequentiality of the sign. In the subsequent chapters these different types of sign analysis will be presented.

5.2.1 A simultaneous model

The first linguistic evidence that the signs of a sign language are something more than gestures which lack internal structure was presented by the American linguist William Stokoe in his seminal work Sign Language Structure: An Outline of the Visual Communication System of the American Deaf in the year of 1960 (Liddell 1984:373). Stokoe (1993 [1960]) argued that the signs of a sign language should be regarded as having an abstract sublexical structure similar or parallel to the one that underlies words in spoken languages.

In his analysis of signs in ASL, Stokoe (1993 [1960]) distinguished three aspects of the sign: dez (articulator), tab (place of articulation), and sig (articulation). According to Wallin’s explanation (1994:3), “[t]he articulator is the acting hand/s assuming different handshapes, and orientation/s, i.e. how the hand/s is/are held in relation to the signer’s body. The place of articulation is the position where the hand is acting and consists of either the space in front of the signer or a position on the signer’s body. Articulation is the action of the hand and consists of various types of movement, directions of movement and types of interaction” (Wallin 1994:3). These three aspects comprise only a limited number of possible elements of each category, and a change in one of the aspects results in the change of the meaning of a sign. As an illustration, see the sign pair CUNNING and CAT from Estonian Sign Language (Figure 2):

CUNNING	CAT
Figure 2. Difference in one aspect: place of articulation (Laiapea 1992:2102).

This pair demonstrates a difference in one aspect, the place of articulation, whereas the handshape (articulator) and movement (articulation) are identical. The difference in articulator (handshape) is demonstrated by the signs MOTHER and WOMAN from ESL (see Figure 3).

MOTHER	WOMAN
Figure 3. Difference in one aspect: handshape (Laiapea 1992:2102).

In this pair the movement (articulation) and location (place of articulation) are the same. Signs FATHER and RUSSIAN from ESL are identical, except for the movement (articulation) (see Figure 4).

FATHER	RUSSIAN
Figure 4. Difference in one aspect: movement (Laiapea 1992:2102).

Stokoe’s model (often referred to as the aspectual model) underlines the simultaneity of the three aspects, and each aspect “is a different way of focusing on the unit that is the sign” (Engberg-Pedersen 1993:38). In addition to this original model of analysing a sign, several other models describing the sign as mainly simultaneously organised have been proposed by linguists (cf. Klima & Bellugi (1979) [American Sign Language]). For instance, the parameter model, which is now regarded as traditional phonological analysis of signs describes the sign “as a simultaneous unit of items from different parameters” (Engberg-Pedersen 1993:36), including the place of articulation, handshape, movement of the hand/s, and orientation of the hand/s. The difference between the aspectual model and the parameter model is emphasised by Engberg-Pedersen (1993:37) as follows:

[w]hereas the parameter model is an attempt of a phonological analysis of signed morphemes parallel to phonological analyses of spoken language morphemes, Stokoe based his analysis of sign forms on a recognition of the difference between a spoken and a signed morpheme.

5.2.2 A sequential model

Until the descriptions outlined by Liddell, the sequential nature of the sign was not given much attention in sign language research. Liddell (1984) introduced a theoretical model of sign structure which segments signs on the basis of movement sequences. The segment types are divided into broad categories: M (movement), where the hand/s move/s along a path, and H (hold), where the hand/s remain/s stationary. Liddell (1984:381) explains that, for example, the sign THINK in ASL was traditionally viewed as a simple sign consisting only of a single handshape (articulator), a single location for contact (place of articulation), and one motion (articulation). Liddell argues, that the sign

Liddell (1984) also draws a parallel between his segmented theory of sign structure and the morpheme structure of spoken languages. In ASL, movements and holds are subject to restrictions on their sequencing just as vowels and consonants are in spoken languages. Liddell (1994:3916) writes that two-segment signs are overwhelmingly ordered MH rather than HM; three-segment signs have HMH as a dominant order; four-segment signs have MHMH as a dominant order. It appears that the morpheme structure constraints in ASL (and obviously in other sign languages as well) are as strict as in spoken languages.

Different models emphasising the sequentiality of the sign have been proposed since Liddell’s original theory, cf. Perlmutter (1990) [American Sign Language], Liddell (1990) [American Sign Language].

5.2.3 Fingerspelling

The signs of sign languages are often confused with fingerspelling. Fingerspelling, it should be noted, is not a sign language, but “represents the standard written language through a series of hand configurations and movement” (Kyle & Woll 1994:3897). It is a manual alphabet representing the letters of a written language directly, that is, hand and finger configurations are used to indicate letters, “such as making a V with the index and middle fingers or an O with the thumb, and index finger” (Steinberg 1982:74). For instance, a word such as about would be signed letter by letter, a, b, o, u, t, following the English spelling. In this letter-by-letter method, words and entire sentences may be communicated whereas the order of letters is exactly the one that appears in the written language (Steinberg 1982:74). The main function of fingerspelling is to represent proper names, and ‘foreign’ words (Toom 1999:77), but it can also serve as a source for loan signs (Kyle & Woll 1985:124; for a discussion of the adaption of fingerspelled words into a sign language, see Kyle & Woll 1985:123-128, Engberg-Pedersen 1993:43-44). There are many different manual alphabets (and some syllabaries) in use over the world. Most of these are one-handed, e.g. American, Swedish, Russian, Estonian, Finnish, but, according to Kyle and Woll (1985:123-124), a two-handed system is used throughout Britain. (For examples of several manual alphabets, see Kyle & Woll 1985:123-128).

In addition to fingerspelling, special sign systems such as Seeing Essential English, Signing Exact English, Signed English, Signed Swedish, etc. are used for making a certain spoken language visible to hearing-impaired persons and for supporting the learning of it. In these systems, signs of sign languages designate an actual word meaning, but follow in signs the syntax of spoken words, and have supplementary signs for providing a particular grammatical device or feature which is not present in a sign language (Kyle & Woll 1985:34). For example, a sentence like I asked mum about my grandparents would have a sign for each English morpheme: I + ask + PAST + mum + about + my + grand + parent + PLURAL where tense and plural have separate signs.