Battison’s dominance condition is theoretically somewhat

Battison’s dominance condition is theoretically somewhat The Weak Hand Rule: A Reformulation of the Symmetry and Dominance Conditions Rachel Channon University of Maryland While most two-handed signs have the same handshape on both hands, some have different handshapes. For example, in the ASL sign CREDIT-CARD, the strong hand has a fist shape, while the weak hand has an open flat handshape. An important question in sign linguistics has been why the shape of the weak hand varies. I argue here that it is primarily controlled by phonetic factors interacting with phonology, specifically weak hand place features. Battison’s Symmetry Condition (1978) states that when both hands move independently, they will have the same handshape, place, orientation and movement. His Dominance Condition states that if the two hands have different handshapes, then the weak hand will be passive and will have one of 5 handshapes: 1) TIMRP-extended with fingers spread or unspread, and the thumb in various postures (also called a B or 5 handshape), 2) TIMRP-curved (C-hand), 3) TIMRP-ring: thumb and fingers touching at tips (O-hand), 4) TIMRP-closed: a fist with the thumb either restraining the fingers or beside the fingers, (A or S hand), and 5) I-extended: the index finger extended from the fist (G or 1-hand). Together, these two conditions create the following sign categories for two-handed signs: Type 1: Signs that obey the Symmetry Condition Type 2: Non-symmetrical two-handed signs where both hands have the same handshape Type 3: Non-symmetrical signs with different handshapes. The Symmetry and Dominance Conditions have generally been assumed to be part of the phonology of sign languages, but there are several problems. First, these constraints appear to be universal. Although these rules were originally formulated for ASL, they appear to hold for all studied sign languages, including NGT and JSL: no researcher to date has proposed any sign language as a counterexample to these two very famous conditions. This suggests that either these constraints are evidence of some innate phonological knowledge, or more probably, that they are the surface manifestations of inherent phonetic constraints on the use of two hands. The Symmetry Condition has been argued to be a cognitive, non-linguistic, constraint by Van der Gijn, Kita and van der Hulst (to appear) because both signs and similar two-handed gestures of non-signers appear to fall under this condition. Second, the two constraints are not explanatory: Why are both hands the same? Why are there only these five possible weak hand shapes? Third, if a sign falls in the domain of the Dominance Condition, the signer is still left with five alternative handshapes, with no way of choosing between them. Fourth, there are some exceptions to the Dominance Condition. For example, the ASL sign SKIP-CLASS strikes the flexed middle finger of the weak hand (M-bent) with the extended index of the strong hand. This bent middle finger handshape is completely different from any of the weak handshapes listed in the Dominance Condition. A number of signs have been mentioned in the literature as having an extended index and middle. For example, in CHOOSE the strong thumb and index together contact in turn the index and middle of the weak IM-extended hand. Other exceptions have also been mentioned for various languages in recent discussions on the Sign Language listserv. These two conditions can be simplified, unified, and explained with the Weak Hand Rule: (1) Weak Hand Rule: The weak hand assumes the phonetically easiest handshape which does not conflict with any underlying weak hand place features. I argue here that there are five explanations for the shape of the weak hand: 1) a cognitively based preference for both handshapes to be the same; 2) an articulatory requirement for access to the weak hand place; 3) a perceptual requirement (which may also be articulatory) that the weak hand place be sufficiently prominent; 4) a small number of signs appear to require a place shape feature, and 5) an even smaller number appear to have two or more simultaneously expressed places. First, for most signs, the weak hand in a two handed sign does not conflict with the phonology because there is no conflicting weak hand place. This is the case for all signs picked out by the Symmetry Condition (Type 1 or balanced signs). Such signs always have the same handshape on both hands. Most of these signs are made in neutral space and probably have no underlying place, but even when they are made on the body, these places have no requirements that conflict with using the same handshape. In Type 2 signs, although there is a weak hand place, it does not conflict with the strong handshape, so the handshapes will also be the same. Some ASL examples are RIGHT (I-extended), and UMBRELLA (TIMRP-closed). The second through fifth explanations determine the weak handshapes of type 3 signs. Second, using the same handshape as the strong hand may conflict with physical access to the underlying weak hand place. The resolution of this conflict is to use the phonetically easiest handshape that allows access. Most commonly this will be TIMRP-extended. For example, in CREDIT-CARD, the weak hand place is the palm side of the hand. If the weak hand were a fist like the strong hand, then it would be impossible to touch the palm side of the hand. The signer therefore selects the phonetically easiest handshape which does allow access, which is TIMRP-extended. Third, the weak hand place must be visible or prominent. Using the strong handshape may interfere with place visibility. If so, the phonetically easiest handshape which still highlights the underlying place will be used. Fourth, some signs have a place shape feature, such as [curved] or [ring] which must be expressed. In GROW, the strong hand moves up through the weak TIMRP-curved hand. The curved handshape does not provide higher prominence or greater accessibility for the palm side of the hand (the apparent weak hand place), but rather the reverse. However, it certainly does express an underlying feature [curved]. Fifth, a few signs have more than one underlying weak hand place on two different digits. ASL examples are CHOICE (index and middle), and THEN (thumb and index). The Symmetry and Dominance Conditions appear to express phonological facts about sign languages, but when examined more carefully, they turn out to be explainable on primarily phonetic grounds. Little about the weak hand is actually a part of the phonological representation for the vast majority of two-handed signs. For a relatively small subset of two-handed signs, a weak hand place must be specified, but even for these signs, many characteristics of the weak hand turn out to be phonetically determined. References Ann, Jean. 1996. On the relation between ease of articulation and frequency occurrence of handshapes in two sign languages. In: Lingua. Special issue on "Sign linguistics: Phonetics, phonology and morpho-syntax" 98: 1-3 (1996) - pp. 19-42. Battison, Robbin. 1978. Lexical borrowing in American Sign Language. Silver Spring, MD: Linstok. Gijn I. van, Kita, S. and H.G. van der Hulst (to appear). The non-linguistic status of the Symmetry Condition in signed language: Evidence from a comparison of signs and spontaneous co-speech gesture.