Towards a Cognitive–Semiotic Typology of Motion Verbs
Viktor Smith2013, The Construal of Spatial Meaning
Setting the scene-the cognitive semiotics of motion The present paper argues that the lexico-grammar of spatial Motion (as a supercategory for dynamic movement and static location, cf. Talmy 1985) cannot be understood except as an integral part of the semiotic triad of reality, mind, and language. M otion in language should thus be explained on the basis of the (Gestaltist) psychology of motion in perception, in that language 'structures' the mind's construction of motion in reality. Accordingly, the typology of motion verbs is based on an experientially founded typology of motional situations in mind. A mental motional situation is perceptual, or 'pictorial': Human beings perceive motional situations in reality by forming (concrete) 'pictures' of them with diverse figureground constellations-and recognize them as belonging to different categories (according to stored percepts). There are two kinds of picture, viz. static, or 'stable', and dynamic, or 'unstable', roughly according as the figure is static or dynamic. Furthermore, we seem to be able to construct only one situational picture at a time. A single situational picture is a simple mental Situation-a stable picture is a 'state', and an unstable picture an 'activity'. So far the notion of M otion has been Perceptual. Now, it goes without saying that the 'mentality' of Situations involves much more than simple perceptual Situations, in that situations may be conceived of as possibly integrated with one another into 'complex' Situations. A "snapshot" of what at first sight might seem to be only a state or an activity may thus show out to be the endpoint or the starting point "window", respectively, on an integrated, complex Situation involving an Activity and a State, what will be called an Action. In the first case, the State in focus would be preceded by a causal Activity; in the second case the Activity in focus would be succeeded by a resultant State, in the normal course of events. The connection between the two simple Situations in a complex actional Situation is a general relation of telicity, the causal Activity tending to actually eventuate in the resultant State. The state-focused Action will be termed an Event, whereas an activity-focused Action will be termed a Process. Illustrating this, we may conceive of a scenario where I am sitting alone in the drawing room, then leave for the kitchen and come back, and lo and behold, you are sitting there! This may be conceived of as a M otion Situation, viz. a M otion Event, where you are sitting here as a result of your, say, returning home from work, and I may second it by the utterance Nå, du er kommet hjem fra arbejde 'oh, you've come home from work'. In this case the motion for me was only conceptual, in that I didn't see, or otherwise witness it, but only inferred it. We may thus talk about Conceptual motion in such cases. When now turning to language (as a system) and the typology of motion verbs in the mental lexicon, we must add the Sign Vehicle, i.e. the phonological expression, as a representation of Percean Firstness. The linguistic Sign Object (Secondness) and Sign Interpretant (Thirdness) then recall the mental perceptual and conceptual structures, respectively, just mentioned. So the sign contents are twofold , the linguistic cognitive-semantic domain being bipartitioned into an (abstract) perception-based 'imaginal' representation (cf. Spatial Structure in Jackendoff 2002) and an (abstract) conception-based 'ideational' representation (cf. Conceptual Structure in Jackendoff 2002). 1. Background, aims, and scope 1.1 Lexicalization typology M otion event research has grown into a well-established and highly productive field. Its theoretical cornerstone are the classic studies by Talmy (1975, 1985; for further refinements, see 2000: 25ff.), supplemented by works primarily by Slobin (e.g. 1996a/b; 2004a/b), but also by others (for an overview, see M ora Gutiérrez 2001). Despite the overwhelming amount of specific works within motion event research and despite the seemingly growing awareness of the need for a more fine-grained, less schematic approach than the Talmy-Slobin framework, the core assumptions and variables of the framework nevertheless are still upheld. Talmy's basic assumption is that even though people's pre-linguistic conceptualization of e.g. a directed M otion Situation appears to be universal-involving the same fundamental components to be lexicalized (apart from Figure and Ground, Motion itself, Manner of M otion, or Cause, and Path (i.e. trajectory), the ways of linguistically lexicalizing it in different languages are not the same because not all the components are able to be colexicalized in the same (verbal) morpheme in a major lexicalization system (Talmy 1985: 76): apart from cases where only M otion is lexicalized in the verb, as in English move, either the M anner component co-lexicalizes with the M otion component in the verb, leaving the Path behind to be lexicalized in a so-called Satellite, as in M anner languages, or it is the Path component that is lexically 'incorporated' into the verb, in so-called verb-framed or Path languages, whereby the M anner component becomes secondary, left for optional expression in a con-verb or adverb. Thus, we have a nice binary typology of major lexicalization patterns, and derivatively of languages, in that it is assumed that at least most languages fit into one of these types: Manner (or, satellite-framed) languages, like e.g., Danish, Swedish, English, German, Russian, and Chinese, where only the Manner of motion is lexicalized in the verb root together with M otion, while the direction or Path of motion is explicated elsewhere when
