Goldin-Meadow & Brentari's (G-M&B's) view of sign and gesture as a unified system is more revolutionary than it might outwardly seem. They part with the prevailing backdrop of contemporary language studies and embrace instead the long-neglected (if not poorly understood) tradition of conceiving language as a human activity instead of a chain of thing-like lexical pieces. The undergoing paradigm shift the authors call for draws much of its force from leaving behind the extant framework that has dominated language research for the past 300 years, namely, the “speech as written language” metaphor (Ingold Reference Ingold2007) that portrays vocal sounds and bodily signs as means of delivering stable word meanings. At the dawn of modernity, the invention of the printing press and subsequent availability of books progressively inspired the idea that human speech was ultimately a variant of printed word production: Our mind uses sound units (i.e., words) whose semantic content does not vary across contexts and users. Thus, contextual cues of utterance, melodic aspects of the voice (such as rhythm and prosody), and certainly the accompanying gestures were dismissed from the framework of language production and comprehension except as peripheral information sources.
Shortcomings of the metaphor of speech as written language become blatant whenever the meaning intended by the user differs from the lexical meaning. Co-speech gestures constitute one of these cases, since they modify or complement verbal expression. This leads to the problem of combining two purportedly antinomical types of meaning: lexical and gestural (McNeill Reference McNeill1992). G-M&B contribute to close the artificial theoretic divide between objective meaning bearers (words and signs) and idiosyncratic and contextually dependent meaning (gestures). This divide, pervasive since Saussure's definition of langue as the subject matter of modern linguistics, fails to reflect that language, first and foremost, emerges organically among the humans who use it. The meaning of words is wholly dependent on “their always insistent actual habitat, which is not, as in a dictionary, simply other words, but includes also gestures, vocal inflections, facial expression, and the entire human, existential setting in which the real, spoken word always occurs” (Ong Reference Ong1982, p. 46). Though the heritage remains tacit in the article, G-M&B are heirs to this contextual and organic conception of language; they reveal so by stressing that in real communicative acts, signs and words often behave like gestures and vice versa.
This tradition can be traced back to Wilhelm von Humboldt, who maintained that language “is no product (Ergon), but an activity (Energeia). Its true definition can therefore only be a genetic one” (Humboldt 1836/Reference Humboldt1988, p. 49). Language is not a set of prefabricated meaning units ready to be deployed for communication; one cannot replace the living utterance with tokens whose content has been described and preserved outside the stream of the real communicative situation. While the language-as-action tradition has known remarkable advocates during the 20th century – Peirce, Bühler, and Wittgenstein among the most prominent – it is in the writings of Heinz Werner, a lesser-known but crucial figure (Valsiner Reference Valsiner2005; Wagoner Reference Wagoner2013), where several of G-M&B's claims find the most relevant support.
Through a variety of experiments, Werner identified “physiognomic features” of words that grant them connotative values in a direct, immediate way (Werner Reference Werner, Barten and Franklin1978b). Just as when we see a face, we perceive words as hard or soft, sweet or dry, energetic or tired, and so on. Such physiognomic features of words do not correspond to their semantic representation; they rather include dynamic contents, imprecise but intermodal and synesthetic. Therefore, when the words “climbing” and “raising” are displayed centered on a monitor, they are perceived upwardly, while “falling” and “plunging” are perceived downwardly (Kaden et al. Reference Kaden, Wapner and Werner1955). For Werner, the physiognomic qualities are what make symbol formation possible: “We submit that even the most conventionalized units of speech – words and sentences – are still part and parcel of an articulatory process, bodily postural activity, which, through its dynamic features, links those conventionalized units to their referents’’ (Werner & Kaplan Reference Werner and Kaplan1963, p. 207). The pervasiveness of the physiognomic qualities consequently blurs the formal distinction between signifier and signified: “the ‘arbitrariness’ of linguistic forms seems to us to be unwarranted” (Werner & Kaplan Reference Werner and Kaplan1963, p. 16). This is particularly salient in the case of gesture, since “the material moment of bodily posture and motion, and the moment of meaning, are an indissoluble unity, i.e., a gesture cannot be significantly analyzed into a bodily displacement and a meaning arbitrarily related to it” (Werner Reference Werner, Barten and Franklin1978a; Reference Werner, Barten and Franklin1978b, p. 424). G-M&B seem to share this view, as they reveal when they approvingly cite Waugh's (Reference Waugh and Violo2000) indictment of arbitrariness in a text that calls for the reconciliation of form and meaning.
But how are we to harmonize the call for unity that pervades the target article with the authors’ claim that it is necessary to distinguish between imagistic gestures and categorical sign (or speech) for the sake of predicting learning potential? While there doubtless is enormous practical value in their experimental insight (Goldin-Meadow et al. Reference Goldin-Meadow, Shield, Lenzen, Herzog and Padden2012), it seems an insufficient ground on which to proclaim a univocal imagistic/gesture, categorical/sign mapping, and particularly so in a field that is just coming of age. Imagistic and categorical are not separate positions across a schism but rather the fuzzy endpoints in a continuum. The brain's responses to uni- and crossmodal mismatches are fundamentally the same (Cornejo et al. Reference Cornejo, Simonetti, Ibáñez, Aldunate, Ceric, López and Núñez2009; Kelly et al. Reference Kelly, Kravitz and Hopkins2004). As the physiognomic nature of words makes clear, imagistic properties are also to be found in linguistic formats. The imagistic-categorical distinction is not an empirical claim but an axiom that assumes that meanings are products instead of processes. It is a tenacious residuum of the inherited metaphor of speech as written language that persists in gesture studies.
Goldin-Meadow & Brentari's (G-M&B's) view of sign and gesture as a unified system is more revolutionary than it might outwardly seem. They part with the prevailing backdrop of contemporary language studies and embrace instead the long-neglected (if not poorly understood) tradition of conceiving language as a human activity instead of a chain of thing-like lexical pieces. The undergoing paradigm shift the authors call for draws much of its force from leaving behind the extant framework that has dominated language research for the past 300 years, namely, the “speech as written language” metaphor (Ingold Reference Ingold2007) that portrays vocal sounds and bodily signs as means of delivering stable word meanings. At the dawn of modernity, the invention of the printing press and subsequent availability of books progressively inspired the idea that human speech was ultimately a variant of printed word production: Our mind uses sound units (i.e., words) whose semantic content does not vary across contexts and users. Thus, contextual cues of utterance, melodic aspects of the voice (such as rhythm and prosody), and certainly the accompanying gestures were dismissed from the framework of language production and comprehension except as peripheral information sources.
Shortcomings of the metaphor of speech as written language become blatant whenever the meaning intended by the user differs from the lexical meaning. Co-speech gestures constitute one of these cases, since they modify or complement verbal expression. This leads to the problem of combining two purportedly antinomical types of meaning: lexical and gestural (McNeill Reference McNeill1992). G-M&B contribute to close the artificial theoretic divide between objective meaning bearers (words and signs) and idiosyncratic and contextually dependent meaning (gestures). This divide, pervasive since Saussure's definition of langue as the subject matter of modern linguistics, fails to reflect that language, first and foremost, emerges organically among the humans who use it. The meaning of words is wholly dependent on “their always insistent actual habitat, which is not, as in a dictionary, simply other words, but includes also gestures, vocal inflections, facial expression, and the entire human, existential setting in which the real, spoken word always occurs” (Ong Reference Ong1982, p. 46). Though the heritage remains tacit in the article, G-M&B are heirs to this contextual and organic conception of language; they reveal so by stressing that in real communicative acts, signs and words often behave like gestures and vice versa.
This tradition can be traced back to Wilhelm von Humboldt, who maintained that language “is no product (Ergon), but an activity (Energeia). Its true definition can therefore only be a genetic one” (Humboldt 1836/Reference Humboldt1988, p. 49). Language is not a set of prefabricated meaning units ready to be deployed for communication; one cannot replace the living utterance with tokens whose content has been described and preserved outside the stream of the real communicative situation. While the language-as-action tradition has known remarkable advocates during the 20th century – Peirce, Bühler, and Wittgenstein among the most prominent – it is in the writings of Heinz Werner, a lesser-known but crucial figure (Valsiner Reference Valsiner2005; Wagoner Reference Wagoner2013), where several of G-M&B's claims find the most relevant support.
Through a variety of experiments, Werner identified “physiognomic features” of words that grant them connotative values in a direct, immediate way (Werner Reference Werner, Barten and Franklin1978b). Just as when we see a face, we perceive words as hard or soft, sweet or dry, energetic or tired, and so on. Such physiognomic features of words do not correspond to their semantic representation; they rather include dynamic contents, imprecise but intermodal and synesthetic. Therefore, when the words “climbing” and “raising” are displayed centered on a monitor, they are perceived upwardly, while “falling” and “plunging” are perceived downwardly (Kaden et al. Reference Kaden, Wapner and Werner1955). For Werner, the physiognomic qualities are what make symbol formation possible: “We submit that even the most conventionalized units of speech – words and sentences – are still part and parcel of an articulatory process, bodily postural activity, which, through its dynamic features, links those conventionalized units to their referents’’ (Werner & Kaplan Reference Werner and Kaplan1963, p. 207). The pervasiveness of the physiognomic qualities consequently blurs the formal distinction between signifier and signified: “the ‘arbitrariness’ of linguistic forms seems to us to be unwarranted” (Werner & Kaplan Reference Werner and Kaplan1963, p. 16). This is particularly salient in the case of gesture, since “the material moment of bodily posture and motion, and the moment of meaning, are an indissoluble unity, i.e., a gesture cannot be significantly analyzed into a bodily displacement and a meaning arbitrarily related to it” (Werner Reference Werner, Barten and Franklin1978a; Reference Werner, Barten and Franklin1978b, p. 424). G-M&B seem to share this view, as they reveal when they approvingly cite Waugh's (Reference Waugh and Violo2000) indictment of arbitrariness in a text that calls for the reconciliation of form and meaning.
But how are we to harmonize the call for unity that pervades the target article with the authors’ claim that it is necessary to distinguish between imagistic gestures and categorical sign (or speech) for the sake of predicting learning potential? While there doubtless is enormous practical value in their experimental insight (Goldin-Meadow et al. Reference Goldin-Meadow, Shield, Lenzen, Herzog and Padden2012), it seems an insufficient ground on which to proclaim a univocal imagistic/gesture, categorical/sign mapping, and particularly so in a field that is just coming of age. Imagistic and categorical are not separate positions across a schism but rather the fuzzy endpoints in a continuum. The brain's responses to uni- and crossmodal mismatches are fundamentally the same (Cornejo et al. Reference Cornejo, Simonetti, Ibáñez, Aldunate, Ceric, López and Núñez2009; Kelly et al. Reference Kelly, Kravitz and Hopkins2004). As the physiognomic nature of words makes clear, imagistic properties are also to be found in linguistic formats. The imagistic-categorical distinction is not an empirical claim but an axiom that assumes that meanings are products instead of processes. It is a tenacious residuum of the inherited metaphor of speech as written language that persists in gesture studies.
ACKNOWLEDGMENT
The writing of this commentary was supported by the Chilean National Fund for Scientific and Technological Development (Grant FONDECYT 1141136).