Cognitive Linguistics

Please download to get full document.

View again

of 15
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
A concise contribution to the Handbook of Applied Linguistics summarizing the field of Cognitive Linguistics. Authors: Hans-Jörg Schmid and Friedrich Ungerer
Document Share
Document Tags
Document Transcript
  Routledge Handbook of Applied Linguistics, Ch. Cognitive Linguistics, Schmid & Ungerer, 2010-02-02, 2nd draft Please cite the published version from  1 Cognitive Linguistics Hans-Jörg Schmid and Friedrich Ungerer 1. INTRODUCTION Like many other notions in linguistics, the term Cognitive Linguistics  is used in a number of ways. What may be special about this notion, however, is that two competing and in many respects incompatible approaches to the study of language go by the same name. While these two approaches share the idea that linguists should consider psychological aspects of speak- ers’ knowledge about language (cf. the Latin cognoscere   ‘(get to) know’) rather than merely describe linguistic behaviour, they differ with regard to how they explain the nature and sources of this knowledge. The first view, very much associated with Chomsky and known as Generative Grammar   (cf. Chapter XY), sees knowledge about language  –   i.e., linguistic com- petence    –   as a very special human ability which is not, or only remotely, related to other cog-nitive faculties such as perception, attention or memory. The second view of Cognitive Lin-guistics takes a completely different perspective and emphasizes the experiential nature of linguistic competence. It is this approach, and its vision of explaining the cognitive founda-tions of linguistic structure and usage, that this chapter will be concerned with. In this ac-count, knowledge about linguistic structures is explained with recourse to our knowledge about the world, and it is assumed that language both reflects and contributes to shaping this knowledge. Introduced by linguists such as Fillmore, Lakoff, Langacker and Talmy in key  publications in the 1980s, this notion of Cognitive Linguistics is today represented, e.g., by the  International Cognitive Linguistics Association  (ICLA) and in the papers published in the  journal Cognitive Linguistics . 2. HISTORY AND KEY ISSUES   2.1 Categorization, prototype theory and basic levels An important starting-point of cognitive-linguistic thinking  –   which actually predates the term cognitive linguistics itself, which was not used before the early 1980s  –   was the empirical research into the nature of conceptual categories carried out by the anthropologists Berlin and Kay (1969) and the psychologist Rosch (1973, 1978). Studying the denotational ranges of  basic colour terms like red, blue  and  yellow  in a large number of languages, Berlin and Kay found that there was a surprising degree of agreement on what informants from different lin-guistic and cultural backgrounds considered as the best examples of red, blue , etc. For the  border areas, e.g., the range of colours from dark-red to purple or from a turquoise-like blue to green, there was much less agreement. Berlin and Kay referred to the areas on the colour spectrum which represented the best examples of basic colours as  focal colours , and Kay and McDaniel (1978) later demonstrated that physiological aspects of the visual apparatus were responsible for the observed inter-subjective and cross-cultural commonalities. This is an in-teresting and very straightforward example of how properties of linguistic units, in this case the meanings of basic colour terms, are influenced by other cognitive abilities, here percep-tion. Berlin and Kay’s work was taken up by Rosch and extended to other types of categ o-ries including geometrical shapes ( SQUARE ,  TRIANGLE ) as well as everyday concepts such as FURNITURE ,  VEHICLE ,  WEAPON  and others. What Rosch found was that just like in the case of colour categories, the members of these object categories could be rated on a goodness-of-example scale by informants in psychological tests. For example, informants agreed that cars and trucks were very good examples of the category VEHICLE , but rated skis, skateboards and  Routledge Handbook of Applied Linguistics, Ch. Cognitive Linguistics, Schmid & Ungerer, 2010-02-02, 2nd draft Please cite the published version from  2 elevators as very poor ones. Rosch introduced the term  prototype  for the best examples of categories and argued that they served as cognitive reference points for the storage and re-trieval of categories. This idea was complemented by the notion of  fuzzy boundaries  between categories (cf. Labov 1973), referring to the observation that conceptual categories such as CUP ,  MUG  and BOWL  are not separated by strict category boundaries, but seem to fade into each other, with objects possibly being named as cup  by some informants and as mug   by oth-ers. In short, rather than being subject to a checklist of necessary and sufficient features as suggested by structuralist semantics (and Aristotelian philosophy), conceptual categories are internally structured in terms of prototypes, good and less good members as well as fuzzy  boundaries to ‘neighbouring’ categories. This idea is one of the cornerstones of what is known as  prototype theory   of categorization  and, since these categories are labelled by words and have conceptual content, as  prototype semantics . Prototypes can be shown to differ from less typical members of categories with regard to the number (and nature) of attributes associated with them. For example, while cars and trucks are associated with crucial attributes of the concept VEHICLE   such as ‘used to transport people and things’, skis and skateboards can i n-deed be used as a means of transport but are much more strongly linked with attributes like ‘sports’ or ‘fun’.  Although the idea of prototype theory first came up in connection with superordinate categories such as FURNITURE ,  VEHICLE and  WEAPON , it soon emerged that the notion of proto-type is even more helpful when it comes to explaining basic level categories or   concepts ( BED  and TABLE ,  CAR and  TRUCK  ,  GUN and  KNIFE rather than  FURNITURE ,  VEHICLE and  WEAPON ,   etc). It is here that we find words which are short, morphologically simple, acquired early in ontogenetic development and introduced into discourse in unmarked contexts. As shown by Rosch et al. (1976), the members of basic level categories have a similar shape which lends itself to perception, and possibly representation, as a holistic gestalt. In addition, we interact with similar motor movements with members of basic level categories, e.g., we sit down on all types of chairs. Superordinate categories, on the other hand, rely on a different principle, also often subsumed under the label  prototype theory , the principle of family resemblances. As Rosch and Mervis (1975) showed, the seemingly different members of superordinate cate-gories such as  FURNITURE  or VEHICLE , rather than depending on large numbers of category-wide attributes as basic-level categories do, are held together by clusters of overlapping at-tributes, just like the members of one family will usually not all resemble each other but have certain sets of characteristics in common. Indeed, the notion of family resemblances had been invoked much earlier to explain the internal conceptual coherence of the category GAME  by the philosopher Wittgenstein (1958). 2.2 Frames, cognitive models and conceptual metaphors Conceptual categories are not only linked in memory with attributes associated with the cate-gory members, but also embedded in a huge conceptual network of more or less firmly stored knowledge structures. One type of these structures is known as  frames  and defined as ’ cogni-tive structures [...] knowledge of which is presupposed for the concepts encoded by the words ’  (Fillmore and Atkins 1992: 75). A classic and very influential example from the pre-cognitive-linguistic era is the so-called commercial transaction frame  (Fillmore 1977) pre-supposed by verbs such as buy, sell, pay  or cost  . The frame is described in terms of the frame components BUYER  ,  SELLER  ,  MONEY and  GOODS , and it is assumed that even though the verbs do not require all of these components to occur on the syntactic surface (cf. examples 1 and 2), mention of any of the verbs will invariably activate the whole frame.  Routledge Handbook of Applied Linguistics, Ch. Cognitive Linguistics, Schmid & Ungerer, 2010-02-02, 2nd draft Please cite the published version from  3 (1) The book [ GOODS ]  cost ten pounds  [ MONEY ]. (2)  Mary [ BUYER  ] bought an expensive book [ GOODS ]. In addition, depending on the verb chosen, certain components of the frame are highlighted to various degrees. For example, while the verb cost   draws attention to the GOODS  and the  MONEY  which fill the subject and object slots in the sentence (example 1), the verb buy  high-lights the BUYER   (subject) and the  GOODS  (object) (example 2). Obviously, this has to do with  putting a certain perspective on a scene and deploying attention to certain aspects, a cognitive ability reflected in other areas of language we will look at later. While frames are conceived of as somehow delimitable knowledge structures, other types of cognitive models are less restricted. Lakoff, for instance, in his treatment of idealized cognitive models   (1987), takes up Fillmore’s discussion of the noun bachelor   and argues that this concept only makes sense within an idealized cognitive model of a society whose mem- bers share certain expectations as to the institution of marriage. Ungerer and Schmid (2006: 49), who opt for a deliberately comprehensive definition of cognitive models as ‘ stored repre- sentations that belong to a certain field”, provide the ex ample ON THE BEACH , which ‘i n- cludes’ closely interrelated person and object categories such as PEOPLE , SAND , SHELLS , BUCKET  as well as action and event categories, e.g., SWIM , SUNBATHE , BUILDING A SANDCAS-TLE  and others. While it may be criticised that these descriptions of cognitive models are to-tally subjective, open-ended and apparently of a somewhat unscientific ad-hoc nature, it may well be the case that this is exactly how our minds work. The psychological reality of these knowledge structures can be tested with priming experiments and other tests and gleaned from language use, e.g., when speakers use definite noun phrases with anaphoric reference to com- ponents of frames that are not explicitly mentioned but still activated, cf. the NP the sea  in example (3) (3) We spent some time on the beach yesterday. The sea was very rough . Cognitive models are not individual, purely subjective knowledge structures, but presumably shared to a large extent by the members of a culture and therefore also seen as cultural mod-els . It goes without saying that frames and cognitive as well as cultural models are also based on our experience of the world around us. One particularly fruitful early field in Cognitive Linguistics which relies on the idea of cognitive models (or domains ) is the conceptual theory of metaphor, introduced in the pio-neering book by Lakoff and Johnson (1980; 2 nd  ed. 2003). In a nutshell, this theory claims that conventionalized metaphorical expressions such as (4) or (5) are by no means dead metaphors  but surface manifestations of deeply entrenched underlying mappings of one domain, the  source domain , onto another, the target    domain . (4)  He got all steamed up . (5)  I almost exploded. In these two examples the cognitive model of a hot fluid in a container is mapped onto the concept of anger, yielding a conceptual metaphor dubbed ANGER IS A HOT FLUID IN A CON-TAINER  . Other examples of conceptual metaphors discussed by Lakoff and Johnson include AN ARGUMENT IS A JOURNEY  (cf  . 6 and 7), IDEAS ARE OBJECTS  (cf. 8 and 9) or COMPANIES ARE PLANTS  (cf. 10 and 11). (6) We have arrived at  a disturbing conclusion. (7)  Do you follow  my argument  ?  Routledge Handbook of Applied Linguistics, Ch. Cognitive Linguistics, Schmid & Ungerer, 2010-02-02, 2nd draft Please cite the published version from  4 (8) We dropped  the idea. (9) They canvassed  a new idea . (10) The company has several  branches. (11) We’ve been growing  continuously over the past years . From an experiential point of view, it is important to emphasize that conceptual metaphors typically use a more tangible and concrete domain as a source, which is mapped onto a more abstract domain in need of conceptual structure. 2.3 Figure and ground, prominence and salience Another experiential aspect related to the cognitive abilities of perception and attention is the gestalt psychological principle of figure and ground. This principle suggests that when view-ing a given scene we will invariably single out certain elements as prominent figures while relegating others to the less prominent ground. For example, looking up into the dark sky at night, we inadvertently select the moon as a salient figure which stands out from the black ground behind it. Arguably, a reflection of this perceptual principle can be identified in the structures of linguistic utterances describing such a scene: while example (12) sounds fairly natural, as it highlights the salient figure in the more prominent syntactic slot of subject, a complementary utterance like (13), although an equally true depiction of the scene, would be decidedly weird, or at least marked. (12) The moon is in the sky.  (13) The sky is around the moon. Perceptual stimuli which are likely to be selected as figures tend to be smaller, more movable, geometrically simpler, more dependent and more prominent (once perceived) than typical ground entities (Talmy 2000: 315); in addition, figures tend to be more relevant, and thus both  perceptually and conceptually more salient, for the language user, and this is all reflected in degrees of prominence awarded to the linguistic material referring to these salient entities in actual utterances. A large part of the early research into figure-ground phenomena focused on preposi-tions (Brugman 1981, cf. Lakoff 1987: 416ff.). Central to these and later studies on preposi-tions is the notion of image-schema , defined as ‘ relatively simple structures that constantly recur in our everyday bodily experience: containers, paths, links, forces, balance, and in vari-ous orientations and relations: up-down, front-back, part-whole, center-periphery, etc. These structures […] are directly meaningful, first, because they are directly and repeatedly e xpe-rienced because of the nature of the body and its mode of functioning in our environment ‘ (Lakoff 1987: 267  –  268). It is common practice to refer to the figure in these schemata as tra- jector   and to the ground as landmark  .   Probably the most powerful aspect of these schematic structures is the potential of schematic mental imagery for specific context-sensitive elabora-tions. This explains, for instance, the wide range of semantic variation for the preposition over   illustrated in examples (14) to (18): (14) They have a horseshoe over their door.  (15) The dog jumped over the fence . (16) Carl cycled over the bridge . (17) The village clouded over  . (18) The wall fell over.  Routledge Handbook of Applied Linguistics, Ch. Cognitive Linguistics, Schmid & Ungerer, 2010-02-02, 2nd draft Please cite the published version from  5 Example (14) describes a stative configuration of a trajector above a landmark, which is con-sidered as representing the fundamental image schema associated with over   in a recent treat-ment by Tyler and Evans (2003: 66). (15) and (16) represent dynamic scenes in which the trajector moves through a stage that corresponds to the central schema, with the trajector be-ing in contact with the landmark in (16). In (17), the trajector is encoded in the verb clouded   and covers the landmark, while in (18), trajector and landmark coincide but perform a move-ment similar to the trajector in (15). Yet image schemas can also be metaphorically extended and then account for the mo-tivation behind figurative, non-spatial or non-visual experiences. For example, Tyler and Ev-ans (2003: 85  –89) trace the meaning ‘excess’ encoded by over   in (20) to more concrete mean-in gs like the one exemplified in (19), and the ‘completion’ sense in (22) to uses of type (21):  (19) The arrow flew over the target and landed in the woods. (20)  Many students wrote over the word limit.  (21) The cat’s jump is over.   (22) The film/game/match is over. Image-schemas and their elaborations and metaphorical extensions thus contribute to account-ing for meaning relation in the complex polysemy networks associated with linguistic ele-ments such as in, over, out   or up , which function, among other things, as prepositions, parti-cles and prefixes in English. Unlike in conventional dictionary entries, which simply list meanings of lexemes, the motivations and links between the wide range of senses become  plausible  –   an effect on the internal conceptual co herence of these ‘radial categories’, which is sometimes considered a further amendment to the prototype theory of meaning introduced in the section on categorization above. 3. MORE RECENT DEVELOPMENTS 3.1 Prototype theory, basic levels and entrenchment Though by no means uncontroversial, prototype theory has gained a firm place in linguistic theorizing. A substantial part of the recent discussion of the prototype model of categorization has revolved around the issues of the theoretical status and cognitive reality of prototypes. This has to do with the question as to whether the results of goodness-of-example ratings are  basically just a superficial effect of the rating task (e.g., with low ratings for an ostrich as a  bird), or whether they reflect a marginal membership of the subcategory OSTRICH  within the category BIRD . Croft and Cruse (2004: 79  –  81) insist on the importance of the distinction, stressing that while OSTRICH may indeed be a poor example of the category BIRD , it is still undoubtedly a fully-fledged member. Taylor (2003), like Ungerer and Schmid (2006: 55  –  56), makes a distinction between folk and expert models of categories and claims that everyday models corresponding to discrete, hard-and-fast expert categories can still show prototypical-ity effects and fuzzy boundaries. Taylor also transfers the notion of prototypicality to techni-cal categories in linguistics, e.g., in the area of phonology and morphology. The notion of basic level has recently come to be viewed as just one manifestation of the more general cognitive process (and product) of entrenchment   (Langacker 1987: 100, 2008: 16  –  17; Geeraerts et al. 1994; Schmid 2007). For example, telling a story of a dog chas-ing a cat, the terms that will first come to mind are precisely these basic-level terms, dog   and cat  , rather than superordinates such as mammal   or subordinates like retriever   and  ginger cat  .
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks