Cognitive Linguistics And Idioms

CHAPTER 2 COGNITIVE LINGUISTICS AND IDIOMS

2.1 Cognitive linguistics. Main principles

Linguistic though and practice are embedded by cognitive linguistics, which is a modern school focused on analyzing the connection between the human language, socio-physical experience and the mind. Cognitive linguistics appeared in the 1970’s. During that period of time, the disciplines of linguistics and philosophy were dominated by the formal approaches to language. This modern school was highly influenced by cognitive psychology, which also emerged during the 1960’s and 1970’s, even though at first it had a philosophical nature. This feature can be seen in the works about human categorization realized by Charles Fillmore in the 1970’s and George Lakoff in the 1980’s.

In the 1970’s and early 1980’s, research was dominated by a large number of scholars who lived in the USA. However, in the 1980’s, the interest towards cognitive linguistics rose among those who inhabited the northern continental Europe. In 1989, due to the large interest of a significant number of scholars, the International Cognitive Linguistics Association was formed, followed by that of the Cognitive Linguistics journal, a year later. Ronald Langacker, a pioneer in the domain of cognitive linguistics, regarded the events that took place in 1989 and 1990 as “marking the birth of cognitive linguistics as a broadly grounded, self conscious intellectual movement”(Langacker 1991:15)

The best term to describe cognitive linguistics is “movement” because it does not stand as a single theory. It is complimentary, containing a pair of commitments as well as guiding principles. The two fundamental commitments are: the generalization and cognitive one, while the two main branches are: cognitive semantics and cognitive approaches to grammar.

“The Generalisation Commitment represents a dedication to characterizing general principles that apply to all aspects of human language” (Evans 2007:3). Its main focus is to find the existing broadcast generalizations.

In the case of formal linguistics, phonology, semantics, pragmatics and morphology have little basis for generalization because its main goal is to form language by means of explicit mechanical devices. In this case, formulations are created with the help of formalisms which have their origin in mathematics and logic. Naom Chomsky’s work is the most notable in formal linguistics.

Cognitive linguistics on the other hand considers that syntax, semantics and phonology should be treated differently. ”However, given the Generalization Commitment, cognitive linguistics does not start with the assumption that the modules or subsystems of language are organized in significantly divergent ways, or indeed that wholly distinct modules even exist”(Evans 2007:4). As a result, this commitment tries to investigate openly the way in which a common set of human cognitive abilities determine the apparition of different aspects of linguistic knowledge. Concrete consequences about studies of language are determined by the Generalization Commitment. Cognitive linguistics attempts to re-use methods and explanations across aspects of language, which have been previously successful, while also focusing on what is common among these aspects.

Furthermore, when studying language, cognitive linguistics adopt a “vertical” approach. Regarding the composition of language, it is considered as being formed of a set of distinct layers of organization. On one hand, the sound structure represents words created by sounds while on the other hand the syntactic structures contain the resulting words. “Vertical approaches get a richer view of language by taking a vertical slice of language, which includes phonology, morphology, syntax and of course a great amount of semantics”(Evans 2007:4). Vertical language is considered to be more complex compared to the horizontal one, keeping in mind the fact that it is more textured while at the same time offering possible explanations.

The Cognitive Commitment is the next one we are going to focus on. It offers a characterization of the general principles of language which relate with the information already provided by other disciplines about the brain and mind..This commitment provides the feature “cognitive” to cognitive linguistics.

“Just as the Generalization Commitment leads to the search for principles for language structures that hold across al aspects of language, in a related manner, the Cognitive Commitment represents the view that principles of linguistic structure should reflect what is known about human cognition from the other cognitive and brain sciences particularly psychology, artificial intelligence, cognitive neuroscience and philosophy”(Evans 2007:5)

As a result, Cognitive Commitment considers that models of language ought to reflect what is already known about the human mind and not specific formalisms.

This commitment contains several concrete ramifications. Structures or processes that do not respect the existing properties of the human cognitive system are not part of linguistic theories. Furthermore, already known properties of human cognition are used to explain the language phenomena. These modes are more willing compared to the ones that are formed from a priori simplicity metrics. It is absolutely necessary for a researcher in the domain of cognitive linguistics to find convergent proof of the cognitive reality “of components of any proffered model of explanation”(Evans 2007:5).

The best developed areas of cognitive linguistics are: cognitive semantics and the cognitive approach to grammar. Cognitive semantics studies the conceptual structure and meaning construction. It is also concerned with the relationship between experience, the conceptual system and the semantics encoded by language. In addition language is used as a means of investigating the cognitive phenomena. “Consequently, research in cognitive semantics tends to be interested in modeling the human mind as much as it is concerned with investigating linguistic semantics”(Evans 2007:5). On the other hand, the cognitive approach to grammar deals with the process of modeling the language system. This approach used the conclusions resulted from cognitive semantics in order to model the system.

The guiding principles of cognitive semantics are the following: conceptual structure is embodied, semantic structure is conceptual structure, meaning representation is encyclopedic and meaning construction is conceptualization. In the following section, these principles are going to be shortly presented.

According to the first principle, we have a species-specific view of the universe. This fact is determined by the nature of our bodies. As a result, our constructions of the world are determined by the nature of our embodiment. This process is clearer in the realm of colour. The human visual system has three kinds of photoreceptors while other organisms have more. “Having a different range of colour channels affects our experience of colour in terms of the range of colours accessible to us along the colour spectrum”(Evans 2007:5). Humans, unlike rattle snakes cannot see in the infrared range. The nature and range of our visual experience is determined by the nature of our visual apparatus.

The second principle states that language refers to the concepts that are produced in the mind of the speaker. The equality sign can be put between semantic structure and conceptual structure, but they are not identical. Semantic structure refers to the meanings that are associated by convention with words.

The third principle deals with the encyclopedic nature of semantic structure. As a result, lexical concepts are not dictionary views, and provide access to large amounts of knowledge relating to a particular concept. In affirming this, however, we cannot deny that words are associated with conventional meaning. Cognitive semantics considers this conventional character as being a prompt for the process of meaning construction in which an appropriate interpretation is chosen against the context of the utterance.

The final guiding principle suggests that language does not encode meaning by itself, and, as mentioned before, words are only prompts used to form meaning. So, meaning is the result of what happens at the conceptual level. The process of constructing meaning is similar with conceptualization in which, in order to recruit background knowledge, linguistic units are used as prompts. Meaning is considered a process not just a thing that can be packed by language.

In the next section, some of the most significant theories in cognitive semantics will be briefly discussed: image schema theory, the categorization and idealized cognitive modes, the cognitive lexical semantics and finally the mental spaces theory.

Mark Johnson developed the image schema. In his book entitles “The Body in the Mind”, he considers that image schemas help in the manifestation of the embodied experience. The concepts: CONTACT, CONTAINER and BALANCE are connected to the human preconceptual experience in which the human body structures the experience of the world in a direct way. “These image-schematic concepts are not disembodied abstractions, but derive their substance , in large measure, from the sensory-perceptual experiences that give rise to them in the first place”(Evans 2007:9).

Jean Mandler, a developmental psychologist, considers that children from an early age are attracted by objects and spatial display and that in this way, they associate them with similar experiences and as a result, they form patterns. The CONTAINER image is a configuration in which an entity is supported by another one which contains it.

Lakoff and Johnson considered that concepts of this type which are rudimentary embodied, can provide the conceptual building blocks and that in this way, structures can be provided in order to build more abstract concepts. The fundamental concept CONTAINER helps us understand abstract concepts like LOVE.

Johnson suggests that due to the fact that activity is constrained by containers, state of beings are conceptualized and therefore understood.

Mandler notes that “one of the foundations of the conceptualizing capacity is the image schema, in which the spatial structure is mapped into the conceptual structure”(Mandler 1992:591), further suggesting that “basic recurrent experienced with the world form the bedrock of the child’s semantic architecture, which is already established well before the child begins producing language”(Mandler 1992:597). In conclusion, experience helps form the basis of the humans’ fundamental concepts.

The second theory is related to Lakoff’s theory of Idealised Cognitive Modes. Lakoff’s ICMs are considered to represent background knowledge structures which have a stable character. He further sustained that categorization is related to ICMs, which are mentally stable representations about the world. In addition, cognitive processes such as categorization and reasoning are guided by ICMs. Different sources can cause the apparition of typicality effects, and they can be determined by the discrepancies between ICMs and how certain concepts are understood. These effects can also arise because of the sorts of ICMs that people have access to. This can appear when an exemplar represents an entire category and it is called metonymy. Typicality effects that appear in this way are called by Lakoff metonymic ICMs.

An example of this type of effect is the stereotype HOUSEWIFE-MOTHER. In this case, the mother stays at home, takes care of the children and does not have a job. According to this stereotype, mothers nurture children and to do so, they have to stay at home.

The two key commitments of cognitive linguistics are embodied by ICMs: the Generalisation and Cognitive one. Lakoff used the new findings from cognitive psychology and tried to construct a model of language that suited the findings. He used principles common to the linguistic and conceptual phenomena and finally laid the foundation for the cognitive approach to language.

The following theory in cognitive linguistics that we are going to present is the cognitive lexical semantics.

Lakoff’s theory of ICMs caused significant consequence in the domain of cognitive lexical semantics, which deals with the way in which the relationship word-meaning is managed by cognitive semantics. According to cognitive lexical semantics, lexical items are conceptual categories. As a result, a word represents meanings that are related but at the same time different and have typicality effects. Lakoff, words are categories that can be changed and researched by using the theory of ICMs. “In particular, Lakoff argued that lexical items represent the type of complex categories he calls radial category”(Evans 2007:15). A prototype is used in order to structure a radial category while its category members are related by means of convention to the prototype and not determined by predictable rules. Word meanings are senses kept in the mental lexicon.

Lakoff’s theory mentioned above has been highly influential, but still a number of criticisms arose especially regarding the “full-specification” view, according to which a large number of distinct senses can be designated to a lexical item. These criticisms were determined by the lack of clear methodological principles regarding the different meanings of an item. However, recent works have provided such a methodology. Due to these works, especially in the use of corpora, cognitive lexical semantics has made serious progress in the analysis of lexical categories.

The final theory is the mental spaces one, which is focused on meaning construction. It was developed by Gilles Fauconnier in his two books entitled “Mental Spces”(1985) and “Mappings in Thought and Language”(1997). Recently, he has extended the theory, resulting in the apparition of a new framework theory called Conceptual Blending Theory. These two try to analyse the hidden conceptual aspects of meaning construction. According to them, meaning is formed with the help of prompts that are provided by language. These prompts are underspecified.

Fauconnier suggests that two processes are invoved in the construction of meaning: the construction of mental spaces and the establishment of mapping between those mental spaces. The local discourse context guides the mapping relations and because of this, the meaning construction is always context-bound. According to him, mental spaces are “partial structures that proliferate when we think and talk, allowing a fine-grained partitioning of our discourse and knowledge structures”(Fauconnier 1997:11). This theory suggests that meaning is divided by mental spaces into separate conceptual regions in the moment we think and talk.

Mental spaces encapsulate specific kinds of information, which are formed by means of generalized linguistics and information acquired through pragmatic and cultural strategies. But, mental spaces determine the apparition of temporary conceptual regions because they are constructed “on-line”. In addition, they are made in order to accomplish purposes that are specific to the ongoing discourse.

Linguistic expressions have meaning potential and they do not just encode meaning, but represent partial building instructions, through which mental spaces are formed. “Of course, the actual meaning prompted for by a given utterance will always be a function of the discourse context in which it occurs, which entails that the meaning potential of any given utterance will always be exploited in different ways depending upon the discourse context”(Evans 2007:18).

Scope builders set up mental spaces. These space builders can be prepositional phrases, adverbs (really, probably) and subject-verb combinations followed by an embedded sentence (Mary thinks that…). They can either prompt for the construction of a new mental space or turn the attention to the mental spaces that have been previously formed.

Elements are contained by mental spaces that can be entities which already exist in the conceptual system, or entities that are formed on line. Already existing knowledge structures, frames and ICMs internally structure the mental spaces. Mental spaces which have been constructed are linked to the others established during the discourse.

This theory shows how the viewpoint changes during discourse and at the same time offers intuitive solutions to referential problems to which formal semantics has not found a solution.

Bending Theory is related to Mental Spaces theory. This is determined by its concern with the dynamic aspects of meaning construction. However, it also has some differences because Blending Theory considers that meaning construction involves the integration of structure from across mental spaces, thus resulting the emergent structure. This structure contains more than the sum of its parts. Researchers approving of this theory suggest that the process of blending also referred to as conceptual integration, is a general and basic cognitive operation.

The second area of cognitive linguistics that we are going to focus on is the cognitive approach to grammar. This type of cognitive linguistics has a diversified set of foci and interests. Ronald Langacker, who is a cognitive linguist, tries to find the cognitive mechanisms and principles that are fundamental for the properties of grammar. Leonard Talmy on the other hand, tries to develop his model. Construction grammar theories are concerned with the characterization of the linguistic units and constructions that are found in a grammar. Other linguists focus on explaining the process of grammaticalisation.

Cognitive approaches to grammar have two guiding principles: the symbolic thesis and the usage-based thesis.

According to the symbolic thesis, the linguistic unit, also called form-meaning pairing is the fundamental unit of grammar. Langacker called it “symbolic assembly” and argued that it had two poles: the semantic one (meaning) and the phonological one (sound). Ferdinand de Saussure originated the idea that language has an essentially symbolic function. He also sustained the fact that the symbolic unit is the fundamental unit of grammar.

Cognitive approaches to grammar do not only investigate the aspects of grammatical structure, but it contain the entire inventory of linguistic units defined as form-meaning pairing. Meaning and grammar are interdependent and complementary. The cognitive approach to grammar assumes the study of the units of language, while the cognitive approach to semantics attempts to understand how this linguistic system in related to the conceptual system.

The usage-based thesis considers the speaker’s knowledge of language (mental grammar)to be created by the abstraction of symbolic units from instances of language use that are situated. According to this theory, there is no difference between knowledge of language and use of language because knowledge of language if knowledge of how language is used.

2.2 Conceptual structures in cognitive linguistics

Cognitive linguistics tries to classify the systematic processes through which people are able to understand abstract concepts. Lakoff’s Idealised Cognitive Model suggested that we organize our knowledge of the world through cognitive mechanisms. The structuring principles of ICMs are: metaphoric and metonymic mapping, propositional structure and image-schematic structure. The image schemas are the result of our sensory motor experience, while propositional ICMs are determined by predicate-argument descriptions. At the same time metaphors are sets of correspondence between discrete conceptual domains while metonymies are one correspondence mappings. On the other hand, Ruiz de Mendoza (1996) has sub-classified the ICMs into: operational (metaphor and metonymy) and non-operational (frames and image-schemas) cognitive modes. By doing this distinction, he presents the processual nature of metaphor and metonymy. These two concepts work on the basis of propositional or image-schematic ICMs.

In the following part of this subchapter, we are going to focus on the conceptual metaphors and metonymies.

In the cognitive linguistic view, conceptual metaphor is viewed as a process which helps humans understand concepts and domains in terms of other concepts. Conceptual metaphors have a very significant role in language due to the fact that they have central roles in the way in which we deduce concepts from one semantic area. In the cognitive tradition, metaphor is seen as a matter of thought and as a result it has a conceptual character, not a linguistic one. Conceptual metaphors are divided into three categories: oriental (through which people associate abstract concepts with experiences that involve spatial orientation just to help them understand concepts), ontological (help people to speak about abstract entities in terms of substances, objects without mentioning the type of the object or substance), and finally structural (presuppose the interrelation between two concepts, in which one is more abstract than the other).

Personification is a specific type of ontological metaphor. It attributes human characteristics to things and in doing so; it implies an understanding of non-human entities in terms of human beings.

These three types of metaphors are determined by the source domain. Conceptual metaphors present several differences from metaphorical expressions. The latter are determined by the terminology of more concrete source concepts. Humans use conceptual metaphors in an unconscious way and also without effort.

An important characteristic of conceptual metaphor is systematicity due to the fact that they provide systematic mappings between domains. Metaphorical expressions are systematically linked to particular conceptual metaphors. Furthermore, they form metaphor systems which are hierarchical. In terms of metaphor systems, different source domains characterize them. In English, the metaphor systems that were discovered are the following: the Great Chain of Being metaphor system, the Complex Systems metaphor systems and the Event structure metaphor system. English is not the only language in which these systems have been discovered. We can also mention Hungarian and Chinese.

The Great Chain of Being is a folk model of nature. It states that entities are organized hierarchically and as a result each level of the chain receives the properties of lower entities and incorporates a new one that makes each level more complex. Also, the defining property of each level is not inherited by the levels which are place below. Humans are rational beings , animals are instinctual, while plants are just entities that live. The classification mentioned above has been improved by Ruiz de Mendoza And Otal (2002). They consider that metonymy has an important role in the cognitive approaches to language. The analysis of metaphor clearly differentiated literal language from the figurative one, while at the same time having the central role in the reaction against the traditional view of conceptualization. Ruiz de Mendoza and Otal (2002) have established three criteria when they realized the classification of metaphor: the nature of the source domain, the complexity of the mapping system and the nature of such correspondences. They further suggested that a distinction ought to be realized between structural and non-structural metaphors according to the number of correspondences involved in the metaphoric mapping. Structural ones contain more than one correspondence while the others have only one. Lakoff’s orientational and ontological metaphors are part of structural metaphors. In addition, structural metaphors are divided into situational and non-situational. The non-situational ones can be image-schematic, image metaphors and propositional ones. Propositional metaphors are non-topological, while the remaining two are topological. In image schematic metaphors, the source domain contains one or more image-schemas, in image metaphors, the source and the same domain are images with a shared degree of similitudes, and in propositional metaphors, the target domain containing the abstract concepts is interpreted according to a non-situational construct in the same domain. A metonymic mapping should be made within the metaphoric source domain, through which a fragment of a situation is extended into a complete one in order to create situational metaphors. If this process can be noticed externally, then we have a scenic situational metaphor, if it is not, then the metaphor is a non-scenic one.

Grady (1997) made a distinction between resemblance and correlation metaphors taking into consideration the nature of mapping. Resemblance metaphors are based on the similarities existing between source and target while correlation metaphors are firmly based in the conflation of concepts.

The following scheme of metaphor types according to the nature of source domains was proposed by Ruiz de Mendoza and Otal in 2002.

According to their level of generality, metaphors can be at the generic or specific level, thus resulting generic-level metaphors and specific-level metaphors. The former ones are on a high level on the scale of generality of conceptual metonymy. They encompass generic level source and target domains. An example may be EVENTS ARE ACTIONS or GENERIC IS SPECIFIC. Specific level metaphors instantiate generic level metaphors. On the other hand, specific level metaphors are instantiations of generic level metaphors. They are also made up of target domains and specific level source.

In terms of complexity, conceptual metaphors are divided by Grady into primary or compound, while Kovecses regards them as being simple or complex. According to Kovecses, simple metaphors map the meaning focus of the source onto the target and in doing so, they provide the major theme of complex sentences.

Many image schemas, as IN-Out, UP-DOWN are mapped by primary metaphors. “Non-decomposable primary metaphors usually have independent experiential motivation, and occur in various metaphorical expressions independently of any single metaphor” (Grady 1997:286).

Lakoff and Johnson (1999) argue that primary metaphors are acquired in an automatic and unconscious way and that they are embodied through bodily experience in the world.

Besides conceptual metaphors, there is another conceptual mechanism presented in Lakoff and Johnson’s Metaphor We Live By, which has a central role in human language and though called conceptual metonymy. It is a cognitive process in which the mental access to another conceptual entity (target) is allowed by one conceptual entity (the vehicle). This process happens within the same conceptual domain. Metonymical expressions are not isolated, but systematically analysed depending on the relationship existing between the given entities and are grouped according to conceptual metonymies.

Metonymies, like metaphors have a conceptual nature. Some scholars argue that metonymies have a more fundamental role to conceptual organization in comparison with metaphor. They have also claimed that metaphors have a metonymic basis

Here are some examples of conceptual metonymies: the entity Hemingway can be both the AUTHOR and the WORK, in the example Last night I read Hemingway. In addition PLACE can refer to an event that happened there or an institution whose headquarters are there. The specific relationship existing between the entities can be observed with the help of conceptual metonymies. Mostly we can notice that metonymy represents PART FOR WHOLE (let’s put our heads together) or WHOLE FOR PART (America is a huge country).

Metonymical relationships contain related entities which are close in terms of conceptual space. The producer is close to the product since it makes the product. Mental access to the target entity is provided by the vehicle entity which is in the same domain as the target entity. “The entities in an ICM form a coherent whole in our experience of the world as they can co-occur repeatedly” (Lakoff 1999:25).

In the following section we are going to present the relationship between conceptual metaphors and metonymies. The basic distinction between these two concepts is the type of mappings created between the domains. In metonymy, the mapping takes place within the same domain while in metaphors it happens across different experiential domains. As a result, metaphor is “a conceptual process which involves the mapping at the structure of one conceptual onto that of another conceptual domain”, while metonymy is “a conceptual process which involves the substitution of one conceptual entity for another conceptual entity within the same domain”(Dirven and Rodden 2001:25).

According to Kovecses (2002), there are several differences between metaphor and metonymy. Continuity is the basis of metonymy, while similarity is the basis of metaphor. On the other hand, metonymy involves one domain, compared to metaphor which involves two distant domains. Metonymy uses a single domain to allow access to a single target entity while metaphor is mainly used to understand a whole system of entities in terms of another system. Metaphor takes place only between concepts whereas metonymy also occurs between concepts, as well as linguistic forms.

Metaphor and metonymy interact sometimes making it difficult to state whether a linguistic expression is either a metaphor or metonymy. An example of this situation is the expression “to be close-lipped”. It has two meanings. The first one “to be silent” which uses metonymy and the second “to say little” based on metaphor. It is also possible to encounter a metonymy within metaphor. This happens in the expression “to shoot one’s mouth off” which means “to talk foolishly about something that one does not know much about or should not talk about”. “In the metaphorical reading, the gun as a source domain item is mapped onto the target domain, speech, more specifically, onto the mouth” (Kovecses 2002:15). So, the foolish use of a firearm is mapped onto foolish talk.

Many metaphors can have their basis formed by certain metonymic relationships. The main type of metonymic relationships which can determine the formation of conceptual metaphors are CAUSATION and WHOLE-PART.

The speaker’s shared everyday knowledge about particular domains is called conventional knowledge and it includes standard information about specific domains and the relationship between the entities. As a result, our conventional knowledge about a specific part of the body, for example the human hand, contains information about the parts, shape, size, and functionality.

Idiomatic expressions are motivated by several cognitive mechanisms. “Thus it is possible that in addition to one or more conceptual metaphor, a conceptual metonymy (or several conceptual metonymies), and/or our conventional knowledge, all contribute to the meaning of an idiomatic expression” (Csabi 2004:18).

Gibbs (1996) suggests that conceptual metaphors have psychological validity. He further argued that they are present in our conceptual system.

Conceptual metaphors and metonymies occur independently of language, due to the fact that there is a large number of non-linguistic realizations of the two conceptual structures. Lakoff and Johnson argue that (not all conceptual metaphors are manifested in the worlds of a language” (1999:57). Apart from figurative linguistic expressions, metaphors and metonymies can also be formed in the structures of movies, cartoons, morality, politics and many others.