10. Issues in Language and Thought

The evidences proved that language and thought are proccessably intertwined and established. The interactive relationship between language and thought has also been demonstrated in most of the studies seeking to address the issue. Such an issue may change with the changing concerns of the two participants in communication.

1. The Evolution of Linguistic Complexity

Human language system has complex structure and physical inadequacies. Its complexity evolves on several levels as people learn the languages and use them to communicate and to express their thoughts.

Not only has language evolved as unique to human beings due to the evolution of the species as a genome -induced long-drawn process of evolution, but this has itself shaped the further course of language -induced evolution. Evolution of language is widely accounted and explained in many theoretical perspectives but they all agree to brain, vocal tract and a genetic endowment of language faculty as the crucial determinants of language as unique to humans. At the same time, it needs to be pointed out that emergence of language, particularly in the light of the vast gamut of typological commonalities and individualities, also shows a credible shift from simple to complex symbolic system of communication.

In fact, people are not free to introduce innovations in the language systems unless these innovations are immediately understandable through metaphorical or metonymical process. It is because languages are so variable and changing. This complex characteristic can be found in the studies of Chomsky (1988) about dialects along the border of Dutch and German which are so similar to be distinguished to each other. Generally, languages are dynamic systems, and increasingly complex both in the individual and in the partially shared systems of language communities.

2. Models of Semantics

An adequate theory of language needs to support a real adequate theory of semantic structure. A major psycholinguistic issue in semantics relates to how meaning is organized and represented in human mind. Traditionally, there are two models of semantics: classical models and new models.

The classical models seek to identify semantic units as discrete, independent entities, detached from human experience and cognitive processes. For this model, words have fixed meanings and direct correspondences to concepts and objects they symbolize; and that how human being named an object or an even is accidental and arbitrary. As they seek to break down language into its barest constituent parts, much as in morphology, the classical models are mathematical in nature. Such a model of semantics looks at the formation of complex words just as a process of combining semantic units, with a catalogue of semantic universals. With this it would be possible to construct semantic units to cover any range of ideas. Through the various models of mental lexicon available today, such as the hierarchical models, protype models, network models and so on.

A constraint of the classical model is that while it inclines mathematically, it seems to be unable to convey and express how humans actually use words. It is primarily because word meanings are more than merely a reference. Language users may also load information of attitudes, feelings, and opinions onto the words for their referents.

Kess (1992) highlights three basic elements in meaning.

i. Symbol / name = the phonetic shape of the word

ii. Thought / reference / sense = the information which the name conveys to the speaker

iii. Referent / thing = the non-linguistic feature or event the speaker is talking about

Words stand in a relationship to the world or one’s mental classification of them allow him to identify parts of the world and make statements about them. The relationship by which language hooks onto the world is usually called reference, and the semantic links between elements within the vocabulary system are an aspect of their sense. In (7) below, for instance, the underlined nominals (the girl and an ant) identify, pick out, or refer to specific entities in the world.

In (8), John (proper name) has reference but no sense while happy does not have any reference but has sense:

(7) The girl saw an ant
(8) John is happy

Thus the relationship between a linguistic expression and an entity in the external world to which it refers also called the referential meaning of the expression. The meaning or sense of an expression is the concept or set of properties that is called to mind. This is the issue of a new model of cognitive semantics.

Of late, however, many new models of cognitive semantics have been tried to deal with the problems of meaning formation by looking towards human experience and the human body as the primary sources for linguistic meaning. In this model, the meaning of words lies not in the feature they have, nor in the truth-conditions they fulfil, but in a construction purportedly achieved between speaker and hearer in a domain of knowledge.

3. Syntax

Syntax is the study of sentence-building, and a grammar of a specific language is a set of conventional rules about how words are combined to carry out structural meanings though a description of the complex abstract knowledge which speakers-hearers have of their language is equally crucial to the cognitive process. It is, however, not a matter of mere juxtaposition of words or morphemes, but sentence patterns structured according to definite restrictions; that is the patterns in which words occur are hierarchically ordered. In the following examples in (9) and (10), the nominals such as John/Marry/flower occur, but the meaningful relationships which they enter into are defined by their relative ordering according to the syntax of English.

Thus, in (9), John is the giver, Marry is the recipient, and (10) is vice versa.

(9) John gives Mary flowers
(10) Mary gives John flowers

Syntactic analysis allows people to determine possible groupings of words in a sentence. Sometimes, in a sentence, there is only one possible grouping, and one should be well on the way to working out the meaning. As in (11), for instant, there are two groups of noun: the squirrel with long tail forms one noun phrase; and a piece of bread forms another noun phrase:

(11) The squirrel with long tail enjoyed a piece of bread

On the view of linguists, syntax is more or less full specified by a learning mechanism, the Chomskyan LAD (Language Acquisition Device) which

significantly constrains the language learner with prior knowledge about the nature of language (Chomsky 1986). Prior constraints on a learner are assumed to be innate and therefore genetically determined. In fact, the best explanation for the structure of language is one which invokes biological evolution. Generally, rules of syntax specify the possible organizations of words in sentences. They are normally specified by writing a grammar for the language. However, just having a grammar is not enough to analyze the sentence–we need a parse to use the grammar to analyze the sentence. We, indeed, recall more than we actually hear, because we not only make inferences from the sentences that we hear but we actively construct chunks of memory. We integrate all this information into a mental model of the situation, and it is this model that is stored and later consulted for recalled (Kess 1992).

4. Evolution without Natural Selection

Kirby (1999) points out that linguistic transmission can be viewed as a repeated transformation of information between two domains: (i) the internal domain (I-domain); and (ii) the external domain (E-domain). The I-domain contains languages that exist as grammatical knowledge in individual’s brains, whereas the E-domain contains languages that exist as sets of actual utterances. This conception has parallels with the conception of genetic transmission in biology. The following figure presented by Gupta (2002:25) will compare the transmission of information between phenotype and genotype domains with the transformation between E-domain and I-domain.

Figure 3. Transformation of Information in Linguistic and Biological Systems

According to Gupta (2002), in both these adaptive systems the biological and the linguistic, the transformations that map between the two domains can be seen as ‘bottlenecks’ on the transmission of information overtime through the whole system.

5. Spoken Word Recognition

The research area of this subsection examines how hearers recognize words in spoken language. Work on continuous speech recognition has brought out some relevant aspects of the issues involved. One example can be found in the related research on segmentation on the phonological structure in comprehension. Another main target of this research area is to map out the information-processing architecture recognition: what information is used and what are different stages in the recognition process, and how does information flow through the recognition system? The emphasis is on phoneme sequencing constraints; hearers’ sensitivity to the relative likelihood of phoneme appearing after each other and a comparison of lexical and phonotactic influences on phonetic categorization (Gupta 2002).