This essay attempts to describe the role of semantics in the study of linguistics, its ability to reveal the connection between cultures, and its ability to describe human experience. Contemporary studies of semantics are also done through cognitive experiments that use imaging technologies. However, the study of semantics may be constrained by the huge variation in word meaning across cultures. The future of semantics lies in explaining such variational discrepancies and accounting for them.
What are semantics?
Expressing and communicating meaning is essentially the function of human language (Goddard 2011). Meaning is the knowledge acquired by the human brain through experience which is represented symbolically in language. The relationship between the knowledge acquired about an experience and the words used to describe them are known as semantics (Binder et al., 2009). Semantics is a core field of linguistics and is crucial in the attempt to describe and account for linguistic competence (Goddard, 2011). Much of the foundation of semantics originates from logic and the philosophy of language (Chierchia & McConnell-Ginet, 1990).
Semantics reveals the connection between languages and cultures, with language being the main way in which children are assimilated into their society’s practices and beliefs (Goddard, 2011). From a scientific perspective, the accessing of this stored knowledge about the world is a cognitive function known as semantic processing. Semantic processing is essential to human behaviour. It allows us to access acquired knowledge, reason, plan, and problem solve. It is also integral to language function, with impairments to processing being part of such disorders as dementia, asphasia, schizophrenia, and autism (Binder et al., 2009).
The study of semantics arose when it became apparent that for most languages syntactic structure relied upon word meaning within an expression (Goddard, 2011). A distinction between language competence and performance is important in linguistics in order to gain critical data through which theories can be tested. For example, linguists often rely upon the intuitive judgements of native speakers to ascertain which string of words composes a grammatical sentence. Therefore semantics adopted three concepts:
1) Generative grammars made in artificial languages can be thought of as models for natural languages;
2) these generative grammars are realised as cognitive systems in the human brain; and
3) they are distinct from observable linguistic behaviours, yet also explain them (Chierchia & McConnell- Ginet, 1990).
However, the main theoretical controversy in semantics is what the nature of the optimal system of semantic representation could be (Goddard, 2011). One of the commonsense views of semantics is that the meaning of a word is structured idea within the user’s mind. As we grow we receive pre-packaged concepts that allow us to quickly communicate and share ideas with those around us (Goddard, 2011). John Locke contended that as all able bodied humans share the same sense organs they also experience the world in similar ways which enables the acquisition of similar ideas about the world. Renee Descarte’s contrasting view was that such concepts as time, causation and identity show that they could not possibly be solely derived from our perceptions. Therefore, one of the main approaches to semantics is the aim of finding a universally applicable system of meaning. As we can translate between languages, it seems that a universal framework underpins our understanding and is shared by all humans (Goddard, 2011).
Some properties of semantic competence are producing linguistic meaning. Presumably, this can be done through understanding the single words within a sentence and the algorithm for combining them. A good example of this is Lewis Carroll’s nonsense poem Jabberwocky. Therefore, the concept of a universal semantic base was trying to discover the categories and rules that combined this universal vocabulary. However, advances in neuroscience have also allowed us to investigate how semantic processing actually operates within the brain through functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) (Binder et al., 2009).
What are the techniques used to study the brain mechanisms involved in semantics?
In the studies investigated by Binder et al. (2009), fMRI and PET experiments were conducted using written and spoken words as stimulus materials. The goal of these experiments was to identify the brain systems used for accessing meaning from words and see how conceptual knowledge is organized and accessed. All the experiments involved healthy adult participants. It was found that ‘all consciously executed, goal-directed tasks require a minimum set of domain- general processes that include maintenance of attention, direction of attention to relevant information (external and internal), maintenance of this relevant information in short-tem memory store, maintenance of the task goal and task procedures in working memory, decision, response selection, and error monitoring (Binder et al. 2009).
The Binder et al. analysis also included studies that made a comparison between human primates and non-human primates. These studies found that the human semantic system a network of parietal, temporal and prefrontal heteromodal association areas, which are greatly expanded in human primate brains compared with non-human primate brains.The Binder et al. 2009 analysis of current scientific studies on semantic processing also concluded that the most dense activation of foci was in the angular gyrus. This region of the brain is virtually non-existent in the lower primates and is greatly expanded in the human brain in comparison with the macaque monkey. The angular gyrus is connected almost entirely with other associated regions and receives almost no direct sensory stimulus inputs.
Some fMRI studies have shown that the angular gyrus is activated in response to semantically anomalous words embedded in sentences, which suggests that it plays a role in integrating individual concepts onto a greater one (Binder et al., 2009). Research done by V.S. Ramachandran, director for the Centre of Brain and Cognition at the University of California, found that the angular gyrus is most likely responsible for the human ability to comprehend metaphor (Kiderra, 2005). Therefore, it is asserted by Binder et al. (2009) that the angular gyrus is a high level area for retrieval and conceptual integration, playing a particular role in sentence comprehension, discourse problem solving and planning.
A further analysis of relevant fMRI studies on semantics by Cappa (2012) confirmed that the underlying knowledge within the neural system is distributed throughout the brain. While brain activity is elicited through these requirements, it was found in one study that the lack of temporal activity was unexpected, especially as this was a role given to Wernicke’s area for word comprehension. However, another study interpreted this finding by concluding that ‘semantic processing is automatically engaged whenever subjects are exposed to a meaningful stimulus’ (Cappa, 2012). Language tasks, such as the assessment of verb processing are associated with perceptual and conceptual tasks and generally associated with the activation of the motor and premotor regions. Further studies based upon sensory and motor verbs accurately predicted a relationship between the cortical area of predicted activation and the verb meaning (for example ‘touch’ and the somatosensory cortex) (Cappa, 2012).
How have studies on the brain constrained or highlighted semantics?
Languages have a huge variation in word meaning across cultures. For example, it could be assumed that all languages would have a word to describe a single hand or one’s hair. Yet the Russian word raka describes the entire arm, and in Yunkunytjajara (Central Australia) mangka describes head-hair while the word yuru describes body-hair as well as fur. One also might think that all languages would have a separate word to describe such pervasive things as mountains, clouds or trees. However, in Yunkunytjajara one word, punu, covers both tree and bush, or apu covers both mountain and hill. The word clouds is covered by several words thatdescribe different types of clouds in Yunkunytjajara. Even concrete meanings are not universal across languages. English words such as privacy, work, shame, or guilt are abstract concepts particular to certain cultures. In this way, language is a unique way of describing and thinking about the world and therefore studies on the brain may constricted by such subjectivity of understanding across cultures (Goddard 2011).
So, is a universal syntax or semantic structure a feasible part of universal grammar? All languages seem to refer to space and spatial relation and regenerate them to refer to more abstract concepts as possession. However, the precise details are not universal. Two types of components are involved in meaningful language- informational significance, which connects language with the world; and cognitive significance, which links language and semantic knowledge. Informational significance is something shared between many individuals. An example is a statement such as ‘the door is closed’, which is easily understood because it can be concretely depicted in that it involves information given to grasp its meaning. Also, language allows us to externalise out thoughts and share our imaginings so that informational significance is not only related to physical phenomena but also mental phenomena. Cognitive significance depends upon matching representations rather than having the same links to situations. For example a statement such as ‘this is yellow’ is subjective and will not necessarily be represented identically as a general understanding but it achieves a degree of consensus of informational significance. Our communication relies upon the fundamental similarities that we experience and some meanings are meaningful even if they cannot be matched concretely. Therefore, this aspect of meaning highlights the importance of how communicated information is processed (Chiercha & McConnell-Ginet, 1990).
Questions for future studies of the understanding of semantic processing?
Binder et al. (2009) proposes that future analyses of cognitive imaging studies should concentrate on establishing an ontological system for defining the activation of cognitive processes. Such an objective system would rest upon axioms concerning stimuli through consensus and would allow for a common theoretical framework.
Goddard (2011) hints at such a system when he states that semantics is about what knowledge people have of their language. It is not enough to know that people use these words; we need to be able to describe what they know about a word’s usage. Goddard also highlights another problem, that of anglocentrism in the study of semantics. Using English as metalanguage to describe other languages necessarily involves bootstrapping English cultural notions onto other cultures that may not have the same types of concepts or equivalents. Cappa (2012) is also concerned over the interpretation of cognitive imaging studies and whether observed activation in some actually reflects semantic processing and not ‘epiphenomena’ rather than an actual correlation of semantic expression with activation of visual or motor imagery. Cappa (2012) also states that further studies could be carries out of the antero-lateral temporal cortex, as it is an anatomically complex region that is a crucial hub of semantic processing and is only beginning to be investigated in finer detail.
The aim of finding a universally applicable system of meaning will assist neuroscience in investigating how semantic processing actually operates in the cognitive system. Human semantic systems are greatly expanded in comparison to non-human primates that infers our use of description and meaning is necessary for our survival. One of the key areas of the human cognitive system is the angular gyrus which is a high level area for retrieval and conceptual integration. In order to explore how such integration can influence human problem-solving abilities the aspect of meaning that relies upon the fundamental similarities that we experience highlights the importance of how communicated information is processed. Therefore, research into the universal understanding of meaning seems as necessary to linguistics as the investigation into universal grammar.
Binder, J.R., Desai, R.H., Graves, W.W., & Conant L.L., (2009), “Where Is the Semantic System? A Critical Review and Meta-Analysis of 120 Functional Neuroimaging Studies” in Cerebral Cortex, December 2009;19:2767—2796
Cappa S,F., (2012), “Imaging semantics and syntax”, NeuroImage,61 (2012) 427–431
Chierchia, G. & McConnell-Ginet, S., (1990) Meaning and Grammar: An Introduction to Semantics, Cambridge : MIT Press,1990
Goddard, C., (2011), Semantic Analysis: A Practical Introduction, Oxford University Press
Kiderra, I. (2005), “Grasping Metaphors: USCD Research Ties Brain Area To Figures of Speech”, UCSD News, University of California, San Diego, retrieved on 16 January 2013 from: http://ucsdnews.ucsd.edu/archive/newsrel/general/RamaMetaphors.asp