112 DEVELOPMENT: ASPECTS OF ENGLISH Most words in Newspeak are removed by selecting a basic term (‘good’, rather than the possible range of ‘fine, nice, excellent, ok, well, lovely, pleasant, superb, brilliant’ and so on) and then allowing a small set of very regular inflections (‘gooder, ungood, goodest, plusgood, doubleplusgood’). The idea on which this is based, as Symes explains above, is that language and thought are directly and causally connected, so reducing the language will reduce the capacity for thinking freely. This strong linkage between language and thought is a highly deterministic view. It asserts that thought is purely linguistic and mental operations are conducted in terms identical or very close to the surface structure of language. It also suggests that ideas are literally unthinkable, or at least very difficult to imagine, without the linguistic means of articulating them. Such a view has been called the Sapir-Whorf hypothesis, named after Edward Sapir and his first student and later colleague Benjamin Lee Whorf. In fact, their theory of linguistic relativity was not as deterministic as has been characterised, and ‘Sapir-Whorfianism’ is largely a misapplied term. However, Sapir certainly took the view that language and what he called our ‘thought-grooves’ are very closely related, and a culture’s perspectives and preoccupations can be discerned in its linguistic patterns. This can be seen, for example, in the overlexicalisation of certain domains (kinship terms for family relationships, technical terms or jargon in certain occupations) where there is a particular cultural interest. Whorf developed his position after noticing, when working as a fire insurance investigator, that people would throw cigarette ends into gasoline barrels labelled ‘empty’, when in fact they were ‘full’ of explosive petrol vapour. Cognition and language Though fictionally terrifying, Newspeak and its extreme determinism could not pos- sibly work as described in the book. People commonly demonstrate an astonishingly creative capacity for articulating new thoughts either by coining and spreading new words and phrases, or by shifting the meanings of existing words and phrases, or by extending metaphorically the meanings of words and phrases. If the deterministic shack- ling of language and thought were true, it would be impossible to invent any new object or concept before inventing the word for it. Translation between languages would be impossible. Miscommunication and the exaggerations and hypercorrections caused by social diffusion are also a challenge to absolute determinism, since accidental inno- vation is often taken up and made meaningful. And of course, words convey more than they simply denote, and meanings are conveyed by collocations and syntax as well as simply by words. Nevertheless, a weaker form of linguistic relativity, closer to Sapir and Whorf ’s actual work, can be discerned in generativism, functionalism and especially cognitivism. For example, the claims in the first of these for universal deep structures in human language, and innate constraints on what can be well-formed are both relatively deter- ministic assertions. Generativists suggest a deepest form of interior language as a sort of ‘mentalese’, which determines all the various surface languages of the world. In the functionalist tradition, too, there is an acknowledgement that some form of linguistic relativity operates through the social practice of language. For example, one of the most powerful applications of systemic-functional grammar is in critical discourse analysis (see strand 5), which typically explores the discursive SWIN|KCrEIB1Qqc8svpQueSEh0w==|1282035956 LANGUAGE AND THOUGHT 113 patterns in politically and socially significant and influential texts. Of course, those texts can only be significant and influential if you first acknowledge the power of their language to manipulate, constrain and alter the thoughts of the readers or hearers to a large extent. The claimed power of advertising, the moving power of political speeches, and the resonant power of literary works depend on a degree of mind- manipulation by authors through language. Lastly, cognitive linguistics is perhaps the closest relation to Sapir and Whorf ’s work in this area. Some of the earliest work in this tradition in the 1980s was in the large-scale conceptual metaphors that underlie not only many common expressions but also every part of speech: even prepositions (‘in, over, through, with, amongst’, and so on) were shown to be spatially metaphorical and schematic of our early infan- tile manipulations of objects (see Lakoff and Johnson 1986). Some conceptual metaphors (such as life is a journey or love is war) seem to be almost universal across all language cultures of the world. Other powerful conceptual metaphors, such as the following, provide a structuring frame through which we commonly under- stand and recognise the world: ideas are objects communication is a conduit ideas are plants time is money attention is money anger is hot fluid in a container knowing is seeing causes are forces categories are containers important is big good is up and many others. The following phrases exemplify each of these: ‘We can share that idea’, ‘Do you get what I’m saying’, ‘He’s been nurturing that notion for a while’, ‘She just wasted an hour of her life’, ‘Pay attention’, ‘He blew his top’, ‘It’s clear’, ‘She made me do it’, ‘It falls into the category of an accident’, ‘It’s a huge development’, ‘Things are looking up’. The cognitive linguistic research into conceptual metaphor is not absolutely deterministic: no one says that it is impossible to conceptualise LIFE in many ways other than as a JOURNEY, for example (though trying to conceptualise TIME in any ways that are not spatial is very difficult for non-physicists). However, researchers have shown that the extended and habitual usage of a particular conceptual metaphor pushes people towards that form of thinking as the most natural, conventional and easiest form, to the point where people don’t even notice the metaphorical nature of the concept any more. For example, to say that something is ‘in front of ’ something else requires a metaphorical sense of the front and back of objects and your perceived reality of which you are probably not consciously aware. And politicians use metaphors of family, the schoolyard and village communities to articulate their views of international forces and politics in ways that in fact are highly selective and manipulative but seem entirely appropriate and natural. 114 DEVELOPMENT: ASPECTS OF ENGLISH Indeed, much of our linguistic life happens below the level of conscious aware- ness. It seems that we do many things and even say things and then report them to our conscious brain, which then reflects on them and claims ownership and motiva- tion for them. This means that – unless we have had our intensity of awareness raised by very close attention or by the sort of analytical training offered throughout this book – normally we don’t even notice the patterns of language that are subtly struc- turing our perspective and thinking. Means of access through language Of course, researchers in language face a significant difficulty because of the fact that language is the material to be observed that gives insight into how language works, and almost our only access to the linguistic workings of the mind is through that same medium. There are several swirls of circularity here that cause real methodological problems for investigators of language in its broadest sense. It is an extreme example of the observer’s paradox, which, as we have seen in B12, is the notion that the object of investigation changes by the very act of the observer being in the context. Since language is so bound up with consciousness itself, the problems posed for linguists by the observer’s paradox are particularly thorny. In general, there are three broad means of trying to investigate the nature of the language/thought continuum, as practised by the three broad movements in modern linguistics. Generativism for most of its history has relied on the intuition of the theoretical linguist or a group of informants to make judgements on the outcomes of predicted rules. Intuition is problematic, of course, from the point of view of transparency and falsifiability, since there is a large element of subjectivity in it that is particular to the circumstances, and the particular circumstances might not be easily repeatable. However, intuitive judgements have the advantage over almost all other methods except neuroanatomical ones of offering a direct means of investigating cognition and perception in language. Aside from the possible interference of false self-awareness – the phenomenon of watching yourself while you are doing something – intuition is very direct. The second broad method is discourse analysis, discussed throughout strand 5. This includes not just the analysis of spoken events, as discourse analysis is traditionally understood, as conversation analysis or analysis of exchange structure, outlined in A5, but also methods such as stylistics (strand 11), where linguistic frameworks are used to explore textual patterning, corpus linguistics, where computational power is used to explore large bodies of language data (see B10), and critical discourse analysis, where public, corporate and institutional registers are explored (also introduced in A5). Discourse analysis, in this broad, inclusive sense, aims to add to our understanding of language/thought by investigating its traces left in texts and textuality. Though more indirect than intuition, the analysis of discourse that the analyst has not had a hand in producing serves to minimise any influence on the data, other than the fact that it has been selected for analysis (see A12 and B12). Furthermore, the development of a linguistic framework in relation to one domain of language, which then seems to work also by application to another domain, suggests a general validity for the framework. Finally, empiricism of various kinds can be used to prise open the nature of lan- guage and thought. In general, any method which involves the analysis of evidence is LANGUAGE AND THOUGHT 115 empirical, and so there is a sense in which intuitions about language data and the systematic analysis of language data are both empirical too. However, truly empirical exploration includes things like psycholinguistic experiments in which a group of informants are presented with a linguistic task and their responses are recorded and then analysed, in order to establish the truth or falsity of a hypothesis about language processing. This form of investigation tends to be conducted with a close attention paid to the verifiability of the method and the transparency of the data, so it is highly open to falsifiability. However, it is the most indirect means of investigating language and thought, and analysts must be constantly on guard for complicating and inter- fering factors in complex situations, and aware of the dangers in treating statistical results with more symbolic importance than they merit. These issues are also discussed in A12 and D12. SECTION C EXPLORATION: INVESTIGATING ENGLISH LANGUAGE . language that are subtly struc- turing our perspective and thinking. Means of access through language Of course, researchers in language face a significant difficulty because of the fact that language. Symes explains above, is that language and thought are directly and causally connected, so reducing the language will reduce the capacity for thinking freely. This strong linkage between language and thought. involves the analysis of evidence is LANGUAGE AND THOUGHT 115 empirical, and so there is a sense in which intuitions about language data and the systematic analysis of language data are both empirical