1 From the Russian ostranenie – ‘making strange’.
2 For a discussion of the labelling of this field, and a defence of the term ‘stylistics’, see Leech (2008: 1–4).
3 For readers not familiar with this term, it refers to linguistic approaches which, for example, describe the grammar or phonology of a language in general terms, without necessarily connecting this description to contextualised examples in particular texts.
4 Note that although this statement appears to ignore the role of the reader in meaning construction, this concern is increasingly present in recent stylistics. See chapters 3–6 for a range of approaches to this topic.
5 In some contexts, such foregrounding may be intentionally humorous, but the reader will have to take on trust that the earnest nature of the remaining text in this article implied a lack of intentionality.
7 For a theoretically informed discussion of the value of linguistic thinking which is not primarily data-based, see Chapman (2006).
8 See, for example, Wynne et al.'s (1998) description of analytical practice in analysing discourse presentation in a corpus of written texts.
1 Nouns are generally categorised as countable nouns (e.g. toy) which may pluralise (toys) and may occur with indefinite articles (a toy) or as mass nouns (e.g. air) which may not pluralise (without changing meaning radically) or occur with indefinite articles (*an air) but may occur with determiners of quantity such as ‘some’ (some air).
2 For further discussion of internal and external deviation, see Short (1996: 59–61); Leech and Short (2007: 41–5) and Leech (2008:14–18).
3 By rules we mean descriptive rules, not prescriptive ones.
4 Some readers may be wondering why the repetition of backgrounded features is not similar to what has been called parallelism by stylistics. This is because the repetition of backgrounded language is not foregrounded as in the cases of explicit parallelism. The latter are noticeable, whereas the former require detailed (often computer-based) study to discover.
6 Though popular opinion would probably ‘place’ meaning at the level of the word, there is much evidence from linguistics that meaning is situated throughout the structural levels of language. See Jeffries (1998) for an approach to meaning at all levels of structure.
7 Mimesis is related to the concept of iconicity, used more generally in semiotics to define the direct connection between a sign and its signifier. For discussion of iconicity in stylistics, see Leech (2008: 114–15, 149–50).
10 Readers not familiar with the lexical/grammatical distinction can find it defined in any introductory English grammar, and specifically in Jeffries (2006: 83ff.).
11 The sibilants are a sub-set of fricative consonants (in English, these are /s/, /z/, /ʃ/ and /ʒ/) which are articulated by making a groove in the centre of the tongue, down which the air flows. This results in a more ‘whistling’ sound, and may therefore be used to represent a particular kind of fricative sound which is more focused than, for example, /f/ or /v/.
12 Note that although sound symbolism is a commonly-used expression for such meaning potential, the relationship between the sound and the meaning is more accurately termed ‘indexical’ or even at times ‘iconic’, since it is not completely arbitrary but is somewhat motivated by similarity.
13 This book, as with many in linguistics, illustrates exclusively from English. Most of the points that we make can be applied to other languages, though there are some areas of structural variation, including morphology, where other languages may draw upon a slightly different range of resources to create foregrounding effects.
14 The field of critical discourse analysis uses the concept of naturalisation to explain how ideological concepts may enter the perceived common sense of a social group by repetition and backgrounding. Thus, if an ideology (e.g. women should be thin) is repeatedly assumed, rather than stated, it becomes more difficult to distinguish it from a given ‘fact’ (e.g. gravity pulls things towards the earth).
16 Note that although apposition is most common with noun phrases, any two or more structures of the same type may be juxtaposed in a text in such a way that they are implicitly co-referential, and thus appositive.
1 For more explanation of the mechanisms of cohesion, see Jeffries (2006: 183–7) and for the original explanation, see Halliday and Hasan (1976).
2 You can read more about the pronoun use in this poem, compared with another poem, in Jeffries (2008).
3 Some lexical semanticists (e.g. Cruse 1986, 2004) make finer distinctions between opposites with different conceptual fields, but the ones included here are those most often recognised.
4 This phrase would be considered ungrammatical by some, since ‘less’ is traditionally used with mass nouns, and ‘fewer’ with countable nouns like ‘emissions’. This appears to be a distinction which is undergoing change at the moment, and ‘fewer’ is disappearing rapidly.
1 Cook (1994), Semino (1997), Jeffries (2001) and Semino (2001) produce a range of arguments about the extent to which these uses of schema theory to characterise literary effect can be supported. The interested reader may wish to pursue in particular the 2001 articles, where some of the issues are resolved.
2 Note that although there is no clue in the poem as to the sex of the speaker, we are likely to assume it is a man simply because this is a first-person poem written by a male poet. Our assumption rests largely on the fact that the discourse structure of poems prototypically consists of a speaker addressing a hearer. This simple discourse structure leads us to assume that the speaker and the poet are one and the same. The more complex discourse structure of plays and prose fiction means that we are less likely to make such assumptions when we read these texts.
3 This supposed inadequacy of cognitive metaphor theory is not accepted by all, since the fact of butchery being a highly skilled profession is not part of its popular image, and there is thus no mismatch between the idea of an incompetent surgeon and a butcher in the popular imagination. This challenge to a particular example may therefore undermine the rationale for blending theory in general.
1 Despite his use of these terms, Werth intended his theory to be able to account for all forms of language in use, and so in this chapter it may be assumed that the terms author, text and reader refer also to speaker, discourse and hearer.
2 We use the term ‘characters’ here to refer also to non-fictional characters. Text world theory uses the more neutral term enactors.
1 Though see section 1.6 for discussion of the guiding principles of stylistics.
2 Researchers intending to embark on a project involving quantitative research methods should inform themselves about the basic principles of statistical analysis and significance testing. Butler's excellent introduction, Statistics in Linguistics, is now out of print, but can be read online at www.uwe.ac.uk/hlss/llas/statistics-in-linguistics/bkindex.shtml. A more recent work, focusing particularly on the use of statistics in humanities research, is Van Peer et al. (2007).
3 This principle of stylistics is even now giving way to a more inclusive principle that will soon include multimodal texts with or without linguistic content.
4 This journal is the official publication of the Poetics and Linguistics Association (PALA), which is the organisation bringing together stylisticians from around the world.
5 BYU-BNC; available at: http://corpus.byu.edu/bnc/.
6 The exceptions to this are the study of Old English and Middle English texts, where close analysis of the text itself is the prime focus, and the practice of close reading generally, which came to prominence in the mid-twentieth century and was championed by, among others, I. A. Richards.
1 See www.pala.ac.uk.