106 DEVELOPMENT: ASPECTS OF ENGLISH is presented is to encourage thoughts of the unsaid opposite (‘best’ then hints impli- citly at the worst). The choice of register is not just simply poetic but highly self- consciously poetic. The poem even begins with a semantic and syntactic trap for the reader: ‘She walks in beauty’ is semantically odd, and its odd distraction is partly what causes the vast majority of people reading the whole line out loud to pause at the end of the line. ‘She walks in beauty, like the night’ is then ambivalent between beauty being like the night or she being like the night, and the night is conventionally the time of mystery, edginess, danger and illicitness. Though the next line rescues the sentiment, the fleeting effect has already occurred. Similar suggestive effects permeate the poem, as in the invocation of darkness and shadow. There is a sense of vagueness and underspecificity in lexical choice (‘name- less grace’, ‘dwelling-place’, and especially suggestively ‘all below’) – these are taken as idealisation of beauty in the traditional interpretation, but they could equally be read as being euphemistic and evasive on the part of the writer. The writer’s voice and perspective provide the content of the first two stanzas, but the observing conscious- ness is not foregrounded. However, this changes in the final stanza, with the distal deictic ‘that’ (‘that cheek that brow’) signalling strongly – again by implicit con- trast – the deictic centre of the writer, where ‘this’ would be located. He draws atten- tion to the inarticulacy of his viewpoint (‘so soft, so calm’) and even highlights the issue itself in invoking the word ‘eloquent’. The writer, in fact, enters into the poem towards the end as more than a mere neutral observer, as an evaluator with his own agenda. The iambic tetrameter, which looks so innocuous, was also prototypically the form of ancient Greek dialogue, and the most famous example of the form is the dialogic and seductive opening to Christopher Marlowe’s The Passionate Shepherd to his Love, ‘Come live with me and be my love’ (c.1585). Dialogue and viewpoint The foregrounding of the poem as a dialogue, in which only the writer speaks and the woman is abstracted and silent, seems to evoke a viewpoint with his own agenda. ‘The smiles that win’ could be read as false, tactical smiles (and the definite article serves to heighten this sense of disembodied artifice); ‘the tints’ are even more cosmetic and superficial. ‘But tell of days in goodness spent’ can of course be read as ‘merely tell of a life passed in goodness’ but it could equally be read as ‘instead confess a life with all goodness spent and used up’. In this reading, the sudden shift to indefinite universal sentiment, ‘A mind at peace with all below’, appears ironically overstated, and the ironic intent of the final line can be strongly inferred in the closing exclamation mark. (It is noticeable that many editors, perhaps intuitively mindful of the underlining potential here, delete the original punctuation in favour of a simple full-stop.) We are suggesting that the idealisation presented on the surface of the poem is too good to be true, and the poem is also more concerned with the writer than might appear at first glance. What might at first appear to be a joyful celebration can under analysis look more sinister and cynical. There are several subtle stylistic cues that might invite this reading, and if the reading is there at all impressionistically, a detailed stylis- tic analysis can help to reveal it clearly. Whether it can be considered valid or not is a matter for your own judgement, and your own production of evidence. TECHNIQUES AND ETHICS 107 TECHNIQUES AND ETHICS One of the most important methodological techniques that needs to be acquired for English language study is the ability to record and then transcribe spoken language data. The need for transcription sets the discipline of language and linguistics apart from others in the social sciences. A number of transcription extracts are used in this book, particularly across the discourse strand, in A5, B5 and C5, as well as in D3, on pragmatics and discourse. Whilst traditional quantitative methods, such as experiments which take place in language laboratories, are always recorded, other social scientists who conduct qualitative research through participant observation rely solely on a notebook and pen to record their observations. However, to produce a systematic study of any stretch of language data, it is important, wherever possible, for recordings to be made and data transcriptions to be produced. Our memories cannot record accurately the details of exactly who said what, when they said it and how they said it. As far as very short exchanges are concerned, researchers may produce a written transcription of the spoken data from memory, writing it down as a record of what took place at the very first opportunity to do so. Of course, if we had a recorded version we could check and double check for accuracy, but on some occasions really fruitful exam- ples of language data will present themselves when we do not expect it. For example, there are occasions when we overhear very short exchanges when we are not in any ‘official’ language-gathering capacity, as in the ‘Carlos Bridge’ example in A10, where recording is not taking place. Ideally, all language data should be recorded, though exceptions are generally made for very short, spontaneous exchanges when analysing discourse features. However, if you were wanting to analyse phonetic language variation, then it is essential that you have a recording of your spoken data – you need to be able to listen to recordings of phonetic features on a number of occasions to obtain any accuracy in your analysis. Likewise, any longer stretch of discourse or any discourse where there are multiple speakers would also be impossible to transcribe without a recording being made. There is no standard system for transcription conventions for English language study and conventions can vary quite dramatically, depending upon which particular aspects of the English language researchers are aiming to analyse. Probably the largest and most comprehensive set of transcription conventions is followed by hard-core conversation analysts, originally designed by Gail Jefferson (see Jefferson 2004). While the lack of a standardised system can be frustrating when you first start off as a fieldworker and analyst in English language studies, the flexibility of developing your own conventions is really useful as it gives you the freedom to choose the con- ventions that are most important and most relevant to the study which you are con- ducting. For instance, if you are not examining accent variation, then you do not need to transcribe phonetically. Transcription conventions Here are the transcription conventions which we have followed in this book. These have been blended together from across the disciplines of sociolinguistics and discourse analysis: B12 108 DEVELOPMENT: ASPECTS OF ENGLISH (.) indicates a pause of two seconds or less (-) indicates a pause of over two seconds (xx) indicates material that was impossible to make out {xx} indicates material that has been edited out for the purposes of confidentiality [ ] closed brackets indicate the starting and finishing points of simultaneous speech %word% percentage signs indicate material was uttered quietly WORD capital letters indicate material was uttered loudly ((laughs)) material in double brackets indicates additional information = equals signs indicate no discernible gap between speakers’ utterances all materials contained within double dotted lines should be read alongside each other, as in a musical score. The observer’s paradox and ethics Should spoken language data be video or audio-recorded? A newly emergent research focus in discourse analysis is on multi-modal communication – where a full range of communicative behaviour, including gesture, is analysed in conjunction with the more traditional approach of investigating spoken discourse by the production of written transcripts. If you are taking a multi-modal approach then you cannot do this without video recording. However, one feature of multi-modal studies at present is that they tend to be restricted to artificial, experimental settings, due to the number of video-recording devices that are required in one room at any one time in order to make accurate recordings of gestures (see Knight 2009, for a good example). It is also well established that video recording is more likely to trigger the effect of the observer’s paradox: that the act of observing (through the researcher’s physical presence and/or recording equipment) is likely to change what it is that you want to observe. Permission for gaining access to data generally becomes far more difficult if you want to use video, as many individuals are far more reticent to be video recorded than audio recorded (Duranti 1997). However, with audio recordings you are restricted to gaining ‘a filtered version of what happened whilst the tape was running’, though this method still has ‘the power to capture social actions in unique ways’ (Duranti 1997: 119). The inescapable fact remains, however, that the presence of any recording equipment, be it video or audio, can change participants’ behaviour. Duranti (1997: 117) argues that the problems raised by the observer’s paradox, if they are carried to their logical conclusion, suggest that ‘it would be better not to be there at all’. There are two ways in which such a proposition can be realised. The first option is to throw in the towel and completely abandon the study of people altogether, whilst the second option is to record covertly, not informing participants that their behaviour is being recorded, which raises considerable ethical problems: The first option is self destructive [as] . . . it implies that we should not improve our understanding of what it means to be human and have a culture (including a language) simply because we cannot find the ideal situation for naturalistic-objective observa- tion. The second proposal is first of all unethical and, second, impractical under most circumstances outside of laboratories with two-way mirrors. (Duranti 1997: 117) TECHNIQUES AND ETHICS 109 The unethical nature of covert recording is also highlighted by Milroy (1987), who argues that it is now essential for language studies to be carried out overtly, so record- ing covertly and then asking for permission afterwards is no longer a viable option. Milroy points out that, as a tape is a permanent record of behaviour, those being recorded are entitled to know that such a record is being made. As a person’s voice is an impor- tant part of their self-image, they may often be unhappy with being recorded without their knowledge. Covert recording can also result in serious practical disadvantages. Surreptitious recording, when revealed, can result in future access being denied. Furthermore, the quality of recording is usually poor if the equipment is concealed, often so that data cannot be properly transcribed. Codes of ethics are now being estab- lished by recognised organisations (such as the British Association of Applied Linguistics) to establish ethical guidelines for data collection. Whilst the observer’s paradox is a problematic concern, this does not mean that researchers should abandon analysing naturally occurring data. For the vast majority of the time ‘people are too busy running their own lives to change them in substan- tial ways because of the presence of a new gadget or person’ and ‘neutral’ observation is an illusion anyway (Duranti 1997: 118). By arguing this, Duranti is not advocating that the observer’s paradox be ignored. Rather, he is quite rightly suggesting that researchers should approach it in full awareness that it is completely unavoidable. Any language student or researcher should include a discussion in their method- ological rationale on how they have tried to minimise the effects of the observer’s para- dox, but ultimately acknowledge that it is impossible to get around. A direct and transparent acknowledgement of the observer’s paradox, followed by a reflexive account of how you have attempted to minimise its effects and then how it may have affected the data collection, is an acceptable way forward. In the vast majority of cases, research subjects forget that they are being recorded as they become absorbed in what they are doing, particularly after the first few minutes that the recorder has been running. Many researchers have interviewed participants afterwards to ask whether they thought the recording equipment had affected their behaviour. The consensus tends to be that, after the initial few minutes, they forgot they were being recorded as they became focused upon the topic and/or task at hand. However, there are exceptions, and the following examples illustrate occasions where the presence of the tape recorder was noted and actually became part of the topic of the conversation outside of the first few minutes of these business meeting interactions: Sharon is talking on the topic of a difficult customer . 1 Sharon: They’re banging on that we’re always late 2 shove that up your arse 3 ((laughter)) that’s recorded 4 ((Laughter from many)) Meeting-Chair Amy, who is the line manager of all meeting participants, is running through a staff rota to confirm employees’ working hours . 1 Mary: Oh excuse me ((yawning)) 2 Amy: What are you laughing at? ((laughter from many begins)) 3 Mary: [I’ve (xxx) been taped (-) taped yawning ((laughs))] 110 DEVELOPMENT: ASPECTS OF ENGLISH 4 [((laughter from many)) ] 5 Mary: Sorry about that 6 Amy: It’s really boring this meeting is it? 7 Mary: [No I just need some food ] 8 Karen: [I I ] The reference to the tape recording equipment here appears to take place when speakers become very self-conscious about what they are saying. In the first instance this is related to using the word ‘arse’ by Sharon in line 2 in a formal business meeting where the established speech norms of this community of practice do not normally include expletives. In the second example, Mary has become self-conscious as she has committed a social faux pas by yawning in a meeting. Part of her embarrassment about this is caused by the fact that her yawn has been recorded on tape and is thus permanent. One of the ethical considerations of recording your own friends is that if you have to leave the room for any reason then it is possible that they may talk about you as they forget the tape recorder is running. So, whilst the example below is a good illus- tration of participants forgetting that the recorder is on, it presented researcher Jennifer Coates with the ethical dilemma of being able to hear her friends talking about her on the rather sensitive topic of her ex-husband: Discussing Jen’s arrangements to get her ex-husband to help with her move to London MEG: I mean I wouldn’t rely on him for something as vital as that/ [. . .] [ Jen leaves the room to answer phone ] SALLY: Your faces when Jennifer said that- that Paul was going to do MEG: (LAUGHS)- SALLY: the move/ .hh I wish I’d got a camera/ (LAUGHING) ((it)) was MEG (LAUGHS) SALLY: sort of- ((xx)) in total disbelief/ I think the most difficult MEG: mhm/ SALLY: is- is that when you’ve loved someone / you- you half the time you SALLY: forget their faults [don’t you? and still maybe love them / MEG: [yeah/ (Coates 1997: 123) Alongside ethics are also practical considerations of gaining access to language data. Data collection can become difficult, for example, during times of political and social unrest/upheaval: some social settings can also be really problematic, if not downright dangerous to enter. It is very important that you do not put yourself in danger whilst doing fieldwork! LANGUAGE AND THOUGHT 111 Sociolinguist Miriam Meyerhoff (2006) reports on a case where an individual named Anibal Otero was arrested and imprisoned for being a spy during the Spanish Civil War as he was discovered with strange ‘codes’ in a notebook. Assuming he was a spy, he was charged with treason. However, the codes were actually phonetic transcrip- tion, as he was employed as a fieldworker to work on the Linguistic Atlas of Iberian Spanish. As a truly dedicated linguistics researcher at heart, he turned his plight to his advantage by collecting data from his fellow prisoners. In C12, you will come across an activity that asks you to think about these access issues in more detail and decide whether in some scenarios data collection should not be attempted at all – for instance, studying the language of drug dealers or criminal gangs. LANGUAGE AND THOUGHT Do we think in language or do we speak after verbalising our thoughts? One of the main reasons why the study of language has proven so enticing and tantalising for thousands of years is that it appears to be intimately bound up with our thought pro- cesses and consciousness itself. Indeed, until the most recent years of psychological experiments, eye-tracking devices, MRI scans and the clinical observation of those who have suffered head injuries, the main window onto the human mind consisted of the patterns of language through which humans articulate their thoughts. Though the observable tip of language can be seen and heard in writing and speech, we can talk to ourselves sub-vocally, and think silently in our own language inside our heads. Whether explicitly or implicitly, much of the modern linguistic study of language has also been a study of the mind. Determinism and language Therefore it is not surprising that the main movements in modern linguistics – generativism, functionalism and cognitivism (see A13) – have all become engaged with philosophical discussions concerning the status of what the mind is, what conscious- ness is, what reality is, and whether the observable patterns of language can tell us anything about our inner lives, our genetically inherited lives, our determination or free will, our adaptability or how far we are constrained within our social systems and human condition. For example, in the Appendix to his 1948 novel Nineteen Eighty-Four, George Orwell set out the principles of ‘Newspeak’, the fictional language of his imagined totalitarian state. The compilers of the 11th Newspeak Dictionary aimed to reduce the number of words available to the population. The lexicographer, Symes, says: Don’t you see that the whole aim of Newspeak is to narrow the range of thought? In the end we shall make thoughtcrime literally impossible because there will be no words in which to express it. Every concept that can ever be needed, will be expressed by exactly one word, with its meaning rigidly defined and all its subsidiary meanings rubbed out and forgotten [. . .] It’s merely a question of [. . .] reality-control. B13 . acquired for English language study is the ability to record and then transcribe spoken language data. The need for transcription sets the discipline of language and linguistics apart from others. for transcription conventions for English language study and conventions can vary quite dramatically, depending upon which particular aspects of the English language researchers are aiming to. studying the language of drug dealers or criminal gangs. LANGUAGE AND THOUGHT Do we think in language or do we speak after verbalising our thoughts? One of the main reasons why the study of language