Amazon S3



Introducing Pragmatics in Use (Second edition). RoutledgeAnne O’Keeffe, Brian Clancy and Svenja AdolphsE-RESOURCE: COMMENTARY ON TASKS___________________________________________________________________________CHAPTER 1: INTRODUCTIONTASK 1.1: CONTEXT AND LANGUAGE IN USEThis task asks you to reflect on Table 1.1 which shows a range of contextual variables that aid understanding of language in use based on Rühlemann (2019: 6-7). There are many ways that this task could be addressed. Here we offer just some suggestions and some prompts. 1) Consider the variables in Table 1.1. What new variables might be added to this list? Or, how might any of these be modified or sub-divided?Perhaps one could add other contexts such as: Cultural context (the range and scope of cultural variables that might be at play in a given context); Medium context (how is the language being transmitted? Face-to-face; over the phone; video conferencing, etc.)Consider whether the category spatiotemporal might be subdivided into the spatial and temporal context. These are two very large categories with many variables.2) In relation to two languages that you are familiar with, discuss how some of these variables might differ across these languages and related cultures (e.g. in relation to the activity context).For example, within the category of multi-modal context, compare how posture, direction of gaze, proximity to the other speaker, head or hand movements that accompany an utterance might differ in two languages. For example Spanish and English. For another example, take social context, across two languages that you are familiar with. Are the role and power relationships exactly the same across the languages? Child - parent; student - professor; patient - doctor; customer - shopkeeper; passenger - bus driver. Is it possible that in one language and culture, there is a closer/more distant or informal/formal relationship across different roles?TASK 1.2: COMPARING WORD FREQUENCY RESULTSThis task requires you to examine Table 1.3 which shows the results for the top 20 most frequent words in three different corpora: the EnTenTen15 (based on internet texts); the BNC1994 (spoken and written British data) and the TED_ed corpus of TED talks.You will notice that from these three very different corpora come three different lists. However, the top five items are the same, though in a different order. This is not surprising because these five grammatical words are essential to constructing most sentences. When you come across a word list that does not have these words in very high ranking frequency positions, it is noteworthy.Circle the words that stand out as being different in their rank order of frequency.The words that stand out will be those that are different when you compare them across the other two lists. You might start by circling any words that only appear on one of the lists. These are: from (only in the EnTenTen15 list); he (BNC1994); we, so, they, have, what (TED_ed corpus).Taking the different types of corpora into consideration, speculate as to why these words that you have circled might be more or less frequent than on the other lists.from in the EnTenTen15It could be speculated that from is more frequent in the EnTenTen15 list because it is drawn from the internet and perhaps there are more people talking about where they are from; perhaps there are more texts that indicate where a document has come from, either a sender or a company. These are speculations and they would need to be followed up using concordances searches.he in the BNC1994We can speculate that because the data in the BNC, which is mostly written data, was collected in the early 1990s there was less awareness of gender neutral language. This would need further investigation.we, so, they, have, what (TED_ed corpus)First of all, it is interesting that this more specific and smaller dataset has more unique words in its top 20. This indicates a more specialised use of language in this corpus. We can see pronouns we and they. Pronouns are always interesting to examine and the fact that we and they are in the top 20 suggests some kind of opposition - between them and us! We also see discourse marker so. This is indicative of spoken language and perhaps informality or perhaps in sequencing an argument or a case. The high frequency of what could relate to its use in rhetorical questions or perhaps in relative clauses. It would need further investigation.If you have access to these corpora, examine the words that you have identified by looking at them in concordance lines. Check if your intuition is correct (see also the section on concordance lines and task 1.4 below where we follow up on this).Test the speculations above in the commentary on (2). Are they correct? Check your own speculations.TASK 1.3: EXPLORING MULTI-WORD UNITSThis task focuses on some high frequency multi-word units (MWUs) which operate as frames (* marks a slot that can be filled by various possible words). A corpus will help you with this task if you enter these items as searches: of the * of it was * at the * of * in the * of on the * of Think of words that might go into the empty slot marked by * and then sort the phrases that you have generated into the following common referential categories. Try to come up with three examples for each category.Here are some possible answers: Referring to timeReferring to placeReferring to somewhere in a textIt was lateAt the end of classAt the beginning of timeIn the midst ofIn the age ofOf the end ofOf the beginning ofAt the time of death It was nearbyOn the top ofAt the centre of townAt the base of treesIt was hereOf the front ofOf the back ofOf the middle ofAt the heart ofIn the back ofon the top ofAt the beginning of chapter 5At the end of page 9In the middle ofOf the table ofOf the list ofOn the back ofOf the middle of Which other category might you add to finish categorising the phrases you have generated?If you look up these phrases in a corpus, you will get a lot of ideas to answer this question. Some examples are: Describing processes and sequencesof the separation ofof the process ofof the flow ofof the first of/of the last of/of the next ofDescribing quantities of the kind of; of the number ofof the mass ofof the hundreds ofEvaluationit was effectiveit was coolit was busyCan you think of longer phrases that these items might frame across these three functions (e.g. It was getting late; In the beginning of the century, etc.)? A corpus will aid this task. Here are just a few examples:at the heart of the matter.at the thought of her son’s dire futurein the age of political correctnessin the midst of her terrorTASK 1.4: CORPUS LINGUISTICS AND LANGUAGE PATTERNSTable 1.6 shows us the most frequent verbs that follow we + have + to in the TED_en corpus. Task 1.4 asks you to examine these verbs and their examples.What hypotheses can you form about possible pragmatically specialised uses of we + have + to + [verb] in TED talks, based on this table?One can observe and hypothesise that we have to + [verb] in TED talks is being used to call for something, to appeal. It is requestive in nature. We could speculate that it is used in calls for change or appeals for action. It is used to try to make change happen on a public forum. In the examples, there are a number of references to climate change and the environment. One could speculate that the pattern we have to + [verb] is being used in public discourse on climate change by activists who are appealing for action and change.How might you follow up on these hypotheses using the TED_en corpus or in another corpus?One could follow up on the hypothesis that we have to + [verb] is a pattern that is found in public discourse in relation to climate change. This could be done searching this pattern in another corpus and comparing the frequencies and the verb patterns compared with the results from the TED_en corpus. It might be possible to observe its use in relation to climate change in a large contemporary corpus. One could take one or two of the actual patterns, for example, we have to do and we have to make and examine the contexts in which they seem to occur in a large general corpus. It may be observed that they occur mostly in public discourse, such as speeches and broadcasts. It may also be observed that we have to + [verb] occur more in spoken that in written contexts. This exploration could be scaled up by building a corpus of speeches by environmental activists.TASK 1.5: COMPARING KEY WORD RESULTSThis task requires you to examine the key word results in two tables. Both lists are generated from the same text but using different reference corpora. The target text was the well-known 1995 BBC1 Panorama television interview by Martin Bashir with Diana, Princess of Wales. You are asked to review and compare the lists and to consider three questions. What are the main differences between these two lists?The list of key words that has resulted for using the corpus of media texts is obviously smaller than the list generated when the Spoken Academic Corpus was used as a reference corpus. This is a good illustration of the importance of choosing your reference corpus carefully and it shows the value of trying out different reference corpora.Apart from the first list being substantially shorter than the second one (which is only an extract of the full list), many of the words in the first list are more specific, especially the nouns.What might account for these differences?The reference corpus used to generate the first list was ‘close’ to the text being examined in terms of its genre. The TV interview between Martin Bashir with Diana, Princess of Wales was of the same genre as the media corpus and so many items did not emerge as key because they are generally frequent in media interviews, chat shows and phone ins. These items include: the pronoun I and related patterns I’m, I’ve, I’d; interviewee use these a lot so they are highly frequent in the media corpus as well and so are not key items.Think. The word think is used a lot in media contexts in patterns such as I think, what do you think, did you think, and so on.Feel. Patterns such as I feel, I didn’t feel are not infrequent in media interviews.Divorce. It might seem surprising that this word is not on the first key word list but divorce is often discussed in media discourse, such as chat shows.Which list is most useful and why?Both lists are useful in different ways. The first list gives a narrow focus in terms of key words. It shows the words that are unusually frequent even when one compares with a corpus of media discourse. The second longer list is useful too because is show what is different about this type of interaction when compared with a genre that is distant, in this case, academic discourse. We can see that words like feel, loved, for example are key when compared with academic data because these are words that are not likely to be frequent in academic lectures. Using these two types of reference corpus allows us to both compare (within the genre of media discourse) and contrast (between media discourse and academic discourse).CHAPTER 2: RESEARCHING PRAGMATICSTASK 2.1: DATA COLLECTION: APOLOGIESIn this task, you are required to reflect on Figure 2.2 which provides a range of options for gathering data. You are asked to consider your options for collecting responses to apologies.The options listed in Figure 2.2 are:Elicitation approachesDiscourse completion tasksRoleplaysQuestionnairesSemi-structured interviewsCorpus approaches based on recorded dataParticipant observation recordingsSpeech recordings (researcher present)Speech recordings (researcher not present)Large corpora of languageGive some specific details on an elicitation method (i.e. what type of task would you use?) and note your thoughts on the main advantages and disadvantages with this approach.See below for advantages and disadvantages commentary. For each of the elicitation approaches, here are some possible task types by way of example. Elicitation approachesDiscourse completion tasksTask: You are walking down the corridor in your university, rushing for a lecture. Someone bumps into you and says: “Excuse me!”. What do you say?Alternative version:Task: You are walking down the corridor in your university, rushing for a lecture. Someone bumps into you and says:, “Excuse me!”What do you say?What do you feel like saying?RoleplaysRoleplay prompt cards for use in pairs. Record the conversation.Role card AYou have used your bank card to pay for your lunch at a local cafe but you notice later in the day that the payment has been changed twice to your account. You go back to the cafe that evening to talk to the manager.Role card BYou are the manager of a local cafe. You have been having problems with card payments. Due to a fault in the card reader, some customers are being changed twice for their bills. Prepare an apology for a customer. QuestionnairesA typical task here is to design a questionnaire that asks participants how they would respond when someone apologies. Provide some scenarios. For example, How do you typically respond when:- someone bumps into you and apologies?- a friend apologises for being late when you have arranged to meet for lunch?- a family member apologises for using borrowing your coat without asking?- a restaurant staff member apologises for your order taking a long time?Semi-structured interviewsA typical scenario would involve getting a small group of people to consent to being involved in a semi-structured interview and to agree to being recorded. In this situation, you could ask them questions about their typical responses. Provide some scenarios. For example, similar prompts to those used in questionnaires (above) could be used:How would you typically respond when:- someone bumps into you and apologies?- a friend apologises for being late when you have arranged to meet for lunch?- a family member apologises for using borrowing your coat without asking?- a restaurant staff member apologises for your order taking a long time?You could ask some more in-depth questions about how they feel about responding to apologies. For example, for each of the scenarios above, you could explore what they would typically say and what they really want to say but do not.In the following grid, find a commentary on the advantages and disadvantages to different elicitation approaches.AdvantageDisadvantageElicitation approachesDiscourse completion tasksHigh degree of researcher control over the variables and the task can be trialed and then improved so that it elicits the language that you are seeking to study.The researcher is controlling the interaction so the control and precision of the task is offset by the degree to which the researcher has ‘manufactured’ the situation that has elicited the language. In other words, many would argue that the language that results from the task will not be natural.RoleplaysResearcher has a good degree of control over the task so as to ensure that the language is elicited. The task can be tested and improved.As aboveQuestionnairesResearcher can elicit the language that they want to study with a good degree of control and they can gain additional contextual and cultural information if they wish.As aboveSemi-structured interviewsResearcher has a reasonably good degree of control and they can gain additional contextual and sociolinguistic information.With multiple speakers, semi-structured interviews will lead to a lot of material to transcribe and sift through. While the researcher has a reasonable degree of control over the content of the interviews, the participants will be talking about the language and introspecting on what they would typically say and how they would say it rather than offering naturally occurring language.Give specific details on a corpus approach to finding examples of responses to apologies (e.g. which corpus would you use?; would you build one?).For each of the corpus approaches, here are some possible task types by way of example. Try to add some more.Speech recordings (researcher present)Set up recordings in a call centre that deals with complaints from customers (with consent and ethical clearance). Record phone interactions over a period of time and observe the interactions that you are recording and take field notes. Record a number of differ speakers over time. Transcribe the interactions to form the corpus. Identify all instances of apologies and the responses of customers to these. Then systematically analyse the apologies, across the different calls. Use your field notes to add extra contextual information. Speech recordings (researcher not present)Set up a recording device at a customer services desk (with consent and ethical clearance) in a hardware shop, an electrical appliance shop and a department store. Record interactions over a two-day period.Transcribe the recordings and manually annotate any instances of apologies that occur. Because you were not present, you may be lacking in contextual details. For example, when a customer says, “I bought this yesterday”. What “this” refers to may not be apparent.Large corpora of languageUsing the spoken BNC2014, search for Illocutionary Force Indicating Devices (IFIDs) for apologies that have been identified by existing speech act research. Sift through concordance lines of the results. To make the dataset manageable, safe your concordances to an excel file and randomly select 100 lines for further investigation.Here are some advantages and disadvantages to each approach. Can you think of others?Corpus approaches using recorded languageParticipant observation recordingsThe researcher sits in and observes and records language, usually over an extended period of time. The language so the context of the language will be understood in great depth, including the physical context, who said what to whom, and so on. The field notes taken during the recordings will be of great benefit when transcribing and analysing the data.The researcher has a very low level of control over the language that emerges. If one is interested in a particular speech act, there is no guarantee that it will be performed within the observation period unless recordings are made in situations where the speech act is typical (e.g. requests at reception desk).Speech recordings (researcher present)The researcher as a participant in the recording gains insight into the context (see above). The researcher may prompt conversation where required.Because the researcher is present and has the potential to prompt conversation, there is a degree of interference.Despite being present at recordings, there is no guarantee that the language one is researching will be used (without interference from the researcher) unless recordings are made in situations where the speech act is typical (e.g. requests at reception desk).Speech recordings (researcher not present)Without the researcher present, there is no question of observer bias or interference. The language will be naturally occurring without any interference,The researcher has no control over the language that is recorded.Not being present at the recording event will leave the researcher lacking in contextual details unless they have carefully allowed for the collection of meta-data (information about the participants and the recording event).Large corpora of languageWith a large sample, there is likely to be scope to find many examples of a target language phenomenon.Despite the size of a corpus, it may be challenging to find certain types of language. For example, it is challenging to find instances of responses to apologies. This cannot be done automatically and manual sifting will be essential.TASK 2.2: DISCOURSE COMPLETION TASKSTry to complete both tasks.Consider how they differ from each other.DCT 1 is short and clear. DCT 2 offers much more context though it is much longer. Look at how you have responded to each task. What are the advantages and disadvantages of each based on this?Examine your responses to both. What do you notice? You may find that you have written a much longer response to DCT 2 compared with DCT 1. When the prompt offers more context, it gives the participant more to think about and more to say/write. This may be an advantage depending on what the research is examining. It’s always worth trying out different prompts before you finalise your approach.TASK 2.3: MULTIPLE-CHOICE QUESTIONNAIRESThink of an insult or other face threatening situation and design a multiple choice task questionnaire using the template below.Here is just an example:Your friend says, “I don’t think long hair really suits you.” Which of the following represents what you would say or do:A) That’s not a nice thing to say.B) Yeah. I think you’re right.C) Look who’s talking! D) Like I care what you think! E) [Use an insulting hand gesture to show your anger at what your friend has said]Use a spoken corpus to search for each of the responses that you have come up with. If the corpus data offers more enhanced wording of your responses, revise them accordingly.Using the BNC2014, to search for That’s not a nice thing to say, you will find, Well that’s not a nice thing to say. Based on this, option A could be amended to Well that’s not a nice thing to sayUsing the BNC2014, to search for like I care, you will see the pattern Like I care I don’t. Based on this we could enhance option D by changing it to, “Like I care. I don’t.”TASK 2.4: ROLEPLAYSIn this task you are required to try out a roleplay prompt so as to see how it works and improve it. This is the prompt. By trying it out, you might, for example, add more context.Prompt A friend of yours is leaving to return to China after an academic year abroad. He has invited you to his farewell party. Unfortunately, you can’t make it. What do you say? Enhanced prompt One of your closest friends is leaving to return to China after an academic year abroad. The friend has invited you to a farewell party. Unfortunately, you can’t make it because you have a family occasion that you have to attend. What do you say?TASK 2.5: INTERVIEWSThis task asks you to find out about the last time someone you know experienced one of the following situations and the aim is to use the details as the basis for a DCT or roleplay prompt. The options are:received a compliment;made an apology;refused an invitation;congratulated someone;made a complaint;insulted someone;apologised for a mistake they had made at work.Here we supply one example of a response and a DCT that is designed based on this experience.Received a complimentMy friend complimented me on my new white sweater. I said, “Oh, it’s just something I got a sale. It was only €15. I love it.”DCT based on this:You are wearing a new sweater that you bought for a bargain price. Your friends says, “I love your sweater”.You say: __________________________________________________________________________________________________________________________________________________________________________________________________________________________________TASK 2.6: DESIGNING A SPOKEN CORPUSIn this task you are asked to consider the following research scenario: You want to build a corpus to address your research question. You have a timeline of seven months for the project. All of the work will be done by you. First, consider the research question below and then review the design matrix for the collection of your data in Table 2.2. Research question: Do pragmatic markers differ across age and gender in Canadian English? Do you think the design matrix adequately meets the needs of the research question? Is there going to be enough data to explore the research question properly?The research question is ambitions because it seeks to look at both age and gender. Two males and two females across three age groups, across a total of one and a half hours of recordings, will not give enough data to address this question. How might you tweak the research question and/or the design matrix so as to ensure that the data will be suitable and to the right scale to answer the research question and to ensure that the project will be completed within the timescale?Narrowing the research question is wise given the timescale. For example, one could focus only on one gender and one could refocus the age ranges. For instance:Do pragmatic markers differ across different age groups of female Canadian English speakers?More speakers are also needed for more robust conclusions. The design matrix in Table 2.2 could then be amended as follows.VariableAge range 120 - 25Age range 270+Gender (number)8 female participants )8 female participants Conversation format4 dyads:4 dyads:Setting(s) of recordingsChatting in the home of one of the participantsChatting in the home of one of the participantsRelationshipFriends chattingMinimum/maximum length of each recording30 minutes per dyad(4 hours in total)Estimated number of words24,000+ words24,000+ wordsLanguage varietyCanadian EnglishRecording typeAudio recordings only.TASK 2.7: TRANSCRIPTION AND THE BNCweb CORPUSThe task asks you to listen to a BNCweb sound file while observing the original transcript and then you are required to compare it with the re-worked transcription:Once the file transcript pops up, listen to the recording of the actual extract (focus in particular of the lines 7746 to 7751). Then compare the broad transcription (Figure 2.9) with the narrower version offered by Rühlemann (2019: 94), based on Conversation Analysis, in Figure 2.10.As we note in the paragraph below this task, when one listens to the recording of this extract one can hear the visceral nature of the interaction, especially when the speakers use the word hate. The speaker really emphasises this. The broad transcription fails to capture this entirely whereas the CA-based re-transcription administered by Rühlemann (2019) adds so much more for the analyst. Note the advantages and disadvantages of the approaches to transcription in Figures 2.9 and 2.10 approach.The main advantage to a broad transcription is that it is quicker to do. When building a large corpus like the BNC, there is neither sufficient time nor funding to do a detailed transcription.The CA transcription, on the other hand, captures spoken language more accurately and in this example we can see that the emphasis that it marks in important to the full understanding of this interaction. In an idea world, all transcription of spoken corpora would be detailed in this way but at least if a solid and accurate broad transcription is provided, and, crucially, the sound files are made available, then any researcher is free to add more detail to the transcript to aid their research.TASK 2.8: PRAGMATIC ANNOTATIONIn this task you are required to consider the extract in Figure 2.12, from a recording of a cookery instruction, led by a chef trainer, taken from the CLAS corpus, a one-million spoken word corpus collected in the context of university-level Hotel Management degree programme. How might you annotate the use of directives with angle brackets tags? See suggestions below in transcript. We have used <dir> to mark the opening of a directive and </dir> to mark where it ends.Are there any other pragmatic items that you might like to annotate? If so, what tag would you use? Discourse markers have also been marked using <dm> to open the tag and </dm> to close it. Notice how many there are in such a small extract!<Chef Trainer> <dm> Now </dm> guys, it’s not Jess’s responsibility to make yer sauce. Ye should be doing it, <dm> so </dm> Jess, <dir> you go back over and look after your bench </dir>. That’s how these boys are going to learn by making mistakes. <$pause/$><Chef Trainer> <dm> Now </dm> <dir> you can grate some cheese </dir>, a small bit of <dir> no don’t use that </dir>, this cheese here is the wrong one, <dm> okay </dm>. You see this here?<Student> Yeah<Chef Trainer> Cloves can be put in a small bowl and left inside in your fridge, the water thrown away and the skins into the bin, <dm> okay </dm>. <dir> You need to clean as you go along </dir>, guys, and <dir> keep it tidy </dir>, it’s how you work <$pause/$>.<Chef Trainer> <dm> So </dm> fifty grams of cheese grated using the small side of the grater <$kitchen noise/$> <$pause/$>. </dir> Keep that knife down by your side when you’re walking with it </dir>.CHAPTER 3: CORPUS PRAGMATICSTASK 3.1: CONCEPTUALISING CORPUS PRAGMATICSIn the context of the statement that, “Corpus pragmatics brings together the fields of corpus linguistics and pragmatics”, this task asks you to consider three questions:List the key methodologies that come to mind in relation to a) corpus linguistics and b) pragmatics.Sample answers:Word and phrase frequency lists; concordances; key word listsDiscourse completion tasks (DCTs); roleplays; questionnairesLooking at this list, discuss how you think there might be links or synergies between these methodologies, i.e. can any of these be used in combination?Sample answers (based on the responses to 1) above):The results from a DCT study can be analysed using corpus software so words and phrases can be quantified and compared automatically; items can be concordanced and key word lists can be generated using a reference corpus to compare the DCT data.Roleplay recordings can be transcribed and analysed using corpus software.Results from a DCT can be searched for in a large corpus so as to recall speech act data. These can then be manually checked.Corpus data and help in the design of DCTs, roleplays and questionnaires by offering real examples on which to base tasks and prompts.Discuss whether any methodologies on your list are challenging to use in a combined way.Questionnaire data is not very amenable to corpus analysis unless one can gather some examples of what people typically say in certain situations and then use these as search items in a corpus.TASK 3.2: A FORM-TO-FUNCTION APPROACH TO SORRYIn this exercise, examine the 20 concordance lines of the word sorry in the BNC2014.How many of these instances of sorry are functioning as apologies?To what degree of certainty can you draw your conclusions in each case?The following functions appear in the 20 lines. An asterisk (*) marks those that are categorised with less certainty:Sorry in hesitations, making clarifications and self-repairs: lines 1, 5, 8*, 10, 11, 13*, 16, 17, 20*;Sorry when seeking clarification: 2, 9, 19; Sorry in apologies: 3, 4, 6, 7, 12, 18*;Sorry in line 15 is a reference to an apology made in text message.TASK 3.3: GETTING STARTED ON FUNCTION-TO-FORM RESEARCHYou want to use a spoken corpus to examine thanking.Think of possible search items.The most obvious items that come to mind are: Thank you; Thanks; Cheers; Thanks a million; Thanks so much.On what basis did you select search items? How do you know these are the right search items?The items listed are the first items that come to mind in relation to thanking (of course, each individual may come up with different variations). Can one really be sure that these are the best candidates? It is best to approach is to look at some previous research on Illocutionary Force Indicating Devices (IFIDs) for expressing gratitude. By looking at existing research, we could add the following to our first intuitive list:Thanks a lot; Thank you so much; Cheers mate; Ta very much; Ta!Try out for these items in a spoken corpus and reflect on your choices.Here are the results for the above search items in the spoken BNC2014. Note that these are just the basic frequencies. One would need to sort these. For example:* marks items that are also counted in the frequency of Thank you;** marks those items that are already counted in the frequency of Thanks; Cheers includes uses which are not expressions of gratitude;Ta includes uses which are not expressions of gratitude.Thank you 3,783 Thanks 1,267Cheers 410Thanks a million 8Thanks so much12**Thanks a lot17**Thank you so much49*Cheers mate21Ta very much 1Ta220Reflection on my choices: Some of the search items have recalled a multitude of occurrences. A way of reducing this number to a manageable size will be needed so as to check through the concordance lines manually to find instances of thanking. The approach shows also that some items that I thought were frequent actually were not.Design an elicitation task that you could use to create a list of search items for thanking.A possible elicitation task: Discourse Completion TaskYou and your friend arrange to meet in the park and eat your lunch. When you get there, you realise that you have forgotten to pack your lunch. Your friend kindly says:Friend: Don’t worry. I have extra here. Have one of my sandwiches and some of my fruit.You are very grateful for this and you say: _________________________________________________________________________________________________________________TASK 3.4: CONCEPTUALISING CORPUS PRAGMATICSRead the extract and note any queries or confusion that arise for you in the absence of the original sound recording.Perhaps you may find the following confusing:Turn 2: Why is someone mentioning a passport and a diary. What do these refer to? The teacher seems to understand perfectly so we can guess that they are referring to materials within the class that are regularly used;The <pause> after turn 3: Did something happen here? Is this when the latecomers entered? How many latecomers were there? Line 4 says, “We had to do, do jobs”. That suggests more than one latecomer. Then turn 6 says, “I’m sorry I’m late”, which suggests one person; There are two pauses at the beginning of turn 7. What is happening in these pauses? Is the teacher issuing a stern look or a stare in reprimand to the latecomers? Is there a hand gesture accompanying the second pause in this turn. What is the tone and pitch of the teacher’s voice for this part of turn 7, You don't look as if you mean it, sit down please because my lesson started ten minutes ago <pause> there's one over there <pause>;When does the teacher turn her gaze back to the whole class and stop addressing the latecomers? Is it at this point in turn 7: “Now in a minute I'm going to give you a printed sheet and it's going to ask you how well you think you're doing”.Review your list (from above) and consider how these issues might affect how you interpret this extract pragmatically.[E.g. How could lines 5, 6 and 7 be interpreted pragmatically in different ways depending on facial expression, intonation and gesture? Could they be interpreted sarcastically? Humorously? Ironically?]The points of ambiguity and possible confusion listed above make it difficult to interpret the force of teacher’s reprimand. Turns 5 and 7 could have been uttered a) with a smiling face in a soft tone of voice, suggesting a humorous reprimand or b) in a high pitch with an angry face and pointing finger indicating a harsh reprimand. Alternatively, its force may have been somewhere in between a) and b). TASK 3.5: COMPARING CORPUS RESULTS IUsing two spoken corpora, search for the forms listed in Table 3.2 (from Schauer and Adolphs, 2006). Remember to use normalized pare your results (you could also compare with those in Chapter 7).See below the results for searches, per million words (PMW), on the BNC2014 and the original spoken BNC (1994). The results were calculated automatically using the Sketch Engine interface.FormFrequency in Corpus 1 (PMW)BNC2014 Frequency in Corpus 2 (PMW)BNC spoken (1994) Thanks 112 .84Cheers 36 .8Ta 19.5 1.9Thank you 335 3Thanks a lot 1.5.06 Thanks very much 4 .2Thank you so much 4 .01Nice one 21 .07Cheers sweetie 0 0 Reflect on the usefulness of this approach and design another study where a pragmatic phenomenon could be explored in a corpus based on an elicitation task. It could be based on the results of an existing study (see Chapter 2 for ideas on elicitation tasks and existing studies).In the case of some of the forms, for example, Thanks, cheers, and thank you, there are many results in the corpus. On one hand, this tells us that these items from the DCT were solid representations of what people frequently say but it will mean a lot of work in sifting and sampling within the corpus. For other language items, there are very few results, for example, Thanks a lot is very low in both corpora and cheers sweetie occurs in neither corpus. This tells us that these results from the DCT are not representative of what people really say. There are also instances where there is a substantial difference in frequency between the two corpora. For example, nice one. This is much more frequent, per million words, in the BNC2014. This raises the question of whether this is a phrase that was not commonly used at the time of the collection of the BNC around 1994 or whether, perhaps, the method used in gathering the BNC2014 data through crowdsourcing resulted in more encounters of thanking.Another issue that these corpus results raise is that some of the items might not be used in thanking. We have not searched through them to find out. It is possible that many of the instances of nice one would be part of structures such as This is a very nice one. Also, the result for Thanks on its own will also count all of the occurrences of Thanks in other phrases such as thanks so much, thanks a lot, and so on.An idea for an elicitation task for thanking is given in the commentary on task 3.3 (part 4).TASK 3.6: COMPARING CORPUS RESULTS IIThis task requires you to look in more detail at the results in Task 3.6. Scroll through some of the concordance lines in the corpora that you have used and examine some examples of thanking. Look closely at the turn taking. Here we look at just three examples.Examine some of the corpus results from Task 3.5 in the concordance lines, with expanded contexts. What do you notice about how turn-taking and expressions of gratitude?Here is an extract from the BNC2014 where three friends are on a train and they are chatting. The the ticket inspector asks to see tickets and then thanks passengers for showing them. In this extract, notice how the thanking is repeated but because it is part of a service encounter, the three friends just continue their conversation amid all of the thanking. S0326.M: railcard is really good S0383.M: I might buy one on Ebay S0328.F: can you get one on Ebay? S0383.M: yeah like a fake ones S0326.M: it's they're yeah they're letting the erm the UNKMALE.M: tickets please thanks very much yeah lovely cheers thank you that's great cheers thanks a lot S0326.M: they've erm they're been outsold by the voice of the Tesco points vouchers for Tesco S0383.M: yeah S0326.M: you get like five hundred pound Tesco voucher for like three hundred and fifty pounds S0383.M:so lame there's loa there's a whole there's so much of that shit on Ebay as well like get like buy one get one free tickets to Alton Towers and shit like that S0328.F: Tesco points The second and third examples come from the spoken component of COCA. Both extracts are from the end of media interviews. Notice how the thanking spreads across turns. This is something typical of naturally occurring examples of thanking that is not picked up by DCTs. It is also context-specific and commonly found in closing sequences in media interviews.Date2017 (171020)TitleAnthony Bourdain On “Appetites,” Washing Dishes And The Food He Still Won’t EatSourceSPOK: Fresh Air 12:00 AM EST DAVE-DAVIES# Anthony Bourdain, thanks so much. It's been fun. ANTHONY-BOURDAIN# Thank you. Date2017 (171203)TitleInterview with Gen. H.R. McMaster; Differences between House and Senate Bill; Interview with Sen. John Barrasso; Pressure Building to Address DACA; Trump^s Foreign Policy; Reflecting on Reagan PresidencyWALLACE# General McMaster, thank you, thanks for your time, and what a great weekend here at the Reagan Library. MCMASTER# A wonderful weekend. Thank you, Chris. WALLACE# Thank you, sir. MCMASTER# Thanks so much. (END VIDEOTAPE) The second part of this task:Reflect on this approach to function to form research. What are the merits? What are the challenges?In this approach, a list of items in task 3.5 were searched in two corpora. The list originated from a DCT task that focused on expressions of gratitude. Looking back at the results in task 3.5 and in part 1 of this task, we can say that:The DCT gave us a range of items to search for. We were able to see how these items differed across different corpora and that was informative in terms of their ‘naturalness’ or in relation to their relative frequency across different varieties of English. It did show up some forms that are not very frequently used;By looking at some of these items more closely, we were able to recall some interesting examples of expressions of gratitude by sifting through concordance line. This showed us the valuable insight that thanking can be repeated / reciprocated and that it can spread across turns;Overall, this approach of administering a DCT and then using the results as the basis for searching a corpus seems to have something constructive to offer. It seems even more useful if more than one corpus is used as it may bring to light variational (e.g. American v British English) and contextual differences (thanking in a service encounter v thanking in the closing of a media interview, as seen above).TASK 3.7: OOPSThe task asks you to use the spoken component of COCA to search for collocates of Oops and then to reflect on whether there is a case for using Oops as an Illocutionary Force Indicating Device (IFID) for apologies outside of blogs.At the time of writing, the top two collocates for Oops in the spoken component of COCA are as follows (see below). From a superficial scan of these results, there seems to be a strong indication that:Oops is a feature associated with spoken language (as one would expect as it is a vocalisation); The word sorry is the 15th most frequent collocate;The first collocate of Oops is a full stop. Note: the software counts punctuation marks as well as words so the most frequent collocate is a full stop;When we click on the full stop collocate, the concordance lines of the full stop collocate shows many apologies as well as examples of self-repairs;By clicking on the first actual lexical collocate, the pronoun I, you will see that the pattern, Oops I did it again, Oops I’m sorry, Oops I made a mistake have multiple occurrences. Based on this initial survey of the collocates listed below, there seems to be evidence that Oops could be a plausible candidate for an IFID for apologies in spoken language. Explore some more of these collocates yourself.TASK 3.8: COMPLIMENTSIn this task, take any four of the following adjectives and multi-word units relating to positive evaluation based on Taavitsainen and Jucker (2008) and search for them in the spoken component of COCA: beautiful, nice, great, lovely, and multi-word units, such as really nice, really great, well done, like/love your, what a, you look/’re looking.Here we take the following four and show a compliment that we recalled from each item:A: You look gorgeous. B: Thank you.CALLER: Yes, good evening PRESENTER: HiCALLER: Sylvia, I love your book PRESENTER: It's a beautiful book, a beautiful book. GUEST: Thank you. Thank you.PRESENTER: You wrote a very lovely book that I read years ago on writing about your own life and Which of the four items appears to be most successful in recalling compliments?From the process of searching, you may have found that the multi-word items seem to recall compliments more accurately. From the four examples used in this commentary, You look and love your were much more productive in recalling compliments. High frequency single word adjectives like beautiful, nice, great and lovely are multi-functional (e.g. as response tokens as well as modifiers of nouns) and so you may have to sift through quite a number of concordance lines to arrive at an example of a compliment.Based on your results, reflect on the usefulness of these this approach to retrieving compliments.It certainly has merit and generated examples of compliments. As discussed above, high frequency single word items are multi-functional and so you may have to sift through quite a number of concordance lines to arrive at an example of a compliment whereas the multi-word units seem to have more specialised usage, at least in this case. It points to the importance of using search items that have proven to be good IFIDs. There is much research into speech acts that can give you direction on the forms that are most likely to be used to perform a speech act and these can form the basis for recalling speech acts in large corpora. The other dimension that is worth considering in this approach is that if the metadata from a corpus is available, then one can explore variables such as age and gender (do certain age groups compliment more than others? do men compliment women more? What is being complimented?) or speech event (e.g. are compliments part of closings in interviews with celebrities?). Looking at compliments in different varieties of English, using different corpora would be interesting in terms of comparing whether compliments are accepted or rejected. In some cultures, it is common to reject a compliment.CHAPTER 4: REFERENCE TASK 4.1: CLASSIFYING REFERENTIAL ITEMSIn this task, we ask you to classify a range of referential items, marked by square brackets [], in an extract from Jane Austen’s novel Emma. It consists of two parts - the first asks you to both classify the items according to their word class (pronoun, verb, adverb, etc.) and then reflect on whether each item is used to reference person, place or time. The second part of the task requires you to consider where we determine the meaning for each item - can the meaning be determined from the text itself or do we have to move ‘outside’ the text to a situational context that we have, in this instance, created in our imagination while reading the novel. Reading a novel requires us to construct an alternative reality where characters move in an imagined space constructed in our minds by the skill of the writer. So, when Jane Austen writes that Mr. Knightley was on horse-back ‘not far off’, we build a picture in our minds of where Mr. Knightley is in relation to Miss Bates. This recourse to an imagined space for the creation of meaning, rather than to the text itself, leads us to the classification of ‘not far off’ as referring to place and having a situational meaning. We have provided the first 10 items as a sample for you in Table 4.2 and some commentary after them. As you can see, some patterns are already beginning to emerge in the classification of items. It is also a useful exercise to compare the classification of items in this task (written language) with the majority of the other tasks in this chapter which involve spoken language. Consider the following extract (4.2) from Jane Austen’s novel Emma. Items that refer to people, space and time have been marked with square brackets.Build a table, such as that illustrated in Table 4.2, to classify these items in terms of their reference to person, place or time and considering how this reference is accomplished linguistically, for example, through the use of pronouns, determiners, adverbs, etc.Consider whether the meanings of the items on your list are determined from the text itself (textual meaning) or from our knowledge of what is ‘outside’ the text (situational meaning). The first two instances have been done for you.(4.2)[Shortly afterwards] Miss Bates, [passing] [near] the window, [descried] Mr. Knightley on horse-back [not far off]."Mr. Knightley [I] [declare]! [I] [must] [speak] to [him] if possible, just to [thank] [him]. [I] [will] not [open] the window [here]; [it] [would] [give] [you] all cold; but [I] [can] [go] into [my] mother's room [you] [know]. [I] [dare] [say] [he] [will] [come] [in] when [he] [knows] who [is] [here]. Quite delightful to [have] [you] all [meet] so! [Our] little room so [honoured]!"[She] [was] in the adjoining chamber while [she] still [spoke], and [opening] the casement [there], immediately [called] Mr. Knightley's attention, and every syllable of [their] conversation [was] as distinctly [heard] by [the others], as if [it] [had] [passed] within the same apartment."How [d'] [ye] [do]? How [d'] [ye] [do]? Very well, [I] [thank] [you]. So [obliged] to [you] for the carriage [last night]. [We] [were] just in time; [my] mother just ready for [us]. [Pray] [come] in; [do] [come] in. [You] [will] [find] some friends [here]."Table 4.2 Classification of referential items according to person, place or time referenceItemPersonPlaceTimeTextual meaningSituational meaningShortly afterwards ? ?passing ?? near??descried??not far off??I??declare??must??speak??him??(task adapted from Biber et al., 1999: 231-232)There are some interesting points to be made when classifying referential items in this instance. The first of these is in relation to the first-person pronoun I. Generally in analysing spoken language, I has person reference, as it has in extract (4.2), and situational meaning. I has situational meaning in spoken discourse as the conversational participants are present and the presence of items such as Brian in utterances such as I, Brian, am going to the shop are unnecessary as the other conversational participants can see from the situation that I = Brian. However, in a novel such as this we encounter a third person narrator and a range of different characters, many of which are given opportunities for direct speech. Therefore, when we encounter Mr. Knightly I declare! In extract (4.2), we must refer backwards in the text to determine who is talking - in this instance it is Miss Bates. For this reason, I has been classified, in this instance, as having textual meaning. It is also worth considering whether the following, in particular, have textual or situational meaning - you in It will give you all cold and we in We were just in time; my mother just ready for us. Then compare whether or not you and we have textual or situational meaning in extracts such as (4.4), (4.5) and (4.9).TASK 4.2: DEIXIS AND THIS STORMConsider extract (4.3), discuss how this storm operates with deictic reference in this stretch of discourse.(4.3)[Context: Recording takes place in the home between intimates. Speakers are numbered according to their appearance in the extract.] <$1> well the various cross channel ferries have been cancelled for today into tomorrow because of this storm<$2> oh right<$1> yeah(Spoken BNC2014: Text S263)This storm can be considered deictic as, although we can infer from the co-text that it might be accompanied by strong winds, given that <$1> states that various cross channel ferries have been cancelled, we have no idea how strong the winds are or the nature of the storm itself unless we experience it first-hand.Tasks 4.3-4.9 examine the traditional categories of deixis and ask you to identify and analyse linguistic items in the context of different language corpora. TASK 4.3: FREQUENCY DISTRIBUTION OF WE IN COCATable 4.3 shows the frequency results for we in the overall COCA corpus in addition to its frequency in the spoken, fiction, magazine, newspaper and academic sub-genres (generated using the ‘Chart’ function on the COCA interface). As can be seen, we occurs almost 4,500 times per million words in the corpus overall.1) Note your observations about the frequencies of we across COCA genres.2) How might the frequency differences be explained? Table 4.3 Frequency counts for the occurrences of we in COCAWith the exception of ‘Spoken’, we can see that we is relatively evenly distributed across the other sub-genres in COCA. Personal pronouns in general are more frequent in spoken language given our concern as conversational participants with one another and our thoughts and actions. These concerns are also present in fictional dialogues as we have seen in Task 4.1, so it might have been expected that the frequency of we would be higher than in Table 4.3. This result may be due to the texts that were chosen for the ‘Fiction’ sub-genre in COCA, which is composed of literary magazines, movie scripts and some first chapters from novels. A corpus comprised entirely of novels, for example, might yield a higher frequency of we. Magazines and newspapers, while concerned with issues such as establishing an in-group identity, especially in opinion pieces, also cover areas such as business, sport, lifestyle, etc. where the use of we may be less prevalent. Finally, we has what might be considered an unusually high frequency is academic texts. This is due to its multifunctionality within academic writing to make generalisations, and exclusively refer to the author(s) or inclusively to the author(s) and reader(s). TASK 4.4: IDENTIFYING PLACE DEICTICSThis task is similar in nature to Task 4.1 and it might be useful to go back and forth between the two, comparing and contrasting your answers. The speakers in extract (4.10) are looking at the drawing or diagram of <$2>’s allotment. Extended context for this extract is available from the CQPweb interface. Identify the linguistic item(s) that represent spatial/locational deixis in the following extract (4.10).(4.10)[Context: <$1> = female (age 34); <$2> = female (age 33)]<$1> oh wow look at that<$2> it’s quite big actually<$1> what your allotment or the plan?<$2> yeah a bit well both it's not to scale erm so the gate would be here<$1> okay<$2> and then<$1> is it all fenced off then?<$2> so there’s a fence round the outside<$1> is there? <$2> yeah<$1> oh that’s very London<$2> erm well yeah there’s a gate with a padlock<$1> is there? <$2> and but that’s cos so the allotments are all dotted around er this estate so there's m-mine is here there's one behind<$1> mm<$2> and then there’s two next to us so <$1> okay<$2> so (.) erm the gate's here(Spoken BNC2014: S2TP)This extract contains a number of spatial deictics. For example, here is used spatially in the gate would be here and m-mine is here and the gate’s here where the speaker points to the plan. There are also a number of prepositions and prepositional phrases used spatially such as behind in there’s one behind and next to in there’s two next to us, the meaning of which can only be determined by looking at the plan itself. TASK 4.5: IDENTIFYING TIME DEICTICSHere, we return to the realm of written language in order to explore markers of time deixis.Identify the linguistic item(s) that represent time deixis in the following extract (4.12). (4.12)That points to a complex evolutionary story. Two cognitive scientists at Lund University in Sweden, Can Kabadayi and Mathias Osvath, conducted a series of experiments with five captive, hand-raised ravens. Obviously, that's not a lot of ravens, and hand-raised ravens do not behave like wild ravens. But when it comes to figuring out the outer bounds of cognitive abilities for a species, those aren't the most important problems to worry about. Testing more ravens, and wild ravens, comes later. Jelly bean now, or burger tomorrow? First, the ravens had to be given an experience they could plan for. They learned how to operate a puzzle box that opened to yield a reward, but the box could only be opened with a specific tool. The next day, they were shown the puzzle, loaded with food, but no tool. Only an hour later they were shown a tray of objects, including the tool, and given the opportunity to choose just one thing. Fifteen minutes later, the puzzle came…(COCA Magazine Section: Ars Technica (publication date 17/07/2014)) Again, when considering this piece we have to bear in mind that it is written, as opposed to spoken, language. This means that the coding time (CT) and reception time (RT) are markedly different given the time lapse between when something is written and when it is read. When reading the piece, although our understanding remains unaffected, we do not know the exact time reference in relation to items such as later, now, tomorrow or an hour later. TASK 4.6: DAYS OF THE WEEKUsing a spoken corpus, such as COCA or the Spoken BNC2014:1) Search for the frequencies, per million words, of the following and note the totals:a. the lexicalised days of the week.Day of the weekLCIECOCASpoken BNC2014Monday11116105Tuesday1001365Wednesday85986Thursday1061081Friday17223146Saturday20215125Sunday16923125TOTAL945109733b. the items yesterday, today and tomorrow.Day of the weekLCIECOCASpoken BNC2014today 549208299yesterday 38936146tomorrow29638216TOTAL 12342826612) Compare the results to those in Table 4.4. Even a rudimentary frequency search such as this one yields a number of results that might be worthy of further analysis. For example, in relation to the lexicalised days of the week, although Friday, Saturday and Sunday occur most frequently in all three corpora, how might the frequency discrepancy in COCA be explained?In a similar vein, today is most frequently used in all three corpora. However, the frequency of occurrence is again much lower per million words in COCA. TASK 4.7: IDENTIFYING DISCOURSE DEICTICSExtract (4.17) is an extended piece of writing from an essay contained in the British Academic Written English (BAWE) corpus (see Appendix 1). The extract is from a history essay that explores the rise of modern China.Read the piece and identify the discourse deictics. Remember, the realm of discourse deixis moves beyond anaphora and cataphora, to the level of sentence, idea, or in the case of written academic discourse, argument. (4.17)Despite the consolidation of power during the reign of the Emperor Kangsi to Qianlong, there remained however, certain underlying grievances that were never fully eradicated…Though westerners have had a presence in China dating back to the Yuan dynasty, the Opium Wars and the corresponding treaty system marked a watershed for China-West relations, resulting in the change of status quo, of perceptions; and manifested itself in the increasingly intrusive presence of the West…It is in this context that the above question should be addressed, and the writer will examine firstly the internal conditions and entrenched corruption which made it conducive for the outbreak of social unrests, followed by the implications of change brought about by the West, such as the desire to trade, introduction of new technology…and in turn produced a virulent anti-foreign nationalism which threatened the social order during the period in question. Hence the writer will argue that the social unrests at the beginning of the 19th Century was due more to the internal rather than external factors; but increasingly, the presence of the West is crucial to the emergence of violent xenophobic nationalism, which account for the social unrests in the later part of the 19th Century.(BAWE: File 0019j)Discourse deictics in this piece of academic writing include items that have featured in Chapter 4 such as this and/or that such as the use of this in it is in this context. Other discourse deictics commonly used in academic writing include the pragmatic markers however, in turn and hence as these markers indicate a relationship between the sentence in which they occur and prior sentence(s).What category of deixis does the writer (used on two occasions in extract (4.17)) belong to?The writer is used in (4.17) to refer to the author of the essay, rather than using say, an author pronoun such as I or we. Therefore, the writer is an example of person deixis.Examine the use of items identified as discourse deictic (in this text) in other corpora, such as the academic component of COCA, and explore whether or not these items are used consistently as discourse deixis.This task is largely a self-directed one with no right or wrong answers. It is designed to encourage the use of corpora and corpus interfaces such as COCA in order that you become comfortable with both the mechanics of using these resources and then progress to the analysis of the results generated. TASK 4.8: SOCIAL DEIXIS AND CONTEXTExamine this extract (4.19) which is part of a conversation in the Spoken BNC1994.Based on the social deictics that are used (marked in bold), what might be the context of the speech situation?(4.19)<$1> Chrissy?<$2> Mm?<$3> Want some turkey?<$1> Do you want some turkey?<$2> No thanks.<$1> Did you David?<$2> I've got hazelnut and ham haven't I?<$1> Mm mm. [unclear], one of those. And some of that. I didn't wash all this so there's<$3> Well I must admit<$1> There's more if you need it.<$3> It tastes better if it's washed and put in that, in that cling film.<$1> Mhm. <$3> It keeps better than it's left in the bottom of the fridge. Has David got some coleslaw? You will eat your coleslaw honey won't you?<$4> Mhm.<$1> He's got some but he's not very enthusiastic.<$3> I thought you said you liked it?<$4> I go off and on it, but<$3> Oh.<$4> Sometimes I do, and sometimes I don't.<$3> Right.<$1> Ow!<$3> But it, it doesn't keep forever. When did I make it? Saturday. So it wants using.<$1> I'm doing my best, you know ma=<$3> Well, I've made some more since then.<$1> You know me and coleslaw. Oh, so I don't have to rush it?<$3> Well, not that. I'd prefer it to go.<$1> Have you got everything you want kiddies?<$4> Yeah.<$3> Yes. And a bag of crisps, got the same packet. Boys, when we next see Gwyneth will you be nice and quiet and kind to her, her Mummy's just died?(Spoken BNC1994: File KCH)The answer to question 1 is a pretty straightforward one given the social deictic items - the context is obviously a family at the dinner table.Use the social deictics, and any other linguistic clues, to determine ‘who’s who’ in the conversation. This question is slightly trickier but there are some clues for us in addition to the social deictics. <$1> and <$3>, in addition to using the social deictics kiddies and boys, are also performing a number of roles such as ‘host’ and ‘behaviour monitor’ and these indicate that <$1> and <$3> are the parents. The final utterance in the extract where the children are instructed on how to behave when they next see Gwyneth, sees <$3> in the role of ‘behaviour monitor’ or ‘moral guardian’ and these roles have commonly been associated with the mother of the family. <$2> and <$4> are the children and, using the social deictics Chrissy (short for Christopher in this instance) David and boys, we can see that both <$2> and <$4> are male. The items kiddies in particular points toward younger children (the Spoken BNC1994 metadata reveals that Chrissy is six years old and David is nine years old).TASK 4.9: THAT + NOUN IN COCAWith the findings from section 4.5 in mind, examine Table 4.6 which presents the top 20 most frequent that + noun patterns in the spoken component of the COCA corpus (these results were generated using the search that _nn*). Some of the results, for example, that students or that women, are obviously part of a that-clause. Some of the results, for example, that kind, that sort or that case are most likely part of larger multi-word units. However, many of the patterns are similar to those evident in Table 4.5, such as that + day, night or money, for example.Use the online COCA interface to analyse a downsample of 100 occurrences from one of the patterns from Table 4.6.Determine whether or not there might be an empathetic deictic dimension to the use of the item you have sampled.Table 4.6 Top 20 frequency results for that + noun (that _nn*)NFORMFREQUENCY1that way279852that people270173that time239324that night181305that day174796that kind155447that point136238that moment88019that students765010that women711211that year696612that question573713that part552314that money540415that person535616that morning530517that sort471618that children437319that information432820that case4293In order to provide a sample answer for this task, we have focused on a sample of occurrences of that person in COCA and some patterns of use do seem to emerge from the data. There seems to be a lot of language associated with law and legality surrounding the use of that person in the spoken component of COCA: A quick scan of the lines generated reveals the presence of language associated with law enforcement in the US. For example, in the first concordance line we see the FBI director is investigating someone and that person fires them. Similarly, we encounter lines such as that person is not reported to the feds; even if they arrest them, that person when convicted and they’ve already made one arrest of that person who was responsible for renting this van. That person also co-occurs with language of violence and loss - that person would have been taken somewhere and roughed up; incredible hatred toward that person; once that person lose their rights and I’ve talked to the family of that person. They lost all of their life times. Obviously, this is a limited sample and more concordance lines need to be examined but, in these instances at least, it does seem that there is the possibility of a negative empathetic dimension associated with the use of that person. CHAPTER 5: POLITENESSThe first task in Chapter 5 is designed to test your intuitions about polite language and the different situations in which different linguistic items associated with politeness are used. This task can be performed for any language, and is not restricted to English. TASK 5.1: POLITE LANGUAGEWhat words do you most commonly associate with politeness? In what situations do you most frequently/least frequently use them?Very generally, some of the words that might be commonly associated with politeness in the English language include please, thank you, thanks, cheers, excuse me or sorry. Modal expression such as could, may and would are used to construct, for example, indirect requests. There are also a variety of strategies such as vague language and hedging that are used in politeness. Finally, the past continuous tense is sometimes connected with politeness, for example, I was wondering if you could tell me… or I was hoping to get an appointment… (there are a number of verbs associated with this grammatical structure such as wonder, hope or think).In terms of the situations in which these items are used, there are a number of factors that need to be considered: the power difference between speakers (boss-employee, staff-customer, teacher-student, etc.), the length of time speakers have known one another, age difference, the medium being used (spoken, written and computer mediated communication may have different politeness requirements) and there are also inter- and cross-cultural norms to be considered both within the same language and between speakers of different languages. There may also be other factors that you have taken into account that we may not have listed here.TASK 5.2: THE USE OF PLEASEExamine the frequency results for the occurrences of please (per million words) in the Spoken BNC2014 and the spoken component of the ICE-Ireland corpus in Table 5.1.Table 5.1 Frequency counts per million words for the occurrence of please in Spoken BNC2014 and ICE-Ireland Spoken BNC2014ICE-Ireland (Spoken)please244.51116.2Consider the differing results for please. What might explain the difference? Does it mean that British people are more polite than Irish people?It is unlikely that, based on a broad stroke comparison such as this, that we can come to any conclusions about whether or not one culture is more polite than another. The difference is more likely explained when we examine the design of the two corpora. The Spoken BNC2014 is comprised of conversations between speakers in primarily informal settings whereas the spoken component of the ICE suite of corpora contains conversations from both formal, for example business transactions and legal cross-examinations, and informal settings. It is possible that some of the more formal spoken categories do not require the use of please. This demonstrates the importance of considering corpus design before making any generalisations either within or between corpora.ICE-Ireland was collected from the early 1990s (from both Northern Ireland and the Republic of Ireland) whereas the Spoken BNC2014 was collected between the years 2012-2016. What might this tell us about the frequency differences in the use of please?Interestingly, as we discuss in Chapter 7, please has shown an on-going rise in frequency of use since the 1980s. This may have had an impact on the results in Table 5.1 given that the Spoken BN2014 is approximately 20 years ‘younger’ than ICE-Ireland. Can you think of any ways in which corpus tools could help to further investigate the difference between the results? (see our discussion of cheers in Chapter 7)One way of further investigating the results is to expand on whether or not some of the more formal spoken categories in ICE-Ireland feature the item please. This can be done by running frequency counts for a selection of categories:Category Frequency of pleaseBroadcast discussion0Business transaction5Classroom discourse0Face-to-face conversation41Legal cross-examination 6For example, we can see that please does not occur in either the category of broadcast discussion or classroom discourse and is at relatively low levels in both business transaction and legal cross-examination. In contrast, the category of face-to-face conversation contains 41 instances of please. The Spoken BNC2014 is exclusively comprised of face-to-face conversation. TASK 5.3: POLITE LANGUAGE USE IN DIFFERENT SITUATIONSTask 5.3 is linked to both Tasks 5.1 and 5.2 but focuses on the use of a specific linguistic construction or act (see Chapter 6), that of a request. We have seen how different speech situations and categories require different politeness strategies. Consider how you make a polite request in different situations.How do you make a request in your workplace, as opposed to at your family dinner table?We have at our disposal a range of syntactic structures available to us in order to construct a request and we can vary these according to the situation:Get me a cup of tea.Get me a cup of tea please.Get me a cup of tea, love/mum/John.Would you mind getting me a cup of tea? Would you mind getting me a cup of tea, honey/dad/Kate?Are you by any chance passing the canteen? In that case, could you possibly get me a cup of tea please?Consider the person to whom you are making the request. How might the language of the request vary depending on whether the request is to your boss, your colleague, your parent, your sibling or your child?Generally speaking, more indirect requests are considered more polite and are thus associated with more formal situations such as if an employee was making a request of their boss then the choice might be at the more indirect end of the spectrum. However, imagine if you have known your boss for a long time and are quite friendly with them outside the workplace. This might affect the choice you make. Similarly, the more direct strategies are often connected to the more informal situations, and, therefore, you might find yourself using a direct strategy with a sibling, for example. However, imagine a parent who wishes to teach their child politeness through the use of an indirect request strategy. Although there is no reason why they cannot use a more direct one, they choose, for different reasons, to use an indirect one. Therefore, it depends on a large number of factors as to the linguistic choice we actually make. The remainder of the tasks in this chapter consist of applying the models of politeness to corpus data. The first of these, involves identifying a range of Brown and Levinson’s positive politeness strategies in an extract from the OANC.TASK 5.4: POSITIVE POLITENESS STRATEGIESExtract (5.8) is taken from the Switchboard component of the Open American National Corpus (OANC) (see Appendix 1). Read the extract and work out how many distinct positive politeness strategies are being used by the speakers. (5.8)<$1> could you hang on one minute Jim thank you. <$2> yes ma'am. <$1> I am so sorry to keep you on hold are you at work okay I am too. <$2> no problem yes yeah um I think what you have is the way the. <$1> um. <$2> the justice system works is they bend over backwards trying to protect the guilty so many things in their back pardon yes I know. <$1> oh it's insane it's insane. <$2> my wife participated in a jury trial several years back wherein the individual after it was over and they had came up with the maximum sentence. <$1> yes yes. <$2> in the jury form they found out that uh the gentleman involved had a long history of the same type offense which was theft of uh property.<$1> right uh-huh oh yes.<$2> and yet the district the attorney for the prosecution could not enter these uh the good man's background into it it was like this was a first time offender. <$1> yes.<$2> whereas uh I think that's those that should be in jail should be in jail.<$1> it seems huh that's correct and there's too much leniency there's too much uh um underhanded uh things going on that the public are not aware of. <$2> yeah.<$1> and it seems as though we are uh giving the criminal the benefit more than the victim the victim has been victimized twice not only by the perpetrator of the crime but also the courts that try to do the justice.<$2> well I think this is one of the reasons that also that attorneys have such a a unsavory reputation shall we say. <$1> I do not have any I don't have an ounce in my body of credibility toward the the judicial system in the United States.<$2> yeah well what you got is the situation wherein uh if you ever get in trouble you want to hire the smartest crookedest lawyer you can find. <$1> isn't that terrible.<$2> and uh that's not the idea it's the idea is not to get the guy off of the crime that he committed but to punish him for the crime he committed.<$1> well to punish him for the crime but also to try and reeducate the man's thinking.<$2> yeah. <$1> to get him out of that uh uh criminal mind that he has and to direct him into a more productive life<$2> yeah well we have uh a situation again that I am familiar with where uh the son of an acquaintance of mine was killed while trying to stop a robbery. <$1> um oh yes how sad.<$2> and the person that killed him was a young woman who had left the house with full intent to commit crimes carrying a gun in her purse.(OANC Switchboard: File sw-3095-ms98-a-trans)We can see a range of positive politeness strategies at work here, especially on the part of <$1>:Strategy Example from Extract (5.8)Pay attention to a hearer’s interests, wants, needs or goods<$1> I am so sorry to keep you on hold are you at work okay I am too.Exaggerate interest in, approval of or sympathy with a hearer<$1> oh it's insane it's insane. Use in-group identity markers (solidarity address terms, dialect features, slang, contraction)<$1> could you hang on one minute Jim thank you. Seek agreement/make small talk<$1> I am so sorry to keep you on hold are you at work okay I am too. Assert or imply knowledge of and concern for a hearer’s wants<$1> it seems huh that's correct and there's too much leniency there's too much uh um underhanded uh things going on that the public are not aware of. Use inclusive ‘we’ forms<$1> and it seems as though we are uh giving the criminal the benefit more than the victim the victim has been victimized twice not only by the perpetrator of the crime but also the courts that try to do the justice.TASK 5.5: HEDGING IN DIFFERENT CONTEXTSFigure 5.3 illustrates the cumulative results for the occurrences of the hedges just, really, actually, probably, I think, a bit, kind of, sort of, you know and I suppose in different context types in the Limerick Corpus of Irish English. As the figure demonstrates, there is a difference in frequency of occurrence across the five different context-types. Examine the frequency differences and answer the questions that follow.Figure 5.3 Distribution of hedges in LCIE (normalised per million words) (Farr et al., 2002)Why do service encounters contain the lowest number of hedges?The aim of many service encounters is to fulfil a transactional goal and it is likely that the speakers do not previously know one another (cf. Task 5.7, however). Therefore, this goal, and the relationship between speakers, creates a situation where an efficient transaction characterised by low levels of mitigation does not cause any threat to face.Why are there fewer hedges in conversation in the family than in conversation between participants in radio phone-in?The clue to the answer to this question again lies in the nature of the speaker relationship. Similar to service encounters, radio phone-in participants do not usually know one another but it appears that, in contrast to service encounters, face is important in radio phone-in. For example, questions need to be asked by the presenter and these threaten face and so require downtoning or mitigation. In addition, these questions need to be answered and callers frequently mitigate any offence that might be caused. Families, on the other hand, are characterised by fixed, stable, established speaker relationships. This does not mean that face is unimportant to family members, but that face needs are attended to in different ways than those evident in Figure 5.3. Why are hedges most frequently used in teacher-trainer feedback?Teacher-trainer feedback contains the most hedges because, although the relationship between teacher-trainer and trainee is asymmetrical, the teacher-trainer actively seeks, using hedging, to mitigate their power in order to encourage the trainee. At the same time, the trainee also hedges to acknowledge the power differential that exists in the relationship - for example, a trainee will not often directly disagree with the teacher-trainer but instead will soften the disagreement. TASK 5.6: MARKERS OF IMPOLITENESSThe purpose of this task is to investigate Culpeper’s list of personalised negative vocatives connected with insults. The idea is to examine their use in face-to-face interaction in order to determine how frequently they are used by one speaker to directly insult another. In Culpeper (2011a), a number of personalised negative vocatives are connected with the impolite formula type insult. Table 5.5 lists a number of these items and their frequency of occurrence in the Spoken BNC2014. We have only included counts for the singular use of each and have limited the items to those with a frequency of between 10 and 100 occurrences.Use the CQPweb interface () to search for some or all of the remaining items and, using the table below, determine the number of times they function as markers of impoliteness or whether they have different functions (we have classified the 26 occurrences of moron as an example).By way of illustration for this task, Table 5.5 shows that there are 22 instances of the item slut in the Spoken BNC2014 and the concordance lines for each of these occurrences are presented here. In terms of applying our categories from Table 5.5 to these concordance lines, we get the following results marked in bold and shaded in the table.Table 5.5: Negative vocatives that can be used in insultsItem (frequency)Impoliteness markerTalking about a third person not presentMarker directed at (hypothetical) selfOtherbastard (93)berk (10)dickhead (56)liar (56)loser (32)moron (26)41714plonker (10)slut (22)2956sod (56)squirt (11)We can see that, as with moron, a speaker calls another speaker conversational participant a slut on only two occasions (lines 13 and 20). Again, talking about a third person who is not present in the conversation is the most frequently used category (for example, lines 2, 4 and 5). In relation to this category, it might be that the speaker, although talking about a third person, uses the word slut to deliberately be impolite as they be aware that, for example, another speaker in the conversation disapproves of the use of the word. Another feature of the concordance lines containing slut is that there are a number of instances of reported speech - for example, lines 4, 6 or 16 - where people appear to use slut as an insult in face-to-face encounters but these events have not been captured by the corpus. Again, perhaps we are looking at issues in corpus design here - it could be unlikely that people who know one another well and have built up a relationship with one another over a period of time are using insults such as slut or moron, at least not very frequently. Determine whether or not the categories provided in Table 5.5 are suitable for the classification of some or all of the other linguistic items. Our categories are by no means definitive. Feel free to expand on these categories or add others where necessary. Although we have only analysed two insults thus far, the categories appear to be more or less suitable for our purposes. However, in our results from the analysis of slut, the ‘other’ category has six entries. These are lines 12, 14, 16, 17, 18 and 21. Lines 16-19 are all from the same conversation which reports on a conversation where someone has equated pansexuality with being a slut. We could argue that in this instance, although there is no evidence that the speaker is being directly impolite towards another speaker, it is insulting to call a pansexual person a slut and, therefore, we might add these instances to the ‘impoliteness marker category’.TASK 5.7: POLITE VERSUS POLITIC BEHAVIOURTable 5.6 illustrates the frequency counts for the use of thank* (thank* includes occurrences thanks, thank you, thanks very much, etc.) in service encounters in the LCIE corpus where the employee and customer are strangers (not acquainted) or familiar with one another (acquainted). The frequency is given in the form of a percentage value which means that, for example, in the ‘not acquainted’ category, thanks* represents 4% of the total frequency of the words used in the corpus. Table 5.6 Frequency counts for thank* in service encounters in LCIENot acquaintedAcquaintedWordFrequencyWordFrequencythank*4%thank*1.2% As we can see, there is a notable drop in frequency between when the employee and customer are strangers as opposed to when they are acquainted with one another.How might this finding be reflected in the politic and/or polite language used between customer and employee?What has probably emerged in this instance is that politeness is renegotiated between the customer and the employee as their relationship develops and this results in the relatively infrequent use of thank* becoming equated with politic behaviour in their interactions. Does this mean that they are less polite with one another as the relationship changes?It seems unlikely that the employee and customer become less polite with one another as the relationship changes. Other norms of what constitutes (im)politeness most likely emerge as the relationship develops.How might this be extended to other (im)polite behaviours in service encounters? (Think back to Task 5.3 and our consideration of polite language use in different situations.)It would be interesting to examine the occurrence of other items in service encounters such as those outlined in Tasks 5.1 and 5.3 in order to ascertain what constitutes politic and polite behaviour in this context. CHAPTER 6: SPEECH ACTSIn the first task you are asked to differentiate between the use of the verb suggest as an explicit performative in an academic context. In order to do this, you should consider both the explicit meaning of suggest, to put forward a plan or idea for someone to think about, and also the meaning of suggest in an academic context which is often used as a reporting verb to indicate tentativeness which, in turn, mitigates the presence of disagreement or the presentation of new knowledge. This tentativeness is a conscious act on the part of the speaker/hearer as both disagreement and the presentation of new knowledge are often mitigated in an academic context. TASK 6.1: PERFORMATIVE SUGGESTHow many of these examples of suggest are functioning as explicit performatives, as defined by Austin?Figure 6.1 20 randomly generated concordance lines for suggest in the BASE (sorted 1L)Instances of suggest functioning as an explicit performative include lines 4-7 where the personal pronoun I occurs one item to the left of the node. Indeed, the presence of a pronoun before the verb often indicates its use as an explicit performative. In contrast, the use of suggest as a reporting verb, for example, lines 2, 3 or 19, are preceded by noun phrases such as figures or group(s) where suggest, rather than functioning as a performative, indicates tentativeness in presenting disagreement or knowledge. What other linguistic forms might be used to make suggestions in an academic context (see Task 6.5)?There are a variety of forms available to make suggestions in an academic context. For example, Why don’t you ...?; Let’s…; How about… or You should...TASK 6.2: FELICITY CONDITIONS FOR PROMISESOn the surface, this method seems to have the potential to distinguish one speech act from another. However, as you may discover while reading about speech acts in general, the boundaries between different acts tend to become blurred (see Task 6.5). For example, the same act could be interpreted as a request, suggestion or threat depending on speaker intention. Nonetheless, felicity conditions are useful in understanding aspects such as speaker and hearer roles in speech acts. In Figure 6.3, we have begun the process of distinguishing the felicity conditions for promises by identifying the propositional content and the sincerity condition.What are the preparatory and essential conditions that must be met in order to distinguish an act as a promise?Figure 6.3 Felicity conditions for promises (adapted from Culpeper and Haugh, 2014)Felicity conditions for promises[A = act, H = hearer, S = speaker]Propositional Content:Preparatory: Sincerity:Essential: Future A (act) of S (the speaker)H (the hearer) wants S to perform AIt is not obvious that S will do A in the normal course of eventsS intends to do ACounts as an undertaking of an obligation on the part of S to do ATASK 6.3: SPEECH ACT CLASSIFICATIONAlthough things are changing, pragmatically tagged corpora are still the exception to the rule, and commercially available ones even more so. Therefore, the chance to explore a corpus tagged in this way provides us with an excellent opportunity to examine the frequency of speech act categories in different speech situations. Consider Table 6.1:Table 6.1 Distribution of speech acts in three speech situations in SPICE-Ireland (normalised per 10,000 words)Speech situation<rep><dir><com><exp><dec>BRN(broadcast news)996.544714.52CLD(classroom discourse)920391.122.831.70.5FTF(face-to-face conversation)1670.7629.127.355.40.4With the exception of declaratives, why might the other speech act categories be more frequent in face-to-face conversation than in either classroom discourse or broadcast news?What characteristics of the speech situation classroom discourse might make it the ‘middle ground’, in terms of speech act frequency, between broadcast news and face-to-face conversation (see also Chapter 8)?In essence, although the two questions are different, they can be addressed with one answer. Broadcast news is usually characterised by a one-way flow of information from presenter/anchor to audience. Therefore, it can be reasonably expected that category of representativeness dominates the speech situation as it is not necessary for the presenter/anchor to, for example, get the audience to do something (directives). Classroom discourse, on the other hand, is generally characterised by interaction that alternates between teacher and students(s). This relationship is an asymmetrical one and directives are expected in this situation, though they may primarily originate from the teacher. The presence of some commissives, which commit the speaker to some future action, and expressives, which express the speaker’s psychological state, is to be expected. Finally, face-to-face conversation is generally dialogic in nature and often characterised by more symmetrical power relations. In many cases, face-to-face conversation in spoken corpora takes place between people who know one another well such as family, close friends or work colleagues. Often, although arrangements may be made and stories narrated, the only goal of face-to-face conversation is the conversation itself. These characteristics result in the distribution of speech acts evident in Table 6.1. The dialogic nature and symmetrical power relationships result in a high degree of both representatives and directives as no one participant has control of the conversational floor. Also, it is reasonable to expect commissives and a wider range of expressives as speakers may have more opportunity to perform this speech act category. TASK 6.4: CAN YOU…? VERSUS COULD YOU…?Aijmer (1996) states that in addition to sentence type, felicity conditions and the type of subject (for example, the choice of you as subject in an interrogative rather than I in a declarative), other strategies are used to mark a request as polite or indirect. Using the examples of Can you…? and Could you…? taken from LCIE in Figures, 6.4 and 6.5, determine what other strategies, if any, are used in conjunction with Could you…? to mark it as a more indirect request strategy than Can you…?.Figure 6.4 20 concordances lines for Can you…? in LCIE (sorted 1L)As we can see from figure 6.4, using the Can you…? strategy, although more indirect than a strategy such as Do X, still results in a fairly bald question type with little or no mitigation present, either before or after can.Figure 6.5 20 concordance lines for Could you…? in LCIE (sorted 1L)The Could you…? request, on the other hand, features a range of strategies used in conjunction with the structure which marks it as more indirect. Vague language is present in line 1 (a few beers), line 3 (some examples) and line 7 (do something about that). The pragmatic marker (PM) like occurs as part of the request in lines 8 and 17 and the Irish English PM now occurs in line 20. In Task 5.4, we have seen that slang can be used as a marker of positive politeness and this strategy is evident in line 6 where the speaker uses the term snaz. The politeness item please is also present in lines 18 and 19. Also of note is that could begins a speaker turn on six of the 20 occurrences in Figure 6.5 whereas can begins 16 of the 20 turns evident in Figure 6.6. The strategy of a request occurring further from the beginning of a turn has the effect of mitigating the face threat associated with directives. TASK 6.5: SUGGESTIONS IN MICASEThis task is designed to encourage you to question whether or not the use of why don’t you or why don’t we constitutes a request or a suggestion. As we have seen in Task 6.4, we can use corpus tools such as concordance lines to help us find patterns or strategies associated with the use of a particular structure that allows us to differentiate one strategy from another. Table 6.2 illustrates that the frequencies of the multi-word units why don’t you and why don’t we in MICASE are quite similar. Table 6.2 Frequency counts for why don’t you versus why don’t we in MICASEFormFrequencywhy don’t you60why don’t we77Using the instances of the units in Figures 6.6 and 6.7, determine why these units illustrate some of the difficulties associated with Searle’s identification and classification of speech acts. In other words, building on our use of corpus tools thus far, what we would like you to do is find patterns that might distinguish request-why don’t you/we from suggestion-why don’t you/we.Figure 6.6 15 randomly chosen concordance lines for why don’t you…? in MICASEFigure 6.7 15 random concordance lines for why don’t we…? in MICASEIn terms of patterns, there are one or two potential patterns to the left of the why don’t you/we node in Figures 6.6 and 6.7 such as the discourse marker so and the co-occurrence of like with why don’t we. The more frequent patterning, however, occurs to the right of the node. For example, the pragmatic marker just appears immediately to the right of both why don’t you and why don’t we. Just has generally been shown to collocate with suggestion-why don’t you/we in order to mitigate the imposition associated with the suggestion. However, as we have discussed in Task 6.4, requests are also frequently mitigated. Therefore, the issue raised in this task is whether or not we are looking at instances of request-why don’t you/we or suggestion-why don’t you/we? In addition to the pragmatic marker just, verbs that involve interaction, such as ask, get, tell or use, have also been shown to collocate to the right of suggestion-why don’t you. Although these particular verbs do not occur with any great frequency in the concordance lines here, verbs that have interactive meaning such as share, talk, explain or say, do occur. These, coupled with the occurrences of just, could point towards a suggestion-why don’t you/we classification. On the other hand, verbs that indicate internal dispositions such as volition or obligation, for example like, want or have to, have been associated with request-why don’t you/we. The verb write, the writing process does not involve much social interaction, might be a clue to the classification of an instance of why don’t you/we as a request. These are the corpus analytic processes that we should engage in in order to ensure that the process of differentiating between particular speech acts becomes clearer. These processes also have a lot of pedagogic implications. CHAPTER 7: PRAGMATICS AND LANGUAGE VARIATIONTASK 7.1: UM AND TURN POSITIONFigure 7.1 illustrates 20 randomly generated concordance lines for the occurrence of um in the Spoken BNC2014.Figure 7.1 20 random concordance lines for um in the Spoken BNC2014Based on these concordance lines, what initial hypotheses might be generated in relation to the position of um in turns in contemporary spoken British English?Based on these concordance lines, if we apply the rule that, in order to be classified as turn-initial, um must be the first item in a turn, then 6 out of the 20 occurrences of um are turn initial. Therefore, a tentative hypothesis might be that in the Spoken BNC2014, um is less likely to occur in turn-initial position than in other positions in the turn.How do these initial, tentative results compare to those in Table 7.3?When we compare these results to those in Table 7.3, we see that in Figure 7.1, 6 out of 20 instances of um are turn-initial. This gives us a percentage figure of 30% of occurrences at turn-initial position. This is in contrast to the 13% of um in turn-initial position in the face-to-face component of the OANC. This finding is, perhaps, worthy of further investigation.TASK 7.2: WELL, TURN POSITION AND FUNCTIONFigure 7.2 illustrates 20 randomly generated concordance lines for the occurrence of well in the Spoken BNC2014.Figure 7.2 20 random concordance lines for well in the Spoken BNC2014Based on these concordance lines, what hypotheses might be generated in relation to both the turn position and function of well in contemporary spoken British English?Based on the concordance lines, see can see that if we extend the classification of turn-initial to include well in second position after other backchannels such as in oh well (line 10 and 16) and mm well (line 2), then 10 of the 20 occurrences of well are turn-initial. It is important to note that the lexical use of well in well done (line 12) has been excluded here. Therefore, based on these concordance lines alone, it is difficult to formulate a hypothesis in relation to either the position or function of well. How do these results compare to those in Table 7.4?However, when we compare the results to both Tables 7.3 and 7.4, we can see that 50% of the occurrences of well are in turn-initial position. This compares to 16% of occurrences in initial-position in the face-to-face component of the OANC. In addition, all of these turn-initial occurrences are ‘floor grabbing’. Only 14% of the occurrences of well have a floor grabbing function in the OANC. Return to Task 7.1, what are the functions of um? Can any connection be made between turn position and function based on the instances of well and um in these tasks?Based on the instances of um and well in Figures 7.1 and 7.2, it seems that um and well, when in turn-initial position, have a predominantly turn or floor grabbing function. It should be pointed out that this finding is based on a very small sample of concordance lines and that more analysis is necessary in order to further examine this connection between position and function. TASK 7.3: CLUSTERS IN ACADEMIC SPOKEN ENGLISHTable 7.7 illustrates the top 10 most frequent 2-, 3- and 4-word units in BASE. Using Tables 7.5 and 7.6 as reference points, discuss the similarities and/or differences between spoken academic English versus written academic English and/or Irish English casual conversation.Table 7.7 The 10 most frequent 2-, 3- and 4-word chunks in BASEBASE2-wordFreq.3-wordFreq.4-wordFreq.1of the9,958going to be975the end of the2182in the7,766one of the880at the end of1953going to4,184a lot of872is going to be1914you know3,722in terms of733if you want to1705and the3,659I’m going to649to be able to1676to be3,612we’re going to605if you look at1527sort of3,601this is the509in terms of the1368if you3,595if you like494at the same time1349to the3,471you can see494the way in which12910this is3,053and this is492going to talk about123Let’s apply our definition of a PM to the results shown in Table 7.7. PMs are defined by their optionality, their textual and interpersonal functions and the fact that they have little or no semantic meaning. Therefore, the 2-word you know (highlighted) is the only item to meet the criteria as a PM. All the other items in Table 7.7 are syntactic fragments frequently used in the construction of a phrase, clause or sentence. This result echoes both written academic English (our analysis of which similarly yielded only one PM on the other hand). However, you know is the most frequent 2-word unit in the Spoken BNC1994 and LCIE corpora. In addition to this, spoken British academic English contains a number of units that are interactive in nature signalled by the presence of personal pronouns, for example, I in I’m going to, you in if you like and we in we’re going to. Finally, the BASE features two items sort of and a lot of which can be classified as highly interactive vague language items. It is to vague language that we next turn our attention to in Task 7.4. TASK 7.4: IDENTIFYING VAGUE LANGUAGE ITEMSExtract (7.7) is part of the transcript from a cabinet meeting held in the White House on February 12th, 2019 (source: ). Identify the vague language items used in the extract.(7.7)[Context: <$1> = US President Trump.]<$1> They've already announced, in some cases - and in many cases, they have announced - they're moving back into the country. They want to be a part of the United States. It's like a miracle in the United States, what's happening. But we have a lot of companies that have left. In many cases, they left our country and they're moving back. And that means a lot of jobs. Speaking of jobs, we have to have more people coming into our country because our real number is about 3.6, 3.7. It took a little blip up during the shutdown and went up to 4. And 4 - any country would take a 4. But we're about 3.7; probably going lower. We need people. So we want to have people come into our country, but we want to have them come in through a merit system, and we want to have them come in legally. And that's going to be happening. We're doing very well in that regard. But we have tremendous numbers of companies. And you've been reporting on them. A lot of car companies are coming back to the United States. We want to keep the job boom going strong, and we must protect our economy.(Cabinet Meeting, White House, 12/02/2019)There are a lot of vague language items in (7.7), the majority of which are concerned with approximation and quantification. Among these are: in some cases; in many cases; a lot of; more people; about; a little; tremendous numbers.TASK 7.5: ADJUNCTIVES IN COCATable 7.9 illustrates the frequency results for the search item * that type of thing in the COCA corpus overall and in the individual spoken, fiction, magazine, newspaper and academic sections. As can be seen, * that type of thing occurs predominantly in the spoken component of COCA, with only one occurrence of the adjunctive in the academic component. This appears to contradict the findings of Table 7.8. Table 7.9 Frequency counts for the item* that type of thing in COCAHow might this apparent contradiction be explained? Pay particular attention to corpus design when considering the answer to this question.In Table 7.8, we can see that (and) (all) (that) type of thing has a frequency of occurrence in the BASE corpus of 7 times per million words. This is in contrast to a single occurrence of * that type of thing in the academic component of COCA. In terms of corpus design, it should be noted that the academic component of the COCA corpus is comprised of academic journals and is, therefore, a written academic corpus. Vagueness, in the form of adjunctives, may be acceptable in spoken academic language in order to build solidarity and the feeling of in-group membership. However, it appears that this particular type of vague language is unnecessary in the written medium. Instead of indicating any sort of vagueness, academic prose is largely more concerned with certainty of information.TASK 7.6: DISJUNCTIVES IN COCATable 7.11 illustrates the frequency results for or whatever in the overall COCA corpus and its component parts. Although the frequency counts appear to support the hypothesis that disjunctives are more frequent in spoken discourse than academic discourse, there is a much greater frequency difference between or whatever in the spoken and academic components of COCA than evident in Table 7.10. How might this frequency discrepancy be explained? Table 7.11 Frequency counts for the item or whatever in COCAAs Table 7.11 demonstrates, or whatever, a disjunctive, is more frequent in COCA than * that type of thing, an adjunctive. This supports our findings that disjunctives are notably more frequent than adjunctives - we demonstrated this through a comparison of Irish English intimate discourse and British English spoken academic discourse. In terms of the answer to the task, again, what we are dealing with here is the difference between spoken academic language and written academic prose. A selection of fifteen concordance lines for the occurrence of or whatever in the academic component of COCA can be seen here:A quick scan of the lines reveals that on occasion, or whatever is used in conjunction with language that is informal in nature: for example, I'm kind of just alone in my room or whatever (line 25); just kind of hit them or whatever (line 28) and like, I think I really stick out in crowds or whatever (line 40). One explanation for this might be that these are participant responses taken from research instruments such as questionnaires or interviews and reduplicated verbatim in the journal article. However, more analysis is needed in order to ascertain the reasons for the increased frequency of use of or whatever compared to adjunctive * that type of thing in COCA. TASK 7.7: SPEECH ACTS AND VARIATIONBased on your observations and language intuition, what are the three most commonly used expressions of gratitude in English. Go to the Google N-Gram Viewer () and search for these items by typing each of the three, separated by a comma, into the search box. This task can, of course, be done using other languages – this tool allows the user to search a variety of language corpora such as Chinese, French, German or Hebrew.How might the results generated from Google N-Gram Viewer based on your choice of expressions of gratitude be interpreted?Do they confirm or refute your intuitions about the most commonly used expressions of gratitude?This task is largely dependent on the expressions of gratitude that you have observed or intuited as commonly used in English or, indeed, if English is not your L1, then in your own language. What is important for this task is that you enter these items in the Google N-Gram Viewer and compare the results to your observation and intuitive ones. TASK 7.8: SPEECH ACTS AND VARIATION OVER TIMETable 7.12 presents the frequency results, normalised per million words, for thank, thanks and cheers in the Spoken BNC1994 and the Spoken BNC2014. Table 7.12 Frequencies of thank, thanks and cheers in the Spoken BNC1994 versus the Spoken BNC2014 (normalised per million words)Spoken BNC1994Spoken BNC2014ItemFrequencyItemFrequencythank512.55thank348.26thanks140.95thanks110.92cheers13.52cheers35.89 How have the frequency results for the use of the items changed over the twenty year time period between the Spoken BNC1994 and the Spoken BNC2014?Working from Table 7.12, we can see that the frequency of use of thank and thanks have decreased in the twenty year time gap between the collection of the two corpora. This decrease in the use of thank and thanks is in contrast to the increased use of cheers. Why might this be the case?There are any number of reasons as to why this might be the case and more detailed analysis may reveal more insights than can be offered here. One of the reasons is probably, yet again, linked to the issue of corpus design. The Spoken BNC1994, in addition to face-to-face casual conversation, contains more formal speech situations such as classroom discourse, business transactions and legal cross-examinations (see also Task 6.3). The Spoken BNC2014 is entirely made up of casual, informal face-to-face conversations, predominantly between people who know one another well. It might be that this contrast between formality and informality and conversational participants who know one another well and those who do not, accounts for the frequency differences. It could also be that we are witnessing the process of language change where the polysemic cheers is increasing in its frequency of use. TASK 7.9: CHEERS AND SOCIOLINGUISTIC VARIATIONIn the Spoken BNC2014, cheers mate occurs 20 times. Access the concordance lines using the CQPweb interface and, using the corpus metadata, determine the predominantly gender, age and socioeconomic bracket of the users of this phrase.Rather than just give the answer, it might be good to instead explore the process that will lead to the answer. First of all, go to the CQPweb interface for the Spoken BNC2014 corpus. In the text box under ‘Standard Query’, type ‘cheers mate’. There is no need to change any of the settings in the drop down menus under the text box. Press the ‘Start Query’ button. As we mentioned, there should be 20 concordance lines generated for cheers mate. Here are the first ten of these: In order to access the corpus metadata and determine the gender, age, etc. of the speaker, click on the speaker tag that appears to the left of cheers mate. For example, in line 1 here, to the left of cheers mate is the tag ‘S0572’. When you click on this tag, a range of information appears, which looks like this:Using the metadata for speaker S0572, we can see that he is male, aged 14 and, amongst other things, was born in Co. Durham in the UK. We also discover that he is a student and has been assigned the social grade ‘E’ by the corpus developers. According to this system, a categorisation of ‘E’ represents state pensioners, casual and lowest grade workers and unemployed people with state benefits. As the speaker is a student, he has received this classification. In order to fully answer the question relating to the speakers that most commonly use cheers mate, simply click on the remaining speaker tags and build a profile.CHAPTER 8 PRAGMATICS AND VARIATION AT THE LEVEL OF REGISTERTASK 8.1: IDENTIFY THE REGISTER CHARACTERISTICSUsing Table 8.1 as a template, identify the register characteristics of a telephone helpline.Register characteristicTelephone helplineMode Spoken Interactiveness and online planningThere is a degree of advance planning in that the caller has a reason for contacting the helpline. The advisor has probably received some training on how the call should proceed. There may be room for some spontaneity also. Shared immediate situation noneMain communicative purposeinformation/explanation Audience individualDialect domainregional/globalTASK 8.2 CONFLICT SEQUENCES AND TURN STRUCTUREExamine the following conflict sequence, characterised by at least three consecutive turns where speakers mutually challenge one another, between two intimates (intimates are either family members or close friends). Conflict sequences are also characterised by dispreferred responses. (8.4)<$1> yeah but it's still (.) you probably don't need to clean it that much do you?<$2> no but you don't need to clean it <$1> you do need to clean dishwashers<$2> you don't<$1> yes you do<$2> you don't <$1> that's why they have the dishwasher tablets and stuff as well for cleaning<$2> you just put it in and it's done<$1> I know it cleans the dishes but you do need to clean that as well<$2> no<$1> you do<$2> you're lying (.) you're a liar you're a liar oh<$1> rude you're a liar<$2> >> well you're a liar what? straight up do I care?(Spoken BNC2014: File S5MM)What is the structure of the turns in extract (8.4)?Assigning a structure such as the adjacency pair to (8.4) is not as straightforward as it might appear. We know that, for example, a typical adjacency pair structure is the question-answer. At the beginning of (8.4) we can see what might be labelled a question-answer adjacency pair:Question <$1> yeah but it's still (.) you probably don't need to clean it that much do you?Answer <$2> no but you don't need to clean it (dispreferred response) However, the pre-start yeah but connects this turn to the previous one where yeah but signals disagreement with what has been said in the previous turn. Therefore, the turn that we have labelled question here appears to function simultaneously as the second pair part of a previous turn and the first pair part of the next adjacency pair. We also see adjacency pairs such as assessment-disagreement (the preferred response is agreement in this pair):Assessment <$1> you do need to clean dishwashersDisagreement<$2> you don't (dispreferred response)This adjacency pair then triggers a side sequence of disagreement between the speakers. Another adjacency pair evident in (8.4) is insult-insult return:Insult <$2> you're lying (.) you're a liar you're a liar ohInsult-return <$1> rude you're a liarIn extract (8.4), this adjacency pair appears to have a third part which is a response to the insult-return:Insult-return (2)<$2> >> well you're a liar what? straight up do I care?Issues such as these have lead to some criticisms of the usefulness of the concept of the adjacency pair as a description of conversational organisation. How does this structure differ to, say, the prototypical turn structure in extract (8.1) or the example of a dispreferred response in (8.3)?The turns in (8.4) differ from the prototypical turn structure in (8.1) in that there are very few pre-starts, aside from yes and no. The extract also demonstrates that a dispreferred response does not necessarily entail a longer speaker turn. Both the preferred and dispreferred response can be achieved using a short speaker turn.TASK 8.3: ANALYSING THE PRAGMATICS OF TELEPHONE CALL OPENINGSRead extract (8.8), an extension of extract (6.3), the opening of a call to the NHS Direct helpline. (8.8) 1. phone ringing2. NHS advisor: Good evening. This is N H S Direct. My name’s [name]. I’m a health advisor here. Could I just start by taking your telephone number please?3. Caller: [personal details removed]4. NHS advisor: Okay. How can I help then [name removed]?5. Caller: Right. I went to see the doctor yesterday.6. NHS advisor: Mm.7. Caller: I’ve got erm what they’re calling a middle ear infection.8. NHS advisor: Yeah.9. Caller: Erm it’s a new doctor cos I’ve erm recently moved into the area.10. NHS advisor: Mhm.11. Caller: Erm when I’ve been before [pause 0.5 sec] I’ve been prescribed something different. This time she’s given me antibiotics and basically the problem is that I’ve collected them. I’m a bit worried about taking them cos I’m not actually sure if I might be pregnant and erm obviously I I know about [inhales] you know not taking medication when you’re pregnant. So I just wanted to really check what I could do and really whether I should be taking antibiotics anyway with [inhales] you know all these things about you're not supposed to take them unless you really need them.12. NHS advisor: Yeah. Okay. Erm [pause 1 sec] okay. Best off that you talk to a nurse about this.13. Caller: Right.14. NHS advisor: Er so you’ve got a middle ear infection.15. Caller: Yeah.16. NHS advisor: What you been prescribed then?17. Caller: Augmen= Aug=18. NHS advisor: Augmentin?19. Caller: Something. Yeah.20. NHS advisor: Yeah.21. Caller: Yeah. Something like that. [laughs]22. NHS advisor: Yeah. It’s Augmentin.23. Caller: That sounds about right yeah.(NHS Direct Corpus) Consider the CA interactions that you have just examined and label the structural organisation of the call opening.The structure of the call opening might be labelled in the following way:SequenceCall openingSummons1. phone ringingAnswer + Greeting + Identification sequence 2. NHS advisor: Good evening. This is N H S Direct. My name’s [name]. I’m a health advisor here. Could I just start by taking your telephone number please?3. Caller: [personal details removed]Business of call4. NHS advisor: Okay. How can I help then [name removed]?5. Caller: Right. I went to see the doctor yesterday.6. NHS advisor: Mm.Consider the call from a pragmatic perspective in comparison with extracts (8.5) and (8.7). How might any similarities and/or differences be explained from this perspective?The nature of the interaction is institutional and the participants are strangers, however, the ‘summons-answer’ sequence is longer than in, for example, an emergency phone line;The use of the greeting good evening and the NHS advisor’s name are positive politeness strategies and in part echo an unmarked relationship sequence, where the participants are not very friendly, not neither are they strangers. Therefore, there is an immediate reduction in the formality and institutionality of the context, perhaps in an effort to help the caller feel comfortable speaking about their health issues;The NHS adviser also attends to the caller’s negative face with the indirect request could I just start by taking your telephone number please?;Finally, as the business of the call gets under way, the NHS advisor uses the caller’s name, again lessening the face threat associated with any request for the sharing of personal details or information. TASK 8.4: EXPLORING THE INTERFACETable 8.3 compares the top 20 most frequent turn initial items in two corpora – English language teacher meetings (C-MELT) and casual conversation between family and close friends (LINT). Table 8.3 Comparison of top 20 most frequent turn initial items in C-MELT and LINT C-MELTLINT1YEAHYEAH2I/’M/’LL/’VE/’DI/’D/’M/’LL/’VE3ANDOH4BUTNO5NOAND6MMWHAT/’D/’LL/’S7OKAYIT/’D/’LL/’S8SOYOU/’D/’LL/’RE/’VE9OHTHAT/’D/’LL/’S10[NAME]HE/’D/’LL/’S11UM HMMAH12RIGHTBUT13THAT/’SSHE/’D/’LL/’S14WELLTHE15IT/’SWELL16THEY/’LL/’VE/’RETHEY/’D/’LL/’RE17WE/’LL/’RE/’VE/’REIS18THEDO19YOU/’LL/’VE/’REDID20BECAUSESO We can clearly see the similarities between the language of meetings and that of family and close friends – a comparison only made possible through the application of a corpus methodology:Both lists contain initiators such as yeah, no, so, but, well and and demonstrating their importance in many different registers;The pronouns I and you and the demonstrative that feature prominently on the two lists;C-MELT contains the pronoun we and the syntactically independent items mm, okay, um hmm and right;LINT is distinct from C-MELT with the syntactically dependent items he, she, is, do and did (all shaded) unique to this particular dataset. How might the similarities and differences between the two registers be explained from a pragmatic perspective?In terms of similarities, response tokens such as yeah and well and pronouns such as I and you highlight the interactive nature of both registers. The first difference of note is the presence of the pronouns he and she on the LINT list. He and she are very frequent in intimate data because family and close friends are naturally concerned with others known to them. In addition, there is a large amount of shared knowledge of other people between intimates and so these others are frequently the topic of intimate talk. The second difference is that we features on the C-MELT list but not on the LINT list. The pronoun we has been connected with the creation and maintenance of group identity. This is necessary in the workplace discourse exemplified in C-MELT but, due to an underlying recognition by family and close friends that a group already exists, the creation or maintenance of a group identity is not as necessary in this register. The third difference is the presence of the verbs is, do and did which, when they occur as turn-initiators, function to form questions. In the intimate data, is typically co-occurs with the pronouns he, she and it. Questions are often used by conversational participants to show interest in one another, explore common ground, elicit reciprocal information and drive the conversation forward. The interpersonal nature of questions coupled with the unique contextual factors present in intimate discourse allows conversational participants in this register to question one another more directly than in other registers. Direct questioning strategies, in fact, serve to index the intimacy of the relationship and, hence, the friendliness of the requests.TASK 8.5: EXPLORING KEYWORDS AND THEIR CONNECTION TO PRAGMATICSTable 8.6 also shows maybe as a keyword in C-MELT. Figure 8.3 illustrates a sample of concordance lines for maybe + we in C-MELT. Figure 8.3 Sample concordance lines for maybe + we in C-MELT (unsorted)What are the patterns associated with the use of maybe + we?The pattern associated with maybe + we occurs to the right hand side of the node in Figure 8.3 and is maybe + we + modal verb. Can, should, might, have to and could all appear to the right of maybe + we. How might these patterns be connected to, for example, politeness strategies (see Chapter 5) employed by the teachers?Modal verbs are associated with a wide range of functions. For example, in the making of requests or suggestions, both could and might are considered polite options. They are also associated with tentativeness which, in this instance, allows colleagues to avoid the face threat connected to these actions. Have to is used to refer to obligations, specifically those that come from outside the speaker. Therefore, this allows the speaker to distance themselves from any personal involvement in obliging a colleague to do something. This again makes the conversation less threatening and direct. TASK 8.6: MULTI-WORD UNITS AND THE NOVELS OF JANE AUSTENTable 8.8 lists the top 10 4-word units from a corpus of Jane Austen novels comprised of Pride and Prejudice, Emma, Persuasion, Sense and Sensibility, Northanger Abbey and Mansfield Park. Table 8.8 The 10 most frequent 4-word units in the Jane Austen corpus Jane Austen’s novels UnitFreq.1I do not know1332I am sure I853the rest of the824a great deal of805in the course of716at the same time677I am sure you578I do not think559and I am sure5410it would have been49 For this task, build a corpus of the novels of Jane Austen (those listed here) in order that the functions of some of the 4-word units in Table 8.8 might be further analysed using, for example, concordance lines. Building a rudimentary written corpus is a relatively straightforward exercise. Follow these basic steps to get started:Go to the Project Gutenberg homepage ();Use the ‘search for books’ option to locate the individual novels;Choose the ‘Plain Text UTF-8’ version of the novel and open the file;Copy and paste the content into Notepad, taking care to avoid ‘noise’ such as the website’s terms and conditions detailed at the beginning and end of each novel; Save the Notepad file onto your hard drive (a separate file can be created for each novel or all six novels can be copied into the same file);Download the freely-available corpus analysis software package AntConc ();There are a number of excellent tutorials on YouTube explaining how to use AntConc – choose one that deals with the concordance tool;Generate concordance lines for the item(s) in Table 8.8. In order to give this final, more analytic step (8) some focus, answer the following question:Based on the analysis of both the Sherlock Holmes and Shakespeare corpora, what pragmatic features emerge from the list and what might these tell us about the register represented by Jane Austen’s novels?One of the features that emerges from an analysis of these multi-word units in comparison to, say, the Sherlock Holmes corpus, is the interactive nature of these units. The personal pronoun I features in five units, and the second person pronoun you in one unit - I do not know, I am sure I, I do not think, and I am sure and I am sure you. These personal pronouns are not present in the top ten units in the Sherlock Holmes corpus. The presence of I do not know, also one of the most frequent units in the Shakespeare corpus, and frequent in the form of I don’t know in LCIE and the Spoken BNC1994, is often used as a softener or mitigator. Similarly, I do not think, or its 3-word unit equivalent, I don’t think, is a frequent cluster in LCIE and also functions as a mitigator. Multi-word units such as these demonstrate Austen’s awareness of frequent linguistic patterns in spoken language and, therefore, provide us with a testament to both her knowledge of the function of these items and the accuracy of her representations of direct speech in her novels. TASK 8.7: ANALYSING CORPUS MARKUPWhen we look at extralinguistic information in the NHS Direct helpline corpus, we find a lot of caller inhalations noted by transcribers. We therefore treat these as important to this register. When we look at the concordance lines for <$E> inhales <\$E>, we find the following as displayed in Figure 8.5. When examining the concordance lines, bear in mind that <$2> is the caller to the helpline. Figure 8.5: Examples of concordance lines of <$E> inhales <\$E> (unsorted)oea that you know of?<$2> <$E> inhales <\$E> Erm no. His stools were a lithich ones are you on?<$2> <$E> inhales <\$E> Oh. Er erm Amoxycillin? <$at about painkillers?<$2> <$E> inhales <\$E> Er well I had some sort of on any health problems?<$2> <$E> inhales <\$E> Erm I've been ill. <$=> Erm I colour is the rash?<$2> <$E> inhales <\$E> Erm <$E> 1 sec <\$E> well it''ve been scratching?<$2> <$E> inhales <\$E> Oh yeah. Oh yeah. <$3> <$= how is he with that?<$2> <$E> inhales <\$E> I mean he's not complained abre blisters are they?<$2> <$E> inhales <\$E> Well they look like them a biess water than usual?<$2> <$E> inhales <\$E> <$E> extends following syllab. How can I help you?<$2> <$E> inhales <\$E> Well I've come to babysit for What patterns emerge in relation to this extralinguistic information?As Figure 8.5 demonstrates, the extralinguistic information <$E> inhales <\$E> is associated with two distinct patterns. The first is to the left of the node where we can see that the physical act of inhalation by the caller <$2> is preceded by a question. The other pattern to emerge is that the inhalation is followed, in the majority of the examples presented here, by a response token, for example erm or I mean or well, or a series of response tokens, for example, Oh. Er, erm or Er well or Oh yeah. Oh yeah. This pattern of inhale + response token(s) on the part of a caller might indicate their willingness to answer the question as comprehensively as possible given that the caller does not, in the main, provide a closed answer even when they have the opportunity to do so in, for example, the case of a yes/no question. CHAPTER 9: PRAGMATICS AND LANGUAGE TEACHINGTASK 9.1: EXPLICIT VERSUS IMPLICIT FEEDBACKThis task requires you to imagine you are an English teacher of an A2 Elementary level (CEFR) who has presented and practised some examples of invitations in the class textbook with the students. Then, by way of follow up, you set up a paired role play task. The task involves one student inviting another student to their house for a birthday party. You are asked to consider the following interaction between two of the students: Alex and Sarah. Alex is given a prompt card and Sarah is his role play partner.Role card for AlexYour birthday is on Saturday! You are having a party in your house at 6pm. Invite a friend to your party. Extract from the roleplay interaction: Alex: I’m having a birthday party on Saturday at 6pm and please you must come!Sara: Yes, thank you. I would like to go. Where is your house? The task asks: if you were the teacher, would you see a need to address anything about the Alex’s use of language in his invitation? Would you: Explicitly: tell Alex that saying I am having a birthday party on Saturday at 6pm and please you must come! is too direct as an invitation and suggest alternatives, then practise these and re-do the role play task;Implicitly: after the roleplay, show all of the learners some more authentic examples of invitations and then ask them to undertake their role play again (and pay close attention to how invitations are performed). There is no right answer to this question and the goal of the task is to promote reflection. If you believe that overt explicit instruction is important, then you are likely to opt for 1 above. You see the importance of feedback to the student and the class as a learning opportunity as well as the need to practice the task more than once.If you believe that learning takes place subconsciously, you will see option 2 as more plausible because you believe that by showing the learners more examples of invitations and performing the role play again, you will offer a better opportunity for acquiring the pragmatic norms for invitations. Option 2 does leave open the possibility that many in the class will still think that “please you must come” is an appropriate way of making an invitation.TASK 9.2: IMPLICIT LEARNING OPPORTUNITIESCan you think of classroom interactions where a learner might pick up (or intuit) pragmatic norms of the target language? For example, requests: a classmate asking to borrow an eraser. Make a list of possible learning opportunities.There are many possible answers to this question. Here are just some examples:Requests to use the bathroom;Invitations to play a game with other classmates;Requests to go to the whiteboard and perform some action in front of the class;Directives to tidy up the classroom;Directives to pay attention;Directives to perform some action in relation to the physical space of the classroom (open window, close door, move a chair, hang up a coat, stand in a line, etc.).Requests to distribute books or collect books;Requests to put up hands if...Greetings and leavetaking routines when guests that come into the classroom;Routines and formulaic language around birthdays or holiday rituals.TASK 9.3: IMPERATIVES IN ACADEMIC WRITINGConsider Figure 9.3 which illustrates the results for the verb note in the MICUSP (per 10,000 words) across all of the disciplines represented in the corpus (with the disciplines in Neiderhiser et al.’s, 2016 study shaded): Choose any two of the disciplines and examine a sample of instances of note:What percentage of the uses of note in your sample are imperative?The result here will depend on the discipline you have chosen. It is best to go through at least 20 examples. You will need to go through the results, example by example, and decide whether note is being used as an imperative. To calculate the percentage, divide the total result for note as a directive, divide it by the total number of the examples you have searched and then multiply this number by 100. How many are at the beginning of a sentence? What is the impact of note at the beginning of a sentence?This result will depend on which discipline you have chosen to explore. You will probably notice a connection between note being at the beginning of a sentence and its functioning as a directive. Note, as a directive can also be placed in the middle of a sentence. Compare the following examples. Note at the beginning of the sentence usually carries more force as a directive.ECO.G1.03.1What Drives the Relationship between Education and Fertility?EconomicsResearch PaperNote that educational level will be treated as a categorical variable here for the convenience of interpretation.LIN.G1.06.2Perceptions of Speaker in Computer-Mediated CommunicationLinguisticsProposalHe indeed found results that indicated that eye dialect effectively rendered speakers as substantively socially different; however, note that although eye dialect is a written modality, What other non-imperative uses of note do you find?You may notice that note is used to mark an important point in phrases like:We note that…It is interesting / important to note thatI note thatHe notes thatApart from being used as a verb, the noun form of note can also be found, as the following examples illustrate. What is interesting about these uses is that they are both figurative and based on two separate noun meanings, note meaning a small piece of paper in the first example and a musical note:As a side note,[referring to a play] ...inject the play with a rather optimistic note.TASK 9.4: CONSCIOUS AND SUBCONSCIOUS LEARNING OPPORTUNITIESIn this task, you are asked to look again at the list of imperative verbs from Neiderhiser et al. (2016) in Table 9.1 and then undertake the following three tasks. The verbs in Table 9.1 are:notenoticetakeletrecallinstallseeassumeimagineconsiderreferdefinesupposecallcompareTake any one of the verbs and explore its use more closely in the MICUSP. Note your observations and prepare a mini-presentation on it.There are many possible observations that could be made about the verbs used as imperatives. For the purposes of illustration, we take one example from the list above, the verb imagine. First we enter it as a search and we see how its use is distributed across the disciplines in the corpus, and across types of writing. To compare like-with-like, make sure that the “per 10,000 words” button is clicked at the top of the screen. From this our first observation is that, imagine is much more frequent in philosophy. This first calculation is based on all uses of imagine and some of these may not be imperatives. We therefore need to look at the examples and pick out those that are used as imperatives. On closer examination, it is clear that most of the uses of imagine are in declarative clauses rather than imperatives.Here are some examples of imagine being used as an imperative verb. From these examples and further observation, it seems that calls for the reader to hypothesise is most common in Philosophy.BiologyNow imagine such a metropolis spreading to cover every last crevice on Earth.Imagine and analyze the results:NursingImagine a start-up clinic run by a physician and nurse practitioner. PhilosophyImagine the following situation. Imagine the following individual, whom I dub Francis. … imagine a case where the hardship faced by the gambler produces a vastly worse state of affairs than does the employee's loss. Consider how this kind of information could be used to develop classroom materials and how it might aid conscious or subconscious learning.Doing hands on observation makes the user of the corpus data notice patterns of use both consciously and subconsciously. If a learner were to undertake this activity, looking at a number of different verbs, they would first of all notice the different patterns of its use. For example, a learner might consciously notice some patterns such as Imagine + determiner while they may subconsciously notice the pattern It is + adjective + to imagine that while the are sifting through examples so as to find imperative uses. If they have not used these imperative verbs in their writing before, it may give them confidence to try them out. If they are unsure about how to use them, they can go back and look up the corpus to reassure themselves that this is something that is used in successful writing. An example of a possible classroom activity:With a partner, use the MICUSP to search for the following three verbs across all of the corpus: imagine, assume and consider. Once you have answered the following questions, prepare a mini-presentation for your classmate.Identify examples where each verb is used as an imperative. What is the function of these imperative verbs in the examples that you have selected? Do certain disciplines appear to use more imperatives than others?Are some of these verbs more likely to be used in certain disciplines than in others?Look at the percentages of “Distribution across paper types” for each verb (in the pie chart). Does this suggest that certain types of papers are more likely to use imperative verbs?Using the Academic genre component of COCA, design a task for advanced learners of English based on some of these verbs.A possible task using COCA:The verbs consider and notice are commonly used as imperatives in academic texts, usually at the beginning of a clause to draw the readers’ attention to an important point. Explore these two words in the Academic section of the COCA corpus using the Key Word in Context (KWIC) function. Here are the settings to use for consider. When searching for notice, use the same settings apart from changing the search word. Look at the KWIC results for each word.Make a list of some of the main patterns in which the two words are used and decide whether they can be used interchangeably, i.e. as synonyms;If you were advising a classmate on how to use each of these words as imperatives, what would you say?Compare your answers with someone else in the class.TASK 9.5: THE USE OF I THINK IN WRITTEN ACADEMIC ENGLISH Using the MICUSP, conduct a search for I think across disciplines and note any frequency differences (across per 10,000 word results).Here is what the result looks like at the time of writing:Looking at examples, find instances of I think in clause initial, medial and final position and discuss their function.Here are some examples:Initial clause positionBIO.G1.08.2Drosophila neuroblastsBiologyResponse PaperGenerally speaking, I think the paper is fairly convincing in demonstrating the effect of Hb and Kr.[Note that this example, the discourse marker, generally speaking precedes the instance of I think so it might appear to be an example of a media clause use however, Generally speaking is not part of the main clause. Clause initial items are those that come at the beginning of a clause even if the clause is preceded by a discourse marker.]ECO.G0.08.1Transportation Costs and the Non-Agricultural Sector in 19th Century BrazilEconomicsResearch Paper... I think that the evidence presented by Leff to support his estimate of the size of the domestic agriculture sector is very weak, and his argument could have been vastly improved if he had done additional research to produce a more accurate estimate of the amount of the population invested in activities other than domestic agriculture from 1822-1913. In addition to the actual size of the domestic agriculture sectorMedial clause positionEDU.G0.03.1School Study: Abbott Middle SchoolEducationReport... Their body language is telling of how they feel about their friends and also of some of their insecurities. For example, some of the girls physically latch on to each other, which I think is a visual example of their need for acceptance in their peer group.ENG.G1.05.1Yeats and SpenserEnglishArgumentative Essay1 of 2 hits Show all1. Allegory and, to a much greater degree, symbolism are a natural language by which the soul when entranced, or even in ordinary sleep, communes with God and with angels. They can speak of things which cannot be spoken of in any other language, but one will always, I think, feel some sense of unreality when they are used to describe things which can be described as well in ordinary words.Final clause positionAfter much searching, the following instance of I think in the pattern, I think so was the only potential example but on examining the fuller context (by clicking “show all”), it was clear that this was referring to a language usage in an example that followed. This tells us that in academic writing, clause final instances are rare.N.G0.10.2The Distribution of Anaphoric 'So' in MICASELinguisticsResearch Paper1 of 15 hits Show all1. Pro-clause complement with finite antecedent: I think so.Make observations about the multi-word unit patterns of I think across disciplines (Tip: note what words typically come before or after it; do you notice any consistent patterns? Are these specific to particular disciplines?).The disciplines of Philosophy, Linguistics and Education show the highest frequencies of I think.In Philosophy and Education papers, I think seems to be used almost always in the pattern: I think + that clause (with or without that) whereas in Education, it also appears in medial clause position and this use seems to downtone the force of the opinion being expressed. It seems to personalise it more. This is just a tendency based on a small sample. It might be interesting to follow up on this in another dataset and across other disciplines.PhilosophyPHI.G1.02.2The Problem of Necessary / a posteriori and Contingent / a priori TruthsPhilosophyResearch Paper1 of 5 hits 1. (i) is an assumption that a good number of philosophers share nowadays. I will not comment on that in this paper, although I think it might be a problematic tenet. LinguisticsLIN.G1.01.1Recognizing Textual Entailment With a Modified BLEU AlgorithmLinguisticsResearch Paper… However, I think that there is a lot of room for innovation in this approach and it may offer significant progress in the future. See (Kouylekov & Magnini, 2005), (Raina et al. 2005), and others.EducationEDU.G0.01.1Learning In EconomicsEducationReportThe theory of development that I think is most relevant to discuss is sociocultural theory, specifically the framework proposed by Vygotsky.EDU.G1.11.1Rethinking Academic Writing About Sexualized ViolenceEducationProposal... While I may be ideologically and ethically resistant to having these biases challenged, I am even more concerned with allowing women’s voices to speak through and out of the data. Therefore, even if the data I collect does not confirm my political or ideological standpoints I think I will be able to assimilate it into my understanding of the research.Search for I think in another academic corpus (e.g. COCA). Note any observations in relation to clause position and multi-word unit patterning and prepare a mini-presentation.Here are some observations about I think based on searches of the Spoken BNC2014:Clause positionI think appears regularly in all three clause positions. Unlike what we found when looking at part 1 of this task, there are many examples of I think in final position. Notice how initial position I think asserts an idea. Medial position I think downtones an opinion or a point while a clause final position I think downtones the assertion to the point of casting doubt on what the speaker has just said. This probably explains why writers in the MICUSP rarely use I think in clause final position. Compare the following examples from the BNC2014:Clause initial positionI think that was pretty much the first thing I did when I got here I think I've been put off both of them just by my gymI think it's nice try that I think that's the right measurement Clause medial positionthese ones I think will be more your cup of tea yes yes yes it's because I think they are your old gloves these ones this is probably a bad bad idea I think opening it right now is probably a bad idea but I'm going for it got it now Clause final positiontrue that's probably the right way I think no I'm gonna have a shower I think my one's about the best I think cos that's really how I learnt how to play drums I think I only surpass her by one inch I think it was in Luton I think TASK 9.6: THE USE OF I THINK IN SPOKEN ACADEMIC ENGLISH For this task you need to use MICASE and search for I think across two different speech events types (e.g. lecture, dissertation defence or colloquium). A tip: to see the options for speech events, use your arrow to scroll down through the options as there is no a drop-down menu.Here we compare Advisory sessions and Dissertation Defence, by way of one example.How is I think used differently across user roles (i.e. lecturers and students)?Observations on I think patterns in Advisory sessions As one would expect, the advice giver typically uses patterns such as the following, with I think in initial position:i mean that could be a good balance i mean i think you should think about, your strengths and interests i think it's a matter of_ it's not even a matter of energy like i think i just need to manage my time better i think your sophomore year is probably a better time to kinda get started on thati think it really is something that you should, think about.i think you should if you're gonna stick with the class you should be hounding your G-S-I a fair amount The advisor also tries to soften the force of the advice and promote reflection using patterns of I think ifyou know i think if you're failing if you have a thirty-three percent overall and you need to get you know a hundred on the next, [mhm] all the next homeworks they're not even collecting them but the next exam in order to get a C-minus then you're being kind of unrealistic ...i mean i think if you're afraid of math or you dislike it or you're you know just, not interested not good at it you're not gonna get better but if it was something where you know the first six weeks of the term you're just kinda like hm hm hm hm hm this is fun and then you just kinda realize that you hadn't been doing the work ...The student in the advisory session also uses I think to reflect. Notice how the speaker’s tentativity is reflected in the clause positioning of I think:i could finish my concentration like by end of next year i think? by_ if i did summer term?i was thi- i i think that if i... i i really enjoyed teaching elementary language [mhm] because, it was something that was so, um, obvious when, eh- students had grasped something or made a breakthrough, you know what i meani wouldn't wanna be in a career or studying a major that, is not that interesting to me i'm just doing it because it's_ i could be successful, or whatever i think i'd rather, stick with something that's more, um i could be more involved inObservations on I think patterns in Dissertation DefenceThe panel members on dissertation defences use I think both to assert strong opinions and also, sometimes, to soften the force of what they are saying. Notice the role of clause position and other hedging items.Asserting opinion with I think in clause initial position and no hedging: i think that needs to be in the abstract Paul Bley's influence uh on on Jarrett has been underestimated, uh i think you might want to probe a little deeper into this, [S2: mhm ] because Bley is also a kind of chameleon-like figure. Asserting opinion with I think in clause initial position but mitigating its force through the use of hedging devices (underlined in the examples): on the technical side mhm i think it would be a good idea to, mhm to, let your readers know, since, there's a somewhat kaleidoscopic, uh, impression created by reading the dissertation …i think it would be worth your while to put a a fair bit of time into this. [The more direct way of saying this might be: I think you need to put a lot of time into this]There are also some but not many instances where the panel members use I think in the clause medial and final position giving opinions in a slightly more downtoned way:[At the opening stage of the defence] S3: i mean cuz there're various points of which, i think i might wanna want clarification and or comment on certain thingsi i mean i i'm not, i think Don was asking something, more broad than just that measure but but it, i mean that's a measure but, [right.] what you have is only an inkling of what those curves look likeyou've got to sort of embrace that i think Students also use I think both to assert and defend ideas. For example:[S5 is a defence panel member; S2 is the doctoral candidate]S5: okay, here's another thing about keys. page seventy-two... uh, this piece... uh, seems to have practically no B naturals in it... [S2: mhm ] right? [S2: right ] now why is_ why didn't you just notate this with one flat then? is this is this, do you_ is this somehow a, a, a Lydian F or a Dorian D or something it doesn't sound anything like it to me, [S2: mhm ] but here again [S2: mhm ] i mean, you're you're hearing this with different, and better informed ears than i am, but i i just don't see why this isn't in on E-flat? and why these all these accidental B-flats [S2: yeah ] have disappeared?S2: i i i think what's what's happening is that um, uh, you know that i- it's, it's the same uh, principle that we talked about with the the, A-flat i mean i accept your argument that the A-flat wasn't uh, uh the best solution to to uh, take as a as a start- harmonic starting point [S5: mhm ] but, uh the A-flat a- A-flat with the sharp four, would eliminate the D-flat [S4: mhm ] and it's, it's a similar situation here, with a B uh flat. i, i think of D minor, now so predominantly as not the Aeolian but the, uh the Dorian mode [S5: mhm ] that's and and and when i, uh when i compose, uh th- this is how it started, i i would save myself the hassle of putting, the B-flat in. and maybe that's a little too shorthandQuite often, in examples from the students’ discourse, we find instances of clause medial uses of I think:… this graph is again mean annual temperature against percent entire, percent entire margin species. and what i have is that CLAMP database which are these open circles, and then many of the other databases that, um create_ that were used to create a- the equations that i tested. and i'll point out, um these, green diamonds which are the Australian, data set, shown there, and i'll just, show you one thing which is, why i think, all of these, especially low-temperature sites are overestimated, is that in general you can see that at any given temperature, these South American sites tend to have a higher percentage of entire margin species, than other sites, from, different databases. and because of that these, sites tend to be overestimated when you run them through equations based on this data here. and i'll just point out as well that ...S2: how many species, [S9: yeah ] you would find in general? [S9: yeah ] um... i can't give it a really great estimate of that i will say that, lowland areas in South America can have, upwards of three hundred and fifty tree species, and when you get up into the montane regions that the diversity decreases, [S9: mhm ] a lot. and so some of the areas that i looked, at when i was compiling my data list you would find, maybe a hundred and fifty, or so, species in the submontane regions, which is sort of in between, but it depended also on the area you were in so if they had, low soil nutrients you wouldn't expect as many species even if you're at, the submontane region or, cloud forest which can occur at many different [S9: mhm ] elevations, those tha- you know the, that particular precipitation environment also might, um change but i think in general you'd find maybe a hundred a hundred and fifty species on the high end in submontane forest.In summary, what we find is that I think is used in different ways in different contexts of use. Variables such as the role of the speakers, the power asymmetry and speech event all have an influence on its use. We see that clause position is a very good indicator of how I think is operating pragmatically.What are the multi-word unit patterns of I think (i.e. examine the most common word(s) to the right or left of I think). Do these vary across speech events (or discipline)?Here are some patterns from different speech event types in MICASE, across the disciplines of Anthropology and Chemical Engineering.AnthropologyI think + that clause [with and without that] is the predominant pattern to the right of I think:i think morals are really central to this but,i think that, in bands tribes and chiefdoms, they had to do that, i think that they, are ex- they are explicitly saying that they want to show children, that there is an African-American community.To the left of I think we find conjunctions so and but preceding I think as well as pragmatic marker I mean. These are all good examples of speakers linking with ongoing discussions in coherent ways by marking their discourse for their listeners:so i think it's also, social, social norms and that kind of thingso i think that susceptibility, is every bit as malleable, uh and as changeable as as risk is, so i think in one community in New Jersey that they're having, Gujarati as one of thebut i think you're right that they're um, and and the the point you made about the Focus Hope in Detroit umbut, i think even like the way they view politics is just different from, where they came from like, the family, ... almost the exception but i think that's partially just their background yeah that makes sense i mean yi mean i think i think you're right in thinking about, the way that they're different. i'm not sure it's gonna to be useful to you to think about whether i mean i think with the whole multiculturalism, you know it's, that they don't really necessarily, feel that, they need toi mean i think we could talk about like the six year old boy with like the gun obviouslyChemical EngineeringI think + that clause [with and without that] is the predominant pattern to the right of I think:i think they i mean they have like tanks and reactors but.i think we went through and and derived, the concentrations yeah i think that's when you seal it i think that's when you put the carbonation in i think that you added wrong you have a what? To the left of I think we also find the use of conjunction but to introduce contrary ideas where I think downtones its force:but i think that just meant that it wouldn't lose, the carbonation,but i think if you add, twenty liters of yeast to what do you have four hundred liters of, beer?We also find this example of I mean functioning in a similar manner:i mean i i think it's the same thingAnother common pattern to the left of I think is a speaker’s response token (underlined in examples below). That is, they respond to what another speaker has said and then continue the turn:yeah i think that's when you seal it i think that's when you put the carbonation yes, yes i think i saw this yesterday yeah. okay, but then the finalabsolutely right. um, i think i (mighta boxed that) now, check that calculation though because i'm not sure i mean Apart from the obvious disciplinary differences that might prevail between Anthropology and Chemical Engineering, the main different that we can observe in the patterns to the right and left of I think is that in the Anthropology data, especially through exploring more extended examples, is that speakers are expressing ideas less collaboratively. In the Chemical Engineering data, the examples appear to be more interactive.What do you notice about the clause position of I think?See commentary on part 1 where this is exemplified in detail. In summary, examples show that the clause initial position can be used to make strong assertions or these can be downtoned by adding mitigating devices (also called hedges). Both those in roles with more and less institutional power use clause initial I think and depending on the situation, they may or may not soften it with a mitigating device.Both clause medial and clause final examples are also found; these can show degrees of uncertainty.Prepare a mini-presentation based on your main observations.Your presentation will depend on the contexts that you have explored in MICASE.TASK 9.7: SUCCESSFULLY PERFORMING SPEECH ACTSConsider the examples of elicited responses to a request task based on booking a study room in a university context from the SPACE corpus in Figure 9.4.A) Typical non-native speaker (B2 Upper-Intermediate level) response: I want to find out how to book a study room please. Can you help me? B)Typical native speaker response: Hi. Could you help me book a study room please? How do the two requests differ?Response A appears somewhat more direct because it begins with I want. Response B seems more polite because it begins with an informal greeting Hi and the request is expressed in a downtoned manner using Could you help me… please? However, Response A is ultimately just as polite because it also uses please and it ends with a more polite request form Can you help me?Are they both successful?Yes. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download