 Okay, we might get started. The last few weeks of the departmental seminars, where we were showcasing what's going on here in this department. So we've had, we had the talk, like, Indeed and Henrietta and Marie and Carolina. And today, we have another talk from our SOS person. This is Sam, Samantha the Child. First met Sam five years ago when she... What? Six years ago. Oh, seven. It's 2017, I started in 2010. So she did her BA studies at Nottingham and then came here to do the MA's in documentation and description. During the MA she did a really interesting study of Mauritian Creole use here in London. So looking at how families from Mauritius were dealing with the three, at least three, different languages that were available to them, English, French and Mauritian Creole. I guess they're interested in multilingualism, which really played out in that thesis. And so she's since gone on to do further work on the multilingualism and particularly where you've got situations where people are choosing to use different languages and mixing and trans-language as it's called. So her research on the PhD has been based in Senegal, in the southern part of Senegal, and she's been looking there at the village level, particularly the role of gender in issues of language choice and language use. And today we're going to hear about these practices in southern Senegal and from languages to trans-languaging in a very post-modern field. Thank you. Well, thanks for having me and thanks for the introduction. You did steal a little bit out of my intro slide, but that's fine. So, yeah, to be said today I'm going to be talking about multilingual practices in the Casamans in Senegal, showing you a bit of data from my PhD research. And I also talk a bit about how I collect data, the various methods that I use. I'll show you a couple of, well, one clip actually, and some transcripts and we'll look at how I go about analysing the different multilingual practices before coming on to wider discussion about languages and trans-languaging and what the data that I've collected as part of my PhD and also our research team can contribute to the discussion. So I'm going to give a bit more background about the Casamans in Senegal, for those of you who don't know, sorry for those who've heard it numerous times from numerous team members. So it's a very picturesque setting in West Africa in Senegal and where we work it's very close to different national borders. So the Casamans is the southern part of Senegal and above that you have the Gambia and beneath Guinea-Bissau. And these were all different colonies. So Senegal was a French colony, Guinea-Bissau was Portuguese and the Gambia was a British colony. And all of these national languages also play into the multilingual situation that we have. So as is pretty typical for West Africa, Casamans is highly multilingual and we have official languages, national languages and then on a more regional level different languages that function as lingua francas and then numerous local languages. Some of these are associated with particular villages and some have a bit of a wider spread. And then on the individual level of the person, people generally have quite diverse linguistic repertoire and this aligns with mobility so it depends where they've been throughout their lives, different villages, towns, countries where they pick up different languages. So more specifically you can see on the map here so this is the area of the Casamans and it's super in the border with Guinea-Bissau it's only about 20 kilometres and this here in the middle of Senegal is the Gambia but it's only a couple of hours by road so people frequently cross different borders to see friends and family members and the like. I work as part of a research team I'll say more about them in a minute but the red dotted line is the area that we work in. I work in Essil here and this is the sort of like peninsula area is a group of 10 villages and those are all associated with a language which is referred to as Jola Banjal or Jola Egima and then other team members work in the villages of Bran and Jibon Khair and those two villages are just 500 metres side by side and they're each associated with their own local language as well. So I'll just give you a wee bit of very brief history before speaking a bit more about the different languages in the area. So not much really is known about the history of the Casemonts pre-colonial times most of it comes from oral history reports so in the 16 to 1700s there was a large by-nuke kingdom and then there were different population expansions of Mandinka and Jola people then in the 1700s well first of all actually the British tried to establish a base in the Casemonts but it wasn't successful so they went a bit further north to the Gambia then the Portuguese established garrison in Zigan Show and this has resulted in Portuguese Creole which is still spoken today as one of the regional lingua francas but the country of Senegal was part of French West Africa so part of the French colony and it remained that way until independence and after independence French was retained as the official language as with many different ex-colonial African nations in the 80s to the early 2000s there was an armed separatist conflict in the area that we work in the Casemonts and this led to quite a large population movement so a lot of people moved between different villages and they also crossed borders into the Gambia and Guinea-Bissau as well as a result of the conflict so there's been a high amount of mobility in the region which has also contributed to a highly diverse linguistic setting so we'll have a look now at just how multilingual the area really is so on the macro level, as I've said the official language of Senegalese French it's the only language which is used as a medium of communication medium of education in schools and it's spoken alongside Senegalese French as well People vary depending in how well they understand or speak French according to how they've learnt it and their life history, etc There are also various languages which have the status of national language in Senegal among these Wolof which is the most widely spoken some 95% of the population speak or understand it to a certain degree then we also have Jola which I'll be coming back to quite a lot throughout my talk it's key for the points I want to make and we also have numerous other languages some of them are spoken less in the area I work in more of them are spoken to the north of Senegal So, in addition to the different national and official languages operating on the macro scale all of the languages on the slide here are tested in our corpus we have a Jola group and the village that I work in, Essil is mostly associated with Jola Egima and Jola Banjal and some of these languages such as Jola Foni and Jola Kassa operators regional lingua francas a lot of them share quite a high proportion of the lexicon so, Kujoerai and Egima share about 75% of the lexicon then we also have other languages from different groups within the same Atlantic family Bainut-Guba, Wolof, as I've mentioned before and in addition to that we have even more languages so we have some from the Mandé family numerous Indo-European languages French is the official language at school in high school children learn both English and Spanish and numerous participants have travelled quite widely and they've learnt German, Italian and some other European languages as well and as I said there's Portuguese Creole because of a lot of population movement and as I mentioned quite a lot of participants report speaking Guinea-Bissau Creole and some report speaking Kazamos Creole there's some work which is just being done on this at the moment by Biagi, Biagi et al and there's a paper coming out in a new volume quite soon about that so we'll see if the language is considered to be different or not whether it's just the participants so who are the people investigating all of these languages at once I think this talk was originally built as a crossroad showcase but I think I'm the only one in the country so I'm representing here today so before we go on to talk about the other things I just have to name check everyone in the project if you want more information you can always pop on our special website and blog so the crossroads project started in 2014 and it's running till the end of next year we have numerous people in the UK based here at SOWAS and also at the University of Schechanter Job in Dakar and we're researching all different types of topics so phonology, gesture, in addition to social linguistic topics like myself just take a look at us oh it's blurry and that's good because it's not a good picture embarrassing picture and just a quick shout out to all the team members so our principal investigator is Frida Rico Lubica who's based here at SOWAS and then we had three postdoctoral researchers at SOWAS Alex, Abi and Rachel then that's three UK based PhD students and Law is our admin assistant who keeps everyone sane and coherent and then this is the Senegal team here the Senegal team leader, two PhD students and our team of transcribers without whom we would not be able to do very much so my role in the team and what I'm going to be concentrating a bit more on for the rest of the talk I look at the link between multilingualism and mobility Tweet the PhD a bit since Peter's introduction so why I was particularly interested in this village is that in such a multilingual area Sanya and Sanya and Bethan have described the village as monolingual where Jola Egema is dominant so that already stood out to me that there's purportedly this monolingual village within a highly multilingual area so I wanted to have a look at what people are actually doing in naturalistic conversations so I focused a lot on linguistic repertoires and the link with mobility so after people have travelled when they come back to the village do they then use the different languages that they've acquired with people or do they revert back to using Jola Egema and in my work I've found that it's quite useful to use the translanguaging approach so I'm introducing it briefly here now so when we go through and have a look at some of the data and discussions you can just sort of have it in your mind so translanguaging came about I think the term came about in Wales and it originally started off as a pedagogical technique in bilingual schools so in English, Welsh bilingual schools generally they found that people weren't necessarily mixing languages but still rigidly would use English in a certain class or to discuss a certain topic and Welsh for another one and people found that children learnt better when they were able to move across and within languages so Garcia and Lee Wei have done a lot of work on it and they describe translanguaging as people using their full linguistic repertoires and all resources regardless of defined boundaries of named languages Jorgerson et al named the same practice polylanguaging and they say as well that language users use whatever linguistic features are available to them in order to achieve their communicative aims and that doesn't really matter how well they know any of the languages involved they can do this fluidly and at ease so I'll also be using this to ask what Jola is so we've already seen that the state, the Republic of Senegal considered Jola to be a national language and also certain linguists, Poznanukov and Cesare consider it to be a group of related languages it's also worth pointing out as well that lots of the Jola varieties listed on the slide I just showed aren't yet documented and in my own data it varies whether people report Jola as a language that they speak or whether they list the separate varieties so keep that in mind, we'll come back to it so before I show you a bit of data I just wanted to mention about how I collect my data and what methods I use so coming from a social linguistic background if you hadn't guessed already I work in qualitative methodology and I prefer this as I don't think that the data that I collect is necessarily easily quantifiable and I find it quite difficult to make generalisations across groups of people when people have such individual repertoire and also there's such a high rate of mobility I often don't know who's going to be living in the village at any one time then it's quite difficult for me to generalise about certain language patterns and extend them across different groups of people so I prefer a fine grained in-depth approach focusing on a few case studies and I also find that using this sort of methodology allows for a lot more reflexive practice so I think a lot about the different methods I'm using and I also include myself and my own presence in the analysis of my data and in my write-up as well so I use a variety of different methods from different fields so I've done a fair amount from an ethnographic approach to include different viewpoints and voices in my work so what do people think about their own language use and to do this I take detailed field notes and do a lot of participant observation engaging in people's daily lives a lot of my recorded data is naturalistic I call it naturalistic because obviously your presence as a researcher is going to affect the data that you collect but the mere fact that you're there shouldn't necessarily be a problem if you take it into account in your analysis so I've divided up my recordings into staged and observed communicative events and basically how I consider them different is if I've asked a participant to do something specifically take time out of their day to come and do an interview with me then it's staged and if I've just tagged along with them and I'm not interrupting their day in any way and they would normally be doing some washing up or something and I just happen to be recording their conversation while they're having up while they're washing up, that's observed and I've also done a lot of semi-structured interviews to get linguistic biographical data to we've collected an awful lot of data with the team I have personally over 30 hours of different types of data and I've currently got about half of it transcribed so far most of that is actually the naturalistic data and the interviews which weren't conducted in French for the other ones I've gone through and just annotated some of them there's a lot of different issues which have come up through doing this research particularly in such a highly multilingual setting I've found that the languages which are used to carry out the research can have quite a significant effect on the data which people give how they describe the different languages in different languages so I have a little look at some data I have two excerpts I wanted to show you I can only show you one though due to consent issues so we'll watch a clip of one video and then we'll just have a look at the transcript of the other one people agreed to share within our research team but not to the public I'm afraid so to have a look at what Jola is and what I think is happening in the clips we're going to focus on one participant, DS4 and her language use the other participants in the clips have different repertoires but they all share a Jola variety so the two different contexts one of them is a women's work group who are harvesting rice in the rice field and the second one is a meeting of the Women's Catholic Association and beforehand I've been told that both of these would be completely monolingual Jola eggman situations and I asked why people are coming from lots of different areas and villages to attend these meetings would it not be easier to use some other language and people would describe it as like an inclusive way of including everyone because most people are likely to speak one of the Jola varieties okay so let's have a look at the clip the new version of Prezi makes you pay to put a video in so I think I'm just going to go back to PowerPoint next time okay it's only 20 seconds so it's very short this here this is DS4 and it will be clearer when we have a look at the transcript subtitles worked a little bit better earlier on there's always something okay so it's a pretty short clip will you be able to follow anything? shall I play it again? I'm not sure okay so the subtitles not quite as snazzy as I'd hoped but we'll have a look at the transcript anyway so in the clip we just have the transcription and then the translation so when we look at our data we don't start out by us the researchers looking at it and wondering okay what languages are in here what languages are they switching between so what we do me and my colleague Miriam Vidal presented a slightly different version of this last year at LDLT5 so we've called it the triangulation of analysis which makes it sound nice and sciency and we think that in order to analyse any bit of natural data like that we need to have three things which is a speaker's report which we'll come on to in a bit but that's essentially what does the speaker say that they're doing about their language use an observer's report so that's the second party and that could be someone such as a local research assistant or transcriber and lastly our report about what we think is going on so in the middle of the triangle we have things like experience, knowledge and attitudes so there could be anything there but these are all things which will influence people's analyses of what's going on with the data and then we also have these arrows of influence because we feel that people can influence each other so particularly the researcher and the speaker if I'm doing an interview with someone I've found that quite often I've prompted particular language names and the transcribers and the researchers work a lot more closely together and I've found that we can influence them as well this is a slightly thinner arrow because I've found that there's a bit less influence there but I think it really depends on perhaps how many people you're working with but it's a possibility so I've put a thin arrow in there so to have a look at the S4's language use and what's going on we have to have a look at the three different analyses so what does she say she does. For this we need linguistic biographical data, information about her linguistic repertoire and this is the part of it which doesn't really require any expertise in linguistics obviously from the speaker to their self-report on what they're doing so here at the top we have DS4 and the different languages which she says she speaks in her linguistic repertoire so she says French Jola which she lists as a separate language Jola Banjal, Jola Fourni and Walloff. It was quite interesting during one of the interviews that I did with her which I did in French she just told me before line one that she comes from Tangyem which is a village to the north in the Casamans so it's not in the same area so I didn't know about this village, first time I'd heard of it so I asked her okay Tangyem is that the same Jola as is spoken here so the brackets here indicate overlapping speech when she's finished speaking she interjects with yes it's the same Jola and then I finish speaking at the same time and then because I've said is it the same as here then she corrects herself and says no no no it's not the same Jola as here it's completely different but she feels that she can't speak one or the other without mixing them together she's still now for 16 years speaking Jola Egimona daily basis but she still feels that it's quite difficult to separate out the languages so that's what she thinks she does, she thinks she mixes the languages and we also have the idea of Jola being similar and different when we have a look at what the transcribers think so the observer like I said it could be a transcriber or a research assistant and generally they would have a wider background knowledge about the social linguistic setting and they would often have received topic related training so perhaps from the researchers so in our case the postdocs trained up the team of transcribers in multilingual transcription techniques using the Senegalese alphabet for writing in national languages so even though all of the languages in the area haven't been documented or codified they still transcribe them using the Senegalese alphabet so when we have a look at the transcript of the clip I just showed you this is from the transcriber DS so he says in the first segment she's speaking Jola Banjal and Fony and in the second segment Jola Banjal, Fony and also French in the first section though the only Fony word which he identifies is Jakum and the rest is Jola Egima where as in the other segment there's it's blocked out anyway the first bit is Jola Egima and then the second bit is Jola Fony and there's French under the blue bit because the lines moved okay so with the transcriber DS he consistently transcribes DS for language use as Fony mixed with Jola Egima and occasionally some French lexical items and that's consistently across contexts across transcription files in lots of different contexts then fairly recently I had another transcription back from the transcribers and it was from the transcriber ACB and in a cut of 15 minutes he'd only marked one instance of Jola Fony and the rest of it he transcribed as Jola Egima bear in mind she says that she has great difficulty in separating the two languages and is from a Jola Fony speaking area so we have this phrase here which is the only segmental phrase that the transcriber says is in Jola Fony and for the rest of the clip everything is apparently in Jola Banja so it's not the same clip we were looking at before but it really stood out for me across my files so I made me really think okay what's really going on here so finally after getting all of the other information from the researcher sorry from the speakers and the observers then I have a look at the file so for us it may be the case that we have the least experience in the setting not always and we've obviously received linguistic, social linguistic training which is influencing our methods and analyses so as I said when I come to look at it we have well I don't necessarily think it's conflicting information but we have different information from different parties so just to recap DS4 reports not being able to separate the two languages and she also reports Jola as a separate language in the second excerpt we only have one instance of Jola Fony but also other participants in the clip most of them were speaking Jola Fony as part of their repertoire so in theory there wouldn't have been some sort of comprehension problem if she had chosen to speak Jola Fony but for me I think the key for the different transcriber ACB I realised that DS actually knows DS4 and it's obviously probably would be quite a common occurrence if you're working in small communities with small language groups a lot of people are going to know a lot of your participants so it transpired that he did know her and he knows that she's from Jola Fony speaking area whereas ACB didn't know her at all and she didn't have any idea about her background or linguistic repertoire so for me that was also quite interesting because the transcriber has he speaks all of the languages in his repertoire as well so in theory he should be able to identify when Jola Fony is being spoken then when I have a look at the transcription all I can see is French which is not marked by the transcribers so we have communion and serre which means type but the transcribers don't know instances such as this so we've got various different takes on the same data so what I wanted to have a look at now is to consider then if the data is the same why should we perhaps change our approach to analysing and thinking about the data so moving away from looking at discrete and bounded languages towards a more practice oriented approach and for me it's quite important to question sort of who decides where the boundaries of various languages are whether it's the speakers or documentary linguists whose opinions are being represented in the analysis and also if we're going to use a translanguaging approach does this mean that we have to totally disregard concepts of bounded languages so within the past ten or so years particularly in social linguistics there's been sort of a turn from looking at the structure of languages to more of a focus on people the context of situated language use focusing on peoples, repertoire biographies and their histories to sort of incorporate as well peoples perspectives into a more holistic understanding of language use and this also fits with different approaches so people for example who might be analysing similar data using code switching or code mixing would be using much more of a structuralist approach whereas a lot of the work done on translanguaging and similar terms comes from a post-structuralist idea so a lot of work has been done on translingual practice by Kanagaraja there's lots of different terminology kicking around translanguaging so we have terms such as polylanguaging translingual practice, languaging I think there's even more but they're all essentially describing the same thing which is peoples linguistic practices regardless of these bounded languages and Kanagaraja himself says it's quite difficult to speak about new approaches and paradigms like this without using terminology from older paradigms so even if we're talking about linguistic practices it is quite difficult to do that without speaking about named languages and he says that obviously labelled languages and language varieties still have a reality for social groups so we can't dismiss them sort of offhand they form important parts of peoples linguistic practices and this move as well towards translanguaging also considers as well spatiality so we're again concentrating a lot on the context and the situation in which different communicative practices take place so this also allows a lot of accommodation of diversity and unpredictability but quite a nice way of conceiving of languages has come out from our research group so Rachel Watson is one of our post-doctoral fellows and she's been considering languages as categories using prototype theory so instead of there being a category of French she would argue that there are certain words that are more prototypical of French than other ones so Taylor says that humans make categories around the conceptual core of a category so not using language as an example many people have an image of what a cat looks like and that would be your prototypical cat maybe it's a black Halloween cat then there are other types of cats which are further away from what you consider the prototype to be so perhaps one of the sphinx hairless cats for me that's a bit further away from the prototype or a lion, it's still a cat we can class them as cats but depending on your culture or individual people we will have a different idea of what the best example or a prototypical example of that category is so Rachel's decided to apply this to the concept of languages in our research setting so for her that means different linguistic elements perhaps it's a word or pronunciation or some form of construction will be more prototypical of certain languages than others so this allows for a lot of variation between speakers so I don't know if this might help but this is an adaptation of a Venn diagram from one of her papers which is coming out quite soon so for example for Jola Egima which I've put in the blue circle here for example there's a feature which isn't shared with any other Jola languages so they use the voiced and all of the other Jola varieties use the unvoiced so you would put in the Jola Egima part at the top and then something for Jola Kijiroi which is more typical would be here and then sort of the shared 70% of the shared lexicon would be sort of this bit here between Jola Egima and Jola Kijiroi and then you can add many other different languages on and then this overlap here which I've added in Jola is what I'm interested in so where do things overlap and what does this mean so I posit that the bit in the middle of all of the languages of all of the Jola languages is not necessarily a language but is rather a language in practice so how people communicate so if people do want to they can speak the separate languages using phrases or words or pronunciations which are more prototypical of the others but for a lot of forms it's quite difficult to tell which Jola language is being used and this is also why I think in the second clip we have everything a banjal for DS4's language use the instance of Jola Foni because in that instance she's using a form which is prototypical of Jola Foni whereas many of the other forms in the rest of the clip are shared but because the recording took place in Essil where people expect Jola Egima to be spoken the transcriber is primed for hearing Jola Egima and then only when DS4 uses the really prototypical Foni phrase does he then transcribe it as such so think that you don't although some people when they talk about trans language have gone so far to say that there's not necessarily these separate languages I don't think that's necessarily the case in our research area and also for my sort of my idea about Jola as a language practice there's a little bit of support from a couple of other research settings so Casper Euphemans works in the Gambia so it's very close to the Casamance and it's a very similar multilingual setting to what we work in so he did this 2015 study where he looked at multilingualism literacy and linguistic practices and the quote here was taken from a series of interviews which he did with people when he asked what language would you like to see taught in schools and the respondents were very resistant to naming any particular language and instead they preferred to talk about black peoples language which wasn't any particular language but was more a representation of how they communicate using different languages without necessarily assigning them a particular name and well Euphemans even go so far to say that now the field of social linguistics has sort of changed from the fishmanian paradigm who speaks what to who where as to who languages what to whom when with what resources and under what conditions which is again taking the focus away from the separate structured languages and moving on to peoples practices another study which is coming out quite soon I had a preview of the paper on academia is by Isabel Leglis so she works in the French Guiana Surinam border area which is in South America which is again highly multilingual and there are lots of different languages from different language families spoken there so it's sort of a fairly similar situation to ours where there's a group of related languages ones from different language families she has some really interesting data so in a couple of her conversations there's data and you can't tell which language is being spoken it could be one of two and that's for every single linguistic element and she found that actually speakers prefer to use these ambiguous forms which you can't associate with one language or another particularly when communicating with people who they don't know so they don't know their linguistic repertoire so they actually prefer to choose these forms which are shared among different languages which I think is what is happening with Jola so I think when people report speaking Jola they report being able to speak different Jola varieties but also to use these ambiguous forms which are shared among the different varieties and they're doing this as an inclusive linguistic practice as people said to me before not everyone speaks French or Wolof but many people will speak one Jola variety so there's a fair chance that they speak some of the overlapping forms in the middle so it's just a brief conclusion to try and bring everything back together so Serge Sanyer describes Essil as being a monolingual area so even though I don't necessarily agree with him I think that no that's not true it's not that I don't agree with him I think that we can include Essil as a monolingual area if we take a different approach a translanguaging approach so Essil can be seen as a prototypical area in which to speak Jola Egima and this forms part of a wider Jola linguistic practice which is spoken in Essil the crossroads area and beyond the prototype approach then we still have separate languages with social meaning for speakers and I think that all of these different situations exist together in the same highly monolingual area and it's sort of one of the reasons why I think that conducting social linguistic research in this area is so interesting and challenging and that's me, customary flick through the references thank you Comments or questions for Sanyer if I can start off how do you think children figure out what the hell is going on in this kind of context do you have any information about it? there's been a project on child language acquisition so we should have some ideas soon but there's a particular type of language acquisition strategy which tends to be used in that area so Calve and Dreyfus refer to it as sejo-languistic a language trip or a language stay so fostering is quite widely practised in the region so if a family there's an aunt and uncle who don't have any children then quite often their brothers and sisters will send children to go and live with them for a while but particularly in the Casemonts area they either need to a village or an area that has a different social linguistic setting so different languages are spoken there so children will pick them up and also this is also practised by adults as well so one of the friends that I made in Bram in one of the villages she just went to one in Jamboreng a village which is on the coast and they speak Jola Quartai she's like I want to learn Quartai so people just go for a month to a different village to learn different languages and it's sort of cemented as a practise from childhood but how they learn to differentiate but also to move between them we don't have the data yet I guess the question is not so much the as you say you've got this overlapping core is how people actually separate them off so this one is funny well that's the thing where the prototypes really work because for a lot of it it doesn't seem to be able to be separated but when there are the distinct features then people can otherwise I think we get the instance like in the second transcript where you get oh that's default Jola Egima and only things that diverge from that will be marked differently of a particular group or in general so I don't necessarily know an awful lot about the historical social linguistic side of it but the original inhabitants of the region were the Bynook population and then there was sort of like a Jola expansion and a lot of the original Bynook people were incorporated into other ethnicities or language groups so in some of the languages you see there's some similarities between Bynook and Jola and in some they're a bit further away and on the list that I showed there's a couple Aramae and Bayot which are sort of controversially included within the Jola group because people think well Cesarean and Posnokov posit that we don't really know but they think that those languages were originally a lot more different but they became similar through convergence so through language contact with different groups but we don't really have all the data on it so some of the Jola varieties are quite specific to different villages and some are used more widely and in addition each one will have its own sort of social linguistic setting so that people who speak those varieties will be in contact with different speakers of lots of different languages but more than that I don't know Can you tell us more about this distinction between speaking and say language and how do you interpret this with your own like why would you know that they are speaking in a different language and it's a different approach but really what's the difference? Well essentially the data is the same, people are just doing the same thing so that's I can see that there's not necessarily a difference in the data or what people are doing, people have been speaking like this for a long time but I think for me I still refer to the fact that people speak different languages it's still a concept which people talk about and they're familiar with but before I really had some of my interview data and before I did focus groups I still was quite fixated on the different languages and how many languages do people speak but particularly when I was getting answers such as Jola in my data I really started to wonder what was going on particularly when the postdocs who were looking more at syntax and semantics and things there was so much overlap that they were at one point they were looking at which languages borrowed certain forms from other languages but they just couldn't pull anything apart and I originally was sort of quite not dismissive but if people said to me what languages also do you speak and they would say Jola then I'd always prompt and say well I know there's loads of different Jolas so which Jola, the Jola from where and lots of people would still insist no Jola so that sort of made me think differently about it and then I did a focus group on what people think about sort of multilingualism, Jola and languages and actually I don't think I mentioned it but it fits in quite nicely with the prototype theory of Rachel's the participants were speaking about Jola and they described it as something that's really vast and then the moderator said ok so what about everything else, what do we do with it is Jola and Kimura language and one of the participants said no no no those are all categories so they didn't classify them as language but as categories so that fits in quite nicely with Rachel's but essentially it's still people speaking to each other and communicating in a different way of theorizing what they're doing but putting the focus really on the communicative event rather than starting with these sort of a priori assumptions about where the boundaries of certain languages are but one of our colleagues in Senegal is still working on like code switching from a more structuralist approach so I don't think that they need to be separate it's sort of better to get all of the approaches working together It's very interesting I'm curious about language practice at school if you have any data or you said that language instruction is French but as you know the trans language was a very significant idea to talk about language practice at school so if you don't have any ideas or any data I didn't get any permission to do recorded data but I did observe a few classes in the school so in the village of Essil they just have school the first stage of school so up to the age of 10 and then children have to go to another village to attend middle school generally what I saw children in that area before they start school only have some familiarity with French so generally in the first couple of years it's very much focused on like a transitional bilingualism aiming towards French monolingualism still there are still reports that a lot of the children once they get into the upper two years of school are being like being forced to wear a symbol if they speak French in class but for the younger years it's really the main medium of communication which is used to teach French in the first couple of years to get children ready for the later stages they in theory no but they do in the first few years and they do in the playground and everything as well but a couple of the teachers who come from outside of the area so when I was in my second field trip a week before me a new teacher had arrived and he was meant to be taking the nursery class like the smallest class but he didn't speak any Jola languages at all and just spoke Wolof French and I think Mandinka and he had to swap with another teacher because the children then understand what was going on and the other teacher used Jola as a medium to teach French but that's really still the focus on the education system getting them to acquire French as a monolingual how do they become in the end and in French or do all children and host of course become very proficient in French there? No, it depends really how late they stay in school and if they progress to the secondary school or not I think also recently in the last year or so the school system changed they used to have to do a leaving exam at the end of primary school to be able to progress to secondary school and many students wouldn't pass it but that's now sort of been I think they still have to take the exam but they're allowed to progress anyway so there's quite a lot of variety really and also even though French is mainly acquired in school it's not the only way to acquire French and in the village where I work have acquired French through going to work as domestic workers in the town working for French families or French speaking families from other African countries so it's not the only sort of way to acquire French but it does sort of create lots of different tiers of French speakers when you talk about the kind of triangulation and you have those three different sources how do you store that information and where does that end up to you have the LR files from the transcribers and do you add in the extra information about French do you add in stuff about... on the files? so we have like a there's a separate column I show you the version which is before it gets to me so we have the transcription column then there's one for... well I added this in translation into English for the subtitles then we have a note section so when the transcribers get it they have to transcribe it and then translate it into French and mark it for which languages they think are being spoken in each segment then when we get it we have a note section so we can go through and comment on and add things if we want to and then we would save the two... I saved the two separate versions so before I get it and then after I get it and then yeah you add English for the presentations and then did you mean like storage and things as well or did that answer it? I was wondering I was thinking more about things like where you have job and sometimes that kind of central job area is being interpreted as the default for that area and other times it's being interpreted as the default for what the person knows about that speaker whether that ends up skewing the data in the nonsense yes I think it does but I don't necessarily think it skews the data I think it's like another take on it like you were to just sort of take the data at face value and what the transcribers say about it and I think it is you could say that like the data is skewed in a certain way but because we have like a team and we check it with different transcribers and then also we have our take on it and the speakers take it doesn't necessarily mitigate sort of the bias but it does give you different perspectives on the same data perhaps the postdocs have a different take on it when they're working more with they've got Excel sheets and Excel sheets of all these different paradigms now class paradigms for all the different languages and the overlaps so they deal with their data in a bit of a different way It's really impressive and I think you're right that you need to kind of include all of those things is great I also use Envivo as well which is quite good for tracking qualitative data because you can pull lots of different things in so if you export your LN files as Excel or tab delimited text then you can pull it and then it works in like a node hierarchy which you can create yourself and you can also include things like photos and videos so I pull things in there and then you have to create people and links between people so then in that I would say that DS4 knows DS in this capacity and in each LN file the transcribers have to say who like they put it in the little field who's transcribing it so you can track it and sort of track the relationships between the speakers and transcribers and things As you talk about methodology I think your methodology is very interesting and you mentioned about the fact that you don't exclude yourself from the search and you constantly yourself as a part of it So could you give us some more examples maybe of how this happened throughout your research Yeah So I think it sort of started in the first trip I went on which was quite short and we just did a couple of semi-structured interviews with people and I went through afterwards and they were in French and I transcribed them and I don't think then I pulled out from this it was just a discussion with someone about what language she speaks and I pulled out quotes from it to use in a paper about how she said that Jola is her mother tongue and I thought oh this is really great it's nice data then the next day I went back and revisited it and I watched the clip again and I realised that if you just take her out of it and not me then you actually miss how I've sort of prompted her because if you go a couple of minutes I think it's a minute back then she we're talking about French and she says oh my French is really dreadful and I was like no no no no it's like way better than mine my French is dreadful it's not even my mother language I said like English is my mother tongue and so I've sort of already the idea was in her head and then when you watch the video back you can see when she says it she says because I said it like in a particular way as well I said English is my mother tongue and she repeats the exact same phrase and she says Jola is my mother tongue but taking it out of the context without including me is sort of quite yeah it sort of gave me a different reading on it so I guess that was one instance and then also in English in schools like I said so when we're around people are always trying out English and it just makes the whole multilingual setup a lot more different but yeah I think Mia did a piece on the so using the communities of practice approach so we have a base where the transcribers work in the village of Bran if we're in Bran and we're going to work with the transcribers then we go along and she actually did a little study of our language use among us and the transcribers and how we're all sort of influencing each other and learning all of the different languages which is quite interesting but yeah there are like some serious parts of my research where it's like if I take myself out of this I've got a totally different view on it and also people speak about me so when even if you're doing like a participant observation recording or something then quite often you see people being spoken about or people will say Sam's around you have to speak and then you listen back and you're like all of the languages multilingualism but yeah there are all these different sort of language ideologies to navigate but a few different points really I'm wondering if you are anyone in the team that's looking at say a particular individual going to different locations and then seeing what they do in terms of their language practices according to who's there and who they're talking to well we did as part of the team we did a social network study and as part of that we chose two key participants from each of the three villages and then they were miked up with pretty much something like this a lapel mic and a really small dictaphone and then their language was recorded throughout the day and there was no restrictions on where they could go or who they met or anything and we've used some the because of like consent issues there were big village meetings where we said okay on certain days so and so is going to be wearing the mic and we'll be walking around so just be aware of that and particularly actually in my village my participant would before he started a conversation would always draw attention to it and be like look I'm wearing the mic is this okay which I think some team members weren't that okay with because it's sort of interrupted the flow of conversations but I don't mind because then you definitely know good consent for that and after 30 seconds they forget anyway so on a couple of instances people did go to other villages and we also found that a lot of people from the other villages would come as well so I've got some nice data where I think it's a couple of Rachel's key participants have just come around for lunch in one of my participant's house and we've got some really nice data there where it's sort of again it's the whole JOLA thing some bits have been marked as JOLA Kajira, Somers JOLA Banjao and a lot of it I think is just JOLA but I would like to have a bit more of a defined approach to see different people especially when they go further a field not just within our area but what happens when they go to town as well I think Sam's opened up a whole bunch of different perspectives on the ways of looking at multilingual contexts like this so let's thank her for a really interesting talk