 Okay, sorry, we're a bit late, we'll take a few issues. Welcome as I'm going to go ahead and leave it at the end of the seminar. I'm very happy today to welcome and introduce Chege Di Tiore. You're probably all familiar with him, at least for those who are with us. So Chege Di Tiore has been a senior lecturer in Swahili at Swahili. But how long now? Quite a while. He's also a member of the Centre of African Studies at Swahili as well. He has quite an impressive academic background, very international. He studied in Mexico at the University of Autonomar Metropolitana. In Michigan State University for his PhD. He taught at El Conejo de Mexico, Boston University and Kenyatta University in Kenya. So quite international background. His areas of expertise are broadly language and culture, but most specifically African languages about the use, their spread, especially in Eastern Africa. He's also interested in lexicography and translation from and into Swahili. Kikuyu, it's your native language, Kikuyu. Spanish and English. In 2002 he wrote Swahili Spanish Dictionary, which is probably the first one. He's also an expert in urban youth languages with the study of Cheng, which is a youth language spoken in Nairobi. And you mainly talk about it a little. Yes, I would. So today we're going to explore social linguistic issues related to the linguistic situation and landscape in Nairobi. So the title of Chege Stoke is A Match Guy's Test, Experimental Study of Attitudes Toward Varieties of Swahili and English Spoken in Nairobi. I leave you the floor. Thank you very much, Rosé. And for a wonderful introduction and apologies for the delay. There was a slight hitch here. Right, so let's get started. Right, that title of my talk, that long title I think reveals most of what I'm going to say, is you may have come across, you know, those who are studying linguistics or social linguistics, in particular, psycho linguistics, you may have come across a mention of this experiment that has been used in several places, originally done in Canada, but it has been adapted in various places since then, and with interesting results. So I will explain why I thought this might be a good thing to apply in the case of Nairobi. Right, so a bit of background first. What I'm going to talk about has a lot to do with change in language, linguistic change. And as we all know, at least those who are in linguistics, linguistic change is as inevitable as death, so to speak. Languages must change. Languages always change. And there are two main reasons. One being internal reasons. These are internal dynamics that bring about changes. For example, the articulation of vowels or some consonants may change their quality. But also through contact with other languages. That is also a major reason as to why linguistic change takes place in the system. In the case of Kenya, with 62 languages and dialects, you can imagine there is quite a lot of contact happening and therefore there's quite a bit of mutual cross-linguistic influences. These are just examples from lexical innovations which tend to be the easiest ones to notice. We don't notice grammatical changes as quickly or as readily. We may not even sometimes not quite observe or perceive phonological changes, but when a new word comes in, we take notice immediately. So many speakers of Swahili may or will be able to recognize, maybe not Chahi so much, but they will say Chahi comes from Hindi, although we know it comes from beyond that. Kahawa, some might recognize, comes from Arabic. Computer, you may recognize, comes from English. Maktaba, again some people might recognize this as an Arabic word, especially if they have a bit of Arabic. Hoteli, Hotel, Basi, Burse and so forth. So these are easy to notice, but sometimes the other kind of changes are not so easy to notice. Again, very broad background about Swahili. Some of you may be aware of the language spoken by about 80 to 100 million people, mainly in East Africa, but also Central Africa, DRC, mainly the Democratic Republic of Congo. The vast majority of the speakers of this language speak it as a second language with varying degrees of competence. At this point we'll come up in the later study that I'm talking about. So degrees of proficiency on one hand and also substrate influences will always come into action. The indigenous native Swahili dialects are in great danger, mainly because their speakers are minority groups, even though their language is very big worldwide, but the actual native communities are also endangered alongside other languages, mainly because of the onslaught of standard Swahili, English and Shane, and some of them are likely to be dying. Now there's a view which I want to bring in here because it's relevant to what I'm talking about, that languages can be viewed simultaneously as discrete units or particles, you can list them, language wave, B, C, D, L, 1, 2, 3, 4, or the languages can also be seen as bundles of features across time and space or waves, which are best studied in terms of their variation, how degrees of variation, types of variation, and so forth. And this is a kind of very, very interesting for linguists because in a sense they give us an idea of linguistic change in progress. In other words, because we can't spend thousands of years waiting to see the changes that are happening in language, we have to find other means of trying to anticipate the changes, and one of the ways is through a synchronic study of cross variation across dialects of the same language. Another view is to see language as part of a larger ecological metrics or field where the functional roles and usage of linguistic codes for a wide range of purposes are more in focus. And here we have a couple of examples, I would say Arabic and Chinese offer these type of arrangements where you have very many distinctive varieties of Arabic versus what might we call the standard Arabic, which also depends on the country, so you might have Moroccan standard Arabic or Egyptian standard Arabic and so forth. Same thing with the Chinese, multiple Chinese languages which only share a common written form. So this is one view of language which I would like to see to use and introduce this idea of micro languages which captures all those things that we look at language as comprised of waves of features that extend across time, geography, and space. And these different varieties of the same language, of the same macro language play different roles in different domains, at least in the case of Kenya. And so Shen, Kenyan Swahili, and standard Swahili, or Swahili Sanifu for those who understand the language, these are three different varieties of the same language as part of the macro language which extend all the way outside of Kenya into Tanzania, into the Democratic Republic of Congo where you have all other different kind of varieties of the same language. The key thing is that speakers and others recognize this as Swahili but the variation can be quite big. And so I want to look at Swahili in a sense as a macro language because sometimes the degree of divergence between some of the ways of speaking understood as Swahili sometimes they go beyond dialect. In other words, sometimes the differentiation is fairly extreme to the extent that you might start thinking about a different language. So the course of varieties of Swahili, for example, the southern Tanzania and northern Kenya are geographical. They are regional and these are native speaking communities. They are also part of the macro language but they will be of less focus in this particular talk. The social variations and other regional ones are more interesting in my view especially those ones that are in contact with many other different languages of the region. So in Dubil is one documented spoken in DRC, again youth language and so forth, and Cheng in Kenya, somewhat comparable but also with their own slight differences. And of course the wider regional or national group patients are also of very much interest to me because they now take shape as Tanzanian, Kenyan or Congolese Swahili and speakers can immediately or generally they recognize each other as such. Oh, that sounds like Congolese Swahili. Oh, that sounds like Tanzanian Swahili. Oh, that's definitely Kenyan or something like that. So they are this mutual recognition and this is what I want to explore a bit more. These are just tools of analysis, speech community, I'm sure you are aware of all that, norms and values. So this is important because it's not just understanding and knowing the rules of grammar of Swahili. It is also about the rules of communication within a particular speech community or let's say Kenyan Swahili or Tanzanian Swahili or DRC Swahili. In other words, a speech community must be defined slightly beyond simply the language that one speaks. The English you speak out here in London may identify you easily as a non-member of a speech community, let's say for example in Liverpool, even though you're speaking the same language, but the rules of talking, pronouncing, maybe gesturing even and so forth might be... And this is what we use as a basis for defining a speech community or community of practice which is a term that is increasingly more in use. But speech community as defined by Delheim's way back in the 1960s, I still find it quite useful to use. Right. So this idea of vernacularization of Swahili is what I'm going to explore a bit more because it's where Kenyans in this particular case seek a national identity through the use of Swahili while at the same time retaining an ethnic identity. This interesting play is what I want to examine a bit more. How does one retain, remain, how does one express their own ethnicity in Kenya while at the same time speaking a national language such as Swahili or even English for that matter, but somehow retain elements of that identity. And I'm going to argue that this is done mainly through language. So I did talk about Kenyan Swahilis, right? Here we are. So several distinctive and recognizable varieties which also serve as markers of identity. And I'm talking about native Swahili dialects. We know their names like Kim Vita, spoken in Mombasa Island, Kiyamu, spoken on Lamu Island, and a few other dialects of the region that we might talk about, Kiba Juni and so forth. Then we have what I'm calling vernacularized Swahili, which is recognizable through by its substrate or mother tongue influence. So somebody speaking Swahili, you can tell maybe their first language simply by the way they speak their Swahili, not terribly unusual. Cheng, now of much interest in recent times, is a growing urban Swahili-based vernacular largely associated with the youth. Because this idea of youth languages is becoming a bit contentious, especially when now you find people who cannot really be defined as youth, speaking a so-called youth language. Then what do you do? If a 50-year-old is speaking Cheng, you can insist that this is a youth language? I don't think we should. In any case, maybe you'll agree with me. Age is just a number. Right. And then of course we have what I call educated Swahili. Standard Swahili, which we teach you here at Source and elsewhere. The standard book, the book that you have, the Swahili that is in the dictionaries, the Swahili is in the textbooks where, you know, and that's the formal variety which we all know. Now most Kenyans are able to navigate between some of all these varieties which exist on a sort of a continuum quite easily. And these Swahilis are marked lexically, structurally and even in prosody. Internation, stress patterns are different when one is speaking Cheng as when they are speaking Swahili. For example, Cheng has a way of lengthening the final syllables in ways that Swahili does not. So you can say, well, that word may be Swahili, but you're pronouncing it in Cheng, for example. Right. A few examples. Things have changed. For example, maybe those who are taking Swahili, they have been taught Hujambo. And you say, Sijambo, right? Or if I say Habarigani, Zuri-san. Okay. I don't think I've taught you to say Psasa. Okay. Which is like, what's up? Yeah. Mambo. Maybe you've had that if you've traveled around East Africa. But these are now fairly, what was very informal greetings by youth. Ten, fifteen, twenty years ago, you know, including the president is talking that way, for example. So it's no longer very much youth. So these have been changed. There are a lot of changes in dynamics. I mean, not going to full detail today, but that's part of it. So I look at Swahili in this way. Kenyan is in Kenya, right? There's a little dot there called Sanifu, standard. Sitting in a sea of all these variations of ways of speaking Kenyan Swahili. That's the way we see it. And so I'll be talking about that gray area that largely, not the blue one, which is already nicely fixed in the textbooks and dictionaries and so forth. So I look at it in Swahili a little bit that what I mean by a continuum, a concept that we use a lot in linguistics, in social linguistics anyway, where you're not talking of black or white. You're talking about some kind of possibility of movement along a continuum of speeches. So we are not always stuck on one part here of the standard Swahili. No, even in ordinary speech, in a short conversation, one can easily move back and forth along that continuum. So the idea of continuum is useful because it does not set stuck boundaries, which then limit us from exploring more things. So let's jump into the topic more properly. Language attitudes. These are important. They are part of a community's linguistic culture and they should not be ignored. That's fast forward saying that. People develop the ability to identify a language and make associations which are derived from the surrounding environment, culture and society. So these attitudes are culturally informed and very importantly, they can congeal into stereotypes. Now, attitudes can also be good indicators of social stratification of language and they are relevant to a nation's planning of language and educational policy. You know, we cannot create good policy if we ignore the dynamics of this language. And this has happened in many countries of Africa in fact, that the reality is ignored in favor of an idealistic program, language program, which 40, 50 years down the line, proves not to be to have been the right one. There's an example with many examples including Kenya, Morocco and others. So in order to test these hypotheses of attitudes, I conducted this much guy's test which is well known and I will talk a little bit more about it now. But before I do that, let me just, before I leap into Nairobi, have a quick look at that article because it's also talking about stereotypes. But they're not Kenya and not Swahili stereotypes. Maybe one second to have a look at that. I have highlighted what I think are key words that are relevant to the discussion today. So it's an opinion column in a local newspaper here in the UK which caught my attention, it's quite not too old, it's in July during this summer. And it caught my attention because it was talking about things that I am trying to research about in far away Kenya. And the key words here, diversity, that you can translate that into multilingualism or multi-dialectalism. That's what really this person is talking about. They are talking about prejudices. I've just mentioned stereotypes. I've just mentioned about things, ideas that people have in their mind. That's prejudice, prejudging. There's this new word for me, region of bear. Okay, whatever that means. But I understand that to mean a characteristic way of speaking English in a particular region of the United Kingdom. We know the word thick, we know where stockport is but now we can see a distinctive stockport accent. We can see the word like northerness. We can see words like hybrid accent. Okay, we can see this long sentence there which caught my attention because it appears in Nairobi again. This person gave, I gave the claim, the question that those I am meeting have mistaken me for the man who has come to clear out their drains. Why? Because of the way he spoke. And then he's finally out of it the polished clip from four months who offer very little ales. Why? And the last question, very interesting. Who says that our leaders are not allowed to sound as though they should be? If that's exactly how their constituents sound. In other words, this person is asking, how do they speak differently than how they speak to others? So this will come relevant and I brought it today just for that. So back to the match guys experiment. It started out in Canada. It was a study of bilingualism. Immersion schooling led to, you know, because of the division between the Quebec province and the rest of Canada. So you have two languages. It's a bilingual country, French and English. And there has always been a bit of a tension between the two and the bilingual schooling system took effect and people wanted to understand how this is happening. What's the result? What's coming out of that? And in this study, both English and French subjects ranked English guises. I will explain what guises are in a minute. Although each group differed in other ways. But there was consistency from both groups on the higher ranking of English voices. Just call them voices, the guises, the recorded voices on those traits. But French speakers were judged by French subjects very highly on elements such as religiosity, kindness, sociability, and so forth. Now following this study, it was adopted for many other situations, especially in North America, looking at Spanish versus English or Chicano English and so forth. So let me... I don't know. I want to explain how this works. Okay, it's very simple. You have a recording, okay? Yeah, let's look at this one. Some of the early... The first Canadian was simply testing voices in English and voices in French, okay? But you have the same person reading out those things, whatever passage it is, or whatever statement they're making. But once the judgment has been made, all right, how does that French-speaking person sound to you? So it sounds try a religios, try this and that. They will say what they feel, that voice is telling them. Notice there's no face involved here at this point. Now once you put a face to the French voice and you have a face to the English voice, then in the next stage you switch them around without letting the participants know. And then you ask them again, what do you think of this English voice and what do you think of this French voice? And amazingly, as you can see with the other experiments with African American children and white children, the voices were still judged in the same way, right? Okay? People were basing themselves on the language, on the manner of delivery, not the face or the speaker behind that himself, okay? This will become a bit more clear. It has been applied to a Welsh study, which I also find quite interesting. And among the conclusions of the Welsh study was that speaking in Welsh can create solidarity among the Welsh speakers, but more importantly that a Welsh-accented English is an important marker of identity of a Welsh person, okay? And that the RP-pronouncing person is sometimes seen as alien or as standoffish and so forth. In the case of Norway, again I'm just going through some of these studies that have been in the past, in the case of Norway they wanted to find out how people react to different accented Englishes spoken by Norwegians. Quite a lot like what I'm doing in the case of Kenya because what I'm looking at is how people react to different Swahili accents, okay? As second language speakers, not as native speakers. So here in the same way, the researchers wanted to find out how do people react to a Norwegian speaking with a British accent, for example, as opposed to one speaking with an American accent. Now these things are important, as you know, in places as far away as Japan, there are studies and experiments and policies that are based on understanding of how people react to British English or American English. And in some cases you find that one is preferred to the other. This also is another example of this kind of study. And so what they found very interestingly is that contrary to what they thought at first, if you look at bullet point number two, that a Norwegian accent, at least in their conclusion, was not a turn-off, right? Norwegian accent in English. In other words, a Norwegian speaking with a heavy accent, as we say colloquially, was not necessarily seen badly. Apparently this is what was thought before. So the study showed that actually this is not the case. And that on the other hand also, it does not necessarily make you sound nicer when you speak with a Welsh accent, et cetera, et cetera. So these are just brief looking at those results. So let me talk about the Nairobi test itself. The idea is to gain the true feelings of Kenyans about their languages, right? What do they actually think? Whether they hear somebody speaking Swahili with an X kind of accent, okay? As usual, ethics were involved, so everybody was, there's no deceptions here involved. But obviously, there's a kind of a trick, if you want, because people are not really aware of all that I am doing. However, it's not unethical because it's not interfering at all in any of their way. And informing them also in some cases when I did, we didn't seem to affect their willingness or ability to do the judgment. And so I selected, I obtained a professional voice artist, actually a well-known personality, radio personality, and I'm afraid that drive is not working where I would have played a few more because they are on that particular drive. But I got this man to record 10 different voices, okay? In these ways, in what I call Kenyan English, as defined by variously, of course now we are talking of different Englishes because again, just like the Swahili, it's not very difficult if you are Kenyan to recognize a fellow Kenyan when they are speaking English. There are peculiarities of the way we speak English. You can tell, you can tell a Ugandan, okay? You can tell a Tanzanian, more or less, you know? And so it's not unusual, you know, these things, they are there. And so Kenyan English does exist, as Kenyan Swahili does as well. So he recorded in Kenyan Swahili in Shang, in Swahili Sanifu, and he's very good, by the way, he's very, you know, well, he's a professional, you know, he's in a studio and again I could show a few pictures, but it's not working here. And then he recorded in these vernacular Swahilis, all right? And I selected these particular languages for various reasons, obviously. Number one, their size also. But also their ubiquitness in the city. They are very, very present in the city. These are the main languages, the main ethnic groups of Kenya. Actually, all of these except the Maasai, and the Somalia to an extent, okay? But even though Maasai and Somalia are minority, they are proportionately well represented in Nairobi. Partly because Nairobi is located where Maasai country is, so it's in their home, literally. And Somalia because many of them are Abans, you know, people. And so they're into Aban activities like business, you know, and so forth. So there are a lot of Somalis in Nairobi, as well as these other groups. And I had a team of, obviously, in my research, there are some people helping out, very, very good people, but these were involved in a separate project, not in this particular recording. And so what I got him to record was these texts and I really disappointed if we can't listen to some of these to be... Oh, no, I mean, the... Oh, here they are, sorry. Okay, good. So the guys are here, right? So the idea is that my respondents will listen to these voices without knowing. And it didn't always fully work out. Some people, some respondents figured out that, hmm, that sounds like the same guy, actually. But mostly they didn't. And those who did, they were kind enough and polite enough not to show. And it didn't matter very much, really, because their views were not really affected by that knowledge. In any case, I did explain to some of them that, look, this is an experiment and, you know, you will hear different voices. Some could figure it out. Others actually did not figure out that it was the same person doing this thing. So, I actually presented a little bit of this in Nairobi about a month ago. Because it was a very different audience. It was very interesting because what I did was I played all this and the audience, we spent about 40 minutes discussing them because they became sort of part of the experiment now. Because it is very clear and obvious. And if some of you may be familiar with different ways of speaking in Kenya, you might recognize that. But that was the beauty of it, I thought, because the people could hear, oh, no, no, no, that must be. And there was a bit of argument and so forth. But listen to one of them if we get voice on this. It's on mute. Sorry. Okay, let me play that again. Maybe I should play, let's see, the English one first for your benefit. So you know what it's all about. In particular, those who live in the same plot are the worst and they are the meanest. Those that really make me angry are those who think that the plots are special, that they don't want women and they don't want the people who are of the chain line. When she asks if her room is vacant in Nairobi, it starts to ask stupid questions. Like, first of all, are you married? Don't tell them you're back. Who's is it? And your husband, where? Where does he work? What does he do? Another one says he's a virgin too, but having a child is supposed to be for her because he doesn't want to keep him with his children. Some of you learn as much as you're being stupid. Okay, so it's a really short extract which I did not create. I actually obtained it from a column in this weekly magazine called the Nairobian. And it's a very interesting magazine because it is the first one to recognize Sheng and Swahili and English as the languages of Kenya. So they are publishing in the three ways. So they have a column in very good standard Swahili, written by an... And on top of that, they always have one in Sheng, written by a very well-known radio and TV personality who has made his name from simply from Sheng language by starting up a radio station, not himself, but being hired as one. And so anyhow, I selected that article for several reasons. First, it resonates very easily with the Nairobians because the question of rent and bad housing and bad landlords, it's emotive and always is good when you're doing a linguistic study to bring out a topic that brings out excitement. And it did actually because people shake their heads and feel bad about these landlords. So that was one reason because he has many columns talking about social commentary and various things. Secondly, I did not have to create something that is stilted, that is not really running well. So I thought this would... At the same time, I also wanted to have a sort of standardized piece that did not involve too many, let's say, hesitations or backtracking or things like that. And therefore, this worked out very well because A, it's published. And B, it can be easily read out by the speaker and so forth. So he read that in those 10 different ways. I don't know if you'd like... So the first one was Sheng and if you recognize that, you might... Okay, so that's basically the text. You've listened to it. It's in English here. And so it's in these three languages now and then in these different ways. So I subjected this to a variety of people. And, you know, younger students, younger kids, and then adults. And among the results of that, here's that. I will not go into much detail because of time. So here are some of the findings that came out of this particular experimentation. I mean, what I did was I encouraged them to give... You know, you can see we are listening in a group. I would play on my laptop and I have an external speaker, quite loud, you know, nice and loud. And I would get the teachers to give me a room, spare room in the school. And then I worked in groups of five because working individually, you know, each one of those is about a minute and a half long. Okay, times 10, right? And then the questioning, the, you know, filling out the... If you did it individually, very, very long and tedious. Which is fine if I had all the time in the world. And also if the teachers allowed me to stay in the school all the time for hours and hours and interrupt their classes and so forth. That's what I opted to do in groups of five. And then I would encourage them to respond and give back feedback. You know, we sort of a little session together in groups of five. So I encourage them to give a shortest description possible of each voice, such as this one you can see. Very short adjective, like good, sincere person and so forth. And so here is what we are looking at here. What I ask these kids, after we play the Shang one, I ask them, how does this sound like? Okay. And these are the responses. Oh, they will say, oh yeah, definitely he's tall. Do you see the interesting thing about this thing? People make judgments about physical looks based on a voice. And it's amazing that the entire group agreed that he must be tall. Yeah. But he's not very honest. You know, he doesn't sound very honest. Okay. Not so smart. Okay. No, no, no, he's not a leader. No way. He's a tall guy, but very sociable. Yeah. Sounds like a nice guy. Yeah. No, he doesn't go to church. Nope, nope, nope. I don't think so. And yeah, he's quite self-confident. And he's a courageous fella. You know, that, as Europa means, he's not afraid. He's fearless. And definitely you can, you know, yeah. Occupation. What do you think he does? Ah, this guy's a hustler. Okay. He's probably, he probably sells scrap metal. That's one of them says. One of the, one of the respondents. But the Kenyan Swahili guy, first of all, is seen as a bit older. He's about 25. Don't ask me how they judge the age. All right. And don't ask me how they decided the English speaker is about 28. Okay. But the Swahili speaker, Kenyan Swahili speaker, the first comment is, ah, he does, he's Swahili is not very good. So there's an implicit recognition that Kenyan Swahili is different and is not great. You know, unless you're speaking, you know, in a, you know, in a formal situation. But in formal setting, Kenyans will speak in a different way. Not to say that they can't, you know, they don't know how the standard language. He's tall. You know, he's quite honest, intelligent. Yeah. He's a leader. Yes. Why? Most leaders they see on TV and MPs and ministers, they speak it as Swahili or English. So yes, if you speak Swahili or English, you can be a leader. That's what I'm extrapolating. Remember the conclusions that I will read out to you here. They are not really what they said. It's about what I extrapolate from their discussion. And that's important to remember. So I would ask the respondents to clarify an answer. Did you mean he sounds tough? All right, for example. And also engage in brief discussions about a particular point or response. I also paid attention to non-verbal language, including hand gestures and facial expressions. There was a lot of snickering, a lot of laughing, a lot of, you know, the kids, you know, they're saying to each other, you know, laughing, you know, sort of hiding, you know. All those things were meaningful because they would give some idea of what's going on. Even if they give different responses, I, you know, I sort of try to judge what would be which. So what do the young females think about these guises? The first thing I found was Shanks speakers are understood to be young on average 19 to 20. They are sociable and self-confident, but they're not very religiously inclined. They are not seen as honest or intelligent and have only secondary school level of education. Shanks speaking man is courageous, and he's a hustler as well. Kenyan, Swahili, and standard Swahili speakers are evaluated as of the same age, slightly higher than the Shanks speaker, but they are both honest, but the Kenyan Swahili speaker is not as intelligent as the standard Swahili speaker, according to their responses. So again, these are interesting judgments. But perhaps the most interesting observation is that teenagers found the standard Swahili speaker very eloquent for having a sweet language. While the Kenyan Swahili speaker is viewed as not fluent in the language itself, that's making a clear line of understanding between native and fluent coastal speaker of grammatically correct Swahili with all the relevant intonations and the broken Swahili that typifies Kenyan Swahili. So very easy to recognize that. Finally, the Kenyan Swahili speaker has leadership abilities. Again, not surprising. Our leaders mainly converse with us in Kenyan Swahili. So it's easy to make that association that there's a leadership. Right. Now, another group here, let's see. The other interesting thing was that the young female teenagers evaluate all vernacular Swahili speech samples as unsophisticated. What is that? Yes, here. This group here was very keen to indicate. In every case, they just listened to that and they would start laughing. Ah, you know, Mshamba. Mshamba in Swahili means, you know, you know what it means. It means up-country banking somebody who is unsophisticated, you know, from out there in the bush. Each one of them without, you know, without favoring any of these. So there is clearly the vernacular Swahili is associated with the rural area, with the countryside. Not unsurprising. Right. None of them is seen to have more than secondary school education and most are primary school leavers. Right. So again, the level of education is captured by the way the person speaks. He can't have gone beyond secondary primary school if they are speaking that Kenyan Swahili. Interesting that stereotypically the Maasai, the Lu and Somali are imagined as tall. Very interesting because once the Somali guys came on, for example, or the Maasai one, they would just say, oh, that guy must be tall. Well, as I said earlier, stereotypes about, you know, and it's true, you know, many Maasai and many Somali people are tall, but obviously you shouldn't be able to tell that just by listening to a voice. So we are talking of stereotypes that are transmitted through language. And Gikuyu and Kanba speakers are understood to be short. Such me, okay. Kanba, Luya and Somali understood to be honest, but there is a hesitation and divided opinion about Gikuyu and Lu speakers. Overall, none of the speakers are seen to be outrightly intelligent, but there is no doubt about their sociability and self-confidence, which are seen to be very high. There is, again, about character that's very interesting stuff, like the Lu speaker likes to fight. You know, some respondents were saying that, he likes to fight. The Gikuyu one is funny, whatever that means, okay. The occupations are also illuminating about the socio-economic division of labor roles and the trades as understood by Nairobi teenagers. So it's very, very clear that the Gikuyu speaker, whether he's either unemployed or, you know, pushes some cockatani, you know, sells things on the street. He's a street vendor, basically, according to the stereotype of the voice, okay. The Lu speaker is meant to say, oh, he must be a border-border, you know, border-border is a motorcycle taxi. So just listen to the voice say, oh, that guy must be a border-border taxi driver. The Kanba is a shoemaker. So just by listening to the voice, the speaker will say, oh, yeah, most likely a shoemaker. So again, we are looking at stereotypes associated with different trades. And of course, we know there's an ethnic division of labor of sorts for historical and other cultural reasons, perhaps. The Somali man is straightforwardly imagined to sell either in a clothes shop, right, or sells camels, right, or Mira. Mira is the cat, which is very popular with many people in Kenya and Somali in particular. But he may also own a hotel, which in Kenyan English means actually a cafe. If you go to Kenya, someone invites you to a hotel. You're going to a cafe to eat some food, yeah. That's Kenyan English for you, okay. Right, and so forth. So now I know time is running, but a couple more slides. We are now older people. These are university, some of these were from the university, University of Nairobi, others are from Kenyatta University, others I pulled up on the street. And so this was also interesting that a couple of interesting differences is that first of all, these grown-ups, they actually filled out the forms themselves, not as I did with the kids. So we sat in a group, but they would listen and I asked them to fill out the questionnaire. So that's a bit different because in a sense, it's more accurate in a sense because they can actually say more, they can expand on little things here and there. Ideally, this experiment should be done in a laboratory. With headphones, every individual with a sheet of paper, you listen to the ten voices and you make your remarks. I did not have a laboratory. I did not know how to take people out to a laboratory. So this is my improvisation in this particular case. So it's experiment that says. But interesting, therefore we obtained a more personalized profile of responses. Interestingly, two of the students at the university that were in one of the groups, one of them was from Zambia and spoke Bemba. The other one was from Nigeria and spoke Hausa. But they still had very interesting opinions about it. They are studying there, so they've been living there for a few years. And so they have already internalized those things. Five minutes, thank you. So most of them labeled the Shang voice as that of a Matatu town. You know, the public mini buses which are sort of rough sometimes. Some of them, the young men are associated with a lot of things. And so most of them were associated Shang with that language. But again, just like the others, they indexed these vernacular varieties of Swahili to various trades. Not any professional, not professionals. The vernacular Swahili were all indexed to small petty trade. Not university work, not accountants, not a creative, not an office job guy, not a cyber cafe owner or operator. No, farmer, and so forth. On average, the age they started was 25 for Shang, higher than the teenagers. But Shang was negatively evaluated in general except that all of them agreed about it having high self-esteem. Again, the Shang is not a very good leader. He's arrogant, he's unreliable, but frank and opinionated. And so forth. Kenyans and Swahili speakers are also positively evaluated except that they're not seen as very great leaders. Now these opinions given by fairly mature adults working on either undergraduate or postgraduate degrees offered an interesting repertoire, a wider repertoire of descriptive words. Obviously they have more vocabulary, so they were able to use words like rugged or shaggy or arrogant or free spirit, words that are more descriptive. Right, and interestingly, some of them, at least one responded, or pinned that the voice on the tip of the Shang speaker must be a very dark-skinned person. Ask me how? How do you tell somebody's skin color by an African, that is, by the voice they speak? So it's quite revealing of very, very many things. In the interest of time, let me just draw some conclusions of observations. That most Nairobians have a positive attitude towards Swahili. That Swahili is widely accepted in many registers, but the same cannot be said of any of the other Kenyan languages of Kenya. But also many see Swahili as difficult. This is a textbook Swahili that is shoved down their throats in a classroom, which is in the main national newspapers, which and so forth. And usually people just want to have an easier time in life, isn't it? They just want to speak, you know, as easily as possible, not as if they're in a classroom or in a lecture like this one. Many people also believe that the best, or even the only way of expressing nationalism, is through Swahili. Nobody says that we should speak English to become more Kenyan, or that we should speak other languages to become more Kenyan. Swahili still is seen as that. But of course, English is still seen as the best language for upon mobility, and in some Nairobians have attitudes and feelings, even beliefs about their own languages and those spoken by others. For example, Dolbo sounds nice or sweet. These are quite commonly, I came across this. Luo is one of the Nailotic languages of western part of Kenya. And many also say, oh, Somali is difficult, okay, and so forth. Now these, I think about attitudes, you may ask yourself, what is it all about? Well, we can tell many things. First of all, reveal the motivations, for example, of learning L1 or L2. One of the most amazing things I find in Kenya is that hardly anybody consciously goes ahead to learn another Kenyan language. They might acquire it along the way, because they live in certain areas also, but very few people. And so, why is that so? Is it because of these attitudes? Is it because nobody, somebody doesn't want to sound like a peasant, doesn't want to sound like, you know, whatever attitudes are being revealed? So they can help us understand this and inform policy. The status of language and the status of the speakers, you know, so we can now talk about a high variety of Kenyan Swahili and a low variety of it, which is the ordinary one spoken on the street, and the difference can be quite large now. And so we have to question the standard itself. What are we doing by giving news, reading news on the radio and television in a language that maybe the gap is, you know, so big that maybe the ordinary person actually cannot follow what you're saying in a standard national language? These I have seen happen, at least from literature. I understand this happens also in many Arabic-speaking countries. We can see questions of language shift, if it's happening, not the scientist's particular one, but also the question of loyalty to languages, the effective filter. And what do people then feel? This particular example, I was intrigued because there was one girl I remember very much who, when I played one of the voices, which was the voice of her first language, Luya, I didn't know that at the first. But the others just burst out laughing and looking at her. And almost, you know, sort of, she didn't look upset, but she was also smiling back and also, so I dug up a bit more. I said, why? What's so funny? Oh, she's Luya. So I wasn't too sure how to interpret that. Is it a good thing or a bad thing? How does this person, individual, feel being pointed at? Why are people laughing? By the way, she was also very harsh in describing the Swahili accent with her own mother tongue. And so it's very interesting what these kind of attitudes and feelings do to the speakers themselves, especially the minority language speakers. How do they feel being singled out? And of course these kind of attitudes can help us avoid the law of unintended consequences because a lot of resources are wasted and no change is a result. I've just been studying a similar issue in the case of Morocco. And sure enough, after 50 years of pursuing a policy of Arabization, they have reached a conclusion that that's not the right thing to do, that actually they are alienating a very large group of people, that the two have not quite met, and that English and French maybe offer even more possibilities. So that's 50 years down the drain. Now, in the case of Kenya, Swahili has been compulsory since 1984 in secondary schools. How much have we achieved on that? That's a question I would like to know. If we still have these kind of attitudes towards Swahili and to the vernacular, what is the meaning of a policy, an expensive one too, if it's not producing results or the desired results? The results maybe were not anticipated. One of the reasons why a lot of younger generation people, the millennials in particular, they are neither into English, high English or high Swahili, rather they want Shang, which sort of comes in between the two. And this is why such a magazine or newspaper becomes a lot more successful than the Swahili only newspaper. And that's why comics such as this one, very good creative stories, very good graphics, but guess what, it's all in standard Swahili. And guess what, it didn't last a year in the market. Today, there are lots of Shang magazines and they're doing quite well. So you can see what I mean, that we may be actually focusing on one thing, but actually we should be focusing on the other. Also, attempts to eradicate or to correct the youth, the young people when they're speaking bad Swahili, they backfire, so forth. Okay. That's right. I think the very final point is simply that the Kenyan Swahili raises questions about the diachronic effect on the official languages, on English and Swahili. I should say here the Shang question, right? What is it doing to Kenyan English and what is it doing to Kenyan Swahili? A lot of interesting things because if you look at the structure, there are really many other things. Kenyan varieties of Swahili and English are gaining acceptability. Nobody is saying, no, let's not talk that. They may evaluate it a bit low, but they are accepted. Now, in my view, understanding and being aware of variation can help us to teach and learn the standard variety better. When I talk to the teachers and they are all up in arms, no, we don't want to hear that thing. And I say, look, can you use a text like the one I showed you here to help the students see this is wrong, this is wrong, this is wrong and this is how you do it. In my view, this can be done and this will be done elsewhere. Finally, maybe all these little changes that may be seen in Shang in particular can help us sort of predict the future of where it's going. So, standard Swahili in Kenya, how real is it? I don't know. Thank you very much. Thank you. Thank you very much, Shigen. Very inspiring to have many little questions, but I keep the phone to you guys. Sorry, I had to rush a little bit because it was a bit late. We have 15 minutes here. I guess one of the reasons why you use recorded voices is that you kind of take yourself out of the experiment a bit, right? Say it again, Swahili. So, one of the reasons why you would use recorded voices is that you take yourself out of the experiment. I'm wondering whether you still have an idea of how they judge you and your use of language and whether you have any differences between the university students and the students at school and so on. Okay. You mean myself as an individual Kenyan speaker or as the researcher? Well, I mean, you were both. Yeah, exactly. That's why I mean that. Yeah, because when I'm in Kenya, I wear two hats, right? Yeah, so I'm an ordinary person on the street and I can flip around between all those ways of speaking with great ease. Obviously some of the new ones. I'm not very much into it. But I can interact with young Shen speakers with a fair amount of ease. Swahili is the same thing and English is so. So, I think I don't think they looked very much at me as in terms of what is my position in that scheme of things. Of course, they would see me as slightly, I'm always a little bit an outsider of course in that particular group because obviously they can see I'm driving a car I'm not really living around with them. I speak good English, I'm a university lecturer so I'm not exactly on the path. But I don't think there was any negative attitudes towards my presence there. I don't know if that's what you mean. Yeah, I mean that's fun but also what do you think could have influenced how I see? It's very difficult. The only way I can avoid that is by not doing the research. But it's a tough one and I think you're referring to maybe I skipped that but they observe as paradox. The more you want to observe language the less people behave normally. So, what do you do? We use all different kinds of techniques and one of them is love's work, which I use quite a lot, the danger of death. Have you heard of this thing? So, if you want to get the vernacular the most natural speech, excite people, make them either angry afraid, no, no, no. And one of the things, the best things I've done it with my research is to ask them can you describe the day you nearly died? You're a car accident, sorry to say but they're quite a lot especially in Nairobi. And I got these real stories of hold-ups and gangsters in and car accidents and definitely the natural voice now comes out. The breathy parts of it, the excitement the intonation changes, the real shame comes through because they need to explain to themselves. So, we use these kind of techniques. Other things, again in your field methods you may have learned that when you do a recording ignore the first five minutes. Don't pay attention too much. Because the people are still stiff, they're still they have not relaxed. So, don't pay too much attention to that. Break it down so that eventually they can bring out the vernacular because linguists, we are always looking for the vernacular which means the language that one speaks when they are most relaxed or their most natural self. If you research if you want to research on English language, you don't come to university classrooms or something like that. You go somewhere else. So, there are different ways to mitigate that because the Observers' Paradox is always there. Yeah, yeah, yeah. You have a question? The people that were identified as Kikui? Yes, my name is too obvious and I don't hide it. Because Chega is a very common Kikui name. We go about it by using the Christian names instead and it's very common, very unfortunate because one of the reasons of doing this work and I'm very interested in this is because Kenya is very ethnically divided. And as you know we've been having political problems based on ethnicity. And so it's always something that you carry around with you and you can't really drop it. As I said, you can try to use the Christian names so that people don't readily identify. But as you have seen it's not just the voice, even the look people make judgments about it. That's how a bit not very good. But yes they do. Do you plan on playing the same recordings to speakers with different linguistic background like of the minority languages or maybe of the Swahili suburban in the neighbouring countries? Or would that be ethically difficult? Ethically, I'm not sure if it's how do you see it. As you said, there are many advice would that be too much looking for standard types to which forcing it. Yes, because some people might react actually negatively. They say oh you're making fun of our language. I always anticipated that. But happily, with about 152 people I never got that this is not in an overt way. But I sort of always kept half expecting to be challenged. And I can see more or less some people say this is stereotyping. But that's not my aim of course. Now about expanding it further the main reason I may not do that immediately or in the near future is because this study focuses on Nairobi actually. Because Kenya is quite big and diverse I would have a bit of time going. So I'm looking at Nairobi because it's a microcosm of the country and all speakers are found there. As you can see it's actually somewhat advantageous in the sense that I'm actually including what are nationally regarded as minorities Somali and Masai. But in the city of Nairobi they actually are not minority. So actually it works out a bit that way. But for now I'm sticking to Nairobi and maybe the surrounding areas. Hi that was really interesting. I'm doing something very very similar about India actually. I'm interested in what you say at the end about how Kenya varieties of English are giving them acceptability. If I am not wrong you have just the one English variety in there. And that was a very standardized English variety because it came from the other newspaper that you talked about. Well it might be our translation of it. Right. But did you consider at all including as electoral varieties of English? Because obviously when the English that you used if you were to see that on paper there's not really anything within that that could identify that as a Kenyan speaker. Oh no but they're not looking at it on paper. No no exactly. What I'm saying is did you consider or would you consider perhaps including another variety of English which was perhaps maybe less intelligent to somebody who was not familiar with Kenyan English for example or something a bit more vernacularized English. Yeah I mean the closest I would go there is vernacularize the English using one of the Kenyan background because there isn't really in Kenya it's hard to talk of a best or acrolecto. Well you can talk of the acrolecto variety of English. But because English in Kenya does not really exist on a continuum as let's say in the Caribbean. It's really one or the other. And the Kenyan English is only the difference is that they are influenced by different mother tongues so the pronunciations and so forth will come through in that way. So the furthest I would go in that direction which is interesting is maybe another way of speaking this one that I got was what myself and the professional artist thought is the regular round of the mill kind of Kenyan way you know it would be quite vague but basically someone who has done some secondary school that's our definition. Someone who has done up until GCSE can speak English competently in Kenya and that's the standard more or less that we're using here. But it would be interesting to see now you know in Shang you there's a lot of transatlantic influence particularly from the global black culture you know of hip hop and reggae and so forth. They might recognize a bit of that but most of them don't understand. I don't think they would understand any kind of say but select of Caribbean dialect of English or American English for that matter. Another question? Yeah. That was very interesting I know I came here before this talk was also because I tried this test for my PhD when I did my field work. I failed twice so I couldn't use it and I was wondering because you also mentioned some people the speakers were actually the same person did anyone come to you to ask and how did you answer if they came to you? To ask me about if they were the same person? Not directly. No. But I do recall at least one or two instances two instances actually where one of the kids says you're not the same guy? Something like that and so I didn't press it I sort of stayed away from it and then I think some of the other grownups I interviewed they had an idea of that but I sort of used my research privilege to sort of keep it because I know if we now got into discussing oh that's not real that's not real now the other alternative I had by the way which I considered very carefully was to record natural speech samples from the television or from the radio okay and there are plenty and I kept feeling tempted when they're doing interviews on the TV I could just get a recording and say what do you think that is all about the problem with that is that for this test to work you need the same text yeah and so now if you get this recording of this person saying this obviously you can't get a similar one in another language somewhere else so but I don't know because all the studies I have looked at they all must use the same voice in your study and this was my parent question you asked the same guy, the journalist to record the same text these 10 different ways of speaking don't you think this is a limitation it is a limitation in a way the stereotypes that the hearers heard may have been conveyed first by the speaker I don't know could be, possibly if you ask him to talk with, to talk shame for instance or to talk with the rule or accent you must have an idea of how it is to speak with the rule or accent in his own way and that's the whole point that he like most Kenyans know how a rule is accented so he sounds not imitating I was not really looking at whether it's a good imitation of this language or that I'm looking at the views behind that so that would still not damage the experiment it's not the language that I'm interested in it's the views behind that and so whether you realize it's a fake voice or the same but it doesn't really matter the idea is that you understand that the speaker any other question did you play the different languages in the same order each time or did you swap which one they heard first? good question, with the groups I swapped the order precisely so that to avoid the monotony as well and also to avoid being repetitious with the same although the groups were isolated I would have them in one class in another private room only the five of them rearranging them was partly also for me I was getting very listening to the same thing with five groups of 20 people so that was the only main reason I don't know what you had in mind so if you say people who speak under this name and the first one you hear is Masai then you're not going to understand about the rent situation do you think they would have enough knowledge to understand what the meaning behind the voice is if they were the speaker of that variety mm-hmm as I said and one of the things I noticed all the time was whenever one of the respondents identified their language in that they would start smiling even the speaker themselves would recognize that yeah the Somali kid heard the Somali excited voice and one interesting comment one of them made was oh thank you Dr. you for having us there that's very interesting because as I said before the smaller minority groups I think they used to be ignored but this time they are in the experiment and that was what worked out very well I'm not sure if I answered your question is that for you the Masai accent or does it have Masai words it's just Swahili spoken as if the guy is Masai so everyone would understand oh yeah everybody understands the same it's the same text the one I showed you here repeated 10 different ways by this same speaker the lack of other variables in this experiment other than the voice is very important because that's what makes the experiment you know effective if you change it off everything else then it might be my fact some people have tried for example using different speakers but then now there's a danger of people reacting to the voice the particular individual voice rather than the content of what they're saying so you're still advisable to use the same at least the ones that have seen the same voices yeah well unless there's a burning question you know what comes in the hand okay and it's time for Steve Grigan thank you very much for coming and this part of an ongoing project so if you have any feedback in addition to those questions I'll be quite happy to know