Friday 25 May 2012

No Way!

no, not me, no way, no how!

It’s well known that negation is more frequent in spoken language than in writing. This is hardly surprising: after all, speech acts such as denials, refusals and rejections are typical of face-to-face interaction. It seems more surprising, though, to learn that in spoken English teenagers use negatives more frequently than adults do.

This is what Ignacio Palacios Martínez discovered in his analysis of the Corpus of London Teenage Language (COLT). He compared the use of negation in the teenage corpus with adult speech, mainly using the Diachronic Corpus of Present-Day Spoken English. Whereas the teenagers used about 33 negative words and structures in every 1000 words, the adults used only about 23 – a difference that was statistically significant.

The reason, Martínez suggests, lies at least in part in the nature of teenage language. About one third of the teenagers’ negatives occurred in imperatives, commands, orders, strong suggestions, directions, instructions and refusals. Martínez argues that this is because teenagers like to be direct, straightforward and spontaneous when they are talking to each other.  The adults, by contrast, tended to use more roundabout expressions to carry out the same kinds of speech act, which resulted in a lower number of negatives overall. The teenagers were just as direct when expressing an opinion, and this also often led to their using negation. Adults were more inclined to hedge their opinions using indefinite or vague expressions.

A further characteristic of the teenagers’ speech was the use of several negatives in a string, which increased the force of the negation. The example in the first box shows a string of this kind.

this geezer from Bedlam yeah got stopped the other day in this car, yeah, he was pissed, he was tripping and he was speeding yeah, no not, no licence; no tax, no ruddy insurance yeah

The second box contains an example of another tendency found in the teenage corpus: this is the use of negatives as a kind of ritualized play, where one speaker immediately contradicts the other.

S: no I never
J: yes you did
S: <laughing> I never
J: he saw your body
S: I never
J: and ever since then, face it S!
S: no shut your mouth! Shut up!

Although both adults and teenagers sometimes used fixed negative expressions to strengthen the negative force of what they were saying, the particular expressions differed. Adults tended to use not at all as a stronger term than no, whereas teenagers preferred no way. Teenagers also used idioms such as I couldn’t give a toss, and negative forms such as nope, nah and dunno. They also used innovative forms like uncool (meaning the opposite of being trendy or fashionable). Negative concord (e.g. I didn’t get no matches) was more frequent in the teenage corpus too, as was nonstandard never (e.g. Vernon never called for me yesterday), though Martinez points out that this may reflect the social class or regional origins of the speakers, which he could not take into account in his analyses.  Social factors such as these may interact with the psychological development of adolescents, which accounts for their tendency to be direct and to experiment and play with language, but all these issues need further investigation.
Martínez, Ignacio M. Palacios (2011) The expression of negation in British Teenagers’ Language: A Preliminary Study. Journal of English Linguistics 39: 4-35.

doi 10.1177/0075424210366905

This summary was written by Jenny Cheshire

Thursday 17 May 2012

Okay, that's my cue

Will computers sound completely human-like in the future?

Words like okay, alright and right pose a particular challenge for automatic speech recognition because they have a wide range of functions in conversation.  As a result, words with such a variety of uses can be ambiguous in their interpretation.  In natural conversation, their meaning can be interpreted through their position in the utterance and, importantly, various auditory and acoustic cues, such as the intonation pattern.  However, in computational systems such as TTS (Text-to-Speech), how can the correct intonation pattern be assigned to a word such as okay if the system cannot decide which function it is serving in the utterance?

Agustin Gravano, Julia Hirschberg and Štefan Beňuš analysed a group of words classified as Affirmative Cue Words (ACWs) for their acoustic and prosodic similarities and differences, to see how these properties could help with computational disambiguation. The functions of ACWs include showing agreement, showing interest and signalling the beginning or end of a topic. Gravano, Hirschberg and Benus found ten potentially different functions of ACWs , though only okay and alright are versatile enough to be used in all ten ways. 

The data they used came from a recorded corpus known as the Columbia Games Corpus which represents 13 Standard American English speakers.  They identified 5,456 ACWs which represented 7.8% of the speech. Of these, the 6 most common were alright, okay, yeah, mm-hm, uh-huh and right.   However, in order to make sure the data wasn’t dominated by one speaker (and therefore skew the results), the data was levelled so that it represented all speakers as equally as possible.

Looking at the position of the words, they found that alright and okay were used in similar positions in the utterance, and that mm-hm and uh-huh also showed similar distribution patterns (frequently when there was a pause either side of the word so that it stood alone), with their primary functions either as backchannels (showing that the listener was following) or to show agreement.  They suggest that this means that the members of each pair can be used interchangeably.

In addition to position and function, they looked at factors such as intonation pattern, intensity, duration, pitch and voice quality.  In doing this, they could identify the acoustic and prosodic features which were more likely to indicate a particular function of a single ACW.

Using this data, the researchers conducted a number of experiments which tested the ability of computational systems to recognise and correctly classify the function of the ACWs in their data.  They noted that their approach allowed for the inclusion of a wide range of available information, which led to greater accuracy in classification.  However, of all the factors incorporated into the experiments, data related to the ACW’s position in the intonational phrase turned out to be the most important factor in disambiguating the function of these words. For example, right was the only ACW that could be used with a checking function (e.g. through its use in tags – “it’s there, right?”) and this was one of the few instances where an ACW could be found at the end of an intonational phrase.

They conclude that, by analysing and incorporating a wide range of acoustic and prosodic characterisations in computational testing and subsequent programming, spoken dialogue systems (such as those which aim to both recognise and produce speech) will be able to improve their performance and take one step closer to emulating human speech.

Gravano, A., Hirschberg, J. and Beňuš, Š. (2012) Affirmative Cue Words in Task-Oriented Dialogue. Computational Linguistics 38:1-39

This summary was written by Jenny Amos

Monday 14 May 2012

How four-year-olds use language to make friends

How do children initiate friendships?

Research studies have shown that friendships are important to children and that they offer them emotional support. Some studies have shown that children who have good quality friendships have higher self-esteem, better peer interactions generally and do better at school. But how do children as young as four years old know how to initiate and maintain friendships? What linguistic strategies do they use? Amanda Bateman has been looking at the way children use pro-terms ‘we’ and ‘us’ in the co-construction of new friendship networks when they first start primary school.

Bateman’s study took place in Wales, where children start school at four years old. The research was conducted approximately three weeks after the children’s first day at primary school. Thirteen four-year-old children (6 girls, 7 boys) were videoed and recorded during their morning playtime, one child per morning over a three-week period. Interactions in which the children used the words ‘we’ and ‘us’ in their social alignments with each other were transcribed and analysed.

In one example, Bateman demonstrates how a four-year-old girl initiates a friendship with a boy firstly using the collective pro-term ‘we’ ‘why don’t we just..’ and then follows with the use of a possessive pronoun to invite him to her party ‘will you come to my party’ successfully achieving her desired affiliation with him as he accepts her offer.

Bateman also shows how once friendships have been initiated, they are protected and maintained, as in the following example:

Emma:                       kerry do you want to play mums and dads
Kerry:                        no we’re playing families
Kathy:                        no

In this interaction Emma approaches the other two children and asks Kerry if she wants to play mums and dads. Kerry rejects the offer and uses the collective pro-term ‘we’ to show that she is already affiliated to Kathy. This establishes Kerry and Kathy as an exclusive friendship, from which Emma is excluded. Kathy’s use of emphatic ‘no’ also serves to protect their exclusive friendship.

The study shows that the pro-terms ‘we’ and ‘us’ help children to verbally affiliate themselves with another child in order to establish a friendship. Equally, though, the same pro-terms can also be used to exclude other children who are not accepted as being part of the friendship group. Interestingly, the study aligns with findings from studies conducted years and countries apart, leaving Bateman to suggest that children worldwide use the same strategies when co-constructing social alignment on an everyday basis. The researcher also concludes by drawing attention to the creativity and social competencies in social organization processes in children as young as four years old.
Bateman, A. (2012). Forging friendships: The use of collective pro-terms by pre-school children. Discourse Studies14 (2): 165-180.

DOI: 10.1177/1461445611433630
This summary was written by Sue Fox

Thursday 10 May 2012

I got oot of the car and I gied inside this tiny, peerie hoose! *

The Shetland Islands provide an interesting site for studying dialect death

What is happening to the Shetland Islands dialect? Is this distinctive dialect dying out along with many other traditional dialects as research on British varieties in recent years has indicated? Researchers Jennifer Smith (University of Glasgow) and Mercedes Durham (University of Aberdeen) conducted a sociolinguistic study in order to test such claims as they relate to the Shetland Islands, focusing specifically on the main town of Lerwick, the commercial and industrial centre of Shetland.

The Shetland Islands are situated in the North Sea, between Norway to the east and Scotland to the south. The Vikings invaded Shetland in the 9th Century and while the dialect is described as a variety of Scots, there are still traces of the Viking language, Norn, in evidence. The dialect has a number of features which are unique to the Shetland Isles but also others which are used more widely throughout Scotland.

Thirty adults were sampled in the study (15 male, 15 female), divided equally into three different age groups - 17-21 years, 45-55 years, and 70+ years old – to represent three generations of speakers. The study used spontaneous speech, elicited during sociolinguistic interviews lasting 1-2 hours conducted in each participant’s home.

The researchers report on six features. The first two are lexical items, peerie (a Shetland-specific word to mean small, tiny, little as in the title above) and ken (a Scotland-wide word to mean know). The second pair of features are grammatical structures: the Shetland-specific use of the verb Be as in ‘they were been coopers as well’ where standard English would have the verb Have i.e. ‘they had been coopers as well’, and secondly, the more Scotland-wide use of yon to mean that, as in the example ‘what’s yon?’ Finally, the researchers consider two phonological features: the Shetland-specific use of [d] to replace the <th> in words such as that, then, those and the use of the more Scotland-wide feature of pronouncing words such as all, ball and call as a’, ba’ and ca’.

The results of the analysis of all six features reported on in this paper showed that there is a steady decline in the use of the local, traditional forms in favour of more standardised forms across the three generations of speakers. The results therefore seem to confirm reports that the local dialect is disappearing in the Shetland Islands. However, there is an interesting twist in the tale. The speakers in the two older generations of speakers all seem to pattern in the same way but the younger speakers show a sharp divide; some of the younger speakers show high rates (in some cases even higher rates than the older generations) in the use of local forms while other young speakers have very high rates of the newer, standard forms. The researchers considered all kinds of reasons why this might be the case, for example gender, networks, time spent away from the island and attitudes towards Lerwick and the Shetland Isles generally but none of these influences seemed to provide the answer as to why there was such a split among the young speakers. These results highlight the fact that a complex array of factors are involved in the process of language change and dialect attrition and the change does not neccesarily follow a regulated or gradual pattern. More and more, sociolinguists are focusing on the use of language by individuals in the community and this study would seem to highlight further need for this kind of close investigation. In fact, the researchers suggest that future research by way of a more in-depth ethnographic study of the Shetland Isles may reveal the reasons for the split among the younger age group. It would also provide close monitoring of a dialect undergoing attrition while it is actually happening.

* The title of this article can be glossed as ‘I got out of the car and I went inside this tiny, little house’
Smith, J. and Durham, M. (2011). A tipping point in dialect obsolescence? Change across the generations in Lerwick, Shetland. Journal of Sociolinguistics 15/2: 197-225.

DOI: 10.1111/j.1467-9841.2011.00479.x

This summary was written by Sue Fox

Friday 4 May 2012

Yerr ‘aving a laarf aintcha?

It can be a noun or an adjective; it’s used throughout the media (usually negatively) to describe a certain type of person or behaviour; and it was Oxford University Press’ ‘word of 2004’.  It is, if you haven’t guessed already, the word chav. The Oxford English Dictionary defines chav as:

a young person of a type characterised by brash and loutish behaviour and the wearing of designer-style clothes (esp. sportswear): usually with connotations of low social status

Indeed, it’s this social stereotype that Joe Bennett investigated to see if there was a correlation with the language that’s known as chavspeak.  One point of particular interest, he notes, is that the identity of ‘chav’ is not something with which people voluntarily align themselves. Instead, chavs are always viewed as ‘other people’.  This in itself suggests that there is a very negative evaluation attached to the term.  So what about chavspeak?

Bennett compared two ‘humour books’ that aimed to describe not only the supposed social aspects and cultural practices of so-called chavs, but also what is thought to be characteristic of chavspeak.  In doing this, he wanted to investigate the links between linguistic forms and social practice – why were certain linguistic forms chosen? And, how do these linguistic forms work to characterise and strengthen the view that chavs are representative of an underclass in society?
At the discourse level, chavspeak is presented in the humour books as demonstrating communicative incompetence and rudeness. Bennett points out that this is enhanced by the type of ‘translation’ provided.  For example:

Giss a fag, mate?   ‘Excuse me, but can you let me have one of your cigarettes please?’

Yerr ‘aving a laarf aintcha?   ‘Are you jesting?’

This is also the case at the lexical level, where regular English words said to be typical of chavspeak are ‘translated’, drawing on themes such as crime and public disorder:

Shoplifting: ‘the favoured hobby, or indeed job, of most Chavs’

The social connotations exploited by the writers of the humour books in these examples serve to underline the lower class status of anyone who attracts the label of ‘chav’.  However, the most linguistically salient exploitation of social connotations can be found in the phonological features said to be characteristic of chavspeak, shown in the two books by the spelling the writers use. Both the books examined by Bennett mention TH fronting (spelling brother as brovva), glottal stops (with aunty  spelt aun’ie), h-dropping (have spelt ‘ave), <n > at the end of words like runnin and monophthongisation of the MOUTH vowel (mouth spelt maaf).  Each of these phonological features has urban and/ or working class connotations and many are (rightly or wrongly) associated with Cockney English by non-linguists. These spellings therefore link with the assertion in one of the books (Chav!) that chavspeak is some sort of Cockney hybrid.  Bennett notes that these phonological features are the ones that British speakers tend to strongly  associate with people from the lower social classes.

Bennett concludes that chavspeak is a social construct that is continually reinforced by those who draw on existing social stereotypes in order to create it.  He cautions that, while those who exploit the chav stereotype in the media do so under the guise of humour and innocent intentions, this only serves to strengthen what is a negative stereotype. This humour could therefore incubate the type of prejudice that exists in relation to other social stereotypes, and can be a resource for the maintenance of social relations of power.
Bennett, Joe (2012) ‘And what comes out may be a kind of screeching’: The stylisation of chavspeak in contemporary Britain’. Journal of Sociolinguistics 16:5-27
doi: 10.1111/j.1467-9841.2011.00521.x
This summary was written by Jenny Amos