Why we use emoji: Written gestures in online writing

When we talk to each other, we don’t just rely on words. Emotion is embodied, and our expressions, our body language, our tone of voice are all used to convey our feelings and affect how our words are interpreted. But for online written communication, we can’t rely on these details. As discussed in the previous post, punctuation can be helpful to represent tone of voice, but often there is still something missing. In the fifth chapter of her pop linguistics book Because Internet, Gretchen McCulloch explores how emoji became popular as a way of replicating gestures in online communication.

Emoji cannot be considered a language: there is a limit to what can be expressed, and most languages can handle meta-level vocabulary about language, which emoji cannot. But they clearly do something. However, many popular emoji use hand and facial gestures, which, McCulloch says, inspired her to begin treating them as gesture.

There are two types of gesture which emoji can represent: the first are called emblems. These are nameable gestures, and have precise forms and stable meanings, and are often culturally specific, such as winking, giving a thumbs up, and obscene hand gestures. Many of these have directly equivalent emoji, for example, fingers crossed 🤞, rolling eyes 🙄, or a peace sign ✌️. Some emoji are more metaphorical, such as the eggplant emoji as a phallic symbol, but, with knowledge of internet norms, they still have fixed meanings. Emoji are not the only way to express emblems online: reaction gifs and images are also used to express specific moods or actions, many of which we can refer to by name (for example, most internet-literate people will know what I mean by Michael Jackson Eating Popcorn.gif).

The second type of gesture with corresponding emoji are illustrative or co-speech gestures. These gestures are dependent on surrounding speech, and highlight or reinforce the topic. You often make these without realising, and at times when they make little sense, such as waving your hands around when on the phone and your conversational partner can’t see you. These gestures don’t have specific names but can be described. Think of the way you move you your hands when giving somebody directions or describing the size of something. These gestures are also represented in emoji. The example McCulloch uses is the range of emojis possible in a ‘Happy Birthday’ message, perhaps a combination of the following 🎂🍰🎁🎊🎉🎈🥳. In these contexts, the order doesn’t matter, these emoji aren’t telling a story, they are adding to the current one. Illustrative emoji are also more likely to be taken at face value, and don’t necessarily require knowledge of internet culture that, for example the eggplant emoji might require. If emblems are for the benefit of the listener, then illustrative gesture are for the benefit of the speaker, used to help them get their message across.

McCulloch also examines common sequences of emoji, finding that, unlike words, emoji are often repeated, both as a straightforward sequence of the same emoji multiple times (the most common being 😂), and sequences of different emoji that are linked thematically, such as the series of birthday related emoji above, or a series of love emoji such as 💕💓😍💗🥰💖. This is another reason why emoji can be considered gesture: repetition does not generally occur in our words, but does occur in hand gestures.

Repetitive gestures are known as beat gestures: they are rhythmic, and if you stutter while you speak, your gestures also do the same. Emoji also do this: we type 👍👍👍 to represent a sustained or repeated thumbs up gesture in real life. We can even repeat emoji which don’t have a literal gesture attached, because, as a whole, emoji can be repeated. The ‘clap back’ is a common beat gesture among African American women, and this is often represented through emoji as a form of emphasis: 👏 WHAT 👏 ARE 👏 YOU 👏 DOING 👏

Emoji serve an important purpose in informal written communication, filling in for expression and gesture which otherwise are hard to convey. For more from McCulloch on the topic of emoji and gesture, Episode 34 of her podcast Lingthusiasm with Lauren Gawne, discusses the content in this chapter, and provides several further links on the topic of emoji and gesture.

------------------------------------------------------------

McCulloch, Gretchen. 2019. Because Internet: Understanding the New Rules of Language. New York: Riverhead Books.

This summary was written by Rhona Graham

Accent Bias: Responses to Accent Labels

Continuing our series of posts related to the 'Accent Bias in Britain' project, in this blog post we discuss some findings from our research which investigated current attitudes to accents in Britain.

In the first part of our study, we replicated Coupland & Bishop's study (2007, summarised in an earlier blog post) to see whether the accent attitudes that people held 12 years ago still persist today. A similar study was conducted by Giles in 1970, giving us a further time point to compare our results.

We recruited a sample of over 800 participants aged between 18 and 79 via a market research firm. The group of participants was intended to be a representative sample of the UK population, so was balanced for gender and region (England, Scotland, Wales, and Northern Ireland) and included all major ethnicity groups.

Once participants had been recruited, they were asked to respond to 38 British accent 'labels', such as 'Estuary English', 'Received Pronunciation', 'Multicultural British English', and 'Birmingham English'. You can listen to some of these accents here. The participants were asked to rate each accent label on a scale of 1-7 - where 1 is the lowest and 7 is the highest - for the prestige and pleasantness of the accent.

After they had completed the survey, we collected social information about the participant, including their gender, ethnicity, age, region of origin, highest level of education, occupation, English accent, languages spoken. We also asked them to complete a short questionnaire about their exposure to different UK accents, the diversity of their own social networks, their beliefs about bias in Britain, and respond to a series of questions designed to measure how much they were concerned about being perceived as prejudiced.

As the image above shows, when compared with Giles' results in 1969, Coupland and Bishop's results in 2004, our findings (2019) demonstrate that whilst there are some minor differences, overall, attitudes to accents in the UK remain fairly stable. Standard accents, such as Received Pronunciation (RP) remain very highly rated, whereas ethnic and urban accents, such as Birmingham English, are rated much less favourably. These findings appear to be stable across the three time points.

Want to replicate this study? 

We've developed a series of Language Investigations and Teaching Units that helps students and teachers develop a research project of their own! Head over to Teach Real English! to access these resources.  

However, all is not lost it seems. Although we see similar patterns across the three studies, we do see a gradual improvement in the ratings of the accents that are rated the lowest (Afro-Caribbean, Liverpool, Indian, Birmingham). In fact, our 2019 study reports quite the improvement in overall ratings of these accents. It's therefore possible that people view these accents much more positively than they did 50 years ago.

However, this study examines only responses to 'accent labels'. What would we find if we played actual audio recordings of these accents to participants? Would we see the same results? In the next blog post, we introduce the findings from the second part of our study. In the meantime, you can find our more about the project by visiting the project website. 

This summary was written by Christian Ilbury

n 

There ain’t nowt wrong with accents

What do you think of when you hear someone speak with a Brummie accent? How about when somebody speaks with a West Country accent? Do you think that some accents are more attractive or prestigious than others? If so, it’s possible that your judgements of these accents are influenced by accent bias.

As part of the Accent Bias project led by academics at Queen Mary University of London and the University of York, over the next few weeks we’ll be uploading a series of Digest posts that discuss the effects of accent bias.

In the first post of the series, we focus on a 2007 study by Nikolas Coupland and Hywel Bishop that investigated how people perceive different types of British accents, looking specifically at whether some accents were evaluated more positively than others.

Credit: https://www.voices.com/blog/quiz-british-slang-uk-colloquilisms/

In their 2007 study, they report on a BBC survey that collected 5010 respondents’ evaluations of 34 different accents. To assess these evaluations, they created an online survey where participants where asked a series of questions about the prestige and pleasantness of the 34 accents. This included asking participants direct questions such as “How much prestige do you think is associated with this accent?”, and “How pleasant do you think this accent sounds?”. The participants rated their judgements electronically via a digital survey by clicking a seven-point rating scale, where 1 is low rating whilst 7 is high rating. This is what is referred to as a 'label study' in that participants were not asked to listen to a recording of the accent, but were simply asked to respond to different accent 'labels', such as 'Asian English' or 'Southern Irish'.

Participants also were asked to indicate where in the UK they were from, how old they were, and their gender. The researchers also asked a series of questions about whether the respondent liked hearing different accents to test whether their attitudes towards accents and dialects influenced their ratings of the different accent labels. 

Coupland and Bishop find that, for social attractiveness, accents such as Standard English, Southern Irish, and Scottish are generally positively evaluated, whilst accents such as Birmingham, South African, and Glasgow were typically down-rated – that is, they score much lower. For prestige, however, they observe a slightly different pattern. Received Pronunciation (or the ‘Queen’s English’) scores much higher in terms of prestige than it does for social attractiveness. Whilst accents such as Birmingham and Asian English score poorly across the two different scales.

Interestingly, accents such as Southern Irish English, Newcastle English and Afro-Caribbean English are rated far higher for attractiveness than for prestige, whereas London English, North American-accented English, South African-accented English and German-accented English are all ranked higher for prestige than for attractiveness.

Whilst these ratings reveal more general trends of the social evaluation of different UK accents, Coupland and Bishop suggest that these evaluations may be influenced by the respondents' social characteristics, such as whether they are male or female. Focusing just on ‘prestige’, on the whole, Coupland and Bishop find that women are more likely to evaluate a given accent as prestigious than men. They also find that where the respondent is based in the UK appears to play a role in their evaluation of a given accent, with participants more likely to evaluate in-group accents as more favourable than others. In other words, Scottish speaking participants were more likely to evaluate Scottish accents more positively than respondents from other parts of the country. Similarly, they also observe that the respondents’ age is likely to influence their evaluation, with the oldest age group tending to show a preference for their own accents than all other groups. Lastly, they observe that the more liberally-minded respondents who indicate that they appreciate accent variation were more likely to rate non-standard accents as more prestigious than their peers. 

So, what does this all mean? Well, Coupland and Bishop note several implications of this study. The first is that language use is influenced by ideology – that is a widespread system of ideas and values that governs a particular concept or social issue. For instance, they observe that there is a general tendency to rate ethnically linked accents (Asian and Afro-Caribbean) and some of the urban vernaculars (Birmingham, Liverpool, Glasgow) as lower in prestige and attractiveness than non-ethnically linked and rural accents. They argue that this is because there is a widespread belief that people should ‘speak properly’ and so accents that are further away from more ‘standard’ varieties are typically perceived to be less attractive and less prestigious than ones that are closer to the standard.

Whilst these findings may not at first seem very encouraging for speakers of non-standard varieties, Coupland and Bishop suggest it seems that there is seems to be a shift towards embracing more liberal attitudes towards accent variation, with younger respondents and those claiming that they like hearing different accents more likely to evaluate non-standard varieties as more prestigious and more socially attractive. So, it seems that although some people might think of certain accents as more attractive or prestigious than others, perceptions are gradually changing. 

Given the fourteen or so years since Coupland and Bishop conducted their study, it’s worth considering whether this liberal outlook has continued. Over the next couple of weeks, we’ll be focusing on the Accent Bias in Britain project which, among other questions, sought to investigate this issue. In the meantime, or more information on the project, you can visit the Accent Bias in Britain homepage. You can also find further educational resources, including a Language Investigation and a Teaching Unit on our Teach Real English! website.

------------------------------------------------------------

Coupland, Nikolas & Hywel Bishop (2007) Ideologised values for British accents. Journal of Sociolinguistics, 11 (1):74-93.



This summary was written by Christian Ilbury
https://doi.org/10.1111/j.1467-9841.2007.00311.x

You are what you Tweet!

In the time that it takes you to read this article, millions of users will have sent a Snapchat, uploaded an Insta Story and updated their Twitter profile. The age of digital culture is very much upon us. For Linguists, the contemporary networked society offers a way to explore language use beyond the traditional method of recording and interviewing speakers. This includes those studies which examine the dialectal distribution of words and features across different parts of the country. One such paper is Grieve and colleagues’ recent Twitter-based analysis of lexical variation in British English.

Traditionally, linguists interested in researching dialectal variation (i.e., linguistic features specific to a particular geographic region or group) have set about researching this topic by conducting surveys and interviews with speakers of a particular variety. For instance, a linguist might ask someone to name the “a narrow passageway between or behind buildings”. If you’re from the south, you might say ‘alleyway’ but northern speakers might call it a ‘snicket’ or a ‘ginnel’.

With the advent of social media, however, linguists no longer have to elicit these words directly. Rather, they can extract massive datasets of social media data to examine where in the country these words are used most.

In their 2019 paper, Grieve and colleagues used a corpus (i.e., dataset) of 180 million Tweets to examine lexical variation in British English. Helpfully, since tweets include what is known as ‘metadata’ that relates to the location in which the tweet was sent, Grieve and colleagues were able to plot these tweets on maps to identify where these words were most frequent. They compared their analysis with the more traditional approach taken in the BBC Voices project.

Their analysis very convincingly shows that the lexical variation observed in the Twitter data mirrors that identified in more traditional analyses! This finding is shown in the graphic below, where for all of the 8 words, the Twitter maps look comparable to those created for the BBC Voices project. For instance, consider the maps for the word ‘bairn’ – a word that means ‘child’ is typically heard in northern UK dialects (second row, right). The BBC Voices project map and the Twitter map are virtually indistinguishable. Across both maps, this word appears largely confined to the north/north-east of the UK – as expected.

Whilst, for the most part, the traditional dialect maps and the Twitter dialect maps look very similar, Grieve and colleagues note some differences. For instance, in the Twitter dataset, ‘bairn’ is observed to account for a maximum of 7.2% instances of the word ‘child’, even in the areas where it is stereotypically associated with that dialect. This is in comparison to the BBC Voices dataset, which reports a maximum of 100% of instances of ‘bairn’ for ‘child’ in some areas. Discussing the reasons for this difference, Grieve and colleagues explore several possibilities. First, they suggest that the differences may be related to a decline in usage of this word. It is possible that 'bairn' has simply become less popular over time. However, the decline in the use of this word also might have something to do with the type of data we get from Twitter and the way it's analysed in large-scale studies such as this. In particular, the authors note that it is impossible to examine the conversational context of the tweet. A such, it’s possible that’s there’s some contexts where users would use ‘child’ for ‘bairn’ even if they use the dialectal term ‘bairn’ in speech. For instance, if a user is reporting someone else’s speech.

Nevertheless, with these issues aside, Grieve and colleagues’ analysis suggests that the findings observed in large-scale dialectal surveys are largely mirrored in the Twitter data. As such, we can expect more and more sociolinguistic research to examine data from social media sites, such as Twitter in the future! So, it seems, you really are what you tweet!

------------------------------------------------------------

Grieve, Jack; Chris Montgomery; Andrea Nini; Akira Murakami & Diansheng Guo (2019) Mapping Lexical Dialect Variation in British English Using Twitter. Frontiers in Artificial Intelligence. 

This summary was written by Christian Ilbury

https://doi.org/10.3389/frai.2019.00011.

"I'm so Fancy"

Remember Iggy Azalea? Well, if you were just about anywhere in 2014, you might recall her smash hit song 'Fancy' featuring Charli XCX. In fact, that song was so popular that it earnt Iggy Ig's a Billboard award for the 'biggest ever hit for a female rapper'. But whilst she might be one of the most recognisable Hip-Hop artists of the current period, you might also recall that she's faced quite a lot of criticism too with many referencing the difference between her ethnicity (as a White Australian) and her distinctive rapping-style which has been referred to as a 'Blaccent' (literally a 'black accent'). 

But, why does Iggy sound 'Black'? And why do people perceive Iggy to have a 'Blaccent'? These are two questions that Maeve Eberhardt  & Kara Freeman decided to investigate in their 2015 paper.

By transcribing Iggy's entire back catalogue of albums, EPs and mixtapes, Eberhardt & Freeman set about analysing her distinctive rapping style. With newspaper articles referring to Iggy's 'Blaccent', the authors examined her use of features typically found in African American English (AAE) in her rap music. As a speech variety, AAE is typically spoken by Black individuals (i.e., African American) speakers who live in parts of Northern America.

One feature that the authors decided to explore is 'copula absence' which describes the tendency for speakers to pronounce the sentence "he's in here" as "he in here" - in other words, the verb 'be' (is/are) is absent from the sentence.

Whilst this feature occurs in many varieties (including some varieties of British English), in AAE, researchers have found certain patterns that seem unique to the variety. In particular, they have found that speakers tend to use higher amounts of copula absence before certain types of words and that this feature is more likely when the verb occurs before gonna as in "she gonna go home" and least likely before noun-phrases "Marie's in there".

Remarkably, by analysing Iggy's rapping style, Eberhardt & Freeman found good evidence to suggest that she wasn't just using copula deletion randomly but, rather, her use of this feature mirrored the same patterns that native AAE speakers exhibit! However, when they analysed Iggy's interviews, they found that she rarely uses copula deletion.

Iggy in an interview - sounds Australian, huh? 

So, why does Iggy use a variety that's typically spoken by Black African Americans in her rap but not in interviews? One such explanation has to do with the music industry and genre that she's working in: Hip-Hop. As an art-form that originated in Black communities in the U.S., many of these artists come from this community and typically those who speak AAE - think of Jay Z or Lil Wayne. As such, the language associated with this genre of music - the 'Hip-Hop Nation Language' (HHNL; Alim, 2004) - is largely based in AAE and shares many features of this variety.

In order to get by and sell records, it seems then that you need to use the 'code' that's typical of the genre and rap in HHNL. But, as a White Australian, Iggy doesn't really look or sound like a Hip-Hop artist... Hip-Hop in an Australian accent doesn't seem to work! Herein lies the explanation for her performance of AAE.

Eberhardt & Freeman argue that she uses AAE to sound like a 'real' Hip-Hop artist in order to sell records. And she does this quite well- as we've seen she uses the same features in the right 'slots' as a native speaker. But, whilst she might be able to speak AAE like a native speaker, it seems that her performance is still pretty problematic. In fact, there are virtually hundreds of articles on Iggy's 'cultural appropriation' of AAE, with many referencing her use of this variety and her lack of authenticity as a White Australian.

So whilst Iggy may be claiming to be a "a white girl with a flow ain't been seen before" it seems that she's not the "realest" after all...

------------------------------------------------------------

Eberhardt, Maeve & Kara Freeman (2015) ‘First things first, I'm the realest’: Linguistic appropriation, white privilege, and the hip‐hop persona of Iggy Azalea. Journal of Sociolinguistics. 19(3):303-327.

This summary was written by Christian Ilbury

https://doi.org/10.1111/josl.12128

Authenticity in the Hood

Whether it’s in friends, designer handbags, or website security, one quality people always want is authenticity. Their fake counterparts are shoddy at best and damaging at worst, whether that be for your emotional health, your belongings, or the contents of your bank account.

The music is real, even if the wool on the collar is not.

In a recent paper, Pia Pichler and Nathanael Williams look specifically at the way identity is authenticated by four young men from South London. Nathanael was one of these four men, and he recorded over five hours of conversation between him and the three others. The conversation covered a range of topics, including class, race, language, fatherhood, and the US; however, the focus of this research specifically covered their discussion of hip-hop.

In order to investigate the links between identity and authenticity, Pichler and Williams draw on Silverstein’s “cultural concepts”, which describe people’s use of linguistic elements that do not have a straightforward interpretation. In order to access the meaning of those elements, you need to be part of a shared cultural sphere with the person who is using them. One example they give is that of a wine connoisseur, who uses certain terms to describe wine which might otherwise mean something different. Using these terms to convey these cultural meanings is therefore a way to indicate your affiliation with that sphere, and make it a part of your identity.

In the conversation that Nathaniel analysed, the four men frequently positioned authentic aspects of hip-hop culture against inauthentic intruders. One example is during a discussion of World Star Hip Hop, a website which features regular content about the genre, from both contributors and users. One of the men, Les, was complaining about some girls fighting on the website, specifically referring to them as “white girls from The Hills”. “The Hills” was a series that focused on the lives of white, upper-middle class women in Los Angeles; by referencing them, Les positions them against the working class and non-white culture of the website. The men also reference cultural concepts within the UK to position things. For example, Les also complains about white kids from Oxford or Cambridge proffering extended opinions on hip-hop. Given the reputation of Oxford and Cambridge as wealthy university towns, this indexes the white person he is complaining about to a middle-class background that is at odds with his supposed knowledge about a predominantly black and working-class genre. This therefore renders the person and their opinions on the best hip-hop artists as inauthentic.

Pichler and Williams also note linguistic features used by the men that are just as important to the construction of their identities. For example, when discussing his brother’s membership in a South London gang, Les says dey and dem as opposed to they and them – a feature known as DH-stopping. He also uses yout and bruv; while the use of such lexical items is not specific to the Englishes typically spoken in hip-hop, they still index an authentic background, as they are features of MLE, or Multicultural London English. As a dialect usually spoken by working class people, often of colour, in inner-city London, it is not at odds with hip-hop culture, which often draws on local dialects.

These are just some of the ways in which the participants authenticated themselves. Now consider the conversations that you have – how do you position things against one another, and what features do you use if and when you do so? You may be doing the exact same thing.

------------------------------------------------------------

Pichler, P., & Williams, N. (2016). Hipsters in the hood: Authenticating indexicalities in young men's hip-hop talk. Language in Society 45(4): 557-581.

This summary was written by Marina Merryweather

Young, old or just ‘emerging’? How does age affect language change?

As times change so too do people’s life stages. Traditionally these were thought of as child > adolescent > adult, but this idea is now called into question with the addition of an ‘emerging adult’  stage before adulthood.  As age is such an important sociolinguistic variable, this is a significant development for anyone interested in studying language variation and change, as Douglas S. Bigham explains in detail.

An emerging adult’ is aged 18 – 25, in higher education, unmarried, moves around a lot and has a large, although not necessarily close, social network. Bigham makes it clear that not all 18-25 year old young adults are ‘emerging adults’.  The ‘emerging’ label is dependent on a particular psychological state, defined by the following factors:  identity exploration, instability, self-focus, feeling in-between and feeling that anything is possible. The first two factors challenge sociolinguistic notions of place and social network whilst the last three challenge ideas about social identity.

For example, traditional models of ‘place’ in sociolinguistics rest on the idea that place is geographic and that geographical and political boundaries (e.g. mountains or rivers) create and maintain linguistic variation.  However, emerging adults move around at an unprecedented rate, rarely remaining in the place where they were born and usually moving several times.  They will also live quite intimately with a variety of people from different social classes, cultures, religions etc (think of university halls of residence).   Ironically, as emerging adults move from community to community, they become more detached from any particular one, so they have no strong community bonds. Identity exploration and instability means that emerging adults’ accents seem to be unrelated to their geographical origins.  Rather than seeing their accent as showing that they come from, say, London, they seem to view accent as another facet of their personality: “just part of who you are, y’know?” 

The rise of virtual online networks has contributed to this sense of not belonging to a place or group, or ‘feeling in-between’.  Emerging adults may have friends from all over the globe but may never actually meet them, only interacting with them virtually. Their social boundaries are therefore blurred, as are their gender and sexual boundaries.  Many US university students replied ambivalently to Bigham’s question about their sexuality, responding “straight, I guess” or “mostly straight”. Emerging adults need to be able to negotiate their language across virtual global communities and seem to ‘self-focus’ on their personal identities rather than forming social ones. 

Another problem for traditional sociolinguistic models is that emerging adults view themselves in classless terms.  Over 95% of Bigham’s emerging adults were ‘very sure’ that they would “get to where they want to be in life” and would have a better life than their parents, regardless of their socioeconomic background.  They felt that ‘anything is possible’.

Emerging adulthood is therefore posing a challenge, as age and class have traditionally been such important variables in sociolinguistic research.  It has always been assumed, for example, that the speech of older people = the speech of an older time period; and that we can study language change by comparing older people’s speech with the speech of the young.  However, emerging adults are a completely new phenomenon, only surfacing in the last two decades.  As this new ‘emerging’ life stage surfaces, so too does a new challenge for sociolinguists.

----------------------------------------

Douglas S. Bigham (2012) Emerging adulthood in sociolinguistics. Language and Linguistics Compass 6 (8): 533-544.

doi. 10.1002/Inc3.350

This summary was written by Gemma Stoyle

A be’er way to analyse t-glottaling?

Most people are familiar with the concept of t-glottaling, that is to say the replacement of /t/ with a glottal stop in a word. Using the symbols of the International Phonetic Alphabet, a word such as better may be represented as either [betə] if pronounced with an audible [t], or [beʔə] if pronounced with a glottal stop. Frequently, the glottal stop pronunciation is also represented as be’er. T-glottaling, once (and still by some) considered a stigmatized feature of speech, is now widespread in Britain and is generally perceived as being stereotypical of British urban speech (although it is by no means confined to urban areas).

The replacement of /t/ with a glottal stop has been the subject of investigation by many sociolinguists in different parts of Britain and these studies have generally agreed that t-glottaling is most likely to occur in certain contexts. Within different social settings, speakers use t-glottaling more in informal styles of speech than when using a more formal style of speech and, generally speaking, you might expect to hear more t-glottaling from speakers lower down the social scale. T-glottaling is also associated more with younger speakers than with older speakers, an indication that this is a feature of language change which is still in progress. In addition to these social constraints on the use of t-glottaling, there are also linguistic contexts which are more likely to trigger its use.  For example, t-glottaling is more likely to occur word-finally, in a word such as bit, than in word-medial position, in a word such as bitter. In word-final position, t-glottaling also follows a pattern in that it occurs most frequently when it is followed by a consonant (put down), less so when it occurs before a pause (stay put) and least of all when it is followed by a vowel (put off).

Researcher Erik Schleef goes a step further in his analysis of t-glottaling. He focuses on t-glottaling among adolescents in London and Edinburgh and finds that although t-glottaling follows the patterns discussed above in both locations, his results show that the Edinburgh teenagers use more t-glottaling overall than the London teenagers. This may come as somewhat of a surprise as t-glottaling is often associated with London speech but, as Schleef points out, the phenomenon actually seems to have had a longer history in Scotland. Schleef also finds differences between the two locations in the teenagers’ use of word-medial /t/ when the /t/ occurs before a vowel (as in the word bitter), which is the context where t-glottaling is least likely to occur according to most studies. In London t-glottaling was most likely to occur before a consonant word-medially but in Edinburgh t-glottaling was just as likely to occur before either vowels or consonants.

Schleef also examines whether the use of t-glottaling is influenced by the grammatical category of words and finds that adjectives and nouns disfavour t-glottaling whereas function words (e.g. prepositions and pronouns), and both progressive and past participle forms of verbs (e.g. waiting, waited) tend to favour glottal replacement. On closer examination, Schleef suggests that progressives and past participles are likely to show high rates of t-glottaling because /t/ appears in word-final position of the root word (in this example, wait) which is a context that attracts high rates of glottal replacement.

Lexical frequency was also investigated and may also trigger the use of t-glottaling. Schleef found that in both locations more frequent words favoured glottal replacement when the /t/ occurred in word-medial position and in London this was also the case for word-final position. In Edinburgh, Schleef also found that words of four or five syllables favoured glottal replacement.

What this research indicates is that the use of glottal replacement may be affected by more than just phonological factors, as has been assumed in many previous studies. It provides a framework for the way forward for future studies of this phenomenon.

___________________________________________________   

Schleef, E. (2013). Glottal replacement of /t/ in two British capitals: Effects of word frequency and morphological compositionality. Language Variation and Change 25: 201-223.

Doi: 10.1017/S095439451000094

This summary was written by Sue Fox

‘Throve’ and ‘dove’ or ‘thrived’ and ‘dived’? Let’s call the whole thing off!

At first, trying to explain the formation of the past tense in English may seem simple – you just add –-ed, don’t you? So that walk becomes walked and help becomes helped, right?  Correct! ... for ‘regular’ verbs.  Unfortunately, there are also many ‘irregular’ verbs like eat (ate) and stand (stood) that do not easily fit into this pattern.  In a logical world, we would expect language to regularize over time and this is true of some verbs – helped was hulpon in Old English. However, some regular past tense forms have actually become irregular – for example, mean > meant.

Lieselotte Anderwald investigated this phenomenon in British and American English.  These two varieties have developed their own national features and peculiarities over time and Anderwald was curious to see if this had happened with irregular past tense forms.  She concentrated on three verbs which have been recorded in both regular and irregular forms.  These are throve vs. thrived, dove vs. dived and plead vs. pled.  She searched a digital database or ‘corpus’ of 400 million American English words called Corpus of Historical American English (COHA) to find occurrences of these irregular past tense forms from as far back as 1810. She also searched the British National Corpus (BNC), containing over 100 million words from British English so that she could compare her findings. Then, she consulted her own Collection of Nineteenth-Century Grammars (CNG), containing 258 British and North American Grammar Books published during the 19^th century, to see if linguists were recommending a particular usage of these verb forms.

Although throve was the main past tense form used during the nineteeth century, it declined rapidly in use from 1910 onwards and seems to have regularized so that American users now only use thrived.  However, British English still uses throve at times.  American English appears to be leading British English by several decades in this regularization process.  In a completely opposite way, the irregular form dove is becoming more irregular in American English.  In fact, dove is used as much as dived in modern American usage today (50% of the time), unlike in British English where it is used just 1% of the time. The irregular form pled also seems to be a new form which has emerged during the twentieth century in American English, although in both varieties it is used very infrequently and mainly in legal contexts.

With this data in mind, Anderwald consulted the CNG.  In the case of thrived vs. throve the American  nineteenth century grammars permitted a lot of variation in usage and started to endorse thrived from the middle of the century, whereas the British Grammars appeared to strongly favour the use of throve. It is hard to know whether these American Grammars were just describing what they observed happening to this verb form or whether their recommendations were in some way influential in the change actually taking place.  Dove was rarely acknowledged as an irregular verb in any of the grammars consulted and dived was the only form accepted.  This is interesting considering how widely used it now is in American English and its lack of acknowledgment in grammars does not seem to have influenced the emergence of this irregular form.  Pled was only mentioned in twelve grammar books, ten of them from America and just two from Britain.  There was a rise in its inclusion in verb tables in  grammars of the 1860s, so that children would have learnt plead-pled-pled by heart.  Looking at her data from COHA, Anderwald noticed that there was a small rise in usage of pled in the 1870s which may have been caused by the generation of 1860 using it in their adult writing: a very small example of prescriptive influence maybe?  If this is so, it is probably because plead was (and is) so infrequent that users needed to consult a grammar book, whereas a more frequently used word is much better entrenched in the memory and therefore perhaps less influenced by grammarians. 

The differences Anderwald found between these changing irregular forms shows how two varieties of the same language can grow, develop and change in different ways. This is what makes language an integral part of a national character and grammarians may only minimally affect this.  Language will develop in its own way and won’t be restrained by rules…in fact sometimes it will even break them – obviously what throve throve to do!

Anderwald, Lieselotte (2013) Natural language change or prescriptive influence? Throve, dove, pled, drug and snuck in 19th-century American English. English World-Wide 34:2 (2013), 146–176.

doi 10.1075/eww.34.2.02

This summary was written by Gemma Stoyle