Diary of a student sociolinguist

by on April 29, 2015

There is a vast difference between knowing a theory and putting a theory into practice. Perhaps that sounds obvious, but as the poet, John Keats, wrote in a letter: “Nothing ever becomes real till it is experienced.” I spent two years learning sociolinguistic theory. By my third undergraduate year I decided it was time to head out into the field. “Into the field” brings to mind images of jungle explorers venturing out into unknown and uncharted territory, or of intrepid sociolinguists documenting endangered languages. Admittedly, my first fieldwork experience was less exotic, but it was exciting nevertheless!

One of the first hurdles that a sociolinguist encounters is finding people to take part in the study. It’s all very well having a brilliant research proposal, but it’s going nowhere if you don’t have participants! After my research proposal was submitted, I began the time-consuming endeavour of recruiting people to interview: making phone calls, asking people in shops, sending emails, and utilising social media. Sociolinguistics is a relatively new academic field which looks at language in relation to social factors, such as gender, social class, age, and ethnicity. Even so, within the 50 years since sociolinguistics really took off (as illustrated below), there have been immense changes in technology. In the 21st century, it is so much easier to recruit participants for studies and reach a larger number of people, using the Internet and social media networks.

This Google Ngram Viewer shows the frequency of words or phrases over time in a corpus (a large, searchable collection) of books. Note the rise of the word “sociolinguistics”, beginning in the 1960s.

Heading out into the field

For my research, I went back to my hometown, or rather, home village: a small village tucked away in a valley set in an expanse of heather moorland in North Yorkshire, England.

(Wikimedia Commons, 2009)
(Wikimedia Commons, 2009)


The North York Moors is a rural area and there are few towns, but the northern edge of the moors is less than 20 miles from the urban areas of Middlesbrough and Stockton-on-Tees—areas with an industrial past. The research questions that I wanted to investigate were whether urban Middlesbrough is an influence on the speech of residents in the rural locality of the Moors, and also whether linguistic features (i.e. sounds in people’s accents) are becoming more standard. I did this by looking at two linguistic variables:

1) The glottalling of /t/, and

2) The CURE and THOUGHT variables (i.e., /ʊə/ ~ /ɔː/)

Understanding these variables is key to understanding the study, so I’ll do my best to provide an explanation of what they sound like. /t/-glottalling occurs when the t-sound in words like “wateror “city” is pronounced as a glottal stop (marked with the IPA symbol: ʔ)—that means “water” and “city” are pronounced wa’er or ci’y. Try it yourself and you’ll see what I mean!

The CURE and THOUGHT variables are relatively straightforward to understand: try saying “cure” and then say “thought”. The vowels in these words sound different, don’t they? However, some regional accents in Britain use the vowel sound in CURE in words like “poor” and “moor”, whereas in my accent I say those words with the vowel that is in THOUGHT. In phonetic transcription, /kjʊə/ and /pʊə/ show that “cure” and “poor” can sound the same. But there is a change taking place: the /ʊə/ sound in “cure” is undergoing a transformation, so that it sounds like the vowel in “thought”. This is called ‘CURE lowering’, because the sound is quite literally moving lower in your vocal space. So when I talk about the CURE variable (which I will be doing a lot in this article!), I’m referring to the vowel sound in “cure”, as opposed to the vowel sound in “thought” (which I would refer to as the THOUGHT variable).

Now, getting back to the study, I’ll briefly review some previous findings: citing other academic literature is a key part of any research study. Middlesbrough has a high frequency of glottalled /t/ among younger speakers (Llamas, 2007), so if I found an increase of this feature in younger speakers on the Moors, it could potentially be linked to the fact that Middlesbrough is one of the closest towns—and a centre for jobs, recreation, and education—to the Moors. Secondly, the CURE variable is the traditional Moors pronunciation for words like “poor”, but THOUGHT is becoming the standard in Received Pronunciation—the most prestigious English accent in the UK, also dubbed ‘BBC English’ or the ‘Queen’s English’.

Interviewing people in the field

To find out whether changes are potentially taking place in the dialect landscape of the North York Moors, I carried out sociolinguistic interviews with eight participants—four older (50 years and over) and four younger (18 – 25 years). These interviews involved casual conversation and reading from a word list and a short story passage, each with varying levels of formality. For example, the casual conversation would ideally elicit an informal speech style from the participant, whereas reading a passage would ideally elicit a more formal speech style. In such interviews, the Observer’s Paradox is a difficult but inevitable problem for many research studies: you want to observe how people behave naturally, but being observed often makes people more self conscious than if they didn’t know they were being watched! To try and minimise this, I made sure that interviews were as relaxed as possible. Participants were aware that they were being recorded, but I put the recording device in a discreet place on a side table and tried to put participants at their ease.

Findings and observations

There is a significant increase in the use of the glottalled /t/ (or [ʔ], represented in the grey columns below) in the younger age group, as seen in Figure 1.

Figure 1: Distribution of /t/ variants by age (Total number of occurrences = 432)
Figure 1: Distribution of /t/ variants by age
(Total number of occurrences = 432)


The difference in the CURE and THOUGHT variables is also significant. I found that the traditional CURE variable is still the preferred pronunciation of choice by older and younger speakers across the three contexts (casual speech, reading passage and word list).

Figure 2: Distribution of CURE ~ THOUGHT by age. (Total number of occurrences = 192)
Figure 2: Distribution of CURE ~ THOUGHT by age.
(Total number of occurrences = 192)


In my discussion of the results, I suggested that the rise of the use of the glottal stop is a reflection of an urban-influenced linguistic trend that has been widely observed in empirical studies across England (cf. Llamas, 2007). Therefore, I argue that it would be simplistic to attribute the increase in the glottalled variant in the speech of people on the Moors to just the influence of neighbouring Middlesbrough. Younger speakers in this area of the UK have grown up with more geographical mobility than the older generation, and thus with more diverse influences on their speech.

Although the traditional CURE variable on the Moors is preferred by both older and younger groups, I found that the younger group showed an increase in the THOUGHT variable when adopting a more formal style of speech—in my study, this means reading from the word list. As I mentioned earlier, the THOUGHT variable is becoming the new standard, which suggests that a shift towards the standard RP accent may be taking place in the speech of North York Moors residents. However, this small-scale project threw up many more questions for further research, for example, how is the identity of young North York Moors residents changing and how is that affecting their accents? I hope to pursue these questions with postgraduate research in the near future.

My experience of conducting fieldwork as an aspiring sociolinguist was at times frustrating and nerve-wracking but most of all, it was rewarding. As an undergraduate student, you simply don’t realise just how much goes on behind the scenes when you read about other people’s studies until you conduct your own study. A multitude of decisions need to be made and it takes hours to transcribe spoken interviews: I spent one week transcribing the four hours of speech that I obtained from my participants. Then, after the arduous business of transcribing, you need to extract the relevant bits. In other words, the data that you’re going to analyse in your study need to be extracted, coded, and statistically analysed. My level of linguistics ‘nerdery’ rose to new heights when I successfully analysed my data and got a statistically significant result! Even for a small undergraduate study, a lot needs to be done before the hallowed final version emerges. The researcher must mould and shape the raw data into a polished version, like a sculptor working with a mass of raw materials. Research takes labour, time, and effort, but sociolinguistics is an exciting academic discipline and I am thoroughly smitten with it!



Llamas, C. (2007). “A place between places”: Language and identities in a border town. Language in Society, 36(04), 579-604.

Comments are closed.