The Spoken BNC2014 early access projects: Part 1

In January, we announced the recipients of the Spoken BNC2014 Early Access Data Grants. Over the next several months, they will use exclusive access to the first five million words of Spoken BNC2014 data to carry out a total of thirteen research projects.

In this series of blogs, we are excited to share more information about these projects, in the words of their authors.

In Part 1 of our series, read about the work of Deanna Wong, Jonathan Culpeper and Robert Fuchs.

Deanna Wong

Macquarie University, Australia

Investigating British English backchannels in the Spoken BNC2014

Have you ever listened to someone listening? While we might expect that listeners are silent, it turns out that listeners have a lot to say. Mostly, this listener speech happens at the same time as when the speaker is talking, but listeners are not talking to interrupt the speaker. Instead, listeners signal to the speaker that they are paying attention, that they agree with what the speaker has to say, and sometimes, that they are ready to have their turn at talking. The words that listeners use to signal these things can range from a simple mm to whole sentences. To make things even more interesting, how listeners listen varies across different parts of the world.

Sociolinguists use the term ‘backchannels’ to describe listener speech. Early research identified backchannels by careful investigations of individual conversations. That analysis took time, though, and it was not until researchers were able to access language corpora that we started to get a sense of the nature of backchannels in conversation on a larger scale.

However looking for evidence of backchannels in a corpus has its own challenges. If the actual language used by listeners is to be uncovered, we cannot assume that they take a specific form. Otherwise, we might miss something important! The key to unlocking this information is to use corpus annotation. Annotation is simply a way of marking what is happening in the talk. For example, corpus annotation can be used to indicate who is speaking, and if they are speaking at the same time so that their speech overlaps.

In my investigation into the Spoken BNC2014 early access subset, I will be using annotation that marks overlapping speech to help identify potential backchannels in conversations from across the United Kingdom. The size of the corpus, and its accompanying information about its speakers will add to our understanding of how British speakers backchannel. It will also help us to compare their backchannels to those produced by speakers of English in other parts of the world.

Jonathan Culpeper

Lancaster University, UK

Politeness variation in England

The stereotype of British politeness is pervasive, and, moreover, it is usually linked to what people say. Take, as an example, this advice on British stereotypes for study abroad students:

The way that British people speak and the language that we use is also considered quite polite. The language that many people use, including lots of phrases like ‘please’, ‘thank you’, ‘pardon’ or ‘excuse me’ and ‘would you mind…’ certainly back this up […]


In fact, the first item in the list, please, seems to be elevated by many English parents to the supernatural – the “magic word” for achieving successful requests. Similarly, in the earthly world of academia, a large number of studies have found evidence that present-day English politeness is often characterised by so-called “off-record” or “negative politeness” –  it’s all about being indirect, showing respect for others’ privacy, freedom from disturbance, and so on. In the example, the expressions ‘would you mind’, ‘pardon me’ and ‘excuse me’ all readily fit this function. To these, one might add could you […], seemingly, the most frequent way in which requests are performed in British English.

But is all this true? For starters, there’s a lingering concern that some British people may actually use other items, perhaps functional alternatives, just as or even more frequently. For instance, thank you is one expression, but what about ta or cheers? More substantially, for anybody living in the north of England, the idea of British indirectness does not entirely ring true. Indirectness is somewhat stand-offish and cold; not reflective of the much proclaimed northern warmth and friendliness. Consider this opinion (written by a Scotsman who has lived in various parts of Britain):

There is definitely a North/South divide when it comes to politeness. Having lived on the South coast of England and then Scotland, it is very noticeable that people are more friendly and polite the further North you go in Britain


Politeness here is connected to friendliness. Maybe academia is orienting to a particular and different cultural stereotype of politeness, one based on a British southern perspective. Or maybe the idea of northern friendly politeness is a stereotype itself and has no basis in what people actually do.

This study sets out to examine these issues. I intend it to be a contribution to one of the newest sub-fields of linguistic pragmatics, variational pragmatics, which combines pragmatics and dialectology. One of the greatest impediments to doing such a study has been the lack of large quantities of spoken, especially conversational, data taken from across Britain. The Spoken British National Corpus 2014 early access subset offers a solution.

Robert Fuchs

University of Münster, Germany

Recent change in the sociolinguistics of intensifiers in British English

As social beings and speakers of a language we are extremely good at putting people into boxes – female and male, young and old, old-fashioned and hip. One of the many clues that allows us to make these (sometimes in fact unwarranted) assumptions is sociolinguistic variation. Whether and how women and men differ in how they speak, for example, is a hotly debated topic in- and outside of academia.

This study approaches this topic from two novel angles. The first is that several sociolinguistic factors, age, social class, gender of speaker, gender of interlocutor and others, are considered in interaction with each other. Secondly, the study also looks at change across time, from the 1990s to the 2010s. For example, given the change in attitudes concerning what roles women and men are supposed to fulfil in society, I expect that any gender differences present in the 1990s will have decreased by the 2010s. The variable that the study investigates is the usage of so-called intensifiers (as in *very* good, *so* cool), which are said to occur more frequently in female than male speech.

Check back soon for Part 2!