i love english language


Posted in Uncategorized by aggslanguage on September 20, 2009


William Labov, born 1927, looked at change in language and how different social factors effected language change, ranging from age to social class to gender. He found language change was either conscious of unconscious, unconscious being when people change their language without noticing, and conscious being when people realise they are changing the way they speak, and actively encourage it.  This investigation looks at conscious speech change and uses the example of Labov’s New York Department Store study with the involvement of PRESTIGE.

Prestige can be separated into ‘overt prestige’ and ‘covert prestige’. Both are used when changing speech to gain prestige – appearing to have a high reputation/standing/success etc- but do so in different ways. If someone uses ‘overt prestige’ they put on an accent that is generally widely recognized as being used but the ‘culturally dominant group’. In England this would be R.P, so putting on a more ‘posh’ accent than their regional one would be using overt prestige, to fit with the ‘dominant group’. This is the traditional definition, although with the rise of other accents such as Estuary English it may be questioned. ‘Covert prestige’ is the opposite, as ‘covert’ means secret. Therefore it means to put on an accent to show membership to an ‘exclusive community’ in the area, rather than to fit with the ‘dominant culture group’. Using covert prestige would therefore be putting on a more ‘street cred’ accent rather than R.P, and even though the ‘dominant culture group’ generally sees it as being inferior, using language fitting with the local community would lead to earning respect with those also in the community.


In 1966 Labov completed the “New York department store study” which examined overt prestige involving both class and gender. Labov investigated the pronunciation of the ‘post-vocalic’ /r/ sound in American speech, which is the /r/ sound that comes directly after a vowel in words such as the middle of the word ‘fourth’ and the end of the word ‘floor’. Labov carried out his experiment by walking into 3 different department stores in New York representing different social classes, being Saks (upper class), Macy’s (middle class) and S. Klein (lower class). He went on to ask shop assistants where the location of departments were that he knew were on the fourth floor, to allow them to spontaneously say the words ‘fourth floor’ which includes the /r/ pronunciation. Furthermore as an added factor he then pretended he had not heard the assistant, making them repeat their answer of ‘fourth floor’ to see if their pronunciation had now changed, as their speech had become careful rather than spontaneous.

His findings were that the sales assistants from Saks used the /r/ sound most, showing that the current overt prestige form in New York was to pronounce the /r/. Those from Klein’s used it least as they would have used more covert prestige, so would not have pronounced the /r/ sound, and said an utterance along the lines of “flaw”. Finally those from Macy’s showed the greatest upward shift of pronouncing “floor” rather than “flaw” when they were asked to repeat their utterance.

Therefore Labov found that the pronunciation of /r/ increased as the class of the store increased, as well as an increase of /r/ in careful speech, and concluded that the more careful the speech was the more likely the /r/ was to be pronounced. Labov found the overuse of /r/, known as hypercorrection, was most common in the lower middle class(Macy’s), as they were most likely to be aware of which speech forms are ‘classy’ and would use these forms in careful speech to improve prestige and appear to belong to the higher middle classLabov also found hypercorrectness to be strongest in the language conscious middle class women, showing that overt prestige seemed more common in women than men, the factor that I chose to investigate in my experiment.


As my investigation was based on Labov’s theory of overt and covert prestige, I would expect that female speech would contain more overt prestige, so contain changes in speech from their regional accent to R.P to appear more ‘posh’ and to fit in with the ‘culturally dominant group’. Therefore, they may adopt phonological features of R.P. such as elongating the /a:/ sound. In contrast male speech would be expected to have more covert prestige, therefore speaking more with their regional accent.


I carried out the investigation by looking into careful speech, as the data would be more straightforward to collect, and decided on four words which have different pronunciations in a local Manchester accent compared to R.P. To make sure the pronunciation of the chosen words weren’t completely unnatural they were placed in sentences, which I asked 20 males and 20 females to read aloud, whilst recording whether the chosen words were pronounced in a regional accent using covert prestige, or in R.P, using overt prestige.

My chosen words were ‘lunch’, ‘book’, ‘cinema’ and ‘garage’ placed into the following sentences:

–          ‘I bought a sandwich for my lunch.’

–          ‘I went to the library and borrowed a book.’

–          ‘I went to the cinema and saw a film.’

–          ‘The car was parked in the garage.’


I identified the two contrasting ways of pronouncing the chosen words, and written using the International Phonetic Alphabet are:

Chosen word Regional accent R.P.
Lunch l nt∫ l      nt∫
Book bu:k b k
Cinema sInIm sInIma:
Garage gærId3 gæra:d3


The Recorded Data:


  said with a regional accent said in R.P

















  said with a regional accent said in R.P.














The results showed that regarding certain words in the female data there was an outright majority in the pronunciation, as everybody chose to say ‘book’ in the style of R.P, rather than with a strong Manchester accent. On the other hand, ‘cinema’ was said almost always with the regional accent rather than in R.P. This is an interesting observation, as it shows in the female group there was no outright result as to which accent was used when asked to read the sentences, it depended on the word. There was a similar result with the male speakers, that everyone said the word ‘book’ in R.P, and ‘cinema’ with a regional accent. Because the results for male and female were so similar, it questions Labov’s theory that women use more overt prestige than men.


However, the words ‘lunch’ and ‘garage’, in particular ‘lunch’, are more in favour with Labov’s theory of prestige and prove my hypothesis. When asked to read the sentence containing the word ‘lunch’ 80% of the males used the regional accent and 20% with R.P. Yet when females were asked to do the same task 65% of them used a regional accent whereas 35% used R.P. This supports Labov’s theory, as there was a 15% increase in the use of R.P. between males and females, suggesting women used more overt prestige than the men, who used more covert prestige. The results for the word ‘garage’ were along a similar theme, as 5% of females used more overt prestige, uttering the word in an R.P accent. This result is not a strong as the result concerning ‘lunch’, but still proves my hypothesis and Labov’s theory.


In conclusion of the results, certain words in my experiment prove my hypothesis that women use more overt prestige than men do. However, other words in the experiment disproved Labov’s theory, not because men used more overt prestige than women did, but because the results were very similar between male and female, which showed that neither sex used prestige that was more overt. When looking at the results though there is an added conclusion, because the words ‘book’ and ‘cinema’ had an almost completely one-sided result. This questions whether prestige was in fact considered at all when uttering the words, and that the differences between pronunciation of the words ‘lunch’ and ‘garage’ that seemed to contain prestige were actually due to outside factors, such as differences in accent due to background, class and age. This theory is reflected in Labov’s work, who also concluded in any language experiment other factors are involved and prestige in fact ‘complicates matters’.


When evaluating this investigation several issues need to be addressed. As explained in the conclusion of results there are other outside variables, suggesting the experiment was not a ‘fair test’. Although the people involved in the experiment were all from the same location, they did not all have the same background, i.e. some people spent large majorities of their childhood in different areas. This suggests their accent may not have fit to the typical ‘regional accent’ and made them appear to be speaking in R.P. and using overt prestige, when they were not at all, therefore skewing the results. Other added factors were that the people were not all from the same class, they were of different ages, and the relationship I had with most of them was one that may have meant speakers were unlikely to feel the need to include prestige of any sort in their speech.

Additional problems involved how the experiment was carried out, effecting the reliability of the data. The data was only collected from 20 members of each sex and, as there were added factors besides prestige, it meant anomalies would have highly effected the overall conclusions. To improve reliability of the data I would increase the number of people asked, and possibly have a larger number of words to test, to leave room to discard anomaly results. I would also consider the words I select more carefully. In my experiment the word ‘book’ was possibly an incorrect word to choose, as 100% of both male and females pronounced it in R.P. This highlights the fact the regional accent pronunciation that I chose, being /bu:k/, may not have been correct for the area I was testing in, as the results for the other words showed over 60% of the utterances were made in the regional accent, male or female.


If I were to review my results again with the knowledge the figures for ‘book’ may be anomalies, my conclusions are different, and more decisive. Even though with both male and female the majority of utterances were made with the regional accent, those made in R.P. are slightly higher in the female utterances for all 3 words of ‘lunch’, ‘cinema’ and ‘garage’. Therefore assuming there were not additional factors, Labov’s theory, and my hypothesis, have been proved correctly, that women use more overt prestige.

H Eakin

6 Responses

Subscribe to comments with RSS.

  1. Said Rizi said, on September 12, 2010 at 11:31 am

    Very useful. Hopefully, I can make use of it in my CLC research as well.

  2. Melissa said, on April 28, 2011 at 2:28 pm

    Since this is about language use, you should know that ‘effect’ is a noun and ‘affect’ is a verb, so you may want to alter your write up. Otherwise, interesting findings.

  3. Eric said, on November 14, 2011 at 1:56 am

    Hi, I have a question. Is there any famous research study linked with covert prestige?

    I understand that covert means secret. From what I have read, Labov’s Department Store study is to be linked with overt prestige involving social class.

    • aggslanguage said, on November 18, 2011 at 5:42 pm

      Labov’s Martha’s Vineyard study is the clearest example

  4. Eric said, on November 14, 2011 at 2:19 am

    Can Labov’s New York Department Store Study be used to link with covert (secret) prestige?

    In the sense, that we look at S Klein (lower class) as the most percentage (80%) of salespeople tend towards vernacular speech varieties without the post-vocalic /r/, followed by Macy’s (middle class: 48%) and finally Saks Fifth Avenue (upper class:38%)? If I am wrong, please correct me. Thanks!

    • aggslanguage said, on November 18, 2011 at 5:41 pm

      possibly. there is certainly an argument to be made that covert prestige was active in Kleins – but i don’t think Labov made that point

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: