1. The linguistic variable and the identification of variants.
  2. This paper shows the tendencies in the palatalisation of alveolar plosives /t/, /d/ and alveolar fricatives /s/, /z/ when followed by a palatal approximant /j/, both word-internally and across word boundaries. Palatalisation is “a secondary articulation in which the front of the tongue is raised toward the hard palate” (Ladefoged 295). As a result of palatalisation, /t/, d/, /s/ and /z/ followed by a /j/ sound are realised as /tſ/, /dz/, /ſ/ and /z/ respectively.

  3. Hypothesis.
  4. Hypothetically speaking, the process of palatalisation will occur relatively frequently in informal and semi-formal contexts, since no attention is paid then to sounding “correct”. Furthermore, informal speech being fast and pro ne to ellipsis, it makes a perfect ground for secondary phonological processes, inculuding palatalisation.

    The second assumption is that palatalisation will affect the most common words and phrases, decoding of which will not cause any difficulty, e.g /t, d, s, z/ sounds followed by “you”, “your”, “year”, “yet”. Therefore, th e process will probably occur more frequently across word boundaries.

    The last hypothesis is that palatalisation is more common in British accents than American ones, since /ju:/ is usually realised as /u:/ in American accents, e.g. BE /tju:n / vs. AE /tu:n /.

  5. Data collection.
  6. a) The method of data collection and informant sampling.

    The tokens have been collected in two ways. A considerable part of the data (154 tokens) has been collected by means of observing television announcers on Sky News and CNN channels, whereas the remaining 52 tokens have been obtained by means of recording the speech of native speakers of English teaching at the School of English in Poznan, Poland. The tokens occurred in semi-formal speech, with more attention paid to the text spoken rather than actual pronunciation.

    b) The speech community.

    As far as the educational and social backgrounds of the informants are concerned, they are all university graduates and members of higher middle class. The British accents speakers come from the British Isles, while the American acc ents speakers are from the United States of America. No distinction as to dialectal varieties has been made.

  7. The results obtained.

  1. Examples of tokens:

  1. Presentation of results:

i) Overall results:

 

Context:

Word-internal

Across word boundaries

Total

all

only phrases including “YOU” or “YOU”

/tj/

8

27

13

35

/tſ/ (palatalised)

13

23

9

36

/dj/

17

1

0

18

/dz/ (palatalised)

16

26

15

42

/sj/

8

19

14

27

/ſ/ (palatalised)

7

11

10

18

/zj/

1

20

10

21

/z/ (palatalised)

1

8

3

9

Total:

71

135

71

206

Tokens palatalised: 105 (50.97% of the overall data).

ii) The results with respect to the position of the sounds undergoing palatalisation.

 

 

 

Graph 1. The occurrence of palatalised forms with respect to the environment.

Graph 1 clearly shows that palatalisation is most common across word boundaries (61.75%), phrases containing the pronoun “YOU” being slightly more prone to palatalisation that other items followed by a word-initial /j/ sound.

 

 

iii) The results with respect to the sounds palatalised.

Graphs 2-5 show tendencies towards palatalisation of each of the sounds tested: alveolar stop and fricatives, both voiced and voiceless. Considering the pair of alveolar plosives, in case of the voiced one (/d/) the addition of high fro nt tongue position is more common than when the voiceless alveolar stop /t/ is followed by the semivowel /j/. Conversely, the voiceless alveolar fricative /s/ tends to be palatalised more often than the voiced one /z/).

Graph 6. Palatalisation with respect to the sounds undergoing the process (number of items).

Out of the total of 206 collected tokens, the proneness to undergo palatalisation differs as follows (descending order):

/d/ - 42 occurences (70% of all voiced alveolar plosives included in the data, 20.4% of all the data),

/t/ - 36 occurences (50.7% of all voiceless alveolar plosives included in the data, 17.5% of all the data),

/s/ - 18 occurences (40% of all voiced alveolar plosives included in the data, 8.7% of all the data),

/z/ - 9 occurences (30% of all voiced alveolar plosives included in the data, 4.4% of all the data).

The share (per cent) of each of the palatalisated sounds in the data collected shows graph 7.

Graph 7. The share of the palatalised sounds in the total of 206 tokens.

Graph 7 illustrates the high tendency of alveolar plosives towards palatalisation (37.9% of the overall data), as compared to the alveolar fricatives, which are less prone to undergo the process (13.1% of the data).

 

 

 

iv) The results with respect to the accent (British vs. American).

Although it is not possible to classify all the accents as either British or American without taking into account both the place of birth and residence as well as the past of the speakers, the informants have been divided into t he two classes on the grounds of the analysis of their accents, the channel of their appearance (in case of data collection by means of observation), or the information available about the informants (in case of the recordings). Since rhoticity does not a ffect palatalisation, Irish accents speakers have been classified as British accents speakers on the basis of locality.

 

Graph 8. Palatalisation with respect to the accent (British vs. American) (unit: number of items).

The patterns of palatalisation shown in graph 8 are similar in case of both accents. Slightly more than a half of the tokens in each groups underwent palatalisation, the percentage being slightly higher in British accents.

  1. Explanation.
  2. Palatalisation is a kind of secondary articulation. “There, the primary place of articulation of a consonant is elsewhere in the mouth; for example a [t] sound, normally made in alveolar position is said to be palatalised if durin g its articulation the front of the tongue is raised toward the hard palate: in the case of [t], the palatalisation would be more noticeable when the plosive was released, as a palatal glide would then be heard before the onset of the next main sound” ( Crystal).

    In her speech on “Palatalisation effects as results of articulatory- and perceptually-driven mechanisms” (Workshop on Phonetics-Phonology Interface, the 33rd Poznań Linguistic M eeting, 27-29 April 2001) Małgorzata Ćavar from the University of Potsdam noted that “different surface results of palatalisation have (...) been claimed to be lexicalised.

    “Another approach to palatalisation takes it to be an auditory-driven phenomenon.

    “Palatalisation effects (...) result from the interaction of articulatory and auditory processes (...) and constraints.” There are auditory constraints, which “aim at the minimisation of the auditory confusion between different su rface forms, and articulatory constraints target a reduction of the articulatory effort involved in the production of a surface form.”

    The results of the research prove that the thesis of the interaction of articulatory and auditory processes is correct. Most of the palatalised sounds (67%) occurred either word-internally or included the initial palatal approximant of “you”, which prevented any ambiguity or confusion in perception. The comprehensibility condition being met, the palatalisation process took place and the effort entailed by sound production was reduced. Alveolar plosives and fricatives were pronou nced as postalveolar fricatives and affricates respectively, the place of secondary articulation being closer to that of the following /j/ sound.

    As far as the frequency of occurrences is concerned, more plosives underwent palatalisation than fricatives did, since the former outnumbered the latter in the overall data collected.

    The results show that only slightly more than half of the tokens were palatalised, which is what had been expected from semi-informal speech. The informants neither stuck to prescriptive norms of pronunciation which opt for maximal clarity and comprehensibility of messages, nor followed the most informal kind of rapid speech with entails a large amount of sound elision and merging. Thus, optimal articulatory output has been achieved, meeting both auditory and articulatory conditions .

     

     

     

  3. Conclusions.
  4. Extrapolating from the results of the research, palatalisation occurs most frequently across word boundaries, the sounds most commonly involved being the alveolar plosives (/d,t/) followed by the second person pronoun “YOU” or p ossessive “YOUR”. Although it was assumed that palatalisation would be rather infrequent in semi-formal speech, the results show that over a half of the data has undergone the process. Finally, accents spoken by the speakers have had hardly any influe nce on the rate of palatalisation; the process seems to be universal and take place regardless of locality.

  5. Bibliography:

Ladefoged, Peter. A Course In Phonetics. Philadelphia: Harcurt Brace College Publ., 1993.

Crystal, David. The Encyclopaedia of the English Language.Cambridge: CUP, 1995.

Ćavar, Małgorzata. Palatalisation effects as results of articulatory- and perceptually-driven mechanisms. Talk given at P LM, Poznań 2001.