Introduction
If we would like to discuss the role of phonological representations in speech, then we need to focus on such phenomenon as the phonological neutralization. This phenomenon actually eliminates a phonemic distinction, especially in a particular context of the phonology. Such an example as a contrast of the word-final voicing is very common in the neutralization.
Main text
In order to understand these, I would like to give you an example, in the context of the minimal word pair (rat/rad (German)). Here we can see the distinction of phonological representations. At the same time we can see here the phonetic representations, which are actually identical.
The first one – “rat” has an “advice” gloss, its phonological (or underlying) form is /rat/, its phonetic (or surface) form is /rat/.
Now, we should see and analyze the word “rad” (German), which has the “wheel” gloss, its phonological (or underlying) form is /rad/, its phonetic (or surface) form is /rat/.
According to this phonological analysis words “rat” and “rad” are different by their phonological representation in such terms as the voicing of the final consonant. Here we can also see that the phonetic forms are totally identical. These words are ending in a certain voiceless alveolar stop.
So, from this analysis we can make a conclusion that the two phonetic forms should be identical if neutralization is already phonetically complete.
Also we can see that the two phonetic forms differ in certain predictable ways if neutralization is actually incomplete.
If we are talking about history of the phonology of the English vowels, then it is necessary to mention that it involves a number of sound changes, such as diachronic. I think that we should be really attentive with its mergers and phonemic splits.
If we are talking about the tense–lax neutralization, we should mention that it
refers to a neutralization not an ordinary way, but in a special phonological context, in a particular required language. We should not also forget about the normal distinction between tense and lax vowels.
Such situation occurs in English mainly before /ŋ/ and before /r/ (in the case when /r/ is followed by a consonant or is situated at the end of a word). It also occurs, when there is a lesser extent, particularly, before tautosyllabic /ʃ/ and /g/. Here I would like to introduce couple of examples of the neutralization of /ɛ/ to /eɪ/ before /ɡ/. They are: egg, beg, keg, Greg, leg and pegs. They are coming to rhyme with such words as: Craig, plague, Hague, and vague.
If we are talking about the American English dialects then the varieties usually have significant and vocalic neutralization before such intervocalic as /r/.
If we are talking about Low front vowels, then there are processes that cause the accents, such as: æ-tensing. In some dialects there is a certain allophonic split, in other dialects it influences all /æ/s. There are also dialects that have the phonemic split.
The short vowel phonemic split can be meet in the Early Modern English and Australian English, where: /æ/ comes into a short /æ/ and a long /æː/.
We can see a new detail in Modern English, which is represented by a new phoneme, /ɑː/.
If we are talking about Early Modern English then we can see here the trap–bath split, where /æ/ merged with the /ɑː/ in certain environments. I refer to such accents as the Boston, the Southern Hemisphere.
Analyzing low back vowels we can name such kinds of them as: the father–bother merger vowels with /ɑː/ and /ɒ/; the lot–cloth split with lengthened /ɒ/ to [ɒː] before voiceless fricatives.
If we are talking about the cot–caught merger then we can see that such words as: cot, rock, and doll are pronounced thin the same way as the words: vowel in the words caught, talk, law, and small.
In the psalm–sum merger, particularly within Singaporean English, the phonemes /ɑ/ and /ʌ/ are both pronounced /ɑ/.
Within the context of Jamaican English, we can see the usage of the bud–bird merger of /ɜ/ and /ʌ/.
If we are talking about the high back vowels then we need to mention the foot–goose merger, where we meet the vowels /ʊ/ and /uː/; the foot–strut split where we can see that Middle English /ʊ/ transforms into two distinct phonemes /ʊ/ (as in foot) and /ʌ/ (as in strut); the dew–duke merger, where the front vowel /y/ transforms into /iu/; the dew–new merger with diphthongs /iu/ and /ɛu/ that is present in all dialects of the modern English.
If we are talking about high front vowels then we should mention some interesting cases of their usage. It is interesting that in the kit–bit split, such words as kit [kɪt] and bit [bət] actually do not rhyme.
We should not also forget about that English-language vowel sometimes changes before such historic letters as: “r” and “i”.
I would like to mention an interesting to my mind fact about mere–mirror merger. This process happens, when /ɚ/ disappears just after intervocalic /r/ in words that are called “monomorphemic”. So, here are some examples of how some words may become homophonous: mirror – mere; terror – tear; error – air; horror – whore.
Conclusion
In a conclusion I would like to say that the phonological representations in speech in the terms of phonological neutralization changes according to the location of the speaker and the certain period of time. So it should be regarded with a view to these parameters and every single case requires a different research approach.
Works Cited
Roca, Iggy, ed. Derivations and Constraints in Phonology. Oxford: Clarendon Press, 1997.
Trask, R. L. A Dictionary of Phonetics and Phonology. London: Routledge, 1996.