Phonetics and phonology, phonemes, allophones, sound combinations, metrical phonology



Linguistics studies primarily the spoken word. To make the research easier linguists use a
specially devised system of notation. In the system one symbol represents one sound. Perhaps the
best known of these systems is IPA (the International Phonetic Alphabet). Many of IPA symbols are
borrowed from the conventional Latin alphabet. In some cases the symbols are variations of alphabet
letters, in some cases they are old letters not used in writing nowadays, some are brand new.
Knowledge of phonetics seems to be basic for linguistic knowledge itself. It is a prerequisite for
linguistics.


Linguists want to be accurate in transcription of sounds. Therefore they divide the utterance into
smaller and smaller units. The smallest segment of sound, which can distinguish two words, is a
phoneme. E.g. sat and sad. Pairs of words such as lock and log, which differ by only one phoneme,
are called minimal pairs. Minimal pairs offer one way to identify the phonemes of any language. The
number of phonemes in the language is relatively small. It varies from language to language. The
average number is around thirty-five. English has forty-four.


British English is a widely spoken variety of English. The standard pronunciation in British
English is known as RP (Received Pronunciation). There are forty-four phonemes used in RP,
consonants and vowels. The number of vowels can be divided into pure or unchanging vowels, as in
let, cat, and diphthongs (gliding vowels) as in lay, boat.


The phonemes have also their variant forms called allophones. The number of variations differs from
phoneme to phoneme. As for variations we distinguish so called free variations (The case when the
variation is random. No two sounds can be exactly the same even if the speaker tries hard.) The
second group of phonemes appear in complementary distribution. It means that the way a sound is
pronounced is conditioned by the sounds surrounding it. It also can be conditioned by its position
in the word.  E.g. pit is pronounced with clear explosive [p]. But in spit it is aspirated. So the
clear variant and the aspirated variant of [p] are both allophones of the phoneme [p].


All languages have certain permitted traditional sequence of sounds and other sequences that are
not allowed. Spanish for example does not permit a word to begin with [s] cluster. That is why
Spanish is translated as Espaňol with [es] at the beginning. In English consonant clusters with
three consonants at the beginning of words obey three strict rules:

1)      The first phoneme must be [s].

2)      The second phoneme must be [p], [t] or [k].

3)      The third phoneme must be [l], [r], [w] or [j].


Many phonemes share common features. They are mostly easily distinguishable. That is why we can
identify a group of voiced and voiceless consonants. Labials, nasals … . Any feature that
distinguishes one phoneme from another is called a distinctive feature. A group of sounds which
share important features in common (labials, nasals …) is known as natural class of sounds.


English phonemes are segmental phonemes as they are segments of sounds. Other languages can perform
non-segmental phonemes. E.g. North Mandarin Chinese distinguishes numerous words by differences in
the rise and fall of tone.


English possesses non-segmental features, too. Each word and group of words has its own rhythm. A
relatively new branch of phonology studies the rhythm as interplay of stressed and unstressed
syllables. It is metrical phonology. Metrical phonology uses tree diagrams to demonstrate an
internal structure of rhythm. It shows how strong and weak syllables alternate.




S = strong

W = weak


Recent phonology pays attention also to the interaction between sound segments and rhythm.