Phonetic or Phonemic ?   C. Collingridge's question is a very frequent one in ELT ... and one that deserves the clearest answer possible.

Phonetics is concerned with the study of human beings' capacity to produce, transmit and interpret speech sounds.   As such, it attempts to represent all speech sounds that human beings have the capacity to produce with their speech organs and so does not focus on the sounds of any particular language.

Phonetics attempts to describe how we use our speech organs (i.e. articulators) in order to articulate sounds in terms of: a) the points in the vocal tract where they are articulated (e.g. bilabial sounds v. alveolar sounds), and b) the manners in which sounds are produced (e.g. plosives v. nasals).

Phoneticians use various different symbols in order to represent speech sounds visually (i.e. transcribe).   Since they aim to make the most accurate and faithful transcription possible, they work with phonetic symbols.   Professionally speaking, in Anglo-American traditions, phonetic transcriptions are made using square brackets to signal that the most accurate representation of what was actually articulated is being attempted.

For instance, if we made a phonetic transcription of the word 'water' as it tends to be pronounced in many regions of North America, the phonetic symbol corresponding to this English phoneme: /t/ would actually be: [ɾ].

Phonology, on the other hand, is concerned with the study of the sound system of specific languages.   This is, the restricted set of sounds (i.e. phonemes) which a sociocultural group of people in contact with one another sanction and consider as meaningful when they engage in communication with each another in the spoken medium in order to create, negotiate, interpret and achieve their intended meanings.   And so, we have areas of study such as Spanish Phonology, English Phonology, Russian Phonology and such like.

Phonology deals, broadly, with two major areas of analysis and study in reference to specific languages:

• Segmental Phonology (i.e. the analysis and study of individual sounds: their articulation, parsing, etc.) ... and
• Suprasegmental Phonology (i.e. the analysis and study of the communicative features which characterise natural uses of language in the spoken medium: a notion referred to as 'connected speech').

In phonological analyses, we only use a restricted number of the many various different symbols available in Phonetics in order to represent visually the sound system of a specific language.   A phonological transcription aims to present a careful, idealised version of how a sound would be rendered, and so we'd work with phonemic symbols.   Professionally speaking, in Anglo-American traditions, phonemic transcriptions are made using slanted brackets to signal that an ideal and careful version is being attempted (e.g. /w/ /ɔ:/ /t/ /ə/ /r/, etc.).

The chart shown on this site is thus: phonemic (as it only represents the 44 individual sounds of the British variety known as a: 'BBC accent' or 'RP accent').

Finally, why are phonetic and phonemic symbols necessary in the first place ?     Well, it all stems from the lack of perfect, one-to-one correspendences between the spoken and written varieties of a given language.

From a theoretical perspective, the discrepancies which exist between the spoken and written varieties of a given language and the degree of cognitive and sociolinguistic effort required for their users to process these discrepancies, languages may be regarded as falling within one of two nraod groups: shallow v. deep languages.

Shallow languages (e.g. Spanish) are characterised by having convergent spoken and written varieties which match each other very closely, and so the depth of the cognitive or sociolinguistic processing required to match and relate written with spoken versions of such languages is not excessively demanding on their users and their communicative resources.

Deep languages (e.g. English), by contrast, have divergent spoken and written varieties which require much deeper cognitive and sociolinguistic types of processing on the part of their users in order to relate written with spoken versions of such languages.   In other words, written forms of words in a deep language cannot be taken at face value for their spellings are not indicative of their pronunciations in the spoken medium.

When native users of a shallow language start learning French or English, for instance, they inevitably suffer from two associated conditions known as: 'language shock' and 'language stress'; especially when they first realise that it is not possible to pronounce words as they see them written and they pretty much have to learn the pronunciation of each individual word they encounter by heart ---at least until they go well beyond the threshold of intelligibility and begin to accept (and continue to discover more stable) phonological features and patterns of correspondence between written and spoken varieties of these deep languages.

The above is of pedagogical significance for Teachers of English in as much as carefully guided, explicit and systematic explorations of segmental phonology in English (and their associated phonemic symbols) will be instrumental in gradually enabling learners to become more and more confident, autonomous and independent users of the English language, for they will know how to pronounce words when they their phonemic transcriptions as they finally come to terms with the fact that English is a deep language.

Cheers !

Eduardo Valdes Garcia Torres, LTCL DipTESOL

Director

Institute for Applied Linguistics

Mexico