In his review of two texts on reading research (Books, March 24), Charles Hulme says phonemes are "the individual speech sound units in words" and that the alphabetic principle is "that the letters in printed words map directly on to the sounds - phonemes - of spoken words".
I don't know whether any of my linguistics colleagues would subscribe to that. Any standard reference work on linguistics states that phonemes are the minimal units in the sound system of a language, that is, sets of similar phones that can be grouped together. Phones are said to be realisations of phonemes. This is not quite the same as Hulme says.
So there are no phonemes in speech sounds, they cannot possibly be units in the speech sounds uttered by real people, and that's why alphabets can be representations of the abstract entities "phonemes". It is also why people speaking with a huge variety of English dialects that seem to have very little correspondence between phonemes and orthography are able to read and write what they call English.
How can I trust researchers in the psychology of reading field when a basic notion from linguistics such as this is misconceived? And how can I trust reviewers outside my field of expertise if I don't know whether they are up to date with the basic know-ledge?
Hans Götzsche
Director Centre for Linguistics
Aalborg University, Denmark