Vowels and Consonants: An Introduction to the Sounds of Languages (review)

This volume is authored by the ‘world’s leading phonetician’, to quote John Ohala on the back cover, and I certainly agree with him. Born in England and educated in Scotland, the long-time UCLA professor and director of the UCLA Phonetics Laboratory (now Emeritus) is practically a household name in many linguistics departments, where countless numbers of students (including mine, ever since the first edition in 1975) have been exposed to the four editions of his A course in phonetics (Fort Worth: Harcourt, 2001), not to mention his numerous other books and articles. This work is less technical than A course in that it omits many technical phonetic concepts and is written in simpler language; however, there is, not surprisingly, some overlap between the two, such as with the vowels of English [(26–30) and A course (29, 81)] and the contrasting stops and nasals in Malayalam [(149) and A course (141)]. At first, it may seem more complex than A course since it starts with acoustic phonetics, a rather unconventional approach (it is Ch. 8 of 11 chapters in A course).

Perhaps the two most fascinating chapters are Ch. 8, ‘Talking computers’ (68–77), and Ch. 9, ‘Listening computers’ (78–95). L does an excellent job in summarizing speech synthesis systems, concluding that much remains to be done. More specifically: ‘It will be a long time before a computer can give a good performance of Hamlet’ (71), and ‘Nobody has yet made a system that can be clearly understood and also sounds perfectly natural’ (77). Speech recognition systems are far ahead of those for speech synthesis to the point where L opines that we will have ‘almost perfect speech recognition systems before we have completely natural-sounding speech synthesis’ (95).

When one thinks of articulatory phonetics, the likes of Ch. 12, ‘Actions of the larynx’ (118–37) immediately come to mind, with tables of contrasts in various languages; e.g. voiced and voiceless nasals and laterals in Burmese (118) (some to be found in A course [265]), breathy voice (= voiced aspirated stops in Hindi [123–26] and breathy-voiced vowels in Gujarati [126–27]), creaky voice (= laryngealized or glottalized vowels in Mazatec [127–30]), voice qualities and tones (minimal pairs of a dozen words /si/ in Mpi, spoken in Thailand [130–31]), ejectives (131–33), and implosives (133–35). L notes that implosives ‘are more difficult sounds to produce than ejectives … most adults have lost the art of producing them spontaneously, and find them hard to make’ (133). Surely, there is a relativity involved here deserving comment. Pull-ups are difficult for most people since they do not consistently do them. Pharyngeal fricatives are difficult for English speakers, since English does not have these consonants, whereas interdentals or uvulars are difficult for speakers whose languages do not have them. And practice supposedly makes perfect.

The final three chapters place the entire volume’s subject matter into proper perspective. Ch. 13, ‘Consonants around the world’ (136–57), contains many interesting observations about the 600 consonants found in the languages of the world. In a table listing the ten most widely spoken languages, number 8 is given as Arabic (all forms) with 161 million speakers (140). Most estimates I have seen put this number well in excess of 200 million.

Ch. 14, ‘Vowels around the world’ (158–68), reiterates what linguists have been taught for decades, viz., ‘probably every language uses at least three distinct vowels’ (159). L goes on record disagreeing with the often-mentioned claim that Kabardian has fewer (ibid.). L begins the chapter by asking how many vowels there are. We only learn the answer (for some inexplicable reason) in the following chapter: approximately 200 (179).

Ch. 15, ‘Putting vowels and consonants together’ (169–83), will shatter the views of many by iconoclastically asserting that speech is not an amalgamation of consonants and vowels, since ‘they are not stored separately … [but rather] as wholes, or at least...