I first heard of John Wells, Professor of Phonetics at the UCL, not as a linguist, but as a geeky teenager who was learning Esperanto. Me, that is, not him. John Wells was, at the time, president of the Universala Esperanto-Asocio (if you can’t guess what that is, go away), and the author of a dictionary of Esperanto that I owned. He’s the author of two blogs, one about questions that arise as he revises his Esperanto dictionary (now completed) and the other about phonetics. I recommend both to you, assuming you know Esperanto and English. If you haven’t learnt Esperanto yet, then at least read his phonetics blog.
Recently he’s been complaining (constructively) about the poor quality of the pronunciation advice in the Guardian’s “Languages for the 21st century” series (here and here). I support his complaints entirely, and have long felt that pronunciation guides aimed at language learners are generally considerably poorer than they should be. As Wells shows, it’s not only that they lack phonetic detail that would be useful to a linguist; they almost invariably resort to ad hoc impressionistic descriptions that are of no practical help to anyone: things like “hard h”. As Wells says: “If you can’t say anything useful, say nothing.”
And we just shouldn’t have to put up with this. The science of phonetics managed years ago to progress to a point where have sensible ways of describing speech sounds. In fact, it got to this point over a century ago. And yet in the very place where this knowledge would benefit the most people, it gets ignored, partly because the amateurs who do the work can’t be bothered to do their research properly (granted, these people may not be amateurs when it comes to language teaching, but they’re being unprofessional in failing to recognise where they don’t know enough). I admit, however, that part of the blame probably lies with linguists having failed to spread linguistic learning outside linguistics.
But we’re here waiting to be asked for advice! If more writers and publishers of these books and booklets would only bother to consult a linguist, we would all be a good deal better served. And I’m not talking about filling these books with linguistic jargon (although I strongly support the inclusion of IPA symbols—chosen by people who know what they’re talking about—in all pronunciation guides, because they take up little space, and they’re very useful for people who know the IPA). However, as Wells says, words like “voiceless” and “voiced” should be within the reach of any half-intelligent adult. And you don’t need to be a linguist to recognise that some sort of sensible articulatory description is better by miles than such meaningless nonsense as “hard h”.