Sometimes, the same sound can be pronounced two (or more) different ways depending on context and where it is in a word. For example, in my dialect of English, the L sound in “lamb” is different from the L sound in “wool”. Specifically, in my dialect, L sounds in the syllable coda [=when they come after the vowel] are pronounced with the tongue pulled back farther than usual: [læm] vs. [wʊɫ].
But that difference doesn’t really “matter”. Yeah, they’re technically two different sounds, but they *act* as if they’re two *versions* of the same sound. For example, there’s no word where, if I swapped one of the two L sounds for another, it would become a different word. [ɫæm] might sound kind of weird, but it’s still recognizably the same word “lamb”. By contrast, you can’t, for example, switch out the T and D sounds this way: “tent” and “dent” are definitely different words.
The “different versions of the same sound” are called **allophones**, and the “overarching sound” they’re both “versions” *of* is called a **phoneme**. So we would say English has a phoneme /l/, which has allophones [l] and [ɫ].
Transcribing something in slashes / / is **phonemic** – it only distinguishes sounds on the phoneme level. Transcribing in brackets [ ] is **phonetic** – you’re claiming to show *exactly* which sounds are being used, or pretty close to it, with all the dialectal and allophonic variations and everything considered. The former is much easier to transcribe and generally sufficient for most purposes. The latter is harder and generally an unnecessary level of precision unless you’re specifically trying to make a point that the former doesn’t capture – like how a phoneme is pronounced differently in two different words or places or dialects.
Latest Answers