Fundamental concepts:

Phoneme:

A speech sound

Phonemic:

Said about a sound or group of sounds that is meaningfully distinguished from another sound in a language. For example, English has phonemic (meaning that using a [b] or a [p] can change a word’s meaning), but not phonemic (meaning that using [p] or [ph], without or with a puff of air, does not change the meaning).

Contrast:

Two are said to contrast if they are considered two different sounds in a language. The above example could be phrased as [b] and [p] contrasting in English, while [p] and [ph] do not. Can also be used for grammatical features or words, not just phonemes.

Allophone:

A that is not meaningfully distinguished from another phoneme in a language. For example, aspirated [ph] and unaspirated [p] in English are allophones and do not contrast (i.e. [tap] and [thaph] and all variations in-between are the same word), but they do contrast in Scottish Gaelic and are therefore in that language.

Allophonic vs. free variation:

Technically, allophones are used in different contexts. For example, English has a tendency to use aspirated [ph] at the beginning of words or and unaspirated [p] at the end of words, in or at the beginning of unstressed syllables. This is a true allophone. In some cases, the variation is more random, where even the same speaker might use more than one pronunciation of the same phoneme, like the pronunciations of “either” and “neither” as [’(n)i:ð] or [’(n)að]. This is called free variation, but is often included in allophonic variation as well.

IPA:

The International Phonetic Alphabet, a mostly Latin- and Greek-based way to encode all possible speech sounds in human languages. Contrary to simply using, say, the Latin alphabet, the IPA denotes the exact pronunciation of a sound, whereas the Latin letter “j”, f. ex., stands for very different sounds in English, French, German or Spanish.

Broad transcription:

Includes only the most notable phonetic features, leaving out non- distinctions

Narrow transcription:

Includes more details of exact realisation of each phoneme, even if these are not ; broad to narrow is a spectrum.

Phonemic transcription:

Focuses on the abstract mental distinctions that speakers make, results in a particularly broad transcription that ignores allophones and (most) dialectal differences, i. e. will transcribe a Krotmag speaker using ta Hol.

Phonetic transcription:

A transcription that focuses more on how something is actually pronounced, f. ex. by including dialectal differences, but the level of detail may still result in either a broad or narrow transcription. Phonemic vs. phonetic is a conceptual and therefore binary distinction.

Transcription delimiters:

IPA is usually either marked by slashes // or square brackets []. Technically, slashes denote phonemic notation and square brackets phonetic notation, but the distinction may also be used for broad vs. narrow transcription. Some also use them interchangeably.

Example:

The word “pity” is phonemically transcribed as /pti/, no matter how a speaker pronounces it. A broad phonetic transcription would be [’pti] in RP and [’pi] in GA. A narrow transcription would be [’phti] in RP and [’phi] in AE. A phonemic transcription in AE of the word “hamster” is /’hæmst/, while the phonetic transcription would include the “p” that most speakers insert (see ), so [’hæmpst].

Phonemic inventory:

The full list of used by a language. May be split into and inventory.