This
dictionary is an alphabetized list of words in spelling. As explained above, it
is alphabetized according to the root of the word. Following the word is a
phonetic form of the word, and in a number of cases a
playable recording of the word – see more details about the recordings below.
This is followed by part of speech (n=noun, v=verb, adj=adjective), when the word
is a noun it gives the noun class number, then a translation into English, and
further explanation as appropriate.
The actual prounciation of words varies considerably in the language, therefore many entries are supplemented with playable recordings and a phonetic transcriptions to familiarize you with these variations. The spelled noun ichaage ‘grain storage area’ is accompanied by a phonetic form, ɪcháage which adds vowel quality and tone information based on the dominant pattern in my data. This does not mean that everybody pronounces the word that way, it means that most examples in my data are pronounced that way. In ɪcháage the phonetically-enhanced main entry is followed by transcription of 4 specific pronunciations: ɪcháágɪ, ɪcháage, cháagɪ, cháage. A play-button precedes each transcription and is enclosed in brackets with it. As you can see from this, there can be differences in tone, the presence of the initial vowel, and the choice of the final vowel. These are quasi-random samples of recordings, intended to acquaint a listener with some of the pronunciations of words. There may be multiple recordings from a single speaker either because of a substantial difference in token pronunciations, or because there is little difference, and a listener should get acquainted with ‘the same thing’ said at different times.
Transcriptions are my judgments based on the work that I have done with these speakers. They do not include every minute noticeable detail, and often involve difficult threshold questions. In the recordings of lugada ‘smoking pipe’, it is unclear whether the specific utterances [ʊrʊgáda] and [lʊ́gáda] differ in the tone of the prefix /lʊ/. This is because the H tones before the stem come from a tone assimilation rule which is variable, therefore many actual pronunciations are possible which extend the tone from /gá/, so in [ʊrʊgáda] the H tone may be ‘somewhat’ extended to the left. In many examples, the syllable to the left of the basic H toned syllable is half-way raised relative to the syllable before that. The pitch range of a final syllable varied substantially, and sometimes a speaker will end with a higher than usual pitch just because of recording circumstances, leading to some higher-sounding word-final un-toned syllables. The citation tone of words can be very difficult to determine, since words with no H and words with one H at the beginning of the stem often sound nearly the same, differing only in pitch range and perhaps rate of pitch fall. Through contextual analysis, the difference becomes easy to hear, this is only a problem with citation forms.
As for the r / l difference, clean classification in terms of the International Phonetic Alphabet is impossible. The reason for this is that “l” probably was originally [ɭ] which is a mix of r and l, so anything in that general range is accepted and actually produced. No note is made of the different pronunciations of y, these samples expose you to the continuum of pronunciations, where you may or may not notice the Swahili/English-like version versus the flat-tongue version. In the two pronunciations of ʊmʊgádi ‘bread’, g is pronounced differently, again another variable feature of pronunciation in the language.
Sometimes,
the recordings vary a bit in grammatical form, especially noun class, perhaps
because a speaker only uses the noun in the plural, or applies a different noun
class to the same word (for example people differ in whether to give the word
for “tradition” in the singular or plural: ʊmw-iima vs. im-iima;
‘law’ could either be given in the singular, il-lago or the plural ama-lago).
Borrowed words are especially variable, so if you pronounce the word for
‘loquat’ as ipalapaandi, you may not find the word listed under a root palapaandi,
since it is listed as balabaande which is the more common version (from ilibalabaande).
It
may help to know what the most common patterns of variation are. Many Logoori
speakers are aware that words with p, t, k in
languages such as Tiriki or Nyore correspond to Logoori words with b, d, g
respectively, which is why Swahili posta ‘post office’ is widely
pronounced ebóósta. Some speakers pronounce the word as epóósta. The
direction of change seems to be one-way: p, t, k, ch from outside
Logoori become b, d, g, j but b, d, g, j in a word from outside
Logoori (ɪdííni ‘religion’, from Swahili ultimately Arabic) keeps that
consonant as is.
The
main purpose of the recordings is to give the user, especially one who is not
highly experienced in the language, an idea how words are pronounced and what it
means if I write “ɪmbárabará” versus “ɪmbárabára”. Or, if one personally pronounces
“knife” as [ʊm’banʊ] and you doubt that anyone would say [m’bano], you can listen to a recording of both pronunciations.
There is no implication about what pronunciations are “correct, standard,
typical”, nor even necessarily “typical for that person”.
In spelling, I write l and not r, but usually I phonetically transcribe the sound as [r] when the individual uses phonetic [r] (Swahili-style r, technically, [ɾ] in the IPA), otherwise [l]. Transcribed [l] may be technically pronounced phonetic [ɭ], which is a mix of r and l. See the grammar for more discussion of pronunciation of r, l, ɭ. Many speakers have a pronunciation that varies along a continuum of [ɾ] to [ɭ]. This is only relevant to understanding spelling – always spelled l, transcribed as [r] or [l] depending on speaker. The dictionary does not generally engage in phonetic nuances.