Different languages of the same grouping often have similar sounds. This can give the family a distinctive 'feel'. While trying to sort out the aesthetics of a phonology, I found myself wanting to compare some groupings against each other. Since I couldn't find a paper with my desired data tabulated for me, I decided to do it myself. It turned out to be rather informative and helped guide some of my decisions. As a "hello" to the board and a first post, I thought I'd share the results.
Link to the Google Spreadsheet of Phoneme Frequencies
Disclaimer: Much of this can't be taken as statistically significant and this compilation it isn't scientific in any way.
I've ranked the top 100 phonemes from PHOIBLE against a number of language groupings from the the database of Eurasian phonological inventories. The fraction of languages within that grouping which contain the phoneme is given and heat-mapped. The over-all PHOIBLE frequencies are also superimposed on an IPA chart representation.
Phoneme Frequencies Spreadsheet
Re: Phoneme Frequencies Spreadsheet
Thank you, I bookmarked it. I can have a lot of fun with this.
I agree with you, though, that it's not scientific or at least that there's room for debate on some of these. One thing that caught my eye right away is that they seem to believe icelandic has phonemic /ɲ/. I dont speak Icelandic at all, but Wikipedia considers it just an allophone of /ŋ/ and also rejectes the voiceless nasals. Looking through the rest of the chart I see examples of phonemes that could well be correct but would be rejected by other people's analysis.
I agree with you, though, that it's not scientific or at least that there's room for debate on some of these. One thing that caught my eye right away is that they seem to believe icelandic has phonemic /ɲ/. I dont speak Icelandic at all, but Wikipedia considers it just an allophone of /ŋ/ and also rejectes the voiceless nasals. Looking through the rest of the chart I see examples of phonemes that could well be correct but would be rejected by other people's analysis.
Sunàqʷa the Sea Lamprey says:
Re: Phoneme Frequencies Spreadsheet
That spreadsheet says that occurs in only two Celtic languages, it occurs in Welsh, Cornish, Breton and Manx. Four. But for some reason the spreadsheet is only counting four of the six Celtic languages anyway...
My conlangery Twitter: @Jonlang_
Me? I'm just a lawn-mower; you can tell me by the way I walk.
Me? I'm just a lawn-mower; you can tell me by the way I walk.
Re: Phoneme Frequencies Spreadsheet
Jonlang wrote:That spreadsheet says that occurs in only two Celtic languages, it occurs in Welsh, Cornish, Breton and Manx. Four. But for some reason the spreadsheet is only counting four of the six Celtic languages anyway...
The data is pulled from the database of Eurasian phonological inventories. The database is a beta version and there are a number of contended points, as Soap pointed out, regarding how the curators have chosen to incorporate the inventory data. Nevertheless, I've used the data without altering it or adding to it. The database is not a complete listing of all the languages of Europe and Asia. Cornish and Manx are simply absent. The database contains phonologies for one dialect each of Irish, Gaelic, Welsh, and Breton. is listed for Welsh and Breton. As noted, Irish has secondary articulation, so for Irish, [bˠ] and [bʲ], not , are given in the database and therefore no hits come up on the exact phoneme search for . (and, of course, Gaelic doesn't contain voiced stops in any event.)