aspiration in source data about dev

[bambooforest]

Are you assigning me this issue? If so, please assign and I'll take care of
both the segments in the original resources and the features in the feature
look up table.

On Thu, Apr 16, 2015 at 3:39 AM, Daniel McCloy [email protected]
wrote:

There should never be a standard aspiration diacritic ʰ on a base glyph
that is voiced. All such instances of ʰ in the data sources should be
converted to ʱ. The features table will need to be updated as well, to
make sure that the revised phonemes are still getting assigned a feature
vector.

—
Reply to this email directly or view it on GitHub
#73.

[drammock]

done. FYI, you can also self-assign from the right-hand column of the issue page.

[bambooforest]

but i thought maybe you wanted to take this on ;)

On Thu, Apr 16, 2015 at 10:14 AM, Daniel McCloy [email protected]
wrote:

done. FYI, you can also self-assign from the right-hand column of the
issue page.

—
Reply to this email directly or view it on GitHub
#73 (comment).

[bambooforest]

@drammock please advise regarding:

cɕʰ ok shouldn't this be t̠ɕʰ ?

for this I get three rows all from UPSID:

nbf upsid cɕʰ
gio upsid cɕʰ
cnh upsid cɕʰ

where we have the rows:

416 cɕʰ cɕʰ cC,h voiceless aspirated palatal sibilant affricate corrected from cʃ̠ʰ
417 cɕ cç cC, voiceless palatal sibilant affricate corrected from cʃ̠

in the correspondences file:

https://github.com/phoible/phoible/blob/master/data/UPSID/UPSID_IPA_correspondences.tsv#L417-L418

[drammock]

If it's "palatal", it should be cç
if it's "alveolopalatal", it should be t̠ɕ

On Fri, Apr 24, 2015 at 5:58 PM, Steven Moran [email protected]
wrote:

@drammock https://github.com/drammock please advise regarding:

cɕʰ ok shouldn't this be t̠ɕʰ ?

for this I get three rows all from UPSID:

nbf upsid cɕʰ
gio upsid cɕʰ
cnh upsid cɕʰ

where we have the rows:

416 cɕʰ cɕʰ cC,h voiceless aspirated palatal sibilant affricate corrected
from cʃ̠ʰ

417 cɕ cç cC, voiceless palatal sibilant affricate corrected from cʃ̠

in the correspondences file:

https://github.com/phoible/phoible/blob/master/data/UPSID/UPSID_IPA_correspondences.tsv#L417-L418

—
Reply to this email directly or view it on GitHub
#73 (comment).

[drammock]

linking this issue to #84

[bambooforest]

I've tracked these down in the UPSID segment / feature set:

cɕʰ
cɕ
cɕʼ
ɲcɕ

which are all:

pal = TRUE
alv = FALSE
palalv = FALSE

So I assume this OK.

[drammock]

not OK. If these are palatal, should be c-cedilla, not c-curl.

On Sat, Apr 25, 2015 at 5:23 PM, Steven Moran [email protected]
wrote:

I've tracked these down in the UPSID segment / feature set:

cɕʰ
cɕ
cɕʼ
ɲcɕ

which are all:

pal = TRUE
alv = FALSE
palalv = FALSE

So I assume this OK.

—
Reply to this email directly or view it on GitHub
#73 (comment).

[bambooforest]

Ok. Please advise on how to move forward. I've extracted all features from
both PHOIBLE and UPSID for the affected segments -- emailing it to you.

On Sat, Apr 25, 2015 at 11:32 AM, Daniel McCloy [email protected]
wrote:

not OK. If these are palatal, should be c-cedilla, not c-curl.

On Sat, Apr 25, 2015 at 5:23 PM, Steven Moran [email protected]
wrote:

I've tracked these down in the UPSID segment / feature set:

cɕʰ
cɕ
cɕʼ
ɲcɕ

which are all:

pal = TRUE
alv = FALSE
palalv = FALSE

So I assume this OK.

—
Reply to this email directly or view it on GitHub
#73 (comment).

—
Reply to this email directly or view it on GitHub
#73 (comment).

[drammock]

ahhh. Palatal sibilant fricatives/affricates. For UPSID, we decided long ago that "palatal non-sibilant" was c-cedilla, but "palatal sibilant" was c-curl. So the preceding stop should become retracted t. Sorry that I missed the word "sibilant" earlier in the issue thread.

[bambooforest]

Before I update, I did a quick and dirty query

select LanguageCode, Source, Phoneme from phoible-phoneme-level-no-trump.tsv where Phoneme like '%ɕ%'

and get this (minus the lots of rows that return simply <ɕ>:

phl ph tɕ
pol ph tɕ
aka gm tɕ
mij gm tɕ
cav ph tɕ
jya ph tɕ
pbr gm tɕ
tsj ph tɕ
kig ph tɕ
cmn ph tɕ
hrv ph tɕ
afu gm tɕ
kxm ph tɕ
buu gm tɕ
bdh gm tɕ
kmw aa tɕ
aja gm tɕ
bod ph tɕ
afo gm tɕ
lno gm tɕ
knc aa tɕ
krs aa tɕ
iri gm tɕ
pww ph tɕ
aka gm tɕᶣ
afu gm tɕᶣ
cko gm ɕᶣ
aka gm ɕᶣ
bod ph tɕʰ
kxm ph tɕʰ
cmn ph tɕʰ
jya ph tɕʰ
tsj ph tɕʰ
pww ph tɕʰ
knn upsid cɕ
cnh upsid cɕ
tel upsid cɕ
teh upsid cɕ
nbf upsid cɕ
gio upsid cɕ
brv upsid cɕ
nbf upsid cɕʰ
gio upsid cɕʰ
cnh upsid cɕʰ
gio upsid ɲcɕ
iri gm tɕʷ
kjq upsid ɕʼ
snk aa ɕː
teh upsid cɕʼ

UPSID could be updated as suggested:

cɕʰ
cɕ
cɕʼ
ɲcɕ

t̠ɕʰ
t̠ɕ
t̠ɕʼ
ɲt̠ɕ

But anything else that <tɕ> all be retracted and the rows updated in the features file?

[bambooforest]

New issue created for this last comment about t and curl. See issue #89

[drammock]

Closing because #90 is merged.