Environment Dependant Affix Process Allomorphs

[I. S.]

Question 1

In Classical Arabic, the 𝟥.ꜱɢ.ᴍ perfective of the G-stem of JYʔ is

jāʔ-a

go.ᴘꜰᴠ-𝟥.ꜱɢ.ᴍ

We can create this in FLEx by creating an Affix Process Rule that takes CyC and produces CāC to get the stem.

But before consonant initial suffixes like 𝟤.ꜱɢ.ᴍ we get a different allomorph:

jiʔ-tu

go.ᴘꜰᴠ-𝟤.ꜱɢ.ᴍ

With normal affix allomorphs, we can define environments where they can be used. How can this be replicated with affix processes?

Question 2

Is defining syllable structure dependent (rather then morpheme or word boundary dependent rules) phonological rules possible in FLEx, like closed syllables shortening, a very common rule cross-linguistically? A rule like Vː -> V / C __ C cannot do this because it ignores syllable boundaries. We want something like Vː -> V / $C __ C$.

[Kevin Warfel]

Hi, I.S.

Are you aware of the existence of a FLEx Parsing Google Group that exists for questions specifically about the use of the XAmple or Hermit Crab (HC) parser? This question would be very appropriate there. If you're not yet a member there but would like to be, let me know.

In response to your Question 1, I believe that what you are wanting is not possible in the context of an Affix Process Rule (APR). The equivalent to the Environment field (normal Allomorph) in an APR is the matrix containing X to the left of the arrow. That is where the "context" is specified. This has limitations.

1) A prefix cannot reference any material to its left as part of the conditioning environment.

2) A suffix cannot reference any material to its right as part of the conditioning environment.

(In the context of a RTL script, this version of those statements might be clearer: A prefix cannot reference any prefix that is further away from the stem than itself. and A suffix cannot reference any prefix that is further away from the stem than itself.)

3) A prefix cannot reference any suffix material as part of the conditioning environment.

I may be wrong, but it is my understanding that these limitations are due to the way that HC processes affixes (starting with prefixes and working outward from the stem, then processing suffixes, again working outward from the stem). 

In response to your Question 2, I don't think this is possible either, as there is no syllable boundary marker available for insertion in FLEx, and even if there was, I'm not sure how FLEx would know where the syllables divide. If there is a way to specify a closed syllable via CV patterns, then what you're requesting might be possible, but the way you worded the question gives me the impression that you're not finding a way to do that. Maybe someone else can suggest a way to do what you're wanting, but I've come up empty in my attempt to think of one.

Best wishes,

Kevin Warfel

--
"FLEx list" messages are public. Only members can post.
flex_d...@sil.org
http://groups.google.com/group/flex-list.
---
You received this message because you are subscribed to the Google Groups "FLEx list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to flex-list+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/flex-list/7c2514a8-253f-4a20-a026-4363d269ccden%40googlegroups.com.

[Michael Maxwell]

Just a couple brief comments here, since as Kevin says this may be
better handled in the parsing group.

I can think of a couple ways to handle these issues: you could order the
stem allomorphy rule after the suffixing rule; or you could treat the
two allomorphs using stem names.

As for the syllable issue, I think you could create a syllable boundary
marker "phoneme", epenthesize it at some point in the derivation, and
delete it later. (This might make for slow parsing, however.) Or you
could re-cast the syllable boundary environment as a sequence of two (or
more) consonants, or a consonant followed by a word boundary. That's
essentially what Chomsky and Halle did in 'The Sound Pattern of
English'. (Whether that's correct or not is of course a different
question--a linguistic question--but it's doable. And FWIW, I think the
right question is not whether syllables exist, but whether syllable
boundaries exist. Or putting it differently, whether syllables are
particles or waves--to reference a famous SIL linguist.)

Mike Maxwell

On 12/11/2024 5:09 PM, 'Kevin Warfel' via FLEx list wrote:
> Hi, I.S.
>

> Are you aware of the existence of a FLEx Parsing Google Group <https://
> groups.google.com/u/2/a/groups.sil.org/g/flex-parsing-list> that exists

> <mailto:ishd...@gmail.com>> wrote:
>
> _Question 1_

> In Classical Arabic, the 𝟥.ꜱɢ.ᴍ perfective of the G-stem of JYʔ is
> jāʔ-a
> go.ᴘꜰᴠ-𝟥.ꜱɢ.ᴍ
> We can create this in FLEx by creating an Affix Process Rule that
> takes CyC and produces CāC to get the stem.
>
> But before consonant initial suffixes like 𝟤.ꜱɢ.ᴍ we get a different
> allomorph:
> jiʔ-tu
> go.ᴘꜰᴠ-𝟤.ꜱɢ.ᴍ
> With normal affix allomorphs, we can define environments where they
> can be used. How can this be replicated with affix processes?
>

> _Question 2_

> Is defining syllable structure dependent (rather then morpheme or
> word boundary dependent rules) phonological rules possible in FLEx,
> like closed syllables shortening, a very common rule cross-
> linguistically? A rule like Vː -> V / C __ C cannot do this because
> it ignores syllable boundaries. We want something like Vː -> V / $C
> __ C$.
>
> --
> "FLEx list" messages are public. Only members can post.

> flex_d...@sil.org <mailto:flex_d...@sil.org>
> http://groups.google.com/group/flex-list <http://groups.google.com/
> group/flex-list>.

> ---
> You received this message because you are subscribed to the Google
> Groups "FLEx list" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to flex-list+...@googlegroups.com

> <mailto:flex-list+...@googlegroups.com>.

> To view this discussion visit https://groups.google.com/d/msgid/
> flex-list/7c2514a8-253f-4a20-a026-4363d269ccden%40googlegroups.com

> <https://groups.google.com/d/msgid/flex-list/7c2514a8-253f-4a20-
> a026-4363d269ccden%40googlegroups.com?
> utm_medium=email&utm_source=footer>.

>
> --
> "FLEx list" messages are public. Only members can post.
> flex_d...@sil.org

> http://groups.google.com/group/flex-list <http://groups.google.com/
> group/flex-list>.

> ---
> You received this message because you are subscribed to the Google
> Groups "FLEx list" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to flex-list+...@googlegroups.com <mailto:flex-
> list+uns...@googlegroups.com>.

> To view this discussion visit https://groups.google.com/d/msgid/flex-

> list/CAHs8okXE9xb76vtCfiUOReWt44-
> O9rq8aH92skGqvXPH70nMPA%40mail.gmail.com <https://groups.google.com/d/
> msgid/flex-list/CAHs8okXE9xb76vtCfiUOReWt44-
> O9rq8aH92skGqvXPH70nMPA%40mail.gmail.com?
> utm_medium=email&utm_source=footer>.

[I. S.]

Thank you Kevin and Michael,

I was not aware of the existence of the parsing group, and I would very much appreciate it if you could invite me!

About the first question:

If stem names (by which I assume you mean different lemmas) are is it possible to avoid forms like **jāʔta being parsed as correct along with jiʔta without resorting to ad-hoc rules? I'm mostly asking out of curiosity. Your comment about ordering the stem allomorphy rule after the suffixing rule has also piqued my interest. I was not aware it was possible. How can that be done?

About the second:

Using a syllable marker is a clever idea that I honestly haven't thought of! Deleting it afterwards can be done using a phonological rule (something like $ -> __ / __?) after applying the syllable dependent rules, right?

Regards,

I. S.

[Sarah R M]

@kevin I would also like to be invited to the  FLEx Parsing Google Group  . Please use my work email smoe...@ufl.edu