Extra column for irregular plurals?

66 views
Skip to first unread message

ast...@gmx.net

unread,
Mar 28, 2015, 3:08:18 PM3/28/15
to flex...@googlegroups.com
I am wondering how to best add the plurals to my flex file.
There are around 14 different plural suffixes in this language, and sometimes 2 or 3 of them would be correct for the same word.
Does it make the most sense to open another column for those plurals? Can I name it "plurals"?
If yes, how is that done?

If not, what are your recommendations how to keep the records of those plurals best?

Thanks for replying soon

David Baines

unread,
Mar 30, 2015, 12:10:50 PM3/30/15
to flex...@googlegroups.com
There are a variety of ways that you can represent plurals in FLEx. 
If the language uses affixes such as -s in English then you would add the affix as an entry in the lexicon. 

If you can come up with a list of noun classes then adding those classes to the grammar area in the Category edit tool is probably the best way to deal with this in flex.   

Or, if there are no rules for how to make a plural from the singular then we could treat each plural as a variant of the singular form. Doing so will create the plural as a variant-entry in the lexicon, as with all variant forms.

Another method of dealing with plurals is to add 'plural' as a custom field. That may well be useful in the analysis stages so that you can sort them and search for them by common features, sort from the end for example to see similar suffixes together.   There may (or may not) be ways of using the bulk edit tools to update FLEx once that analysis is complete. For example the Bulk Edit tools could be used to filter the list of plurals so that only a certain noun-class is showing, then the grammatical category of all those nouns could be changed from 'Noun' to the new noun class.  However, there is no way to bulk edit into the Variant Field.  You can custom fields in the Lexicon Area using the menus Tools->Configure->Custom Fields.
In this case the settings for the custom field would be 
Location: Entry
Type: Single Line Text
Writing System: A vernacular WS

Jeff Shrum

unread,
Mar 30, 2015, 1:00:12 PM3/30/15
to flex...@googlegroups.com

What language family are you working on?  I did some modeling of Nilo-Saharan languages that sound similar.  In many of those languages  nouns fell within classes and were further distinguished by gender.  I created inflectional features for each class, then added each prefix marking it for singular/plural and gender.  The Morphosyntactic gloss builder was a big help in getting the glosses (grammatical info) correct and consistent.  Sometimes the affixes changed because of morphophonemics and I added allomorphs for those.  There were some cases where the noun stem changed when it was plural.  In those cases I created an "irregularly inflected variant".   If you have multiple affixes for plurals, then you will need to create an affix template to show the correct order of those affixes. All this assumes that you plan on using the default parser and not the phonological rule based parser.

 

I personally like creating custom fields for singular and plural and entering those forms in my database.  Even if I don't plan to publish all the plural forms in a dictionary, I like having the data there to double check my analysis later.

 

Jeff Shrum

Language Technology Consultant

SIL-Southern Africa

--
You are subscribed to the publicly accessible group "FLEx list".
Only members can post but anyone can view messages on the website.
---
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 post to this group, send email to flex...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/09e0fb2b-9d37-4f68-a296-7a8ac5c55490%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Astrid Bechtholdt

unread,
Mar 30, 2015, 4:35:44 PM3/30/15
to flex...@googlegroups.com
I am working on an isolated language in South Central Asia. In the meantime I found out that there seems to be some correlation between the plurals and the noun classes in the sense that the same plural affixes usually go with the same noun classes, however one can't say based on the noun class how the plural is going to be. There are 4 noun classes for singular, but some words are plural-only words and therefore would probably need extra noun classes, would make for 7-8 noun classes in total. I also have morphophonemic changes, for the stems as well as for the affixes (some affixes take away the last letter from the stems, others fit in an extra vowel for connection).

My in-between solution is that I added a field in the entry details page on the right side named plural by following the instruction of David Baines - which I thought would launch another vertical column but it didn't. At least I have now a place where I can leave this info. Interesting is also that homonyms can have different plurals accompanying different meanings. I am still finding out how improve my grammar description and to create the affix template - therefore I would like to have more data in order to better decide what is regular and what is variant or allomorph.

astrid bechtholdt
You received this message because you are subscribed to a topic in the Google Groups "FLEx list" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/flex-list/rhinpxzzIZg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to flex-list+...@googlegroups.com.

To post to this group, send email to flex...@googlegroups.com.

maxwell

unread,
Mar 30, 2015, 4:47:03 PM3/30/15
to flex...@googlegroups.com, Astrid Bechtholdt
On 2015-03-30 16:35, Astrid Bechtholdt wrote:
> ...There are 4
> noun classes for singular, but some words are plural-only words and
> therefore would probably need extra noun classes, would make for 7-8
> noun classes in total.

Noun classes typically refer to the affix allomorphs that nouns in each
class take, provided those allomorphs are not phonologically
predictable. If some nouns simply don't have forms for some cells of
the declension, in your case for the plural cell(s), that usually does
not need to be treated as a noun class distinction. Rather, you can
just say that the nouns in question are inherently plural. In other
words, it's a morphosyntactic fact about those nouns, not a class
distinction. At least that's how I'd treat it as a linguist; what it
looks like in a dictionary may be a different matter.

Mike Maxwell
University of Maryland

Jeff Shrum

unread,
Mar 30, 2015, 5:59:15 PM3/30/15
to flex...@googlegroups.com

It sounds like you may be planning to use Fieldworks to make a grammar sketch or use one of the automatic parsers.  You mentioned changes in the stem.  For languages like English accounting for changes in the stem such a "speak" and "spoke" can be handled by creating an "irregularly inflected variant".  In many other European and Afro-Asiatic languages it is necessary to use the "stem name" feature of FLEx because there are a complex declinations that can be attached to the different stems.  This approach was needed to model Amharic.  I am currently using it to model some Portuguese verbs. When you need to work on complex stem allomorphy, look up "insert stem names" in the Help. That should get you started.

 

 

Jeff Shrum

Language Technology Consultant

SIL

 

Jeff Shrum

unread,
Mar 30, 2015, 5:59:15 PM3/30/15
to flex...@googlegroups.com

Astrid,

 

If you want a custom column to appear in the Lexical edit browse view, go to Tools-->Configure-->Columns and pick the name of the custom field for plural off of the list.  This will add your custom column to the browse view.

 

Jeff Shrum

Language Technology Consultant

SIL

 

From: flex...@googlegroups.com [mailto:flex...@googlegroups.com] On Behalf Of Astrid Bechtholdt


Sent: Monday, March 30, 2015 3:36 PM
To: flex...@googlegroups.com

Astrid Bechtholdt

unread,
Mar 31, 2015, 7:33:43 AM3/31/15
to flex...@googlegroups.com
Jeff,

Thank you, that is helpful. I will wait a bit until there is more data.

astrid


On 3/30/2015 10:58 PM, Jeff Shrum wrote:
Reply all
Reply to author
Forward
0 new messages