New Lexical Relation: Distributve plural and Iterative plural

[Rogers Katelem]

I'm having problem with making a new Lexical Relation to behave in a certain way.

Data:

root – lap

distributve plural (dpl) - làak

iterative plural (itpl) - lilaba

The fact:

"Làak" is distributive plural of "lap", but "lap" is not the distributive plural of "làak."

Similarly, "lilaba" is the iterative plural of "lap", but "lap" is not the iterative plural of "lilaba."

What I want:

1. The entry "lap" in the dictionary, should show the dpl "làak" and itpl "lilaba."

2. The entry "làak" should show "làak" as "dpl of lap." [It should not show "làak" is dpl of "lap" because it is not].

3. The entry "lilaba", should show "lilaba" as "itpl of lap." [It should not show "lilaba" as itpl of "lap" because it is not].

Question:

How do I create this kind of Lexical Relation?

[Beth-docs Bryson]

There are several kinds of lexical relations.  There are two varieties of “pair” relations.  One of them is a “pair - 2 relation names”.  That will allow you to use one label for one direction of the relationship, and a different label for the other direction.

I’m assuming this relationship applies at the entry level rather than the sense level (since it seems to be about the forms rather than about the meaning—I’m assuming that different senses of the same form would not have different plurals?).  In this case, you want to choose “Entry Pair - 2 relation names”.

-Beth

--
"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 on the web visit https://groups.google.com/d/msgid/flex-list/CAMutP%3DcDNicb7s4SUh-qA4S44kEpnGnWC%2BU27WkD7%2B2Bbpykvw%40mail.gmail.com.

[Kevin Warfel]

Rogers,

 

Are you aware of the online Lexicography and Dictionary-Making course (https://sites.google.com/sil.org/dls-course/) that our department, Dictionary & Lexicography Services, is working on? There are elements that we’ve not yet dealt with in the course, but the content is already quite extensive.

 

In the intermediate-level course (LEX201), there are lessons that explain the different underlying structures that lexical relations can have, as well as how to create custom lexical relations for situations like yours.

 

 

This course is intended as a help for developing lexicographers and FLEx users of all ages and levels of experience. Please let us know how it helps (or has helped) you, and take advantage of the feedback form at the end of each lesson to suggest ways in which you feel it could be improved.

 

Best wishes,

 

Kevin Warfel

Associate Dictionary & Lexicography Services Coordinator

Rapid Word Collection workshop consultant

--

[Rogers Katelem]

I'm aware of the course. I've read through the topics on Lexical Relations but I couldn't figure out how to set up this kind of relation. Maybe I will read through the topics again, and give feedback.

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/8d54edd7d379ad1d09706276b5a74b3a%40mail.gmail.com.

[Rogers Katelem]

Your suggestion is helpful. The relation is an Entry/Sense relation. I noticed that my problem was in the area of selecting "Reference set type."

But as I fix that, I discovered another issue: It seems I will need a relation that has 3 relation names because the distributive plural and the iterative plural belong to the same verb stem (the singular imperative). If I set up two "Entry/Sense Pair - 2 relation names" relations, I end up having four items instead of three. And when I go to "configure dictionary" menu, I can't tell which is which between the two "Imperative singulars." See attached image.

Is there any way to set up a custom lexical relation that has 3 relation names, or is there any other way to handle this?

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/86BED739-582F-427B-BC84-831C76E97183%40sil.org.

[Beth-docs Bryson]

When using an Entry/Sense relation, you want to be sure that you really do need the relation to map sometimes to entries and sometimes to senses, or sometimes from entries and sometimes from senses.  It can be easy to get tangled up when allowing all of those combinations.  Unless you really need it, it's better to choose either an Entry relation, or a Sense relation.  The determining factor is:  "is it possible for different senses to have a different value for this relation?"  If so, then it is a Sense relation, not an Entry relation.

If you are using Lexical Relations to link these entries, it sounds to me like you need two relations, one for the Iterative Plural and one for the Distributive Plural.

However, as I'm thinking about it more, I'm a little surprised that you are using Lexical Relations for this.  It suggests that you have the plurals as independent entries, rather than as Variants.

Normally plurals don't have an entire full entry, since the only difference in meaning is a piece of grammatical information.  It is more common to list them as Variants (and you could have a Variant Type for each one).  In this case they would get a Minor Entry, so users can find them according to their own spelling.  However, if the spelling isn't very different from the plural and the singular, you may just want a custom field for each.  Then there would be no Minor Entry for the plural forms.

Maybe you have them as separate entries because you want to include the definition and example sentences and notes, etc--that would be a reason not to treat them as a Variant.  However, even in this case, you still might consider treating them as a Complex Form (with the potential to display as a Subentry), rather than a completely separate entry.  That's not typical for Plurals, but it is certainly possible.

You can treat them as separate entries and use Lexical Relations to link them, but it would be good to have a clear reason for why the other approaches aren't suitable for them.

-Beth

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/CAMutP%3Dc%3D4d0nOiQ5erUk5_8rzkWbF_HUQLUFg_tL%3DhJgiSkHMw%40mail.gmail.com.

[Rogers Katelem]

Please see this JWAL article: 

https://www.journalofwestafricanlanguages.org/index.php/downloads/download/75-volume2601/360-the-category-of-number-in-obolo-verbal-morphology

Also available in SIL archive at:

https://www.sil.org/resources/archives/34977

If you can just read through the first three pages and then glance through the appendixes at the end of the article, I think that will give you an idea of what I'm trying to do. Please when you are done looking at the article, kindly come back here and help me.

Thanks.

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/CAEj%3DFHr93ERDOSbzijLSgiXZFuLT2Zq%3DYeXQ_KhOdq%2Bb8EkVdA%40mail.gmail.com.

[Andreas_Joswig]

Hi Rogers,

Having looked at Uche Aaron's article, I can affirm Beth's advice that lexical relations are not the best tool to accomplish what you want. Variants are what you want to employ here to best model the relationships between the various forms. Create variant types (in the list area of FLEx) for distributive and for iterative verbs, and create variant entries for each distributive and iterative forms that you have in the language. Then link them to each base (singular) form by using the variant field, and by assigning the correct variant type. In the cases where there is no singular form, you can make the more frequent iterative form the base form.

Warm greetings,

Andreas

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/CAMutP%3DfiMV9Qghpm6%2B1dp55hB0OfxffutqqkTcw-%2BOAwa%2BkCzQ%40mail.gmail.com.

--

Andreas Joswig, PhD

Senior Linguistics Consultant

Hambühren, Germany / Addis Ababa, Ethiopia

+49 1512 3581686 (Ger) or +251 910 497594 (Eth)

[Rogers Katelem]

@Andreas_Joswig

Sir, I've tried variants but it doesn't solve the problem because there's no option to specify sense when using variants to link two items.

For example: "lap" has many senses, but only one of the senses has the distributive plural "làak", and even this "làak" has many senses, but only one sense refers back to "lap." [Even "làak" itself, when the tone mark is removed, it becomes another verb "laak" (sleep) which has a different distributive and iterative plural.]

That the plurals themselves have multiple senses, is not limited to the verb lap/làak. That is the case with many other verbs. The meanings listed in the JWAL article are just the primary meaning.

Help, please.

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/d4c1fa1e-382a-4670-a4d2-2058b5e247fa%40sil.org.

[Andreas_Joswig]

Dear Rogers,

It is possible to link a variant to a particular sense of the main entry. The following is the dialogue where you assign a variant:

In the red-circled are, check "specific sense", and then choose the sense of the lexeme that you want to connect.

It is more difficult to make the link to just one sense of a variant. It would be my recommendation to have the variant as a separate entry from the other senses. By default, if you create a variant, it does not have a sense to begin with, but it inherits all its information from the main entry. The other uses of the same morpheme you can then have as different senses of a different entry. I hope this works for you!

Andreas

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/CAMutP%3DcMFL9xk-2viS5GDDPjc%2B_x3pD00UtJf4vxs57GtvGTug%40mail.gmail.com.

[Rogers Katelem]

Sir, what version of FLEx are you using? I'm using 9.1.25.877, and in this version I can't find any option to add variant to a sense; that option is available only at the entry level. So one can add variant from entry to sense, but not from sense to entry, or sense to sense. On my own version of FLEx, this screen you have shown is available only when linking entry to sense.

I need a way to link one verb to two plurals (distributive and iterative plural), and the linking can be entry to entry, sense to sense, entry to sense or sense to entry depending on the meaning and usage.

Your suggestion to list the variant as a separate entry doesn't really go down well with me because that will create so many homographs (between 94 to 223 plus) which in fact are not homographs.

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/c3b522dd-bda3-43d8-8a1a-4acd6f3ab7e5%40sil.org.

[Beth-docs Bryson]

It might be worth setting up a consulting session with either Andreas or Kevin, or possibly someone else.  It seems there are many factors that can best be explored interactively in a live session.

I will comment on a couple things though:

 - When creating a Variant of a Sense, it is only possible to create the link FROM the Variant itself, not from the entry that it is a variant of.  So you need to go to the entry for the plural form, and create a link from there to the main form.

 - When you are in the Iterative Plural form, you would choose the variant type "Iterative Plural".  When you are in the Distributive Plural form, you would choose the variant type "Distributive Plural".  Both can be linked to the same verb, but there would be different relation types.

 - When telling FLEx that something is a variant, it is the whole entry that is the Variant.  It is not possible to say that only one sense of something is the variant.  This is because the variant relationship relates to the FORM of the word.  Things related to the form are usually at the entry level, and things related to the meaning are usually at the sense level.  This is not always true, but it's a general principle.

 - If you have a case where only one sense of something is the variant, it may be that you will want to extract that sense into its own entry, separate from the other senses.  Then you will be able to say that that isolated entry is a variant of something else.

But this is only part of the puzzle--there are clearly a lot of linguistic factors that aren't coming out in the short space that is possible in an email.  I would want to explore more about your situation where you want to make links to/from both entries and senses--there is clearly more to the picture.

And your concern about homographs is an important one as well.  It would be good for someone to look with you to understand that better, and figure out the best approach.

-Beth

To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/CAMutP%3Ddm%3DPfOvRDzF6pTmjCPXwzqkM_-2VEOpsqk-7TaHkHy-w%40mail.gmail.com.

[Beth-docs Bryson]

In this video:

FW-Variants 01-Introduction

if you go to about 6:20 in the video, there is a demonstration of how to be in an entry that should become a variant, and link it to a different entry.  There are several steps, but it will eventually take you to the dialog for choosing what it is a variant of, and in that dialog, you'll see the option for linking it to a sense, even though I didn't use that option in that specific demonstration.

-Beth