Features of personal and reflexive pronouns #4
Comments
|
@dan-zeman Sorry for the extremely slow response. ;) These are good suggestions that should be straightforward to implement (except possibly Reflex=Yes for other lemmas than "sig"). What is the recommendation for the Person feature? Should it go on all pronouns that have PronType=Prs? Or is it more subtle than that? |
FWIW here's what we do in English: https://universaldependencies.org/en/pos/PRON.html The one unresolved issue with personal pronouns is whether "you" should be marked for number based on context. (GUM does, EWT does not.) |
|
@nschneid Thanks! I think most of that could be adopted directly for the Scandinavian languages (modulo the fact that there is grammatical gender in addition). Currently, however, possessive pronouns are also subsumed under the corresponding personal pronoun for lemmatisation (that is, the correspondent of "my" would have the same lemma as "I" and "me"). In addition, I think Case=Gen is relevant only for a small subset of this, but I think I remember Amir saying that the same is really true for English as well (where only "his" and "her" are real genitives). |
I'm not sure it's written anywhere in the guidelines (and it would really be a recommendation at best) but I would typically expect non-empty |
|
Thanks! This makes perfect sense, and the “exception” you describe seems to be regular given the recommendation not to include features whose value would be the union of all possible features.
From: Dan Zeman ***@***.***>
Reply to: UniversalDependencies/UD_Swedish-Talbanken ***@***.***>
Date: Monday, 20 November 2023 at 17:31
To: UniversalDependencies/UD_Swedish-Talbanken ***@***.***>
Cc: Joakim Nivre ***@***.***>, Comment ***@***.***>
Subject: Re: [UniversalDependencies/UD_Swedish-Talbanken] Features of personal and reflexive pronouns (#4)
What is the recommendation for the Person feature?
I'm not sure it's written anywhere in the guidelines (and it would really be a recommendation at best) but I would typically expect non-empty Person value whenever I see PronType=Prs. That said, I actually know that I break this expectation with Czech reflexive pronouns. I give them PronType=Prs because Reflex is not a PronType, I want a non-empty PronType for all pronouns and this seems to be the closest match – BUT: Unlike Germanic and Romance languages, the Czech reflexive sebe/se/sobě/si/sebou is used in all three persons, so it does not make sense to put anything in the Person feature.
—
Reply to this email directly, view it on GitHub<#4 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ABZ7ZVWO3CIPLJVU4I7CEKLYFOAWXAVCNFSM4F4HFZH2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOBRHE2DCMBQGQZQ>.
You are receiving this because you commented.Message ID: ***@***.***>
VARNING: Klicka inte på länkar och öppna inte bilagor om du inte känner igen avsändaren och vet att innehållet är säkert.
CAUTION: Do not click on links or open attachments unless you recognise the sender and know the content is safe.
När du har kontakt med oss på Uppsala universitet med e-post så innebär det att vi behandlar dina personuppgifter. För att läsa mer om hur vi gör det kan du läsa här: http://www.uu.se/om-uu/dataskydd-personuppgifter/
E-mailing Uppsala University means that we will process your personal data. For more information on how this is performed, please read here: http://www.uu.se/en/about-uu/data-protection-policy
|
Could the pronoun sig be annotated with
Reflex=Yes? Plus, the pronouns mig etc. could maybe be annotated as reflexive when they are attached to a verb whose subject has the same person and number (this is done in German-GSD; an example where I think it would apply in Talbanken is the line 76579 of the train file, current dev branch). Also, maybe the personal pronouns could have thePersonfeature? Thanks. DanThe text was updated successfully, but these errors were encountered: