[Lingtyp] Definition of “personal pronoun" Back to basics

Fri Jul 9 09:04:18 UTC 2021

Dear all,

At this point of the (very general) discussion, I think it can be helpful for Ian to go back to one or two original mistakes or confusions he made in the way he started the discussion. These pertains essentially to the disctinction between morphosyntactic categories and semantic categories or classes.

1) Ian, "personal pronouns" cannot be simply assimilitated to or constrasted with "demonstratives" and "interrogatives" (or indefinite pronouns).
cf. "But then again, demonstratives, interrogative, and indefinite pronouns also refer to the 3rd person".

Demonstratives or interrogatives refer each to some classes of terms sharing some specific (and distinctive) semantic properties (deictic pointing vs. interrogative value) which are absent from personal pronouns, even for the 3rd (and only problematic) "person".

Moreover, these terms (demonstratives and interrogatives, and also indefinite) do not refer by themselves to morphosyntactic categories: e.g. demonstratives (as well as interrogatives) have pronominal vs. modifying (i.e. modifying a noun) and also adverbial USES (there, where ?), which means that they can be used in different morphosyntactic categories.
NB. Depending on the language, a same or a different form is used in these different syntactic uses:
e.g. same form in Wolof:
xaj 'dog', xaj boobu 'this dog (mentioned previously)', boobu 'this (previously mentioned) one' (talking about a dog)
e.g. different forms in French:
chien 'dog', ce chien 'this dog', celui-ci 'this one'

So demonstrative pronouns (and others) always refer to a "3rd person" in a broad sense of a non-dialogic person or referent but they are not personal pronouns, their semantics is different: the access to the referent is construed in different ways in each case (more simple or direct in the case of the personal pronoun).

2) You again mix up referential value and morphosyntactic categories in your answer to Martin:
But as for (b), Korean can express (i) with any noun:
(b) "Does mom/dad/brother/Ian(i) think that mom/dad/brother/Ian(i) has an answer?"
So that would classify any noun as a pronoun.

Of course not. You cannot base your definition of a morphosyntactic category on the translation of a construction from one language to another. As Guillaume Segerer reminded us : a pronoun (be it personal or demonstrative or anything else) stands for a noun (it is a pro-noun), it is NOT a noun.
That is : a pronoun lacks all the semantic richness of the lexicon (what I have called the 'semantic depth') and also often displays different syntactic properties. It is a discursive index targeting an element (previously expressed by a noun) in the discourse. That is why the same pronominal form (except for gender agreement when required) is used to refer to all kinds of nouns (e.g. mom, dad, brother, Ian....).

In the type of construction above, German use a personal pronoun but Korean repeats the noun. This does not mean that all nouns qualifiy as pronouns but that Korean does not use pronouns in this constructions, and maybe does not 3rd person pronouns (?).

I hope this helps.

Best

Stéphane

Stéphane Robert
CNRS-LLACAN
https://llacan.cnrs.fr/p_robert.php
________________________________

Gesendet: Dienstag, 06. Juli 2021 um 11:03 Uhr
Von: "JOO, Ian [Student]" <ian.joo at connect.polyu.hk>
An: "lingtyp at listserv.linguistlist.org" <lingtyp at listserv.linguistlist.org>, "Martin Haspelmath" <martin_haspelmath at eva.mpg.de>
Betreff: Re: [Lingtyp] Definition of “personal pronoun"
Dear Martin,

thank you for your definition.
But as for (b), Korean can express (i) with any noun:
(b) "Does mom/dad/brother/Ian(i) think that mom/dad/brother/Ian(i) has an answer?"
So that would classify any noun as a pronoun.
The difficulty of defining a personal pronoun seems to suggest that it’s not a good category to begin with. Perhaps “definite pronoun”, including “personal pronouns” and demonstratives, would be a clearer category? It would be typologically more meaningful since many languages don’t distinguish demonstratives from (3sg) personal pronouns.
I’m trying to make cross-linguistic matrices of personal pronouns (see below), and for the moment I’m including demonstratives in the matrices.

[cid:E2530C591C8E49D9AB00BD45CD312A03]

Regards,
Ian
On 6 Jul 2021, 4:49 PM +0800, Martin Haspelmath <martin_haspelmath at eva.mpg.de>, wrote:
Maybe the following will work:

"A personal pronoun is a free form that (i) denotes a speech role (speaker/producer and/or hearer/comprehender) OR that is used as an anaphoric form AND (ii) that can be used in a complement clause coreferentially with a matrix clause argument."

This is a disjunctive definition that brings together locuphoric forms ('I', 'we', 'you') and 3rd-person anaphoric (or "endophoric") forms, following the Western tradition (but not following any kind of compelling logic).

It seems that personal pronouns need to be delimited from three types of somewhat doubtful forms:

– person indexes (I do not include bound forms under "personal pronoun" here, following my 2013 paper on person indexes: https://zenodo.org/record/1294059)
– demonstratives
– titles like "Your Majesty"

I think that if a language has a form like "that-one" or "your-majesty" that can be used coreferentially in a complement clause, one will regard it as a personal pronoun:

(a) "My sister(i) thinks that that-one(i) has an answer."
(b) "Does your-majesty(i) think that your-majesty(i) has an answer?"

In German, the polite second-person pronoun "Sie" (which has Third-Person syntax) can be used in (b), but the demonstrative "die" can hardly be used in (a), so it would not count as a personal pronoun (yet). However, in Hindi-Urdu and Mongolian, as mentioned by Ian, the demonstrative can be used in this way (I think), so it would count as a personal pronoun.

I don't think we need the general notion of "person" to define "personal pronoun". Wikipedia's current definition is therefore quite confusing (https://en.wikipedia.org/wiki/Personal_pronoun).

Thanks for this interesting challenge, Ian! It seems to me that quite a few of our traditional terms CAN be defined, but their definitions are not obvious at all (and the textbooks don't usually give the definitions).

Best,
Martin

Am 06.07.21 um 06:53 schrieb JOO, Ian [Student]:
Dear typologists,

I’m having a hard time trying to find a definition of a “personal pronoun”.
One definition is that a personal pronoun refers to a literal person, a human being. But then again, non-human pronouns like English it are also frequently included as a personal pronoun.
Another definition seems to be that “personal” refers to a grammatical person and not a literal person. Thus, it refers to the (non-human) 3rd person, therefore it is a personal pronoun.
But then again, demonstratives, interrogative, and indefinite pronouns also refer to the 3rd person. (This is a book, who is that man, anything is possible) Then are they also personal pronouns?
What’s the clearest definition of a personal pronoun, if any?

>From Hong Kong,
Ian

<http://listserv.linguistlist.org/mailman/listinfo/lingtyp>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20210709/11c93e8b/attachment.htm>