t/v pronoun typology

[TzirTzi]

Hello all

My understanding of t/v pronoun systems is that given a particular speaker and addressee, there will be an unmarked choice: t or v. The unmarked choice may be determined by social properties of individuals (rank, age, gender, caste, etc.) or properties of relationships (familial relationships, friendships, etc.). Speakers then have the option to use the marked form to achieve particular discourse effects: marked politeness or formality; marked intimacy or friendliness; rudeness or offhandness; etc.

My question is: is anyone aware of a language (with a t/v distinction) in which one pronoun is unmarked for all possible speaker-addressee pairs? Or alternatively, a system where there is a pronoun which is only there to create particular discourse effects, and is never the unmarked form?

The reason I ask is that this is what I seem to be seeing in the data I'm analysing, and I was wondering how likely it was cross-linguistically. It's a historical variety, so I can't go and ask native speakers' opinions

Thanks!

[Astraios]

I'm guessing the one you're already looking at is English.

[TzirTzi]

No, Old Icelandic .

[Astraios]

In that case, English, kinda. It has unmarked 'you' for all pairs, and always-marked 'thou'. Though I wouldn't really say English "has" a T/V distinction, because 'thou' is always marked and never used in ordinary speech.

[TzirTzi]

I'd agree that English doesn't have a t/v distinction as the you-thou distinction (in living speech outside Yorkshire at least) serves only to mark normal from ritual speech. But OI is pretty different - analyses of different corpora have found it to be a very typical t/v system sensitive to social rank and discourse factors/politeness. What I was wondering is whether the analysis I have in mind for the data I'm looking at (t pronoun always unmarked, v pronoun always explainable as a politeness feature specific to the context of the particular utterance) is at all cross-linguistically likely.

[linguoboy]

What I was wondering is whether the analysis I have in mind for the data I'm looking at (t pronoun always unmarked, v pronoun always explainable as a politeness feature specific to the context of the particular utterance) is at all cross-linguistically likely.

Sounds kind of like the system of contemporary Cajun English.

I feel like some dialects of Spanish are moving in this direction, too, generalising either tú (Spain) or Usted (America) to practically every instance. (I've been told that in some Venezuelan varieties, Usted is used even with children.)

[chris_notts]

You might want to look at the Basque hi vs zu distinction. Because the distinction is lost in the urban dialects I don't know that much about it, but I did read a paper once that had interesting claims like the following:

1. hi (and associated verbal agreement markers) is a marked form for solidarity with the addressee, not a generic informal pronoun. Therefore it has quite a different distribution to Romance tu pronouns which don't particularly require the speaker to feel any solidarity with their addressee

2. It was claimed that husband and wife by convention don't use hi with each other

I'll have to try to find the paper... but the basic point is that I think zu is functionally a default in rural dialects that retain the distinction.

The basque case is messy because there is continuing innovation in the area of 2nd person pronouns. For example, zu used to be a plural pronoun but has now been replaced by zuek in that role (with new 2nd plural verbal agreement messily innovated) and some dialects have other forms like xu and maybe even xuek. Xu is derived from zu via palatalisation, which is used expressively to form diminutives in Basque, and I think is an attempt to form a new less formal pronoun.

[2+3 clusivity]

Question, how do you determine which is marked or unmarked? Is the unmarked one simply the one most often used? Also how do you deal with a system with more than two layers?

[Astraios]

This paper?

[zompist]

That paper is very interesting... I like the idea of a pronoun marking solidarity.

As to the original question-- I'd ask, what happens if you don't make the assumption that one pronoun is unmarked? I'm not sure I buy this description

given a particular speaker and addressee, there will be an unmarked choice: t or v. The unmarked choice may be determined by social properties of individuals (rank, age, gender, caste, etc.) or properties of relationships (familial relationships, friendships, etc.). Speakers then have the option to use the marked form to achieve particular discourse effects: marked politeness or formality; marked intimacy or friendliness; rudeness or offhandness; etc

as applying to the Romance languages I know. My impression is that the vast majority of the time, there's a very clear right answer about which pronoun to use; the area of optionality is small.

[Penelope]

What I was wondering is whether the analysis I have in mind for the data I'm looking at (t pronoun always unmarked, v pronoun always explainable as a politeness feature specific to the context of the particular utterance) is at all cross-linguistically likely.

Sounds kind of like the system of contemporary Cajun English.

I'm Cajun and grew up in Cajun country and I'm not sure what you're talking about. Explain?

[clawgrip]

In Japanese, a lack of any pronoun is the default, and is thus the unmarked form. Avoidance of pronouns of any kind is preferable, and a lacking of pronouns is suitable for any social relationship. This means that any use of a pronoun/name/title is marked, because your choice will always be based on your social relationship with that person. I'm not sure if this fits what you want since the unmarked pronoun is actually no pronoun, but there you have it.

[TzirTzi]

Question, how do you determine which is marked or unmarked? Is the unmarked one simply the one most often used? Also how do you deal with a system with more than two layers?

Well, if you can ask native speakers then they'll probably tell you that there's a particular pronoun choice which they'd expect given a particular speaker-addressee context - and they'll probably be able to tell you that a particular pronoun choice would be rude/patronising/whatever in particular circumstances. That would give you your unmarked and marked options for those circumstances. In corpus linguistics, it's harder... In my case I've been looking at frequency of usage indexed against various social properties of speaker and addressee which might conceivably be relevant and then looking at every single case of v usage and looking for discourse factors which might be relevant (in this case, every single one occurs in a context where face-threatening acts are occurring).

I imagine--but have no expertise in this area--that in systems with more than two pronouns all the same sorts of generalisations apply..

You might want to look at the Basque hi vs zu distinction. Because the distinction is lost in the urban dialects I don't know that much about it, but I did read a paper once that had interesting claims like the following:

This paper?

Thanks! Basque sounds like a very interesting system. I'll look into it more...

That paper is very interesting... I like the idea of a pronoun marking solidarity.

As to the original question-- I'd ask, what happens if you don't make the assumption that one pronoun is unmarked? I'm not sure I buy this description

given a particular speaker and addressee, there will be an unmarked choice: t or v. The unmarked choice may be determined by social properties of individuals (rank, age, gender, caste, etc.) or properties of relationships (familial relationships, friendships, etc.). Speakers then have the option to use the marked form to achieve particular discourse effects: marked politeness or formality; marked intimacy or friendliness; rudeness or offhandness; etc

as applying to the Romance languages I know. My impression is that the vast majority of the time, there's a very clear right answer about which pronoun to use; the area of optionality is small.

Well, if there's a very clear right answer about which pronoun to use, then that's the unmarked one in that situation. It might be that the marked one is very marked (meaning that it's very rude/patronising/overly friendly/formal to use it) - but if there are any situations in which you can imagine the marked one being used with the intention of having such an effect, then that kind of does fit the model. If not, then that's also a known type of system, though not the one I thought applied to Romance languages generally...

In Japanese, a lack of any pronoun is the default, and is thus the unmarked form. Avoidance of pronouns of any kind is preferable, and a lacking of pronouns is suitable for any social relationship. This means that any use of a pronoun/name/title is marked, because your choice will always be based on your social relationship with that person. I'm not sure if this fits what you want since the unmarked pronoun is actually no pronoun, but there you have it.

Interesting, thanks - I'll look into the Japanese system, as it does sound potentially relevant.

[2+3 clusivity]

So, given the social context T or V will be the "default" pronoun and therefore unmarked, but if used in an atypical situation it is marked?

I'm confused. Are you suggesting certain pronouns are always marked, or just marked in certain situations?

[TzirTzi]

So, given the social context T or V will be the "default" pronoun and therefore unmarked, but if used in an atypical situation it is marked?

I'm confused. Are you suggesting certain pronouns are always marked, or just marked in certain situations?

The latter.

Given social context - including, most importantly, the identity of the speaker, the identity of the addressee, and the relationship between them - there will be an unmarked pronoun choice. This will be the choice that the speaker will default to. So, to make up some examples for a fictional variety, perhaps if it's a school pupil talking to a teacher in class, the unmarked pronoun choice will be v, whereas if it's the child talking to a peer, the unmarked choice will be t. There will then always be particular effects associated with choosing to used the marked pronoun instead - such as being rude or being especially polite - and when the speaker wants to create such an effect, they might choose to use the marked form.

That's - as I understand it - the expected analysis for t/v systems.

Whereas in the data I'm looking at, v always appears to be being used with the intent to create such an effect, and no speaker-addressee pair uses v consistently - implying that v is always marked.

Does that make sense?

[2+3 clusivity]

Hmm, okay.

If you have free time and are looking for reading material, I might suggest you read: "Terms of address and second person pronominal usage in Hindi : a sociolinguistic study" by K. S. Misra. It discusses strategies of using three levels of T/V pronouns in nonstandard Hindi in a wide variety of social contexts, eg. sister to sister, employee to employer, in both marked and unmarked senses as you have described it.

[TzirTzi]

If you have free time and are looking for reading material, I might suggest you read: "Terms of address and second person pronominal usage in Hindi : a sociolinguistic study" by K. S. Misra. It discusses strategies of using three levels of T/V pronouns in nonstandard Hindi in a wide variety of social contexts, eg. sister to sister, employee to employer, in both marked and unmarked senses as you have described it.

Thanks, that sounds interesting - I've requested it from my library and look forward to having a read.