Talk:Pro-drop language

From Wikipedia, the free encyclopedia

This article is part of WikiProject Theoretical Linguistics, an attempt to build a comprehensive and detailed guide to theoretical linguistics and theories of language on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks.

1 Generalizations across languages
2 False dichotomy?
3 "You want to come with?"
- 3.1 NW Oregon
4 Topic-comment
5 Avoidance of repetition
6 Japanese script
7 English Imperative as prodrop?
8 Intro
9 Non-pro-drop-language centric?
10 Brazilian Portuguese is not pro drop

[edit] Generalizations across languages

The article says: "Portuguese, Spanish, Italian and Romanian can elide subject pronouns only". I'm a Brazilian Portuguese speaker, and we elide object pronouns as well.

In casual Brazilian Portuguese, the example given for Japanese can be translated as follows:

Eng.: "This cake is tasty. Who made it?"

Port.: "Este bolo é muito bom. Quem fez?" (This cake is very good. Who made it?)

Eng.: "I don't know. Do you like it?"

Port.: "Não sei. Gostou?" (I don't know. Did you like it?)

I can't comment on Spanish (and the other languages) because it usually has subtle differences regarding what is usual and what sounds strange.

How could the article be improved? --Yuu en 21:35, 19 November 2005 (UTC)

I'd heard of this. Do you know if this happens in European Portuguese? Spanish certainly doesn't do this kind of thing, for sure, at least not on a regular basis in any major dialect that I'm aware of. Feel free to fix the article with a parenthetical note regarding Br. Port. --Pablo D. Flores (Talk) 22:05, 19 November 2005 (UTC)

-- Thanks for your reply :). I don't know much about European Portuguese. I rembember reading something saying that object pronouns are not dropped, but it could as well be that they are not dropped as often... So i cannot comment on this. --Yuu en 20:55, 17 December 2005 (UTC)

I think this happens as well in Spanish from Spain (in some cases). For example, when talking about a cake someone's offering to someone else:

Eng: "Do you want it?" (yes, I know it sounds weird. It should be "some" instead of "it". But anyway, some word needs to stand after the "want")

Spanish: "¿Quieres?" (literally "Do you want?")

That's not however the case with Catalan or French, where the thing that the someone might want needs to be mentioned:

Cat:"En vols?"

Fra:"Tu en veux?

My guess would be that Spanish might also elide some object pronouns, whereas that's impossible in Catalan (see:Weak pronouns in Catalan) and French. But then again, I'm no native Spanish speaker, just a Catalan who's been forced to learn it, so I might be wrong. What I do know is that it's easy to detect native Spanish speakers amongst people speaking in Catalan (or trying to) because they make the mistake of not using the object pronouns.--80.32.95.147 00:19, 27 February 2006 (UTC)

In colloquial English, we do this, too: "You want?" is quite common. (There seems to be a limit to the cropping, though; while both "You want?" and "Want some?" sound reasonable to me — provided the context is informal — simply "Want?" does not.)

The thing about Spanish — and I must admit here that I'm not a native speaker, or even a particularly good one — is that I'm not sure what pronoun one could include there. AFAIK Spanish has no pronoun corresponding to French en. There are some places where Spanish drops pronouns that could be included for emphasis or other reasons:

Lo quiero. (I want it.) ↔ Yo lo quiero.

But there are others where I don't think any pronoun could be included:

Me gusta. (I like it.) ↔ *Me gusta él. (The latter could only mean I like him.)

Me gusta. (I like it.) ↔ Me gusta esto. (I like this.)

(This makes some sense to an English speaker, since it can't really be stressed in English, so it feels natural that Spanish wouldn't allow this, either, even though in Spanish this means that there's essentially no subject pronoun form of it.) Your example seems to be of the latter type. I'm not sure if there's really an essential difference from a linguistic standpoint, but I think it bears mention. Ruakh 02:19, 27 February 2006 (UTC)

[edit] False dichotomy?

This page seems to be committing a false dichotomy. What about languages that omit dummy pronouns, but which otherwise don't drop pronouns? For example Esperanto. -- Smjg 16:52, 15 Apr 2005 (UTC)

As I see it, those are two different subjects. The definition given first says:

Pro-drop language (from "pronoun-dropping") is a language where pronouns can be elided (deleted) when considered unnecessary or redundant by the speaker.

This definition emphasizes pragmatic usage, that is, the speaker's practice. In Spanish, for example, you can drop your subject pronoun, but you can also keep it. It depends on you (it's not a matter of getting syntax right or wrong, but of sounding redundant or not). Under this definition, Esperanto is not a pro-drop language.

The dummy pronoun thing is an addendum: if a language is pro-drop, then it doesn't use dummy pronouns (since those are always redundant, or nonsensical, by definition). A dummy pronoun is not a "real" pronoun in that it doesn't really refer to anything. It's not the same to drop a "he" as in "he made the cake" and to drop an "it" as in "it rained yesterday" ("it" doesn't mean anything at all, it's just a syntactic requirement of English).

--Pablo D. Flores 13:54, 16 Apr 2005 (UTC)

Exactly. I began to read the article to the effect that only and all non-pro-drop languages use dummy pronouns. And I'm pointing out Esperanto as a counterexample (there are probably many others). Ekz-e, your two examples would translate as "Li faris la kukon" (not "Faris la kukon") and "Pluvis hieraux" (not "Gxi pluvis hieraux") respectively.

Well, looking at it again, it says: "Pro-drop languages deal naturally with these, where non-pro-drop languages such as English and French sometimes have to fill in the syntactic gap by inserting a dummy pronoun." For that matter, what does it mean by "sometimes"? That there are certain conditions in which English or French may omit a dummy pronoun? Or that it applies to some non-pro-drop languages? This should be clarified and maybe expanded on. -- Smjg 15:38, 19 Apr 2005 (UTC)

[edit] "You want to come with?"

I've never heard this. In which lects does it appear? -- Hoary 02:57, 2005 Jun 10 (UTC)

I attend school in Ohio, and hear it all the time. The non-Ohioans make fun of the Ohioans for saying it, so I guess it's not so common elsewhere. (I never noticed it until people commented on it, so I can't say whether people use it where I'm originally from. I simply say, "You want to come?", but I guess some people like their prepositions.) Feel free to revert my change if you think it was better the other way; I won't be offended. Ruakh 22:57, 11 Jun 2005 (UTC)

No, no: it's just that I'd never heard it and found it hard to imagine/believe. But your comment here satisfies me, thanks. -- Hoary 02:08, 2005 Jun 12 (UTC)

I lived in Ohio for a while and I heard this often. Mga 19:06, 13 Jun 2005 (UTC)

It's a midwest thing--I've heard this throughout Chicagoland. Meelar (talk) 19:07, Jun 13, 2005 (UTC)

It also appears in informal speech in California, esp among teenagers. 66.218.46.190 19:11, 14 January 2006 (UTC)

It's definitely a recent development in California. I've lived there all my life, and never heard "Can I come with?" until my teenage kids began using it. In fact, I'm more likely to say "Can I go with you?" or "Would you like to go with me?" February 27

This has been hypothesized to derive from German ("komme mit"). Does it extend to Yiddish speakers in English also? --Dpr 05:47, 12 September 2005 (UTC)

The same construction is common in Swedish (and possibly other Scandinavian languages) as well. I recall it being suggested that it may derive from any of these. EldKatt (Talk) 14:40, 21 December 2005 (UTC)

I've heard of the usage extending as far east as the Pittsburgh area, and the hypothesis that it comes from German has always appealed to me, for one, because a) there were many, many, many German and Swiss settlers in the region from roughly Pittsburgh to Chicago, and b) in German this is a separable verb mitkommen that's usually translated as "to come along". Don't know about the Scandinavian languages. --Darrell Manrique, 12.107.67.3 12:37, 7 June 2006 (UTC)

[edit] NW Oregon

The first time I ever heard "come with" or "go with" was in NW Oregon, where I grew up, probably twenty years ago. It sounded odd then, and it sounds odd now (though it's understandable). There are a couple of issues here:

1. It requires us to end a sentence with a preposition, which in itself isn't so bad (it's something up with which I can put :^) ) if the "object" of the preposition is still expressed somewhere. Unfortunately, here the preposition ends the sentence while its object is lost (well, it's supposed to be "understood," depending on the context).

2. In English, some prepositions can also function as adverbs: He went in seems to end in a preposition, but in can be taken as an adverb modifying went, since in tells us, at least in a vague way, where he went. OTOH, with doesn't seem to work as a preposition, since it provides no adverbial information, such as how, where, or when. 4.243.206.196 02:27, 26 December 2006 (UTC)

[edit] Topic-comment

Are pro-drop traits more common in topic-prominent languages? --Dpr 05:47, 12 September 2005 (UTC)

I don't know if a hypothetical correlation has been extensively researched. But I did find this:

One view: empty objects in certain languages (at least) are variables bound by empty topics, e.g. Chinese (Huang 1984, 1989), European Portugese (Raposo 1986) not pro. Subject-object asymmetries in the distribution of and interpretations open to empty subjects and objects may be captured in this way.

A second view: empty object positions in other languages (e.g. Quechua, Korean, Thai: Cole 1987) may not seem to show the restrictions found in Chinese/Portugese, and so can/should be analysed as occurrences of pro.

Also mentioned here.

In short, it looks like this:

In general, one takes "pro-drop" to mean "null subject" (or "null object"), that is, a syntactic constituent of the sentence is deleted.
However, in some cases one can consider this deleted thing as a topic: first you topicalize something (a subject, object or whatever) and then you delete it. So this could be better termed "topic-drop". This seems to be the case at least in some cases in Mandarin and Japanese.

I get the impression, too, that a topic-prominent language sh/would be naturally more adept at pro-drop, since the basic idea of TP is that you get to say things pragmatically, i. e. with less emphasis in syntactic requirements and more on getting to the point. But this is of course a very general idea of mine. In any case, all TP languages that I know are pro-drop (the opposite is not true).

--Pablo D. Flores (Talk) 11:00, 12 September 2005 (UTC)

This connection has been most clearly suggested in Yan Huang´s "Anaphora" from 2000 (Oxford). He suggests a distinction between syntactic languages (English, German, French) and pragmatic languages (Japanese, Chinese, Korean, Vietnamese), and two of the diagnostics are topic-prominence and zero anaphora. The typology is still quite controversial and not that well known in the general linguistic community, but is definitely on to something. Grape, 26 June 2006

[edit] Avoidance of repetition

We now read: Spanish is pro-drop too, although only in the subject. This is usually to avoid its repetition. How is this avoidance of "repetition" different from the usual pro-drop avoidance of what can be pragmatically inferred? -- Hoary 07:08, 25 November 2005 (UTC)

I think there's a difference. In Spanish you usually do NOT need to pragmatically infer what the subject is, because it's already morphologically marked in the verb. I wouldn't use that explanation. Spanish is subject-pro-drop, period - you don't have to give a cause for the feature. --Pablo D. Flores (Talk) 12:05, 25 November 2005 (UTC)

Of course! How sleepy I must have been when I wrote that. Yes, pragmatics need have nothing to do with it: morphology explains all. But we agree: a desire to avoid repetition has nothing to do with it. Well, I'd normally make a small fix to the article; but because I'm even sleepier now than when I wrote the question above, I'll forgo this. -- Hoary 13:41, 25 November 2005 (UTC)

[edit] Japanese script

I've just removed the Japanese kanji/kana added to the romaji transliteration. This is only an example (not of phonetics, least of all of writing) and Japanese script here is not needed. There was a long discussion about this in the main Japanese language article, no less. The kanji/kana were also incorrect. The topic marker wa is written with the hiragana for ha, and there are common kanji for ki ni itta; however, oishii and dare I've seen more commonly spelt fully in hiragana, おいしい and だれ. Oishii in particular is one of those odd ad hoc combinations that have nothing to do with the historical readings of the kanji in question...

このケーキは美味しい。　誰が作ったの？

知らない。　気に入った？

--Pablo D. Flores (Talk) 14:35, 5 December 2005 (UTC)

Yes, good move. Incidentally, something about t(s)ukutta jarred slightly with my (prematurely fossilized) L2 Japanese, so I asked a native. She told me that it's OK, but rather hints at an assembly of a prefabricated cake, e.g. adding icing to something bought in a store. So I changed it to yaita. (Somebody may now object that this is better translated as "baked"; I think either "baked" or "made" is fine.) -- Hoary 08:47, 6 December 2005 (UTC)

Another question about the Japanese--I only have a small amount of formal training (and a lot of anime), but shouldn't the question end in a "ka"? I'm hesitant to change it based on a single semester's worth of grammar lessons, but I've never heard of "no" as a question particle. -- 69.122.154.184 03:50, 31 January 2006 (UTC)

I'm no Japanese buff myself, but Japanese particles#no says that no can serve as a question marker. Indeed, that article mentions not just ka and no but also kai and ne as question markers (though the four are not all interchangeable, so I recommend seeing that article for details). Ruakh 04:11, 31 January 2006 (UTC)

Haven't heard of kai (this is two years of Japanese, not much). An informal question can end with no particle (only intonation), or with no to show some personal interest; ne is more like a tag question. These are gender-neutral I think, but no without interrogative intonation is typical female speech. 私の二円。--Pablo D. Flores (Talk) 10:28, 31 January 2006 (UTC)

Kai corresponds less to ka than to desuka, I believe. -- Hoary 13:21, 31 January 2006 (UTC)

[edit] English Imperative as prodrop?

The section reads "Nonetheless, subject pronouns are almost always dropped in commands (e.g., Come here)." As these are commands, ie imperative mood, is this relevant? While English tends to use the same word form for imperative and indicative mood, as opposed to some other languages, the lack of subject is the mood marker isn't it? Even French, which as I understand is a decidedly non-prodrop language, would not require a pronoun for an imperative mood, would it? (I do not know any languages outside of Latin and English, hence why these are questions.) --Frick898 19:36, 14 January 2006 (UTC)

French is about as pro-drop as English is: i.e., not very. (There are a few instances where English uses a dummy pronoun where French wouldn't - e.g., English "I find it bizarre that you think that" vs. French "Je trouve bizarre que tu penses ça" - but then, there are some going the other way, too - e.g., English "It is as I thought" vs. French "C'est comme je l'ai pensé.")

Nonetheless, subject-implicit imperatives can be considered an instance of pronoun dropping (or perhaps of null subjects) in English. For example, if I say, "Wash the dishes tonight, okay?" and you reply, "No, you wash them," you're demonstrating that an imperative's subject can be included for emphasis. If you Google "go thou," you'll find a number of instances where "go" is imperative. Most of these are Biblical or obviously drawing (snowclone-style) on Biblical wording, and in seemingly all of them, "thou" is an explicitly stated subject. (In the Bible, I'd presume that the "thou" is being included to make the command singular: either one person is being addressed, or each addressee is expected to obey the command independently. This would explain why this never evolved into the use of "go you" as an imperative: "you" does not provide the same, key information that "thou" does.)

All that said, I think imperatives should have a separate section, not tied to any specific language, as I think that most languages (including French, as you say) allow or require subject pronouns to be dropped in the imperative. Actually, in my (admittedly limited) experience, it seems that pro-drop languages more readily permit the inclusion of subject pronouns in the imperative. (Maybe this is because pro-drop languages take it as more natural for a pronoun to present at some times and absent at others, whereas non-pro-drop languages generally need a specific pronoun-dropping rule in order to drop a pronoun? That's my guess, anyway.)

Ruakh 21:16, 14 January 2006 (UTC)

For what they're worth, a few more English examples:

"You come here!"

"Oh, now you stop that!"

"You just sit right down, now, and tell us all about it."

--Darrell Manrique, 12.107.67.3 12:37, 7 June 2006 (UTC)

Those all sound slightly non-standard to me — they seem more at home in AAVE than in SAE — but I could imagine them. Ruakh 13:39, 7 June 2006 (UTC)

Well, how about "Don't you dare"? --Darrell Manrique, 12.107.67.3 13:15, 8 June 2006 (UTC)

Ooh, now that's a good example! I can't even imagine just "Don't dare." Ruakh 13:20, 8 June 2006 (UTC)

[edit] Intro

The language survey mentions that French is an exception amongst Romance languages for not having pro-drop. This is oversimplifying, as the subjects "je", "tu", "il" etc. are now pretty much universally accepted to be affixes or clitics rather than independent (pronoun) words. Although there is a little room for argument depending on your theory of clitics/your analysis of French, there's not much reason to think that the "je" in "je mange" is any different from the "o" in Italian "mangio": both mark first person, neither is an independent pronoun. Extra evidence for this comes from the fact that (i) these clitics never appear anywhere except attached to a verb, (ii) they cannot be coordinated (*il et elle mange) and (iii) you _can_ put a genuine pronoun in subject position for stress (if only in limited situations; e.g. "lui rejette cette idée" = "HE rejects this idea"), just like in other Romance languages. —The preceding unsigned comment was added by 171.64.42.67 (talk • contribs) .

Heh, I've actually heard the exact opposite viewpoint presented, that "Lui rejette cette idée" is one of the few cases where French unambiguously exhibits pro-drop (in that the ordinary subject, il, is dropped). See e.g. Pro-Drop and Subject (Non-)Recoverability: The Case of Nez Perce, which even uses the exact same sample sentence. At any rate, yes, there is some oversimplification there; feel free to rephrase it, but be sure not to cause the opposite problem of an overcomplex or overly informative lead section. Ruakh 03:58, 11 February 2006 (UTC)

[edit] Non-pro-drop-language centric?

I'm not a linguist and this is my first time to deal with this term. But I think the concept of pro-drop language is non-pro-drop-language centric. As for Japanese, speakers do not delete nouns (grammatically, pronouns are just nouns) since they do not have to be presented by default. Rather, it would be more accurate to say that speakers add nouns when they are not inferable from context or speakers want to emphasize them. Is there any professional analysis from this point of view? --Nanshu 10:51, 24 February 2006 (UTC)

Erm... Leaving Japanese aside (because whether it really has pronouns or just specialized nouns is a problem in itself), the fact is that all languages have "pronouns" of some kind (as well as other "pro-forms"), which means that at some level all human languages "need" pronouns to fulfill some function. In the case of languages that don't mark person/number in the verb, the reason is clear; in others they may be considered as built-in redundancy to reduce confusion, and in others it is clear that they're simply "not needed" in order to understand. The goal of linguistics in this case is to research what languages drop pronouns and in which contexts, not to assume that pronouns should or should not be present. The label "pro-drop language" is incidental; it simply reflects the fact that most linguists are Westerners, speaking European languages which do not drop pronouns much and consider them a basic part of speech. For a Japanese, pronouns are much more "special", since they get by in other ways; if this study had been initiated by a Japanese linguist, s/he would've probably called English a "pro-add language", and English-speaking linguists would be enraged. :) 私の二円。--Pablo D. Flores (Talk) 22:44, 24 February 2006 (UTC)

I think the question makes sense. Japanese doesn't really have pronouns at all. There are plenty of generic nouns you can drop in where other languages might use a pronoun, but there's nothing about these words that makes them grammatically different from other nouns. It's not so much that Japanese drops pronouns, as that it has none and only adds in generic nouns when necessary. Also, although Japanese verbs do not inflect for person, there are numerous verbs that imply specific interactions between people. E.g. "kureru" means "I receive from someone" and is used all the time in conjunction with other verbs to indicate the direction of the action; this makes the use of a noun for "I" completely unnecessary. - Paul D.

I think it might be worthwhile to evaluate here what we wish to classify as pro-drop-ness and what we wish to classify as null-subject-ness (though obviously both articles would have to explain that the terms aren't always used exactly the same ways, that a distinction isn't always drawn between the two, and that not everyone agrees on which languages are either, much less which). I think Japanese might be more null-subject than pro-drop. Ruakh 04:20, 25 February 2006 (UTC)

Hmmm. I have to clarify my idea. In my understanding, the distinction between noun and pronoun is not important here. Rather, the point is that Japanese does not give special grammatical status to subject and object. Maybe it has nothing to do with typological comparison but I think it goes against speakers' recognition. Here is an example.

watashi wa senshū no nichiyōbi ni genkin de honya de hon wo katta. (私は先週の日曜日に現金で本屋で本を買った。)

I bought a book with cash at the bookstore last Sunday.

We can create a verb-oriented structure as follows.

bought / katta
- I / watashi wa
- a book / hon wo
- with cash / genkin de
- at the bookstore / honya de
- last Sunday / senshū no nichiyōbi ni

Subject (I) and object (a book) are obligatory and the rest is optional in English. But in Japanese, each element has equal status. All combinations of each element are grammatically correct: katta ([I] bought [it]). watashi wa katta (I bought [it]). hon wo katta ([I] bought a book). honya de katta ([I] bought [it] at the bookstore). etc. When native speakers speak these sentences, they don't recognize that they drop subject and/or object, just as English speaking people don't think they drop "last Sunday" when saying, "I bought it." --Nanshu 23:47, 1 March 2006 (UTC)

I have added some information about the original "pro-drop" claims made by Chomsky (1981), where Japanese was explicitly not included. I guess the page should be re-edited according to this, but I am not sure how it could best be done. Questions in linguistics concerning pronouns and anaphora are very theory-dependent, and as Ruakh says, there is much conceptual variation and lack of concensus in this area. The literature on pro-drop and zero anaphora is enormous, and rather than editors presenting their own tentative analyses, it might be an idea to include more information on the disagreement and controversies? Nanshu´s point about all elements having equal status in Japanese is another controversial topic. Most linguists would say that subjects and objects have a different status than adjuncts in any language. Just think about topicalization - if you topicalize subjects and objects, the particles are deleted (there is no "watashi ga wa" or "hon o wa"), while all other particles stay attached ("honya de wa") - a pretty clear difference. This is more abstract than what native speakers recognize that they are doing in real time conversation, though. Grape, 26 June 2006

[edit] Brazilian Portuguese is not pro drop

We Brazilians normally use the subject pronoun with the verb, similar to Tuscan dialects of Italian:

Eu sei (I know) [normal]; Sei [know] rare

Eu gostaria (I would like to) [normal]; Gostaria [would like to] rare

Ela viu você (She saw you) [normal]; Viu você [saw you] rare

Trabalhos recentes vêm apontando correlações entre esta mudança do paradigma e o uso cada vez menor de sujeitos nulos no PortuguêsBrasileiro. Destacamos o trabalho de Duarte (1993), que analisa peças de teatro populares dos séculos XIX e XX, mostrando que está ocorrendo uma mudança no PB, no sentido de não mais permitir o sujeito nulo (from the Orbilat page)

Most dialects in Brazilian Portuguese exhibit evidence of an on-going loss of the properties of the null subject (NS) parameter, namely loss of the “avoid pronoun” principle and subject verb inversion (cf. Duarte 1995). Connected with this change, there is also the loss of movement of the clitics to the pre-auxiliary position, resulting in generalized proclisis to the main verb (cf. Cyrino, 1993), with the consequence that these dialects now allow clitics in sentence initial position. The changes are attributed to the loss of the second person tu, a fact that led to the impoverishment of its inflectional system. (from the Ohio State Univ. site)

(In Brazilian Portuguese PRODROP is obligatory only with impersonal verbs: Choveu It rained; Ofenderam vocês (You [all] were offended; with direct addressing is optional: Viu? - Você viu?- Cê viu

CÊ functions as a clitic pronoun: [vo]cê entendeu? - did you understand?; and cannot be used as an object: abandonaram você [but never abandonaram cê] = you were abandoned])