Talk:Null subject language

From Wikipedia, the free encyclopedia

Contents

[edit] Japanese requiring an expressed topic

The article states:

Japanese and several other null subject languages are topic-prominent languages: they require an expressed topic in order for sentences to make sense.

Although Japanese is topic-prominent, sentences definitely don't require an expressed topic to make sense. For one, the topic is always omitted when it is the same as the subject, and in other situations the topic, subject, and object can be, and frequently are omitted entirely, whenever they are expected to be implicitly understood. I will try to reword this section. LeeWilson (talk) 05:57, 11 March 2008 (UTC)

[edit] Pro-drop parameter, parameter settings in the brain

this article should also include stuff about chomsky and "pro-drop parameter" and "parameter settings in the brain" Bogdan | Talk 19:03, 20 Oct 2004 (UTC)

Agree. - FrancisTyers 17:01, 3 December 2005 (UTC)

[edit] Pro-drop languages

I think there's considerable overlap (and confusion) between "null subject" and "pro-drop". Modesty aside, I think the coverage of the subject in pro-drop language is fine, though of course this is not exactly the same phenomenon. I'd prefer to see these articles merged into one. In any case, I'm putting this up for attention. --Pablo D. Flores 14:30, 26 Jan 2005 (UTC)

I see what you're saying, but I do think a language can be one but not the other. Esperanto, for example, is non-pro-drop, but it seems that it is null-subject, since impersonal sentences do not require a subject. Conversely, it seems that a language that allowed some pronouns to be dropped, but never subject pronouns, would be pro-drop, but non-null-subject. I think the articles should simply do a better job explaining the difference. Alternatively, perhaps it should be explained that "null subject" and "pro-drop" are two different ways of looking at what is usually the same phenomenon: while the latter sees pronouns being dropped, the former doesn't see pronouns having been there to begin with. Also, perhaps it is better to say that one language can be more null-subject or pro-drop than another, rather than to attempt to categorize languages as either null-subject or not. Ruakh 05:39, 26 May 2005 (UTC)

I agree that this article needs to do a better job explaining the difference between null-subject and pro-drop. As of now, I don't quite get how this is in practice. Also, note that all examples in the article are in fact examples of pro-drop, leaving me wondering what other cases would look like. EldKatt (Talk) 11:39, 21 December 2005 (UTC)

Pro-drop is defined as the possibility (or the idiomatic obligation) of leaving out pronouns when pragmatically inferable. That's something completely different from the null-subject definition, which doesn't mention pragmatics at all but just syntax; and this article also mixes 1) null-subject as an example of pro-drop, and 2) null-subject as the alternative to dummy pronoun usage with semantically impersonal verbs. Then we had a lengthy discussion months ago about those, and settled on a NPOV compromise... but I still think it in it's raining is truly meaningless and an artifact of syntax. I suggest everybody (including myself) gets well into the subject and then we can discuss, so that we don't have neither overlap nor contradiction among articles. --Pablo D. Flores (Talk) 13:52, 21 December 2005 (UTC)

[edit] Slovenian language

I'm a Slovenian, and based on my understanding of null subject languages, Slovenian is one too.

Example:

I am eating.

Jem.


You are eating.

Ješ.


The verb agrees with subject in person and number, and if the subject is omitted, the verb itself signifies who is the object (in a given contex).

Compare:

Jaz jem. would mean It is me who is eating. or Moi, je mange. (in French).

  • Most, if not all, Slavic languages are like this (Russian, for example), and I see no reason for them not to be on the list, as it's a very prominent feature of those languages. Your examples in Russian: I am eating = (Я) ем, You are eating = (Ты) ешь. Another example: - Где Аня? - Ушла = - Where's Anna? - (She) went out. 84.202.199.101 19:02, 18 March 2007 (UTC)

[edit] Merger from Null subjects

It is evident that we don't need both articles. Does anyone see any problems or other issues in merging these two? --LambiamTalk 14:14, 30 May 2006 (UTC)

Nope, sounds good to me. Ruakh 14:30, 30 May 2006 (UTC)
They clearly refer to the same thing and this is better named, so merge away! —Pablo D. Flores (Talk) 16:58, 30 May 2006 (UTC)

So done. --LambiamTalk 05:46, 13 June 2006 (UTC)

[edit] Irish?

I don't know whether Irish should be considered a null subject language or not. In Irish, some finite verb forms are inflected for person and number (synthetic forms), while others aren't and have to have an explicit subject (analytic forms). Using explicit subjects with synthetic forms is ungrammatical. So for example, the 1st person singular present indicative is synthetic: molaim means "I praise", and *molaim mé is ungrammatical (Irish being VSO the verb precedes the subject). But the 3rd person singular present indicative is analytic: molann sé means "he praises" and *molann without an explicit subject is ungrammatical (except in tag questions and as an answer to a question). So what's weird about Irish compared to "canonical" null-subject languages like Portuguese is (1) not all finite verb forms carry person and number information, and those that don't may not have a null subject, and (2) verb forms that do carry person and number information must have a null subject; it isn't optional. —Angr 11:33, 24 October 2006 (UTC)

I think that given what you just said, Irish is a null subject language, but that what you just said bears explanation in the article. Ruakh 15:05, 24 October 2006 (UTC)
Okay, I'll see if I can dig up some sources. —Angr 15:32, 24 October 2006 (UTC)
On second thought, in light of the discussion below, I won't bother. —Angr 08:38, 29 October 2006 (UTC)

[edit] Which languages to mention?

Mentioning a few diverse null subject languages in the article serves two purposes: (A) to give the reader for who this is a new concept a few examples they might be familiar with; and (B) to illustrate that this feature occurs in widely divergent language families. There are some 2000 null subject languages, and it is pointless to list here as many as possible. Additionally, there is a serious problem of unverifiability. If some user adds, say, Cotoname, who is going to verify this, or, alternatively, challenge this? So I propose to whittle and keep the list down to, say, ten widely known (if only by name) diverse languages. Other viewpoints?  --LambiamTalk 00:54, 25 October 2006 (UTC)

I'm with you 100%. Maybe more than 100%, even.
The current text is thus:
Among the null subject languages are Latin, most Romance languages (including Portuguese, Spanish, Catalan, Occitan, Franco-Provençal, Italian, and Romanian – but not French), Slavic languages (Polish, Slovenian, Croatian, etc.), Finnish, Hungarian, Modern and Ancient Greek, Amharic, Arabic, Ge'ez, Hebrew, Chinese, Korean, Japanese, Persian, Tamil, Tigre, Tigrinya, and Turkish. Old Germanic languages also had this feature, for instance Old Norse regnir ("it rains") [citation needed].
How is this?:
There are some 2000 null subject languages in the world, from a wide diversity of unrelated language families. Null subject languages include Spanish, Hindi, Arabic, Chinese, and Japanese, as well as most of these languages' relatives, and many other languages besides.
(with references, of course). Is that still too long? I went with five languages I figured everyone would have heard of; I included two Indo-European languages because I assume it's the world's largest language family (in terms of number of speakers), and it's certainly the one most familiar to English Wikipedians, so it seemed worthwhile. The three others are all thought to be genetically unrelated to each other and to the Indo-European languages, though obviously the languages have influenced each other to varying extents.
Thoughts?
Ruakh 02:14, 25 October 2006 (UTC)

Bah! If you want to have such a cut, you should name language families, and not languages. Velho 04:11, 25 October 2006 (UTC)

It is in general not a property of language families but of individual languages.  --LambiamTalk 07:07, 25 October 2006 (UTC)
True. Even languages as closely related as French and Spanish can differ on whether or not they allow null subjects. —Angr 07:23, 25 October 2006 (UTC)
Lambiam — Is that true? True, French is the only major Romance language that doesn't allow null subjects; but in general, I think null-subject-allowing might well be a property of language families. (Obviously it depends on big a family you're talking about, though; it's not obvious to me that Indo-European languages can, as a group, be considered null-subject.) Ruakh 15:16, 25 October 2006 (UTC)
Well, it's somewhat of a truism to say that the tighter the relationship between two languages, the more likely they are to share a typological or other grammatical feature, and being null subject follows that pattern. It is not a very enlightening observation; it holds likewise for the preferred branching direction, the presence of a dual, or having a vocative. I can't claim to be an expert on the matter, but specifically for null subjects it would appear that distinction for person in the conjugation is a prerequisite. Since this was largely lost in French (except for being stubbornly retained in the orthography), it would explain the French exception to the Romance family. Modern Germanic languages tend not to inflect for person in the plural forms, which may explain their mandatory subject. I conjecture (but cannot check) that conversely languages having a clear person+number distinction in the verb will tend to be null subject. Changes of inflectional paradigm are among the more common in language evolution, possibly making the null subject feature one of the less stable ones.  --LambiamTalk 17:46, 25 October 2006 (UTC)
Velho — The problem with that is, it's hard to name language families that people have heard of. If we say, "Spanish, Hindi, Arabic, Chinese, and Japanese", people know what that means; if we say, "Romance languages, Indo-Aryan languages, Semitic languages, Sino-Tibetan languages, and Japonic languages", that's a lot more verbose and a lot less clear. Ruakh 15:16, 25 October 2006 (UTC)

I've replaced the text in the article by Ruakh's text, but instead of the absolute number 2000 (which is an undocumented estimate) I've put the vague but unassailable "a considerable part" (of the world's languages).  --LambiamTalk 04:55, 29 October 2006 (UTC)

[edit] Aren't imperative sentences independent clauses?!

Why shouldn't we mention that imperative sentences often come without subject?! Velho 23:34, 4 December 2006 (UTC)

We should; it's just that that wasn't the place for it. I've now added that information in a section where it's more relevant (Null subject language#Null subjects in non-null subject languages). —RuakhTALK 01:55, 5 December 2006 (UTC)
Also, the difference between independent clauses and dependent clauses has nothing to do with anything. —Angr 15:32, 5 December 2006 (UTC)

[edit] "Ele está a chover"

I've corrected the section which said that the construction "ele está a chover" doesn't exist in portuguese based on: [1] and [2](both links are in portuguese) --LeRobert 18:06, 26 December 2006 (UTC)

I'm reverting your "correction", because "ele está a chover" gets zero hits on Google aside from this article (compared with "está a chover"'s 27,100), which casts a great deal of doubt on your sources' claim, and "ele chovia" gets 44 hits (compared with "chovia"'s 174,000, though that's a bit hard to interpret because it looks like "chovia" is a noun as well), which means that even if one person claims it's correct, the vast majority of Portuguese speakers apparently feel otherwise. —RuakhTALK 02:28, 27 December 2006 (UTC)

One of those links referred to a very famous Portuguese song. All I wanted to show was that it's perfectly fine to use that construction even if it's not a very common one (maybe that sentence is not a very good example for this article). --LeRobert 10:57, 27 December 2006 (UTC)

I don't think existing in a single very famous song makes a construction "perfectly acceptable" for normal speech, other than in (explicit or implicit) reference to that song. —RuakhTALK 18:24, 27 December 2006 (UTC)

You only very rarely hear that in Portuguese. It's either done for literary purposes, or (perhaps) characteristic of a few dialects. I think it's fair to say that it's not a part of the standard language, and I would not count it as a counterexample, myself. There is no comparison between those examples and English "It rains" or French "Il pleut", where the pronoun is absolutely mandatory. FilipeS 19:34, 27 December 2006 (UTC)

You're absolutely right Ruakh! However I only mentioned the song to show you that it's not such an unseen feature. It is rare (I agree with you Filipe) but it isn't wrong as it is stated in the article. --LeRobert 21:15, 27 December 2006 (UTC)

But c'mon! The article shouldn't point to the possibility of "ele" appearing. It looks now as if it was kind of an option or so. Whenever "ele" appears in such constructions, it is always somehow non-standard or even anti-standard. The article isn't giving the right idea! Velho 02:01, 28 December 2006 (UTC)

I disagree. With Ruakh's rewrite, the article is fine. FilipeS 13:26, 28 December 2006 (UTC)

Hear! Hear! --LeRobert 13:40, 28 December 2006 (UTC)

I propose to take out the example sentence Está a chover entirely from the Examples section:
  • One way or another, it is somewhat anomalous since this is an impersonal construction.
  • Impersonal constructions are dealt with separately in the next section.
  • There is also a more general treatment of this in the article Impersonal verb.
Then we can drop this subject. :)  --LambiamTalk 07:20, 28 December 2006 (UTC)
I actually agree with Lambiam's suggestion. "Ele está a chover" is a very, very marginal construction. FilipeS 20:37, 27 June 2007 (UTC)

[edit] Over-representation of constructed languages

Here and in other linguistics articles, constructed languages are over-represented. That a given constructed language has a given characteristic is really irrelevant for most purposes, because it doesn't say anything meaningful about the characteristic (all it says is that the language's constructor thought it a good idea). That said, I feel awkward about removing this content without some sort of discussion first … so, thoughts? —RuakhTALK 21:16, 20 March 2007 (UTC)

I totally agree. There are around 2000 Null Subject Languages and such from the 3 languages presented 2 are Interlingua and Esperanto? Mirc mirc 16:10, 27 June 2007 (UTC)

I reintroduced part of that section, for the following reasons. First, the section was nearly blanked. I can see devoting less space to one language type than another, but I see no reason to single out one type and nearly obliterate it. Second, the information was removed under the misunderstanding that this section deals with constructed languages and is therefore not meaningful. While Esperanto is considered constructed, Interlingua is considered a naturally occurring language, so it isn't clear to me how it is somehow less meaningful than French or Japanese. In any case, I think it's a big leap from "constructed" to "meaningless", and it seems an overstatement to suggest that including one constructed language is an over-representation. Finally, several officially null-subject languages are included, just not in the auxiliary language section. I've made this clearer in the text. Cal (talk) 06:05, 18 November 2007 (UTC)

Re: "Interlingua is considered a naturally occurring language": What the heck? —RuakhTALK 06:41, 18 November 2007 (UTC)
Please see Gode's Manifesto de Interlingua, which explains that Interlingua didn't require any construction, and the Introduction to the Interlingua-English Dictionary, which describes how Interlingua is seen as developing naturally over hundreds of years. It's essentially the latent presence of a single language within the world's languages. The basic premise of Interlingua is that an international language doesn't need to be constructed: it exists.
Regarding my recent revision, I see no reason why a particular type of language - the auxiliary type - should be singled out and presented with no examples to facilitate comprehension. To my knowledge, Wikipedia doesn't consider prejudice to be a basis for making edits. Cal (talk) 01:42, 19 November 2007 (UTC)

[edit] Semitic Languages

There is something I want to clarify before I write sections on Hebrew and Arabic. In null subject languages, is it necessary or just optional to omit the subject? I need to know this to finish my section on Hebrew. --Vgp0012 17:56, 14 October 2007 (UTC)

Just optional; Hebrew is indeed a null subject language (especially in the past tense). However, please think about whether these sections are really helpful, and if you do write them, please keep them brief; the last thing we need is for this article to mushroom in discussing the minutiae of fifty million languages. Indeed, we already have an "Auxiliary languages" section that's really just pointless and needs to be either eliminated, or shortened to just a sentence or two. —RuakhTALK 23:30, 14 October 2007 (UTC)
Given Wikipedia's policies and guidelines, I think it would be difficult for any one editor to decide unilaterally which languages should be included in an article. One option would be to have a Specific languages section, with subsections Hebrew, Arabic, and Auxiliary languages. This way, the Auxiliary languages section would be less obtrusive. I've seen this approach used in many linguistic Wikipedia articles. It seems to work well, and I find the diversity of information very helpful. Cal (talk) 02:15, 19 November 2007 (UTC)