Talk:Dual (grammatical number)

From Wikipedia, the free encyclopedia

Contents 1 Dual in Czech 2 Merger with Dual number 3 English both 4 Explanation for the loss of the dual 5 Old English 6 Duals in Russian and English 6.1 Slovenian: Volk (= engl. "wolf") 7 Merger with Grammatical Number? 8 Examples 9 Slovene 10 Dual in Slavic 10.1 Ukrainian 10.2 Rewrite 10.3 Declension table 11 Remnants of dual in Czech 12 Dual and Genitive 13 Pair 14 Splitting off the Dual in the Slavic Languages as a separate article 15 Latin 16 Hebrew transcriptions 17 English: inappropriate example 18 Japanese: dochira is not dual

[edit] Dual in Czech

Question/comment: I am not a linguist, but Russian and Czech are both my native languages. Since I grew up in the United States, I lack much of the formal education in Russian/Czech languages, so this article was very interesting. I was surprised, however, to see the distinction pointed out in Russian about counting, namely that the words change for 1, then for 2,3,4, and then from 5 on, but not in Czech. Czech is identical in this regard (for example, 1 mesto, 2 mesta, 3 mesta, 4 mesta, 5 mest, 6 mest, ...), and thus it should be pointed out in the article that Czech still uses the dual in this sense (if, as in the Russian section, the dual can be extended in this sense to 3 and 4 as well as 2). If I am missing something, I would be very fascinated if someone could explain. Thanks much.

[edit] Merger with Dual number

What about the article "Dual number"? These need a merger.. -Gabbe

Yup. Move the grammatical stuff here, maybe, and leave the other page for maths? -- Tarquin

[edit] English both

English 'both'?
What about English use of 'both' as opposed to 'all'? Is this some sort of dual number too? --romanm 19:47, 3 Nov 2003 (UTC)

It's a word which generally refers only to two, but it's not really part of a morphological paradigm or set of inflectional forms, so it's not the best example of a grammatical dual. It falls between a word like "pair" and a true paradigmatic dual. AnonMoos 03:54, 15 April 2006 (UTC)

[edit] Explanation for the loss of the dual

Is there any theory why the Indo-European languages have all lost their and gone to a system with only singualr and plural (except for Slovene)? This seems odd to have happened separately and near-universally. Xyzzyva 23:40, Jan 5, 2005 (UTC)

The languages tend to simplify. (Even in informal Slovene the Dual number is sometimes replaced by plural.) komar

Right. Similarly, in most modern Arabic vernaculars, plural forms are used instead, although there are some exceptions and proper Modern Standard Arabic does indeed have dual number.

[edit] Old English

Also, I thought Old English made 100% use of the dual, thus giving us such triads as cow - cyne (modern spelling "kyne") - cows, ox - oxen - oxes, eye - eyen - eyes, child - children - childs, brother - brethren - brethres (modern spelling "brothers"), Node 02:03, 2 Feb 2005 (UTC)

I don't think so. AFAIR the -en form (oxen, boxen etc) is older, Anglo-Saxon form (cf. German plural) while -s form is newer, Norman form (cf. French plural). rrw

[edit] Duals in Russian and English

Russian seems to be similar to Serbian in that it uses the genitive singular case for two, three or four of something and then genitive plural case for five or more of something.

Even if it isn't quite a separate dual declension, there certainly is the gramatical awareness of a one/few/many distinction.

Interestingly we use the genitive case in English with collective designations: pair of shoes, herd of cattle, school of fish, class of students, etc. etc. Is this a vestige of duality from proto-Germanic?

I put {{tl:Confusing}} under "Slavic languages". The section now discusses two quite unrelated forms of dual (paucal?):

• AFAIK, all Slavic languages (except Bulgarian and Macedonian) use singular genitive forms with numbers 2-4 and plural genitive for 5+ (Slovenian, in addition, pure dual). That should be mentioned as a common characteristics and an example put inside a table covering all/most relevant languages.
• Other remnants, as with "two eyes" described separately for languages where they're preserved. I did't understand the previous Serbo-Croatian reference to instrumental form, without the example, and I commented it out.

Duja 13:39, 16 February 2006 (UTC)

Now, please help me fill in the table, so that we could put it in the article. I selected "wolf" for the reason that it has distinctive pl. and. sg. genitive in Serbo-Croatian (unlike many others). If someone has a better example, it's welcome. Duja 13:39, 16 February 2006 (UTC):

	English	Eastern	Western			South
		Russian	Polish	Czech	Slovak	Slovenian	Serbo-Croatian
singular	1 wolf	1 волк	?	?	?	1 volk	1 vuk
plural (without number)	wolves	волки	?	?	?	volkovi	vukovi
dual (2-4)	2 wolves	2 волкa	?	?	?	2 volkova 3,4 volkovi	2 vuka
plural (with number)	5 wolves	5 волков	?	?	?	5 volkov	5 vukova

[edit] Slovenian: Volk (= engl. "wolf")

I'm not sure what you mean by "dual (2-4)"? Slovenian, which uses the grammatical dual in full, uses it only for the number two:

1 volk (grammatical singular)

2 volkova (grammatical dual)

3 volkovi (grammatical plural)

4 volkovi (grammatical plural)

5 volkov (grammatical plural)

7 volkov (grammatical plural)

8 volkov (grammatical plural)

...and so on...

Without a preceding number, it's "volk" (singular), "volkova" (dual), "volkovi" (plural). Conversationally, "volka" is often used in place of "volkova", but, once again, only for the true dual (not numbers 3-4) WorldWide Update 23:53, 14 March 2006 (UTC)

Right, I forgot that Slovenian has pure dual. I'll put that in the table.Duja 14:49, 15 March 2006 (UTC)

Thanks/hvala! You got it right. WorldWide Update 23:41, 15 March 2006 (UTC)

Still, plural 5+ is not so simple. For 101, 102; 201, 202, ...; 1001, 1002, etc. is similar as in 1 and 2. Besides, for dual it is much more common to correct into dva volka. Ziga 19:36, 10 May 2007 (UTC)

---

I would like to point out a couple of inaccuracies:

Slovene "očesa" pushed out "oči" instead, becoming the proper dual (and plural) of "oko").

This is somewhat inaccurate, and the situation is a bit more complicated in practice. The dual and plural form is "oči" when referring to human or animal eyes, and "očesi" (dual) / "očesa" (plural) when referring to other eye-like things.

Nominative case of noun "city": singular: 1 mesto dual: 2 mesti plural: 3-4 mesta, 5+ mest

"mest" is not the nominative case of anything -- it's the plural genitive. In Slovenian, the plural genitive is used instead of the plural nominative if the number of things (modulo 100) is 5 or more. 212.235.208.154 07:50, 28 June 2006 (UTC)

[edit] Merger with Grammatical Number?

I had proposed to merge this page with grammatical number. I think there is information here that could just as well be on the other page. However, I withdrew the suggestion, since I can't commit to this edit right now. :( FilipeS 13:08, 8 April 2006 (UTC)

On second thoughts, this article is probably too large to be merged with grammatical number. FilipeS 19:48, 20 April 2006 (UTC)

[edit] Examples

I added the examples from german wikipedia. I didn't really understand the bit about the adoption of plural form in Austro-Bavarian, though, so i may have translated it incorrectly.

[edit] Slovene

The following was in the main article on grammatical number, but I have decided to delete it (see here). I think it's more appropriate for this article, for the reasons explained in the other Talk page. However, since it seems to contain some inaccuracies (see here), I thought it would be better to leave it here, so that it would be corrected by someone who knows Slovene, and added to this article.

Slovene, a Slavic language, is more complicated:

• Barva (a color) (singular), barvi (two colors) (dual), barve (three or four colors), barv (five and so on)
• Hiša (house) (singular), hiši (two houses) (dual), hiše (three or four houses) (plural), hiš (five or more houses) (plural)
• Miš (mouse) (singular), miši (two or more mice) (dual := plural)
• Jaz (I) (singular), midva/midve (we) (dual + [Masculine/Feminine gender]), mi/me (we) (plural [Ma/Fe gender])
• Vrata (one door) (singular), dvoje vrat (two doors) (dual), troje vrat (three doors) (plural), [plural noun with different or same form]
• babine (one afterbirth period) (archaic word) (singular), babini (two afterbirth periods) (dual), babine (three or four afterbirth periods) [plural noun with different or same form], babin (five or more afterbirth periods)
• Človeštvo (mankind) (singular), človeštvi (two mankind) (dual), človeštva (three or four mankind), človeštev (five or more mankind) [collective noun with different form]

Note:

• When a number reaches one hundred and one (two) (or several hundred or thousand and so on), singular and dual are used again. (ena knjiga (one book) (singular),dve knjigi (two books) (dual), pet knjig (five books) (plural), sto ena knjiga (101 books) sto dve knjigi(102 books))

• These and similar examples are very often used incorrectly, even in published or electronic dictionaries.

FilipeS 14:26, 5 November 2006 (UTC)

[edit] Dual in Slavic

It is inappropriate to talk about a "dual" (or even "paucal") number in Slavic languages other than Slovenian. The article claims that BCS and Russian have a dual, but (i) it is only used with the numerals 2-3-4 and (ii) it is identical to the genitive singular. In fact, it is the genitive singular, not a dual form. For example, Rus. "волкa" cannot be used by itself (say, as the subject of a verb) to mean "2, 3, or 4 wolves"; you have to use "волки" and it just means "wolves (plural)". Moreover there are no special verb forms for agreeing with a dual/paucal subject (as there are in Slovenian); there are only singular and plural verbs. Historically, there might be a connection between the special syntax of 2-3-4 and the lost dual (I have no idea), but the current article makes a much stronger claim.

Numeral phrase syntax in Slavic is very complex and probably deserves an article of its own. The discussion of the dual in Slavic in this article should really only focus on Slovenian, and mention lexical vestiges of old duals in some plural forms (hands, eyes, ears) in the other languages.

CapnPrep 16:49, 5 November 2006 (UTC)

I agree. I was a bit surprised to see that the Slovene dual is, in comparison to other Slavic languages, seriosly downplayed - Slovene (along with Sorbic) is afterall the only Slavic language that retains an actual grammatical dual, unlike those other Slavic languages, where some partial vestiges of it remain at best. For starters, I'll move Slovene to the top of the Slavic languages and try to make something out of it's section. Unfortuantely, grammar is not my strong point, so any further help would be more than appreciated. TomorrowTime 09:31, 16 January 2007 (UTC)

Ok, some observations.

-I realise the list of Slavic languages was meant to be alphabetic and my pushing Slovene up to the top will no doubt disturb some deep cosmic harmony, but I think Slovene more than deserves to be first on the list.

-The Sorbian language, one of the three Indo-European langauges with a surviving full dual is not even mentioned. What gives?

-The Slovene entry has been considerably shortened. By me. The ridiculous petty assertions in the oh-so-numerous brackets have been either deleted or moved into the text. No brackets, just freedom!

-I hereby proclaim that I find the adjective "Slovene" to be better sounding than the adjective "Slovenian". Anybody who would want to hang me for high treason because of this should please discuss that here, not in the article. No brackets, just freedom!

-Is there any rule to how examples should be given? I wouldn't mind giving more examples of words and their dual forms, but I'm not sure how to go about this. In fact, all in all this whole article could do with some standardisation - as it is, every language entry is written differently, and it's hard to read.

That's all, folks. TomorrowTime 10:00, 16 January 2007 (UTC)

I deleted the irrelevant sections on Russian and BCS and part of the Ukrainian. If someone can demonstrate (with references) that the use of the gen. sing. with 2-3-4 has anything to do with dual number, this should be discussed as a general Slavic feature, not repeated for every Slavic language individually (the discussion in #Duals in Russian and English above is worth reviving). I salvaged one of the tables for use with Slovene, but I don't know the bare plural form of mesto. CapnPrep 01:25, 17 January 2007 (UTC)

Great, looks better already. As for mesto... Hm, the bare plural in nominative would be mesta - this is then inflected for other cases, of course, but I'm not sure if we should go into such details at this stage of the development of the article.

Mesto is a neutral noun. I suppose examples of a feminine and masculine noun would be in order here. In trying to keep up with the general architectural feeling I propose the following nouns: cesta (road) for feminine form and dimnik (chimney) for masculine. So: cesta (sing), cesti (dual), cest (3-4), cest (5+), ceste (bare plural, nominative); dimnik (sing), dimnika (dual), dimniki (3-4), dimnikov (5+), dimniki (bare plural, nom).

Are the above examples ok? I can always help out with further examples or minor clarifications, but I must confess that grammatical explanations are not my forte. You know how it is with native speakers and grammar :) TomorrowTime 09:52, 17 January 2007 (UTC)

[edit] Ukrainian

I have just deleted the following section:

====Ukrainian====

Ukrainian dialects (e.g., of the Vinnytsya region) have preserved the dual in feminine nouns: кімната - дві кімнаті - шість кімнат (a room - two rooms - six rooms); рука - дві руці - сім рук (a hand - two hands - seven hands) and in neutral nouns: відро - дві відрі - десять відер (a bucket - two buckets - ten buckets).^{[verification needed]} Verbs have completely lost their dual forms.

In essence, the same is true about Ukrainian and Russian languages overall, not just this dialect. There is a difference between the plural forms for a) two objects and b) five or six or seven objects, but that doesn't make a) a dual. See discussion above. Is the form a) not applicable to three or four objects? Why is there no example that shows that this dialect is any different in this respect? If there were one, it would need to be verified. Sergivs-en 18:14, 20 April 2007 (UTC)

I believe the claim is that the form of the noun used with 2-3-4 is distinct from the genitive singular for feminine and neuter nouns. This would be then different from standard Ukrainian (and Russian). But the information would have to be verified (and the text above rewritten, so I agree with your deletion). CapnPrep 21:48, 28 April 2007 (UTC)

[edit] Rewrite

Somme comments on the recent rewrite of this section (which I like, overall):

• What are the "dual characteristics" of the declension of two that "can be verified" in the first table?
• There is no single rule for 2-3-4 in the modern Slavic languages that have lost the dual. The article currently suggests that in all of these languages, the genitive singular form of the noun is used to form a nominative NP. This is not true, for example, in Polish, where the nominative plural is used. This rule is also inappropriate for Southern Slavic languages with no case morphology.
• The "five and above" rule needs to mention higher numerals that end in 1 or 2-3-4 (e.g. twenty-one through twenty-four). Again, there is no single rule valid for all non-dual Slavic languages.
• This section should be moved into a separate article; it is completely out of proportion with the other languages and language families treated.

CapnPrep 22:11, 28 April 2007 (UTC)

[edit] Declension table

I still don't get the significance of the table enumerating the inflectional forms of "two" in more and more Slavic languages. If the claim is that the declension pattern of "two" and "both" displays specific dual characteristics, this has to be shown in comparison to typical paradigms for other numerals and quantifiers. And then, is it necessary to do this for a dozen languages? CapnPrep 12:26, 30 April 2007 (UTC)

Since this is a section on the dual in the Slavic languages, I think that the dual forms in each of the Slavic languages should be mentioned. However, it may be more appropriate to place this section as its own article, where a complete listing can be given. The forms in this section would then be shorten to a respresentative sample (say Russian, Polish, Serbo-Croatian, Slovenian, and Sorbic) with a comparison between the singlular, plural, and dual endings. Woollymammoth 02:26, 1 May 2007 (UTC)

[edit] Remnants of dual in Czech

In Czech, although there is no dual as such, the plural forms of some nouns indicating things which naturally come in pairs (mainly paired organs e.g. eyes / ears) have a different declension from "normal" plurals - these even affects the adjectives which describe them (interesting that it is apparently the other way round in Slovene).

As pointed out in the previous contribution, this should not be confused with the use of the genitive singular for numbers of 5 or more, which is common to most (all?) Slavic languages. This is due to the way numbers 1-4 behave like adjectives, whereas other expressions of quantity behave like nonus.

[edit] Dual and Genitive

To clarify: Slavic languages use the genitive ("of" form) of nouns in connection with most expressions of quantity EXCEPT the numbers 1-4. Thus the expression for "five beers" literally translates as "a five of beers". The use of the dual in Slovene, or remnants of this form in other languages, is a separate issue.

Taking beer as the example, the Czech is: 1 pivo (singular), 2 piva (plural), 3 piva (plural), 4 piva (plural), 5 piv (genitive plural). In Slovene, I believe it would be 1 pivo (singular), 2 pivi (dual), 3 piva (plural), 4 piva (plural), 5 piv (genitive plural).

I apologise if I've got the Slovene examples wrong, but the principle still applies.

You got the Slovene examples right. While in Slovene "five beer" would only roughly (hardly) translate as "a five of beers" it is still true that for 5-and-above the genetive plural form is used. I think this explanation with genetive plural form removes all concerns about implication of more than three (singular, dual, plural) numbers. I might be talking about 10 bears or general plural of bears and I would still say "Medvedje so v kleti." (Bears are in the basement!) and still Five bears in the basement. would be Pet medvedov je v kleti.. Beers might not be such good example, because we would much rather like to talk about beer as a mass of liquid (Pivo je v kleti - Beer is in the basement). 222.152.244.105 13:31, 13 January 2007 (UTC)

[edit] Pair

English utterly lacks a dual, but the semantic issues involved with pairs is interesting. Fashionistas insist on calling it a pant or jean, but they are pants and jeans. In fact, all garments that begin about the waist and extend somewhat around the thighs take the plural: pants, trousers, jeans, boxers, shorts, pantalettes, coulottes, etc.

Compare this to bikini, brief, panty hose (hose is an interesting word). If the garment does not enclose the thigh, it's singular. But notice that whitie-tighties (even though a white cotton male brief) is apparently plural.

The article mentions modern Hebrew. Stereo speakers are spoken of in the dual, along with headlights and eyeglasses. These are natural pairs, as with shoes and gloves. There is the issue of chirality (a bad article). --MarkTwainOnIce 06:55, 23 January 2007 (UTC)

English still retains some use of the dual, and even the trial number, in the use of occasion: e.g. once, twice, thrice, four times, five times etc, Lobbuss 10:46, 12 April 2007 (UTC)

Is it really a dual and trial? I guess they are just numerals, aren't they? Adam J. Sporka 10:23, 19 April 2007 (UTC)

Yes, the dual number is not the same thing as the number two, is it?

I guess it depends on how one defines "grammatical number". Is it just a matter of which quantity you're talking about, or is it a systematic inflection? Traditionally, the term "dual" has been used only in highly synthetic languages that have inflections associated with each number, where hundreds to thousands of words inflect in similar ways to show number. I don't suppose that English does this... FilipeS 12:20, 5 May 2007 (UTC)

[edit] Splitting off the Dual in the Slavic Languages as a separate article

As has been mentioned by others above, I would propose that the section "Dual in the Slavic Languages" be split off as a separate article. A short summary of the Slavic dual should be presented, along with a section on both Slovenian and Sorbic, which have maintained the dual. It would be nice to include the forms for the verbs as well. Woollymammoth 02:29, 1 May 2007 (UTC)

[edit] Latin

I remember from high school Latin that the pre-Golden Age (pre-Virgil) Latin had a dual. I'll look it up, and then work up a section. Has anyone else heard this? samwaltz 22:14, 22 May 2007 (UTC)

• Some languages have a "dual" category, of which Latin has only vestiges, e.g. ambo, duo, where the long -o- is an Indo-European dual. [1]
• The form in -o is a remnant of the dual number, which was lost in Latin, [2]

I'm looking for more. Two lines saying that the dual died out in Latin... well, it's just not enough for my taste. samwaltz 22:35, 22 May 2007 (UTC)

[edit] Hebrew transcriptions

The Hebrew words are transcribed using Ashkenazic pronunciation, even when the text refers to the language "as spoken in the State of Israel". --Smack (talk) 18:50, 23 May 2007 (UTC)

[edit] English: inappropriate example

English has words distinguishing dual vs. plural number, including: both/all, between/among, latter/last, either/any, and neither/none. Japanese, which has no grammatical number, also has words dochira (which of the two) and dore (which of the three or more), etc.

CapnPrep spent a long time over at the grammatical gender article getting it through my thick skull that the instances of gender inflection in English are so insignificant that they're not enough to say that English has grammatical genders. I quite agree. But exactly the same is true of so-called "duals" in English. In short: English doesn't really have dual number. FilipeS 14:30, 29 May 2007 (UTC)

[edit] Japanese: dochira is not dual

As far as I understand, dochira is not necessarily "which of the two", it's just "which way". The set of possible answers has three options: Kochira (this way, near me), sochira (that way, near you), achira (that way, over there). Jpatokal (talk) 06:52, 25 November 2007 (UTC)

Hmm. After putting on my thinking hat and checking WWWJDIC's examples, I'll amend that: dochira can mean "which of two", but it does not necessarily mean that there are only two options:

どちらに御泊りですか。 [T]

Where are you staying? (= infinite options)

紅茶とコーヒーではどちらの方が好きですか。 [T]

Which do you like better, tea or coffee? (= precisely two options)

But you couldn't say "tea or coffee or juice" using dochira in the second sentence. Jpatokal (talk) 06:58, 25 November 2007 (UTC)

The Japanese example is misleading. "Dochira" is part of the ko/so/a/do set with "kochira" and "sochira" and "achira", which are all terms relating to *direction*. That being "this way" (towards speaker) / "that way" (away from speaker) / "the previously mentioned way" (a definite/indefinite distinction) / "which way" respectively. I've heard NS use "dochi" for more than two ("dochi" just being a contraction) and "dore" for two, as well.

I'm very suspicious of this. As far as I know, Japanese has little noun inflection. It doesn't even have grammatical number as a morphological category, so how can it have a dual? FilipeS (talk) 12:09, 11 January 2008 (UTC)