Why Don't Verbs Conjugate with the Object?

[goofy]

If we call "Old Low German" the variety that Old High German developed out of, then you're right. But that's not what "Old Low German" usually refers to - Old Low German is the term for the variety that is first attested around the same time as Old High German AIUI.

But the Old Low German, the language spoken in Northern Germany and the Netherlands at the same time as Southern Germany, Austria, German-Switzerland was doing the above-mentioned sound shift / having completed it, is exactly identical to what was spoken in said area (Southern Germany etc.) before that sound shift. So why call it something different? Or if it's called something different - what is it called?

Well I don't know much about this, but the book I'm consulting says that Old High German is attested in runic texts from 500 AD. And it seems that Old Low German is attested from c800 AD. So the ancestor of both OHG and OLG would have been spoken over 1500 years ago, and would be called West Germanic or Proto-Germanic, depending on which theory you subscribe to. It's possible that Anglo-Frisian and Low German form their own branch, called Ingvaeonic.

I'm skeptical that the Old Low German spoken before the second sound shift, that is about 1800 years ago, is "exactly identical" to the Old Low German spoken after the sound shift, about 200 years later (assuming these dates are accurate). Yes, it was still the same language, so it might make sense to call it "Old Low German", the same way you can call Anglo-Frisian "English". But on the other hand, these terms are useful for labeling languages as they're spoken at particular points in time, and as far as I can tell, the term "Old Low German" is used for the variety attested from c800 to c1100.

What's AIUI?

as I understand it.

[getajob92]

I haven't read all 81 posts, but I'd just like to point out that this exists in Spanish to a certain extent.

In Spanish, one can say lo/la/me/te/le/se/nos/los/las + verb in order to address the object. While these aren't attached to the verb in the example, there are multiple situations where the word identifying the object. For instance, infinitives and progressives can have such a suffix ("tomándote" means taking you and "Yo quiero tomarte" means I want to take you). Such conjugations are also shown when you say "let's <verb>" such as when you say "vamanos" for "let's go."

[gmalivuk]

That's not really conjugating, though. It's just slapping the object pronoun itself onto the end of the verb and writing it as one word.

[dhokarena56]

Nobody's mentioned ergativity?
Ergative languages include Basque and Inuit. In ergative languages, the agent of an intransitive sentence and the patient of a transitive one have the same case, the absolutive; the agent of a transitive sentence is in the ergative.
Contrast accusative languages like English, where all agents are the same.
In ergative languages, the verb agrees with the absolutive (sometimes the ergative too), which in some cases is our "accusative"/"object".
It gets weirder, but that's the general idea. You can read more at http://en.wikipedia.org/wiki/Ergativity.
By the way, this is my first post, so: I'm Dhokarena56 (Dhok), and you may be familiar with me from PCTYD or the ZBB. It's a pleasure to meet you all.

[Monika]

Nobody's mentioned ergativity?

I guess no one speaks Basque or Inuit.

Ergative languages include Basque and Inuit. In ergative languages, the agent of an intransitive sentence and the patient of a transitive one have the same case, the absolutive; the agent of a transitive sentence is in the ergative.
Contrast accusative languages like English, where all agents are the same.
In ergative languages, the verb agrees with the absolutive (sometimes the ergative too), which in some cases is our "accusative"/"object".
It gets weirder, but that's the general idea. You can read more at http://en.wikipedia.org/wiki/Ergativity.

Oh wow, finally some languages which - sometimes - *do* conjugate with the object!

By the way, this is my first post, so: I'm Dhokarena56 (Dhok), and you may be familiar with me from PCTYD or the ZBB. It's a pleasure to meet you all.

Well thanks for joining to let us know about ergative languages!

Now I had to look up PCTYD on the urban dictionary. And then CTY. Now I envy you. Couldn't find ZBB though.

Have you seen the intro thread?

[goofy]

In ergative languages, the verb agrees with the absolutive (sometimes the ergative too), which in some cases is our "accusative"/"object".

The examples on the wikipedia page show how the noun case markings change, but none of them show the verb agreeing with the object.

But it does happen, I gave an example of this in Hindi earlier in the thread.

राजू ने किताब पढ़ी।
rājū-ne kitāb paṛhī
Raju.msg-OBLIQUE book.fsg read.fsg.PERFECTIVE
"Raju read the book" - the verb paṛhī is feminine agreeing with kitāb "book".

Hindi shows split ergativity: sometimes the syntax is accusative, and sometimes it's ergative.

[gmalivuk]

Couldn't find ZBB though.

http://www.spinnoff.com/zbb/

[Monika]

Thx!

[Fryie]

Hm, also I'm always very skeptical of taking irregular forms into account. As for the French examples: verbs like "être", "avoir", "savoir", etc., are very rarely used in the imperative for obvious reasons, so it is not suprising that they still use old forms. It's not the same for many other verbs, though (cfr. "prend!", imp. - "prenne", subj.), although not the best example, since it's also an irregular verb. "saisis", imp., vs. "saisisse", subj., would perhaps be more clear.

You're kidding, right? I can't tell you how many times I've heard "Soyez prudent" (be careful) or "Soyez à l'heure" (be on time - and it's variations, like be there at __ o'clock...). An example in English that I think would work in French is what my mother says to me when we talk about my career goals: "You want to make money? Be a doctor." And as far as savoir goes, you're less likely to hear it, but it still happens. For instance, when exams roll around, your professor might tell you to know certain information. It's used often in English as an imperative ("know this" as in "you think people don't use 'to be' as an imperative? Well know this, mister...").
So I don't think the reasons are obvious at all. Irregular verbs are quite often the most commonly used verbs, generally speaking, at least in the languages I've learned.

Of course verbs like "être" or "avoir" are the most common verbs, but not in the imperative. If you were to compare imperative and indicative uses for, say, "être" and "porter", you would get quite different results. Of course it is possible to use forms like "soyez", "sachez", etc., but they are quite rare compared to the indicative uses. On the other hand, I'm sure a verb like "porter" is used relatively often in the imperative.

The reason is simply that "être" and "avoir" denote actions which cannot be controlled by the subject (as opposed to "porter"). You can use those verbs in the imperative only with an indirect meaning. That is "sachez que ...", e.g., is not a real command but more like "hey, listen, you will want to know what I'm telling you".

As for Spanish (and Italian, as it's the same thing there):

That's not really conjugating, though. It's just slapping the object pronoun itself onto the end of the verb and writing it as one word.

Yeah, but the difference between a clitic (as it is now in forms like "domandarsi" (to ask oneself)) and an inflectional ending may quite subtle. Here, it can be clearly made, since the clitic can be freely moved - as in "si domandò" (he asked himself), and since the pronoun is only attached to the verb in absence of an overt object (that is "domandarlo" (to ask him), but "domandare qualcuno" (to ask someone)).

And as for Ergative languages... yeah, well depends an your notion of "object", really. If the question had been "verbs that agree with the patient" you would of course be right. But I'm prudent to call the absolutive "object", since it is really so pivotal in the language...
And then there's also the fact, just to make things a lot more complicated, that while quite a number of languages are morphologically ergative, there are very few of them that are also syntactically ergative.

[ZLVT]

Ergative languages include Basque and Inuit. In ergative languages, the agent of an intransitive sentence and the patient of a transitive one have the same case, the absolutive; the agent of a transitive sentence is in the ergative.
Contrast accusative languages like English, where all agents are the same.
In ergative languages, the verb agrees with the absolutive (sometimes the ergative too), which in some cases is our "accusative"/"object".
It gets weirder, but that's the general idea. You can read more at http://en.wikipedia.org/wiki/Ergativity.

Oh wow, finally some languages which - sometimes - *do* conjugate with the object!

As I mentioned earlier Hungarian ALWAYS conjugates for object and is not ergative.

Let me see if I get this:
"You kick the dog. I laugh" Here the bolded are in absolutive, but 'You' is ergative? (so you just have a special case for the subject of a transitive sentence basically) and the verb always agrees with whichever is in the absolutive? So "you kick the dog", kick agrees with dog but not 'you' in any way shape or form?

[goofy]

Let me see if I get this:
"You kick the dog. I laugh" Here the bolded are in absolutive, but 'You' is ergative? (so you just have a special case for the subject of a transitive sentence basically) and the verb always agrees with whichever is in the absolutive? So "you kick the dog", kick agrees with dog but not 'you' in any way shape or form?

This happens in Hindi/Urdu sometimes. I don't know if there are any languages that do this consistently. Basque is fully ergative, but as far as I can tell, Basque verbs don't agree with the absolutive.

I've only seen two examples of the verb agreeing with the object in this thread: my Hindi example, and bebboe's Chichewa example. Someone said it happens in Dene.

Way back at the beginning of this thread, you said

Klingon and Hungarian both conjugate for object to various degrees. Persian also can conjugate for object in all person/number combinations.

Could you give some examples in Hungarian and Persian?

[ZLVT]

Hungarian's easy enough. Our verbs in any mood/tense/person/number combination still have 4 forms:
indefinite - used for no object, or an indefinite object -> I eat, I eat somthing, I eat everything, I eat some bread
definite 1st person - used for a definite object in the first person -> you love me, you love us
definite 2nd person - used for a definite object in the second person -> i love you (s/pl)
definite 3rd person - used for any 3rd person definite object.

now the rules aren't that hard really, basically we have a special form for the 3rd person object which is also used for reflexive verbs. and a special form in the singular 1st person for 2nd person objects.

Note the present tense indicative conjugation of the verb "szeretni" to like/love (aimer)

Code: Select all

1st sing szeretek  szeretem  szeretlek  szeretem
2nd sing szeretsz  szeretsz  szereted   szereted
3rd sing szeret    szeret    szeret     szereti
1st pl.  szeretünk szeretjük szeretünk  szeretjük
2nd pl.  szerettek szerettek szeretitek szeretitek
3rd pl.  szeretnek szeretik  szeretnek  szeretik


Now Persian is harder as I've only been studying it for a while, but from my understanding there are a set of endings which are called "Accusative enclitics" who fullfill 2 main roles. Affixed to a verb after the personal ending it denotes the object, and affixed to a noun it denotes the possessor. I'll give the forms as they are affixed to the noun قَلَم [ɢælæm] (pen/pencil) but I've removed the vocalisation on the stem 'cos over tall letters like the l it looks bad.

قلمَم
قلمَت
قلمَش
قلمِمان
قلمِتان
قلمِشان


Though I don't think that really qualifies as conjugating for object since Persian is IE and not really agglutinating so if they really did conjugate for object it would be more integral to the suffix I think

[goofy]

How about a sentence in Hungarian with a gloss?

Yeah, if by "conjugate" we mean "inflectional morphology" then it sounds like Persian doesn't count.

[ZLVT]

I assume you mean a simple SVO sentence so:

kinyitom az ajtót
out-open[indicative active present 1st. sing subj, 3rd. object] the door[acc]

[Monika]

Of course verbs like "être" or "avoir" are the most common verbs, but not in the imperative. If you were to compare imperative and indicative uses for, say, "être" and "porter", you would get quite different results. Of course it is possible to use forms like "soyez", "sachez", etc., but they are quite rare compared to the indicative uses. On the other hand, I'm sure a verb like "porter" is used relatively often in the imperative.

Well, let's check for German.

http://www.google.de/search?q=%22sei%22 ... lr=lang_de 57.900.000 Seiten auf Deutsch für "sei"
http://www.google.de/search?q=%22trag%2 ... &lr=lang_d 2.790.000 Seiten auf Deutsch für "trag"

"Sei" means be as imparative singular as well as I be / he/she/it be in conditional/conjunctive, but the latter is not used very much (though not totally uncommon in writing). It is also found as abbreviation: S.E.I. So the real number of imparatives is not too far away from the 57 million found, I would guess 50-80% of the search results (from looking at the first pages).

"Trag" means carry as imparative singular. It is also often used for "trage", first person singular indicative. 5 of the 10 results on the first page are not "Trag" as verb, but "Trag-" as part of a split compound noun (Trag- und Verformungsverhalten, carry and deformation behavior). Therefore the number of imparatives amount these 2.7 million results is probably less than half.

[Fryie]

Yes - but you would have to compute the relative frequencies, that's the key point (because since "sein" is used much more than "tragen", it's obvious that this applies to the imperative, too). If you were to compare the imperative/indicative proportions for both verbs, however, I suggest you'd find that it is much higher for "tragen" than for "sein".

I did a sample study using the BNC (British National Corpus). I used the verbs "be" (a stative verb) and "throw" (denoting an action that the subject has control over). Unfortunately, it is not (straightforwardly) possible to look for imperatives only, given the current annotation, so in the end I compared frequencies for all base forms ("be", "throw") vs. 3rd person singular present forms ("is", "throws").
The results were the following:
BE
"is" (3rd): 10106.09 instances per million
"be" (base)- 63.9 instances per million

Proportion 3rd/base = 158.155

THROW
"throws" (3rd): 5.28 instances per million
"throw" (base): 7.59 instances per million

Proportion 3rd/base = 0.696

Which supports my hypothesis: The verb "to be" is used insanely more often in the 3rd person singular form than in its base form. With "to throw", it's the base form which is more often used (although the difference is not so enormous).
There is, of course, one methodological problem, namely that base forms can be infinitives, too (some of those cases might be excluded by not allowing a preceding "to") - but I don't see a reason why infinitives should be used more often with "to throw" than with "to be".

[Monika]

Yes - but you would have to compute the relative frequencies, that's the key point (because since "sein" is used much more than "tragen", it's obvious that this applies to the imperative, too). If you were to compare the imperative/indicative proportions for both verbs, however, I suggest you'd find that it is much higher for "tragen" than for "sein".

Oh, I hadn't understood your post to compare these two proportions. And I don't see how they support your point, which I thought was that imparatives of the highly irregular verbs like be and have are used so rarely that they would be forgotten and become regular over the course of centuries, or something like this?

[Fryie]

My original point was more like "imperatives are very often just the bare stem". This was challenged by examples with verbs like "to be", etc., and I said, since those verbs are not used as often in the imperative as other ones, that logic may not apply to them. Also, they are irregular anyway.

[ZLVT]

on tpoic, confirmation that finnish doesn't, but if I rememebr correctly Zulu does to an extent.

[ZLVT]

Ok, turns out wiki has a page on this, though it's incomplete. http://en.wikipedia.org/wiki/Polypersonal_agreement wherein it turns out that Georgian also conjugates for object as does Basque. Also http://en.wikipedia.org/wiki/Seri_language is a language spoken by a few thousands in mexico which also have polypersonal agreement.

[daryuu]

I don't have anything to contribute to the main topic, but I was reading through and wanted to reply to this:

But yeah, I think "be" is irregular in every natural language (including Turkish which I remember reading once only has that one irregular verb, but I could be mistaken), but perhaps especially so in IE languages on account of collapsing the original three into only one or two.

In Turkish "to be" isn't really a verb. Instead one appends certain suffixes to the noun or adjective. This is regular.

How many irregular verbs Turkish has depends on what one considers irregular I guess. Our Turkish teacher taught us 4 irregular verbs. I have seen 5 mentioned, the 4 we learned and an outdated or less used one. I have also read articles that say that only 1 verb is irregular. The irregularities are small things like a shift from t to d.

Chinese doesn't have conjugation, so 是 = "shì" = "to be" is regular, if one can consider "no conjugation, ever, at all" as a case of "regular". I wonder about Japanese.

All five of the Japanese verbs for "to be" (irassharu, iru, and oru for animate existance, and gozaru and aru for inanimate existance) are regular. So far as I'm aware, the only verbs considered to be irregular in Japanese are kuru (to come), suru (to do), and iku (to go).

[Monika]

Thanks ZLVT for the enlightening links and thanks daryuu for answering my old question about Japanese verbs .