Subordination across languages

[mallu]

Is there any   universally accepted criterion for subordination cross -linguistically? Can embedding be considered such a criteria? Is there any language that never use subordination at all?
Could someone clear these doubts,pl.

[Daniel]

Is there any   universally accepted criterion for subordination cross -linguistically?

First, "subordination" must be defined. It is not well defined in general, and there are many competing definitions out there.
You could start here: http://en.wikipedia.org/wiki/Subordination_%28linguistics%29

There are several related definitions, involving ideas like:
1. embedding (one clause inside another)
2. dependencies (one clause relies on another for information or grammatical function)
3. Non-independence: a subordinate clause is one that cannot be used independently.

In some cases, the line is blurry. In the prototypical cases, it's fairly obvious. This is an interesting puzzle and not one that has been solved.

Can embedding be considered such a criteria?

Yes. I think it's a good definition, because if a clause is embedded it is subordinate (I believe this is true; again, there could be room to debate). But it is not necessary that it is embedded-- probably, but maybe not. In some fuzzy cases, that's one question. So embedded => subordinate, but maybe not subordinate => embedded. Usually they're the same thing, though.

Is there any language that never use subordination at all?

Hmm... probably all languages do this. But Daniel Everett has claimed that subordination (and many other things) are not found in Pirahã.
For a while it was said that sign languages did not have subordination; this was wrong-- researchers looking at the data in a different way showed it does have subordination.
Also, if defined based on the typical Latin/European model, some languages (like Native American languages or the oldest stages of Indo-European languages) appear to lack subordination, but this is just if we assume we'll find words like "that" and so forth-- there are clearly still different kinds of subordinated/embedded clauses such as with participles and so forth.

[mallu]

Can a serial verb construction be mistaken with subordination.That too will appear to involve enbedding.Isnt it?

[lx]

I don't think so. I think the idea of a serial verb construction actually forms a direct opposition to what can generally be construed as subordination. (I accidentally clicked on the edit button instead of the reply button when responding, so it has logged it as if I have made a change to the message, but it was unintentional).

[Daniel]

Can a serial verb construction be mistaken with subordination.That too will appear to involve enbedding.Isnt it?

Complicated!
It depends a lot on the theory you're using and also the labels you assign.

"Serial verb" is defined (for example by Aikhenvald) as a structure consisting of two verbs without a marker of coordination or subordination between them. But that's just surface level.
I'm completely unconvinced by that definition, and in fact many researchers do try to relate subordination, coordination and serialization-- is serialization subordination, coordination or (sometimes) both? There's a well known paper by Li & Thompson that discusses Chinese serial constructions in that way, and there are many more of course.

--
My personal opinion is this (and I will be writing about it in my dissertation): serial verb constructions might have unique surface properties (in fact, they probably do!), but underlyingly they represent other familiar structures like subordination and coordination.




Short answer: sometimes.

[mallu]

I don't think so. I think the idea of a serial verb construction actually forms a direct opposition to what can generally be construed as subordination.

Then there claim should be that all languages have serial verb constructions. Isn't it ? Atleast English lack such constructions,I believe?

[Daniel]

Then there claim should be that all languages have serial verb constructions.

No; there's no reason that all languages must have a feature. For example, not all languages have "R" sounds, even though R sounds exist.
There are some major question about why languages have serial verb constructions, but there is no reason to assume that all languages use them!
Again, a lot of this ends up being definitional.

Atleast English lack such constructions,I believe?

Definitional.
There's the well known quasi-serial verb construction "go get" or "come see", and then there are also other sequences of two verbs such as "be running" or "make him go", which fit some definitions of serial verbs.

This is a topic you should read about because there isn't one single answer to it.


Pullum, G. K. (1990). Constraints on intransitive quasi-serial verb constructions in modern colloquial English. In Ohio State University Working Papers in Linguistics (pp. 218–246).
https://linguistics.osu.edu/sites/linguistics.osu.edu/files/osu_wpl_39.pdf

Aikhenvald, A. Y., & Dixon, R. M. W. (Eds.). (2006). Serial verb constructions a cross-linguistic typology. Oxford: Oxford University Press.
https://www.google.com/search?q=serial+verb+constructions+aikhenvald

Li, C. N., & Thompson, S. A. (1973). Serial Verb Constructions in Mandarin Chinese: Subordination or Co-ordination? In C. W. Corum, T. C. Smith-Stark, & A. Weiser (Eds.), You take the high node and I’ll take the low node; papers from the Comparative Syntax Festival, the differences between main and subordinate clauses, 12 April 1973. (pp. 96–103). Chicago: Chicago Linguistic Society.

Lord, C. D. (1989). Syntactic reanalysis in the historical development of serial verb constructions in languages of West Africa (Ph.D dissertation). University of California, Los Angeles, Los Angeles.