correlative conjunctions and independent clauses

[Holmes]

Greetings,

It is common grammatical wisdom that an independent clause is a clause that is capable of being punctuated as a separate sentence. It is also common grammatical wisdom that certain correlative conjunctions (specifically, not only . . . but (also) . . .; and either . . . or . . .) can coordinate independent clauses:

(1a) They will have a party, and they have hired a clown.
(1b) Not only will they have a party, but they have hired a clown.

(2a) You will do as you are told, or things won't go well for you.
(2b) Either you will do as you are told, or things won't go well for you.

However, closely examined, sentences like (1b) and (2b) above do not seem to consist of two independent clauses by the abovementioned definition of "independent clause," since neither *Not only will they have a party nor *Either you will do as you are told is capable of being punctuated as a separate sentence.

Are (1b) and (2b) sentences with only one independent clause rather than two? That is the conclusion I am inclined to draw, though it does occur to me that one might say that the correlative conjunction properly lies outside the clauses it unites; so perhaps that is a possibility as well. What do you think?

Thank you.

 

[5jj]

It is common grammatical wisdom that an independent clause is a clause that is capable of being punctuated as a separate sentence.

Like much 'common grammatical wisdom', it does not stand unto close analysis.

 

[jutfrank]

it does occur to me that one might say that the correlative conjunction properly lies outside the clauses it unites; so perhaps that is a possibility as well. What do you think?

Yes, surely that's the way to think about this, as far as I'm concerned. Conjunctions conjoin two elements.

They will have a party, and they have hired a clown.

Let's call the red bits statements and the blue bit a conjunction that conjoins them. The conjunction isn't actually a part of either statement itself. It's the logical 'glue' that sticks them together. You can't remove the first element to leave: And they have hired a clown. That doesn't make sense on its own. It's no different with correlative conjunctions.

 

[Holmes]

Yes, surely that's the way to think about this, as far as I'm concerned. Conjunctions conjoin two elements.

They will have a party, and they have hired a clown.

Let's call the red bits statements and the blue bit a conjunction that conjoins them. The conjunction isn't actually a part of either statement itself. It's the logical 'glue' that sticks them together. You can't remove the first element to leave: And they have hired a clown. That doesn't make sense on its own. It's no different with correlative conjunctions.

Thank you, Jutfrank. I agree that "And they have hired a clown" doesn't make sense on its own. Syntactically, though, it can stand alone as a well-formed sentential unit. But neither "Not only will they have a party" nor "Either you will do as you are told" can so stand, each being felt to be incomplete.

A lot of stand-alone sentences, such as John 3:16 (KJV), begin with a coordinating conjunction, even though understanding why the conjunction has been used requires a broader context.

Still, I like your sense that the elements coordinated by the two halves of the correlative conjunction may be regarded as independent clauses. Even the first half of (1b), which becomes "will they have a party" when "Not only" is deleted, could be punctuated as an independent interrogative clause.

 

[jutfrank]

I agree that "And they have hired a clown" doesn't make sense on its own. Syntactically, though, it can stand alone as a well-formed sentential unit. But neither "Not only will they have a party" nor "Either you will do as you are told" can so stand, each being felt to be incomplete.

I wish I could follow the thinking on this because I don't really get it at all, if I'm honest. The clause And they have hired a clown is clearly not a complete thought any more than because they have hired a clown or if they have hired a clown. In all three cases, there's some logic missing. If you remove the conjunctions, they're complete. That's because the conjunctions only makes sense in context with the other clause they're there to conjoin.

In what way is Either you will do as you're told incomplete? Isn't the incompleteness only in that there's a logical or-phrase missing? What does that have to do with syntax?

I'm glad I'm not a grammarian.

 

[Holmes]

I wish I could follow the thinking on this because I don't really get it at all, if I'm honest. The clause And they have hired a clown is clearly not a complete thought any more than because they have hired a clown or if they have hired a clown. In all three cases, there's some logic missing. If you remove the conjunctions, they're complete. That's because the conjunctions only makes sense in context with the other clause they're there to conjoin.

I'm glad you've brought up the definition of a sentence as a "complete thought." That's a traditional definition that I've been curious about. It seems so natural, but lately I've been thinking that it's a definition more suited to the paragraph, or maybe even to the short essay, than to the sentence.

Since you find "And they have hired a clown" to be incomplete, I'm assuming that you would say the same about "They have hired a clown, too." Yet, absent context, we can no more understand the "too" in the latter than the "And" in the former, rendering them semantically "incomplete."

The question is, do you find these "incomplete" utterances to be sentence fragments? I don't. But I do find "Not only have they hired a clown" and "Because they have hired a clown" to be sentence fragments, as I'm sure you do as well. What about the other two, though?

In what way is Either you will do as you're told incomplete? Isn't the incompleteness only in that there's a logical or-phrase missing? What does that have to do with syntax?

Syntactically, it could be said that one element of a correlative conjunction needs to be complemented by the other element. Where there is a sentence subject preceded by "Not only did . . .," the sentence must continue after that clause comes to an end; otherwise there will be a sentence fragment.

On the other hand, when a sentence begins with one of the FANBOYS, traditionally called coordinating conjunctions, the FANBOYS element is an addition that does not require a syntactic complement within the sentence -- even though, at the level of meaning, something else is required pre-sentence.

Do you consider John 3:16, arguably the most famous sentence of the New Testament ("For God so loved the world that . . . ."), to be a sentence fragment? If you do, I'm pretty sure that many people around the world would beg to differ. I see that verse, and that verse alone, even on road signs!

 

[jutfrank]

I'm glad you've brought up the definition of a sentence as a "complete thought." That's a traditional definition that I've been curious about. It seems so natural, but lately I've been thinking that it's a definition more suited to the paragraph, or maybe even to the short essay, than to the sentence.

Yes, I know what you mean. I've always thought of the traditional definition of a sentence as a complete thought as ill-defined and very problematic.

Since you find "And they have hired a clown" to be incomplete, I'm assuming that you would say the same about "They have hired a clown, too." Yet, absent context, we can no more understand the "too" in the latter than the "And" in the former, rendering them semantically "incomplete."

Right. I'd rather say they're 'logically' incomplete than 'semantically' incomplete.

The question is, do you find these "incomplete" utterances to be sentence fragments? I don't.

No, I don't, either.

But I do find "Not only have they hired a clown" and "Because they have hired a clown" to be sentence fragments, as I'm sure you do as well. What about the other two, though?

For me, there's no interesting difference between And they have hired a clown and Because they have hired a clown. The point of the conjunctions is to connect logically to another thought, whether that thought is articulated in speech or not.

Syntactically, it could be said that one element of a correlative conjunction needs to be complemented by the other element. Where there is a sentence subject preceded by "Not only did . . .," the sentence must continue after that clause comes to an end; otherwise there will be a sentence fragment.

On the other hand, when a sentence begins with one of the FANBOYS, traditionally called coordinating conjunctions, the FANBOYS element is an addition that does not require a syntactic complement within the sentence -- even though, at the level of meaning, something else is required pre-sentence.

We're thinking about this in very different ways. I don't see the value in a bottom-up approach to thinking of language as a collection of syntactic components. For me, language is a kind of thought itself, expressed by utterances and connected by logic. Yes, syntax does clearly play an important part in linguistic meaning but by no means the greatest part. Conjunctions are not grammatical things and shouldn't be analysed as such. They're logical 'glue' that connects assertions, statements, propositions, etc. in a coherent way.

Do you consider John 3:16, arguably the most famous sentence of the New Testament ("For God so loved the world that . . . ."), to be a sentence fragment?

I don't need to call it a sentence 'fragment'. It's an utterance that cannot be fully understood outside its context. The conjunction connects it logically to the wider discourse. What is incomplete is not any kind of 'structure'—what's missing is the context.

Sorry, I know this isn't helping, but it's interesting to me that so many people are still trying to understand language on the level of the sentence, which just seems to me the wrong way of doing things. But then I do also know that I have a poor understanding of syntax and so I probably don't realise fully the fruit it can bear.

 

[Holmes]

I'd rather say they're 'logically' incomplete than 'semantically' incomplete.

I'm happy to say that instead. The contextual void surrounding the use of "and" or "too" is not informationally incomplete on an abstract level. The words signal that something has been added to something else; we just don't know what that "something else" is.

For me, there's no interesting difference between And they have hired a clown and Because they have hired a clown.

For me, there's all the difference in the world between them on a syntactic level. As a sentence, "And they have hired a clown" is possible; "Because they have hired a clown" is not: it's a fragment.

I realize, of course, that people, including me, frequently use "because"-clauses as sentences in conversation. However, those "sentences" are not true sentence fragments, in my opinion, because the main clause is conversationally understood. This even happens with infinitival clauses:

A: Why did the chicken cross the road?
B: To get to the other side. (= The chicken crossed the road in order to get to the other side.)

However, as a decontextualized unit, "To get to the other side" is not a sentence. The same can be said for "Because they have hired a clown." But it is obvious that "And they have hired a clown" is a sentence, however incomplete it may be at some non-syntactic, or trans-structural, level. Isn't it?

 

[jutfrank]

But it is obvious that "And they have hired a clown" is a sentence, however incomplete it may be at some non-syntactic, or trans-structural, level. Isn't it?

Well, I wonder then what exactly is incomplete about Because they have hired a clown. What exactly do you think is missing?

Isn't the question here a matter of definition? What defines a sentence? Can we define it purely syntactically, without resorting to meaning?

lately I've been thinking that it's a definition more suited to the paragraph, or maybe even to the short essay, than to the sentence.

This sounds like an interesting line of thought to follow. I'd love to hear you develop your thoughts on this, if you'd like to.

 

[Holmes]

Well, I wonder then what exactly is incomplete about Because they have hired a clown. What exactly do you think is missing?

It's missing a main clause! Being just a subordinate clause, it constitutes a sentence fragment. Here's one way it could be made complete:

• Because they have hired a clown, a lot of kids will want to come to the party.

 

[jutfrank]

It's missing a main clause! Being just a subordinate clause, it constitutes a sentence fragment.

Sorry if it sounded like a stupid question. But aren't you saying a fragment is not a sentence if it consists of just a subordinate clause, and what makes it a subordinate clause is that it contains a subordinate conjunction, and what makes a conjunction subordinate is that it connects a subordinate clause to a independent clause? Is this not circular reasoning?

I don't think there's any syntactic structure missing at all. What's missing is the articulation of some kind of logical content. That's to say that Because they have hired a clown doesn't make sense on its own.

If we don't make it, then hey.

I wonder whether you'd consider the above example a complete sentence. Does then hey count as an independent clause?

 

[Holmes]

Sorry if it sounded like a stupid question.

I thought it was an interesting question, actually. I just wanted to see how you'd respond to my blunt, knee-jerk reaction to it.

But aren't you saying a fragment is not a sentence if it consists of just a subordinate clause

I'm saying that a word-string or a phrase is not a sentence is if consists of just a subordinate clause. Without a main clause to which it attaches, it is just syntactic debris, the ruins of what was, or of what might have been, a sentence.

and what makes a conjunction subordinate is that it connects a subordinate clause to a independent clause? Is this not circular reasoning?

I don't find it circular. In syntax, we deal with sentences and their building blocks. Certain combinations of building blocks yield things that native speakers perceive as sentences; others do not. Native speakers don't perceive subordinate clauses as sentences. That's just the way it is.

I don't think there's any syntactic structure missing at all.

And I find that amazing.

What's missing is the articulation of some kind of logical content. That's to say that Because they have hired a clown doesn't make sense on its own.

I completely agree with you there. There's a semantic component as well. But the existence of a semantic component doesn't undermine the existence of a syntactic component. Rather, one might naturally suppose that the semantics here is grounded in the syntax, or the syntax in the semantics.

If we don't make it, then hey.

I wonder whether you'd consider the above example a complete sentence. Does then hey count as an independent clause?

That's a special type of sentence. "Hey" is an interjection, and interjections, being capable of functioning as sentences, can be used as if they were independent clauses. However, I wouldn't want to call "Hey" a clause, dependent or independent, since it lacks a subject and predicate.