Possible Duplicate:
Changing subject and verb positions in statements and questions

Look at the following questions - can anyone give a simple grammatical explanation as to why we put the verb to be at the end of these questions:

Can you tell me where the hospital is?

Do you know where the children are?

Do you know what the answer is?

This is confusing to learners as they are used to seeing the verb to be in questions at the beginning such as:

Is the hospital near here?

Are the children in school?


6 Answers 6


These clauses:

where the hospital is

where the children are

what the answer is

(sometimes called "interrogative content clauses") are not really questions. Although they're introduced by a question word such as "where" or "what" or "when", they don't usually have subject-auxiliary inversion:

I told him where the hospital is.

Where the children are, and where the children are not, has been debated for centuries.

I know what the answer is — Sally told me — but I don't understand why it's the answer.

An "indirect question" is a statement that uses one of the above clauses, and whose purpose is to report on a question without actually asking it:

He wanted to know where the hospital is.

For centuries, gray-haired professors have asked each other where the children are: for they never look out the window to see.

I asked Sally what the answer is, and she told me, but she wouldn't explain why it's the answer.

In colloquial usage, you'll sometimes see subject-auxiliary inversion used in indirect questions — especially when the speaker doesn't know the answer — though this is considered incorrect in formal usage. When this is done, the sentence is sometimes pronounced with question-like intonation or written with a question mark:

(colloq.) He wanted to know where's the hospital?

(colloq.) You want to know where are the children? I don't know, ask someone else.

(colloq.) I asked Sally what the answer is, and she told me, but she wouldn't explain why is it the answer.

One special case of this use is actually to implicitly ask the indirect question, by using it as a complement in a clause that expresses lack of knowledge, desire for knowledge, or a request for knowledge. In formal usage, the same rule obtains as above:

I want to know where the hospital is.

Tell me where the children are.

(perhaps slightly informal) I wonder what the answer is?

The only thing different about your examples from these is that the surrounding clause is, itself, a question:

Can you tell me where the hospital is? (or colloq. Can you tell me where's the hospital?)

Do you know where the children are? (or colloq. Do you know where are the children?)

Do you know what the answer is? (or colloq. Do you know what's the answer?)

with subject-auxiliary inversion in the main clause. Note that questions do not trigger subject-auxiliary inversion in subordinate clauses:

Did you do it because you wanted to, or because you had to? (not *"Did you do it because did you want to?")

Where was it when I needed it? (not *"Where was it when did I need it?")

and this is just one particular instance of that.


The examples you give:

Can you tell me where the hospital is?

Do you know where the children are?

Do you know what the answer is?

do not have the verb at the end of the sentence because it is "put" there (presumably, by the Question-Formation rule).

Rather, they have the verb at the end of the sentence because the rest of the end has been moved to the front of the sentence by the Question-formation rule, exposing the naked auxiliary verbs like worms in the dirt.

You're right, this can be disconcerting to learners.

Here's how it works. You start with a normal statement

  • The hospital is at First and Main Street.
  • The children are over at Aunt Martha's house.
  • The answer is 2x + 3xy - 7y.

Except that you don't know the boldfaced information, and you want to know it. So your sentence is really

  • The hospital is where?
  • The children are where?
  • The answer is what?

Normally when you make a Wh-question, you invert the subject and the first auxiliary verb and also move the Wh-word to the front, but -- and this is the sticky point in these constructions -- you don't invert the auxiliary verb and the subject in embedded questions, but you do move the Wh-word anyway, leaving the auxiliary verb at the end of these embedded questions, with its complement phrase transformed into a *Wh-*word and moved to the front as a marker.

These are all embedded questions -- polite versions, not the real interrogatives themselves.

  • Where is the hospital ~ ~ Can you tell me where the hospital is?.
  • Where are the children? ~ ~ Do you know where the children are?
  • What is the answer? ~ ~ Do you know what the answer is?

(Remember, you did ask "why?".)


A simple answer is that your first questions express two ideas:

[Can you tell me] [the location of the hospital]
[Do you know] [the location of the children]
[Do you know] [the identity of the answer]

whereas your second questions express one idea:

[the location of the hospital]
[the location of the children]

and, rather like German, the second idea has the verb at the end.

However it applies only to questions. For statements the order is conventional English statement-order with the verb second.

[I think] [the railway station is over there]

A limited analysis would indicate that this is a rule for questions with two simple ideas (with certain exceptions, I'm sure) rather than the two-idea questions being an exception from the "rule" mandating "verb at the beginning".

One exception I've thought of is

[Do you know] [if the postbox is red]

which has a conditional clause and consequently isn't two simple ideas.

  • 2
    It doesn't apply only to questions. Example: Do you know where the keys are? I don't know where the keys are. I know where the keys are. Mar 20, 2012 at 11:20
  • Your last (counter) example is that way because english never (in modern, non-poetic usage) places an adjective before the verb is.
    – Marcin
    Mar 20, 2012 at 12:28
  • [Do you know] [if the postbox is red]: There is no conditional clause here. It is an indirect question.
    – Irene
    Mar 20, 2012 at 15:06

Can you tell me where the hospital is?

Do you know where the children are?

Do you know what the answer is?

Each of these sentences contains a main clause, i.e. Can you tell me and Do you know, both of which are in the interrogative form. The rest in each of your examples (where the hospital is, where the children are and what the answer is) are subordinate clauses. They function as objects of the main verbs tell and know. They can't stand alone as independent clauses without the main clauses that precede them (in their current form, that is).

You can explain to your students that it is not grammatical to have both the main clause and the subordinate clause in the interrogative form. Since the main clause is in question form, the rest follows in the affirmative because technically you are asking someone if they know something (that something is substituted by an indirect question).

You can have two question forms in a row in the same sentence only when you join two main clauses which are in the interrogative form, e.g. Where are the children and where is their mother?


I can think of two reasons (although I'm sure there are more).

First, I think

Do you know the answer?

is better than

Do you know what the answer is?

But, as for why we sometimes say the latter - can I tell you what the reason is?

When we speak, we construct the sentence in our minds as it's spoken. If there's a better way to phrase a question, it's a little hard to fix the sentence midstream. People don't generally say:

"Do you know what the..."

and then trail off, thinking, "Oh, no! This is bad... I'm going to end this question with a verb!"

(On the other hand, writers have that luxury. They can proofread, and revise any awkward prose until it reads more smoothly. But not all authors proofread their writing so rigorously.)

The second reason? It's acceptable English. You said, "This is confusing to learners..." Frankly, not all English is written with the "learner" in mind. While I can sympathize with the confusion of the neophyte, English would be a much duller language if we weren't allowed to deviate from a few standard forms.

  • Thanks for that, problem is, if you are an English teacher and learners ask you why in class and you can't answer (as happened to me yesterday!) then it's a bit embarrassing! Thanks again. Mar 20, 2012 at 10:06

In the case of can it might be because it's a modal verb and the sentences follow the pattern of German, and other Germanic languages, where the modal verb can means the main verb be moves to the end of the sentence

Not the answer you're looking for? Browse other questions tagged or ask your own question.