1. Er ist ein Mann, der weiß was er will.
  2. Er ist ein Mann, der, was er will, weiß.

I know that people use the first one more often. But I think 2nd is also correct because "was er will" is an object of the dependent clause "der weiß", where the verb "weiß" should go to the end. But I do not know for sure.

Can the object clause was er will appear between the relative pronoun der and the finite verb weiß?

  • 1
    Welcome to German.SE. What do you think is problematic if you are convinced that it is false? (I could imagine some correctness with it) Apr 20, 2020 at 14:56
  • In my opinion "was er will" ("what he wants") takes acts as the object in the relativ sentence. The object here needs to stand after the verb (weiß). Therefore the second versions is not correct.
    – redleo85
    Apr 20, 2020 at 15:52
  • @ShegitBrahm, I think 2nd is also correct because "was er will" is an object of the dependent clause "der weiß", where the verb "weiß" should go to the end. But I do not know for sure. For the first case, I saw people write this way.
    – Odgiiv
    Apr 20, 2020 at 16:46
  • So then please edit your question to break down your concerns in detail as far as you know. E.G. include your clarification comment. E.g. why you think there is a problem - to solve the puzzle by us needs an idea what is unclear for you. A simple "I don't know please tell me" does not fit with the "philosophy lived here". Apr 20, 2020 at 17:17
  • 2
    A completely valid question about word order closed for no reason at all.
    – David Vogt
    Apr 21, 2020 at 10:32

2 Answers 2


In German Text with a subordinate clause, it is often used to give extra information, that can be left out, but the sentence still makes sense without it.

Example: Mein Auto, das ich sehr gut finde, fährt sehr schnell.

You could leave out "das ich sehr gut finde" and still read "Mein Auto fährt sehr schnell." and understand the sentence.

If you put "was er will" into a nested subordinate clause, the impression could be given that it is extra information, which it is not.

If you read the sentence like "Er ist ein Mann der weiß" the information of the context what he knows is missing.

This is not "the" reason but it may be one reason why #2 it is not used in this case.

Also verbs at the end of present tense are very rare in German as i can think of it, they are usually used in past tense. It creates a "hanging" feeling of not knowing the action before the contextual information to it.

Example: "Er geht zum Einkaufen", "Er ist zum Einkaufen gegangen."

Lastly there may be a (unwritten) rule to put the verb right behind the relation, because with "der" you are relating to the "Mann" and should not rip apart the Man from his action because otherwise the direct relational context is missing.

I can not say this for sure, as i can not point to a definite rule, but as a native speaker #2 really sounds "wrong" and not just uncommon.

  • I agree. The basic point is actually that a phrase limited by commas provides additional information. In the case you quote, one would call them (parenthetical) non- restrictive relative clauses. In the case of the OP, the DO "was er will" is required: "er weiß was? "was er will". A case which sounds more acceptable to me, although not usual, is parenthetical "wenn". "Er ist ein Mann der, wenn er will, weiß". This is obviously different in meaning as compared to the "was-sentence".
    – Nico
    Apr 27, 2020 at 15:58

If you were to ask a native speaker for an acceptability judgement, they would probably reject 2). Complement clauses regularly occur in first position (Vorfeld) or at the end (Nachfeld):

Was er will, hat er schon immer ganz genau gewußt.
Er hat schon immer ganz genau gewußt, was er will.

Putting them in the middle (Mittelfeld) is less acceptable. My impression is that, for these sentences to be judged acceptable, there needs to be some other material between the complement clause and the verb in final position, e.g. schon immer ganz genau.

?Er hat, was er will, schon immer ganz genau gewußt.
*Er hat, was er will, gewußt.

Keeping in mind the constraint just mentioned, findings are rare, but they do exist.

Schlögel ist Historiker; aber er begreift sich auch als einen Schriftsteller, der, was er weiß, zu Literatur macht. SZ

[…] dass der ein Sklave ist, der, was er denkt, nicht sagen darf oder ängstlich verschweigt. disskursiv.de

[…] so sind wir die Töchter eines Mannes, der, was er tut, zum Besten seines Vaterlandes tut. Jud Süß

Er hat, was er gesagt hat, nicht so gemeint. Zwischen den Zeilen

The last example is interesting in that it cannot be dismissed as Gelehrtensprache (it's from popular fiction).

  • These are very interesting examples, but I think the "was-object construction" that you adduce covers different types despite syntactic isomorphism. In (1) the position of the "was-clause" is the canonical one. In (2), the canonical position of the "was-construction" would rather be at the end of the utterance. In (3), I wouls also say the position is canonical, but one would colloquially prepose the "was-clause": "was er gesagt hat...". So there seem to be quite a lot of fine-grained factors involved (I don't know which...)
    – Nico
    Apr 27, 2020 at 17:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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