The recent question about möchten reminded me of something I stumbled across:

Man möchte meinen, sie sei schwanger.

The following are results from COSMAS II (Corpus Database):

man möchte meinen         257
man mag meinen             13
man wollte/will meinen     11
man könnte meinen        2154

Now this is interesting, as können can – and will be – often used in a context, where mögen could be used as well:

Heute mag es regnen
Heute kann/könnte es regnen

Actually, mögen is rarely used today. Yet I believe that möchte within man möchte meinen does not stand for want, but it is the true Konjunktiv II of mögen, denoting that you might think something. It would then be the last domain where möchte has survived in its original sense.

Am I correct about this?

  • I somewhat deopinionated your final question but it could still do with some further specification. (I also replaced “Konjunktiv Präteritum“ with “Konjunktiv II“, because the former is confusing for learners.)
    – Wrzlprmft
    Sep 25, 2014 at 14:27
  • Konjunktiv I and II are gruesome, it hides the beautiful sytematic of german, especially in comparision to other indoeuropean languages. On the other hand, one could use Indikative I / II and Perfekt I/II, which would be systematic in a way that I is present and II is past (although stupid).
    – Veredomon
    Sep 25, 2014 at 14:37
  • But it does not confuse learners into thinking that the both Konjunktivs distinguish temporally.
    – Wrzlprmft
    Sep 25, 2014 at 16:03
  • I don't quite understand what your question is.... you make some conjecture about mögen and können and ask if you are correct.. the key part is that conjecture. I have no idea what exactly you are trying to say....
    – Vogel612
    Sep 25, 2014 at 16:46

3 Answers 3


"möchte" is not rarely used. It is normally used for polite wishes in a shop or restaurant.

Ich möchte ein Paar Sommerschuhe.

Ich möchte etwas essen/eine Tasse Kaffee.

In "man möchte meinen" "möchte" expresses the idea: es ist möglich/ die Möglichkeit besteht. Another variant, as often used, is "man könnte meinen".

  • If you count the appearances of modal verbs in general, "mögen" became very rare, I think it was about 1 % (I did it recently).
    – Veredomon
    Sep 27, 2014 at 13:27

This is an interesting case. In English, one could say something in the lines of "I think she's pregnant" or "I'd say/guess/bet she's pregnant" to have a similar effect. Since the extra part "I think" and "I'd say/guess/bet" in such clauses is equivalent to an adverb such as "apparently" or "probably", Halliday and Matthiessen consider that they are a grammatical structure standing for another grammatical structure. For this reason, such expressions receive the name of "grammatical metaphor" (An introduction to functional grammar, 2004; Halliday's introduction to functional grammar, 2014).

Under this view, all of the n-grams that you have listed would not be clauses but only something that looks like a clause but actually functions as an interpersonal modality for marking how sure the speaker is about what he or she is saying. The "möchte" would be just a part of this structure, and not a representation of the non-actuality of a thought.


If it were a "true" conjunctive II,

Man würde meinen mögen, [...]

would sound natural, but it doesn't (to me), so I'm going for "no".

The corresponding periphrasis of Das wäre dann wohl geschafft, where wäre is not a true conjunctive, too, doesn't sound natural either. (Similarly: Wie viel wäre...?, Das wären dann..., Könnten Sie...? etc.) Although these "false" conjunctives are not all the same, since in the examples something already is (...), whereas in your example you don't quite think she's pregnant (like you maybe think she lives in Zürich). I think one could go so far as to say that it's the same möchte as in

Ich möchte gehen.

You want to go, because you have more reason to than not to, but you're not quite there yet.
You want to think she's pregnant, because you have more reason to than not to, but you're not quite there yet.

  • You can just distinguish what you call “true” and “false” conjunctives II as irrealis and not irrealis. Or the other way round: Whether a conjunctive II can be replaced by a würde construct, is actually a good test to find out whether it’s an irrealis or something else.
    – Wrzlprmft
    Sep 26, 2014 at 12:42

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.