Duolingo states: “In German, conjunctions do not change with the case (i.e. they are not declinable).”1 I started to think of languages I know, and I don't remember any which would have this property.

So, are there languages where conjunctions are declinable? How does this mechanism work there?

1: English->German topic on “Conjunctions”. Seems that the page is unavailable for unregistered users.

  • That is certainly true for German, because only nouns, pronouns, and adjectives can be declined. I suppose it depends on what you mean by conjunction, or more likely, what Duolingo thinks it means. I could think of situations that one might want to describe that way, but generally there are better solutions available.
    – jlawler
    Jun 29 '12 at 22:19
  • 1
    I suspect that the sentence was intended to contrast conjunctions from nouns, pronouns and adjectives; rather than to contrast German with other languages. I doubt whether many Anglophone learners of German would have encountered any languages at all in which conjunctions are declined.
    – user780
    Jun 30 '12 at 0:35
  • It seems reasonable from a functional standpoint: since two conjoined nominals should have the same case, case could just be marked on the conjunction. This situation would only arise, however, if the conjunction grammaticalized from something which was itself declinable in its earlier function. The necessary diachronic path is probably to blame for the rarity of this type of marking possibility.
    – user483
    Jul 2 '12 at 13:30
  • It seems that the cases described of declinable conjunctions tend to not be coordinating but subordinating. Jul 3 '12 at 2:42

This is unusual but possible. For example, Christa König argues that in Ik, a Kouliak language spoken in Uganda, nearly all lexical items can be case-inflected.

See examples (48)-(50) on p. 155 in Coding participant marking: Construction types in twelve African languages or have a look at König 2009 in The Oxford handbook of case:

enter image description here


I have seen claims that some dialects of German [PDF] have conjunctions that exhibit agreement morphology, e.g. the remarkable second-person-singular subordinating conjuction daßst in:

Wir haben gewußt, daß-st du in Salzburg war-st
we have known that-2s you(s) in Salzburg were-2s
‘We knew that you(s) were in Salzburg.’

vs. second-person-plural daßts in:

Wir haben gewußt, daß-ts ihr in Salzburg war-ts
we have known that-2pl you(pl) in Salzburg were-2pl
‘We knew that you(pl) were in Salzburg.’

I do not know much more about this, nor what dialects they are, except for a mention that they are Austrian.

To be sure, these conjunctions exhibit only person-marking, not case-marking.


As Mark Beadles says in a comment, declinability seems relevant mostly for subordinating conjunctions, since these make the subclause into a POS of main clauses or other subclauses (Not sure if any languages use anything similar to cases to coordinate main clauses). Also, declinability seems mostly relevant for (let me know if there is a better word) nominalizing subordinating conjunctions (English 'that') since other subordinating conjunctions almost always head adverbial phrases.

If we stretch the definitions and classify Japanese nominalizers 'koto' (and 'no') as nominalizing conjunctions and also classify (certain) Japanese particles as case markers (although these are usually analyzed as syntactic constructs, not morphological declensions), we could argue that Japanese has declinable conjunctions:

Kare-ga kita koto-ga saiwai datta.
He-NOM came that-NOM luck was.
It was lucky that he came.

Kare-ga kuru no-wo matte-iru.
He-NOM comes that-ACC waiting is(/am/are).
(I) am waiting for him to come.


I don't really have a linguistic background and I'm not even sure if this qualifies as true declension but there's certain compound words that act as conjunctions semantically in Arabic that seem to fulfil what you're looking for.

Let's take the word ma3-'anna (3 representing a voiced pharyngeal fricative ʕ ). It's two separate words ma3 (with) and 'anna (that) acting semantically as a conjunction with the meaning "even though" and it just so happens that 'anna is a word that acts a bit like a verb by taking on a myriad of "pronoun endings" that indicate person, number, and gender:

ma3-annahu (even though-he)
ma3-annahaa (...she)
ma3-annahum (...they)
ma3-annahumaa (...they, dual)
ma3-annahunna (...they, feminine)
ma3-annaka (...you, singular masculine)
ma3-annaki (...you, singular feminine)
ma3-annakumaa (...you, dual)
ma3-annakum (...you, plural masculine)
ma3-annakunna (...you, plural feminine)
ma3-annii (...I)
ma3-annanaa (...we)

an example of its use can go something along the lines of:

لم يذهب إلى المدرسة مع أنه لم يكن مريضا
lam yathhab ila 'l-madrasati ma3-annahu lam yakun mareeDan
not go-he to the-school even though-he not be-he sick
He didn't go to school even though he wasn't sick.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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