Abstract
Most grammars and text books list the element je in its correlative use as a subordinate conjunction. The present study shows that this classification is not correct. Je is rather a word or word-like element which introduces a larger constituent which occupies the clause initial position much like a question word, or more precisely an interrogative constituent. One of the pieces of evidence is the multiple occurrence of je(-constituents) inside the same clause, a finding which has never been recognized before. Furthermore it is shown – contrary to Bech (1964) – that je clauses are compatible with negation. The contribution ends with a speculative part in which je is proposed to be either a particle or else even a bound morpheme, lacking word status at altogether.
