RE: IPA Null Consonant

From: Kent Karlsson (
Date: Mon Jun 02 2003 - 15:26:01 EDT

  • Next message: Kent Karlsson: "RE: IPA Null Consonant"

    > > > SMALL LETTER O
    > > > > WITH STROKE] are both ruled out as their semantics is
    > > > totally wrong.
    > >
    > > Not at all (as seen by example Jarkko quoted!). In Danish
    > > and Norwegian,
    > > yes. But in Swedish and Finnish that vowel is written ö (and Ö).
    > Uhhh, sorry, I must have not been clear enough. The symbol
    > used in the Finnish
    > morphology studies is NEITHER the ö nor the ø! It is used to
    > mark the _absence_ of
    > of a morphological unit:
    > jalka -> jalan
    > shows the change
    > k -> EMPTY SET

    Uhhh, sorry. I must have been less clear than I intended.

    This neither indicates nor proves nothing of the kind. What IS shown
    by your example is that a "slashed circlish shape" is used to explicitly
    denote at least one kind of deletion in at least one context. I have no
    problem with this being used for several different kinds of deletions, or
    similar, or this being used by convention by many linguists; it's just that
    your example does not show that. Note that there in no set here, not
    even an empty one. To be nitpicking: the empty set IS something, it's not
    nothing! The empty string is also something, but this something is a unit
    for string concatenation.

    What character this shape is, is harder to determine. Indeed,
    the reference you give contradicts your statement. And I see no
    problem in principle to have a letter, which in other contexts stand
    for something else, in some specific contexts explicitly denote
    deletion (of some kind, or similar).

    Ok, maybe I'm overinterpreting your "shows" here. I think you mean
    "indicates" rather than "proves". It may still be a borrowing from set
    theoretic notation, and Ken gives an argument that that is at least
    sometimes the case.

    See further my responses to Peter and Ken.

                    /kent k

    This archive was generated by hypermail 2.1.5 : Mon Jun 02 2003 - 16:10:05 EDT