(j3.2006) Interp lurking in Fortran Forum

[Van Snyder]

There's an interp request lurking on page 16 of December Fortran Forum:

        There is some ambiguity as to what the term "unreferenced"
        means.  The relevant clause of the ISO standard [Overby means
        F95] (11.3.2) states: "Two or more accessible entities, other
        than generic interfaces, may have the same name only if the name
        is not used to refer to an entity in the scoping unit."  The
        question us what "refer to" means.  Specifically, (1) is USE M,
        X => A, X => B legal if the name X is never actually used, and
        (2) if M contains a module entity named X, should USE M, X=>A, X
        =>B be permitted (in which case the local name X would
        presumably shadow the module entity X)?  IBM XL Fortran 12.1,
        GNU Fortran 4.4.2, PGI Fortran 10.0, and Intel Fortran 10.1 all
        exhibit different behaviors.

A similar question pertains to F08 11.2.2p8

        Two or more accessible entities, other than generic interfaces
        or defined operators, may have the same local identifier only if
        the identifier is not used.

Is it obvious what this means, or should we process an interp?