*, ** and raised crosses

[bwald]

----------------------------Original message----------------------------
The identity of * in both synchronic and diachronic studies has always
struck me as non-gratuitious.  It means "unattested" in either case.  So,
in diachrony it indicates a reconstruction, as a *relevant* unattested
form.  In synchrony it also means "unattested" in a *relevant* way,
generally on the basis of a "grammaticality" judgment (but, in the best
cases, consistent with lack of documentation of occurrence).  Otherwise it
does not come up (unless you consider "dialect" disagreements, which should
be relevant to diachrony / language change -- otherwise, they're usually
called "idiolect" disagreements) .  Because of the interests of synchrony
it remains to further interpretation as to whether or not the synchronic
use of * has diachronic implications.  It does if the *ed example
previously occurred in the language, but no longer does.  Householder, who
I think is credited with the synchronic extention of the symbol, built in
the possibility of a diachronic consideration.  I like it.  It would be
interesting if ** correlated with second-level reconstruction, as Venemann
suggests, and with "even less expected" in synchronic analyses.  But I
don't really think it's as useful as simply the single *, and it could be
misleading in its pretentions to greater precision.  -- Benji