> On Jun 18, 2026, at 10:41 AM, Bill Degnan via cctalk <[email protected]> > wrote: > > On Thu, Jun 18, 2026 at 7:54 AM Paul Koning via cctalk > <[email protected]> wrote: > . >>>>> Perhaps a better definition of a dead language is one for which there is > > >> >>> The real linguistic definition of a dead language is one which children no >>> longer learn from their naive (NB not "native") environment, i. e., where >>> the >>> language is spoken around them for purposes other than simple paedagogy. >>> What >>> this means for language revitalization is that until the generation *after* >>> the >>> one(s) studying, say, Lushootseed speaks it from infancy, it is still dead >>> (or >>> at best moribund). > > > A computer language is not exactly a native spoken language, but you > can apply this test pretty accurately. For example, Sanskrit. We do > have access to writings in Sanskrit today, it has been preserved, but > there is no country where Sanskrit is spoken natively. It's a dead > language but not a lost language.
Yes, that's what I thought too. And then I found Sanskrit on a list of languages spoken in India today. https://en.wikipedia.org/wiki/List_of_languages_by_number_of_native_speakers_in_India -- about 25k speakers of Sanskrit. Blew my mind. paul
