Christophe Grandsire scripsit:

> That's the problem of literary teaching.

In this case, it is the problem of no teaching at all.  What she knows
about French pronunciation she has basically "picked up" without ever
studying French, and her intuition is good, but shaky under stress.
In general, however, you are quite right.

> LOL, strange that it's the original version that receives a specialised name,
> rather than the offspring. In French it's the contrary: "montre" is normally
> understood as an analog watch. "Montre digitale" is the other kind (or we can
> use the expressions "montre =E0 aiguille" and "montre =E0 chiffres", but then
> both kinds are separately named, and those names are used only to prevent
> confusion).

Well, that is more or less true in English too: the term "analog watch"
is not nearly so often used as "digital watch", but if one must make
the distinction (e.g. "My daughter can read digital clocks, but she hasn't
learned to read analog ones yet") then that is how to do it.  I would
say at this point that "watch" can be understood freely as being of either
kind, though a decade or two ago this was not yet true.

After all, only fifty years ago the word "computer" meant "someone who
does computations", and the extension to machines was strictly metaphorical!
Yet you would have to say "human computer" to recover the old sense today.

> Surprising that English is different in that respect...

Not utterly different, but perhaps more quick to generalize existing terms
and therefore requiring more specialized ones (aided perhaps by the facility
for creating arbitrary nominal compounds like "emergency ground automatic
destruct sequencer", or EGADS, the device the blows up unmanned rockets
when they go off course).

John Cowan        <[log in to unmask]>
        You tollerday donsk?  N.  You tolkatiff scowegian?  Nn.
        You spigotty anglease?  Nnn.  You phonio saxo?  Nnnn.
                Clear all so!  `Tis a Jute.... (Finnegans Wake 16.5)