>  [Barr:] Note that Scott's "phonautographs" were NOT "sound recordings" in
>> the
>> modern sense of that term!
> [Biel:] I disagree.  Sound recording and the reproduction of the recording
> still are -- and have always been -- two separate and distinct things.

I've seen both arguments made before, but I fall on the Biel side.

Let's say I record myself saying "hello" into a computer microphone so that
I can open the WAV file and look at the data spectrographically.  I study it
and learn something about pitch contours or formants or whatever.  Then I
delete the WAV because its purpose has been served.  It was never played
back, and there was never any intention of playing it back -- it was for
visual apprehension only, just like a phonautogram.  Was it therefore not a
"sound recording in the modern sense"?

> Patrick brings up a theory (if I interpret Patrick's article correct;y)
> that Edison thought intelligibility was a function of frequency, so his
> cylinder phonograph might have been usable to examine that (if they didn't
> get sidetracked because the damn thing worked so well!!!)

To summarize: I found that Edison was working in July-December 1877 from the
assumption that articulate speech consisted acoustically of a single
frequency that varied only in peak amplitudes, and not in the "shape" of
complex waveforms.  The significance of "shape" was discussed in virtually
all American publications I'd seen (or have seen since) on the
phonautograph.  Thus, Edison struck me as strangely ignorant of a fact it
should have been impossible for anyone to miss who read even the most
fluffy, popular account of Scott's work in the American press closely enough
to steal the idea.  Instead, his path to the phonograph seems to have begun
elsewhere -- I suggest that the "keyboard talking telegraph" was a major

But I'd welcome solid evidence to the contrary.  I suspect there's a lot
more information to be found in Edison's telephone notes than I've yet
brought to bear on the question.

There are other cases in which influences that seem "obvious" in retrospect
don't seem to have played a role in actuality.  Emile Berliner claimed not
to have known about Charles Cros when he first started work on the
gramophone.  I believe him.

- Patrick