A few comments - without giving any precise answers:

Firstly note that there are no fewer than four surviving clavichords by 
C. G. Hubert with the date 1784. Probably the one under discussion here 
is the one in Edinburgh, which is the best known because a good drawing 
of it has been widely available for many years and it has been widely 
copied. But it is exceptional in that it is considerably smaller than 
all the others. Alfons Huber suggests that it was intended as a 
travelling clavichord. The scalings and fretting distances might not be 
typical of the main body of Hubert's work.

> Short scalings of this clavichord indicate that it should be tuned to
> R +2 a tone above the more common clavichords ...

This comes from the data sheet prepared by Grant O'Brien and 'R + 2' is 
derived from the system he used in his book on Ruckers. 'R' means 
reference pitch, which Grant associates with Ruckers 6-foot virginals 
with a c2 scale of about 367 mm *for iron strings*. The Ruckers workshop 
produced a series of instruments related to this reference pitch by 
simple intervals, e.g. a fourth above reference (R + 4), an octave above 
(R + 8) and so forth. R + 2 in this system indicates a just whole-tone 
above reference.

But what is this reference pitch in terms of Hz? Grant is careful not to 
say. And can the system really be applied to eighteenth-century brass 
strung clavichords in the same way that it applies to 
seventeenth-century iron strung virginals and harpsichords*? Not 
necessarily, I'd say. Clavichords do not necessarily have to be tuned up 
to 'just below breaking point' and the relationship between string 
sounding length and pitch is not so simple as in harpsichords.

I think we really do not know what pitch Hubert intended for this 
instrument, and we cannot be so certain, I'd say, that the shorter 
scalings necessarily indicate a higher pitch. In practice, the Edinburgh 
one was tuned to a1=440 Hz when I examined it - probably still is - and 
it works well at that pitch.

As for the temperament (thanks to Thomas Dent for his useful post), it 
is certainly not quarter or fifth-comma mean-tone. It would be possible 
to achieve any of the common circulating temperaments (Vallotti has been 
mentioned) or even equal temperament with a fairly small amount of 
tangent bending. I am inclined to agree with John Barnes that Hubert was 
prepared to leave it to owners and tuners to adjust the tangents to suit 
their taste and judgment. I don't agree, though, that sixth-comma 
mean-tone is necessarily ruled out: as Thomas pointed out, on a 
clavichord it is possible to adjust the size of intervals with finger 
pressure as you play, so it would be possible to mitigate the severity 
of the Eb-G# wolf and the wolf thirds in this way.

A systematic examination of Hubert's scalings and frettings with a view 
to establishing something about his temperament intentions would be well 
worthwhile. The data has already been collected in Koen Vermeij's 
splendid *Hubert Clavichord Data Book* and I really might undertake this 
- when I have time!

Best wishes to all, Peter.

-- 
Peter Bavington
Clavichord Maker
291 Sprowston Mews
LONDON
E7 9AE
www.peter-bavington.co.uk

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
Note:  opinions  expressed on HPSCHD-L are those of the  individual con-
tributors and not necessarily  those of the list owners  nor of the Uni-
versity of Iowa.  For a brief  summary of list  commands, send mail to
[log in to unmask]  saying  HELP .
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::