How do people measure which tuning freqency were used during Baroque period?

Question

In a website decribing the story of A=415Hz, it points out that

In the Baroque Era, pitch levels as high as A-465 (17th century Venice) and as low as A-392 (18th century France) are known to have existed. A few generalizations can be made:

However, how do people know what tuning freqency were they using? For example, if you found a harpsichord from 16th century, and you measured A=392Hz when trying to play the A4 note. How do you make sure this frequency is not due to out of tune of the instrument?

I am curious about how do people know the tuning in the past, specifically, before the invention of audio recording?

Answer 1

Its mostly wind instruments:

HAYNES, Bruce. A history of performing pitch: the story of'A'. Scarecrow Press, 2002.

writes on page 4:

In terms of numbers, I was able to consult the pitches of many surviving original instruments, thanks to a grant from the Canadian Social Sciences and Humanities Research Council. The present book regularly refers to this information, which is included in summary form in the appendices; these list the pitches of some 127 cornetts, 28 Renaissance flutes, 292 traversos, 317 recorders, 70 clarinets, 540 organs, and 13 pitchpipes, for a total of 1,387 original instruments. The appendices include only instruments whose reliability I trust. Of these instruments, about 222 are Italian, 208 French, 544 German, 192 English, 11O Dutch, 77 Belgian, and 31 Austrian.

A change in temperature from 20 to 24 °C will result in a pitch change of 11.7 cent, (PISANI, U. Effect of a Local Temperature Change in an Organ Pipe. Acta Acustica united with Acustica, 1976, 35. Jg., Nr. 2, S. 132-136.) which is not that much, but there are other factors, e.g. wood may change over ages.

Bruce Haynes gives a number for pitch pipes compensated for wood shrinkage:

To compensate for wood shrinkage, pitches given here are 5 Hz lower than pipes presently play.

But not only wind instruments play a role, e.g. the tuning forks of Handel and Mozart are preservered, but

The problem with tuning forks is to relate them with assurance to a particular place, time, or usage. Unlike pitchpipes (which are often stamped and which give note-names), forks offer few clues to their date of manufacture and use, or even where they originated. (page 31, HAYNES)

And even some stringed (e.g. not bowed) instruments can be taken into consideration because of the physical properties of the used strings:

Wraight found that the most common string-lengths for the note f2 of Venetian instruments made between 1523and 1594were 235, 246, 255, and 265 mm, particularly 235 and 265 mm, which would notes a whole-tone apart from the same key of the keyboard. Since at that time there were two important Venetian pitches a whole-tone apart, mezzo punto and tuono corista, it is logical to associate the two string-lengths with the two frequencies (about 464 and 413 Hz, respectively).

(HAYNES, p 23)