I spend a lot of time these days listening to the oldest recordings I can find, either in original condition or digitally remastered. I have for many years been in search of just what it was that characterized the singing of the late 19th and early 20th century. I always look for the earliest date of birth of the singer, not necessarily the earliest recording. The earliest recordings of anything are useless; simple curiosities such as Brahms hammering away at the piano (it could have been a xylophone) in 1889, or the recently unearthed "Au Clair de la Lune" (l860) which is basically noise. One of the most fascinating things I have found on the web is a 1933 film clip of Charlie Coborn singing his best known song, "The man who broke the bank at Monte Carlo." Coborn was born in 1852, eight years before the American Civil War, which means that at the time of this film he was 81 years old. He had only two songs in his repertoire, this one and a piece called "Two lovely black eyes." Since there were no recording media of any kind back then you could take the same song from town to town and few would already have heard it. The reason this is so fascinating is that with so small a repertoire, he must have started singing this song at an early age, since he was on the stage by the 1870's. Now add to that the fact that this film is a good electric sound recording. We therefore have an excellent opportunity to observe the dress, the acting (or at least comic song posturing), the voice and the English language as it was on the music hall stage of England in the 1870's, because it is unlikely that he changed much, if anything, between 1875 and 1933.
I find this clip endlessly fascinating. It takes us farther back than one might at first think, because it is a safe bet that to a large extent he spoke as his parents spoke, if we allow for the "dramatic" changes necessary for a stage presentation. His recitation in the middle is perhaps a good indication of how he normally spoke. If he speaks as his parents spoke, then that takes us back nearly to the 18th century.
I wish it were that easy in opera. It isn't, because of the human age factor. It would mean very little to hear an 81 year old opera singer in 1933, because their voice would only be a shadow of its peak performing quality. Enter digital remastering, which is fine but remains largely guesswork. What would in my judgment be better would be to restore, faithfully and exactly, an ancient recording machine, such as those used in 1910 (there are some still around) and ask a famous opera singer, at the height of his or her powers, to record an aria exactly as they were recorded in 1910. Comparing the playback to a modern recording made by the same singer should show how much difference there actually is between the clear,"live" voice and the shadowy version that appeared on the old machine, with the same tiny orchestra typical of early recordings; i.e., heavy on the brass because it recorded better. This would tell us a lot. It should then be possible to retro-engineer the old recordings digitally, so that they closed the same perceptual gap evidenced in the modern singer's old recording and his or her new one. Once the difference is clearly understood, and reduced to a formula, it should be possible to digitally remaster almost exactly. I say almost because it is now impossible to tell what kind of shape the old machines were in at the time they were used, or how fast they actually ran. Still, it would be close. It is possible this has already been done, and that I simply don't know about it. I don't think that is the case, however. For one thing, it would be hard to find a famous singer willing to have his or her voice, in antique shadowy mode, at large among the public. (Possibly someone like Netrebko, whose penchant pour la nouveauté seems to know no bounds.)
For the time being, however, we need to make an intelligent compromise, and try to find the oldest singer we can, recording on a fairly decent piece of machinery, while still in the bloom of his or her career, and, in addition, someone classically trained. (Caruso made some recordings while very young, but he was always a belter, even from youth, and he had very little training.) A good candidate would be Antonina Nezhdanova, the Grande Dame of Russian opera, born in 1873, and recording as early as 1910, when she would have been only 37 years old. Happily, there is a such a recording of her, and it is good to the point of adding some real substance to the idea of a golden age of classical vocalism. However, you have to wait until next time to hear it:) Stay tuned, and we can take a good look at the amazing woman most Russians consider to be their greatest soprano, and in a country of such great opera voices, that is saying something!