Some thoughts

I remember several years ago reading an article talking about linguists trying to piece together Proto Indo European. By using language morphing and modeling about how current languages change and several other factors. The paper itself which was very long shows how in depth they went at many different methods. If I remember right they thought they had a decent chance of hitting the original words for finger, water, and some others. They said all combined they thought it exceedingly likely that at least one of the words they came up with would have been actually used in the context around 8,000 years ago.

Of course language is older than that. So recently another linguist I was reading on the formation of languages had a paper he included on the usage of huh? The grunt. It is used in more languages than should be natural. And also used in far too many isolated language groups than would otherwise be expected. He thought that this might be because it is a sound so old that it's one of the first words and might even be getting programmed into us evolutionarily. After all a common meaning for showing that you don't understand would be highly valuable for contacts between strangers.

There's been some work done on the possibility of getting actually spoken language of thousands year old pottery as well. After the fly wheel was invented the pottery was spun at a fairly even rate. As the brush stroked along the pottery voices and other sounds made small differences in the depths the strokes imparted. They proved that these could actually be picked up by machine. Of course what they could not do was to use AI to be able to back compute the language and what was being said. Though they certainly thought it might be possible. After all the scrolls of Pompeii or the burned remains of them left were thought to be unreadable. Using modern AI and imaging techniques we are starting to get words and even sentences from there and they do expect to soon be able to get pretty good at reading them. The possibility of getting the lost history of Fabius Pictor. Or for true gold of history the lost Anales Maximi? I can dream.

All that is to say that maybe there will be some method in the future to get really old language. Going back tens or hundreds of thousands of years. Probably not but you never know. If they ever did even the cutest girl would not be able to pull me away from looking at those results.