How likely is it that any non-Celtic language was spoken in the British Isles when the Romans invaded?

Question

We know from Roman writers the names (or Latinized versions of them) of many ancient British tribes that they encountered, the Iceni, Parisi, Trinovantes etc. but the Romans were rarely interested in recording or classifying the dialects that such barbarian tribes spoke. Romans were more interested in when the barbarians were going to become civilised and learn Latin.

It is often assumed that the Ancient Britons and Irish of the same period all spoke Celtic languages. We do have place or river names across much of Britain that appear to be Celtic survivals even in areas where the Celtic language died out a very long time ago e.g. in Eastern England there is a River Ouse in Bedford and another River Ouse in York, both of which seem to come from a Celtic word for 'water'. There are the surviving Welsh, Gaelic and Irish (and revived Cornish) Celtic languages in the West, and the limited record we have of the extinct Pictish language of North-East Scotland is usually taken to show that it was a Brythonic (related to Welsh) Celtic language.

However, the Celtic languages are a branch of the Indo-European language family that itself most likely began North of the Black Sea around 4,000 - 3,000 BC, the Celtic branch of which probably originated in Central Europe some time later, and eventually spread to Britain and Ireland, perhaps bringing the Bronze Age with them, apparently replacing whatever languages had been spoken in Britain previously.

How confident can we be that only Celtic languages were spoken in the British Isles when the Romans began their conquest of Britain in 43AD?

What is the likelihood that some pre-Celtic or other Continental related language was the native tongue of any tribe or community?

Answer 1

The answer appears to be "We don't know." We're sure that there were spoken languages in the British Isles before the arrival of the Celtic languages, because the archaeology tells us that there were working human cultures.

We cannot say with certainty that there was no tribe or clan that preserved a pre-Celtic language to the time of the Roman invasion. All we know is that there is no solid evidence for it.

There were hypotheses that the Pictish language was not Indo-European, but they were based on the unreadability of Ogham inscriptions, and claims of cultural practices, such as tattooing and matrilineal family structures, which were not considered usual for Indo-European speakers, as of the end of the nineteenth century. There weren't any records of the language indicating that it was non-Indo-European. More recent ideas have portrayed Pictish as Celtic with a non-IE substrate language, but those have been undermined by more recent archaeology, leaving Pictish as "Probably a Celtic language."

There are imaginable ways in which we could discover that a non-Celtic language was still in use by the time that the Romans arrived, but they're all pretty unlikely. For example (made up by me, just now), if we were to learn to read Ogham far better (plausible), discovered a lot more writings in it (unlikely) that could be securely dated to before the Roman invasion (really unlikely), and were to find in them phonetic transcriptions of text that was definitely not Celtic, but resembled Basque, that would be evidence of a non-Celtic language being used. Basque is not an Indo-European language, and is not related to any other living language. It may well be a survivor from the period before the IE languages spread into Europe.