How are proto-languages dated?

Question

This is something I've been thinking about lately. You look up any reconstructed proto-language, you'll probably find an estimate as to when it was spoken. But how can this number be determined?

From my research, these numbers came from an outdated theory that languages all change at the same rate (which has clearly been proven false), thus you can date a proto-language by the number of changes that have occurred.

Well, if this method has long since been debunked, then why are the results of it still widely accepted? That doesn't make any logical sense. Why would you continue to accept the results decades after the method used to determine said results has been debunked?

Answer 1

First, the obvious fact: Any proto-language must precede its descendants by quite some amount of time, this gives a data point for the end of the proto-language.

There are some (heavily debated, but much used) internal linguistic methods that estimate time depths by the amount of language change. These methods are called glottochronology.

In addition, external evidence is used: Does the reconstructed proto-language have a common word for some cultural artifact (domesticated animals, cultivated plants, invented or newly introduced things)? in the case, it must post-date the invention or introduction of that artefact.

Tying a linguistic group to some archeological artifacts (e.g., some style of ceramics) is also often done, although the ceramics does not tell anything about the language of its makers.

There are also attempts to align genetic markers of origin with languages and use the dates from genetic changes to estimate the age of a proto-language (but languages can be learned by genetically unrelated people).

Those methods all put together lead to some estimates that are more or less trusted in the community. There is always a rather large error margin on those estimates as well—you will see ranges covering a millennium or more for the proto-language's age.