How do they know that? Any proofs?
You can’t prove a negative, so there’s no way to prove that a word didn’t have such-and-such meaning before a certain date. What we do is, we look at the earliest mention of a word where it clearly seems to mean X. Then we known that meaning X dates at least as early as the date of that document. In fact it had that meaning earlier (or the writer wouldn’t have used it!), but there’s no way to know how much earlier. All we can say is that by that date it already had the meaning (the technical term is terminus ante quem).
Of course, dating all earliest mentions of all meanings of all words is quite a monumental task! You English speakers are actually spoiled for choice: the Oxford English Dictionary—the big, multi-volume one—is probably one of the best such compendiums in all languages, if not the best; it’s been 160+ years in the making and by now it has accumulated a lot of data.
Whenever you are interested in a particular word, I heartily recommend you head to the nearest public library and consult the OED. The entry for "mark" lists 42 meanings (not counting idioms), all of them with copious attestations of ancient use, and with most of them grouped into two main branches:
- I. a boundary, a border between lands. Earliest English attestation is in the Metres of Boethius (880–950) xi.65: Hæfð se ilca god eorðan and wætere mearce gesette (“having set the border for the same good earth and water”, if I’m not mistaken—someone who actually knows OE please correct this).
- II. an object marking a boundary; a sign. First attestation in this family is the Old English Hexateuch (traditionally said to date around the 950s): xxiv.4. Moyses..getimbrode an weofod æt þam munte nyðeweardon & twelf mearca on twelf Israhela mægðum. (something about Moses building up the mound an altar and twelve 'marks' of the twelve Israeli tribes). It also pops up in Ælfric’s Grammar (late 10th century), denoting ‘a banner, a standard’: ‘Victricia tollite signa‘, nymað þa sigefæstan mearca (=victorious banner).
Modern ‘mark’ has has a quite complicated etymology, too, merging together at least three different cognates: Old English mearc (strong feminine declension) = boundary, borderland, present with this meaning in several Germanic languages; mearc (strong neuter declension), more related to the meaning of ‘sign’ (these may give different reflexes in other languages, e.g. Middle Dutch marke vs. marc/merc, or Old Icelandic mǫrk vs. mark); and, thirdly, as a Scandinavian loan.
This is all scratching the surface. By a rough count, the OED entry for ‘mark’ alone has some 17 thousand words (an entire novella about one word). So if you’re interested in historical details about a word, go check the OED.
