You're still missing the point. Your argument assumes current copying technology. I'm assuming perfect copying technology, so that the copy is identical to the original in every possible detail, as identical as one copy of a software application is to another. One Mona Lisa goes in to the duplicator, two come out. Each of the ones that come out are indistinguishable from the original.
> ... It had been believed to have been painted between 1503 and 1506; however, Leonardo may have continued working on it as late as 1517. Recent academic work suggests that it would not have been started before 1513. It was acquired by King Francis I of France ...
So what we rely on here is custody records, all the way back to King Francis I. Also, why limit the discussion to modern copies? Again, from Wikipedia:
> A version of Mona Lisa known as Mujer de mano de Leonardo Abince ("Leonardo da Vinci's handy-woman") held in Madrid's Museo del Prado was for centuries considered to be a work by Leonardo. However, since its restoration in 2012 it is considered to have been executed by one of Leonardo's pupils in his studio at the same time as Mona Lisa was being painted.
Basically, it was a copy, painted at about the same time. Now imagine that King Francis I liked the Mona Lisa so much that he had one or more copies created. How would we know now which was the original? Only via records.