Someone had claimed that the Bible was a word-for-word translation. Well, mostly. There are stories (humans are ever inventing and copying stories) of 70 elders faithfully translating words about a virgin, or maybe it was supposed to be a maiden? Or it was 72 elders, not 70? Close enough? Or why did 500 arhats get together at the juncture of oral and written sutras? Some of this is doubtless for political and marketing purposes, so that no one tribe (or nation?) is more equal, or this newfangled writing thing, we gone done it right! There is safety in numbers; three copies is better than two and with only one copy, well, who knows what errors it contains. And is someone testing or reviewing the work?
Ladders were made to match the height of the enemy’s wall, which they measured by the layers of bricks, the side turned towards them not being thoroughly whitewashed. These were counted by many persons at once; and though some might miss the right calculation, most would hit upon it, particularly as they counted over and over again, and were no great way from the wall, but could see it easily enough for their purpose. The length required for the ladders was thus obtained, being calculated from the breadth of the brick.
— "The History of the Peloponnesian War". Richard Crawley. Gutenberg eBook #7142.
Stories are not copied atom-for-atom so there can be errors, but do the errors matter? At higher levels of reading much can be changed (in addition to the languages) between Ovid and Shakespeare and "West Side Story" while still telling the same story, except now if you copy one of those too closely without certain rituals you may have to deal with various legal rituals. "Weaver girl, cowherd boy" has also seen variation, speaking of star-crossed lovers. "Ladyhawke" as well, or was that a case of convergent evolution?