I know that sounds troubling, but it's actually not.
Let me give you an example.
1) I can do things through Christ who gives me strength.
2) I can do all things through Christ who gives strength.
3) I can do all things through Him who gives me strength.
Lets assume that these are three different manuscripts. Each sentence has been altered in some way. Number 1 is missing the "all." Number 2 is missing the "me." Number 3 is has "Christ" changed to "Him."
This represents three textual variants. How do we know what the original said? We compare them, and see what the majority of texts say, giving weight to the earlier manuscripts. For example, sentence 2 and 3 say "I can do all things...." so that is the most likely wording of the original.
That's how the KJV translators dealt with textual variants. But they only had three manuscripts. What did they do if they came to a place where they had three different readings?
They guessed at which one was right.
That's it. They guessed, and hoped they were right.
Since 1604, when the KJV was commissioned, or 1611, when it was completed, however, we have found more manuscripts. Like, a LOT more. At last count, there were nearly 7000 manuscripts and manuscript fragments.
"But a fragment can be a tiny scrap of paper!" you might object. And you're right. But what is the average size of the manuscripts we've found?
400 pages. That's the average.
If you staked them all up, the stack would be a mile high.
So now, instead of comparing three different manuscripts, we can compare 7000. And where 500 or even 2000 of them say something different, we can generally rely on the rest to say the same thing.
The KJV is a great translation, don't misunderstand me. But the modern translations are significantly closer to the original writings than it is.
