I can see that the bank ended up with $4 less than it planned to, but I don't see that as money being destroyed. It happened because the original estimate of hypothetical dollars was wrong. (Also if the hypothetical dollars are "money" then you're double counting it: creating $10 in circulation has required creating $17 overall which strikes me as poor notation to say the least). (Also if all had gone according to plan, the extra $4 would have come from the pockets of people buying apples. That $4 is still in circulation, either in the same pockets or it got spent elsewhere).
Suppose I buy a painting. I believe it to be an original Van Gogh so I pay $10 million for it. I then find out it is fake, and worthless. Was $10 million (of money) destroyed? Of course not, I just mis-valued an asset. Suppose it then turns out to be real after all. Owing to the fascinating history of this painting it is now valued at $20 million. Was $10 million of money created (relative to the moment when I originally thought it was a Van Gogh)? No. Was $10 million of wealth created? Yes as the world now has one more thing worth $10 million in it.
Money != wealth, even in the materialist sense where wealth consists purely of goods and services. Money is a metric we use to keep track of wealth, and in general it's considered helpful if that relationship holds, so if we're trying to maintain that relation rigorously the central bank should print another $10 million (or create it by making loans) to reflect our knowledge and appreciation of the Van Gogh - if it doesn't then the existing fixed quantity of money in the system will now be representing a greater quantity of wealth, causing deflation.
As I said in my other post I am not an economist by training. If the economists want to call this thing that got created/destroyed here "money" then I guess I should let them, but I would like to hear a good reason why it makes sense to do so, and I haven't heard one. Absent of a good reason I might as well call it haddock. Or, considering the OP was asking which things that could be explained better, we could acknowledge what any good programmer knows: part of a good explanation is choosing the right names for things.
As I think of it, credit (what got destroyed) and currency (what was used to create the credit) are interchangeable - $10 of credit buys as many seedlings as $10 of currency, hence we think of them as one 'thing' - I might call 'money' the category including both, though I'm no expert in economics either and these might be nothing like the official jargon. But what does seem reliable is that destroying $4 of credit has the same effect on the price of apples as would destroying $4 of currency, though the latter is a more rare event.
