The language or the dictionary? Surely the true purpose of a dictionary is to document a language, not to define it?
That's certainly my view, but some seem to disagree.
But what it documents is the meaning, i.e. the definition of words. So it does define it. The OED has a quasi-official status in that regard - you hear of new words being added to it, and the sense is that it is that process which gives them validity.
Fair enough. However, what I presumed dave1x was getting at was the question of where any changes/additions to the dictionaries originated. He, like myself, seems to feel that dictionaries are, when necessary, updated to reflect changes in common/general usage - i.e. to document ('define', if you wish) the language 'as it is being spoken'. It is not, IMO, for the writers/publishers of dictionaries to, themselves, make the initial decision to change or add meanings unless those changed/new meanings are already in common use. In other words, they document and define the meaning 'as it is being used', but (IMO) they should not themselves attempt to dicate what new meanings should be given to words, if the words are not already being used commonly with those new meanings.
As I've said, a language could never change ('evolve' or whatever) unless people started using words in a way not currently documented in dictionaries.
You have, but you have not been able to show why this is necessary, or even beneficial, or what harms or disadvantages would accrue if it did not happen.
I don't think I've ever expressed such a view. What I have said is that this is a debate for academic linguists, many/most of whom seem to believe that evolution of language is a desirable thing - and, as I think we've agreed, such evolution, if it were deemed desirable, could never happen unless people start, and continue, using words in a manner different from what is documented in current dictionaries.
We of course need new words all the time, to describe new things, but we never need to take an existing word and pervert its meaning.
Agreed - we never "need" to do that - but change of meaning (what you call "perversion") is one of the major ways in which evolution of language has always occurred, and probably always will.
Kind Regards, John