Giving charitable readings is important. But at some point the question arises "Are we talking about the same theory?" A lot of logical positivist work is not at all plausibly read as treating the verification principle as a recommendation, for instance. That's an interesting suggestion, but the logical positivists would turn out to look to be a pretty dishonest bunch if you take them just to be recommending that we exclude unverifiable utterances from the language. Such a reading would take much of the wind out of their polemical sails.

​Likewise if you view them as recommending a new concept of meaning​ which rivals that in ordinary language. For then they seem to be trading on an equivocation. Yes, what folk understand as meaningless, like "’Twas brillig, and the slithy toves / Did gyre and gimble in the wabe," is bad philosophy. But if you want to replace that concept of meaning​ with your own, you aren't entitled immediately to infer from "this is meaningless" to "this is bad philosophy".

​There will be a tension in any account of what the logical positivists are saying which attempts to hold that they are writing about our​ ordinary notion of meaning. For the verification principle is a principle for sorting utterances into the meaningful and the meaningless, on the basis of their verifiability. But how do I sort them? Well, I hear them and understand them; because I understand them, I can judge whether they are verifiable or not. But then my understanding them as meaningful must be prior to and separate from my judging of their verifiability. Relatedly, to judge that something is unverifiable I need to understand it already. There's no such thing as judging Lewis Carroll's nonsense as unverifiable. Because it's nonsense, it doesn't claim anything which may or may not exceed the possibilities of verification. I can only judge "There are undetectable fairies" as unverifiable because I know what it is claiming.

​And as Miguel points out, to judge empirically that people use the word "meaning" as the principle says, I need to have an independent grasp of the content of their claims.

​There are lots of doctrines in philosophy that "look like" verificationism from some distance--sometimes a lot of distance, sometimes not so much. I am not actually convinced that they're all bad or wrong, though because of the "look like" in the previous statement, the claim is exceedingly broad. But the sort of verificationism defended by the actual logical positivists is not, I think, very defensible, even if one is exceedingly generous in interpreting them. (It's not clear that logical positivists could not reply to some of these arguments. For the most part the possible replies do not seem very promising either. But the absence of anticipation of them is what indicates that they don't hold the more sophisticated views.)

Well, I think your answers are pretty much correct.
After thinking about it a little much, I think that basically the three ways that he's using is basically the same way to redefine what metaphysicans call truth, and then doing a fallacy of equivocation, which is pretty much useless. And I also think that Greg's remark for the difference beetween verification and understanding is devastating.

Thank's everyone.