There are two useful tests when evaluating the value of words like this:
-
Do people use it as a form of self-identification? If they do, that’s probably the real definition. If they don’t it’s probably just an insult.
-
Does the word have a consistent definition? If the definition frequently shifts to suit the needs of the speaker, it’s probably not a real definition.
I really like your first point! Second one is a little tricky. It’s not just a fluctuation with an individual, but rather the difference between groups. Bottom line, the consistent definition depends on your own exposure to it, if you’re not going by what others claim to be “is the most frequent.”
The second one is definitely a bit trickier.
I think there are two major forms of inconsistency that matter most.
-
When the parties in a conversation use different definitions for a word, they will just argue in circles. They may both have good points but neither party will understand the other. That’s often fairly easy to resolve, “I can understand your point if we use your definition of X. We can also see how my point stands if we use my definition of X. How about we call them X1 and X2 so we don’t get confused?”
-
When one party uses different definitions of the word it’s fair to ask them to pick one or to be explicit about when they’re shifting definitions. When someone says, “I believe Y because X is TRUE and I believe Z because X is NOT TRUE,” they’ve introduced a huge logical hole which needs to be addressed.
