But the names DO mean things - whether we agree on that meaning or not. This is the reason that the "taxonomy by numbers" approach to cataloging them did not take off - obviously there is more than "just" an identifier there, despite the fact that being an identifier is indeed the most important part of what a name does. If not, why would anyone here be bothered?
To the people who named birds before the era when DNA was even a notion, a "tit" was defined as a little active bird that makes a noise that sort of sounds like that. A "flycatcher" is a type of bird that sallies from a perch to catch flying insects. The linguistic taxonomy of the situation was perfectly sensible.
But like it or not, our understanding has changed as has our language. If enough people decide that the definition of a "tit" is a Parid - then that is what it is because that is how language works. We no longer refer much to mudhens and bullbats, nor pigeon hawks and man o'war birds, nor solitary vireos and hedge accentors. All those names meant something more at one time; but we have words that we have collectively decided are "more" meaningful - enough to be adopted. If a name has doesn't mean anything, people have little incentive to adopt it.
I'll add that meaning can and has come from both the bottom up and the top down - its easy to find examples of both.