Alan McLeod has a post on that perennial question: are styles just more hassle than they're worth? (His actual words: " at what point do beer styles become so diversified as to be useless?") While recognizing right out of the gate that it's a question with no definitive answer (or twenty, take your pick), I'd like to piggyback on the question with a few thoughts.
I'm supposed to be writing a book, the structure of which is based on style. Actually, it didn't have to be, but the virtue of styles are evident the second you consider writing a book about beer based on something other than them. One could re-invent the wheel--group them by broad type or region--but then all the people who already use style would wonder what the hell you were talking about. And of course, breweries in most countries use style to sell their beers. For better or worse, we have to play the cards we're dealt.
I'd also put in a word in praise, too. Beer styles been around as long as beer has been around. Both the Sumerians and Egyptians had different types of beer, and I suspect that per-civilized proto-Sumerians probably distinguished between types of gruel-beer, too. Having something to name beers is really helpful. The difference between lambic and Flanders-style wood-aged beers is not incidental, even if they're both tart beers. Alan suggests that gradations can become too fine (agreed!), but they could obviously get too coarse, as well.
The difficulty is that styles shift and change over time, and sometimes, they're not well defined in the first place. I spent a fair amount of time trying to figure out what to do with saisons and biere de gardes. The fashionable trend is to group them in the genus Farmhouse ales. It's not a terrible decision--they were once a single style that branched, and the regions they're brewed border one another. Yet for the person arriving fresh and new to good beer, they don't appear to have anything in common. Biere de gardes are a lot more like Scottish ales or bocks than saisons. There are a bunch of other rustic-style ales that aren't saisons per se, either, but which have more apparent connection to them in terms of yeast character, rusticity, and so on. (I grouped saisons with other rustic beers and gave French beer its own chapter.)
Or take Irish ale, that poor, neglected style. It dates back 1,500 years--"red ales" were mentioned in the sixth century, no doubt quite a bit different, but part of a single lineage. Yet thanks to the success of stout, there are almost no commercial examples left. Among those that are, tell me exactly how they are distinguishable from other British session ales? Heritage isn't enough to sustain a style if it appears to have vanished in the wild.
I hate to say it, but Rodenbach-style red ales and oud bruins are also on serious life support, too. They've more or less been collapsed into a single style, and there are only a few breweries that still produce the style in the authentic barrel-aged fashion. If Rodenbach went out of business, could we say the style really still existed commerically?
The converse problem involves the proliferation of Frankenstein beers that glue borrowings of various styles together to make beers conforming to no style. Your IPAs made with saison yeast and peaches--that sort of thing. We shouldn't declare a new style every time the idea for a new beer crosses a brewer's mind. On the other hand, if everyone makes a black IPA, at a certain point we're just going to have to accept the damn thing.
Styles are provisional, like grammar. We have to try to keep a semblance of order and be a bit schoolmarmish about enforcement, beating back the worst offenses. But at a certain point, the barbarians will overrun the gates, and then we have to let it go. Businesspeak and imperializing we can resist, sentences beginning "hopefully" and black IPAs, probably not. So it goes.
I am willing to entertain your own style pet peeves, however, particularly if they're entertaining.