advantages to using multiple speakers (or even multiple redundant drivers in a speaker, for that matter), would be added system efficiency - likely yeilding better dyanmic potential from any given channel/system - reinforcement of frequencies (possibly more solidified image), cancelation of any differences between speakers/drivers for less discernable distortion, and "maybe" a more diffuse or larger sound from that channel, possibly.
Balance and quality over quanity and unbalance is always going to be of a higher priority ranking, imo. And the downsides to doing such things would be that you would most likely deal with, yes, comb-filtering and phase cancelation, and holes in the bass response (probably upper and mid bass freq cancellations for guestimated distance from one speaker to the next for any likely center channel arrangement being applied above/bellow or flanking side/side).
Sitting closer to one center vs. the other would surely have you only really hearing sound from one speaker, domminating (haus effect?...forgot). And, at that, you would have to be concerned about filtering from the bass, as I said.
This would be akin to a speaker designer carefully designing a Dappolito configured speaker, where two opposing woofers could possibly canceling each other at certain frequencies, who's distance between themselves were 1/4 (1/2?) wavelengths, corresponding to specific frequencies of that length. (lol- i think this came out right?)
Anyway, if simply putting more speakers for each channel was the solution, we'd all be doing it in our systems. At the very least, adding more speakers means more acoustic issues to deal with, especially in context of multiple seating arrangements. Now if you could adjust phase or time delay for all channels, relative to how far away in proximity to each speaker in relation to each other, you could probably serve up more benefits than down-sides to this approach.
I personally think a more practical approach is to have superior quality speakers (active, hi-efficiency models?) setup optimally for maximum coverage, dynamic range, soundstage pressence, coherence, detail, flat accurate response and tonality,etc. I just think just adding more speakers, just for the sake of adding for "more should be better" kind of a theory is probably going to be troublesome for most, in all likelyhood. And, given that most people have know clue what's doing what acoustically in a system, and or how to properly work with what they got, I think this using multiples should be left alone for most of us. But hey, that's just my perspective.
Leave the tricky stuff to the professionals,and focus on quality over quantity. Yeah, that's my stance, and I'm sticking by it - lol
