Another way to look at this kind of problem is to see it as a set of tradeoffs, rather than a situation with a clear, crisp solution.
On the one hand, making involuntary commitment easier increases the risk of taking away the freedom of people unnecessarily, even opening opportunities for people to game the system to get rid of annoying relatives, enemies, rivals. In the old days, remember, that was the complaint: Too many harmless people locked up. Books, plays, and movies with this theme were common at one time: One Flew Over the Cuckoo's Nest, Harvey with his six-foot imaginary rabbit, The Curious Savage, and even Miracle on 34th Street.
So we changed the rules, and now it's much harder to commit someone involuntarily. This really means that the tradeoffs lie in the direction of not locking up someone who eventually becomes dangerous, and thus innocent people losing their lives -- sometimes just the crazy person, who commits suicide when he could have been helped.
As we move the slider back and forth, we trade security for society on the one hand and concern for the individual on the other. If we choose to change the laws to make involuntary commitment easier, we have to recognize that the tradeoff is that we will have people who will be committed who shouldn't be; it's inevitable. And if we don't change the laws, we will have people die at the hands of insane people; that, too, is inevitable.
Of course, admitting to tradeoffs is not the way to win in politics or in public forums, to we tend to absolutism: Our way is best, not simply better, it is without negative tradeoffs or costs, and the other side's ideas are uniformly bad, period. We win arguments that way; we don't get closer to any useful truths.