Since it takes only one counterexample: I participate (pseudonymously) in G+.
I follow a number of people who associate by their (so far as I can ascertain) birth / given / legal names, and a number with evident pseudonyms (having gone so far as to seek out same). Several of these have noted that the apparent pseudonyms are in fact birth / given / legal names, go figure.
I use circles to categorize both inbound and outbound posts by content and context (NB: this doesn't work very well, and recent improvements in search help).
There are people I do know IRL who are on G+. None AFAIK actually know the link between my pseudonym and my legal identity, though of course, some may have done the math.
I frankly don't trust Google (or by extension its successors and/or business/government "partners") with my personally identifiable data, to this degree.
I've had the batshit crazy conspiracy shibboleth thing thrown at me before. Sometimes it's a valid criticism. Surprisingly often, it's not.
As for the "use another service" shibboleth: network effects and social/commerce expectations currently all but require use of Facebook in some contexts, and similar effects would almost certainly arise for any closed, proprietary successor. So long as social networks are closed, there really isn't a viable alternative. It's the equivalent of trying to exist without a phone or postal service in the 1980s or 1990s. Yes, it's possible, but only in a highly marginal way, and you'll find you're excluded from many interactions and services.
The real solution, the one I've been considering for some time, is an open social networking framework that breaks the silos. If social networking is truly the cat's meow, and sufficiently valuable that people demand it, and systems such as Diaspora can gain traction, I see the third-tier networks and open solutions (Diaspora, etc.) developing an open model. Think very strong analogs to how the GNU project and Linux emerged to the point that no proprietary Unix vendor can credibly be called a market leader, though several do survive (Sun as an Oracle subsidiary, AIX within IBM, Mac OS X as a consumer product, etc.).