I think we've descended out of the fuzzy world of supposition and into the nasty empirical land of testable theories. I'm not going to pretend I have any more insight into this than what I've already typed - but I will say that my pediatrician mentioned that bilingual babies generally have smaller vocabularies in both languages and start talking later, but that these setbacks are offset quickly by other gains. A brief perusal of the web seems to bear out his arguments.
I've read that, as well. Do you plan to have your daughter continue to learn Spanish? One of my regrets is quitting German lessons when I was a kid, much to my father's protest. If/when I have a child, I think the things I'll be that parent about are music, sports and learning a second language (my dad, in retrospect, was way too forgiving). Of course, with the proliferation of immersion schools, I think the language thing is comparatively easier now than it was in the 80s.
Don't forget that child-rearing books in the 1980's advised against bilingualism or multilingualism, as language acquisition was not as well understood. What people could see was that bilingual kids took longer to speak and that they'd often respond in the other language, as they hadn't yet figured out when it was appropriate to do so.
We aren't actively teaching her Spanish. We're having her spend her days at a daycare in Venice where the two main workers speak to each other in Spanish and much of the discussion is in Spanish. As such, certain words she knows in English and certain words she knows in Spanish. For example, we've been trying to get her to say "all done" for six months now and her rendition is "rearrgh." Last night, however, she turned down milk by saying "no mas."