Last weekend I was hanging out with some other recently-tenured friends when the conversation turned to grad school and our relationships with our respective dissertation directors. We're all in different fields, went through different programs, and the nature of our advisor relationships were also quite different. But the conversation made me think, not for the first time, that dissertation directors trail only one's family of origin and certain romantic partners in their emotional and psychological impact. And as in those cases, an advisor's importance has less to do with what they actually do or don't do (i.e., whether they're objectively cruel, thoughtless, or neglectful) than with the fact that they're intimately involved in our lives at a crucial and difficult stage.
At this distance, I feel confident in saying that no one has the relationship she wants with her advisor while a grad student, just as virtually no one has the relationship she wants with her parents while a teenager; the problem is that you're not yet the person you want to be--whoever that is--and you're radically dependent on someone else. This complicates even the best relationships.
But here's the thing: unlike one's parents, the only purpose of an advisor is to get you to the point where you don't need him or her--where you know, experientially, that you can write a persuasive chapter, a publishable article, a dissertation, a book. And any advisor who helps you get there is a good advisor.
Now, don't get me wrong: there are certainly better and worse advisor relationships. Some are objectively bad (an advisor who doesn't read your work, belittles it, steals your ideas, makes sexual advances) and some are just bad-for-the-individual (a personality mismatch). And a bad relationship can do real damage. But few advisor relationships are so good that the advisee is never anxious, embarrassed, playing the suppliant, or terrified of letting his or her advisor down. And everyone has to learn, sooner or later, to trust herself and her own intuition, to find other mentors and collaborators, to do work without the (literal or figurative) voice of her advisor in her ear.
Accordingly, there are "good" advisor relationships that don't serve the advisee well: a close relationship isn't helpful if you depend too much, or for too long, on your advisor's advice or approval.
These days, I'm happy with where my relationship with my advisor is at, and I don't think much about its past. But if there's a bigger lesson here, it's that one can't escape the need for external sources of approval, especially in one's early years (as a child, as a scholar), and it's normal to cathect on those individuals or imbue those relationships with all kinds of magical thinking. But one is happier the more internally-motivated one becomes. Mostly, this just comes with time. But it's never too early or too late to try to separate one's sense of self from extrinsic sources.