Too many good things in this Paris Review interview with Adam Phillips to exhaustively excerpt (is that a thing?), so just go and read it.
On not knowing oneself:
Symptoms are forms of self-knowledge. When you think, I’m agoraphobic, I’m a shy person, whatever it may be, these are forms of self-knowledge. What psychoanalysis, at its best, does is cure you of your self-knowledge. And of your wish to know yourself in that coherent, narrative way. You can only recover your appetite, and appetites, if you can allow yourself to be unknown to yourself. Because the point of knowing oneself is to contain one’s anxieties about appetite. It’s only worth knowing about the things that make one’s life worth living, and whether there are in fact things that make it worth living.
On digression (and the essay):
If one looked into digression, what would begin to fall apart very quickly would be the idea of nondigressive prose and conversation. It seems to me that digression may be the norm, the invisible norm, in conversation. Because if you believe in digression as something separate, you must believe it’s possible to be coherently focused and purposive. What psychoanalysis shows is that one is digressive whether or not one wants to be. Indeed, the digressions one is unaware of are the most telling. Even in normal conversation it’s very interesting how we pick up on each other’s digressions, not only in terms of content, but also in terms of tone of voice, so that it’s actually extremely difficult to stay on a subject. To stay on a subject you’ve got to know what the subject is.
Relationships should make us feel better. Why else bother? But there are different ways of feeling better. And I don’t think the project is to make people feel better. Nor is it to make people feel worse. It’s not to make them feel anything. It’s simply to allow them to see what it is they do feel. And then what redescription might change. It’s done through conversation, but it’s also done through the medium of who the analyst happens to be. In other words, it’s not a replicable technique. In that sense it clearly isn’t scientific, because it’s something to do with what goes on between two people, mostly unconsciously. An analyst should be someone you have an appetite to talk to and who has a desire to listen to you. Not a professional desire, which is a contradiction in terms. Analysts are people who don’t speak on the patient’s behalf, don’t speak for someone, unlike parents and teachers and doctors and politicians.
Here’s a good answer to a question I like to ask, myself and others (i.e. I’m damned):
[I]f you live in a culture which is fascinated by the myth of the artist, and the idea that the vocational artistic life is one of the best lives available, then there’s always going to be a temptation for people who are suffering to believe that to become an artist would be the solution when, in fact, it may be more of the problem. There are a number of people whom you might think of as casualties of the myth of the artist. They really should have done something else. Of course some people get lucky and find that art works for them, but for so many people it doesn’t. I think that needs to be included in the picture. Often one hears or reads accounts in which people will say, Well, he may have treated his children, wives, friends terribly, but look at the novels, the poems, the paintings. I think it’s a terrible equation. Obviously one can’t choose to be, as it were, a good parent or a good artist, but if the art legitimates cruelty, I think the art is not worth having. People should be doing everything they can to be as kind as possible and to enjoy each other’s company. Any art, any anything, that helps us do that is worth having. But if it doesn’t, it isn’t.