That depends on the child. It's not the case that every child would rather know they are loved than that they are intelligent. Love doesn't create any fewer unwanted obligations than intelligence does.
Parental love and fear to disappoint parents in a force that drives child to internalize values and to prioritize them. What that values are depends on parents, it could be compassion, love, greed, competitiveness, hard work, intelligence, beauty, knowledge, it could be anything. It is good because it is the only known way to a person to form. It is a way for parents to pass their values to a child. Child seeks love from his/her parents and avoids rejection, using any tools he/she have. Then tools become values and motives and get an ability to drive behavior by themselves, even without any parent nearby.
But the trouble is, if between those values there are personal traits that cannot be possible altered by a child, then he/she is trapped. If parent constantly shown to a girl that he loves her when she is pretty and rejects her when he is not so pretty, and then she compared herself with models and found that she is not so pretty as them, than she would be unable to meet standards of her parent without a plastic surgery. If parent shown signs of rejection to his child (just a visible disappointment might be enough) when he failed to be smart enough, than child face a risk to grow not smart enough to meet standards of his parent, and he would have no way to meet them: he couldn't be smarter than he is.
Love of parents should be unconditional. It solves the most of problems of such a kind. But if you wish to praise your child, praise for something that is really an achievement, for something that is under control of a child, not for something that he got from a birth and couldn't control. Not for intelligence, but for a hard work. Not for beauty, but for... I don't know... for love, compassion or care.
From a behaviorist point of view, when you praise your child for something that not under his control, you are replicating a famous Skinner's experiment with a random reinforcement[1]. If you are punishing child (for example, showing signs of disappointment) for his failure to be smart (i.e. for the reasons that were not under his control), then you'd replicate learned helplessness[2]. In reality it is more complicated, because child would face as rewards so punishments, but in a short it would make no good for a child.
