It seems obvious to me that there are cases (not only involving learning) in which people quite reasonably do things neither for your reason 1 nor for your reason 2. For instance, people often do things in order to be paid, even though they don't enjoy them all that much for their own sake and they don't do much to build skills. People do things to impress other people. People do things because they consider they have a moral obligation to do them. Etc.
The idea of "learning for its own sake" may be a puzzle even for a steelmanned version of your dichotomy -- e.g., you'd think people do things either because they directly enjoy them or because they are a means to some directly enjoyable end (though I think even that is too narrow, actually) -- but as it stands there are just waaaaay too many obvious exceptions to your dichotomy for the fact that something doesn't fit it to be puzzling at all.
For what it's worth, I think a leading reason for the particular phenomenon of learning-for-its-own-sake is simply that these people have spent years in an education system that consistently praises and rewards them for learning. It would be hard for them not to emerge with a feeling that learning is good for its own sake. (Which it might be -- or it might be consistently valuable in ways that would be hard to predict in advance -- and those may be among the reasons why the education system is that way.)
A priori, it seems to me that one should engage in activities only if they satisfy at least one of these two conditions:
But investigation we did in connection with our research at Cognito Mentoring led my collaborator Jonah and me to notice that a fair number of people seem attracted to learning things for the sake of learning, although they neither have an internal belief that learning the subject is important, nor a deep interest in learning that specific topic. Some of them then get frustrated that they're unable to make progress on their self-set learning goals, and this may harm their self-esteem (and put them off learning more important things later). Others may experience success that encourages them to learn more things, some of which may be interesting or important. Our page on managing your time generally advises against participating in and focusing on such activities, or at minimum critically considering whether the activities are sufficiently important to justify engaging in them even if one doesn't find them interesting.
However, Jonah and I may be missing important perspectives. I've heard claims that engaging in activities that are neither interesting nor important has intrinsic value -- it helps build character, makes one grow as a person, or it just might turn out to be important.
This school of thinking is reflected in diverse quarters. Tiger Mom Amy Chua famously forced her daughters to learn musical instruments to build their character, even though at least one of her daughters found it a terrible experience, and there was no reason to believe that the activity itself is important. The belief that one should try and learn new things is also widespread (albeit in a very different sort of way) in the rationalistic self-help community.
What's going on? Some possible explanations:
What do you think?