When should a supervisor be an author?
Allow me to strongly disagree with eykanal's answer. There is no universal standard. You must ask your advisor in advance what her coauthorship policy is.
In theoretical computer science (and mathematics), it is generally considered unethical to list someone as a co-author who has not made a novel and significant intellectual contribution to the paper. In particular, merely funding the research is not considered an intellectual contribution. Adding a supervisor's name to a paper to which they have not directly, intellectually contributed is lying.
In practice, writing a good grant proposal requires at least as much intellectual novelty as writing a good paper. Most of the good ideas that PIs pour into their proposals also appear in papers; as long as those ideas constitute novel intellectual contributions, the PI merits co-authorship. But that only works once per idea; once an idea has been published, it's no longer novel, by definition.
To be specific and personal:
I am not a coauthor on all of my PhD students' papers. (Of course, I still report my students' independent work back to NSF as outcomes on whatever grants supported them. So I still get credit from NSF for having the foresight to fund the student.) The same is true of all the other theoretical computer science faculty in my department.
My PhD advisor is a co-author on only one of the papers I published as a PhD student.
My PhD advisor doesn't have a single co-authored paper with his advisor.
The following answer is based on my experience in the fields of Neuroscience, Biology, and to a lesser extent, Electrical Engineering.
This is often an unnecessarily touchy subject amongst graduate students. To be clear: As a graduate student, you can expect that your advisor will appear as an author on all of your papers. He is providing your funding, your resources, and (ostensibly) is the Primary Investigator on whatever project you happen to be working on. Even if he does not contribute, you are working on his project, and he wrote the grant for it, not you. There may be situations where you will be the sole author of a paper you published during your graduate career, but those will be unusual circumstances, indeed.
That being said, you can read through the Wikipedia article on the subject, which discusses conflict. This Canadian Medical Association Journal article (thanks, Wikipedia) states that authorship ordering conflicts occurred in over 60% of published papers. To help with things, there's a good chance your university has it's own authorship guidelines (e.g., [1][2]); speak with your department.
Most importantly, speak with your advisor. Clear communication early on can help to stave off future problems, or sometimes communication will alert you to the fact that there may be future problems that you should address early on.
There are two reasons that authorship is a difficult area for students:
- Power dynamics make students suspicious of being exploited, but
- Attribution of ideas is very difficult. Supervisors can spend months or years describing an idea to a student, and when the student finally understands the idea they get the "ah ha!" sensation of discovery and think it's their own.
No one thinks less of you for having your supervisor's name on a paper since everyone knows this is tricky, so I think it makes sense to be generous – at least so long as your supervisor has agreed to have their name on the paper. If they feel they have made enough of an investment to deserve to have their name associated with your work, then I would give them the benefit of the doubt. If you can't even trust them on this, why are you taking their academic advice at all? But if your supervisor has specifically asked that you keep their name off your paper, you should respect that (and be sure that you understand why!)
In general, authorship on any article with any group of people should be agreed well in advance of submission by all involved parties, in writing (email or recorded chat sessions). The generosity principle works so long as whoever did the most work and the most writing is the first author, and efforts by the team should be made to ensure where possible that that is the same person.