I'm a doctoral student who is trying to graduate. The requirement at my university is that students write 3 publishable manuscripts. Because my adviser knows that I don't want to stay at my current university for a post-doc, she wants me to submit my manuscripts before she'll let me graduate. I spoke to the assistant chair of my department to ask if this was normal, and she said that it's understandable because if a student leaves and doesn't publish, the adviser has to "take responsibility for finishing and publishing the papers." This sounds weird to me - shouldn't I be able to decide if I want to publish my dissertation work? Obviously, I'd like to get them published, but it seems unreasonable that my adviser could take my papers and publish them herself if I don't do it in the time frame she wants.
I think there's a disconnect between what the assistant chair means when they say "take responsibility for finishing and publishing the papers" and what you're interpreting it to mean.
What I imagine they mean is that your advisor would have to do all the editing to get the results into a publishable format, tying up loose ends of experiments, submitting the paper to the journal, etc. -- coordinating the myriad things needed to get a body of work into a publishable form. I don't think she's talking about taking the paper, removing your name from the author list and submitting it behind your back. -- You'd likely be informed about the submission, even if it's merely at the level of your advisor emailing "here's the finished manuscript, any comments before we submit?".
You're absolutely correct that you, as author and a person who did the work, have a say in whether something gets published. And you're also correct in saying that it's unethical for your advisor to ignore you and publish without your consent (either with or without your name on the author list). However, it'd also be unethical for you to hold up publication of an article for non-scientific reasons . As such, it's normally expected that any research work you do is probably going to get published, and if you don't do it yourself someone else is going to do the legwork to make it happen. (Giving you the appropriate credit .)
So it's all about how much "with[out] me" you're thinking about. Once the results are done, coauthors can have very, very little to do with the actual publishing of the paper. Other authors can do all the work of collating the results, analyzing them, writing, formatting, etc., and the coauthor can just look at the manuscript and say "Yup, looks okay to me!" and that counts as permission to publish.
 If you have scientific objections to the article as written, that's one thing - work with your advisor to address them. But if you think the article is scientifically sound but disagree with publication because you no longer care about it, or you want to hurt your advisor, etc., that's another.
 "Appropriate credit" drops off the less involvement you have in writing the paper. In certain circumstances, not actually writing the paper could knock you off of "first author" status. (For those fields which order authors by contribution amounts.)
If the requirements at your university are that you write three "publishable manuscripts" (I assume in a peer-reviewed journal or conference?), your advisor can clearly ask for exactly that. Now what makes a manuscript publishable? I see three reasonable ways to ascertain that:
- The manuscript is in a state such that it could be submitted for peer review (evaluated by someone who has a good judgement of that, maybe your supervisor).
- The manuscript has received favorable reviews from peer review.
- The manuscript has been accepted for publication after peer review and revision.
Asking you to have your manuscript submitted instead of just ready for submission does not seem over-stretching the first option, and it is actually already the weakest of the three. In fact it seems understandable that your supervisor would not see a real difference between a manuscript that is ready for submission and one that has actually been submitted. This is a few hours of work apart at most.
It doesn't sound like your advisor would want to finish the papers without you, but if she'd be willing to, I wouldn't see any problems with that. Of course she has to retain your name on the paper, but certainly only one of the authors can handle all the correspondence with the journal in the name of the other authors. Also any revisions in principle should be approved by you, though as long as you find a working modus that is ok for everybody involved it could also be handled more flexibly.
A dissertation is the student's work. The outsourcing of committee work to journal referees provides a convenient way of passing the buck; but is it good for academia? Advisers guide and lead the student. Does one ask for part of the salary of an undergraduate alumnus who benefited from a faculty members guidance and leadership?
The natural course of things, barring human difficulties, is that the student-adviser relationship is a close one and after graduation, they will work together. What we see with the modern trend toward making coauthored publication a requirement of dissertations is a lot of students who do not succeed once they are no longer just implementing the faculty member's next notch in the belt. We should be developing the student's scholarly ability.
I'm a professor. It would be considered unethical even for a thesis advisor to publish the work of a graduate student without the student's implicit or explicit permission. Usually, quite a bit of the advisor's creativity and work goes into advising a student, and that part is ultimately owned by the advisor. If I tell a student of mine to take this and that data set and do this or that to it, if then the student does so, but then refuses to work with me in publishing the results, then I am justified in doing the work myself or let someone else do it, and then publish the results. It becomes involved because when two or more people work together, attribution of ideas and insights is tremendously difficult, but academic custom dictates that all intellectual contributions (unless they are completely trivial) are contributed, usually through authorship. Therefore, in this scenario, I might have to someone acknowledge the possible contribution of the previous student. But this scenario is very far-fetched.
Profs at a research university get evaluated including on the number and quality of published articles with students.
You should talk with your advisor about your plans. It is in both of your interests to get published, independent on whether you are staying at this university. It would be wrong to go to another place and then publish your results without acknowledging your professor's contributions and it would hurt her / his career.
Since both advisor and advisee benefit from publishing, there are usually few problems in this area. In case a student is leaving for the non-academic world, it is normal but not normative for an advisor to see publications through the process.
You have to give your advisor three publishable manuscripts anyway in order to graduate if that is what the department demands. There seems to be at least an implicit expectations that publishable manuscripts are indeed eventually published, so if you object to publishing, the Ph.D. committe is (probably) justified in not promoting you to doctor. (If you have a scientific disagreement with the article or its contents, then you open up a new can of worms. Usually, the department will then assign a committee to sort it out.)