A frequent complaint about scientists is that they are generally bad writers. I was wondering why that is so. One reason is that we simply tolerate bad writing. We let people get away with it. You can imagine that if people's papers were rejected from journals due to poor writing, they would improve their writing in a hurry. The scientific establishment doesn't value good writing enough. Another reason (I speculate) is that scientists don't expose themselves to good writing. They don't read literary works or essays that are the quality of The Atlantic . So they have no idea what good writing is like. The poor quality of scientific writing certainly doesn't help. I've attended some wonderful writing seminars and they seem to attract people who are already decent writers. They know what good writing is like. But where are the people who need help the most?
I think the best way to address this problem is to require students to spend time learning how to write, from a specialized writing teacher. Most advisors are probably bad writers themselves or they have no idea how to teach their students. Advisors have too much work to do; they don't have time to teach people how to write. Frankly, teaching writing is time consuming and tedious. My experience with professors is that their "writing instruction" consists of telling students, "I don't like X. Change it to Y," with no explanation. This is not teaching. Some students who are bright or already good writers will figure out what the professor really meant, but most people need an explanation of "why." Otherwise, they will simply make the same mistake in the future.
 The Atlantic is just an example. I'm not implying it's the best magazine or my favorite.