The other day I wrote about 5000 words in the day. That's maybe fifteen pages, double spaced. In one day.
At that rate, someone could feasibly write a 100-page dissertation in six days.
Of course,when I write that much in a day, I also throw away a large part of what I write. This is ok if I can keep my eye on how things add up. If I write 5000 words in a day, and then throw away 80% of what I write, I finish my 100=page project in 30 days, not 6. And if I throw away 95% of what I write, I've finished my dissertation in 120 days--one semester.
All it takes is to write 5000 words a day, and then find that 250 of them are useful.
If you were to keep every word you write, then you would only need to write 250 words a day to write a dissertation in a semester. But that would be far harder, and would probably lead to a worse dissertation, than trying to write 1000 words a day with the conscious acceptance that you were going to only use 250 of them in the long run.
As a writer, what is harder: to write one coherent, well-formed page or to write four, rambling, poorly written pages? My personal opinion and experience say that I can probably write ten times as fast when I don't care about the quality of the outcome as when I do. When I worry about the outcome, each word is a struggle. When I write, on the other hand, just to put my ideas on the page to see how they work out and to discover if there are any weaknesses, then I write easily. I may hit many dead ends; I may struggle with obstacles, but in the end I produce more.
In her book Bird by Bird, Anne Lamott titles a chapter "shitty first drafts," and says "all good writers write them."
Write and then write more. Don't worry about making it good; worry about making it express your ideas. Worry about it saying all the different things that you want to say. In whatever order it comes out. Just write. And write more. And try to learn as you go.
If you write regularly, the writing adds up. If you're too intent on getting it right the first time and therefore you don't write, you're not doing yourself a favor. Be willing to get it wrong first, and then fix it later using what you have learned.