Iteration doesn't only occur in the writing. There are many aspects of a large project that you can return to as you work through it.
Any project that involves significant learning will lead to new perspectives that force revisiting old ideas and old decisions. It is not unlikely that you will have refine/redefine your terms, redefine your method, redefine your conclusions, and so on. Because you are learning, you are subject to the ideas that you come to value. We would need much less iteration if we never learned anything at all--we could simply do things exactly the way we did them previously. But that probably won't work for writing a dissertation.