It's more theoretical. While it's possible to write a dissertation in a short amount of time (I wrote mine in 2 months), it really should represent multiple years worth of dedicated research. For example, I had a figure in my dissertation which took 4 months to create, in terms of data collection and processing. Just one single figure out of several dozen!
That's one aspect is that most dissertations are not very impactful. The general idea is that a good dissertation should extend the field of knowledge in just a small way. It doesn't have to represent a titanic shift in thought or be revolutionary in any way. So really most dissertations are not supposed to have much thunder at all to steal.
At that level of specificity, it's possible to know all the big players by name and what they're working on. It's easy to find your own corner in such a small group, and it's very rare for a dark horse researcher to enter the field and suddenly steal your topic. It's hard to become an expert in a field without being noticed by already experts.
