9

At the end of my abstract algebra class this spring, we were given an overview of differential algebra and some differential Galois theory. We went too fast to prove anything nontrivial, but I found Liouville's theorem on elementary antiderivatives really interesting and have been trying to get a handle on the proof recently. The theorem is as follows:

Theorem (Liouville) Let $\mathbb{F}$ be a differential field. For $a\in \mathbb{F}$, there exists an elementary differential extension $\mathbb{E}$ containing an antiderivative of $a$ if and only if there exist $u_1,...,u_n,v\in\mathbb{F}$, and constants $c_1,...,c_n\in\mathbb{F}$ such that

$$ a = v' + \sum_{i=1}^nc_i\frac{u_i'}{u_i}.$$

The only proof I've found online is that due to Maxwell Rosenlicht (see theorem 4.3 here). Its logic is reasonably easy to follow but it's long enough that I can't piece together an intuition from it. Does anyone have / know of an intuitive account for why Liouville's differential algebra theorem is true?

JFox
  • 340
  • 2
    It would be preferable if you gave links to the statement of the theorem, as well as the proof online that you found. Asking readers to do detective work about basic context that should be included in the question is a good way to lose their attention. Unless you're keen on shooting yourself in the foot this way, consider adding links. To include a link just [link text](url) Good luck – rschwieb Jun 28 '19 at 13:00

0 Answers0