19

5

So there are easy examples for algebraic closures that have index two and infinite index: $\mathbb{C}$ over $\mathbb{R}$ and the algebraic numbers over $\mathbb{Q}$. What about the other indices?

EDIT: Of course $\overline{\mathbb{Q}} \neq \mathbb{C}$. I don't know what I was thinking.

flag
3 
I bet you were not expecting the answer to be a theorem. It's one of the coolest little theorems in all of Galois theory. – Harry Gindi Dec 13 2009 at 14:58
4 
Incidentally, C is not an algebraic closure of Q, since it contains transcendental elements like e (or more generally, because it has uncountable cardinality). – S. Carnahan Dec 13 2009 at 16:16

2 Answers

34

Theorem (Artin-Schreier, 1927): Let K be an algebraically closed field and F a proper subfield of K with $[K:F] < \infty$. Then F is real-closed and $K = F(\sqrt{-1})$.

See e.g. Jacobson, Basic Algebra II, Theorem 11.14.

link|flag
2 
Sure, you put LaTeX in your answer, but I answered first! =) – Harry Gindi Dec 13 2009 at 14:53
4 
@Harry: You beat him by three and a half minutes only. That is about the time it would take Pete to type in his answer plus look up the exact reference in Jacobson's book. A clear case of independent responses if I ever saw one. – Harald Hanche-Olsen Dec 13 2009 at 16:24
1 
I know that! I just typed it out as fast as possible because I knew the answer and so did everyone else. I wasn't claiming he copied it, I was claiming, "FIRST POST!!!" – Harry Gindi Dec 13 2009 at 16:41
24 
Original reference, for the record: Emil Artin und Otto Schreier: Eine Kennzeichnung der reell abgeschlossenen Körper, Abhandlungen aus dem Mathematischen Seminar der Universität Hamburg 5 #1 (1927) 225–231 (doi:10.1007/BF02952522). Quoth Zentralblatt: β€œIn Ergänzung der Untersuchungen über die algebraische Konstruktion reeller Körper der Verf. (ibidem 5 (1926), 85-99; F. d.M. 52) wird bewiesen, dass die reell abgeschlossenen Körper identisch sind mit den Körpern, die durch endliche Erweiterung algebraisch abgeschlossen werden können, ohne selbst algebraisch abgeschlossen zu sein.’ – Harald Hanche-Olsen Dec 13 2009 at 18:04
39 
Am I the only person who finds this kind of competition to be an unhealthy expenditure of effort? – Boris Bukh Jan 7 2010 at 15:50
show 2 more comments
21

The Artin-Schreier theorem says that every algebraic closure of finite index has index 2, and it's the algebraic closure of a real-closed field.

Page 299 of Algebra by Serge Lang. Google Books Link to the page.

link|flag

Your Answer

Get an OpenID
or

Not the answer you're looking for? Browse other questions tagged or ask your own question.