This is problem 2.1 of Falko Algebra I Fields and Galois Theory. I have finished the proof but it looks very unsatisfying.
E/K is a field extension. L1,L2 are intermediate fields. L1L2 denotes the composite field formed by L1 and L2. Suppose [L1:K],[L2:K]<∞. Prove if [L1L2:K]=[L1:K][L2:K], then L1∩L2=K.
I can show by presenting the basis of L1L2/K. If K⊂L1∩L2 is proper, this means L1∩L2 has non-trivial overlap(i.e. I can start removing redundant basis elements from either L1 or L2). This will show [L1L2:K]<[L1:K][L2:K].
Questions:
This proof lacks elegance though simple. Is there a proof based on purely showing algebraic manipulation to complete the proof(i.e. showing either [L1∩L2:K]≤1, [L1L2:L1∩L2]=[L1:K][L2:K] or other equivalent relations)?
Why should I expect this is the case without looking at the basis?(i.e. Assume I do not know the proof but I want to know the reason why this algebraic relation leads to this special vector space construction.)
If [L1:K],[L2:K] are relatively prime, this becomes trivial. In my view point L1L2=L1⊗KL2, relatively primeness implies for simple extensions L1,L2 over K, the K−algebra can be generated by the simple tensor of the two generators. How do I generalize this statement for arbitrary case by assuming finite extension say L1=K(a1,…,am),L2=K(b1,…bn) where I assume m,n are minimal generator and [L1:K]=m1,[L2:K]=m2 which may not have any straightforward relationship with m and n respectively?(i.e. I want a straightforward proof showing that for [L1:K],[L2:K] relatively prime, I can present an explicit construction of basis of L1⊗KL2
This statement does not say [L1:K],[L2:K] must be relatively prime here.
Answer
This indeed is a bit laboured. Suppose L1∩L2≠K.
Then |L1∩L2:K|=d>1. Let ni=|Li:K|. Then |Li:L1∩L2|=ni/d. Also |L1L2:L1|≤|L2:L1∩L2|=n2/d, so
|L1L2:L1∩L2|=|L1L2:L1||L1:L1∩L2|≤n1n2/d2
and then
$$|L_1L_2:K|=|L_1L_2:L_1\cap L_2||L_1\cap L_2:K|\le n_1n_2/d
Here I used the principle |LM:L|≤|M:K| where L and M are
extensions of K and LM is a compositum of L and M over K.
This boils down to noting that if M=K(α) then LM=L(α).
No comments:
Post a Comment