Splitting it every which Way

Algebra Level 5

Consider a cubic polynomial f ( x ) f(x) with integer coefficients that has three distinct complex roots a , b , c a,b,c . What are the possible degrees of Q ( a , b , c ) \mathbb{Q}(a,b,c) over Q \mathbb{Q} ? Enter the sum of those possible degrees.

Clarification :

Q ( a , b , c ) \mathbb{Q}(a,b,c) consists of all complex numbers of the form g ( a , b , c ) g(a,b,c) where g ( x , y , z ) g(x,y,z) is a polynomial with rational coefficients. The degree of Q ( a , b , c ) \mathbb{Q}(a,b,c) is its dimension as a vector space over Q \mathbb{Q} .

Relevant Wiki : Algebraic number theory .


A complete solution will contain examples of all possible cases.


The answer is 12.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

1 solution

1.- Let the polynomial (x - 1)(x - 2)(x - 3) then Q ( 1 , 2 , 3 ) = Q [ Q : Q ] = 1 \mathbb{Q}(1,2,3) = \mathbb{Q} \Rightarrow [\mathbb{Q}: \mathbb{Q}] = 1 . Q ( a , b , c ) \mathbb{Q}(a,b,c) is the smallest field containing Q \mathbb{Q} and a, b and c. Notice also that Q ( a , b , c ) = Q ( a , b ) \mathbb{Q}(a,b,c) = \mathbb{Q}(a,b) due to Vieta's formula a + b + c Z a + b + c \in \mathbb{Z}

2.- Now, let the polynomial x 3 + x x^3 + x then Q ( i , i , 0 ) = Q ( i ) [ Q ( i ) : Q ] = = 2 \mathbb{Q}(i,-i,0) = \mathbb{Q}(i) \Rightarrow [\mathbb{Q}(i) : \mathbb{Q}] = = 2 .

3.-Let the polynomial p ( x ) = x 3 3 x + 1 p(x) = x^3 - 3x + 1 then if θ \theta is a root of p ( x ) p(x) , θ 2 2 \theta^2 - 2 is also a root of p ( x ) p(x) and Q ( θ , θ 2 2 ) = Q ( θ ) [ Q ( θ ) : Q ] = 3 \mathbb{Q}(\theta,\theta^2 -2) = \mathbb{Q}(\theta) \Rightarrow [\mathbb{Q}(\theta) : \mathbb{Q}] = 3 . Let's see that p ( x ) p(x) is irreducible over Q \mathbb{Q} . If some root θ \theta of p ( x ) p(x) belonged to Q \mathbb{Q} then θ = m n \theta = \frac{m}{n} with m,n coprime integers and θ \theta distinct to 1 and -1, then m 3 n 3 = 1 + 3 m n m 3 = n 2 ( 3 m n ) \frac{m^3}{n^3} = -1 + 3\frac{m}{n} \Rightarrow m^3 = n^2 \cdot (3m - n) but this a contradiction with the Fundamental Arithmetic Theorem.

4.-And finally, the polynomial x 3 2 x^3 - 2 fullfils [ Q ( 2 3 , 2 3 e 2 i π 3 ) : Q ] = [\mathbb{Q}(\sqrt[3]{2}, \sqrt[3]{2} \cdot e^{\frac{2i \cdot \pi}{3}}) : \mathbb{Q}] = = [ Q ( 2 3 , 2 3 e 2 i π 3 ) : Q ( 2 3 ) ] [ Q ( 2 3 ) : Q ] = 2 × 3 = 6 = [\mathbb{Q}(\sqrt[3]{2}, \sqrt[3]{2} \cdot e^{\frac{2i \cdot \pi}{3}}) : \mathbb{Q}(\sqrt[3]{2})][\mathbb{Q}(\sqrt[3]{2}) : \mathbb{Q}] = 2 \times 3 = 6 .

Note.- there aren't more cases. Why?

Thank you, this looks wonderful! (+1)

Clearly, you cover all the possible cases: If f ( x ) f(x) is reducible, then the degree must be 1 or 2 (your cases 1 and 2) and if f ( x ) f(x) is irreducible the degree will be 3 or 6.

In case 3 you can simply use the "rational root test" to see that x 3 3 x + 1 x^3-3x+1 is irreducible.

Now that you are "on a roll": Here is another interesting problem that is still waiting to be solved.

Otto Bretscher - 5 years, 3 months ago

Log in to reply

hmm, interesting, thank you, althought I was wrong in part 1 ,haha. Ok, I'm going to check my book about Euclidean Geometry and Linear Algebra. I hope to do it in a week. I'm seeing that you are posting a lot problems about this...

Guillermo Templado - 5 years, 3 months ago

Log in to reply

What was wrong with part 1? What did I miss?

Otto Bretscher - 5 years, 3 months ago

Log in to reply

@Otto Bretscher Nothing. It's me. I'm a stupid. Can you believe that I failed your last problem knowing that A T A = I 3 A^{T} \cdot A = I_3 and dim ker(A + I) = 2 and dim ker(A -I) = 1 \text{ dim ker(A + I) = 2 and dim ker(A -I) = 1} and eigenvectors with different eigenvalues are orthogonal if A is orthogonal?

Guillermo Templado - 5 years, 3 months ago

Log in to reply

@Guillermo Templado Don't feel too bad... that problem has 22 attempts and 3 solvers thus far ;)

Otto Bretscher - 5 years, 3 months ago

Log in to reply

@Otto Bretscher I appreciate the link to the algebraic number theory wiki, but I tried to keep everything as down-to-earth as possible, so I didn't talk about e.g. the fact that field extension degrees multiply in towers, or Galois theory, etc. I guess it's a good way to make someone read the whole thing in a fruitless attempt to find something that will solve their problem for them!

Patrick Corn - 5 years, 2 months ago

Log in to reply

@Patrick Corn I think the Wiki is still very helpful. If people know any linear algebra, they should be able to figure out that the degrees multiply, based on what they learn in the Wiki. The problem really does not require Galois theory.

Otto Bretscher - 5 years, 2 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...