Bourbaki

Lemma 2

Tag 06OR, permanent. Cite it as [Bourbaki, Tag 06OR].

Algebra, chapter V, § 11. Abelian extensions

Let f, g and h be monic polynomials in Q[X]\mathbf{Q}[X] such that f=ghf = gh. If f has integer coefficients, the same is true of g and h.

Let aa (resp. bb) be the least of the integers α≥3\alpha \geq 3 (resp. β≥1\beta \geq 1) such that αg\alpha g (resp. βh\beta h) has integer coefficients; we put g′=agg' = ag and h′=bhh' = bh and show by reductio ad absurdum that a=b=1a = b = 1. If this were not so, there would be a prime divisor pp of abab. If u∈Z[X]u \in \mathbf{Z}[X], let us write uˉ\bar{u} for the polynomial with coefficients in the field Fp\mathbf{F}_p obtained by reduction mod pp of the coefficients of uu. We have g′h′=abfg'h' = abf, whence gˉ′hˉ′=0\bar{g}'\bar{h}' = 0; since the ring Fp[X]\mathbf{F}_p[X] is an integral domain (IV, p. 9, Prop. 8) we thus have gˉ′=0\bar{g}' = 0 or hˉ′=0\bar{h}' = 0. In other words, pp divides all the coefficients of g′g' or all those if h′h' and this contradicts the hypothesis.

We have the relation

Xn−1=∏d∣nΦd(X). X^n - 1 = \prod_{d|n} \Phi_d(X) .

For we have Xn−1=∏ζ∈μn(Q)(X−ζ)X^n - 1 = \prod_{\zeta \in \mu_n(\mathbf{Q})} (X - \zeta) and the sets SdS_d for dd dividing nn form a partition of μn(Q)\mu_n(\mathbf{Q}).

The formula (6) determines Φn(X)\Phi_n(X) when we know Φd(X)\Phi_d(X) for all the divisors d<nd < n of nn; since Φ1(X)=X−1\Phi_1(X) = X - 1, we thus have a recursive procedure for calculating Φn\Phi_n. For example for prime pp we have

Xp−1=(X−1)Φp(X), X^p - 1 = (X - 1)\Phi_p(X) ,

whence

Φp(X)=Xp−1+Xp−2+⋯+X+1, \Phi_p(X) = X^{p-1} + X^{p-2} + \cdots + X + 1 ,

and

Φpr+1(X)=Φp(Xpr)forr≥0. \Phi_{p^r+1}(X) = \Phi_p(X^{p^r}) \quad \text{for} \quad r \geq 0 .

Let us list the values of the polynomials Φn(X)\Phi_n(X) for 1≤n≤121 \leq n \leq 12:

Φ1(X)=X−1Φ2(X)=X+1Φ3(X)=X2+X+1Φ4(X)=X2+1Φ5(X)=X4+X3+X2+X+1 \begin{align*} \Phi_1(X) &= X - 1 \\ \Phi_2(X) &= X + 1 \\ \Phi_3(X) &= X^2 + X + 1 \\ \Phi_4(X) &= X^2 + 1 \\ \Phi_5(X) &= X^4 + X^3 + X^2 + X + 1 \\ \end{align*}
Φ6(X)=X2−X+1 \Phi_6(X) = X^2 - X + 1
Φ7(X)=X6+X5+X4+X3+X2+X+1 \Phi_7(X) = X^6 + X^5 + X^4 + X^3 + X^2 + X + 1
Φ8(X)=X4+1 \Phi_8(X) = X^4 + 1
Φ9(X)=X6+X3+1 \Phi_9(X) = X^6 + X^3 + 1
Φ10(X)=X4−X3+X2−X+1 \Phi_{10}(X) = X^4 - X^3 + X^2 - X + 1
Φ11(X)=X10+X9+X8+X7+X6+X5+X4+X3+X2+X+1 \Phi_{11}(X) = X^{10} + X^9 + X^8 + X^7 + X^6 + X^5 + X^4 + X^3 + X^2 + X + 1
Φ12(X)=X4−X2+1 \Phi_{12}(X) = X^4 - X^2 + 1

The values of Φ1,Φ2,Φ3,Φ4,Φ5,Φ7,Φ8,Φ9\Phi_1, \Phi_2, \Phi_3, \Phi_4, \Phi_5, \Phi_7, \Phi_8, \Phi_9 and Φ11\Phi_{11} follow directly from (7); now we have Φ1Φ2Φ3Φ6=X6−1\Phi_1 \Phi_2 \Phi_3 \Phi_6 = X^6 - 1 and Φ1Φ2Φ3Φ4Φ6Φ12=X12−1\Phi_1 \Phi_2 \Phi_3 \Phi_4 \Phi_6 \Phi_{12} = X^{12} - 1, whence Φ4Φ12=X12−1X6−1−X6+1\Phi_4 \Phi_{12} = \frac{X^{12} - 1}{X^6 - 1} - X^6 + 1 and finally Φ12=X6+1X2+1=X4−X2+1\Phi_{12} = \frac{X^6 + 1}{X^2 + 1} = X^4 - X^2 + 1. The cases n=6n = 6 and n=10n = 10 may be treated similarly.

Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.

Cited by

Cites

Nothing.

Transcribed from the printed edition, 2003, Springer. Not translated.