Münster J. of Math.

14 (2021), 155–163 Münster Journal of Mathematics

DOI 10.17879/59019521586 c Münster J. of Math. 2021
urn:nbn:de:hbz:6-59019522020

The Weierstrass preparation theorem and

resultants of p-adic power series

Laurent Berger

(Communicated by Peter Schneider)

Abstract. We define the resultant of two power series with coefficients in the ring of inte-
gers of a p-adic field. In order to do this, we prove a universal version of the Weierstrass
preparation theorem.

Introduction
Given two polynomials P and Q with coefficients in a field K, the resultant
Res(P, Q) allows us to determine whether P and Q have a common root in K.
The resultant is a polynomial function of the coefficients of P and Q, and
Res(P, Q) = 0 if and only if P and Q have a common root.
In this article, we consider a similar question for p-adic power series. Let K
be a finite extension of Qp , or more generally a finite totally ramified extension
of W (k)[1/p], where k is a perfect field of characteristic p. Let OK denote the
integers of K, let mK be the maximal ideal of OK , let k be the residue field
of OK , and let π be a uniformizer of OK . Let Cp be the completion of an
algebraic closure K of K, so that mCp is the p-adic open unit disk. A power
series f (X) = f0 +f1 X +· · · ∈ OK [[X]] defines a bounded holomorphic function
on mCp , and may have roots in this domain. Given two such power series, we
would like to know whether they have a common root. The Weierstrass degree
×
wideg(f ) of f is the smallest integer n such that fn ∈ OK , or +∞ if there is no
such integer. If wideg(f ) = n is finite, then f has precisely n roots (counting
multiplicities) in mCp . Our main result is the following.
Theorem A. For all n > 1, there exists a power series
Resn ({Fi }i>0 , {Gi }i>0 ) ∈ Z[Fn , Fn−1 , {Fk }k>n+1 , {Gk }k>0 ][[F0 , . . . , Fn−1 ]]
such that for all power series f (X), g(X) ∈ OK [[X]], with wideg(f ) = n, we
have Y
g(z) = Resn ({fi }i>0 , {gi }i>0 ).
z∈mCp
f (z)=0
156 Laurent Berger

In particular, Resn ({fi }i>0 , {gi }i>0 ) = 0 if and only if f and g have a common
root in mCp .

Note that if wideg(f ) = n, then f0 , . . . , fn−1 ∈ mK so that the power series

Resn ({fi }i>0 , {gi }i>0 ) does converge. The main technical tool for proving
theorem A is the Weierstrass preparation theorem. We use a version due to
O’Malley (see [6]) which allows us to prove the following universal Weierstrass
preparation theorem. Recall that if R is a ring and I is an ideal of R, a
polynomial is said to be distinguished for I if it is monic and all its non-leading
coefficients are in I. If n > 1, let Rn = Z[Fn , Fn−1 , {Fk }k>n+1 ][[F0 , . . . , Fn−1 ]]
and let In be the ideal of Rn generated by F0 , . . . , Fn−1 .

Theorem B. We can write the power series F (X) = i>0 Fi X i ∈ Rn [[X]] as

P

F (X) = P (X)U (X), where U (X) ∈ Rn [[X]]× and P (X) = X n + Pn−1 X n−1 +
· · · + P0 ∈ Rn [X] is a distinguished polynomial for the ideal In .
In addition, P and U are uniquely determined by F .

Theorem B provides a universal Weierstrass preparation theorem, and the

existence part of the classical versions follows by specializing. In particular,
Theorem B shows how the coefficients of p and u depend on those of f when
we write a power series f (X) ∈ OK [[X]] as the product of a distinguished
polynomial p and a unit u.
In Section 3, we give an application of our results to the iteration of power
series in characteristic p. We show that such a power series admits a lift to
characteristic zero satisfying certain properties, which strengthens a construc-
tion of Lubin (see [5]).
We finish this article with a sketch of an analogue of our constructions that
singles out the roots of a power series in a circle, instead of in an open disk.

1. A universal Weierstrass preparation theorem

The classical Weierstrass preparation theorem over OK (see, for instance,
[3, Ch. VII, Section 3, no 8]) says that if f (X) ∈ OK [[X]] and wideg(f ) = n,
there exists a distinguished (for the ideal mK ) polynomial p(X) of degree n and
a unit u(X) ∈ OK [[X]]× such that f = pu. In addition, p and u are uniquely
determined by f . The coefficients of p and u depend on those of f . In order
to make this dependence more explicit, we use the following strengthening of
the Weierstrass preparation theorem, which is [6, Thm. 2.10].

Theorem 1.1. Let R be a ring, and take f (X) = f0 + f1 X + · · · ∈ R[[X]]. Sup-

pose that fn ∈ R× and that R is separated and complete for the (f0 , . . . , fn−1 )-
adic topology.
There exists a distinguished (for the ideal (f0 , . . . , fn−1 )) polynomial p(X)
of degree n and u(X) ∈ R[[X]]× such that f = pu.
In addition, p and u are uniquely determined by f .

Münster Journal of Mathematics Vol. 14 (2021), 155–163

 Weierstrass preparation and resultants 157

Although we do not need this in the remainder of this article, we point

out the following corollary of Theorem 1.1. Note that some even more gen-
eral versions of the Weierstrass preparation theorem can be found, see, for
instance, [4].

Corollary 1.2. Let R be a ring and let J be an ideal of R such that R is

separated and complete for the J-adic topology. Take f (X) = f0 + f1 X + · · · ∈
R[[X]]. Suppose that fn ∈ R× and that f0 , . . . , fn−1 ∈ J.
There exists a distinguished (for the ideal J) polynomial p(X) of degree n
and u(X) ∈ R[[X]]× such that f = pu.
In addition, p and u are uniquely determined by f .

Proof. This follows from Theorem 1.1, and the following assertion [8,Tag 00M9,
Lem. 10.95.8]: if I ⊂ J are two ideals of a ring R, with I finitely generated,
and if R is separated and complete for the J-adic topology, then R is separated
and complete for the I-adic topology. 

If n > 1 is fixed, we can consider the variables {Fi }i>0 and we define

Rn = Z[Fn , Fn−1 , {Fk }k>n+1 ][[F0 , . . . , Fn−1 ]].

The ring Rn is separated and complete for the (F0 , . . . , Fn−1 )-adic topology,
and the following result (Theorem B) is an immediate consequence of Theo-
rem 1.1.
i
P
Theorem 1.3. We can write F (X) = i>0 Fi X ∈ Rn [[X]] as F (X) =
P (X)U (X), where U (X) ∈ Rn [[X]]× and P (X) = X n + Pn−1 X n−1 + · · ·+ P0 ∈
Rn [X] is a distinguished polynomial for the ideal In .
In addition, P and U are uniquely determined by F .

Example 1.4. We give an explicit formula for P (X) in Theorem 1.3 when
n = 1. If n = 1, then P (X) = X + P0 and P0 ∈ R1 = Z[F1 , F1−1 , {Fk }k>2 ][[F0 ]]
is given by the following formula ([1, Prop. 2.2]; here π(j, n) denotes the set of
i1 , . . . , in ∈ Z>0 such that i1 + i2 + · · · + in = j and i1 + 2i2 + · · · + nin = n):
n 
X X 1 n+j+1 X (n + j)!
P0 = F0n+1 − F2i1 F3i2 · · · Fn+1
in

j=0
F1 (n + 1)!i 1 !i 2 ! · · · i n !
n>0 π(j,n)

F0 F2 3 2F 2 
F
=− − F02 · 3 + F03 · 4 − 52 + O(F04 ).
F1 F1 F1 F1
(We have (n + j)!/(n + 1)!i1 !i2 ! · · · in ! ∈ Z if i1 + i2 + · · · + in = j and i1 +
2i2 + · · · + nin = n; indeed, (n + j)!/(n + 1)!i1!i2 ! · · · in ! becomes a multinomial
coefficient and hence an integer if we replace either n + 1 by n or ik by ik − 1
for some k. If ℓ is a prime number, then it cannot divide both n + 1 and all of
the ik . Hence (n+ j)!/(n+ 1)!i1 !i2 ! · · · in ! is a rational number that is ℓ-integral
for every prime number ℓ, and is therefore an integer).

Münster Journal of Mathematics Vol. 14 (2021), 155–163

158 Laurent Berger

2. Resultants and discriminants of p-adic power series

By the theory of Newton polygons, a distinguished polynomial p(X) =
X n + pn−1 X n−1 + · · · + p0 ∈ OK [X] of degree n has precisely n roots in
mCp (counting multiplicities). Let p(X) be such a polynomial. If g(X) =
i
P Q
i>0 gi X ∈ OK [[X]], we can consider p(z)=0 g(z).

Proposition 2.1. There exists a power series

ResPoln (P0 , . . . , Pn−1 , {Gk }k>0 ) ∈ Z[{Gk }k>0 ][[P0 , . . . , Pn−1 ]]
such that for all g(X) = i>0 gi X i ∈ OK [[X]] and all distinguished polynomial
P

p(X) = X n + pn−1 X n−1 + · · · + p0 ∈ OK [X] of degree n, we have

Y
g(z) = ResPoln (p0 , . . . , pn−1 , {gk }k>0 ).
p(z)=0

Proof. Let Z1 , . . . , Zn denote n variables. For each n-uple d = (d1 , . . . , dn ) ∈

Z>0 with d1 6 · · · 6 dn , let Zd = (e1 ,...,en ) Z1e1 · · · Znen , where (e1 , . . . , en )
P

ranges over all distinct permutations of (d1 , . . . , dn ). There exists polynomials

Sd ∈ Z[{Gk }k>0 ] for each d, such that
n X
Y X
Gk Zik = Sd ({Gk }k>0 )Zd .
i=1 k>0 d

If we write ni=1 (X − Zi ) = X n + Pn−1 X n−1 + · · · + P0 , then by the funda-

Q
mental theorem of symmetric polynomials, each Zd belongs to Z[P0 , . . . , Pn−1 ].
We set
X
ResPoln = Sd ({Gk }k>0 )Zd ∈ Z[{Gk }k>0 ][[P0 , . . . , Pn−1 ]].
d

Note that the total degree of Zd is d1 + · · · + dn so that the degree of Zd as an

element of Z[P0 , . . . , Pn−1 ] is at least (d1 + · · · + dn )/n. Therefore, the above
sum converges for the (P0 , . . . , Pn−1 )-adic topology. The proposition follows
by specializing. 

We can now prove Theorem A.

Theorem 2.2. There exists a power series
Resn ({Fi }i>0 , {Gi }i>0 ) ∈ Z[Fn , Fn−1 , {Fk }k>n+1 , {Gk }k>0 ][[F0 , . . . , Fn−1 ]]
such that for all power series f (X), g(X) ∈ OK [[X]] with wideg(f ) = n, we
have Y
g(z) = Resn ({fi }i>0 , {gi }i>0 ).
z∈mCp
f (z)=0

Proof. By Theorem 1.3, we can write the power series F (X) = i>0 Fi X i as
P

F (X) = P (X)U (X) with P (X) = X n + Pn−1 X n−1 + · · · + P0 , where each Pi

Münster Journal of Mathematics Vol. 14 (2021), 155–163

 Weierstrass preparation and resultants 159

belongs to the ideal (F0 , . . . , Fn−1 ) of Z[Fn , Fn−1 , {Fk }k>n+1 ][[F0 , . . . , Fn−1 ]].
The result follows from Proposition 2.1, by setting
Resn = ResPoln (P0 , . . . , Pn−1 , {Gk }k>0 ). 

If f, g ∈ OK [[X]] and wideg(f ) = n, we write Resn (f, g) instead of the more

cumbersome Resn ({fi }i>0 , {gi }i>0 ).
Remark 2.3. We have Resn (f, gh) = Resn (f, g) Resn (f, h).
Definition 2.4. We define Discn to be the power series
Discn ({Fi }i>0 ) = Resn (F, F ′ ) ∈ Z[Fn , Fn−1 , {Fk }k>n+1 ][[F0 , . . . , Fn−1 ]],
and likewise write Discn (f ) instead of Discn ({fi }i>0 ).
By Theorem 2.2, a power series f (X) ∈ OK [[X]] with wideg(f ) = n has only
simple roots in mCp if and only if Discn (f ) 6= 0.
Proposition 2.5. The set of elements of OK [[X]] having only simple roots in
mCp is open in the p-adic topology.
Proof. Take f (X) ∈ OK [[X]] having only simple roots in mCp . We can divide f
by an appropriate power of π and assume that wideg(f ) is finite. Let n =
wideg(f ) and v = valπ Discn (f ).
The fact that Discn belongs to Z[Fn , Fn−1 , {Fk }k>n+1 ][[F0 , . . . , Fn−1 ]] implies
that for every h(X) ∈ OK [[X]], we have valπ Discn (f + π v+1 h) = v, so that if
g(X) ∈ OK [[X]] is such that valπ (f − g) > v + 1, then g(X) has only simple
roots in mCp . 

Note that the set of elements of OK [[X]] having only simple roots in mCp
is also dense in the p-adic topology. If f = pu, with p distinguished hav-
ing multiple roots, then p can be approached by distinguished polynomi-
als having only simple roots. Indeed, the (usual) discriminant of p(X) =
X n + pn−1 X n−1 + · · · + p0 is a polynomial ∆(p0 , . . . , pn−1 ) and its zero set is
closed with empty interior.

3. Lubin’s proof of Sen’s theorem on iteration of power series

In this section, we give an application of the above constructions. In his
paper [5], Lubin gives a short and very nice proof of Sen’s theorem on iteration
of power series. We start by recalling Sen’s theorem and P Lubin’s argument.
Recall that k is the residue field of OK . If w(X) = X + i>2 wi X i ∈ k[[X]], let
i(w) = m − 1, where m is the smallest integer > 2 such that wm 6= 0 (or +∞
n
if there is no such integer). For n > 0, let in (w) = i(w◦p ). Sen’s theorem [7,
Thm. 1] says that in−1 (w) ≡ in (w) mod pn for all n > 1 (where the congruence
holds automatically if one side is +∞).
Lubin’s argument is to show that for each n > 0 such that in (w) 6= +∞,
there exists a finite extension L of K and a power series fn (X) ∈ X · OL [[X]]

Münster Journal of Mathematics Vol. 14 (2021), 155–163

160 Laurent Berger

such that the image of fn (X) in kL [[X]] is w(X) and such that all the roots of
n
fn◦p (X) − X in mCp are simple. We then have
 ◦pn 
f (X) − X
in (w) − in−1 (w) = wideg ◦pnn−1 ,
fn (X) − X
so that in (w) − in−1 (w) is the number of points of mCp whose orbit under fn
is of cardinality pn . This number is clearly divisible by pn , which implies Sen’s
theorem.
Using our methods, we can improve Lubin’s result. We prove that there is
one lift f of w that works for all n, and has coefficients in OK .
Theorem 3.1. Take w(X) = X + i>2 wi X i ∈ k[[X]] and let N ⊂ Z>0 be
P

the set of n such that in (w) is finite. There exists f (X) ∈ X · OK [[X]] whose
n
image in k[[X]] is w(X) and such that for all n ∈ N , the roots of f ◦p (X) − X
in mCp are simple.
Proof. Let W be the set of f (X) ∈ X · OK [[X]] whose image in k[[X]] is w(X),
n
and let Wn be the set of elements of W such that the roots of f ◦p (X) − X in
mCp are simple. We prove that if n ∈ N , then Wn is open and dense in W for
the p-adic topology. Since W is a complete metric space, the T theorem follows
from this assertion and Baire’s theorem, which implies that n∈N Wn is dense
in W and hence nonempty.
Fix an element
P w̃ ∈ W . We have W = {w̃ + h, hP∈ πX · OK [[X]]}.
n
If F (X) = j>1 Fj X j , write F ◦p (X) − X = j>1 Fj(n) X j . Take n ∈ N
and let i = in (w) + 1. Let F (X) = j>1 (w̃j + Hj )X j , where {Hj }j>1 are
P

variables. For all j > 1, Fj(n) ∈ OK [H1 , . . . , Hj ]. Since Fi(n) (0) = w̃i(n) ∈ OK ×
,
(n) (n) −1 (n)
Fi has an inverse (Fi ) ∈ OK [[H1 , . . . , Hi ]]. If j 6 i − 1, then w̃j ∈ mK ,
and so Fj(n) is in the ideal (π, H1 , . . . , Hj ) of OK [H1 , . . . , Hj ]. The power series
n (n) (n) −1 (n) (n) (n)
Disci (F ◦p (X) − X) ∈ Z[Fi , (Fi ) , {Fj }j>i+1 ][[F1 , . . . , Fi−1 ]]
therefore gives rise to an element Dn ({Hj }j>1 ) ∈ OK [{Hj }j>i+1 ][[H1 , . . . , Hi ]].
Let us first show that Wn is open in W . If f = w̃ + h ∈ Wn , with h ∈
πX · OK [[X]], then Dn (h) 6= 0 by definition. If v = valπ (Dn (h)) and g(X) is
in X · OK [[X]], then Dn (h + π v+1 g) ≡ Dn (h) mod π v+1 , so that valπ (Dn (h +
π v+1 g)) = v. Hence f + h′ ∈ Wn for all h′ ∈ πX · OK [[X]] such that valπ (h −
h′ ) > v + 1, and therefore Wn is open in W .
We now show that Wn is dense in W . If this is not the case, its complement
has nonempty interior. Suppose therefore that there exists f = w̃ + h ∈ W
and v > 1 such that Dn (h + π v g) = 0 for all g ∈ X · OK [[X]]. We can write
X
Dn ({Hj }j>1 ) = Pd ({Hj }j>i+1 )H1d1 · · · Hidi ,
d∈Zi>0

where d = (d1 , . . . , di ) and the Pd are polynomials with coefficients in OK .

The fact that Dn (h + π v g) = 0 for all g ∈ X · OK [[X]] implies that for all
fixed values of {gj }j>i+1 , the corresponding power series in H1 , . . . , Hi is zero

Münster Journal of Mathematics Vol. 14 (2021), 155–163

 Weierstrass preparation and resultants 161

on the set (h1 , . . . , hi ) + π v OK

i
. It is therefore the zero power series. This in
turn implies that for each d, the polynomial Pd ({Hj }j>i+1 ) is zero on the set
{(hj + π v OK )}j>i+1 , and therefore Pd = 0.
This implies that Dn is the zero power series, and therefore that for any
n
extension L/K and any f (X) ∈ X · OL [[X]] such that wideg(f ◦p (X) − X) = i,
n
the roots of f ◦p (X) − X in mCp are not simple. This contradicts Lubin’s
result in [5] (the aforementioned construction of the power series fn ). 

4. A universal Hensel factorization theorem

In this section, we sketch an analogue of our constructions that singles
out the roots of a power series in a circle {z ∈ Cp , |z| = r} instead of in
an open disk as in Section 1 and Section 2. Let P OK {X} denote the ring
of restricted power series (power series f (X) = n>0 fn X n with fn ∈ OK
and fn → 0 as n → +∞). An element of OK {X} converges on the closed
unit disk {z ∈ Cp , |z| 6 1}. We are interested in the roots of f in the unit
circle {z ∈ Cp , |z| = 1}. Take f (X) = n>0 fn X n ∈ OK {X}, one of whose
P
× ×
coefficients is in OK . Let µmin (f ) = min{i > 0, fi ∈ OK } and let µmax (f ) =
×
max{i > 0, fi ∈ OK }. If n = µmin (f ) and n + d = µmax (f ), we have the
factorization f = p · u in k[X], with
−1
p(X) = f n+d · (f n + f n+1 X + · · · + f n+d X d ) and u(X) = f n+d · X n .
Hensel’s factorization theorem [3, Ch. III, Section 4, no 3] implies that there
exist p(X) ∈ OK [X] and u(X) ∈ OK {X} such that f = pu, the polynomial p
×
is monic of degree d, p(0) ∈ OK , and µmax (u) = µmin (u). This analogue of the
Weierstrass preparation theorem, along with the theory of Newton polygons,
implies that f has precisely µmax (f ) − µmin (f ) roots (counting multiplicities)
in the unit circle.
Let {Fi }i>0 be variables, take n, d > 0, and let
−1
Sn,d = Z[{Fn+j }06j6d , Fn−1 , Fn+d ][[F0 , . . . , Fn−1 , {Fn+d+k }k>1 ]].
Our definition of aPpower series ring in infinitely many variables is the “large”
one (for instance, k>0 Fk belongs to Sn,d ), see [2, Ch. IV, Section 4]. Let In,d
be the ideal of Sn,d generated by F0 , . . . , Fn−1 , {Fn+d+k }k>1 . The following
result is a universal Hensel factorization theorem.
Theorem 4.1. We can write the power series F (X) = i>0 Fi X i ∈ Sn,d [[X]]
P

as F (X) = P (X)U (X), where P (X) ∈ Sn,d [X] is monic of degree d, P (0) ∈
×
Sn,d , and U (X) ≡ Fn+d X n mod In,d .
In addition, P and U are uniquely determined by F .
Proof. The ring Sn,d is separated and complete for the In,d -adic topology. The
−1
polynomials P (X) = Fn+d ·(Fn +Fn+1 X+· · ·+Fn+d X d ) and U (X) = Fn+d ·X n
×
generate the unit ideal in Sn,d /In,d [X], since Fn , Fn+d ∈ Sn,d . Indeed, a
m
descending induction on n−1 > m > 0 shows that X ∈ (P , U ) by considering
X mP .

Münster Journal of Mathematics Vol. 14 (2021), 155–163

162 Laurent Berger

The theorem therefore results from Hensel’s factorization theorem (see [3,
Ch. III, Section 4, no 3], and the discussion at the beginning of no 5 of ibid). 
We now give an application to the slope factorization of polynomials. Take
a nonzero polynomial Q P (X) ∈ K[X] and let c be its leading coefficient. We
can write P (X) = c · r Pr (X), where for each r, the polynomial Pr (X) is
monic and all of its roots are of valuation r ∈ Q. ByQGalois theory, each
Pr (X) belongs to K[X]. The decomposition P (X) = c · r Pr (X) is the slope
factorization of P (X), and Pr (X) is the slope r factor of P (X).
×
Corollary 4.2. Given F (X) ∈ OK [X], one of whose coefficients is in OK ,
there are universal formulas, depending only on µmin (F ), µmax (F ) and deg(F ),
for the coefficients of the slope 0 factor of F in its slope factorization, in terms
of the coefficients of F .
Proof. Let F = F0 F6=0 be the factorization of F as the product of a monic
polynomial of slope 0 and of a polynomial of slopes 6= 0. The polynomial F6=0
has no roots in the unit circle, so that if we view F as an element of OK {X},
then P = F0 and U = F6=0 . 
Theorem 4.1 can also be used, as in Section 2, to produce resultant power
series Resn,d , that will detect whether two restricted power series f and g, with
µmin (f ) = n and µmax (f ) = n + d, have roots in common in the unit circle.
We end this article with the following question.
Question 4.3. The classical resultant
Q of two polynomials P and Q can be de-
fined using either the product P (z)=0 Q(z) or the determinant of the Sylvester
matrix. Both approaches give the same formula, after a suitable normaliza-
tion. In this article, we follow the first approach. Is it possible to view our
resultants as the (generalized) determinants of some operators on some p-adic
Banach spaces?
Acknowledgements. I thank Sandra Rozensztajn for her help when I was
writing the paper, and the referee for several useful remarks.

References
[1] D. Birmajer, J. B. Gil, and M. D. Weiner, On Hensel’s roots and a factorization formula
in Z[[X]], Integers 14 (2014), Paper No. A47, 18 pp. MR3259300
[2] N. Bourbaki, Éléments de mathématique. Algèbre., Springer-Verlag, Berlin, 2006.
[3] N. Bourbaki, Éléments de mathématique. Algèbre commutative., Springer-Verlag, Berlin,
2006.
[4] J. Elliott, Factoring formal power series over principal ideal domains, Trans. Amer. Math.
Soc. 366 (2014), no. 8, 3997–4019. MR3206450
[5] J. Lubin, Sen’s theorem on iteration of power series, Proc. Amer. Math. Soc. 123 (1995),
no. 1, 63–66. MR1215030
[6] M. O’Malley, On the Weierstrass preparation theorem, Rocky Mountain J. Math. 2
(1972), no. 2, 265–273. MR0289500
[7] S. Sen, On automorphisms of local fields, Ann. of Math. (2) 90 (1969), 33–46. MR0244214
[8] The stacks project authors, Stacks project, 2018.

Münster Journal of Mathematics Vol. 14 (2021), 155–163

 Weierstrass preparation and resultants 163

Received January 27, 2020; accepted March 24, 2020

Laurent Berger
UMPA de l’ENS de Lyon
UMR 5669 du CNRS
Lyon, France
E-mail: [email protected]
URL: perso.ens-lyon.fr/laurent.berger/

Münster Journal of Mathematics Vol. 14 (2021), 155–163