We prove joint universality of the Riemann zeta-function for shifts ζ(s+idjτ) with arbitrary nonzero real parameters dj whose absolute values are pairwise distinct, and show that these conditions are optimal. As a consequence, we establish generalized strong recurrence for every finite family of nonzero real parameters, meaning that the corresponding shifts of the Riemann zeta-function simultaneously approximate each other. The only cases for which generalized strong recurrence remains unproved are those involving both zero and nonzero shift parameters, for which it is equivalent to the Riemann hypothesis. Our proof combines Emsalem's generalization of the six exponentials theorem with a generalization of Pecherskii's rearrangement theorem that may be of independent interest.
Introduction
In 1981 Bagchi [1] discovered an interesting connection between the Riemann hypothesis and Voronin’s universality theorem [13] for the Riemann zeta-function ζ(s). Namely, he showed that the Riemann hypothesis is true if and only if, for every compact set K⊂D:={s∈C:21<Res<1} with connected complement and every ε>0, we have
for every such compact set K and every ε>0. This property is usually called generalized strong recurrence or self-approximation. A natural approach to proving it is to show that the shifts ζ(s+idjτ) simultaneously approximate arbitrarily given non-vanishing analytic functions. Nakamura proved such a joint universality phenomenon when d1,…,dn are algebraic real numbers linearly independent over Q.
For n=2, the problem of joint universality for ζ(s+id1τ),ζ(s+id2τ) was completely solved in [10, 11]. In [10], the case of irrational d2/d1 was treated using the six exponentials theorem, which shows that possible linear relations among the numbers djlogp, j=1,2, p∈P, involve at most finitely many exceptional primes. The remaining case of nonzero rational d2/d1=±1 was established in [11] by modifying the classical denseness argument based on Pecherskii’s rearrangement theorem. Consequently, for two nonzero real numbers d1,d2 with d1=±d2, the shifts ζ(s+id1τ) and ζ(s+id2τ) are jointly universal. This implies that the only case for which (2) with n=2 remains unresolved is d1=1,d2=0, which is equivalent to the Riemann hypothesis (see [11 Theorem 1.4]).
The situation for n≥3 is much more difficult. In [12], the author proved joint universality of ζ(s+id1τ),…,ζ(s+idnτ) when K is a singleton and d1,…,dn are distinct positive rational numbers, namely for any fixed s=σ+it with 1/2<σ≤1 and t>0, the values logζ(s+idjτ) can simultaneously approximate arbitrary complex numbers. Very recently, Mishou [7] obtained joint universality for compact sets and positive integer parameters d1<⋯<dn satisfying certain additional separation conditions. Moreover, at a conference in Vilnius in September 2026, he announced that the separation conditions can be removed, so that joint universality holds for arbitrary distinct positive integers d1,…,dn. Nevertheless, the problem of characterizing the set of possible d1,…,dn satisfying joint universality of ζ(s+id1τ),…,ζ(s+idnτ), or at least (2), remained open.
The main difficulty is related to the linear dependence over Q of the numbers djlogp, where p runs over the primes. In the case considered by Nakamura, these numbers are linearly independent, and the classical Kronecker–Weyl theorem can be applied directly. For general real dj, relations can occur both among the numbers d1,…,dn and between djlogp corresponding to different primes. The former can be taken into account by choosing a Z-basis β1,…,βm of the additive group generated by d1,…,dn. To deal with the latter, we use a result due to Emsalem concerning linear relations among logarithms of algebraic numbers, which generalizes the six exponentials theorem (see [14, 15]). As in the case of two shifts considered in [10], this allows us to isolate a finite set S of exceptional primes such that the numbers βℓlogp, 1≤ℓ≤m, p∈/S, are linearly independent over Q.
Another crucial ingredient of our proof is a generalization of Pecherskii’s rearrangement theorem. Let H be a complex Hilbert space. In Voronin’s original approach [13], one first fixes a sequence (vp) in H and requires that, for every 0=g∈H,
p∑max(0,Re⟨vp,g⟩)=+∞,p∑min(0,Re⟨vp,g⟩)=−∞.
Together with ∑p∥vp∥2<∞, these conditions allow one to approximate any element of H by rearranging the series ∑pvp.
On the other hand, in Bagchi’s approach (see [1] or [6 Chapter 6, §6.1, Theorem 1.16]), one considers vectors zpvp, where ∣zp∣=1 and vp∈H is fixed for each p. The corresponding condition is
p∑z∈TsupRe⟨zvp,g⟩=p∑∣⟨vp,g⟩∣=∞(0=g∈H).
For example, in the proof of joint universality for Dirichlet L-functions, one takes vp=(χ1(p)p−s,…,χn(p)p−s), where χ1,…,χn are non-equivalent Dirichlet characters.
We generalize this criterion by replacing the sets {zvp:z∈T} with arbitrary nonempty bounded sets Ap⊂H. More precisely, assuming
p∑a∈Apsup∥a∥2<∞,
we prove that the condition
p∑a∈ApsupRe⟨a,g⟩=∞(0=g∈H)
implies that the finite sums ∑p∈Map, where ap∈Ap, are dense in H.
Our main result shows that joint universality of ζ(s+id1τ),…,ζ(s+idnτ) holds under the optimal conditions dj=0 and ∣dj∣=∣dk∣ for j=k. In particular, no linear independence assumptions on d1,…,dn are needed.
Theorem 1.1. Let n≥1 and let d1,…,dn be nonzero real numbers satisfying
∣dj∣=∣dk∣(1≤j<k≤n).(3)
For each 1≤j≤n, let Kj⊂D be compact with connected complement, and let fj be continuous and non-vanishing on Kj and holomorphic in its interior. Then for every ε>0,
The conditions on d1,…,dn in Theorem 1.1 are optimal for joint universality. Indeed, if dj=0, the corresponding function does not depend on τ. If dj=dk, the two corresponding shifts are identical. Finally, if dj=−dk, then for every real σ∈(1/2,1),
ζ(σ+idjτ)=ζ(σ+idkτ)(τ∈R),
so the values of ζ(s+idjτ) and ζ(s+idkτ) cannot approximate two independent functions.
As an immediate consequence of Theorem 1.1, we obtain generalized strong recurrence for every nonzero real parameters d1,…,dn.
Corollary 1.2. Let n≥2 and let d1,…,dn be arbitrary nonzero real numbers. Then for every compact set K⊂D with connected complement and every ε>0, we have
Proof. Let a1,…,am be the distinct numbers among ∣d1∣,…,∣dn∣, and put K∗={s:s∈K}. Choose a closed rectangle K0⊂D containing K∪K∗. By Theorem 1.1, applied to the positive parameters a1,…,am with all approximated functions equal to 1 on K0, there is a set of τ of positive lower density such that
1≤h≤mmaxs∈K0sup∣ζ(s+iahτ)−1∣<2ε.
Since ζ(s)=ζ(s), the same estimate holds on K for every shift ζ(s+idjτ), including those with dj<0. This implies (5). □
Corollary 1.2 states that Nakamura’s generalized strong recurrence problem is settled for any finite family of nonzero real parameters. Let us emphasize that if at least one parameter dj is zero and at least one is nonzero, (2) is equivalent to the Riemann hypothesis by Bagchi’s theorem and Theorem 1.1.
Linear independence and Emsalem’s theorem
As can be seen from the proof of the main result of [9], the proof of Theorem 1.1 is quite standard if the set of numbers {djlogp:j=1,2,…,n,p∈P} is linearly independent over Q. In such a case one can use the Kronecker-Weyl theorem to show that the quantities
2πdjτlogp(mod1),1≤j≤n,p∈Q,
can simultaneously approximate arbitrary given real numbers (mod 1), for any finite set of primes Q. This is a crucial step in many known proofs of universality and, simultaneously, the main obstacle in proving Theorem 1.1 for d1,…,dn when the numbers djlogp are linearly dependent over Q.
The case when ∑j=1ncjdj=0 for some numbers cj∈Q, not all zero, which is equivalent to ∑j=1ncjdjlogp=0 for every prime p, we shall treat by choosing a Z-basis β1,…,βm of the additive group generated by d1,…,dn, as in (6) below. Then, for each fixed prime p, the numbers β1logp,…,βmlogp are linearly independent over Q, and every djlogp is a linear combination of them with integer coefficients.
However, linear relations involving different primes may still occur. For example, if d1=log2 and d2=log3, then d1log3−d2log2=0. Such relations for n=2 were investigated in [10] using the six exponentials theorem.
Our first main observation is that, by applying a generalization of the six exponentials theorem due to Emsalem (see Theorem 2.1), one can show that only finitely many primes can occur in nontrivial linear relations among the numbers
βℓlogp,1≤ℓ≤m,p∈P.
More precisely, we prove that there exists a finite set of primes S such that if
p∑ℓ=1∑mcℓ,pβℓlogp=0,cℓ,p∈Q,
and only finitely many cℓ,p’s are non-zero, then cℓ,p=0 for p∈/S, 1≤ℓ≤m. Thus, the numbers βℓlogp with p∈/S are linearly independent over Q and are also independent of all the numbers βℓlogq with q∈S.
The set {βjlogq:q∈S,j=1,2,…,m} may still contain linear dependencies. We therefore choose a Z-basis λ1,…,λr of the additive group it generates. Consequently, the numbers
λ1,…,λr,βℓlogp(p∈/S,1≤ℓ≤m)
are linearly independent over Q, which enables us to apply the Kronecker-Weyl theorem.
Exceptional set and Emsalem’s theorem
L0=spanQ{logp:p prime}⊂{loga:a∈Q>0}=:L.
Theorem 2.1. Let m≥1 and let V be a complex hyperplane in Cm such that V∩Qm={0}. Then
dimQ(V∩Lm)≤m(m−1).
In particular, V∩L0m is of finite dimension over Q.
This follows from Emsalem’s theorem in the form stated in [14 Chapter 13, p. 13-2] (see also [15 Corollary 11.10]). The cited result concerns the larger Q-vector space {z∈C:ez∈Q×} of complex logarithms of nonzero algebraic numbers.
Before we apply Theorem 2.1 to our case, let us introduce some notations. Assume that d1,…,dn∈R∖{0} and put
G=j=1∑nZdj⊂R.
As a finitely generated torsion-free abelian group, G has a Z-basis β1,…,βm for some 1≤m≤n. In particular, β1,…,βm are linearly independent over Q and
dj=ℓ=1∑majℓβℓ,for suitable ajℓ∈Z.(6)
Define
Vβ={z∈Cm:ℓ=1∑mβℓzℓ=0}.
It is a complex hyperplane, with Vβ∩Qm={0}.
Theorem 2.2. There exists a finite set S of prime numbers such that for any finitely supported family bp∈G,
p∑bplogp=0⟹bp=0(p∈/S).(7)
Proof. Applying Theorem 2.1 to Vβ gives that the vector space
Vβ∩L0m⊆Vβ∩Lm
has finite dimension over Q, since Vβ∩Qm={0} and
dimQ(Vβ∩L0m)≤dimQ(Vβ∩Lm)≤m(m−1).
Choose a finite basis v(1),…,v(r) of Vβ∩L0m. Each coordinate of every v(a) is a finite linear combination over Q of logarithms of primes. Since the numbers logp, with p running over all primes, are linearly independent over Q, such a representation is unique. Thus, writing
vj(a)=p∈P∑cj,p(a)logp,cj,p(a)∈Q,
where only finitely many coefficients are nonzero, we define
S=a,j⋃{p∈P:cj,p(a)=0}.
Clearly, S is finite. Moreover, every vector of Vβ∩L0m has all its coordinates in spanQ{logq:q∈S}.
Suppose now that bp∈G and ∑pbplogp=0. Write bp=∑ℓcℓ,pβℓ with cℓ,p∈Z. We obtain
ℓ=1∑mβℓ(p∑cℓ,plogp)=0.
Therefore the vector (∑pc1,plogp,…,∑pcm,plogp)∈Vβ∩L0m. By the construction of S, every coordinate has a representation involving only logq with q∈S. Since the logarithms of distinct primes are linearly independent over Q, we have cℓ,p=0 for every p∈/S and every ℓ. This proves (7). □
Parametrization of the factors p−idjτ
Theorem 2.2 allows us to choose real numbers that are linearly independent over Q and hence to apply the classical Kronecker–Weyl theorem. However, the proof of Theorem 1.1 involves the factors
p−idjτ,1≤j≤n,p∈Q,
where Q is a finite set of primes, and the numbers djlogp need not be linearly independent over Q. We therefore describe how to parametrize these factors using the independent numbers obtained from Theorem 2.2.
The group ∑q∈SGlogq is finitely generated and torsion-free, and hence admits a Z-basis λ1,…,λr. We allow r=0 when S=∅. Although the numbers βℓlogq, 1≤ℓ≤m, q∈S, generate ∑q∈SGlogq, they may satisfy nontrivial linear relations over Q (see Remark 2.3). For every q∈S and 1≤j≤n, write
djlogq=h=1∑rbj,q,hλh,bj,q,h∈Z.(8)
Let Q⊇S be a finite set of primes. By (7), the numbers
λ1,…,λr,βℓlogp(p∈Q∖S,1≤ℓ≤m)(9)
are linearly independent over Q. Indeed, consider a linear combination over Q among the numbers in (9). Multiplying by a common denominator and expressing the terms involving λh as integer combinations of djlogq with q∈S show that, by (7), the coefficient in G at each p∈Q∖S must vanish. Since β1,…,βm are linearly independent over Q, all the coefficients at such primes are zero. The remaining coefficients are zero as well, because λ1,…,λr are linearly independent over Q.
Thus the Kronecker–Weyl theorem applies to the numbers in (9). To recover the factors p−idjτ from (9), we introduce the following parametrization.
Put NQ=r+m∣Q∖S∣ and identify
[0,1]NQ=[0,1]r×p∈Q∖S∏[0,1]m.
For u=(u1,…,ur)∈[0,1]r and vp=(vp,1,…,vp,m)∈[0,1]m, p∈Q∖S, define
ΨQ:[0,1]NQ⟶(Tn)∣Q∣
by
ΨQ(u,(vp)p∈Q∖S)=(ψS(u),(ψp(vp))p∈Q∖S),
where
ψS(u)=(e(h=1∑rbj,q,huh))j=1nq∈S
and
ψp(vp)=(e(ℓ=1∑majℓvp,ℓ))j=1n(p∈Q∖S).
Define HS=ψS([0,1]r) and
Hd=⎩⎨⎧(e(ℓ=1∑majℓvℓ))j=1n:v∈[0,1]m⎭⎬⎫.
The image of ψp is Hd for every p∈Q∖S. Consequently,
HQ:=ΨQ([0,1]NQ)=HS×p∈Q∖S∏Hd.(10)
Now let us put
αQ=−2π1(λ1,…,λr,(βℓlogp)p∈Q∖S,1≤ℓ≤m),
and, for τ∈R, define
γQ(τ)=((p−idjτ)j=1n)p∈Q.
Since the coefficients ajℓ and bj,q,h are integers, (6) and (8) give
γQ(τ)=ΨQ({ταQ})∈HQ,
where the fractional parts are taken coordinatewise. By (9), the coordinates of αQ are linearly independent over Q. The Kronecker–Weyl theorem therefore implies that {ταQ} is uniformly distributed modulo 1. In particular,
HQ={γQ(τ):τ∈R}.
In the proof of Theorem 1.1, we shall apply [4 Theorems A.8.1 and A.8.3] directly to {ταQ}.
Remark 2.3. The following example shows that HS does not have to be a product over primes in S.
Take n=m=2 and
d1=β1=log2,d2=β2=log3.
Since d2log2=d1log3 and d1,d2=0, we have {2,3}⊆S. Moreover,
2−id2τ=3−id1τ(τ∈R).
Since
HS={γS(τ):τ∈R},
every element ((zq,1,zq,2))q∈S of HS satisfies z2,2=z3,1.
On the other hand, since d1,d2 are linearly independent over Q, the Kronecker–Weyl theorem implies that, for every fixed q∈S, the projection of HS onto the coordinates corresponding to q is the whole T2. Consequently, HS cannot be the product of these projections.
Denseness lemma
For a bounded smooth Jordan domain U⊂U⊂D, let B2(U) denote the Bergman space of analytic functions g:U→C satisfying
∥g∥B2(U)2:=∬U∣g(s)∣2dσdt<∞.
We equip B2(U) with the inner product
⟨f,g⟩=∬Uf(s)g(s)dσdt.
It is well known (see [4 Theorem A.2.7]) that, for a compact set K⊂U, we have
s∈Ksup∣g(s)∣≤CK,U∥g∥B2(U).
We will work with the Hilbert space H=B2(U1)×⋯×B2(Un), where Uj⊂Uj⊂D are bounded smooth Jordan domains and
⟨u,g⟩H=j=1∑n∬Ujuj(s)gj(s)dσdt.
Our aim is to show that, for every v=(v1,…,vn)∈H and every ε>0, there exist a finite set M⊂P∖S and real numbers ϑp, p∈M, such that
v−p∈M∑k=1∑∞kpkse(kdjϑp)j=1nH<ε,(11)
where S denotes a finite exceptional set from Theorem 2.2. Choose 1/2<σ1<σ2<1 such that σ1<Res<σ2 on all Uj. For p∈/S and x∈R, define
A standard approach to proving (11) uses Pecherskii’s rearrangement theorem (see [4 Theorem A.6.1]). To apply this theorem directly, one would first need to choose a sequence of real numbers (ϑp)p∈/S, independently of g∈H, such that the fixed sequence (up(ϑp))p∈/S satisfies the assumptions of Pecherskii’s theorem. Note that the vectors up(x) cannot, in general, be written as zvp, where z∈T and vp∈H is independent of x. Thus, the version of Pecherskii’s theorem with unimodular coefficients mentioned in Introduction does not apply directly here.
More precisely, since H may be considered as a real Hilbert space with the inner product Re⟨u,g⟩H and
p∈/S∑x∈Rsup∥up(x)∥H2≪p∑p−2σ1<∞,(13)
in order to use Pecherskii’s theorem, it suffices to check (see [4 p. 243]) that, for every 0=g∈H,
Thus, a direct application of Pecherskii’s theorem would require a single sequence (ϑp) for which the real series
p∈/S∑Rej=1∑ne(djϑp)Δj(logp)
has both divergent positive and negative parts for every 0=g∈H. One possible approach, related to those used in [7, 10, 11], would be to choose ϑp so that, for suitable primes p,
Re(e(djϑp)Δj(logp))>0for all j with Δj(logp)=0.
Such a requirement is unnecessarily strong and may impose additional restrictions on d1,…,dn. Instead, we shall estimate the entire sum
Rej=1∑ne(djϑp)Δj(logp)
directly, without imposing separate sign conditions on its individual terms.
Since the values e(d1x),…,e(dnx) need not vary independently as x ranges over R, finding a single sequence (ϑp), independent of g, satisfying the required divergence conditions for every 0=g∈H is difficult.
We therefore use a more flexible variant of Pecherskii’s theorem (Lemma 3.2 below) with Ap={up(x):x∈R}. In view of (13), to obtain (11) it suffices to prove that
Let MA=supx∈RhA(x) and mA=−infx∈RhA(x). By (17), mA,MA>0 if A is nonzero. On the other hand, by the Cauchy–Schwarz inequality, we have MA,mA≤∑j∣Aj∣≤n∥A∥2. Moreover, since (hA(x)+mA)(MA−hA(x))≥0 for every x∈R, we have hA(x)2≤(MA−mA)hA(x)+MAmA. Hence, using (17) gives MAmA≥∥A∥22/2 and, in consequence,
MA≥2n∥A∥2,mA≥2n∥A∥2.
Hence, both inequalities in (16) hold with cd=1/(2n) when A=0. The case A=0 is trivial, and the proof is complete. □
Now, we prove the following variant of Pecherskii’s theorem.
Lemma 3.2. Let H be a complex Hilbert space and let Ak⊂H be nonempty bounded sets. Put bk=supa∈Ak∥a∥. Assume that
k∑bk2<∞,k∑a∈AksupRe⟨a,g⟩=∞(0=g∈H).(18)
Then, for every N≥0, the set
{k∈M∑ak:M⊂{k:k>N} is finite,ak∈Ak}
is dense in H.
The proof uses a Hahn–Banach separation argument similar to those in [3 Theorem 4(ii)] and [6 Chapter 6, § 6.1, Theorem 1.16]. To pass from sums with terms in conv(Ak) to sums with terms in Ak, we use the convergence of ∑ksupa∈Ak∥a∥2. A similar argument appears in [6 Chapter 6, § 6.1, Lemma 1.15], where sums with coefficients in the unit disc are approximated by sums with coefficients on the unit circle.
Proof. By replacing Ak with Ak∪{0}, we may assume that 0∈Ak. Indeed, the first condition in (18) remains unchanged, while the second is preserved since the corresponding supremum can only increase. Moreover, the set of finite sums in the conclusion is unchanged.
For each k, put bk=supu∈Ak∥u∥. For integers M>L, let
BL,M={k=L+1∑Mmk:mk∈conv(Ak)}.
where conv(Ak) denotes the real convex hull of Ak. Obviously, the sets BL,M are convex. Moreover, the fact that 0∈Ak implies BL,M⊆BL,M+1. Hence
BL=M>L⋃BL,M
is closed and convex.
We claim that BL=H. Otherwise, by the real Hahn–Banach separation theorem and the Riesz representation theorem, there exist g∈H, g=0, and α∈R such that
By omitting the indices for which uk,rk=0, we obtain a finite sum of elements of the original sets Ak, with all indices greater than N. This completes the proof. □
Thus, we are ready to formulate the following denseness lemma.
Theorem 3.3. Assume that dj=0 for all j and ∣dj∣=∣dk∣ whenever 1≤j<k≤n. For every v∈H, ε>0 and y>0, and for any real ϑp at the primes p≤y outside S, there exist a finite set M⊂P∖S containing all primes p≤y outside S, and real numbers ϑp for all p∈M, such that
Choose j0 with gj0=0. We claim that ∑p∈/S∣Δj0(logp)∣=∞. By [5 Lemma 3,4, and p. 229], there are xν→∞, constants B,c>0, and intervals Jν⊂[xν,xν+1] such that
∣Jν∣≥Bxν−2,∣Δj0(t)∣≥ce−σ2xν(t∈Jν).
The prime number theorem gives ∣{p:logp∈Jν}∣≫xν3exν. Hence, since σ2<1,
logp∈Jν∑∣Δj0(logp)∣≫xν3e(1−σ2)xν⟶∞,
which proves our claim.
Thus, we have
p∈/S∑a∈ApsupRe⟨a,g⟩H=∞,
and we are ready to apply Lemma 3.2. Let us fix y>0 and choose y′≥y such that S⊂{p∈P:p≤y′}. Write Py′′={p∈P∖S:p≤y′}. For each p∈Py′′, retain already fixed ϑp for p≤y, and choose ϑp∈R arbitrarily for y<p≤y′. Put vy′′=v−∑p∈Py′′up(ϑp). Then, by Lemma 3.2, one can find a finite set of primes M′⊂{p∈P:p>y′} and real numbers ϑp for each p∈M′ such that
p∈M′∑up(ϑp)−vy′′H<ε.
Therefore, taking M=Py′′∪M′ yields (19). □
Proof of Theorem 1.1
We now complete the proof by following the standard approach originating with Voronin. The only point requiring special attention is the finite exceptional set S.
Proof of Theorem 1.1. Fix compact sets Kj, functions fj, and ε>0 as in the statement. By Mergelyan’s theorem, since fj has no zeros on Kj, one can find polynomials Pj such that
1≤j≤nmaxs∈KjsupePj(s)−fj(s)<ε/3.(20)
Choose bounded smooth Jordan domains Uj with Kj⊂Uj⊂Uj⊂D, and fix real numbers 1/2<σ1<σ2<1 such that σ1<Res<σ2 on every Uj. Write H=B2(U1)×⋯×B2(Un) and choose CB≥1 so that
Since C is finite and ∣ωp,j∣=1, these series converge absolutely and uniformly for s∈Uj. Moreover, let us recall (see Section 2.2) that for the map γC(τ)=((e(−2πdjτlogp))j=1n)p∈C, we have γC(τ)=ΨC({ταC})∈HC, where
We next verify that this estimate remains valid for vectors sufficiently close to ω∗. Recall that Res≥σ1>1/2 on each Kj. For any z,w∈T and k≥1, the factorization
zk−wk=(z−w)ℓ=0∑k−1zk−1−ℓwℓ
implies
∣zk−wk∣≤k∣z−w∣.
Therefore, one can easily notice that for sufficiently small ρ>0 and for every
ω∈V:=⎩⎨⎧ω∈HC:p∈C1≤j≤nmax∣ωp,j−ωp,j∗∣<ρ⎭⎬⎫,
we have
1≤j≤nmaxs∈Kjsup∣Fj(s,ω)−Pj(s)∣<4η.
Finally, using (23), we conclude that
1≤j≤nmaxs∈Kjsup∣Fj(s,γC(τ))−Pj(s)∣<4η.(24)
provided γC(τ)∈V.
Recall that ω∗∈HC, so there is t∗∈[0,1)NC such that ΨC(t∗)=ω∗. More precisely, we may take
t∗=(r0,…,0,({βℓϑp})p∈M,1≤ℓ≤m).
Since ΨC is continuous and V is an open neighborhood of ω∗ in HC, there exists 0<δ0<1/2 such that
ΨC(Bδ0)⊂V,
where
Bδ0={t∈[0,1]NC:1≤ν≤NCmax∥tν−tν∗∥≤δ0}.
Here ∥⋅∥ denotes the distance to the nearest integer. In particular, the Lebesgue measure of Bδ0 is
meas(Bδ0)=(2δ0)NC>0.
For T>0, define
AT={τ∈[T,2T]:{ταC}∈Bδ0}.
Then, by (9) and the Kronecker–Weyl theorem (see [4 Theorem A.8.1]), we have
T→∞limTmeas(AT)=(2δ0)NC>0.(25)
Moreover, for every τ∈AT, we have γC(τ)=ΨC({ταC})∈V, so (24) holds for every τ∈AT.
Let z≥2 be fixed, with C⊆Qz:={p:p≤z}. Recalling (10) gives
t=(u,(vp)p∈Qz∖S)∈[0,1]NQz,
where u=(u1,…,ur)∈[0,1]r parametrizes the exceptional block HS, while vp=(vp,1,…,vp,m)∈[0,1]m parametrizes Hd with p∈/S.
[1]B. Bagchi, The statistical behavior and universality properties of the Riemann zeta-function and other allied Dirichlet series, Ph.D. thesis, Indian Statistical Institute, Calcutta, 1981.
[2]B. Bagchi, Recurrence in topological dynamics and the Riemann hypothesis, Acta Math. Hungar. 50 (1987), 227–240.DOI
[3]A. Good, On the distribution of the values of Riemann’s zeta function, Acta Arith. 38 (1980/81), no. 4, 347–388.DOI
[4]A. A. Karatsuba and S. M. Voronin, The Riemann Zeta-Function, Walter de Gruyter, Berlin–New York, 1992.
[5]J. Kaczorowski and M. Kulas, On the non-trivial zeros off the critical line for L-functions from the extended Selberg class, Monatsh. Math. 150 (2007), 217–232.DOI
[6]A. Laurinčikas, Limit Theorems for the Riemann Zeta-Function, Mathematics and Its Applications, vol. 352, Kluwer Academic Publishers, Dordrecht, 1996.DOI
[7]H. Mishou, Joint universality for shifts of the Riemann zeta function with integer multiplicities, preprint, 2026.
[9]T. Nakamura, The joint universality and the generalized strong recurrence for Dirichlet L-functions, Acta Arith. 138 (2009), 357–362.DOI
[10]Ł. Pańkowski, Some remarks on the generalized strong recurrence for L-functions, in: New Directions in Value Distribution Theory of Zeta and L-Functions, Ber. Math., Shaker Verlag, Aachen, 2009, pp. 305–315.
[11]Ł. Pańkowski, Joint universality and generalized strong recurrence with rational parameter, J. Number Theory 163 (2016), 61–74.
[12]Ł. Pańkowski, Joint value-distribution of shifts of the Riemann zeta-function, Results Math. 77 (2022), article 76.arxiv.org/abs/2010.08332
[13]S. M. Voronin, Theorem on the universality of the Riemann zeta-function, Math. USSR-Izv. 9 (1975), 443–453.DOI
[14]M. Waldschmidt, Linear Independence of Logarithms of Algebraic Numbers, The Institute of Mathematical Sciences, Madras, IMSc Report 116 (1992), Chapter 13, p. 13-2, available at https://webusers.imj-prg.fr/~michel.waldschmidt/articles/pdf/LIL.pdf.
[15]M. Waldschmidt, Diophantine Approximation on Linear Algebraic Groups: Transcendence Properties of the Exponential Function in Several Variables, Grundlehren der mathematischen Wissenschaften, vol. 326, Springer, Berlin, 2000.