CounterexampleConstruction: Literature-sourcedVerification: AI-adaptedPipeline-generatedjudge pass (gpt-5.6-terra)audited 2026-09-07

How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

The Carleson maximal operator is not strong type (1,1)

Statement refuted

There exists a finite constant bounding $∥ C f ∥_{1}$ by that constant times $∥ f ∥_{1}$ for every $f \in L^{1} (T)$ .

In fact, for every $A > 0$ there is a nonnegative trigonometric polynomial $f$ on the period-one torus, with Haar mass one, such that

$∥ f ∥_{1} = 1, ∥ C f ∥_{1} > A .$

As a further consequence under DC, there exists a real $h \in L^{1} (T)$ with $\sup_{N} ∥ S_{N} h ∥_{1} = \infty$ .

Facts & Assumptions

Given: $A > 0$ and normalized Haar measure on $T = R / Z$ . DC is assumed only for the further norm-divergence consequence.

[F1]

The measurable maximal operator is $C f = \sup_{N \geq 0} ∣ S_{N} f ∣$ for $f \in L^{1}$ (Carleson maximal partial-sum operator).

[F2]

The Fejer kernel satisfies $F_{K} (t) = (K + 1)^{- 1} ∣ \sum_{j = 0}^{K} e^{2 π i j t} ∣^{2} \geq 0$ and $\int_{0}^{1} F_{K} (t) d t = 1$ for every $K \geq 0$ (The Fejer kernel is a positive approximate identity).

[F3]

Fejer means of a continuous one-periodic complex function converge uniformly to that function (Fejer means converge uniformly for continuous periodic functions).

[F4]

For every one-period integrable $f$ and every $x$ , $S_{N} f (x) = (f * D_{N}) (x)$ (Fourier partial sums are Dirichlet convolutions).

[F5]

The quantities $∥ D_{N} ∥_{1}$ equal the norms of the partial-sum operators on continuous functions and are at least $(3 π)^{- 1} \log (N + 1)$ for $N \geq 1$ (Fourier partial-sum operator norm equals the Lebesgue constant).

[F6]

For every measure space and $1 \leq p \leq \infty$ , $L^{p}$ is complete in its norm (Riesz-Fischer completeness of $L^{p}$ for $1 \leq p \leq \infty$ ).

[F7]

Under DC, pointwise bounded families of bounded linear maps from a Banach space to a normed space have uniformly bounded norms (Uniform boundedness principle).

[F8]

For two sigma-finite measure spaces and a nonnegative product-measurable function, the product integral equals both iterated integrals (Tonelli's theorem for nonnegative measurable functions on a sigma-finite product).

Counterexample

technique · Fejer-kernel tests, with an additional uniform-boundedness consequence

1.1

Choose an integer $N \geq 1$ with $∥ D_{N} ∥_{1} > A + 1$ , possible from the logarithmic lower bound. For each $K \geq 0$ , the function $F_{K}$ is a real nonnegative trigonometric polynomial of $L^{1}$ norm one.

F2F5given

1.2

Expanding the square formula for $F_{K}$ gives coefficient $1 - ∣ k ∣ / (K + 1)$ for $∣ k ∣ \leq K$ and zero otherwise. Orthogonality of finite characters therefore gives $S_{N} F_{K} = F_{K} * D_{N} = σ_{K} D_{N}$ : either side has coefficient $1 - ∣ k ∣ / (K + 1)$ for $∣ k ∣ \leq \min {K, N}$ and zero elsewhere. Since $D_{N}$ is continuous, $∥ S_{N} F_{K} - D_{N} ∥_{\infty} \to 0$ as $K \to \infty$ , with $N$ fixed.

F2F3F4

1.3

For the additional consequence, let $h$ be a real integrable periodic representative. The convolution formula, after a periodic change of variables, gives $∣ S_{N} h (x) ∣ \leq \int_{0}^{1} ∣ h (u) ∣ ∣ D_{N} (x - u) ∣ d u$ . The integrand is product-measurable: $h (u)$ depends on one coordinate and $D_{N} (x - u)$ is continuous. Both measure spaces are finite, so Tonelli and translation invariance give $∥ S_{N} h ∥_{1} \leq \int_{0}^{1} ∣ h (u) ∣ \int_{0}^{1} ∣ D_{N} (x - u) ∣ d x d u = ∥ D_{N} ∥_{1} ∥ h ∥_{1}$ . Thus $S_{N}$ is a bounded real linear operator on real $L^{1}$ .

F4F8

2.1

Choose $K$ large enough that this uniform error is less than one. The measure has mass one, so $∥ S_{N} F_{K} ∥_{1} \geq ∥ D_{N} ∥_{1} - ∥ S_{N} F_{K} - D_{N} ∥_{1} > A$ . Since $C F_{K} \geq ∣ S_{N} F_{K} ∣$ , the polynomial $f = F_{K}$ is the required witness. Its maximal function is finite and bounded: when the partial-sum index is at least $K$ , the partial sum equals $F_{K}$ , so only finitely many distinct continuous polynomials enter its supremum. Thus the strict norm inequality is an ordinary finite integral, not an artifact of an infinite value.

F1step 1.1step 1.2

2.2

For each fixed $N$ , step 1.2 and the unit-norm tests $F_{K}$ imply $∥ S_{N} : L^{1} (T, R) \to L^{1} (T, R) ∥ \geq \lim_{K \to \infty} ∥ S_{N} F_{K} ∥_{1} = ∥ D_{N} ∥_{1}$ . These operator norms are consequently unbounded.

F2F5step 1.2step 1.3

3.1

Now assume DC. Real $L^{1}$ is Banach by completeness. If $\sup_{N} ∥ S_{N} h ∥_{1}$ were finite for each $h$ , uniform boundedness would contradict step 2.2. Therefore some real integrable $h$ has unbounded $L^{1}$ norms of its partial sums. This is a norm-divergence conclusion and does not assert almost-everywhere divergence.

F6F7step 1.3step 2.2given∎

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

29 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources

Laugesen, Harmonic Analysis Lecture Notes (standard reference, not scraped)