It is known that the Fourier transform $\mathcal F$ maps $L^2 \to L^2$ as an (isometric) isomorphism and $L^1 \to L^\infty$ as bounded operator. Via Riesz-Thorin this result can be extended to give that $\mathcal F$ also maps $L^p \to L^{p'}$ where $1 = \frac{1}{p} + \frac{1}{p'}$ as a bounded operator, i.e. the Hausdorff-Young equality holds: $$ \|\mathcal F(u) \|_{p'} \leq \|u\|_p $$ for all $u \in L^p(\mathbb{R}^n)$, $p \in (1,2)$.

How do we know that the Fourier transform is only an isomorphism from $L^2 \to L^2$? One could argue that no $L^p$ space is isomorphic to an $L^q$ space for $p \neq q$ using the invariants type and cotype as suggested here but, considering that I only want to show that $\mathcal F$ is not an isomorphism from $L^p \to L^{p'}$, I suppose there is a more simple approach directly related to the Fourier transform. I am thinking of showing the fourier transform from $L^p \to L^{p'}$ is simply not surjective.

Can you think of an easier approach to this problem?

  • $\begingroup$ I'd start with the same approach; the non-surjectivity of $\mathcal F:L_1\to L_\infty$ follows from an easy observation that constant non-zero functions do not belong to $\mathcal F(L_1)$. Maybe in your case you can find a very slowly decreasing function in $L_{p'}$ that can not be obtained as an image of a $L_p$ function. $\endgroup$ Aug 29, 2017 at 14:34
  • $\begingroup$ I want to point out that also injectivity may fail, the continuous extension of an injective operator need not be injective anymore. Concerning your suggestion, some remarks/possible issues: Since the Fourier transform is an isomorphism on the Schwartz functions, its range is always dense on $L^p$ despite in the case $p=1$, so the situation there is quite different. I would rather try to show that its range is not closed. But all of this is not so easy because the Fourier transform on $L^p$ for $p\in(1,2)$ is only given by interpolation and no representation is available. $\endgroup$ Aug 31, 2017 at 7:54
  • $\begingroup$ @SebastianBechtel: In general, the extension of an injective operator need not be injective. But since we have $L^p \subset \mathcal{S}'$ (tempered distributions), and since the Fourier transform is injective on $\mathcal{S}'$, the extension is also injective in this case. Nevertheless, I think your idea is the only sensible one, i.e., one needs to show that we do not have $\mathcal{F} f \|_{L^q} \gtrsim \| f \|_{L^p}$. $\endgroup$
    – PhoemueX
    Aug 31, 2017 at 8:15
  • $\begingroup$ @PhoemueX Good point with the distributional Fourier transform! $\endgroup$ Aug 31, 2017 at 8:35

2 Answers 2


Here is a (modified) argument which I used repeatedly in a paper of mine. Possibly, this is some form of type and cotype argument (I am not too familiar with these notions), but maybe it is elementary enough for you:

I will make extensive use of the Khintchine inequality, which states that if $\varepsilon_1, \dots, \varepsilon_N$ are independent and uniformly distributed in $\{1, -1\}$, then we have $$ \mathbb{E} \left|\sum_{n=1}^N \varepsilon_n x_n\right|^p \asymp \left(\sum_{n=1}^N |x_n|^2\right)^p , $$ for arbitrary $x_1,\dots, x_N \in \Bbb{C}$. Here, $\Bbb{E}$ denotes the expected value with respect to $\varepsilon = (\varepsilon_1,\dots, \varepsilon_N)$. It is crucial that the implied constants above are independent of $x_1,\dots, x_N$ and of $N$; they only depend on the choice of $p \in (0,\infty)$.

Now, assume that $\mathcal{F} : L^p \to L^q$ is an isomorphism. Let us first rule out the case $q = \infty$: If we had $q=\infty$, then there would be some $f \in L^p$ with $\widehat{f} \equiv 1$, where I write $\widehat{f} = \mathcal{F}f$. But we also have $\widehat{\delta_0} \equiv 1$, where we are using the formalism of tempered distributions, and where $\delta_0$ is the Dirac measure at the origin. But this implies $f = \delta_0$ as tempered distributions, which is absurd.

Hence $q \in (0,\infty)$. Now, let $\varepsilon_1,\dots, \varepsilon_N$ as above. For $n \in \Bbb{N}$, let $y_n := (n,0,\dots,0) \in \Bbb{R}^d$, and set $f_n := T_{y_n} 1_{(0,1)^d}$, where $(T_y f)(x) = f(x-y)$. Note that $\widehat{f_n} = M_{-y_n} \widehat{f}$ for $f := 1_{(0,1)}$, where $M_\xi f(x) = e^{2\pi i \langle x, \xi \rangle} f(x)$ denotes the modulation of $f$. Finally, observe that the supports of $f_n, f_m$ are disjoint for $n \neq m$. All in all, these properties imply \begin{align*} N^{q/p} = \Bbb{E} \, \left\| \sum_{n=1}^N \varepsilon_n \cdot 1_{(n,n+1)} \right\|_{L^p}^q & \asymp \Bbb{E} \, \left\| \sum_{n=1}^N \varepsilon_n \cdot \mathcal{F}[ 1_{(n,n+1)} ] \right\|_{L^q}^q \\ & = \Bbb{E} \int \left| \sum_{n=1}^N \varepsilon_n e^{2\pi i \langle y_n, \xi\rangle} \right|^q \cdot |\widehat{f}(\xi)|^q \, d\xi \\ (\text{Fubini's theorem}) & = \int \Bbb{E} \left| \sum_{n=1}^N \varepsilon_n e^{2\pi i \langle y_n, \xi\rangle} \right|^q \cdot |\widehat{f}(\xi)|^q \, d\xi \\ (\text{Khintchine inequality}) & \asymp N^{q/2} \cdot \| \widehat{f} \|_{L^q}^q \\ & \asymp N^{q/2} \cdot \|f\|_{L^p}^q = N^{q/2}, \end{align*} where the implied constants are independent of $N \in \Bbb{N}$. This can only hold for $p = 2$.

Now, let us show that also $q = 2$. To see this, note for arbitrary $f \in L^p = L^2$ by Plancherel's theorem that $$ \|\widehat{f} \|_{L^2} = \| f \|_{L^2} = \| f \|_{L^p} \asymp \|\widehat{f}\|_{L^q} . $$ Thus, we get for arbitrary $L^2$ functions $g \in L^2$ that $\|g\|_{L^2} \asymp \| g \|_{L^q}$, which easily implies $q = 2$, by considering e.g. $g = \sum_{n=1}^N f_n$ with $f_n$ as above.


EDIT: There seems to be an error in my calculations.

The fact that $\mathcal F$ is no isomorphism between $L^p$ spaces for $p \neq 2$ can be proven using the same argument that is used to determine the sobolev conjugate.

Suppose that for $p,p'$ with $\frac{1}{p} + \frac{1}{p'} = 1$ the Fourier transform is an isomorphism between $L^p$ and $L^{p'}$. Then for all $f \in \mathcal S$, where $\mathcal S$ denotes the Schwartz space, we have $$ \|f\|_p = \|\mathcal{F}^{-1} \mathcal{F} (f)\|_{p} \leq \|\mathcal{F}^{-1}\|_{L^{p'} \to L^p} \|\mathcal{F}(f)\|_{p'} = C \, \|\mathcal F (f)\|_{p'}. $$

Now, consider for $\lambda > 0$ the dilation $\delta^\lambda f$ defined by $$ \delta^\lambda f(x):= f(\lambda x). $$ We then have $$ \lambda^{-\frac{n}{p}} \|f\|_p = \|\delta^\lambda f\|_p \leq C \; \|\mathcal{F}(\delta^\lambda f)\|_{p'} = C \, \lambda^{-n} \|\delta^{\frac{1}{\lambda}} \, \mathcal{F}(f) \|_{p'} = C \, \|\mathcal{F}(f)\|_{p'}. $$ Letting $\lambda \to 0$ shows that there is no such constant $C > 0$.

Here is the error I made...

The second line of inequalities should read $$ \lambda^{-\frac{n}{p}} \|f\|_p = \|\delta^\lambda f\|_p \leq C \; \|\mathcal{F}(\delta^\lambda f)\|_{p'} = C \, \lambda^{-n} \|\delta^{\frac{1}{\lambda}} \, \mathcal{F}(f) \|_{p'} = C \, \lambda^{-n + \frac{n}{p'}}\|\mathcal{F}(f)\|_{p'}, $$ which after reordering of the $\lambda$-terms just gives the first inequality. Just in the case of the sobolev conjugate, this proof only shows that if $\mathcal F$ is an isomorphism then $p,p'$ necessarily have to be conjugate.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.