MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

For a sphere $S^n$, the diagonal map $\Delta:S^n\to S^n\wedge S^n$ sending $x\mapsto x\wedge x$ is null-homotopic. This is the homotopy group $\pi_n(S^n\wedge S^n)=\pi_n(S^{2n})$ is trivial.

I'm wondering if there are other examples of this phenomenon. That is, is there a non-contractible based space $X$ which is not a sphere, such that the diagonal map $\Delta:X\to X\wedge X$ is null-homotopic?

share|cite|improve this question
Any contractible space? – Daniel Litt Feb 22 '11 at 6:52
Thank you Daniel. I will edit my question. – Gao 2Man Feb 22 '11 at 7:16
up vote 16 down vote accepted

Questions like this are well studied by homotopy theorists, in particular those working in Lusternik-Schnirelmann category (warning: nothing to do with the objects and morphisms kind of category!). In particular, what you are asking for is examples of spaces with weak category equal to 2. The weak category of a pointed space $X$ is the least $k$ such that the reduced diagonal $\triangle_k\colon X\to X\wedge X\wedge\cdots\wedge X = X^{\wedge k}$ is null-homotopic. The weak category was introduced by Berstein and Hilton in this paper, as an approximating invariant for LS-category: note that $\mathrm{wcat}(X)\leq\mathrm{cat}(X)$ for any space. (Warning 2: Nowadays, the prevailing convention seems to be that these invariants are normalised, so that the weak category of a contractible space is zero.)

Note that the weak category of any (non-contractible) suspension $\Sigma Y$ is $2$. In the paper of Berstein and Hilton linked above they construct spaces $Z$ with weak category $2$ and LS-category $3$ (which therefore cannot be suspensions).

share|cite|improve this answer
To make the conclusion explicit: if $X=\Sigma Y$, then $\Delta:X\to X\wedge X$ is nullhomotopic. – Neil Strickland Feb 22 '11 at 11:06
And the converse is false. – Mark Grant Feb 22 '11 at 16:45
@Mark: But see John Klein's answer for a case when the converse is true. – Tom Church Feb 23 '11 at 2:18

Take an $n$-connected CW complex $X$ of dimension $<2n$. Then $X \wedge X$ is $2n$-connected, thus the diagonal map has to be null-homotopic.

share|cite|improve this answer

To add to the above: an $r$-connected CW complex $X$ of dimension $\le 3r$ has the homotopy type of a suspension $\Leftrightarrow$ its reduced diagonal map $X \to X \wedge X$ Is null-homotopic. This is the Berstein-Hilton-Ganea theorem.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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