Mac Lane homology is a homology theory for (not necessarily commutative) rings. Given a ring $A$, Eilenberg and Mac Lane define its cubical construction $QA$ to be a certain connective chain complex, whose homology is isomorphic to the stable homology of Eilenberg-Mac Lane spaces: $$H_i(QA)\cong H_{i+j}(K(A,j)) \text{ for } j \gg i.$$In fact, $QA$ is a dg-ring, and it comes with a ring map $QA \to A$ inducing an isomorphism on $H_0$. You can find the construction and basic properties of the cubical construction in [1] or [2, Chapter 13]. One then defines the Mac Lane homology of $A$ to be the Hochschild homology $$HML_i(A):= HH_i(QA,A)$$and similarly the Mac Lane cohomology is $HML^i(A):=HH^i(QA,A)$.

In 1992, Pirashvili and Waldhausen [3] proved that $HML_i(A)\cong THH_i(A)$, where the right-hand side is topological Hochschild homology. The proof goes by identifying both of them with a functor homology group. Subsequently in 1995, Fiedorowicz-Pirashvili-Schwaenzl-Vogt-Waldhausen [1] outlined a `brave new algebra' proof of this fact. In modern terms, they noticed that the Pirashvili-Waldhausen result would follow if one has a stable equivalence of $H\mathbb{Z}$-algebra spectra $$HQA \simeq HA \wedge_{\mathbb{S}} H\mathbb{Z}.$$Here, $H$ denotes the Eilenberg-Mac Lane functor taking dgas to $H\mathbb{Z}$-algebra spectra, and $\mathbb{S}$ is the sphere spectrum. (For concreteness, my preferred model of spectra is symmetric spectra in simplicial sets.) Indeed, if one has such a stable equivalence, then it follows by base change results that $HML(A)\simeq THH(A)$, for both homology and cohomology.

My question: Is it known that there is a stable equivalence $HQA \simeq HA \wedge_{\mathbb{S}} H\mathbb{Z}$ of $H\mathbb{Z}$-algebra spectra?

This may be a fact known to experts in $THH$, but I was unable to find anything in the literature. One can almost get there: because the homology of $QA$ is the stable cohomology of Eilenberg-Mac Lane spaces, it follows that one has $$\pi_i(HQA)\cong \pi_i(HA \wedge_{\mathbb{S}} H\mathbb{Z}).$$Moreover, both are $H\mathbb{Z}$-module spectra, hence wedges of Eilenberg-Mac Lane spectra, and so an isomorphism on $\pi_*$ actually lifts to a stable equivalence $HQA \simeq HA \wedge_{\mathbb{S}} H\mathbb{Z}$ of $H\mathbb{Z}$-module spectra. This argument already appears in [1], but it wasn't clear to the authors then, and it's certainly not clear to me now, how to lift this to a stable equivalence of $H\mathbb{Z}$-algebra spectra. I guess because $QA$ is connective, in theory one could write down both sides of the equation and match up the algebra structure somehow, but I am not sure if this is very tractable.

Thanks for your time.


[1] Fiedorowicz, Z.; Pirashvili, T.; Schwänzl, R.; Vogt, R.; Waldhausen, F. Mac Lane homology and topological Hochschild homology. Math. Ann. 303 (1995)

[2] Jean-Louis Loday, Cyclic homology, Springer 1998

[3] Pirashvili, Teimuraz; Waldhausen, Friedhelm. Mac Lane homology and topological Hochschild homology. J. Pure Appl. Algebra 82 (1992)

  • 5
    $\begingroup$ Great question, and great write-up of a question. I would give more than one +1 if I could. $\endgroup$ – Theo Johnson-Freyd Aug 26 '20 at 21:07

The answer to your question is yes, these two $H\mathbb{Z}$-algebras are equivalent (and if $A$ is commutative, they are equivalent as commutative $H\mathbb{Z}$-algebras). I am in fact in the process of writing this up with Maxime Ramzi. Hopefully the paper will be finished in a week or 2.

Update : The paper has now appeared on arXiv


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.