MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

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

Let $\cal C$ be a model category which is also additive. Suppose that the homotopy category $\operatorname{Ho}\mathcal C$ is additive, for example this is true when the weak equivalences in $\cal C$ is closed under biproducts (see this question).

If we take a cofibrant object $X$ and a fibrant object $Y$ then there is a natural isomorphism $$ \operatorname{Ho}\mathcal C(X,Y) \cong \mathcal C(X,Y)/\sim $$ where $\sim$ is the homotopy relation. Is this always a group isomorphism?

share|cite|improve this question
up vote 4 down vote accepted

If by "additive" you mean an $\mathbf{Ab}$-enriched category with a zero object and biproducts, then yes. Let $\mathcal{M}$ be model category that is additive in this sense, let $\mathcal{M}_c$ be the full subcategory of cofibrant objects, let $\mathcal{M}_f$ be the full subcategory of fibrant objects, and let $\mathcal{M}_{cf} = \mathcal{M}_c \cap \mathcal{M}_f$. Here are the relevant facts: $\DeclareMathOperator{\Ho}{Ho}$

  1. The coproduct of a family of cofibrant objects is automatically a homotopy coproduct, so the localising functor $\mathcal{M}_c \to \Ho \mathcal{M}$ preserves coproducts. Dually, the localising functor $\mathcal{M}_f \to \Ho \mathcal{M}$ preserves products.
  2. Hence, the localising functor $\mathcal{M}_{cf} \to \Ho \mathcal{M}$ preserves the zero object and biproducts. (Note that $\mathcal{M}_{cf}$ is an additive subcategory of $\mathcal{M}$.)
  3. A category with a zero object and biproducts is automatically enriched over commutative monoids in a unique way, and a functor that preserves zero objects and biproducts is similarly enriched. Thus, there is a unique enrichment of $\Ho \mathcal{M}$ over commutative monoids that makes the localising functor $\mathcal{M}_{cf} \to \Ho \mathcal{M}$ an enriched functor.
  4. Since $\mathcal{M}_{cf}$ is actually $\mathbf{Ab}$-enriched and the localising functor $\mathcal{M}_{cf} \to \Ho \mathcal{M}$ is full and essentially surjective on objects, $\Ho \mathcal{M}$ is also $\mathbf{Ab}$-enriched.

Now, let $X$ and $Y$ be any two objects in $\mathcal{M}$. In order for the hom-set map $$\mathcal{M}(X, Y) \to \Ho \mathcal{M}(X, Y)$$ to be a group homomorphism, it is sufficient that the localising functor $\mathcal{M} \to \Ho \mathcal{M}$ preserve either the coproduct $X + X$ or the product $Y \times Y$. (We already know that it preserves initial and terminal objects.) Thus it suffices to take either $X$ cofibrant or $Y$ fibrant.

share|cite|improve this answer
Why did you start with "if by additive you mean..."? Is there any other definition of additive commonly in use? – Omar Antolín-Camarena Nov 14 '13 at 3:54
Not that I'm aware of, but it's always good to be explicit! (By contrast the phrase ‘preadditive category’ seems to have at least two common definitions.) – Zhen Lin Nov 14 '13 at 8:08

Yes, because the projection functor from the model category to the homotopy category preserves coproducts of cofibrant objects. That is actually the way of showing that the homotopy category has coproducts.

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.