Questions: Opposite Categories and Duality

Question 1

A student has just proven the universal property of a product in a category C. They now need to understand coproducts. What is the most efficient approach given the duality principle?

• A. Look up the coproduct definition separately, since products and coproducts are defined differently and independently
• B. Systematically reverse all the arrows in the product's universal property to obtain the coproduct's universal property
• C. Check whether products and coproducts happen to coincide in C, then use whichever definition applies
• D. Study coproducts in the category of sets first, then generalize

Answer: (B)

This is the duality principle in action. Products are defined by a universal cone with maps INTO the product object; reversing every arrow gives an object with maps OUT of it — exactly the coproduct. The student gets the coproduct definition for free because it is literally the product definition in C^op, restated in C. Option A is the inefficient approach that treats dual constructions as independent when they are logically the same construction in mirror image.

Question 2

Which statement correctly describes C^op, the opposite category of C?

• A. C^op contains only those morphisms in C that have inverses, making it a subcategory of C
• B. C^op has the same objects as C and all the same morphisms, but with every arrow's direction reversed, and it is always a valid category
• C. C^op exists only when C is isomorphic to its own opposite category
• D. C^op has different objects than C because reversing arrows changes the 'type' of each object

Answer: (B)

C^op keeps every object of C and keeps every morphism, but flips the direction: a morphism f: A → B in C becomes f^op: B → A in C^op. Composition is redefined as f^op ∘^op g^op = (g ∘ f)^op to preserve the category axioms. The result is always a valid category — no conditions on C are needed. Option A confuses 'opposite' with 'inverse'; not every morphism needs an inverse to appear in C^op. Option C confuses C^op with the concept of a self-dual category.

Question 3

A contravariant functor F: C → D is mathematically the same thing as a covariant functor F: C^op → D, so the opposite category provides a unified framework where all functors can be treated as covariant.

• T. True
• F. False

Answer: True

A contravariant functor reverses the direction of morphisms: where f: A → B in C, it sends F(f): F(B) → F(A) in D. But in C^op, morphisms already run in the reversed direction — so a functor that is 'contravariant on C' is simply covariant on C^op. This unification is important: it means the theory of functors only needs to be developed for covariant functors, and contravariance is handled by passing to the opposite domain category.

Question 4

In the category of sets, the product A × B and the coproduct A ⊔ B are the same object, because any category with finite products also has products and coproducts coinciding.

• T. True
• F. False

Answer: False

Products and coproducts coincide (as direct sums) only in additive categories — categories where Hom-sets carry abelian group structure and composition is bilinear. Set is not additive: there is no natural way to add two functions between sets. In Set, the product A × B is the Cartesian product (pairs of elements) while the coproduct A ⊔ B is the disjoint union — these are very different objects in general. The collapse of products and coproducts into direct sums requires the special algebraic enrichment of an additive category.

Question 5

Why does the duality principle 'halve the work' in category theory? Illustrate with an example of a theorem and its automatic dual.

Model answer: Every theorem proved about a category C holds dually in C^op by reversing all arrows. Since C^op is itself a legitimate category, the dual statement is a genuine theorem — and when translated back into C using dual terminology, it describes a new construction in C for free. Example: the theorem 'a terminal object is unique up to unique isomorphism' dualizes immediately to 'an initial object is unique up to unique isomorphism' — no separate proof needed. Similarly, the universal property of limits (cones converging to a limit object) dualizes to the universal property of colimits (cocones emanating from a colimit object), giving the entire theory of colimits as a corollary of the theory of limits.

The savings compound as theory deepens. Monomorphisms dualize to epimorphisms, kernels to cokernels, projective objects to injective objects, direct products to direct sums. In homological algebra, the duality between injective and projective resolutions — fundamental to computing Ext and Tor — is entirely a consequence of this principle. Without duality, every construction would require an independent development.