Applicative Laws for Maybe Type

Applicative functors come with a set of laws that apply for all Applicative instances. These laws are as follows:

For more information about these laws, check out this Haskell wiki post.

Maybe is an Applicative functor, and this post will verify that the applicative laws hold for the Maybe instance.

Before we begin, let’s review the definition for Applicative type-class:

class Functor f => Applicative f where
    pure :: a -> f a
    (<*>) :: f (a->b) -> f a -> f b

Maybe is made an instance of the Applicative class as:

instance Applicative Maybe where
    pure x                   = Just x
    (<*>) Nothing  _         = Nothing
    (<*>) (Just f) someMaybe = fmap f someMaybe

Note that the last line above can be re-written as:

    (<*>) (Just f) Nothing  = Nothing
    (<*>) (Just f) (Just x) = Just (f x)

Now, let’s test the Applicative laws for the Maybe type one by one:

Identity

We need to prove that the following holds for the Maybe type:

pure id <*> v = v

Let’s start from the left side of the equation:

pure id <*> v

-- applying definition of pure for Maybe
(Just id) <*> v
-- v could be either Nothing, or (Just x)
case v of
    Nothing  -> Nothing
    (Just x) -> Just (id x)
-- applying id to x is simply x
case v of
    Nothing  -> Nothing
    (Just x) -> Just x

v  -- the right side of equation

Homomorphism

The law’s statement is:

pure f <*> pure x = pure (f x)

To prove this for Maybe type, let’s begin from the left side of the equation:

pure f <*> pure x

-- applying definition of pure
(Just f) <*> (Just x)
-- applying defintion of <*>
Just (f x)
-- un-applying the definition of pure
pure (f x) -- the right side

Interchange

The law’s statement is:

u <*> pure y = pure ($y) <*> u

Starting from the left side:

u <*> pure y

-- applying definition of pure
u <*> (Just y)

-- u can be either Nothing, or (Just f). Let's evaluate both
case u of
    Nothing  -> Nothing
    (Just f) -> Just (f y)

-- re-writing (f y) using function application operator
case u of
    Nothing  -> Nothing
    (Just f) -> Just (f $ y)

-- Using the section representation of the $ infix operator
case u of
    Nothing  -> Nothing
    (Just f) -> Just (($y) f)

-- Un-applying the definition of <*>
case u of
    Nothing  -> Nothing
    (Just f) -> Just ($y) <*> (Just f)

-- Replacing the Nothing clause with an equivalent (via
-- the defintion) of <*> for Maybe
case u of
    Nothing  -> Just ($y) <*> Nothing
    (Just f) -> Just ($y) <*> (Just f)

-- Un-applying the definition of pure and u
pure ($y) <*> u -- Equal to the right side

Composition

The law’s statement is:

pure (.) <*> u <*> v <*> w = u <*> (v <*> w)

Starting from the left side:

pure (.) <*> u <*> v <*> w

-- applying definition of pure
(Just (.)) <*> u <*> v <*> w

-- u, v & w can all be either Nothing or (Just something).
-- From the definition of <*>, we know that if either
-- argument to <*> is Nothing, the entire expression reduces
-- to Nothing.
-- Therefore, I am only going to focus on the Just cases
(Just (.)) <*> (Just f) <*> (Just g) <*> (Just x)

-- Appying <*> to the first two terms
(Just (.) f) <*> (Just g) <*> (Just x)

-- Appying <*> to the first two terms again
(Just (.) f g) <*> (Just x)

-- Applying the function composition
(Just (f.g) ) <*> (Just x)

-- Applying <*> one more time
(Just (f.g) x)

-- via the definition of the function composition
-- operator
(Just f (g x))

-- Applying Homomorphism law
(Just f) <*> Just (g x)

-- Applying Homomorphism law to the 2nd term
(Just f) <*> ( (Just g) <*> (Just x) )

-- Un-applying the definitions of u, v, w
u <*> (v <*> w)    -- Equal to the right side