# What Is Recursion? How Does It Work?

**Recursion** is a method by which a problem gets solved through iteration.

In other words, a recursive function is a function that repeatedly invokes itself infinitely (or until something stops it).

## Important Stuff to Know About Recursive Function

Keep these two essential pieces of info in mind whenever you choose to use recursive functions.

### Info 1: Recursion is not an IIFE

A recursive function is different from an *Immediately Invoking function Expression (IIFE)*.

An IIFE automatically invokes itself once.

However, a recursive function automatically invokes itself repeatedly for an unlimited amount of time or until something stops its re-invocation.

### Info 2: A recursive function needs a base case

The code written to discontinue the re-invocation of a recursive function is called a *base case*.

It is always important to define a base case when creating a recursive function—so that the function will not run endlessly, thereby crashing the browser.

## Example of a Recursive Function

Below is a JavaScript code that returns a concatenation of all the values returned through the `countDown()`

function’s recursive invocation.

// Create a recursive function: function countDown(num) { // Define the base case of this recursive function: if (num < 0) { return "Recursion Stopped!"; } // Define the recursive case: return num + ", " + countDown(num - 1); } // Invoke the countDown() recursive function: countDown(2); // The invocation above will return: "2, 1, 0, Recursion Stopped!"

**Note:** In the recursive algorithm above, the `countDown(num - 1)`

code makes the whole function a recursion because it is the code that makes `countDown()`

recall itself repeatedly.

## Events Behind the Scenes

When we invoked the `countDown`

function and passed in `2`

(that is, `countDown(2)`

), the algorithm started running as follows:

### Step 1: Check if `2`

is less than `0`

The computer checked if the value `2`

—that we passed to the `num`

parameter of the `countDown`

function—is less than `0`

.

Since `2`

is not less than `0`

, the computer didn’t execute the `if`

statement’s code. Instead, it skipped to the next code after the `if`

statement—which is the *recursion code*.

### Step 2: Execute the `return`

statement

After skipping the `if`

statement, the computer executed the `return num + " " + countDown(num - 1)`

code—but substituted the `num`

parameter with the parameter’s value (that is, `2`

) like so:

return num + ", " + countDown(num - 1); return 2 + ", " + countDown(2 - 1); return 2 + ", " + countDown(1);

### Step 3: Execute only the recursive statement

In step 2’s code above, notice that the `return`

command can’t return any value because the `return`

statement includes a recursive code (`countDown(1)`

) recalling the `countDown`

function.

Therefore, while retaining the other parts of the `return`

statement (that is, `2 + ", " +`

), the computer will execute *only* the recursion code (`countDown(1)`

).

In other words, the `countDown(1)`

code will automatically invoke the `countDown`

function while passing in `1`

. Then, the algorithm will start running again by checking if `1`

is less than `0`

.

Since `1`

is not less than `0`

, the computer skipped to the recursion code like so:

return 2 + ", " + num + ", " + countDown(num - 1); return 2 + ", " + 1 + ", " + countDown(1 - 1); return 2 + ", " + 1 + ", " + countDown(0);

### Step 4: Invoke only the recursion code

Again, notice that the `return`

command (in step 3) cannot return any value because the `return`

statement includes a recursion code (`countDown(0)`

) that recalls the `countDown`

function.

Therefore, while retaining the other parts of the `return`

statement (that is, `2 + ", " + 1 + ", " +`

), the computer will execute *only* the recursion code (`countDown(0)`

). So, the `countDown(0)`

code will automatically invoke the `countDown`

function while passing in `0`

.

Then, the function will start running again by checking if `0`

is less than `0`

.

Since `0`

is not less than `0`

, the computer skipped to the recursion code like so:

return 2 + ", " + 1 + ", " + num + ", " + countDown(num - 1); return 2 + ", " + 1 + ", " + 0 + ", " + countDown(0 - 1); return 2 + ", " + 1 + ", " + 0 + ", " + countDown(-1);

### Step 5: Execute only the recursion code

Yet again, the `return`

command (in step 4) can’t return any value because the `return`

statement includes a recursion code (`countDown(-1)`

) recalling the `countDown`

function.

Therefore, while retaining the other parts of the `return`

statement (that is, `2 + ", " + 1 + ", " + 0 + ", " +`

), the computer will execute *only* the recursion code (`countDown(-1)`

). So, the `countDown(-1)`

code will automatically invoke the `countDown`

function while passing in `-1`

.

Then, the function will start running again by checking if `-1`

is less than `0`

.

At this point, `-1`

is less than `0`

. Therefore, the computer will execute the code of the `if`

statement by returning the value `“Recursion Stopped!”`

like so:

return 2 + ", " + 1 + ", " + 0 + ", " + "Recursion Stopped!";

At last, the `return`

statement now has values it can validly concatenate and return. Therefore, the returned value from `countDown`

will be:

"2, 1, 0, Recursion Stopped!"

## Wrapping It Up

In this article, we learned that a recursive function is a function that repeatedly recalls itself until something stops the recall.

Thanks for reading!