Connecting Differentiability and Continuity: Determining When Derivatives Do and Do Not Exist
Why this matters
Imagine you're planning an epic cross-country road trip from Boston to Seattle. You pull up two possible routes on your map. Both routes get you from Point A to Point B without any gaps—they are both continuous paths.
Route 1 is all smooth interstate highways. Your speed is consistent, the ride is comfortable, and you can always tell your exact direction. Route 2, however, is a mess. It has a section where the road abruptly becomes a 90-degree turn in downtown Chicago, another spot with a bizarrely sharp, pointy bridge in the Rockies, and a stretch in a canyon that momentarily goes straight up a wall. Even though this route is technically "connected," could you really drive it? At those specific weird points, could you even define your car's direction?
In calculus, this is the exact difference between continuity and differentiability. This lesson will show you why every "smooth road" (differentiable function) must be a "connected road" (continuous), but not every connected road is smooth enough to travel.
Concept overview
flowchart TD
A[Start: Function f(x) at point c] --> B{Is f(x) continuous at c?};
B -->|No| C[Not Differentiable];
B -->|Yes| D{Does the limit of the slopes from the left and right exist and are they equal?};
D -->|No<br>(Corner, Cusp, or Vertical Tangent)| C;
D -->|Yes| E[Differentiable];
Core explanation
Hey there, future calculus champs. Let's dive into one of the most fundamental ideas in this course: the connection between being continuous and being differentiable.
To start, let's quickly refresh. A function is continuous at a point if there are no interruptions—no holes, jumps, or vertical asymptotes. You can draw it without lifting your pencil. A function is differentiable at a point if you can find its instantaneous rate of change there—in other words, if you can draw a unique, non-vertical tangent line with a defined slope.
The Golden Rule: Differentiability Implies Continuity
Here is the most important takeaway, so let's put it right up front:
If a function is differentiable at a point x = c, then it must be continuous at x = c.
Think of it like a contract. To be "differentiable" is to be "smooth." To be "continuous" is to be "connected." The smoothness contract has a clause in the fine print that says you automatically have to be connected.
Why is this true? The definition of the derivative is based on a limit. For that limit to exist, the function has to behave nicely around the point in question. It can't have a hole or jump away. If the point f(c) isn't even there, or if the function approaches a different value from the left and right, how could we possibly agree on the slope of a single tangent line at that point? We can't. The very act of calculating a derivative forces the function to be continuous at that spot.
This also covers a simple but important point: if a point isn't even in the domain of a function f (like at a vertical asymptote), then it's impossible for it to be in the domain of its derivative f'. You can't find the slope of a tangent line on a part of the graph that doesn't exist.
The Plot Twist: Continuity Does NOT Imply Differentiability
Just because a function is continuous at a point does not mean it is differentiable there.
Your road can be connected, but that doesn't guarantee it's smooth. Let's look at the three classic ways a continuous function can fail to be differentiable. These are the "potholes" and "sharp turns" of the math world.
1. A Corner
The most famous example is the absolute value function, f(x) = |x|, at x = 0.
The graph is clearly continuous. You can draw that "V" shape without lifting your pencil. But what's the slope of the tangent line exactly at x = 0?
- If you approach from the left, the slope is always -1.
- If you approach from the right, the slope is always +1.
Since the slope from the left does not equal the slope from the right, there is no single, unique tangent line at the origin. The function is not differentiable at x = 0. Think of it as arriving at a sharp intersection in a city; your direction abruptly changes.
2. A Cusp
A cusp is like a corner's sharper, more aggressive cousin. Consider the function f(x) = x^(2/3).
Again, this graph is continuous everywhere. But look at x = 0. As you approach from the left, the tangent lines get steeper and steeper, approaching negative infinity. As you approach from the right, they also get steeper, approaching positive infinity. The slopes "disagree" in the most extreme way possible. There's no way to define a single slope at that infinitely pointy tip. The function is not differentiable at x = 0.
3. A Vertical Tangent
Our third case happens when the road momentarily goes vertical. Consider f(x) = ³√x (the cube root of x) at x = 0.
This function is continuous everywhere. But if you try to draw a tangent line at x = 0, you'll find it's a perfectly vertical line. What's the slope of a vertical line? It's undefined. Since the slope is undefined, the derivative does not exist at that point. The function is not differentiable at x = 0.
Summary of the Relationship
- Differentiable? Yes, it's definitely Continuous.
- Continuous? Maybe it's Differentiable, maybe not. You have to check for corners, cusps, and vertical tangents.
- Not Continuous? No way, it's definitely Not Differentiable.
Always check for continuity first. If a function is broken, you don't even need to worry about its smoothness. If it's connected, then you have to put on your detective hat and hunt for those three non-differentiable culprits.
Worked examples
Let's put this into practice. The key is to be systematic.
A Piecewise Function
Consider the function f(x) defined as:
f(x) = { x² + 1, if x < 1
{ 2x, if x ≥ 1
Is f(x) continuous and/or differentiable at x = 1?
Step 1: Check for Continuity at x = 1
For a function to be continuous at a point, the limit from the left, the limit from the right, and the function's value must all be equal.
- Left-hand limit
lim (x→1⁻) f(x) = lim (x→1⁻) (x² + 1) = 1² + 1 = 2 - Right-hand limit
lim (x→1⁺) f(x) = lim (x→1⁺) (2x) = 2(1) = 2 - Function value
f(1) = 2(1) = 2(we use the≥piece)
Since the left limit, right limit, and function value are all 2, yes, the function is continuous at x = 1.
Step 2: Check for Differentiability at x = 1
Now we check if the slopes match. We need to find the derivative of each piece and evaluate them at x = 1.
- Derivative from the leftThe derivative of
x² + 1is2x. Atx = 1, the slope is2(1) = 2. - Derivative from the rightThe derivative of
2xis2. Atx = 1, the slope is2.
Since the derivative from the left (2) equals the derivative from the right (2), yes, the function is differentiable at x = 1.
The Absolute Value in Disguise
Let g(x) = |x - 3|. Where is this function not differentiable?
Step 1: Understand the Function
This is the absolute value function |x| shifted 3 units to the right. We know that |x| has a sharp corner at x = 0. Therefore, we should expect |x - 3| to have a sharp corner at x = 3.
Step 2: Analyze using a Piecewise Definition
Let's rewrite g(x) as a piecewise function.
g(x) = { -(x - 3), if x - 3 < 0 (i.e., x < 3)
{ (x - 3), if x - 3 ≥ 0 (i.e., x ≥ 3)
Simplifying this gives:
g(x) = { -x + 3, if x < 3
{ x - 3, if x ≥ 3
Step 3: Check Differentiability at x = 3
First, you can verify it's continuous at x=3 (both pieces equal 0). Now let's check the slopes.
- Derivative from the leftThe derivative of
-x + 3is -1. - Derivative from the rightThe derivative of
x - 3is +1.
This is where students often make a mistake. They see the function is continuous and stop. But we must compare the slopes. Since -1 ≠ +1, the slopes do not match at x = 3.
Try it yourself
Ready to try a couple on your own? Don't just guess—walk through the steps we discussed.
Problem 1:
Let h(x) be the piecewise function:
h(x) = { 3x - 2, if x ≤ 2
{ x² , if x > 2
Is h(x) continuous at x = 2? Is it differentiable at x = 2?
Hint: First, plug x=2 into both pieces. Do they give you the same value? Then, find the derivative of each piece and plug in x=2. Are the slopes the same?
Problem 2:
Find all x-values where the function k(x) = (x + 5)^(2/3) is not differentiable.
Hint: This function is continuous everywhere. The problem must be a corner, cusp, or vertical tangent. Find the derivative k'(x) using the power rule and chain rule. Are there any x-values that make the derivative's denominator equal to zero? That's your trouble spot.
In simple terms, this topic is about why a function must be "smooth" to have a derivative at a point, and why just being "connected" (continuous) isn't always enough.
- FUN-2.A: Explain the relationship between differentiability and continuity.
- FUN-2.A.1
- If a function is differentiable at a point, then it is continuous at that point. In particular, if a point is not in the domain of f, then it is not in the domain of f'.
- FUN-2.A.2
- A continuous function may fail to be differentiable at a point in its domain.
flowchart TD
A[Start: Function f(x) at point c] --> B{Is f(x) continuous at c?};
B -->|No| C[Not Differentiable];
B -->|Yes| D{Does the limit of the slopes from the left and right exist and are they equal?};
D -->|No<br>(Corner, Cusp, or Vertical Tangent)| C;
D -->|Yes| E[Differentiable];
Read what Saavi narrates
(Sound of a car driving smoothly, then a gentle, warm musical intro)
Hey everyone, it’s Saavi from Shrutam. Let's talk about something that sounds complicated but is actually pretty intuitive.
Imagine you're on a road trip. A continuous road is one that's all connected... no sudden gaps or bridges out. You can get from point A to point B. A differentiable road is a *smooth* road. No sharp 90-degree turns, no infinitely pointy ramps. You can always tell your exact direction.
The big idea today is simple: every smooth road has to be connected. But... not every connected road is smooth. In calculus terms, differentiability implies continuity, but continuity does not guarantee differentiability. A function can be connected, but still have a sharp corner that makes it impossible to define a single slope at that one point.
Let's look at a classic example. Imagine the function g of x equals the absolute value of x minus 3. We want to know where this function isn't differentiable.
First, picture the graph. It’s that V-shape we know from the absolute value function, but shifted over so the point of the V is at x equals 3. The graph is definitely continuous... you can draw the whole thing without lifting your pencil.
But what about being differentiable, or smooth? Let's check the slope at that point, x equals 3. If we approach the point from the left side, the line has a slope of negative one. But if we approach from the right side, the line has a slope of positive one.
This is the key. Since the slope from the left, negative one, does not equal the slope from the right, positive one, there's no single, defined slope right *at* x equals 3. It's a corner. So, the function is not differentiable at x equals 3.
And that brings us to a really common mistake: thinking that if a function is continuous, it must be differentiable. Please, please remember that this is a one-way street! Continuity is the first hurdle. If a function passes that, you still have to check for those sharp corners, cusps, or vertical tangents.
This idea is a foundation for so much of what's to come. Take your time with it, practice identifying those tricky spots, and you'll build a really strong understanding. You've got this.
(Musical outro)
This is the biggest misconception. Continuity is necessary for differentiability, but it's not sufficient. The function `f(x) = |x|` is the classic counterexample; it's continuous at `x=0` but not differentiable there.
After confirming continuity, always explicitly check for the three failure points: corners, cusps, and vertical tangents.
A derivative of zero means the tangent line is horizontal (like at the top of a parabola). A non-existent derivative means you can't define a single tangent slope at all (like at a corner). These are very different situations.
Remember that zero is a number. "Undefined" is not a number. If the slope is 0, it's differentiable. If it's a corner, it's not.
That only confirms continuity. To check for differentiability, you must also check if the *derivatives* (slopes) of the pieces are equal at that boundary point.
Follow the two-step process from the worked example. Step 1: Check continuity. Step 2: Take the derivative of each piece and check if they are equal at the boundary.
Corners are easy to spot, but vertical tangents can be subtle. A function like `f(x) = (x-4)^(1/3)` looks smooth, but its derivative `f'(x) = 1/(3(x-4)^(2/3))` is undefined at `x=4`, indicating a vertical tangent.
When a problem asks where a function is *not* differentiable, always calculate the derivative and look for x-values that would cause division by zero. This is the algebraic signal for a vertical tangent or cusp.