Two Theorems, Two Different Guarantees
On a closed interval, a continuous function has an absolute maximum and an absolute minimum. Under a different set of conditions, the Mean Value Theorem guarantees an interior point where the instantaneous rate of change equals the average rate of change across the interval. Both theorems are about what happens on an interval, but their conclusions answer different questions.
The distinction matters because an attained extreme value is a statement about the function's values, while a Mean Value Theorem conclusion is a statement about its derivative. The point where an extreme value is attained might be an endpoint, where the derivative condition need not apply. Conversely, the point supplied by the Mean Value Theorem need not be an extremum.
The Two Theorems
Attaining a value means that the function actually takes that value at a point in its domain; merely approaching a bound is not enough. The theorem guarantees existence, not a formula for locating the points, and it does not say that either extremum occurs in the interior.
The Mean Value Theorem guarantees an interior point with a particular derivative value. It does not guarantee that this point is unique, nor that it is a maximum or minimum. Notice also that it requires differentiability on the open interval, while the Extreme Value Theorem does not.
Compare the Conclusions Carefully
| Question | Extreme Value Theorem | Mean Value Theorem |
|---|---|---|
| What does it guarantee? | An attained absolute maximum and minimum | An interior point whose derivative equals the secant slope |
| Where is the guaranteed point? | Somewhere in \([a,b]\), possibly at an endpoint | At least one point in \((a,b)\) |
| What hypotheses are required? | Continuity on the closed, bounded interval | Continuity on the closed interval and differentiability in its interior |
If a function meets the hypotheses of both theorems, both conclusions hold. But they generally identify different points for different reasons. One concerns the largest and smallest function values; the other concerns a slope equal to the average change across the interval.
Worked Example: A Minimum Inside and a Maximum at an Endpoint
Worked Example: A Minimum Inside and a Maximum at an Endpoint
Let \(f(x)=x^2\) on \([-1,2]\). Find its absolute extrema and a point guaranteed by the Mean Value Theorem.
First, \(f\) is a polynomial, so it is continuous on \([-1,2]\) and differentiable on \((-1,2)\). The Extreme Value Theorem guarantees an absolute minimum and maximum, and the Mean Value Theorem applies as well.
To find the extreme values, compare the endpoint values with the interior point where the derivative is zero. Since \(f'(x)=2x\), the interior critical point is \(x=0\). The values are
Therefore, the absolute minimum is \(0\), attained at \(x=0\), and the absolute maximum is \(4\), attained at \(x=2\).
For the Mean Value Theorem, the secant slope is
We solve \(f'(c)=1\):
This point is interior because \(-1<\frac12<2\). Checking, \(f'(\frac12)=2(\frac12)=1\), exactly the secant slope. The minimum point \(0\), maximum point \(2\), and Mean Value Theorem point \(\frac12\) serve different roles.
Why Interior Extrema and Mean Value Points Differ
There is a useful connection between extrema and derivatives, but it applies only under specific conditions: if a differentiable function has a local maximum or minimum at an interior point, its derivative there must be zero. This is not the general conclusion of the Mean Value Theorem.
Here is the reason. Suppose first that \(x_0\) is a local maximum. For \(h\) sufficiently close to zero, \(f(x_0+h)\leq f(x_0)\). If \(h>0\), the difference quotient is nonpositive; if \(h<0\), it is nonnegative, because the denominator changes sign. Since the derivative exists, the two one-sided limits of that quotient must be the same number. The only number that is both nonpositive and nonnegative is zero. Hence \(f'(x_0)=0\). For a local minimum, the inequalities reverse, and the same reasoning again forces the derivative to be zero.
This result helps explain why an interior differentiable extremum may have a horizontal tangent. But an absolute extremum on a closed interval can occur at an endpoint, which is not an interior point. Also, the Mean Value Theorem asks for \(f'(c)\) to equal the secant slope, which need not be zero.
Worked Example: Endpoint Extrema and a Matching Derivative
Worked Example: Endpoint Extrema and a Matching Derivative
Let \(f(x)=x\) on \([0,1]\). Compare its absolute extrema with its Mean Value Theorem conclusion.
The function is continuous on \([0,1]\) and differentiable on \((0,1)\), so both theorems apply. Because \(f(x)=x\) increases from \(0\) to \(1\), its absolute minimum is \(0\) at \(x=0\), and its absolute maximum is \(1\) at \(x=1\). Both extrema occur at endpoints.
The secant slope is
Since \(f'(x)=1\) for every \(x\), every \(c\in(0,1)\) satisfies the Mean Value Theorem equation. For instance, \(c=\frac12\) works because \(f'(\frac12)=1\). The endpoints are where the extremes occur; the interior points match the average slope. The same function illustrates both guarantees without making them the same guarantee.
Worked Example: Continuity Alone Does Not Ensure the Mean Value Conclusion
Worked Example: Continuity Alone Does Not Ensure the Mean Value Conclusion
Let \(f(x)=|x|\) on \([-1,1]\). Determine its absolute extrema and examine whether the Mean Value Theorem applies.
The absolute value function is continuous on the closed interval. Its values satisfy \(0\leq |x|\leq1\), with \(f(0)=0\) and \(f(-1)=f(1)=1\). Thus the absolute minimum is \(0\), attained at \(x=0\), and the absolute maximum is \(1\), attained at both endpoints. This is consistent with the Extreme Value Theorem.
However, \(f\) is not differentiable at \(x=0\), so the Mean Value Theorem hypotheses are not satisfied. In fact, the endpoint secant slope is
For \(-1<x<0\), \(f'(x)=-1\); for \(0<x<1\), \(f'(x)=1\); and \(f'(0)\) does not exist. Thus there is no interior point where the derivative equals \(0\). This does not contradict the Mean Value Theorem, because its differentiability condition fails. It does show why the Extreme Value Theorem's continuity condition cannot be substituted for the Mean Value Theorem's full set of hypotheses.
A Common Misreading
A tempting but incorrect claim is: “The Mean Value Theorem gives a point where the function reaches an absolute extreme.” Consider \(f(x)=x\) on \([0,1]\). Its absolute maximum is at \(x=1\) and its absolute minimum is at \(x=0\), while \(f'(c)=1\) at every interior point. For example, \(c=\frac12\) satisfies the Mean Value Theorem conclusion, but \(f(\frac12)=\frac12\) is neither the absolute maximum nor the absolute minimum.
The correction is to keep the equations and their meanings separate. The Extreme Value Theorem identifies attained largest and smallest values. The Mean Value Theorem gives an interior \(c\) satisfying \(f'(c)=(f(b)-f(a))/(b-a)\). Neither conclusion replaces the other.
When Each Guarantee Is Useful
The Extreme Value Theorem is useful when a problem asks whether the highest or lowest value is actually attained on a specified interval. Before using it, check that the function is continuous and that the interval is closed and bounded. The theorem guarantees existence, but further work is needed to identify the points or values.
The Mean Value Theorem is useful when a problem asks about how a function's change over an interval relates to an instantaneous rate somewhere inside it. Its hypotheses include differentiability at every interior point. The conclusion concerns a derivative and a secant slope, not necessarily an extreme value.
For either theorem, the hypotheses are part of the claim, not optional formalities. When a hypothesis fails, that theorem supplies no guarantee. Another argument might still establish a particular conclusion, but it cannot be inferred from the theorem alone.
Check Your Understanding
For each question, identify which theorem applies and what its conclusion does—and does not—guarantee.
- State the hypotheses and conclusion of the Extreme Value Theorem on \([a,b]\).
- State the hypotheses and conclusion of the Mean Value Theorem on \([a,b]\).
- For \(f(x)=x^2\) on \([-1,2]\), what are the absolute extrema, and which point satisfies the Mean Value Theorem equation?
- Why does an interior differentiable local extremum have derivative zero?
- For \(f(x)=|x|\) on \([-1,1]\), which theorem applies, and why does the failure of the other theorem's hypothesis matter?
