Why Check Before Applying a Theorem?
A calculus theorem is not a formula that applies whenever its conclusion seems plausible. It is a conditional statement: if specified hypotheses hold, then a stated conclusion follows. In an AP free-response solution, identifying the right theorem is only part of the justification. You must also show that the function and interval satisfy the theorem’s requirements.
This matters because the conditions are often precise. Continuity may be required at the endpoints as well as between them, while differentiability may be required only in the open interval. A function can meet one condition and fail another. Endpoint signs can also be part of the argument rather than a detail to assume.
A Theorem-Use Checklist
Start by writing down the theorem’s conditions in the form relevant to the problem. For the main existence and slope theorems used in calculus, the required checks include the following.
| Theorem | Conditions to verify |
|---|---|
| Intermediate Value Theorem | \(f\) is continuous on the entire closed interval \([a,b]\). For a target value \(L\), \(L\) lies between \(f(a)\) and \(f(b)\). |
| Mean Value Theorem | \(a<b\); \(f\) is continuous on \([a,b]\) and differentiable on \((a,b)\). |
| Rolle’s Theorem | \(a<b\); \(f\) is continuous on \([a,b]\), differentiable on \((a,b)\), and \(f(a)=f(b)\). |
| Extreme Value Theorem | \(f\) is continuous on the closed, bounded interval \([a,b]\). |
These statements are not interchangeable. For example, the Mean Value Theorem requires differentiability in the interior but does not require a derivative at either endpoint. The Intermediate Value Theorem requires continuity at both endpoints too. Always use the conditions of the particular theorem you intend to apply.
The function’s domain is part of the check. To be continuous on \([a,b]\), a function must be defined throughout that interval and continuous there, with the appropriate one-sided continuity at the endpoints. To be differentiable on \((a,b)\), it must be defined and differentiable at every interior point. An expression that simplifies algebraically may still have a point excluded from its original domain.
What “Verify the Hypotheses” Means
A strong justification gives evidence for each condition, not just a list of theorem vocabulary. For a polynomial, you can state that it is defined and continuous for all real \(x\), and differentiable for all real \(x\). For a function involving a denominator, logarithm, or radical, check where the expression is defined before making those claims. Then check that the entire interval needed by the theorem lies in the domain.
The interval notation matters as well. The closed interval \([a,b]\) includes both endpoints, whereas \((a,b)\) does not. A statement about continuity on \([a,b]\) includes the endpoints; a statement about differentiability on \((a,b)\) concerns only points strictly between them. If \(a=b\), the usual interval requirements for the Mean Value Theorem are not met, which is why the theorem assumes \(a<b\).
Some theorem applications also require a sign condition. For a zero-existence argument using the Intermediate Value Theorem, one checks that the endpoint values are on opposite sides of zero, or that an endpoint value is already zero. For a derivative-sign argument, the relevant derivative inequality must hold throughout the interval specified by the result. A value checked at one point does not establish a sign condition on an entire interval.
Why the Checklist Is Logically Necessary
The logic behind checking conditions is short, but important. Suppose a theorem says that for every function and interval, if a collection of hypotheses \(H\) holds, then a conclusion \(C\) follows. The theorem does not say that \(C\) follows whenever the situation resembles the theorem’s conclusion. It says \(H\) implies \(C\).
Here \(x\) represents the particular function-and-interval setup. To use the theorem for one specific setup \(x_0\), first substitute \(x_0\) into the universal statement. This gives \(H(x_0)\Rightarrow C(x_0)\). If you verify \(H(x_0)\), then the implication gives \(C(x_0)\). That is why a solution must establish the hypotheses before stating the conclusion.
If a hypothesis is not verified, this reasoning cannot be completed. That does not prove the conclusion false. It means only that this theorem does not justify the conclusion from the information established so far. The conclusion might still be true for another reason.
Worked Example: A Missing Domain Point Blocks an IVT Argument
Worked Example: Checking Domain, Continuity, and Endpoint Signs
Consider \(f(x)=\frac{x^2-1}{x-1}\), with its original domain, on \([0,2]\). A student claims that \(f(x)=2\) for some \(x\) in the interval because the endpoint values are on opposite sides of 2. Check whether the Intermediate Value Theorem supports the claim.
First check the domain. The denominator is zero when \(x-1=0\), so \(x=1\) is excluded. Since \(1\in[0,2]\), the function is not defined throughout the closed interval and cannot be continuous on all of \([0,2]\).
Now calculate the endpoint values:
Thus \(1<2<3\), so the endpoint values do straddle the target. But the required continuity hypothesis fails at \(x=1\). The theorem cannot be invoked.
Indeed, for \(x\ne1\), the equation can be simplified as follows:
The only algebraic candidate is the excluded point \(x=1\), so there is no solution in the function’s domain. The endpoint sign information alone was not enough; continuity on the full interval was essential.
Check: Substitution of the only candidate into the original expression gives a zero denominator, confirming that \(x=1\) is not an allowed solution.
Worked Example: Verifying the Mean Value Theorem Conditions
Worked Example: A Square-Root Function on a Closed Interval
For \(f(x)=\sqrt{x}\) on \([0,4]\), check whether the Mean Value Theorem applies and verify the resulting interior point.
The domain of \(f\) is \(x\ge0\), so the function is defined throughout \([0,4]\). The square-root function is continuous on this closed interval. Its derivative is \(f'(x)=\frac{1}{2\sqrt{x}}\), which exists for every \(x\in(0,4)\). The endpoints satisfy \(0<4\). Therefore all the Mean Value Theorem conditions hold.
The average rate of change across the interval is
The theorem guarantees an interior value \(c\) with \(f'(c)=\frac12\). Solve that equation:
Since \(0<1<4\), the value is in the required open interval. Checking the slope directly, \(f'(1)=\frac{1}{2\sqrt{1}}=\frac12\), matching the calculated average rate of change.
Worked Example: Continuity Alone Is Not Enough
Worked Example: A Corner Point on the Interval
Consider \(f(x)=|x|\) on \([-1,1]\). Can the Mean Value Theorem be used to guarantee a point where the derivative equals the secant slope?
The function is defined and continuous on the whole closed interval. The secant slope is
However, \(f\) is not differentiable at \(x=0\), which lies in the open interval \((-1,1)\). In fact, \(f'(x)=-1\) for \(x<0\), and \(f'(x)=1\) for \(x>0\); the derivative at \(0\) does not exist. Thus the differentiability condition fails.
There is no interior point where the derivative equals the secant slope \(0\): where the derivative exists, it is either \(-1\) or \(1\). The theorem cannot be used to guarantee such a point because one of its hypotheses is false.
Check: The endpoint values give the stated slope, and neither possible derivative value, \(-1\) or \(1\), equals \(0\).
Worked Example: Checking the Extreme Value Theorem
Worked Example: A Polynomial on a Closed, Bounded Interval
Check whether the Extreme Value Theorem applies to \(g(x)=x^2-2x+3\) on \([0,3]\).
The expression is a polynomial, so it is defined and continuous for every real \(x\), including every point of \([0,3]\). The interval is closed because it includes 0 and 3, and bounded because \(0\le x\le3\). Therefore the Extreme Value Theorem applies: \(g\) attains both an absolute minimum and an absolute maximum somewhere on \([0,3]\).
This check justifies the existence conclusion; it does not by itself identify where the extrema occur or what their values are. Those require further analysis.
Check: Both endpoint values are defined: \(g(0)=0-0+3=3\), and \(g(3)=9-6+3=6\). The interval and continuity conditions needed for the theorem have been verified.
Common Mistakes and Corrections
A frequent error is to check only the part of a hypothesis that seems most relevant. For instance, opposite endpoint signs do not establish the Intermediate Value Theorem’s continuity requirement. Similarly, continuity on a closed interval does not establish the differentiability required by the Mean Value Theorem.
Correction: First verify that the function is defined and continuous on the entire closed interval. Endpoint signs are not a substitute for continuity.
Another mistake is to conclude that a theorem’s result is false whenever a hypothesis fails. That conclusion is too strong. A failed condition means the theorem does not provide the justification; it does not automatically settle whether the result happens to be true. The absolute-value example showed that no point had the desired derivative, but in other problems a conclusion may hold for a different reason.
In written work, a concise but complete check often follows this order:
State the specific result you intend to use.
Make clear which endpoints and interior points are involved, and check the function is defined where required.
Address continuity, differentiability, endpoint values, or sign requirements separately as appropriate.
Only after the hypotheses are established, apply the theorem to the particular function and interval.
Check Your Understanding
For each prompt, identify the hypothesis that must be checked before the stated theorem can be used.
- For the Intermediate Value Theorem on \([a,b]\), what must be true about the function at every point of the interval?
- For the Mean Value Theorem, where must continuity hold, and where must differentiability hold?
- A function is continuous on \([-2,2]\) but not differentiable at \(x=0\). Which hypothesis blocks using the Mean Value Theorem on that interval?
- Why is checking opposite endpoint signs insufficient by itself to justify an Intermediate Value Theorem argument?
- If a theorem’s hypothesis fails, what can you conclude about whether its conclusion is true?
