LSAT 120 - Section 4 - Question 14

Question Stem
This is a Flaw or Descriptive Weakening question. The key words in the question stem are "grounds for criticizing... reasoning."

Foundational Skills
Phenomenon-hypothesis
Contrapositive
False positive v. false negative

Stimulus
The question is hard because the correct answer (A) is stating something implied by what you might have anticipated while (B) brings up a novel though irrelevant consideration and (E) masquerades as an assumption that you might have spotted.

Someone played a practical joke on Franklin. He doesn't know who did it though he suspects Miller because Miller "has always been jealous of me." Okay, that's motive. But what evidence does he have? Just one piece: a handwritten note where the handwriting does not match Miller's.

So, what conclusion can we draw? Well, you and I are thinking, it's unclear. The fact (phenomenon) that the handwriting doesn't match Miller's could be explained by a number of hypotheses:

1. It was Miller and she disguised her handwriting.
2. It was Miller and an unwitting accomplice wrote the note.
3. It was Miller and she had a willing co-conspirator write the note.
4. It was not Miller.

Franklin jumps to hypothesis (4) as the explanation. He assumes that if the handwriting doesn't match Miller's, then Miller didn't do it. Hypotheses (1) - (3) reveal why that assumption is problematic. It could be the case that the handwriting doesn't match Miller's and (yet) it is still Miller who did it.

Now that we've identified the issue with this argument both using the framework of assumptions and alternative hypotheses, we now can think about how to "criticize Franklin's reasoning."

At a very abstract and general level, we could say something like "It (Franklin's reasoning) fails to consider alternative hypotheses." That would capture hypotheses (1) through (3). That could be a correct answer, if all the other answers are bad. But the correct answer could also be (and in fact turns out to be) more specific.

Answer Choice (A)
(A) says that Franklin's reasoning "fails to consider the possibility that there was more than one practical joker." On first blush, you might think that this is merely descriptively accurate yet doesn't get to the weakness in the reasoning. True, you think, Franklin did not consider that this could have been a conspiracy (a plot involving more than one person). But how many people involved is not the issue. The issue is the identity of those involved.

Ah, but the number of people involved is related to the identity of those involved. Why did Franklin write off Miller? Precisely because he didn't consider that Miller could have had an accomplice, that there could have been more than one practical joker.

Do you see how the test writers made (A) subtle? They could have said "It fails to consider the possibility that Miller had an accomplice." That would have been the blunt and obvious way to state the weakness in Franklin's reasoning. But they didn't. Instead they stated something implied by the blunt version of the hypothesis. If it's true that Miller had an accomplice, then it must be true that there was more than one practical joker.

Answer Choice (B)
(B) says that Franklin's reasoning "fails to indicate the degree to which handwriting samples should look alike in order to be considered of the same source." This is true. It's descriptively accurate. Franklin merely asserts "the handwriting is not hers" without providing any reason for us to believe that assertion, e.g. just how closely must the curve on an "r" match or what angles of "v" or "w" are considered close enough?

All true. But, notice that that assertion "the handwriting is not hers" is being used as a premise. And as a rule of thumb, premises get the benefit of the presumption of truth. If Franklin asserts it, then, unless we have reasons to doubt him, we accept it as true. (B) is asking us to question this premise.

If we were detectives and if this were an actual investigation or criminal trial, then, yeah, this assertion absolutely would come under attack: we'd get experts to explain their methodology and testify so we can ascertain whether to believe Franklin's assertion that "the handwriting is not hers." But, we're doing an LR question. We're just being asked to "criticize Franklin's reasoning" and the rule of thumb is that you do not attack premises. Plus, as we already saw in the stimulus analysis, even granting Franklin the truth of this premise still leaves his argument vulnerable because the issue isn't in this premise. The issue is in the support relationship between this premise and the conclusion.

Answer Choice (C)
(C) says that no explanation was provided for why Miller should be the prime suspect. This is false. It's descriptively inaccurate. Franklin does give an explanation. He says that Miller "has always been jealous" of him. Now, you might think that's a weak motive, but that doesn't change the fact that Franklin gave an explanation. You're just judging that explanation as insufficient. No explanation is different from an explanation that you don't believe.

Answer Choice (D)
(D) says that no explanation was provided for why only one piece of evidence was obtained. This is true, it's descriptively accurate. Franklin did not explain why there isn't more evidence. But so what. From the sole existing piece of evidence Franklin drew a conclusion. That reasoning is present and weak. Our job is to attack that reasoning. That's it. It's not to ask questions that would have been relevant had we been actually investigating this case. Had we been actual detectives actually trying to solve the case, then yeah, we wonder why there was just one piece of evidence.

If you chose (D), you might have been thinking that Franklin's argument is weak because there was only one piece of evidence provided. That is true. But that's not what (D) says. In order to capture your justified concern, (D) should have said something like "It draws a conclusion unsupported by the only piece of evidence available."

Answer Choice (E)
(E) says that Franklin's reasoning "takes for granted" which is just "assumes" that - and here comes the conditional - "if the handwriting on the note had been Miller's, then the identity of the joker would have been ascertained to be Miller:"

match → Miller

(E) is testing your conditional logic with a classic sufficiency-necessity confusion. It also helps if you're familiar with the distinction between false positives and false negatives.

We figured out that Franklin does "take for granted" that if the handwriting didn't match Miller's, then Miller didn't do it:

/match → /Miller

What's the contrapositive of that? If Miller did it, then the handwriting would have matched:

Miller → match

(E)'s match → Miller is just not the same as Miller → match. You don't want to confuse sufficiency for necessity because, well, they're different. In the context of using a test result to determine identity, here's why they are different.

Let's stipulate that the handwriting analysis test that Franklin used was one that contains high false positives but low false negatives.

A false positive is when the test says "match!" but it should not have. The positive result ("match!") was false. Like if a doctor pronounced a man "pregnant!" That's a false positive. Or if a DNA analysis said "match!" but it shouldn't have because there was an error. We'd be rightly suspicious of positive results from tests that contain high rates of false positive. "Sure, test, you say it matches, but you always say that. I don't believe you."

But don't confuse that for when the test gives you negative results. The test could be highly reliable for negative results even if it's unreliable for positive results.

What's a false negative? It's when a test says "no match!" but it should have matched. The negative result was false. Like if a doctor pronounced an obviously 8 month pregnant woman "Not pregnant! Just lay off the chips." That's a false negative. Or if a DNA analysis said "no match!" but it should have said "match!" instead.

Since we stipulated that our handwriting analysis test contains few false negatives, we don't have these concerns. That means whenever the test says "no match," we should believe it. Truly there is no match.

That means it's reasonable to hold the position that /match → /Miller while rejecting the position that match → Miller. The two positions are not the same. The first position requires a test with low false negatives. The second position requires a test with low false positives.