#15 - A standard problem for computer security is that

[Administrator]

Complete Question Explanation

Flaw in the Reasoning. The correct answer choice is (D)

The stimulus observes that there are currently difficulties with computer security, and concludes that a voice-recognition system would be a way of giving access to authorized users without giving access to others, because in trials the such systems never incorrectly accept someone.

The reasoning is flawed because the conclusion implies that the system is a good way of granting access, but the conclusion considers only evidence about denying access. The system could be very good at denying access at the expense of efficient use by authorized users.

You may have also realized that it might be possible to circumvent the system with sound recordings. In that sense, this stimulus presents very much the same situation as did the six-pack wrapping argument (wildlife suffocation) that you no doubt remember. There are many technical critiques you can make relevant to the particular situation, but the inherent flaw is what you should probably focus on. Computer security is a two-part problem (at least), and the parts to consider are access and denial. The stimulus concerns information about only denial, and makes a conclusion about the whole problem of computer security, and that is a standard flaw. Since you are asked to identify the flaw, you should focus on that.

Answer choice (A): The fault in the argument was one of incomplete evidence, not one of an improper analogy. Possibly there are serious differences between voice-recognition and keyboard entered passwords; however, the argument made no comparison-- it drew a contrast in one area. This choice is wrong.

Answer choice (B): It is unclear what kind of data the argument utilizes, so you should not assume that the argument has over-generalized. This choice is wrong.

Answer choice (C): It is not a failure to stay on topic, so the fact that this argument does not consider the other applications of voice recognition is irrelevant, and this choice is incorrect.

Answer choice (D): This is the correct answer choice. There are two parts to the problem of computer security: access and denial. The argument fails to consider the problem of access, even though the argument draws a conclusion about access, and this choice correctly points out that the argument ignores the possibility that the program denies access to authorized users.

Answer choice (E): You may have wanted to select this choice, because you have learned that a qualified conclusion is a weakness. However, the truth is that qualification is not inherently flawed, and done correctly is not a weakness and is actually a means of improving an argument, because it is a means of taking more factors into account. An argument cannot be flawed merely because it is "heavily qualified."

[UnimelbLsat]

So the conclusion was that there will be a way of giving access to those ppl who are entitled to access and to no one else.
I don't see how the system 'occasionally' denies access to ppl who are entitled undermines that conclusion.
Doesn't the conclusion still hold? since it still provides a way for entitled people to access (just unsuccessful occasionally), and never let unauthorized people to access.

The real answer for me would be answer D without the word "sometimes".

[Adam Tyson]

Let's take a look at that conclusion, Unimelb, and see what's going on:

Clearly, if this result can be repeated in an operational setting, then there will be a way of giving access to those people who are entitled to access and to no one else.

There are two parts to it: 1) the new system will give access to those people who are entitled and 2) it will not give access to those who are not entitled.

We already established in the premises that #2 appears to have been accomplished. Great, you cannot get into my computer. The problem is, we have no evidence in the stimulus that #1 has been accomplished. None. So maybe I can't get into my computer either?

As to "sometimes" in the correct answer, it may help to remember that "some" includes the concept of "all". If all people cannot access their computers, then it is true that some people cannot access their computers. But the problem doesn't have to be as bad as "all" in this case. If even one person can't get into their own computer because the voice recognition system fails, then #1 in our conclusion has not been fully accomplished. That wouldn't be as bad as nobody getting in, but it's still a problem for the proposed system, right? Our author gave zero consideration to that side of the analysis, so we have insufficient evidence for the conclusion.

I hope that helps!

[lathlee]

Is this one of Use of Evidence error? or uncertain term usage flaw?

[Adam Tyson]

Definitely not uncertain use of a term, as that would involve some term or phrase changing in meaning over the course of the argument. I would throw it in the general "use of evidence" category, if I cared to label it, but as we have mentioned to you before, you should not be so focused on labeling the flaw. Understanding the flaw is the goal, and the labels are just there to help. As soon as you start worrying about what to call it, instead of just trying to understand it, you are focusing on the wrong thing, wasting time and effort. No more labels, lathlee!

[yournoona]

Hello.
I have a problem in identifying the answer between B and D. I understand why (D) is correct but i do not understand why (B) as an option is wrong.
For me, the reasoning went on like:

Option (B): As the stimulus itself mentions they took small amount of data into consideration. It clearly shows that maybe enough testing wasn't done to fully establish whether or not the computer system will provide incorrectly provide access to someone else.

Regards
Yournoona

[Adam Tyson]

You're right that so far the testing hasn't been very widespread, yournoona, but look at the conclusion again. The author is saying IF we can repeat this result in an operational setting - in other words, IF it works on a larger scale. It's a conditional conclusion, not an absolute one. The author isn't saying "it worked in this small trial, so it's going to work for everyone." She's saying "if we can repeat the results of the small trial on a large scale, then we know it will work." That's why answer B isn't the problem. The author is NOT basing the conclusion on a small amount of data, but is basing it on a conditional claim about possibly having much more data.

[grunerlokka]

I understand why D is the correct answer, but I had trouble eliminating Answer A. Answer A does seem to capture a common kind of flaw in the LSAT, one that does occur to an extent in this stimulus: a flawed analogy. In the initial explanation above for why A is wrong, the explanation is 'the argument made no comparison-- it drew a contrast in one area'. What is the distinction between comparison and contrast? Does this impact the extent to which something is an analogy?

[SammyWu11201]

Hey! For your explanation of AC E, you said that it's not inherently a flaw to heavily qualify a conclusion if it is done properly. What would a qualified flaw in the conclusion look like, then?

[Dave Killoran]

Hi Sammy,

It would look the same--limiting the circumstance described in the conclusion--but doing so in a way that's not applicable to the facts given. That would typically be by adding some new limitation or restriction that wasn't mentioned in the remainder of the argument. The extra qualifications would then operate as unwarranted assumptions.

The issue above isn't in the qualification ("if this result can be repeated in an operational setting") of the conclusion, but that they overlooked something about how the login acceptance was being done.

All that aside, as stated in the answer choice, this isn't often an issue and I don't recall an answer that was correct that used this language.

Thanks!