#13 - Columnist: Computer voice-recognition technology

[sodomojo]

Why is the answer (A) and not (D)?

I can't recall any other questions where you have a conditional conclusion with a premise that doesn't link to either of those conditional elements in the conclusion.

With the clock ticking, I went with (D) as it seemed to have in-scope elements that performed a "supporter" function by linking two rogue elements of the argument together.

I'm having a hard time seeing why (A) would be a necessary assumption.

[James Finch]

Hi Sodomojo,

The premise given in this stimulus is that voice-recognition technology cannot distinguish between homophones. Then we are given the conclusion as a conditional statement, that until voice-recognition tech can recognize and utilize grammatical and semantic relations among words, these programs won't accurately translate spoken words into text (Accurate translation Grammar & Semantics). The phrase "as a consequence" serves to link the original premise as a conditional statement to the sufficient condition of accurate translation of speech-to-text.

As an assumption question, there is a clear logical break between distinguishing homophones, which has been tied to accurate translations, and recognizing grammar and semantics. The phrase "as a consequence" serves to link the original premise as a conditional statement to the sufficient condition of accurate translation, so the conditional relationship we actually have is:

Accurate Translation Distinguishing Homophone Grammar & Semantics

with the logical break occurring between the homophones and the grammar and semantics conditions. We have an A B C conditional relationship, but B C doesn't exist, yet the conclusion still draws the inference that A C.

Answer choices (A) and (D) both link homophones and grammar and semantics, but in opposite ways. (A) makes grammar and semantics the necessary condition for the sufficient condition of distinguishing homophones, bridging the gap in the stimulus by linking B C, allowing us to then draw the conclusion's inference of A C.

(D) on the hand makes distinguishing homophones the necessary condition for recognizing grammar and semantics, essentially reading the stimulus as:

Accurate translation Recognize Grammar & Semantics Homophones

while tacking on an unnecessary B C conditional relationship. However, this isn't the reasoning in the stimulus, and not the assumption that is required for the logic to hold true, but a Mistaken Reversal of it. So (D) is an incorrect answer choice.

Hope this clears things up!

[Khodi7531]

I was between A and D and wasted too much time so I went with D.


I get the gap of the argument and what it was stating...but I have a question. the assumption is the disconnect between distinguishing homophones and recognizing grammar relations right? If so...I'm never comfortable on how you chose what goes where


If the Premise is A, and conclusion is B > C .... why are you doing B > A > C ? Why not A > B > C ?


And if something is necessary for an argument, isn't it "conclusion > assumption" ? So it would be B > C > A ... no?


I'm just confused on placement otherwise I got it down.

[harvoolio]

I am paddling in a boat with Khodi, as I am equally flummoxed.

I am going to make an analogy and please correct me where I make the mistake in choosing the analog of answer (D) instead of (A).

Stimulus: John Doe’s account with LSAC Credential Assembly Service currently does not have an official LSAT score. As a consequence, until John Doe’s account with LSAC Credential Assembly Service creates a school admission index, John Doe’s account with LSAC Credential Assembly Service will be unable to create complete a Law School Report.

Which one of the following is an assumption required by the above argument?

(A) In order for John Doe’s account with LSAC Credential Assembly Service to have an official LSAT score, it must be able to create a school admission index.
(D) Unless John Doe’s account with LSAC Credential Assembly Service has an official LSAT score, John Doe’s account with LSAC Credential Assembly Service cannot create a school admission index.

So, D would be right in this instance because LSAT scores are necessary for admissions indexes, but not the other way around; some schools might not even use an admissions index.

Now I can anticipate the response is that to be analogous the stimulus should read "John Doe’s account with LSAC Credential Assembly Service currently does not have an admissions index. As a consequence, until John Doe’s account with LSAC Credential Assembly Service has an official LSAT score, John Doe’s account with LSAC Credential Assembly Service will be unable to create complete a Law School Report."

But isn't that then a form of circular reasoning? Or is "as a consequence" truly a structural indicator (without exception), that indicates indirect causality (or conditionality?) between the necessary and sufficient condition in that sentence, such that the preceding sentence serves as a direct cause (condition?) of the sufficient condition in the sentence with "as a consequence".

Thanks. (Nice Confused Emoji)

[LSAT2018]

Premise: Not A
Conclusion: Not B → Not C
Assumption: Not A → Not B or Not A → Not C

Not Distinguish Homophones → Not Recognize Grammatical Relations → Not Accurately Translate
Accurately Translate → Recognize Grammatical Relations → Distinguish Homophones


Am I taking the approach for a Sufficient Assumption, not a Necessary Assumption ?

Why is answer (D) a Mistaken Reversal?

[Lsat180Please]

Hi. I need some help with this one. Im confused as to how you could assume that grammar and semantics was the necessary condition for distinguishing between homophones? In my head, I thought they assumed that knowing that it couldn't recognize between homophones was sufficient to know that they could not recognize grammar and semantics. But I believe that this is a sufficient rather than a necessary one.

I also thought okay so if the conclusion is true then premise is true bc of the necessary assumption:

not grammar and semantics --> not accurately translate --> not distinguish homophones

this gets us to answer A in a different way. Please explain how you could understand that if homophones --> grammar and semantics in the way that you explained. Is this just something to memorize with conditional conclusions and a premise? or did all the nots make this problem extra confusing?

Thank you!!!

[Brook Miscoski]

Hi all. Looking through the comments, it looks like the question that remains is why (A) and why not (D).

Look at the second sentence of the stimulus: "...until...grammatical and semantic relations..." The word "until" indicates a necessary condition, so the stimulus is claiming that the ability to recognize grammatical and semantic relations is necessary. However, the stimulus fails to link this to distinguishing homophones, so our assumption should show that recognition of grammatical and semantic relations is necessary for distinguishing homophones.

Looking at answer choice (A): "...must be able to recognize grammatical and semantic relations..." The word "must" indicates a necessary condition, and this choice makes the link to homophones.

Looking at answer choice (D): "unless...distinguish between homophones..." The word "unless" indicates a necessary condition, so this choice is a mistaken reversal. The recognition of grammatical and semantic relations is what we want as the necessary condition.

Personally, I did this question based on a reaction. When I read this, the leap to "recognize grammatical and semantic relations" as the obligatory solution was completely unsupported, so my reaction was that the stimulus is assuming there isn't another way to achieve the goal. Answer choice (A) plainly says there isn't another way to achieve the goal, whereas (D) reverses the relationship by starting at the goal. If you consider what I'm saying, you will realize that my reaction isn't logically different from the more step-by-step approach I outlined above. If you had this sort of reaction, your intuition was good.

[CPA2lawschool]

Hi Sodomojo,

The premise given in this stimulus is that voice-recognition technology cannot distinguish between homophones. Then we are given the conclusion as a conditional statement, that until voice-recognition tech can recognize and utilize grammatical and semantic relations among words, these programs won't accurately translate spoken words into text (Accurate translation Grammar & Semantics). The phrase "as a consequence" serves to link the original premise as a conditional statement to the sufficient condition of accurate translation of speech-to-text.

As an assumption question, there is a clear logical break between distinguishing homophones, which has been tied to accurate translations, and recognizing grammar and semantics. The phrase "as a consequence" serves to link the original premise as a conditional statement to the sufficient condition of accurate translation, so the conditional relationship we actually have is:

Accurate Translation Distinguishing Homophone Grammar & Semantics

with the logical break occurring between the homophones and the grammar and semantics conditions. We have an A B C conditional relationship, but B C doesn't exist, yet the conclusion still draws the inference that A C.

Answer choices (A) and (D) both link homophones and grammar and semantics, but in opposite ways. (A) makes grammar and semantics the necessary condition for the sufficient condition of distinguishing homophones, bridging the gap in the stimulus by linking B C, allowing us to then draw the conclusion's inference of A C.

(D) on the hand makes distinguishing homophones the necessary condition for recognizing grammar and semantics, essentially reading the stimulus as:

Accurate translation Recognize Grammar & Semantics Homophones

while tacking on an unnecessary B C conditional relationship. However, this isn't the reasoning in the stimulus, and not the assumption that is required for the logic to hold true, but a Mistaken Reversal of it. So (D) is an incorrect answer choice.

Hope this clears things up!

Hi James,

Thanks for the explanation.


Quick clarification requested:


My original shorthand interpretation of the stimulus was:


Since "(not) distinguishing homophones", [Conclusion] If "accurately translate", then "recognize relations."


You notated the related premise as "distinguishing homophones" rather than "(currently not) distinguishing homophones."

Should we infer from the "as a consequence" explanation you provided above that the "currently not" of the premise is removed in this logic chain as a result of this being linked to the conditional conclusion?


If so, will you check my math on the rationale for why AC A is correct and why AC D is incorrect:



AC A appears to be the correct answer because the sufficient condition of AC A "distinguishing homophones" matches the premise in the stimulus and the negation of AC A's necessary condition "NOT recognize relation" effectively destroys the logic above.




AC D says: (If (not) distinguishing homophones, then (not) "recognize relations")

The negation of AC D: (If (not) distinguishing homophones, then "recognize relations")

This appears to be incorrect because the sufficient condition does not align with how the premise interacts with the conditional conclusion (see reasoning for why AC A's sufficient condition does align, i.e. why it should be read as distinguished rather than NOT distinguished) AND because the negation of the necessary statement of AC D aligns with the necessary condition concluded ("recognize relations").



Thanks!

[James Finch]

Hi CPA,

I find it easier to think about necessary conditions when taken as a positive rather than a negative, as it makes it easier for me to decipher which will be the prerequisite for the other, which in this case means that I would think through the stimulus as saying:

"Accurate translation requires recognizing relationships because we need to distinguish homophones in order to have accurate translation."

This means that there are two assumptions going on in the stimulus:

First, the one I noted above, that distinguishing homophones is a necessary condition for accurate translation
(AT DH). Without this (and this seems to be a difficult inference to make), the argument falls apart as it has no premise to use a basis.

Secondly, the conditional conclusion requires us to link the necessary condition (recognizing relationships) to the premise about distinguishing homophones. The only way to do this and have it make logical sense (ie that the premise is actually relevant to the conclusion) is for "recognizing relationships" to be necessary to distinguishing homophones, which in turn would be necessary to accurate translation, or:

AT DH RR,

which also means that no RR means no DH, which in turn means no AT:

RR DH AT

which would create the RR AT conclusion.

This means we have to have a linkage of DH RR, as well as AT DH, both of which are assumptions required by the argument. As it turns out, we're only given a realistic choice between DH RR and RR DH, with (A) being the former and (D) being the latter. (D) is the mistaken reversal of the missing link we need in the chain, while (A) is the correct relationship.

Hope this helps!

[Lily123]

Does (D) make the argument?

AT RR DH

If that's correct, does this mean DH becomes the necessary condition for AT instead of RR (even though RR is still necessary for AT in the above)?

In other words, does the relationship imply we could get accurate translations by teaching computers to distinguish between all homophones WITHOUT teaching it to recognize relations?

I understand that when we have: A B C we can infer A C
but I didn't realize that meant we could bypass B altogether. I always thought that B remained necessary to A no matter what.

The reason I picked (D) was because (A) gave us:
AT DH RR
Which I thought made DH the necessary condition for AT. Have I had it backwards this whole time?