Q8 - A translation invariably reflects

[cyruswhittaker]

Could you please explain why C is correct rather than B? Is it because the issue of distinct writing styles is only a problem for human translators due to several translators completing a work? So this wouldn't be an issue for computer translators (assuming one computer translation is uniform)?

[bbirdwell]

Yep. The stylistic difference are due to the multiple translators. Seems relatively safe to assume that a single computer program would do a document, and that it would be consistent.

However, for B, the case is clear. That's a relevant question, because what if the whole "80% accuracy rate" is total nonsense? What if 80% accuracy by computer is less readable than the stylistic differences of the human translators?

[tzyc]

I did not quite get what the question is asking...
If we answer what is asking in each answer choise, we would be able to evaluate whether the argument sounds or not...so find the one which is the least important (or meaningless??) to asnwer for evaluate the argument...is this what it is asking?
Can I ask why (A) and (B) are wrong answers then?

Thank you

[sportsfan8491]

I know that the last post in this thread is a little dated, but if I can revive it and provide my "two cents" on this question, hopefully my comments will be helpful. Experts, please feel free to correct anything that I've said, if you believe it is inaccurate.

Pardon the slight paraphrasing, however, in a nutshell, the argument says that there are some problems with the way that humans translate certain document (numerous translators, working on "long" documents, results in varied and incompatible writing styles). It then tells us that there are some benefits provided by computer programs for language translation (lacks human translator bias, finishes the job quicker, produces stylistically uniform results, gives pretty high accuracy rate). From these points, the argument concludes that we should go with the computer translation programs over human translators.

This seems like a pretty strong conclusion, as there might be considerations that the author has forgotten to address. I think this is where the question stem makes this a pretty tricky question, especially for the early stages of a section. I saw the question stem as asking me: "What has the author already accounted for in his argument?" The point that the author has already accounted for will be the correct answer. On the flip-side of this, the four incorrect answers will be points that the author hasn't necessarily considered or mentioned, but whose investigation would be worthwhile for determining the cogency of the author's argument, given the evidence he/she has provided.

(A) is wrong because the author hasn't accounted for this possibility and it is pertinent to address. Stylistic variety is one of the drawbacks that the author presents for the human translation method. However, the author fails to address whether the problem of varied human translation can be overcome. If it can be overcome, the argument is slightly weakened, so this would be an important factor to consider in evaluating the author's argument.

(B) is wrong because the author hasn't accounted for this possibility and it is pertinent to address. The author mentions a high accuracy rate for computer programs as one of the benefits that they can provide, but as I believe was mentioned in a previous post, we don't know whether this figure is even remotely justified here. If a numerical comparison of accuracy isn't relevant or justified, then the argument is slightly weakened because the numerical accuracy rate is one of the main pieces of evidence that the author provides in making his/her argument.

(C) is correct because the author has already accounted for this point (I put this in italics, in my points above, for this very reason). I think the tricky thing to watch out for is that we are given the plural form of "programs," so one might be tempted to think that different computer programs will be used and having this variety, in terms of programs, will produce varied writing styles. However, the author used the plural version of "programs" in the stimulus and said that they would produce the same results stylistically. So, the author would probably state that this isn't something that they forgot to address, as it is explicitly accounted for in their argument.

Another way I validated this answer is through an assumption that I made that since these are computer programs, the software is probably going to be programmed before the programs are put to use ("work"). I made this assumption based on something that I noticed in the stimulus: we are told that they will "'work' without intervention," but notice that we aren't told about the creation stage (the stage(s) before they are put to "work").

So, I took the word "work" to refer to the stage where the actual translation takes place, and not to the initial creation/set-up stage for the programs. This made answer choice (C) seem like a pointless inquiry to me, as I could see the author stating that even different programs could be written to contain the same code and, thus, produce stylistically uniform results when they were actually "working". If different programs are programmed to "work" exactly the same, how can they possibly produce a distinct writing style from one another? I know this might be a bit of a stretch, as I brought in some outside knowledge here, but this is the thought process that I used to choose this answer with a fairly high degree of confidence.

(D) is wrong because the author hasn't accounted for this possibility and it is pertinent to address. I think this could call into question the accuracy rate: a "yes" answer to this particular inquiry would somewhat undermine the connection between the evidence and conclusion.

(E) is wrong because the author hasn't accounted for this possibility and it is pertinent to address. We are told about a general accuracy rate, but what if this doesn't hold true for niche users of the software? What if it doesn't even hold true for general users? If computer programs just don't stack up to human translation for these users, it could be problematic for the argument. This seems like a relevant comparison that needs to be scrutinized.

[ohthatpatrick]

That was an awesome and eloquent post and I agree with the lion's share of what was said, but I want to try to write a slightly pithier one and clarify this really weird question stem.

"Evaluate" questions are pretty rare, definitely less than 1 per test. Evaluate questions just ask, "which of these answer choices is relevant to strengthening/weakening the argument?"

This is an Evaluate-EXCEPT question (this might be the only one I've ever seen). Remember when you do EXCEPT questions to be careful about describing the correct answer in unfair terms.

In Games, it's totally fair to flip a Could Be True - EXCEPT into "Which Must be False?"

But in LR, it is not fair to flip a Strengthen-EXCEPT into "Which one Weakens?" Instead, we would just think, "Which one DOESN'T Strengthen?" (Remember, something could just be irrelevant ... irrelevance DOESN'T strengthen or weaken)

So 4 of these answer choices will be relevant to strengthening/weakening the argument, and 1 correct answer will be irrelevant.

The core issue is:
Should we use a computer program or human translators to translate a long document?

What we know so far is:
HUMAN TRANSLATORS
- they have different writing styles, which can sometimes be incompatible

COMPUTERS
- faster than humans
- stylistically uniform
- 80% accuracy rate

(A) This is relevant, because if we can solve the problem associated with human translators, then human translators become a better option than before.

(B) This is relevant, because the 80% accuracy rate was one of the premises. If the 80% accuracy rate is meaningless, than the author's endorsement of COMPUTERS is weakened by having one less leg to stand on.

(C) This is irrelevant, because whether different programs have different styles or not, that wouldn't enter into our debate between using A COMPUTER PROGRAM vs. SEVERAL HUMAN TRANSLATORS. If we're using a computer, we're using ONE computer program, which will produce a "stylistically uniform translation". So it doesn't matter whether different programs have distinct writing styles, because we'd only ever use one at a time. We'd never encounter the problem we do with several human translators, who can have incompatible writing styles.

(D) This is relevant because this digs into how trivial / important that 20% inaccuracy rate of the computer program is. Is the computer messing up words that are inconsequential or is it messing up incredibly integral parts of the text?

(E) This is relevant because we don't know how accurate human translators are. Are they 99% accurate? Are they 50% accurate?
And do the type of mistakes computer programs make more/less important to users than the type of mistakes humans make?

[hyk1310]

I realize this is an old post, but I wanted to ask something that I may not be quite grasping just yet.

The criteria for determining whether we should choose the computer translation program over human translation seem to be: 1. speed 2. stylistic uniformity. Accuracy doesn't seem to be a criteria, although 80% accuracy rate is mentioned as an addendum in the passage.

I can see why C is the answer, but I fail to see why E doesn't work (E first of all brings up users' needs, which is unmentioned in the passage; and the accuracy part seems to be an irrelevant criteria).

Thanks for your help!

[jeanne'sjean]

I realize this is an old post, but I wanted to ask something that I may not be quite grasping just yet.

The criteria for determining whether we should choose the computer translation program over human translation seem to be: 1. speed 2. stylistic uniformity. Accuracy doesn't seem to be a criteria, although 80% accuracy rate is mentioned as an addendum in the passage.

I can see why C is the answer, but I fail to see why E doesn't work (E first of all brings up users' needs, which is unmentioned in the passage; and the accuracy part seems to be an irrelevant criteria).

Thanks for your help!

I have a similar question re AC (A) and may be my understanding of the argument has some twists.

The conclusion said:" Therefore, when a long document needs to be translated quickly..." So in my opinion, doesn't the author emphasize the criterion of speed? Therefore, even if the stylistic variety in human translation could be solved as (A) said, what if the author only care about the SPEED and don't care the stylistic problem?

What confuses me is the "quickly" mentioned in the conclusion. Doesn't that exclude the consideration of other relevant criteria?

Thanks in advance!