Question 90·Hard·Cross-Text Connections
Text 1
Some linguists argue that written corpora compiled from newspaper archives can reveal how grammatical constructions evolve over time, but relying on such corpora is problematic because editors and style guides deliberately filter out many of the informal usages that drive genuine language change. Consequently, the corpora underrepresent innovations that appear first in speech or unedited writing.
Text 2
A team of computational linguists created a large dataset of geotagged social-media posts spanning fifteen years. Because these posts are user-generated and seldom copy-edited, the team claims their dataset captures informal linguistic innovations at the moment they appear. The researchers compared frequencies of emerging constructions in their dataset with those in traditional newspaper corpora and found that the newspaper sources documented the new forms only after a significant delay.
Based on the texts, how would the team described in Text 2 most likely respond to the bolded claim in Text 1?
For cross-text connection questions, first restate in your own words the key claim or concern in Text 1 (especially if it’s highlighted or bolded). Then, summarize Text 2’s main method and conclusion, and decide whether Text 2 would agree, disagree, or modify Text 1’s claim—and why. Finally, go to the choices and eliminate any that contradict specific details (methods, evidence, or attitudes) from either text, and select the one that accurately captures how the perspective in Text 2 would respond to the specific idea from Text 1.
Hints
Clarify the criticism in Text 1
Re-read the bolded sentence in Text 1. What specific problem does the author see with using newspaper corpora to study language change?
Connect Text 2’s method to Text 1’s criticism
Look at how the social-media dataset in Text 2 is described. How is it different from newspaper corpora in terms of editing and informality?
Use the comparison result in Text 2
Text 2 reports a delay in when newspapers show new forms. Does this make newspapers look more reliable or less reliable for studying early stages of language change?
Eliminate answers that don’t address filtering
Be cautious with choices that shift the issue away from editorial filtering (for example, toward geography or topic coverage) instead of addressing whether editing causes newspapers to miss early informal innovations.
Step-by-step Explanation
Understand the bolded claim in Text 1
Focus on the bolded part: Text 1 says that using newspaper corpora is problematic because editors and style guides filter out informal usages that actually drive language change. So the key idea is: editorial filtering makes newspaper corpora bad at showing early, informal innovations.
Summarize what the team in Text 2 did and found
Text 2 describes a dataset of user-generated, seldom copy-edited social-media posts. The team claims this data captures informal innovations as they appear. They then compare these innovations in social media vs. newspaper corpora and find that newspapers show the new forms only after a delay.
Decide how Text 2 relates to Text 1’s criticism
Text 2’s evidence supports Text 1’s criticism: newspapers don’t show innovations right away, exactly what you’d expect if editors and style guides are filtering informal usages. Text 2’s team is not defending newspapers; instead, they are offering an alternative dataset that better captures early, informal change.
Match this relationship to the best answer choice
We want the answer that shows the team in Text 2 accepting the criticism of newspapers and highlighting how their own dataset avoids that problem. The option stating that their social-media dataset avoids the editorial filtering that makes newspaper corpora unreliable for studying early stages of language change best matches their method and findings. So the correct answer is: They would argue that their social-media dataset avoids the editorial filtering that makes newspaper corpora unreliable for studying early stages of language change.