Questions: Stylometry and Quantitative Textual Analysis

Question 1

Stylometric analysis attributes a disputed text to Author A with 94% confidence. A literary scholar argues the text's themes are inconsistent with A's known work. Which statement best characterizes the relationship between these two forms of evidence?

• A. The statistical evidence is stronger because it is objective; the thematic argument is subjective
• B. Both are interpretive, and they answer different questions — stylometry addresses surface linguistic habits, thematic analysis addresses meaning
• C. The thematic evidence overrides the statistical evidence because literature is about meaning, not statistics
• D. The conflict means neither result is reliable and the attribution must remain unknown

Answer: (B)

Stylometry and thematic analysis answer different questions. Stylometry asks: whose unconscious linguistic habits produced this surface? Thematic analysis asks: whose conscious ideas shaped this content? These can diverge — ghostwriting and collaboration are common, and authors change. Neither method is simply 'more objective.' Understanding their different claims prevents false conflicts and enables more nuanced attribution arguments.

Question 2

Why does the choice of which linguistic features to measure in stylometry count as an interpretive decision rather than a neutral technical choice?

Model answer: Different feature sets (function words, sentence length, punctuation, rare words) rest on different assumptions about what constitutes 'style' — whether it is unconscious habit or deliberate choice, whether it is lexical or syntactic. Different assumptions produce different results for the same texts. The choice of features is therefore a theoretical claim about the nature of style embedded in the methodology, not a value-neutral technical setting.

This connects to the broader epistemological point that quantification doesn't guarantee objectivity — it relocates interpretive decisions into the design of the method. Stylometry's power comes from consistency (the same features are measured the same way across texts), but consistency is not the same as neutrality. The scholar who chooses which features to measure is making an argument about what style is, and that argument shapes all subsequent results.