6: Conclusion
Text analysis must be constantly interrogated on a methodological level. First, how complete is the text being analyzed and what is its quality? Particularly regarding machine-translated text, what is the accuracy of the OCR process by which analog pages turn into digital text? See, for instance, Danny Sullivan’s colorful critique of Google’s Ngram tool.
More fundamentally, how effective is large-scale text analysis in drawing meaning out of words? Simply counting the number of occurrences of certain words or phrases can often lead to a simplistic interpretation of literary or historical trends. What should the relationship be between traditional “close reading” and “distant reading” and how do they work in tandem?
Despite questions such as these, the increasing rate of digitization and deluge of data will force humanities scholars to learn how to effectively access and interpret text on a previously unimaginable scale.