English verb regularization in books and tweets.

Area of Science:

• Linguistics
• Computational Linguistics
• Sociolinguistics

Background:

• The English language undergoes continuous evolution, impacting grammatical structures like verb conjugation.
• Verb regularization, exemplified by the shift from irregular to regular past tense forms (e.g., 'helped' instead of 'holp'), is a key linguistic change.
• Analyzing large text corpora can reveal patterns and rates of language change.

Purpose of the Study:

• To quantify the extent of verb regularization in contemporary English.
• To compare the rate of regularization in digital social media versus published books.
• To investigate regional and demographic factors influencing verb regularization.

Main Methods:

• Utilized two large, distinct datasets: Google Books (2003-2008) and Twitter messages (2008-2017).
• Employed computational methods to identify and quantify irregular and regular past tense verb forms.
• Analyzed geotagged data to explore geographical variations in regularization.

Main Results:

• Verb regularization is significantly more widespread on Twitter than in published English fiction.
• Regularization rates differ between American and British English contexts, with higher rates in US Twitter data compared to US books, and the reverse for UK data.
• Regional variations in verb regularization were observed across US counties, but not strongly correlated with population-adjusted socio-demographic factors like education or income.

Conclusions:

• Social media platforms exhibit a higher degree of linguistic regularization compared to traditional published media.
• Geographical location plays a role in the rate of verb regularization, with distinct patterns in the US and UK.
• While regional differences exist, socio-demographic factors do not appear to be primary drivers of verb regularization rates in the analyzed datasets.