Cre8n Txt: A Rule-Based Approach to Textese

Abstract

This study investigates whether textism generation can be described through a set of rules – specifically, whether Textese has a context-sensitive, rule-based grammar that describes how textisms are created from their Standard American English counterparts. This was done by assessing 103 participants using a Textese translation task, a grammaticality acceptability judgment task, and a text-message submission task. Of 11 textism categories, rules were able to be created for eight of them. The remaining three categories had insufficient data on which to base substantial generalizations, although they were not inconsistent with a rule-based approach. Along with this, a singular predictable textism form was generated for 63% of English forms, while only 12.5% of English forms had more than one textism form, showing that textism generation is predictable considerably more often than it is not. With a majority of textism categories having generalizable rules and a relatively low rate of multiple textism forms for a single English form, it is likely that Textese has a rule-based grammar for generating textisms.