Recent advances in NLP show that language models retain a discernible level of knowledge in deontological ethics and moral norms. However, existing works often treat morality as binary, ranging from right to wrong. This simplistic view does not capture the nuances of moral judgment. Pluralist moral philosophers argue that human morality can be deconstructed into a finite number of elements, respecting individual differences in moral judgment. In line with this view, we build a pluralist moral sentence embedding space via a state-of-the-art contrastive learning approach. We systematically investigate the embedding space by studying the emergence of relationships among moral elements, both quantitatively and qualitatively. Our results show that a pluralist approach to morality can be captured in an embedding space. However, moral pluralism is challenging to deduce via self-supervision alone and requires a supervised approach with human labels.

Moral rhetoric influences our judgement. Although social scientists recognize moral expression as domain specific, there are no systematic methods for analyzing whether a text classifier learns the domain-specific expression of moral language or not. We propose Tomea, a method to compare a supervised classifier’s representation of moral rhetoric across domains. Tomea enables quantitative and qualitative comparisons of moral rhetoric via an interpretable exploration of similarities and differences across moral concepts and domains. We apply Tomea on moral narratives in thirty-five thousand tweets from seven domains. We extensively evaluate the method via a crowd study, a series of cross-domain moral classification comparisons, and a qualitative analysis of cross-domain moral expression.

Moral values influence how we interpret and act upon the information we receive. Identifying human moral values is essential for artificially intelligent agents to co-exist with humans. Recent progress in natural language processing allows the identification of moral values in textual discourse. However, domain-specific moral rhetoric poses challenges for transferring knowledge from one domain to another. We provide the first extensive investigation on the effects of cross-domain classification of moral values from text. We compare a state-of-the-art deep learning model (BERT) in seven domains and four cross-domain settings. We show that a value classifier can generalize and transfer knowledge to novel domains, but it can introduce catastrophic forgetting. We also highlight the typical classification errors in cross-domain value classification and compare the model predictions to the annotators agreement. Our results provide insights to computer and social scientists that seek to identify moral rhetoric specific to a domain of discourse.