Commenting Fast and Slow on the Guardian Online: Incivility, Impoliteness and Speed of Response

Hateful communication online is a significant societal problem. It can negatively affect mental well-being, including of vulnerable citizens (e.g. Vogels, 2022), and has been a factor in suicide. It can also harden the polarisation of online communities (e.g. Allison & Bussey, 2020). Academic theorisations of incivility and other types of hateful communication still leave many open questions (e.g. Hopp, 2019). A better understanding of the workings of such phenomena, then, is particularly important.

Drawing on a diverse body of literature, including from the communication sciences (e.g. Papacharissi, 2004; Muddiman, 2017) and from psychology (e.g. Kahneman, 2011; Meier & Gross, 2015), we have hypothesised that at least some types of hateful communication in dialogic user-generated spaces online may at least partly be explained by the speed of response. To test this hypothesis, we collected and analysed all newsreader comments posted in response to 'Opinion is Free' news items on The Guardian Online, from March 2006 (when the site first allowed commenting) until March 2023 (when we collected the data). This is a dataset of 38 million reader comments, 2.4 billion words in size.

In the first test of our hypothesis (Clarke & Thompson, 2025), we used comments removed by The Guardian Online moderators as a proxy for incivility, given The Guardian Online moderation standards (The Guardian, 2009). Our analyses of the timestamps of approximately 1 million such blocked comments as compared to the timestamps for the approximately 37 million visible comments showed convincingly that blocked comments are posted more quickly and to a statistically significant degree across all conditions (e.g. both parent and child comments). They indicate a small to medium effect size. We performed post-hoc checks to ensure bot and commercial activity did not explain the time difference. With suspect bot and commercial activity posts removed, the difference in commenting time between blocked and visible comments actually became more stark.

In a second test of our hypothesis (Klamm, Clarke & Thompson, in prep.), we analysed the occurrence of less serious communicative transgressions: nuisance comments indicative of poor etiquette but that do not sacrifice principles of democracy (Papacharissi, 2004). Informed by Impoliteness Theory (e.g. Culpeper, 2016) and using a corpus-informed approach comparable to Oliver's (2025), we found no statistical difference in speed of response between comments containing impoliteness triggers and those free of such insulting, dismissive, condescending behaviour. Taken together, the results of our two studies provide further support for a fundamental distinction between communicative transgressions that sacrifice core democratic principles and the rule of law, and those that are a nuisance but do not have such serious implications (e.g. Papacharissi, 2004; Muddiman, 2017).

We end the talk by briefly explaining the next stage of our research: experimental studies which test the effectiveness to commenting behaviour of various time-delay interventions in the design of digital commenting forums.

References:

Allison, K. & Bussey, K. (2020) Communal quirks and circlejerks: A taxonomy of processes contributing to insularity in online communities. Proceedings of the International AAAI Conference on Web and Social Media 14: 12-23.

Clarke, B.P. & Thompson, W.H. (2025, under review). Fast and Furious: Temporal patterns of incivility in online comments.

Culpeper, J. (2016). Impoliteness strategies. In Capone, A & J.L. Mey (eds.) Interdisciplinary Studies in Pragmatics, Culture and Society, London: Springer, 421–445.

Hopp, T. (2019). A network analysis of political incivility dimensions. Communication and the Public, 4(3), 204-223.

Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.

Klamm, A., Clarke, B.P. & Thompson, W.H. (2025, in prep.). Commenting fast and slow: Speed of response and impoliteness in online newsreader comments.

Meier, L.L. & Gross, S. (2015) Episodes of incivility between subordinates and supervisors: examining the role of self-control and time with an interaction-record diary study. Journal of Organizational Behavior, 36(8) 1096–1113.

Muddiman, A. (2017). Personal and public levels of political incivility. International Journal of Communication, 11, 3182-3202.

Oliver, S. J. (2025). A corpus-based analysis of (im) politeness metalanguage and speech acts: The case of insults in Shakespeare's plays. Journal of Pragmatics, 235, 132-144.

Papacharissi, Z. (2004) Democracy online: civility, politeness, and the democratic potential of online political discussion groups. New Media & Society, 6(2) 259-283.

The Guardian (2009b) Community standards and participation guidelines. Available at: https://www.theguardian.com/community-standards

Vogels, E.A. (2022) Teens and cyberbullying 2022. Pew Research Center, 18 December. https://www.pewresearch.org/internet/2022/12/15/teens-and-cyberbullying-2022/