I have been wondering what it is that we mean by trolling and went hunting for sensible discourse upon the matter.
My preliminary position, before reading around the topic, was that presumably there are several forms of trolling and that this is a label with a spectrum. So I expect to find a range of prefix adjectives. I also see trolling as a form of bullying and so I expect to find an equivalent of victim.
I am quite sure that whatever it is that forms malevolent trolling (which may be one of those labelled subsets), a lot would be cured by (i) fair curation (ii) forms of registration or signature (iii) post frequency limitation. I'm also pretty sure that what we should be doing is encouraging discussion and allowing disagreement but having an environment that produces clarification.
Trolling is malicious online behaviour, or disruptive online behaviour, intended to provoke and distress others for amusement. [2]. Trolls are generally male and show traits of both psychopathy and sadism.
Trait psychopathy is characterised by deficits in empathy and self-control, and a deceitful, callous, and unemotional interpersonal style and impulsive behaviour. 11,12 Trait psychopathy has been shown to be a significant positive predictor of trolling,1,7,8 with researchers theorising that the empathic deficits and thrill-seeking characteristics associated with trait psychopathy may lead these individuals to engage in trolling behaviours. 9
Trait sadism has also been established as a significant positive predictor of trolling.1,7,8,13 An individual high in trait sadism derives enjoyment and pleasure from inflicting physical or emotional pain and humiliation onto others.14 In a recent experimental study, researchers found a parallel pattern for trolls and individuals with high trait sadism to underestimate others' pain intensity and minimise the suffering of others.13 Researchers have even suggested trolling manifests itself in the form of online sadism, and trolls use the Internet as their virtual “playground” to derive enjoyment at the expense of other users'.1 In sum, this combination of psychopathy and sadism paints the Internet troll as an individual who is callous, lacks empathy, and enjoys causing harm to others.7 [2, academic references there. Grammar left as is, though it was tempting...]
I think what this says is that someone who trolls, or who is likely to, looks upon the internet as a playground without censure for bad behaviour. So, like the kid in the playground who goes around with the sole objective of disrupting anyone who is having fun, the 'fun' for the troll lies in disruption. If that is a correct description then I am already aware of people who do this in any group conversation, where, as soon as two people manage to begin exchanging views in any constructive way (such as explaining how they reached what appear to be opposite positions and not becoming heated about this), the conversational troll will intrude to prevent any accommodation being reached. This might be done by claiming to others that an argument is brewing when the opposite was true, by declaiming that 'this is getting serious' as if that must be disallowed (and demonstrating that, as far as they as concerned, this is so), or by simply making more noise than the discussion, drowning out the conversation. Thus the troll-like activity acts to prevent any positive occurrence, every conversation of consequence. A partial solution is available in face-to-face physical presence situations because the quasi-troll can be confronted and cannot escape. The equivalent action on the internet is far harder to challenge. I notice, though, that even a challenge to poor behaviour can be thwarted. Take an example where someone has made an ad hominem ¹ attack, is called on it and where the response is not a backing down but further escalation to name-calling; this then counts as 'successful' trolling, since the thread of what may have become useful conversation has been disrupted. What one needs to be able to do is move the constructive part of the conversation off-line; that has advantages, but notably moves the conversation out of the group – again, this means the troll has succeeded.
We might characterise a troll as being callous, low in empathy, as above. The references [1] and [2] indicate that someone with high self-esteem and a high level of sadism is likely to troll. What they are not doing is showing low self-worth, but the opposite; the better they feel about themselves, the more likely they are to troll. Which makes it a reinforcing behaviour, surely. Particularly, if you respond to the troll and show you're hurt, that reinforces the behaviour; that makes the easily found advice correct; don't feed the troll. The solution then has to be to understand the mechanism and defeat that.
I've recently had advice from my local police about public bullying; one is to become an active bystander and go ask, directly, "Are you okay?". Presumably by asking the question one is indicating the observed opposite, that things are not 'okay'. So the same applies to trolling, one must indicate that this is not okay.
However, that explains what trolling is, but not how to recognise it. Too many people are offended (=hurt) by anything that disagrees with their fondly held belief. Which means then that some of the "I'm offended' needs to be challenged, even if one is to be accused of trolling. What passes for cynicism and satire will often be mistaken for trolling. ² Therein lies a wholly different problem; for meaningful discourse to occur between two parties holding differing opinions challenge and exchange must occur; what must not occur is demonisation of the opposition. It is all very well falling into tribal groups but to find solutions to issues we have to cuase the tribes to be collective and to be co-operative nit merely within the tribes but between the tribes. that calls for conversation during which each party learns how the other(s) reached their conclusions. If one starts with an approach that wants to understand, it is more likely that both sides will find some value in the needs and wants of the other, leading to useful consensus and positive progress. But have any of that, the aggression must be removed and, lostly, be replaced with listening. A troll is not interested in allowing such positives to occur, only in disruption.
From the perspective of an online media organ, like say the Guardian or The Conversation, curation becomes the solution and the problem. Curation amounts to policing the responses to an article and, obviously, soaks up resources. At the very same time, if the standard of writing is high, then the responses to an article can be seen as adding value to the article, and doing this at merely the cost of the curation. So then the issue changes to a consideration of what level of curation is acceptable. If the topic for discussion has nothing to do with trolling itself then it is possible for fairly heavy curation to act more like the letters editor of a newspaper and for posting to occur only after some vetting, so that what appears is not only free of error, it raises a variety of points that don't degenerate into argument. Where the curation is after the fact of posting it is much more difficult to curate—"this post was removed for transgressing our site guidelines" becomes the flag that something was removed. There is another position where complete threads (of argument) are removed. What goes wrong with curation after the fact is to do with the measures of 'success' of the site. For example, if the metric is that there are more clicks (willingly interpreted as interested readers), that will encourage argument. The click-count might well count attempts to respond, so counting multiple times those who make many responses. A metric that counts responses falls more heavily still that way, counting (I suspect) something different from what the site owner thinks of as volume of interest.
One consequence of this is that media sites, in response to the need for curation, then reduce the range of articles that permit response. Partly this is because their chosen form of curation serves the idea that we have instant media service, so that the curator is 'live' and they cannot serve many streams simultaneously. That suggests to me that where instant posted response is not available, some other (slower) form should be on offer, equivalent to Letters to the Editor.
Floundering through a variety of sites, there are some where the poster of a response appears to be using their proper name; Bill Smith rather than WildAngry&50. This implies that there was some form of registration, some confirmation that the respondent has a list of properties (e.g., confirmed name, home address, email, accurately described and somehow ratified); this goes a long way toward keeping the responses clean but still needs curation. Yet such sites are more nearly self-correcting. There's a trade here; you want to be able to post but you don't want to be receiving hate mail any more than you want a brick through a window; you are to an extent trusting that your registration is secure. It might also be that the site owner then enters into some form of understanding with frequent respondents, but that is a slippery slope that can only trend to the dark side; as soon as you are rewarded you're going to be tempted to steadily more extreme views because those will cause more reaction. That is divisive, where what we need in terms of social cohesion is to have nudges towards consensus, quite an opposite position. This then supports the Facebook emoji response, though I think we might have a comment stream variety such as a range of Agree to Disagree alongside Like (which doesn't automatically mean one agrees, it might well be that you like the way a point has been made).
There is a curious equivalence with sites that attempt to help, such as any wiki, any forum to which one goes to solve problems such as technical ones. Here is a place where the troll is less likely to play (but I don't see why, maybe they haven't been driven to find less easy pastures), and respondents (helping, or agreeing that they have a similar issue) have aggregated data that indicates how many posts they've made, with a rating perhaps from novice to expert so that you have an idea how to rank a response. I would like to see an equivalent on comment forums.
Meanwhile, the role of the comment space needs to become clearer and for that we need better, more consistent curation. Which in turn means that it would be helpful to be able to read and understand the objective of what little curation takes place. I am increasingly of the opinion that many should be diverted to Twitter et al, so that those who want to make slower and more considered responses have an appropriate place. I think I'd quite like to have a curator response equivalent to "this post was moved to our Twitter account".
DJS 20220916-20
The ad hominem attack is not the only troll attack. [3]
1 Argumentum ad hominem attack; the respondent attacks the writer, not the content. This diverts the argument from the content to the provider. It classes as a logical fallacy. The simplest I can offer is "Why would you listen to someone who dresses like that?" Which has nothing to do with the logic of the argument and so is diversion; the challenge has to be made in such a way that the attack is identified as such and recognised by the audience as a negative move that is evidence of a lack of argument. That must not be seen as a degeneration to name-calling. It would help hugely if the term ad hominem was widely understood.
2 Poe's Law: on the internet there will be at least one who will take satire at face value. And post accordingly, missing the point.
² ³ ⁴ ⁵ ⁶
[1] https://theconversation.com/new-research-shows-trolls-dont-just-enjoy-hurting-others-they-also-feel-good-about-themselves-145931 but read the lengthy and extended comments, which exhibit things that might well be trolling, though one suspects that the contributors are demonstrating what trolling is rather than doing it, but I'm not at all certain.
[2] https://www.liebertpub.com/doi/10.1089/cyber.2019.0652 High Esteem and Hurting Others Online: Trait Sadism Moderates the Relationship Between Self-Esteem and Internet Trolling, Evita MarchGenevieve Steele 20200710 https://doi.org/10.1089/cyber.2019.0652
Results corroborated previous research showing gender (male) to be a significant predictor of trolling, and trait psychopathy and sadism to be significant positive predictors. Although self-esteem had no additional value on top of trait psychopathy and sadism in explaining trolling, there was a significant interaction between self-esteem and trait sadism. A moderation analysis indicated a positive relationship between self-esteem and trolling, but only when trait sadism was high. These results portray the troll as a callous individual may enjoy causing psychological harm, particularly if their self-esteem is high. These results contribute to building the psychological profile of trolls and provide future directions for research exploring trolling behaviours.
[3] https://samanthanorth.com/internet-trolling-tactics though I had trouble loading this at all.
Whataboutism or whataboutery, accusing the opponent of somethign even worse. Still ad hominem. Often utter fabrication.
Sealioning, which is asking endless questions (not caring what the answer is) - the objective is still to derail the discussion and to enrage the other participants.
Debate obsessives claim that all they are interested in is the debate and in freedom of speech. What they're doing is wasting your life and time, not debating at all.
Sockpuppetry occurs when someone uses several accounts (all fake). These are used to make it look at though general opinion is something it isn't. In effect this looks identical to gang activity by trolls.
'Do your own research', they say, a common phrase among conspiracy whackos. This is a cop-out for provision of hard data.
Strawman fallacies occur when a situation is exaggerated to the point where it can be defeated (easily). The assumption is that if the weak position is defeated, so is the far stronger one. Not true.
Tribalism picks on any group that can be labelled as outside. The idea is to create conflict.
Hate mongering - pretty direct, easily spotted, against the rules in social media. Easily reported and stopped.
Actions:
Ignore. Unfriend, Mute; reply only once and switch off notifications. Report (object) if possible; grab a screenshot first.
[4] https://bulliesout.com/need-support/young-people/trolling/
Private trolling is not trolling, but bullying. The internet is governed in the UK by the Communications Act 2003 and also covers messages sent by text, e-mails and mobile phone calls.Under Section 127 of the Act, it states that it is an offence to send messages that are ‘grossly offensive or of an indecent, obscene or menacing character.’
Mind, until we have internet police, no action will be taken.
[5] https://blog.hootsuite.com/how-to-deal-with-trolls-on-social-media/ Same advice as [4], but more colloquial and more wordy. Aimed at social marketers.
[6] https://www.esafety.gov.au/young-people/trolling A personal attack is not trolling, but flaming, online hate. Trolling is when someone makes a deliberatively provocative comment or post and waits for people to take the bait.