Internet Trolling

Whenever you hear something dumb—and I mean really, unbelievably, absurdly dumb—there is always something that top that: the YouTube comment section. xkcd seemed to think there is nothing “quite as bad”:


But a couple of months ago, when our Internet overlords forced YouTube comments to require a Google+ account, the Internet itself responded with a massive backlash. We could no longer remain anonymous. It was the end of the world, almost as if we had to adapt to writing comments using a real-identity profile system that almost anyone can see and where you have no real privacy because all of your information is stored by a multi-billion dollar company.

One of the most significant reasons the YouTube comments were so bad was that everyone could be anonymous, and thus have no accountability like they would have in real life if they uttered something so dumb in public. As someone who has managed forums before, with first-hand experience in dealing with flamers, spammers, and downright immature idiots, I found myself on Google’s side of the fence when the Google+/YouTube controversy started.

Internet Trolling

My last post on trolling was over three years ago, titled “A Sociological Perspective on Internet Trolling.” I looked at the origins of Internet trolling and attributed it to the combined factors of decline of impression management and decline of social control, both of which resulted from from the interaction of a large number of anonymous, unconnected people. (By “decline of impression management,” I mean that people on the Internet have less of a need to maintain a reputation. And “decline of social control” is what it sounds like: there are fewer influences and incentives for civilized behavior.)

The Origins of Trolling

In just a little more than three years later, however, the landscape of the Internet has changed significantly. Social media was still fighting a campaign of conquest back then, but now it is the established empire. This forces a reevaluation of Internet trolling. Social media has significantly reduced, if not negated, the effects of anonymity in regions where most of Internet interaction takes place. But trolling still exists, albeit in a different form.

Additionally, the definition of trolling needs to be reexamined. From Wikipedia on Dec. 8, 2010 (given the topic, I think the Wikipedia definition gives a more practical, if not more accurate, definition than from a formal dictionary):

In Internet slang, a troll is someone who posts inflammatory, extraneous, or off-topic messages in an online community, such as an online discussion forum, chat room, or blog, with the primary intent of provoking other users into a desired emotional response or of otherwise disrupting normal on-topic discussion.

Note that I have highlighted the phrase “with the primary intent.” This made sense in an age when forums, chat rooms, and blogs were indeed the primary methods of mass online communication. It was difficult to imagine that someone could be a troll “accidentally.”

But here is today’s Feb. 2, 2014 definition:

In Internet slang, a troll is a person who sows discord on the Internet by starting arguments or upsetting people, by posting inflammatory, extraneous, or off-topic messages in an online community (such as a forum, chat room, or blog), either accidentally or with the deliberate intent of provoking readers into an emotional response or of otherwise disrupting normal on-topic discussion.

Today, accidental and intentional disruptions are both considered trolling. In addition, the 2010 definition started by saying that a troll is “someone who posts inflammatory, extraneous, or off-topic messages in an online community,” i.e. an action-oriented definition. If you do X, then you are a troll. By contrast, the 2014 definition starts by being results-oriented, a troll is “someone who sows discord on the Internet.” Hence the possibility of accidental or intentional trolling.

So what does the change in definition say? It means that trolling has evolved, and we may trace the evolution of trolling to multiple factors including:

  1. A change in mode of mass Internet communication, e.g. with social media, it is now easier to be in a situation where you might be a troll, even if unintentionally.
  2. A change in usage of the Internet, e.g. computers vs smart phones.
  3. A change in Internet policies and rules, e.g. the decline of anonymity imposed by Google.

Mode of Communication

In 2010, Twitter was still in its relative infancy, not quite the publicly traded company it is today. But the Twitter style of communication is very different from that of a “forum, chat room, or blog” that both definitions of trolling included as examples of online communication. On Twitter, the 140 character limit severely limits the amount of context, explanation, and room for justification. Thus it is extremely easy to misinterpret posts on Twitter that would not be misinterpreted in a more classic online community.

While factually true and supportive of an arguably legitimate point, this tweet by itself looks incredibly offensive, and the “point” that it seems to make is not the point it was intended to make. Indeed, it generated a lot of controversy. Indeed, here is a link by the same author, Richard Dawkins, talking about the same subject, but using a different medium. Interestingly, he was very aware of Twitter’s inflammatory nature, and in fact the aggressive-looking post is actually a toned-down version of one that could have compared Jews and Muslims:

Twitter’s 140 character limit always presents a tough challenge, but I tried to rise to it. Nobel Prizes are a pretty widely quoted, if not ideal, barometer of excellence in science. I thought about comparing the numbers of Nobel Prizes won by Jews (more than 120) and Muslims (ten if you count Peace Prizes, half that if you don’t). This astonishing discrepancy is rendered the more dramatic when you consider the small size of the world’s Jewish population. However, I decided against tweeting that comparison because it might seem unduly provocative (many Muslim “community leaders” are quite outspoken in their hatred of Jews) and I sought a more neutral comparison as more suitable to the potentially inflammable medium of Twitter. It is a remarkable fact that one Cambridge college, Trinity, has 32 Nobel Prizes to its credit.

In any case, is Richard Dawkins a troll on Twitter? My answer would be a qualified yes. The main caveat is that trolling usually derails the topic, i.e. “disrupt[s] normal on-topic discussion.” However, Dawkins is not trying to bring an existing discussion off topic, but instead, responding very relevantly to a false claim (details in the link). On the other hand, Dawkins is indeed an intentional provocateur, so it would seem that the general context of Richard Dawkins overrides the specific context to this incidence.

Now, if a public intellectual can be a troll, what about the average person? Or the below-average person? What about the “[The Moon landing] is so obviously faked its unbilevable, why r people so gullible??? Morons” type of people?

Of course, Twitter is just one example of many of the changes to social media over the past four years. It may be part of the shortening trend, i.e. the average length of what people read and write becomes shorter over time (books, to articles and letters, to blog posts, to Facebook statuses, to tweets). But it is clear that the Twitter-style confusion is partly due to the platform itself, that the lack of context would not be as extreme as in a forum, chat room, or blog.

Physical Usage of the Internet

The way people access the Internet is changing as well. While 21% of cell phone users used a smartphone by the end of 2009 [Nielsen], the number was 56% in 2013 [Pew]. In addition, a much higher percentage of activities are being done on smartphones now, thus the phone interaction with the Internet has increased by multiple factors since 2009/2010.

In addition, Internet usage is different on the smartphone vs on the computer [Harris]. This is probably tied to various factors including ease of typing (typing longer documents is more difficult on a smartphone), convenience/situational (e.g. using navigation or checking in on a smartphone), security (sensitive work material would probably belong on a computer, as would online shopping with credit card info), etc. As the Harris poll suggests, many of the top uses are device-sensitive.

The length of a piece of text is quite related. While the Harris poll reports social media used roughly evenly on computers and smartphones (with computer usage slightly more overall), there are also different kinds of social media. I might almost exclusively use WordPress on a computer to have better access to faster typing and to resources like research, links, or images; on the other hand, for Twitter, I might use a smartphone more. I would claim that part of the recent change in Internet trolling is due to the rise of smartphones.

Internet Policies

External regulations usually play a part, and in the case of the Internet, the story is no different. While the Internet is generally a rule-free, anti-establishment institution, it can be governed, often with massive backlash. In the midst of backlash against Google for being too much in control and for taking away our privacy, don’t forget when the government tried to pass SOPA, Google was on our side.


It is also interesting that the recent integration of YouTube commenting with a Google+ account has a significant precedent in the online gaming world: the integration of Blizzard forum posting with a RealID. This mid-2010 event provoked massive controversy and was eventually rescinded.

According to Blizzard, one of the reasons was the rampant trolling [previous link]:

The official forums have always been a great place to discuss the latest info on our games, offer ideas and suggestions, and share experiences with other players – however, the forums have also earned a reputation as a place where flame wars, trolling, and other unpleasantness run wild…. Removing the veil of anonymity typical to online dialogue will contribute to a more positive forum environment, promote constructive conversations, and connect the Blizzard community in ways they haven’t been connected before.

Now replace “Blizzard” and “the forums” and “games” with “YouTube” and “the comment section” and “videos” It’s interesting how this is echoed by Google did three years later. Of course, companies always have financial incentives as well, but it is uncanny that both list troll-fighting as reasons for the changes. On the other hand, it looks like Google succeeded where Blizzard failed.

Accidental Trolling

Accidental trolling is a new enough phenomenon that deserves its own section. While trolling doesn’t seem to happen on Facebook as much (though a couple of incidents did occur to me last year), we have already seen what it can do on forums as well as on Twitter, the difference being that on Twitter, the veil of anonymity is replaced by the shortness of speech.

Trolling from anonymity:

  • Identity not connected to real life, thus no accountability, no external reason to maintain reputation.
  • Others are anonymous as well, thus there is even less reason to refrain from insults or flaming.
  • No police or authority figure to enforce rules. There are admins and moderators but this is often insufficient when the amount of posts is absurdly large (e.g. Blizzard forums or YouTube comments).

Trolling from short speech:

  • No facial expressions or body gestures, thus no information about the tone or mood of a contextless sentence. Even sarcasm has to be extremely carefully crafted, otherwise it fails to be noted as sarcasm.
  • Generally little context about the statement. If someone makes a one-sentence claim about physics, it could be: (1) a physicist saying some profound statement about physics, (2) a physicist clarifying some commonly misunderstood notion, (3) a physicist responding to a claim, (4) some physics student writing about what they learned, (5) someone just making stuff up about physics, (6) a fiction writer making a life analogy to a basic concept in physics, (7) a fraud making a New Age claim about quantum healing, (8) a comedian making a joke about something in physics, etc., and it would not be trivial to discern these based on a short, aphoristic saying.
  • Limited space. For example, in a single tweet, I might address part of some large issue simply because I can’t address all of it in 140 characters, and then I get called out on tunnel-visioning to that one part and ignoring everything else. This is less of an issue with longer posts, but it still happens.

Note that the second category, trolling from short speech, is much more prone to accidental trolling, hence why accidental trolling is a more recent issue.

Accidental trolling also occurs from numerous cognitive fallacies. Confirmation bias and selection bias are the king and queen of online flaming. Using the advantage of the limitless information available on the Internet, it’s very easy to find information supporting your own position as well as to ignore contradictory evidence. Thus it’s easy to find articles all over the Internet that seem quite well-researched at first, but then you realize all the research is completely one-sided. This can cause a lot of trolling as well: given an obviously biased article, some commenter might point to some contradictory evidence to hint at how biased the original article was, but the contradictory evidence given happens to also have fallacies, and thus begins a flame war of commenters talking past one another.


Some sites have measures in place that (intentionally or not) reduce trolling, though sometimes at other costs. Tumblr, for instance, does away with the classic thread responding, thus eliminating the very possibility of how most trolling starts; on the other hand, it’s hard to have discussions at all, and it is often a pool of groupthink, with too similarly minded people repeating each other with little influence on the outside world. Reddit has a voting system that should theoretically send the trolls to the bottom; on the other hand, sophisticated trolls (again, whether intentionally or not) might make some statement that seems good at first, getting many upvotes, but in actually just derails the discussion or is a strawman argument, thus getting people who agree to upvote, making the voting system based often on how much people already agree with something, rather than based on the merits of one’s argument or position. If anything, Facebook is the most anti-troll of the major social networks, simply because it uses very real identities, thus people generally don’t want to say anything bad that others close to them can see.

As we have also observed, one method of actively fighting trolling is to remove anonymity, as in the cases of Blizzard and Google. However, this will generally (and perhaps unsurprisingly) produce significant backlash, and even if implemented successfully, it doesn’t fully solve the problem, because anonymity is not the only cause of trolling. In addition, we would need to resolve the factor of lack of context as well.

Thus two important theoretical measures for countering trolling are: (1) lifting the veil of anonymity, to instill the maintenance of reputation and social control, and (2) the increase of context, letting others judge something for what it is, rather than for a distorted vision of what it is.

As the differences between 2010 and 2014 show, trolling is a rapidly evolving issue in the online world that becomes ever more important as the world, not just America, becomes increasingly online. Progressing forward with what we know is the best course of action. And as one of my favorite YouTube commenters ever said, “It’s 2011. Can we get some color photographs of the moon already?

One thought on “Internet Trolling

  1. I think there is distinction between good and bad trolling. To me good trolling is like a literary, satirical art-form that is often humorous (often utilized in cases where the original context or question posed was absurd to begin with). However, the problem begins when people ‘troll’ in situations which don’t really merit trolling. From my experience, most trolling that is intended to provoke the other party seems like it’s conducted by teenagers who just are curious on human behavior (particularly on how the other person will react — or they could just be seeking some form of attention, in which case responding is reinforcing their behavior). I think the people who react strongly to certain claims need to remember to take everything with a grain of salt. Point of view does help, but then it becomes problematic when a person becomes paranoid and begins to think that everyone on the internet is a troll. I usually bypass that by saying “Not sure if trolling, but the scenario is interesting so I think it merits a response nonetheless”.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s