Constructive internet dialogue, is it possible?

Years ago I read an article in New Scientist about road rage. A psychologist said part of the problem was people in cars are depersonalised. You don’t think of the other driver as a person, you think of them as a car. You can’t see their face, read their body language, hear their voice.

By contrast, if you bump into someone as you are walking down the street, there are all sorts of signals – from apologetic body language, to saying, “Sorry!” – that help to defuse the situation.

In a car, there’s none of that. The other driver doesn’t look like a real, breathing person, they look like an object in your way. Just you, and your irritation, feel real to you.

I’ve often thought something similar happens on the internet.

Once upon a time we thought the internet was a big shining hope. Wow, look at that, people from all over the world can talk to each other! Talking is good, so more talking is better, right? Let a new era of peace, love and human understanding reign!

OK, so maybe I’m exaggerating a bit. But in the ’90s there was an optimism about what the internet could do for democracy and for humankind. That it could be a space for conversations that couldn’t or wouldn’t happen easily in real life. That people talking and sharing would naturally come to understand each other better.

But what do we actually see when we look around the internet? We see blowhards shouting at each other on newspaper comment threads and nobody listening, nobody learning anything, nobody understanding each other better. We see flame wars, trolling, polarisation.

Not to say there’s no constructive dialogue on the internet, but it often seems like the exception. Far more often a difference of opinion seems to spiral off into anger, insults and viciousness.

I think a big part of that is, a bit like with road rage, it’s much easier to be rude and unreasonable on the internet than it would be in real life. Words on a screen become a place to vent your frustrations. The other people you’re talking to don’t seem that real. With the added factor that it’s technically public – encouraging some to show off and posture for the imagined invisible audience.

I do see some constructive conversations. I see a lot more situations where constructive online conversations could be useful but aren’t happening – from people of different backgrounds falling out on twitter, to contested areas of science like GM. If we could work out what makes constructive conversations happen in one place, and not another, wouldn’t that be handy?

Maybe sometimes people just want to fall out and let off steam, and we shouldn’t worry too much twitter spats. But it would be useful to know – for democracy, for consultations, for civil society – how we can have public conversations online, about issues that matter, in constructive ways.

I’ve been asking people this question for a while now, and no-one seems to have any answers. We know some stuff about creating constructive dialogue in real life (and it’s still a lot harder than it looks), but I don’t know of any real research on online dialogue. And I speak as someone who’s spent ten years involved in running online dialogue events.

I’ve got some thoughts, and I think our events were usually pretty constructive, partly because of the way we set them up, which we developed by trial and error over the years. To be honest though, those events nearly all involved school students talking to either scientists, or local councillors. Which is a fairly specific situation. Not everything I’ve learned can be applied elsewhere.

For example, I can tell you that live chats tend to be more polite when the teacher is in the classroom. But, sadly, I can’t send a teacher round to look over the shoulder of every bitter misogynist about to take potshots at Mary Beard.

However, here are a few thoughts I’ve had:-

  • I think an ongoing relationship makes a huge difference. 

I’ve seen constructive conversations emerge online, even from fraught or snarky beginnings, when people have an ongoing relationship. Even if it’s just as twitter pals. Where people place some value on a continuing relationship, they are more likely to make an effort to acknowledge valid points, compromise and accommodate. 

Whereas, of course, a drive-by commenter is like the car driver you’ll never see again. It’s easy to flick them metaphorical Vs and shout ‘WANKER!’ after their departing back.

It’s not practical to expect an ongoing relationship from all newspaper commenters or dialogue participants. But are there ways we can use this effect? Give people status within a community for constructive behaviour, a bit like stack overflow?

  • I think photo avatars make a difference.

They remind us we’re talking to real human beings. Are there other little nudges we could use to humanise people?

  • I think effective and proactive moderators or facilitators make a difference.

But that’s resource intensive. Communities can self-moderate, or have volunteer moderators, but I’m not sure I’d hold up reddit as a great example of a constructive online space…

  • I think ground rules make a difference. 

Other people must have more thoughts on this. I imagine there’s a great deal of tacit knowledge out there – in people like me, in online community managers, moderators and all sorts of forum administrators. I suspect there’s also many bits and pieces of relevant research in all sorts of disciplines – in psychology, in sociology, in IT, in politics – but no-one’s pulled all this stuff together.

So I’ve got some questions:-

  1. Does anyone know of any research on this? Do we know of ways that we can set up online spaces that make constructive conversations more likely? Is anyone actually studying it? Is anyone asking all the people I mentioned above what they know? And then is anyone putting this all together to try to design effective online dialogue spaces and testing them out?
  2. Even if there isn’t any research, what are your thoughts on this? What do you think makes some online conversations or spaces more constructive than others?
  3. If there isn’t any research (that we know of), does anyone want to fund me to do a PhD investigating this question? Because it seems like it would be a damn useful thing to find out.

24 responses to “Constructive internet dialogue, is it possible?

  1. Hi
    These are really good questions, but I don’t have the answers. It would be great to find some answers though and a PhD would give you the space to find them (if only one could get the funding). There must be some research… – psychologists etc. are not stupid, they generally pounce on such novel research topics. Ha, I just found a new thing called cyberpsychology! And a journal devoted to this topic. But I am not sure whether you want to go down that road! I really don’t know whether there is anybody else investigating how online communication can be made ‘better’ or rather what formats, factors, management etc. of online spaces can turn them into the spaces of mutual learning, dialogue and engagement that we dreamed of back in the 90s….

  2. Good post – had few of these ideas rolling around my head for a while. Haven’t looked into any research on this I’m afraid, but could offer you some of my own thoughts.

    Agree about avatars. I think that does seem to help. What I am pretty sure about is the ‘location’ of the discussion. In particular, Twitter is very good as a ‘trawler net’, and one can meet a lot of interesting people there. BUT I think it’s very poor for communication. The pressure on space makes it too easy to make comments which are then misconstrued, or simply come across in a more brusque tone than was intended. Blogs have often been the location in which I have seen people moving from antipathy towards greater understanding. This is often facilitated by the blog owner, who has a personal stake in the success and perception of their site. Conversely, Twitter is basically a free-for-all without a gatekeeper (Comment Is Free, although moderated, is also more like this – there is not one person who is identified as being ‘in charge’).

    Finally, there is an issue about self-editing. Twitter’s immediacy and simplicity is a weakness as well as a strength. There is an expectation that discussions are fast-paced, and this leads to people posting first, thinking later.

    If you get any tips on research, please let me know!

    • I’d agree it’s easy for things to blow up on twitter. Although because the cost of following someone is low, I follow some quite disparate people I probably wouldn’t have got friendly with in real life. That means I’m exposed to other points of view. I have seen people have v constructive conversations on twitter where they both moved considerably, or at least understood each other’s take on things better. But it takes both people making a real effort. I’ve seen a lot of unedifying slanging matches too.

      I agree, blogs can feel a bit more like you’ve gone round to someone’s house, so you try not to get too rude.

      There’s also things like messageboards, which I think vary enormously in feel and etiquette. That’s where I think it gets really interesting. The ways that norms of behaviour get established are probably quite subtle, but make a huge difference.

      I guess what I really want to know is, what can we take from all this in order to design online dialogue spaces – for example for consultations, or exercises in mass deliberative democracy? If we wanted to do a project like the re-writing of the Icelandic constitution, but online, how would we do it? Or is face-to-face the only way to keep it civil?

      (I also want to know how can we make public conversational spaces like Comment if Free less aggressive, misogynistic, etc, but maybe that’s just asking too much…)

  3. Hi Sophia,

    I wrote an article in a publication a couple of years back where I looked at the evidence on this topic (it is probably a bit outdated now)

    I agree with you on what you listed as helpful. In my role on the board I’ve also learnt a number of approaches they use to deal with conflict online; among other things they insist on people using real names (hard to police, but the principle is important) and they limit the number of posts that people can make per day (which forces people to think before they post).

    They’ve produced some useful guidance:

    Hope this is all helpful and thanks for a great post.


    • That’s really useful, thanks Ed. I think I may even have seen that local issues forum doc, when Shane and I were evaluating the local e-democracy national project, but I’d forgotten about it.

      Printing out the localism doc to digest later – I love the name of your chapter already, ‘The information society: community catalyst or local liability?’, well, exactly…

  4. Edward is right in that Steven Clift at has done a lot of work on this. Real names, limited posting, moderators. They all contribute towards making a conversation civil and constructive.

    But it doesn’t work in every situation. Pseudonymity is important in many situations, limiting posts can be frustrating and counter-productive to conversation and moderators (as I know well) aren’t always free.

    Refusing anonymous commenters is an achievable compromise. Perhaps don’t insist on real names, but do insist on some ID, preferably something consistent that is used elsewhere. Photo avatars too help.

    Clear objectives for a forum and groundrules to deal with posts that stray from those objectives help. Our live chats function much better if any misbehaviour gets dealt with quickly before it becomes the norm.

    The internet continues to be a fabulous place for conversation and discovery. IMHO the trick is to avoid newspaper comment sections in particular. And those funny sites where people try to be clever. And some blogs have comment sections where everyone is so very self-promoting 😉

    Visit for more about the greatest science engagement that ever was.

    • Cheers Shane. I agree that pseudonymity is useful in some circumstances, but that either real names or a stable pseudonym can be helpful. But I’ve still seen people have very aggressive and pointless conversations under their real names, and where they have a real world relationship too – e.g. on facebook, or messageboards.

      But then I guess people have conversations like that in real life too, and we can’t organise that out of existence. Although we can teach kids how to debate more constructively… (end of advert)

      And yes, the trick with live chats is to get in early and establish the right tone. I’ve been told that on CiF they reckon if the article author joins in below-the-line in the first half hour, it usually stays civil, but if not, it’s more likely to veer off. It also helps with live chats to visibly reward constructive behaviour, more than reacting to bad behaviour. That removes the incentive to act up. But we can’t treat the whole internet like a school classroom…

      I know what you mean about avoiding newspaper comment threads. BUT, these are public spaces. If public spaces become dominated by the shoutiest voices, then I think we’ve lost something.

  5. Pingback: Constructive internet dialogue, is it possible? | Public engagement, only digitally |

  6. How to encourage civilised discussion online? Very hard to say.
    I’ve found the softly-softly approached has help encourage civilised discussion here and on my blog… I used to get a lot more of the rage I felt coming through into the words I wrote, some years back, and think I have got quite good at reining that in nowadays.

    If people see an accommodating atmosphere, other people not getting attacked with insults etc. even if they make a stupid or rude comment, I guess they’re more likely to join it. As you also pointed out, that will sometimes require moderation, which isn’t always available.

    Depends on the person, though. A lot of people love the shouty conversations on their own merits (I like one now and then!) so is that necessarily a bad thing?

    Life is made up of important and unimportant conversations. We can’t all be serious about everything all the time! I agree it’s frustrating to see something descend into ad homs and all manner of nonsense when a good, productive conversation could be happening instead.

    But at least people are talking about things – who knows, in a few years they might remember that stupid comment and realise they’ve changed their minds since then. That’s happened to me a few times!

  7. Hi,

    Not much to add to what’s already been said. These are good questions and answers would be great!

    Intuition suggests that Benetta’s point about tapping into existing communities which form around a topic or campaign is perhaps the best situation for online dialogue to work. A clear common aim or problem to solve – something to work towards.

    Here’s one paper on this sort of thing:
    (super-summary: deliberation shifts participants’ views on youth anti-social behaviour, but that participation in online deliberation tends to reinforce extant political inequalities.).

    I’m slowly collecting bits and pieces about digital engagement here: any suggestions welcome…

    • Thanks Patrick, and your looks like a useful resource.

      Yes, I think you’re right that existing communities, or ones which form around a common aim, will dialogue well. People are then motivated to work together, listen to each other, etc.

      But what about issues where we want different – perhaps opposed – communities to come together? Is dialogue in the real world better in those kinds of cases? Sometimes though we’d want to turn to online dialogue, because people are geographically spread out and it’s difficult for them all to meet. Or because there are other obstacles to them all getting together in the real world – no time when they are all available, or too many people to easily fit in a room together, or simply because online dialogue is cheaper and resources are tight.

      I’m not expecting you to have an answer to these questions, I’m just thinking out loud really:-).

  8. Hi Benetta!

    Yes, I definitely agree that forums (and other internet locales) can be dream territory for obsessives. Remind me to tell you about the time I was accused of being a mind control agent for the New World Order…

    People with very fixed points of view can be extremely difficult to deal with online. They’re impossible to have dialogue with, because they refuse to listen, and as you say, they’ll just keep insisting on their point of view and derailing every discussion – often at great length – until everyone else is bored to tears.

    And yes, online communities will generally have weaker ties. One consequence of which is that pissing off the rest of the community or offending people is lower cost than it would be in the real world.

    I’m not sure what the answers to these problems are. Partly, I think our social norms and habits haven’t quite caught up with the online environment. Although I recently discovered the concept of hellbanning, which I now wish we could do in real life too…

  9. Pingback: Twitter is the conference pub « Alexander Brown .info

  10. Hi, No comments from me but you might want to look at :

    Papacharissi, Zia (2002) “The Virtual Sphere: the internet as a public sphere”. New Media & Society 4(9). Perhaps more of a big picture but I think its interesting. I have it if you cant get it.

    Delborne, J. et al (2011) “Virtual deliberation? Prospects and challenges for integrating the Internet in consensus conferences”. Public Understanding of Science 20(3):367-384. specifically about consensus conferences (clue’s in the name) so has a specific purpose and is handling large numbers of participants, you can just read the last paragraph but they pull their punches to be honest (IMO). I have it if you want it.

    Michinov & Michinov (2008) Face-to-face contact at the midpoint of an online collaboration:its impact on the patterns of participation, interaction, affect, and behaviour over time” Computers & Education 60(4).
    Dont have this but can try tomorrow. Its classroom based but is online and looks at the process in the kind of detail you’re after – uses exciting phrases like “emotional regulation”. maybe useful to see hwo it cites/who has cited it?

    Good luck! When someone introduces a law for online anonymity to be compulsory you can sign me up :o)

    • I meant outlawed, not compulsory!!

      • Cheers Bev, that’s really useful. And yes, please send me the PDFs!

        I seem to have just given myself a shedload of reading to do by asking this question…

        Anonymity is an interesting question. There are obviously some situations where anonymity is a good thing (whistleblowers, people discussing their experience of domestic violence, etc). But I can’t help feeling it’s a bit like freedom of the press – when it’s threatened, people pull out these worthy examples of why it’s important. But the vast majority of the time it’s actually used to bully, harrass, print intrusive pictures of people’s knickers… and not for any worthy reasons at all.

  11. Hi Sophia
    this is one of the central questions of eDemocracy and communication studies and there is more literature on this than 100 PhD students could hope to read.
    It’s a subject much discussed on the Association of Internet Researchers’ (AIR) email list and you can search archives.
    Recent threads on this subject have been “physical/digital worlds” (use as a search term).
    You might also want to look at the recent Onlife project: (I haven’t looked, but it’s one the latest resources.
    So -a lot of research done, but not many conclusions.
    My own research (so far) indicates that different people are comfortable with different media and topics.
    When people discuss e.g. political topics offline, they do so in a relatively free way -with value statements interspersed with news/factoids. They also seem to do this in relatively short bursts (unless it’s part of their job).
    Plus, the idea that we base our opinions on rational deliberation is exploded. (You might as well suggest the government base policies on evidence!)
    An online discussion makes the discussion process more difficult and boring -many people are unsure of their writing and deliberation skills in a public environment (surprising people). You have the time to check, improve, delete, try to be more informed, cleverer.. Plus following a long thread is time consuming and boring without the energy of shared value statements (which attract trolls online).
    I think we really have to look beyond text threads and get over ourselves about rational deliberation if we want to involve people online.

  12. Cheers Ben, yep, I think the Watercooler is a great example of an online campaign/discussion space making a difference to real world practices. Now you just need to set your sights on workfare…

    And yes, hellbanning = awesome:-)

  13. This post was crossposted on the Demsoc blog too, where it got the following comment, also containing some useful sounding papers:-

    Hi Sophia,

    You’re probably already familiar with these, but worth checking out the Kettering Foundation, and a couple of their papers. In particular:

    Framing for Democracy: Exploring the Impacts of Adversarial and Deliberative Framing (2012), Understanding the Longer-Term Benefits of Deliberation


    The Promise and Problems of Online Deliberation (2011)


    You can see all comments on this post here:

  14. I know it’s not exactly a conversation tool but David Price the creator of the site, which is all about trying to get nuanced dialogue, has thought long and hard about how to structure their interface for constructive debate without getting hijacked by nutcases. He may be worth talking to.

  15. Hi Sophia,
    Unlike some other commenters I don’t have useful studies to link to, but rather two “experience based” ideas about why Twitter is particularly bad for constructive debate. I cut my teeth in the ’90s on bulletin boards, where everyone was routinely anonymous, and it’s quite noticeable how debates don’t work on twitter that do on discussion threads. Length is an obvious reason, but over and above that

    1. It’s impossible to effectively ‘follow’ a conversation on twitter, once more than two or three people are involved. Not only does this mean you miss things, but it also means you can’t read the tone of a debate, which leads to a lot of misunderstandings.

    For example, there have been plenty of times when I’ve been voicing Pro-X opinions, and been surprised by fervent and abusive Anti-X replies, which seem entirely out of proportion & uncalled for.
    What I *don’t* see is the fact that elsewhere or at another time these same Anti-X people have had bruising encounters with other Pro-X people. Obviously, the failure of our communication is partly their own fault for assuming that I’m somehow associated with all the people who express a similar view, and also their fault for being defensive – but it means it’s easy for me in turn to become defensive about Anti-X people, who I assume are aggressive and rude because that’s been my only experience.

    2. There are huge power imbalances on Twitter.

    For example, when I say “Anti-X people are rubbish”, that will be read by just over 1000 people, if that. When a popular twitter user says “HPS_Vanessa is horrible because she’s Anti-X”, that might be read by tens or even hundreds of thousands of people, many of whom might want to tell me their opinion of me in turn. Because of the huge imbalance of influence, or range/loudness of voice, I think the more followers you have, the thicker you should grow your skin – but sadly that does not seem to happen!

    I realise these two problems aren’t unique to twitter, but I think they contribute a great deal to the unnecessary unpleasantness of the medium.

    • Hi Vanessa, I know what you mean about following conversations. In a way that doesn’t happen on message boards, people can see one tweet, but be unaware of the earlier conversation leading up to it. Out of context, they can take the tweet in a very different way.

      And I know what you mean about the pro-X, anti-X thing. People can hear a criticism of something, which is meant in one way, as being meant in a much larger way because of another argument they are used to having. But I don’t think that is unique to twitter. Maybe things escalate quicker on twitter though because of the combination of impersonality and short messages.

      With power imbalances, I see your point and I agree people shouldn’t abuse their power. But I also try to remember that people with a high profile are probably getting a lot more shit than I am from random strangers. I’ve had a look at the @mentions of well-known people (especially women) and been horrified at what they sometimes have to put up with. It must be hard to keep your cool.

      I don’t know what the answers to any of this stuff are, but I wish I did.

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 )

Google+ photo

You are commenting using your Google+ 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