The meaning of silence in an online context

May 13, 2009

I’ve been pondering a few of the discussions I’ve seen on online versus offline research approaches:  Zebrabites’ thoughts on the need to understand users in person as well as hypertext;  Matt Rhodes of Freshnetworks’ recommendation that we understand people in natural online communities as well as specially-constructed research communities.   People share themselves and their opinions differently in different contexts.

We tend to treat an online conversation (on a message board, a blog reply, a social network) as an honest record, right there on the screen, of what Audience A thinks of Topic B.  Or, at least, we might elaborate, what Willing Contributors from Audience A think about Topic B.   As Katie and Matt suggest, it’s not quite that simple.

Why wouldn’t people say what they think, in an online network? Why might they keep quiet?  In my view, social pressures shape our conversations online just as much as they do in everyday life.  Indeed, some of those pressures to behave may be even stronger online.

The other day, I touched on discourse or conversation analysis, where you analyse a given conversation and you look at the way the participants frame their discussion.   You also look, as far as you can, at what’s missing.

What is the meaning of silence, online? (Or: have I gone mad in trying to express what I mean?)

Let me try and explain.  Online, we have talk (the posted conversation which we can analyse to our heart’s content) and we have silence (the large majority who read but don’t contribute).   The posted conversation may not be an accurate reflection of the whole, in the same way that conversations in a group discussion may not always be good representations of individual opinions.

What silences people who might otherwise contribute? Here are my top three.

The silencing power of strong opinions

If I read a blog with lots of passionate contributions, my decision to contribute will depend on my viewpoint and how I feel about throwing it into the conversation.   If I dislike loud online argument, and those who have already commented are arguing passionately with each other, I may choose not to contribute my dissenting comment.

Another example: a couple of years ago, I helped troubleshoot an online community that wasn’t working too well.  It was pretty unmoderated: new people joined but they rarely participated.  It turned out that the community was dominated by a small group of older, vociferous and rather right-wing users.   New users arrived, took one look at some of the ‘hanging’s-too-good-for-them’ rants, and ran away.    They didn’t feel welcome.

Perceived social disapproval

I may choose not to talk about something that actually excites me because I suspect that other people will be negative about it.   To take a minor example, I use the blog network, Livejournal, and I’ve recently become a convert to Twitter.  People on LJ are routinely scathing about Twitter.   When I first started using Twitter seriously, I didn’t mention it on LJ.   Eventually, I broke down and talked about it, and various of my LJ mates broke out and said ‘Ooh, yes, it’s great, I have one too and here it is! ‘

I would never have known if I hadn’t asked.

Online conversation persistence

In stable online networks, our conversations can be very, very easy to follow.   As a result it can be very hard to gossip: nothing is hidden.  In real life we can sit through a boring presentation and then have a whispered chat with our neighbour about how dull it was, before putting on a bright smile and congratulating the speaker as they breeze up to us and ask what we thought.

If we thought that the entire conversation with the neighbour would be retrievable later by the speaker, we might act rather differently. Online, a third party may be able to do exactly that.

The persistence of online conversations also leads us to be careful about expressing our opinion of someone who doesn’t appear to be present.   They might come along in 3 hours or 3 days (or 3 years), and get upset.

‘The lurkers are supporting me in email!’

They might be.

Silence is not necessarily agreement.  It can be curiosity, boredom, or dissent, at the very least.  All of us are lurkers (readers), as well as participants.  Sometimes, we watch the debate and we don’t join in publicly.

If a counter-opinion can’t be expressed openly for fear of social consequences, back channels can help participants let off steam.  Sites like PostSecret help some people express socially unacceptable views.   Private messaging options and quick polls (the kind where you can’t see what everyone else put) can make it safer to express yourself.

You can do all sorts of things to make a network a safe space in which to express opinions.   Ultimately, though, you might need to triangulate your findings by mixing methods: networks and one-to-one, online and in-person.    Lifting an online conversation straight from a ‘natural’ network  may lead you to some highly unreliable conclusions.

Thoughts? Have I managed to express myself on this one?

Email This Page

{ 22 comments… read them below or add one }

John Griffiths May 13, 2009 at 1:36 pm

this is very close to your ruminations about selfcensorship on twitter. the online self is a presented self. What we instinctively admire is when what we do feels spontaneous and unforced. When usually it is anything but.

Qual research trades a lot on the betrayed self – this is where discourse analysis excels. the pauses, changes in direction and downright evasions as respondents dodge the elephant in the room despite the best efforts of the moderator – clearly revealing the elephant as outline as they do so. The internet has a big problem in portraying the betrayed or involuntary self. Because you can’t see how many times I have retyped this line or fixed the spelling or changed a word.

The first law of forensics is that a crime always leaves a trace. And researchers are expert in spotting the traces. I like to spot why respondents are participating in research. They almost always find a way to tell me (and its never just about the money). But when I have told a partial truth online I suspect my compensatory mechanism may be offline, offscreen so you never see it. If the silences are offline I really don’t think you’ll find them.

I have one suggestion for finding silences. And that is through soliloquy. Long used by dramatists to reveal and celebrate the inconsistencies and selfblindness of characters – the confessional character of blogs and bulletin boards it seems to me to allow room for silences I am stepping around as I try to present an authentic online self. Ask them a tricky question – and give them plenty of space and time to answer it.


Alison May 13, 2009 at 3:13 pm

This is a great reply – I love the description of traces and partial truths. I think the online self is usually a highly edited one – I think there’s a concept somewhere of ‘performing’ yourself online.

I’ve been ruminating about this for a while – recently I realised I was watching a discussion in one community, participating very heavily through reading the debate and talking about it to people in other spaces – but I did not comment on the original discussion itself because I thought my comments could easily get a very hostile response.

Yes, I do think that the ideal is simply to get people talking. The more people talk, the more they’re likely to reveal, especially if you can show that you genuinely support different views.

We don’t think about the respondent experience very much in online qual or any other type of qual, but the pressure on an online participant is pretty heavy. Hah, I will save that for my post on user experience….:)


Tom May 13, 2009 at 2:12 pm

Excellent comment John – I’d snip that and make it a post if I were you ;)

Yes, this is the lurking problem: another factor is that people feel the strongest contributors have already said everything there is to say and are happy to ‘sit back’.

On the LiveJournal pop community I co-moderate we have a poll every week where people can tick any new chart entries they think are good. The most vociferous commenters are into dance and R&B but there’s a strong silent majority more into indie. Here the combination of poll and comments creates an outline of an elephant – a level of friction within the community about what makes for ‘good pop’


Alison May 13, 2009 at 3:17 pm

Yep, I’ve seen it solved to some extent by remuneration policies on online research comms, which means you have to talk or you won’t get paid.

All sorts of things inhibit people. I’ve always thought that those elaborate ‘seniority’ systems that you see on certain message boards were terrifying – you know, the ones that say ‘Grand Poohbah, 15,000 comments, online since 1929.’

Polls are good. People just trot by and click them.

I like the idea of the elephant’s outline slowly being sketched…


Tom May 14, 2009 at 12:48 am

In general the pro community management types are way ahead of us here – they know all the tricks for getting people to contribute. (And most of them would be horrified by the very IDEA of remuneration – or so they’d claim ;) )

Agree though that hierarchy systems are really flawed – this is what my “Red Shift” part 2 is going to be on, if I get the time to write it!


Katie Harris May 14, 2009 at 11:14 am

Great post Alison!

Tom; “The pro community management types are way ahead of us here, they know all the tricks for getting people to contribute”…

Interesting. But not sure I think it’s a good thing.

(Warning, about to go off on a bit of a tangent here).

Notwithstanding that participants getting to know each other/developing relationships with each other seems to be one of the key selling points for ongoing online market research communities, I’m not convinced about the utility of this dynamic in terms of the output.

Why? Because when participants feel they know each other and/or they know you (the moderator), there’s far more to lose through being honest.

These relationships mediate and shape responses; they bring a whole new dynamic into play. And they add yet another confounding variable to the analysis, eg did X say that because Y had a go at him/her on the last post? etc.

Bulletin board focus groups are better. Not only is it easier, from a moderator’s point of view, to keep an eye on the interactions, participants don’t have time to ‘bond’ in the same way. When they’re all strangers, and know that they’ll remain so for the duration of the research, it’s much easier to speak from the heart.

Collaborate? Yes. Discuss and debate things? Sure. But without the emotional attachment.

I’ve found that a short burst of relatively anonymous interaction – as is possible in a BBFG – can work wonders to allow people to express themselves without the usual (or unusual) online constraints.

This kind of raw dialogue gives me far more insight than a discussion between old mates or ‘crowd polished’ consensus.


Tom May 14, 2009 at 12:51 pm

I think we’re back at the idea of a dualism between performed/social reality and something more ‘authentic’ or ‘honest’ hiding underneath it – which I kind of reject (I think the two are in collaboration not opposition) and I certainly reject the idea that researching the ‘authentic’ is more important than researching the performed. Maybe I’m in Alison’s “critical thinking” basket :)

I totally agree we should think twice before wholeheartedly adopting community manager practices though – the aims of a community manager aren’t the same (and might be at odds with) the aims of a researcher. We can learn a lot from them but as you say we need to keep the effects they’re aiming for in mind.

From the point of view of generating interesting material I think the best formula is a balance between people who know each other and are relaxed enough to chat, and people who are prepared to come in and disrupt or divert that conversation. But my actual recommendation would completely depend on the topic area I suspect!


Katie Harris May 14, 2009 at 1:33 pm

You’re right; it depends on the topic area and variable/s of interest.

Sometimes researching ‘the performed’ is the objective. If you’re researching the dynamics of an online community over time, then that’s the perfect forum for it!

; )

On the critical thinking; well kind of, but not quite what I was getting at. It’s very subtle, but I think there’s a difference between something more authentic or honest ‘hiding underneath’ vs something authentic and honest being *inhibited* by the environment itself


Alison May 14, 2009 at 3:47 pm

Mmm, interesting.

I think the chief obstacle I see both online and offline is the (British?) tendency of politeness to stifle honest dissent, whether it’s politeness towards the moderator, the sponsor, or to each other.

The social contract you create, in inviting or rewarding participation, means that people will often try to please you.

Of course, online this is somewhat countered by the Mr Angry tendency… ‘this is all very well but you’re all completely £$%^ missing the point!!1′


Alison May 14, 2009 at 3:40 pm

Interesting – I think this might lead to my next post, on forming storming and norming in an online context.

I think therecould well be a kind of golden period between the uncertainty of contributing your first few posts on a community, and the stasis that sets in once you all know each other. (Community set-up can mean making sure there is a sweet spot in the first place – content there that makes people feel welcome, moderating policies that respond positively to different kind of opinions.

On the one hand you don’t want the kind of hell that you get on newspaper comment sites or certain tech blogs; on the other, you don’t want it so cosy and settled that no one will say anything different. I suspect it takes a while to get to that place.

Of course, face to face groups can be even worse in terms of suppressing opinion, depending on the characteristics of the dominant types; but at least there, your skilled moderation can open things up.


koganbot May 15, 2009 at 1:32 am

Commenting for my first time. Wish I could say I’d been lurking for months and was finally taking the plunge, but actually I followed Tom’s link in. Have never done systematic research in my life, but have been part of message boards and online communities and their pre-Internet fanzine equivalents going back to the ’80s, most of them with at least a nominal focus on music. Almost all of these allowed anyone to play, and one thing that’s been consistent over the years is that far more guys than gals actually do so, actually speak up.

Also, I’ve done a little bit of volunteer work with teenagers in groups, and my experience is that the girls will tell you how they feel and what they believe (how honestly they’ll tell you is a different matter, but they’ll speak in the language of feeling and belief), whereas boys will put on a show, tell stories, jokes, try to impress you, defy you, etc. The show can tell you a lot, but what it tells won’t be along the lines of a direct response to a question concerning what they feel or think about something. And I assume (though I don’t know) that a girl’s response to, e.g., “would you be likely to listen to a radio station that plays this type of song?” would have more predictive value than a boy’s.

When they’re all strangers, and know that they’ll remain so for the duration of the research, it’s much easier to speak from the heart.

I’d expect that “speaking from the heart” is a rhetorical device that’s more common among females than males, but also may not occur among teens in the same frequency or infrequency as among adults (and the makeup of the group – the mixture – may make a huge difference). And of course it will differ from person to person. But I wonder if there’s a good way to test your generalization about interaction among strangers versus interaction among those who have bonds. It may make sense that relative strangers will think that there’s less at stake, and less to lose from conflict, and in that sense less to lose from saying something from the heart that others will disagree with. But there’s also less to lose from getting caught in a lie, or from sarcasm and taunting and bullying. And there’s equally less to gain from, e.g., speaking your truth if you’re not speaking to people who will potentially care about you one way or another and whom you may want to know your truth, and less motive to rethink your ideas or to explore what really may be on your mind as opposed to going with the first thought that pops up. And there’s less motive to draw out what may be on someone else’s mind. That is, I’d expect that the risks of the heart would be less but the gains of the heart would also be less. But again, how would we test this?

Also, the results may depend on what you want to know. It’s one thing to ask “Are you likely to listen to a station that plays this song?”; it’s another to ask, “Why do you hate the Jonas Brothers?” The respondent is more likely to know the answer to the first than the second (though I find the second a million times more interesting). In any event, my own experience online is that anonymity doesn’t lead to honesty (and also, fwiw, I don’t think I know myself better when I’m alone than when I’m among others). Here’s a link to a comment thread on the question “Busted Versus The Jonas Brothers” where only two of the participants (me and skyecaptain) had any contact with each other online or off. Most of the rest are utterly anonymous, people who’d googled there way onto my lj and had no bonds of acquaintance or friendship to enhance or destroy. We can certainly learn plenty from reading it, but that doesn’t make the answers from the heart, or connect them much to honesty.


Alison May 15, 2009 at 9:02 am

the girls will tell you how they feel and what they believe, whereas boys will put on a show, tell stories, jokes, try to impress you, defy you, etc

Oh, I love that.

Anonymity and honesty. I don’t know. My part of Livejournal is heavily female-dominated and there’s an ongoing discussion of what’s called ‘The Cult of Nice’ where people in a fixed friendship network get attacked if they act critically towards other people (work that one out…) The relatively stable relationship mean that over time it becomes difficult to talk about some topics in public (this may be a female thing of reluctance to offend) YET equally, amongst good online friends or relatively trusted strangers (as in the comments here), you get some terrific (from-the-heart?) conversations happening.

Busted versus the Jonas Brother. The Jonas Brothers are in 3D!! Yes, you can see those comments in all sorts of ways and I like the way they define the bands’ appeal.

(There’s a whole other think about performance and Netspeak and placatory exclamation marks but I won’t go there…)

I’m also struck by the differences between talking one-to-one and talking in a group: different things come out, different conclusions could be drawn. Which one is the one to base a judgement on? That depends.


koganbot May 16, 2009 at 1:57 pm

There seem to be a few separate issues here: (1) whether someone speaks at all, (2) whether he or she speaks honestly, and I’d add a third, (3) the usefulness of what someone says – e.g., it’s not always certain what honesty means: sincere doesn’t always mean truthful, and truthful doesn’t always mean right: if someone says sincerely that she intends to do something, but doesn’t do it, how useful is her report to you?

Doing all this as a thought experiment, I don’t get a clear sense of whether knowing the people you’re talking to better would encourage one or discourage one. As I suggested, boys are more likely to speak but less likely to speak on topic or from the heart.

But if you don’t know the people you’re speaking to, you may be less likely to speak because you don’t know what will go over and what will offend or even what might just be considered irrelevant, and this could well override the fact that you’re less likely to care whether you offend. (But also, you’re also less likely to know the consequences of offending.) Whereas with people you know well, you’re more likely to speak because, among other things, you already know what you can get away with saying, and what not to say. But also, I’m suggesting, speaking with people you know, there may be crucial instances where you’re more willing to risk conflict than with relative strangers, for the very reason that how you behave may have a long-term affect on the relationship, and so the fight my be worth risking, for the sake of a better relationship, and also for the sake of letting these people know you better. So what I’m saying here, since I’m speculating (though on the basis of my experience) I can come up with any old answer, as to whether knowing people or not knowing them inhibits or encourages the response, and how it affects the nature of the response.

Knowing you only from reading a few of your posts, I nonetheless know you better than I know “the general reader,” or whoever happens to be reading my reviews and articles., and my froze is more likely to freeze and go dull there (i.e., when writing articles and reviews) than here. That’s just my experience. But if what I knew about you were different, perhaps I’d be writing differently.

And in regard to freezing, “speaking from the heart” or “speaking with honesty” are genres, and someone may freeze where this seems to be the requirement, or limit his or her answers to appropriate comments. Whereas when just fucking around, that person may end up revealing a lot more, simply because it doesn’t seem to count, and that person not worry about being held to account.


koganbot May 16, 2009 at 1:59 pm

my froze is more likely to freeze

as is my prose, which is what I meant to say.

Alison May 20, 2009 at 12:34 pm

I meant to reply to this last week if only to say that this particular post had provoked so many throughtful responses that I’d be returning to them again and again. There is so much to complicate this, including the nature of the subject being discussed, and the place or set up of the discussion.

For someone setting up an ‘artificial’ community, the main concern tends to be ‘getting any response whatsover’. There are ways and ways, I feel, but the typical comment conversation on one site may be disparate yet harmonious while the conversation on another (let’s say a newspaper comment site) might be more like a bear pit. Perhaps it comes down to knowing a bit about who you’re talking to
and what kind of conversation you want to create.

koganbot May 16, 2009 at 2:04 pm

Also, my guess is where people posting here are similar to one another but different from the average lurker is that we think it’s more important that an idea be put forth for discussion than that we know for sure that it’s right. My experience is that the opinionated person falls into two categories (or maybe the same person can fall into two modes, depending on the situation), one trying to ram his opinions down everyone’s throat, the other trying to put forth opinions in order to test them but being very willing to change. Often the most vociferous and contentious are in the latter category, and what’s at first intimidating about them becomes inviting, when you realize that they want your response and are willing to modify their own views in response to the new information or ideas you provide.


Katie Harris May 15, 2009 at 9:51 am

Hmmm…maybe instead of “speak from the heart”, I should have said “speak without worrying too much about offending someone else” or to Alison’s point, speak at all, rather than keep quiet for fear of offending someone else.


Alison May 14, 2009 at 3:57 pm

Oh, Wordpress. Look at all the pretty nested comments.


richshaw May 14, 2009 at 9:06 pm

Great post Alison, really got me thinking about why people choose not to post. My thoughts echo those of Toms. It seems a lot of the time we’re trying to dig down past what people portray and discover some great truth by uncovering the authentic or real self.

However I’m not sure I want this real self. The betrayed or performed self which we use when we’re online, with our friends, in the supermarket etc surely has just as much value as anything else. After all we’re social animals most of our behavior is influenced by other people by some extent and sometimes we will want to take in account that social influence when we draw our conclusions about peoples behavior. (Then again it’s been awhile since I’ve read any research on the concept of self so I could be chatting rubbish).

The other interesting thing is how online conversations change when we alter the context or parameters in which they take place. It’s quite interesting to see how conversations change between members of a forum or community when you increase or decrease the number of members, and how conversations on open and private forums can differ. Conversations on IM are also interesting because IM encourages an immediate response and it’s also relatively private when compared to things like twitter.

As well as altering the relative privacy and immediacy of conversations and seeing how they effect whether people post, you can also encourage people to post by trying creating a community where people feel they can have an opinion, sometimes I post and opposing opinion to see if anyone agrees with that viewpoint (not an ideal solution, but sometimes it’s fun to play devils advocate). Have you seen the ‘I like’ thing on facebook next to peoples status updates? I think that is quite good at encouraging people who would otherwise be silent.

Hmmmmm I think I could mummble about this stuff for ages, time for a cup of tea.


Alison May 15, 2009 at 2:15 pm

So many interesting points in there. I think there is an inhibited self, for certain purposes and in certain contexts – I remember doing groups last year which were dominated by Iphone users or Blackberry users, and users of more standard handsets were really inhibited by the ‘cool kids’…but equally, I think we often have different personas, whether this simply a work identity versus a private identity, or a whole host of slightly different versions of ourselves that we act at home, at work, with our mates etc etc.

I like the distinction between private/public. I’m very fond of qualitative phone interviews and i’ve done a few IM ones and both of them have similar characteristics: they really tap private thought and people can be extremely revealing in them. Sometimes it’s because the IM or phone stream-of-consciousness encourages a daydreamy, semi-therapeutic conversation. Social context is still important though: people might badmouth the Iphone, for example, but want it desperately for its geek status.

Silent, simple add-ons: yes, like those.

Could also ramble for ages, have lots more to think about so thank you!


diana September 5, 2009 at 12:39 pm

As a seasoned lurker, who rarely comments or joins in on discussions, I have sometimes wondered what it is that stops me, so it was very interesting to read this analysis of silence. All the reasons rang bells for me and so I thought I’d break the habit. (Although I nearly didn’t, because the other comments are far more thought through and interesting than this!)

p.s. I particularly enjoyed the phrase ‘placatory exclamation marks’ !


Peter Singleton February 1, 2012 at 6:17 pm

Other reason for silence is that it might simply take too long to compose a response that made sense and cover the point; sometimes you can think of a quick response, but realise it is very unlikely to have any effect at all, so why bother – or am I showing my economics background rather than applied psychology?



Leave a Comment

{ 1 trackback }

Previous post:

Next post: