Web breaks echo-chambers, or, ‘Echo-chamber’ is just a derogatory term for ‘community’ – my remarks at #AAASmtg

As you probably know, I was in D.C. last week, attending the annual AAAS meeting. This was my second one (funny, back when I was a member of AAAS I was still in grad school and I could never afford to go – now that I am out of science, invitations are finally happening). It is an enormous meeting (about 8200 people this year, I hear) and I missed even seeing some of the friends as the space was so enormous and the program so rich.

Unlike last year, when I was in a session that made quite a splash, this year I was a part of a much more academic panel on Social Networks and Sustainability.

Organized by Thomas Dietz of Michigan State University, the panelists were Mrill Ingram (University of Wisconsin), Ken Frank (Michigan State University) and Adam D. Henry (West Virginia University). These are people from areas like sociology, people who make graphs like this one and understand how to properly interpret it:

My role on the panel was as a ‘discussant’, i.e., someone who does not give a separate talk but comments, at the end, on what the other panelists have said.

I am glad I got the materials from the panelists in advance as this was quite dense stuff.

Every scientific discipline invents new words – the terminology (or jargon) with precise meaning that is necessary for practitioners to talk to each other. For the most part, natural sciences tend to stick to agreed definitions, and counter-examples are relatively rare thus usually quite well known (e.g., the different use of the term “gene” by population geneticists vs. molecular geneticists).

Social sciences, on the other hand, tend to appropriate words from the existing English vocabulary and give those words new, precise definitions. Thus, possibility of misunderstanding by non-experts is greater. Also, some of the terms are defined differently by different sub-disciplines, research communities or even individuals, which makes it even harder to be sure one got the meaning correctly.

This all made reading the materials, as well as listening to the panel, quite challenging for me, the outsider in this field. I am also not a researcher of social networks – I am a user and observer, perhaps an amateur student of them. My thoughts could not be supported by numbers and graphs, but had to, by necessity, be more impressionistic – what I learned from my experiences using, living in, and running online social communities.

As all the speakers went substantially over their allotted times all I had left was seven minutes. Fortunately for me, I had all seven (not 3.5) as the other discussant’s flight into D.C. was canceled. Also fortunately for me, this was the very last time-slot of the meeting, so nobody was in a rush to go to another session and thus everyone let me talk a few minutes longer and then remained in the room asking even more questions.

As I tend to do, and in this case particularly, I decided not to prepare too much (OK, at all) in advance. Instead, I listened to the panelists carefully and made the decision what to say only once I climbed onto the podium in the end and knew how much time I had at my disposal. I decided what to say in the first couple of sentences – the rest came out on its own, pure improvisational theater.

As I was reading the materials and listening to the talks, I realized that a couple of examples were clearly discussing real-world, meat-space, offline social networks, but that all the other examples were ambiguous: I could not figure out if those were online, offline, or combined/hybrid social networks.

So, I decided to use my seven minutes to compare and contrast online and offline social networks, how they differ (more important than how they are similar, which is the default thinking), and how they interact and potentially strengthen each other due to such differences.

This is, roughly, what I said – or at least what I meant to say but had to speed up, i.e., this is an (very) expanded version:

Social norms build and enforce echo-chambers

You want to remain in a friendly relationship with the people you see (or potentially can see) often: neighbors, family, colleagues and friends. Nothing makes for a more unpleasant interaction than discussion of politics, ideology or religion with the people you disagree with.

Thus, there is a social norm in place: politics and religion are taboo topics in conversation. It is considered bad manners to start such conversations in polite company.

This means that most people are not exposed to views other than their own in their day-to-day interactions with other people.

In a small tightly-knit community where everyone’s politics and religion are the same (and people tend to move to such places in order to feel comfortable, on top of most likely being born in such a community to begin with), there is no need to discuss these topics as everyone already agrees. If the topic is discussed, there are no other opinions to be heard – it’s just back-slapping and commiserating about the evil enemies out there.

In mixed communities, the taboo against discussing politics and religion is strongly enforced. Again, as a result, there is not much chance to hear differing opinions.

There is no more airtight echo-chamber than a small community which interacts predominantly within itself, and not so much with the outside world.

Mass media builds and enforces echo-chambers

If you are born and raised by parents with a particular set of beliefs, you will also inherit from them the notions of which media outlets are trustworthy. If you were raised in the reality-based community, you are unlikely to waste much time with the media of the fantasy-based community (and vice versa). If your parents read Washington Post, you are unlikely to read Washington Times. You’ll prefer New York Times and not New York Post. MSNBC rather than Fox News. NPR rather than Limbaugh show on the radio.

But it is even worse than that – the choice is really not as broad. The media shapes the public opinion by choosing what is and what is not respectable opinion, i.e., ‘sphere of legitimate debate’ – what opinions to cover as serious, what opinions to denigrate and what opinions to ignore. There are many ideas that people hold that you will never see even mentioned in the US mass media and some of those are actually very legitimate in the Real World.

Furthermore, the press then divides the ‘respectable opinion’ into two opposites, gives voice to each of the two, and will never actually tell you which of the two is more reasonable than the other – “we report, you decide”, aka, He Said She Said journalism.

By presenting every issue as a battle between two extremes (and the fuzzy, undefinable “middle” is reserved only for them, the wise men), the mainstream press makes every opinion something to be sneered at, both those they deem worthy of mentioning and the unmentionable ones.

By refusing to acknowledge the existence of many stands on any issue, by refusing to assign Truth-values to any, by looking down at anyone who holds any opinion that is not their own, the mainstream press fosters the atmosphere of a bipolar world in which enmity rules, and the wagons need to be circled – the atmosphere that is so conducive to formation and defense of echo-chambers and yet so devoid of airing of any alternatives.

The Web breaks echo-chambers

When an individual first goes online, the usual reaction is shock! There are people in the world who believe what!?!?

The usual first response is anger and strenuous attempts at countering all other ideas and pushing one’s own.

But after a while, unbeknown to the person, all those various novel ideas start seeping in. One is not even aware of changing one’s own mind from one year to the next. Many ideas take time to process and digest and may quietly get incorporated into one’s gradually enriching and more sophisticated worldview.

We all learn from encountering all those other opinions even if we vehemently disagree with them. And we cannot help bumping into them all the time. There are no taboo topics online, no social norms preventing people from saying exactly what they think.

Forming, finding or defending a vacuum-sealed echo-chamber online is extremely difficult, if at all possible.

Your Facebook friends will post stuff that reveals their politics is different than yours (and you did not even know that about them before – they seemed so nice in real life!). By the time you get around to blocking them…it’s too late – the virus has already entered your head [this one sentence added 2-27-11].

People you follow on Twitter because of some common interest (e.g., food or knitting or parenting or technology or geographic area) may be very different from you when concerning some other interest, e.g., religion, and will occasionally post links to articles that contain opinions you have never heard of before.

If you are, for example, a liberal and tend to read only liberal blogs, you will constantly see links to conservative sites that are being debunked by your favourite bloggers – thus you will be exposed to conservative ideas daily.

If your interest is science, you are even luckier. The mainstream media, if it links to anything at all, tends to link either to each other or to governmental sources (e.g., CDC, USDA, etc.). Political bloggers link a lot more, but again the spectrum of sources is pretty narrow – they link to MSM, to governmental pages, and to each other (including the “opposition” bloggers).

But science bloggers link to a vastly broader gamut of sources. If mass media is linked to at all, it is usually in order to show how bad the coverage was of a science story. Linking to each other is important (and that includes linking to anti-science sites when needed to counter them), but what science bloggers do that others do not is link to scientific papers, documents, databases, even raw data-sets (including some Open Notebook Science bloggers who pipe data straight from their lab equipment onto the web).

What echo-chamber? Contrary to what some uninformed op-eds in the mass media like to say, the Web breaks echo-chambers that the social norms and mass media have previously built.

The online and offline social networks can work synergistically to affect real change

Many curmudgeons like to say that the Web does not do anything on its own. They (unlike behavioral biologists) do not understand the distinction between Proximal Causes and Ultimate Causes. Web is a tool that allows, among other things, many more people in much shorter time to organize to do something useful in the real world.

Release of Tripoli 6 was an instance in which massive outpouring of support online forced the mainstream media to cover the story which then forced the hand of politicians to do something.

Likewise, in the case of resignation of George Deutsch from NASA, it was investigative work by a blogger, Nick Anthis, that energized the blogosphere, which pushed the MSM to finally report on the story, which forced the event to happen.

PRISM was an astroturf website built to counter the pro-open-access NIH bill in the US Senate. Outpouring of online anger at the tactics by the publishers’ lobby inundated the senatorial offices – as a result the bill passed not once, but twice (GW Bush vetoed the first version of the large omnibus bill it was a part of, then signed it with no changes in the language on this particular issue) and the Senate is now educated on this issue.

But probably the best example is the Dover Trial (Kitzmiller v. Dover Area School District) that made Intelligent Design illegal to teach in US public schools. The ruling by Judge Jones (pdf) is one of the most powerful texts in the history of judicial decisions I am aware of.

There are anti-evolution bills popping up somewhere in the country seemingly every week. But because of the Dover ruling, they are all illegal. Most don’t make it to the committee, let alone to the floor of the state legislatures. Others are soundly defeated.

Before Dover, both Creationist sites and pro-evolution sites, when linking to me, would bring approximately the same amount of traffic to my blog. After Dover, getting a link from PZ Myers, Richard Dawkins, Larry Moran or Jerry Coyne brings substantial new traffic. Links from Creationist sites? Essentially undetectable by traffic trackers – I discover them only when I search my blog URL to specifically see if there are new links out there. Creationism, while still popular with the people, is politically essentially dead. The Dover ruling castrated it.

But Dover Trial would not have gone that way, and would not result in such a gorgeously written document by the Judge, if it was not for a small army of bloggers who contribute to the blog Panda’s Thumb. A mix of scientists from different disciplines, lawyers, etc., this group has been online – first on Usenet, later on the blog – for a couple of decades before the trial.

This is a group of people who battled Creationists for many years, online and offline, in courtrooms and political campaigns, in classrooms and in print. They know all the characters, all the usual creationist “arguments” (and provided all the answers to them in one place), all the literature, etc.

It is one of them who discovered that the new Intelligent Design “textbook” is really just a reprint of an old Creationist book, in which the word “Creationists” was replaced by “Intelligent Design proponents” throughout the text….except in one place where they made a typo: “Cdesign proponentsists”.

Ooops – a huge piece of evidence that Intelligent Design Creationism is just a warmed-up version of the old-style Creationism masquerading as something new. The Panda’s Thumb bloggers were at the trial as expert witnesses who provided all the expert evidence that Judge Jones needed to make his decision. People who organized on the Web have helped a meatspace history come to pass.

The online and offline social networks can work synergistically if the ecology is right

When looking at the role of online communities and networks in meatspace events, counting the numbers of networked citizens (or ratio of networked to non-networked citizens) is not sufficient – one also needs to know their geographic distribution, and their connectiveness with non-networked citizens. The most fresh example are the so-called “Twitter revolutions” in the Arab world.

There are at least two possible scenarios (or thought experiments) that demonstrate the importance of ecological thinking about social networks:

1) There are 10 people on Twitter in a country. All in the same city, all in the same college dorm, good friends with each other. No communication with other people. No Twitterati in other cities. Nobody knows that other people in other cities have the same negative feelings toward the government.

2) There are 10 people on Twitter in a country. One each in 10 different cities. They communicate with each other via social networks continuously. Each is also a center of the local community of thousands of non-networked people using offline methods of communication. Through this connection, they become aware that there are millions of them, all over the country, and that a revolution is feasible.

In scenario 1, there are 10 buddies dreaming of revolution. In scenario 2, there are thousands of people in ten cities organizing revolution. In both, there are only 10 people on Twitter. Yet, the outcome is likely to be very different.

Thus, the ecology of the networkers, their spatial and temporal distribution, and their effectiveness in informing not just each other but many non-networked citizens, are important data one needs for this exercise.

‘Echo-chamber’ is just a derogatory term for ‘community’

I shamelessly stole this sub-heading from someone on Twitter (let me know who said it first if you know). Edit: Thank you – it was Chris Rowan,

A great example of a case where the Web produced a community (aka echo-chamber) but that was a good thing, is the case of American atheists.

Before the Web, each atheist in the USA thought he or she was the only one in the country. The social norms about the impoliteness of discussing religion, as well as the real fear of reprisals by the religious neighbors, made atheism completely invisible. No need to mention that the media never mentioned them – they were outside of the “sphere of legitimate debate”.

But then the Web happened, and people, often pseudonymously, revealed their religious doubts online. Suddenly they realized they are not alone – there are millions of atheists in the country, each closeted before, each openly so after! It is not a surprise that “no belief” is the fastest-growing self-description in questions about religion in various nation-wide polls and censuses.

President Bush Senior, himself not very religious, could say that atheists are not real American citizens. A decade later, his son GW Bush, himself a fundamentalist, could not say that any more – his speechwriters made sure he mentioned atheists in the listings of all the equally American religious groupings.

Not all online communities need to be politically active. Discovering people with the same interest in knitting is nice. Exchanging LOLcat pictures is fun. But such interactions also build ties that can be used for action in the real world if the need arises.

Without the Web, I would not know many people whose friendship I cherish. Without the Web I would not have this job. Without the Web, me and many of my friends would have never gone to a meeting like AAAS. There would be no such meetings as ScienceOnline, Science Online London, SciBarCamp, SciFoo, and others.

Every time I travel I make sure that people I know online – from blogs, Twitter, Facebook etc. – know I am traveling. I say on which date, at which time, I will be in which restaurant in which city. Twenty people show up. Most I have never met in real life before. But after sharing a meal, a beer, a handshake and a hug, our weak ties become strong ties. Superficial relationships become friendships. If there is a need to organize some real-world action – we can rely on each other to participate or help.

I have a separate Dunbar Number in each city I visited. And I try to connect them to each other even more than they are already connected via online communication. Which is one of the reasons we organize conferences and one of the reasons I am online all the time.


As Science Bloggers, Who Are We Really Writing For? by Emily Anthes.

Are science blogs stuck in an echo chamber? Chamber? Chamber? by Ed Yong.


14 responses to “Web breaks echo-chambers, or, ‘Echo-chamber’ is just a derogatory term for ‘community’ – my remarks at #AAASmtg

  1. I’m all for chalkenging the pejorative term ‘echo chamber’. Great post with a wonderfully warm and inspiring sign-off. Thanks for having taken the time to compose it so thoughtfully.

  2. I’m quite surprised the topic of echo-chambers apparently wasn’t one of the foci of the panelists. Also as a user and observer and maybe an amateur student, I’d actually claim the opposite: the web enables echo-chambers.

    We now have communities that didn’t even exist before the web or consisted of isolated individuals. Now we have flat-earthers, anti-vaxers, birthers and probably an endless list of similarly deranged delusions colluding somewhere in the web. These echo-chambers would not even exist without the web. If the web is supposed to break echo-chambers, how can that be? At least to some extent, the web must enable echo-chambers. I think the impression that the web breaks echo-chambers is an illusion, brought about by the less disputed and more problematic characteristic of the web: it breaks the mainstream. What you describe in the relevant section of this post as a breaking of echo-chambers is actually the breaking of the mainstream by delusions that rightfully have been isolated and quiet before the web.

    I think there are psychological reasons for the misconception in your post.
    Psychologically, most of us feel nervous when talking publicly. The degree of nervousness varies of course, but it probably starts when there are more then 5-10 people in the room, even if you know all of them, and it gets pretty tense when you talk in front of 200 strangers. Moreover, people’s average social network in meatspace rarely covers more then about 100 people, if that. So if you start interacting with a few thousand people online, you subconsciously assume that there must be something to this if such a huge number of people all say the same thing and reinforce your view. That’s the reason why the ‘otherwise nice’ people feel emboldened to now all of a sudden mention some of their outrageously silly views: they feel it’s ok, because thousands of others have the same opinion. Before the web, these people would not feel secure or confident enough to boldly state some of these silly views, for fear, rightfully, of ridicule.

    What needs to seep in, not just consciously, but much deeper, is that a few thousand people in a community on the net is nothing. That kind of number is so far to the right of the decimal, that nobody in their right mind would even start to consider it in any discussion that is supposed to concern the entire world – which is exactly the potential audience every social community could reach. What are a few thousand compared to billions? In today’s connected world anyone could get a few thousand followers ardently claiming the sky is green.

    There are a ton of ridiculous (and to be ridiculed) ideas floating around only because of the web. For the psychological reasons mentioned above, some of the people in the communities reinforcing these ideas become unpersuadables, meaning resistant to debunking, another phenomenon which was isolated before the web, but now seems to be spreading.

    Thus, I think, on the contrary, the web is breaking the mainstream and enabling echo-chambers. Being within an echo-chamber and then being exposed to opposing views only has the horizon-widening effect you described on people before they become unpersuadables. Afterwards, it only serves to strengthen the barricades of the unpersuadables against the consensus reality. Often enough, the outside world comes in too late (see, e.g., anti-vaxers).

    I’m not convinced, yet, that this may not just be a passing issue as the global population adapts to this technology. If this phenomenon is here to stay, though, the radicalization of some parts of the US communities we see now is nothing compared to what we will see in the future – and world-wide.

  3. I second the rebuttal. And lest anyone propose this rhetoric beast, let me add that the answer does surely not “lie in the middle.” It is a question of delienating the functional constraints of a medium and their impact on social behavior. Thus I find a more thorough examination of what constitutes an echo chamber paramount. Then we can hypothesize on cause and effect of phenomena like “unpersuadables” which need not be a threshold matter but could just as well be a continuum of scales of self confirmation bias and other biases.

  4. Bora,
    You had an interesting take on the proceedings and it brought into focus, for me at least, the need to broaden the study of online social networks since we tend to study the ones we have access to. Another takeaway is the necessity to keep both ANT and sociological explanations at hand to explain what we see.

    Other posters – For those that want to argue about “echo chamber” as a term, please refer to decisionmaking theory and not network analysis. Cass Sunstein and Yochai Benkler: http://web.mit.edu/comm-forum/forums/our_world_digitized.html

    Interpenetration is controlled if one network has a mandated trust relationship that can be represented by a hub topology. It can be enforced by mutual agreement initiated by an irrational fear that prevents members from engaging with the normal world. I fear I’m stepping off into speculation at this point, so I’ll quit.

    Second point – I got to listen to Gavin Schmidt! (coolness) The observations on “scient-i-fy-ing” (whatever!) political discourse by adopting science as what amounts to a bargaining chip in a discussion that allows counters to be valued or devalued in a non-rational fashion continues to be a revelatory experience for me.

  5. Hmm. I understand and agree with much of this. I think an interesting example of breakage is actually science on political blogs. I spend a lot of time on lefty US blogs battling anti-science. Most of this is in the form of anti-vax, alt-med, and anti-GMO stuff.

    However, that’s all still within the larger lefty community. And–in my experience–what’s happening there is two other dynamics: 1) These subgroups are finding material in their own echo-chambers/communities that they are bringing to the larger community. And then the part of that sub-community that agrees with them comes in to agree with them. Although dissenting voices also appear, not much progress is made between the groups. And then 2) those folks retreat back to their sub-community sites for reinforcement, and the cycle starts over again.

  6. Thanks for writing this, Bora. I enjoyed reading it from beginning to end – I was unable to attend this session and enjoyed the opportunity to see what came out of the discussion.

    I have read about the echo-chamber concept (most recently in Ed Yong’s posts: http://blogs.discovermagazine.com/notrocketscience/2011/01/18/are-science-blogs-stuck-in-an-echo-chamber-chamber-chamber/) and have had mixed feelings on the concept since. In reading your post, I liked your discussion of echo chambers actually being communities – it makes sense. In my work, I find myself surrounded by people who care about the same topics that I dedicate my day to. They study it all day, in their research and writing. When I step away from that group, it can be surprising to find that others don’t immediately know what I’m talking about.

    I also appreciate your comments about different sets of jargon in different fields. It’s interesting when you get scientists and engineers together – we can work on the same topics, but with completely different vocabularies.

    Thanks again for the post.

  7. I will come back and respond to comments tomorrow. Piled-up work-load first….

  8. Pingback: Society + Science Weekly blogs round-up - 27/28 February 2011 - science, culture, philosophy, religion, atheism - blog by Gurdur - Blogs on the Heathen Hub

  9. Pingback: The Best of February | A Blog Around The Clock

  10. Pingback: Link Loving 04.03.11 « Casper ter Kuile

  11. Pingback: Scienceblogging Roundup – February 27-March 5

  12. An “echo chamber” is, generally speaking, an audio loop that generates “echoes” without acoustics or reverberating surfaces. Similarly, the voices appearing on many websites are synthetic. They do not represent individual personal opinions.

    It’s nice to think the web breaks echo chambers. But if it is to do this, how do citizens engaged in discussing serious matters avoid being baited, diverted, and drowned out by a zombie apocalypse of sock puppets?

    Recently leaked material from security contractor H.B. Gary includes:
    “Persona management entails not just the deconfliction of persona artifacts such as names, email addresses, landing pages, and associated content. It also requires providing the human actors technology that takes the decision process out of the loop when using a specific persona. For this purpose we custom developed either virtual machines or thumb drives for each persona. This allowed the human actor to open a virtual machine or thumb drive with an associated persona and have all the appropriate email accounts, associations, web pages, social media accounts, etc. pre-established and configured with visual cues to remind the actor which persona he/she is using so as not to accidentally cross-contaminate personas during use. “

  13. Pingback: Attentive bee colonies and social graphs «

  14. I must second the options of Bjoern and Jacob. We all certainly agree with the notion that “most people are not exposed to views other than their own in their day-to-day interactions with other people.” But why would the Internet be any different? There’s nothing inherent with the web that persuades people to seek out the opinions of other. Like Bjoer, I think the web often serves to reinforce the echo chamber—much less obliterate it. Put another way, it’s not the access to information that’s the problem, it’s what people choose to do with the access that makes all the difference.

    This gets me thinking about a rather extreme example: Paul Haggis and the Church of Scientology. Alex Wright recently wrote a profile of Haggis* in which Haggis talks about his “discovery” of dissenting information. Haggis is a smart guy, but during this decades within the church it had never occurred to him—not once—to seek out such information. To me, it’s inconceivable that someone could use the web on a daily basis without ever encountering anything that could damage Scientology (even less so, I would think, when one is actively engage in the church), but that’s exactly what had happened. Haggis had never heard or read dissenting information until something prompted him to actively seek it out.

    * Paul Haggis vs. the Church of Scientology (http://www.newyorker.com/reporting/2011/02/14/110214fa_fact_wright)

    If I may I’d also like to mention a pet peeve of mine: Google Social Search. I won’t go into details but my basic premise is that Google with this feature is working to reinforce the echo chamber. I’d much appreciate any thoughts ya’ll might have on this topic: http://www.buildingtothink.com/2010/08/motivated-reasoning-and-social-search/