Confused by Party ID fluctuations in polls? They Are Probably Telling Us Something about Confounding Variables

There is a lot of hand-wringing and second guessing about the polls this election season. One point of contention seems to be whether the samples of some pollsters are skewed towards one party or the other. A lot of this speculations stems from the fact that voters who self-identify as one party or the other seem to fluctuate from poll to poll, even by the same pollsters. How can something fluctuate this much and still produce useful data?

To understand this, we need to untangle sources of error and fluctuations in poll results and also understand confounding variables. (It is not difficult but requires a bit of information and bit of thinking through).

First, all polls based on samples (rather than surveying the whole population like a census) have a “margin of error“, usually around plus or minus 2-3 percent for national polls such as Gallup, Pew. This margin of error should be understood in light of the “confidence interval” which is often not reported but is almost always 95%. This means that, 95% of the time, the sample results will be plus or minus 2-3 percent of the actual results of the whole population. So, even if everything was perfect and the sample was completely random, we would be off by more than 3 percent in either direction 1 out of 20 times (5% of the time).  (Why don’t we have a smaller margin of error? Because the margin of error moves with the square root of the sample size–the relationship is exponential, not linear so you don’t get three times improvement when you go from a sample of 1500 to a sample of 4500. See rough chart or play yourself with the calculator here. So it would be really costly have a sample large enough (about 10,000 in the US) to get the margin of error under one percent. ).

Hence, a few percent fluctuations are to be expected from poll to poll even if everything is done just  right.

Second, there are systemic errors because samples are rarely purely random. All our regular (frequentist) statistics depend on the Central Limit Theorem which assumes a purely random sample.

Random in the context of polling  means that every single potential respondent in the population has an exactly equal chance of being included in the poll. Well, this never works, no matter the method. Different kinds of people may have different odds of being at home. Note that the problem isn’t that some people aren’t at home. That would be fine. The problem is being at home is associated with being a different kind of voter. For example retirees may be more likely to be at home compared to younger respondents. Young respondents are more likely to have cell-phone only homes and are underrepresented in landline only polls. Again, this would be okay if we could assume that cell-phone only voter patterns look exactly like voting patterns of landline people. There is good reason to think not.

Pollsters deal with the fact that their sample is not purely random by weighting their samples by known quantities. In the United States, we know the pretty reliably the  percentages of  gender, race and age distributions–at a pretty fine level, too.  So, if my sample is 55% female and I know the voting age US population is about 52% female, I can adjust my sample calculations so that a woman in my sample counts slightly more *as if* my sample had the expected distribution of gender. Most pollsters and surveys do this and this is a well-tested and reliable method as long as the samples are not too small and the weighting is done according to well known population parameters.

Third, there can be fluctuations because of underlying changes to the actual variable (which you want to measure) *or* the willingness of respondents to answer polls (which you usually want to discount but in the case of voter surveys, you should not because willingness to answer pollsters is likely predictive of election outcomes as well).

So, here comes Party-ID. First question is whether to treat it as a stable population parameter by which to adjust our samples, or whether to see it as a variable. There is good reason to argue that it may well be stable and there is some work that suggests that it is indeed durable. On  the other hand, since it is indeed fluctuating in so many polls, and we know that it has fluctuated between elections as measured by more reliable exit polls, it is reasonable to treat it as a changing variable measuring something. In other words,  if it is fluctuating so much, does it indicate regular measurement error due to statistical fundamentals of sampling (i.e. you can’t escape some level of variation) or could it be measuring something else? I vote for the latter, given available evidence of fluctuation.

I believe that a reasonable argument is that party ID measures a combination of enthusiasm, willingness to respond to polls and willingness to be persuaded by a party so that you self-identify as a supporter rather than an “independent.” In other words, let’s assume you are lukewarm towards candidate A from political party AA but then something happens and you are excited about candidate A. Your phone rings. “I’m calling from Survey firm XYZnonsleazy pollster and I’d like to ask you a few questions about the upcoming elections.”  You are probably more likely to take that minute to answer that question if you are excited about your candidate and the election compared with “your uninterested in the whole thing” state. Plus, maybe, before you got excited, you identified as an independent.  Now you are excited about candidate A so you self-identify as party AA.

Hence, persuading fence-sitters and exciting supporters can easily result in a sample that has a higher party-ID proportion compared with one taken a few days ago among a less excited, less-persuaded crowd. In other words, partisans and fence-sitters closer to the other side may well be sitting out the poll.

Okay, you say, that means that the poll is less reliable because the sample is skewed and is leaving out some folks. Ah, yes, but it is quite likely that the effect on the election is just as real as party-ID may be a proxy for a confounding variable, excitement and likelihood of voting, which is well-known to strongly influence election results.

What are confounding variables? They are the reason you should run, most of the time, when someone says “correlation does not equal causation” without saying anything more substantive. If there is persistent correlation, something is going on and people should try to figure it out before deciding that something is substantive and of interest to you or not.

A classic example of a confounding is that ice-cream sales and murder rates are correlated. It’s not because we all scream for ice cream and pull out guns in frustration, but it is because there is a confounding variable: summer. Summer is correlated with both ice cream and higher murder rates. In this case, it is highly unlikely that you are interested in this relationship and likely not going to track ice cream sales to predict murders because you may as well look at the temperature–you have direct access to the causative variable.

But, in many cases, the confounding, hidden variable is of great interest because we don’t have access to the confounding variable–which is sometimes referred to as “hidden variable” exactly because we don’t have access to it.

Let’s say you find that people who drink carrot juice regularly are less likely to suffer from cancer. Shall we break open the juicer? Probably not because the likely confounding variable is people interested in a healthy lifestyle. People who are drinking carrot juice are likely taking better care of their health in other ways too because that’s just the kind of people they are. So, carrot juice is a proxy (stand-in) for healthy behaviors. But what if all you could measure was whether people drank carrot juice? I’d argue that it would be perfectly reasonable to use it as a predictive variable while also understanding that the relationship was not causative, but went through a confounding variable. What you should NOT do is prescribe carrot juice. (This stuff is serious and compplicated. This what went wrong with the infamous error when women were recommended hormone replacement therapy thinking it reduced cancers when, in fact, it increased them and the previously observed effect was due to a confounding variable–women who took HRT were probably more health-conscious to begin with).

In polling for elections, it is quite likely that party-ID is like drinking carrot juice. It signals something and, as a signal, it is useful. Hence, if party-ID trend goes in consistent direction from poll-to-poll from the same pollster, it is perfectly reasonable to assume that it is a proxy for a hidden variable, excitement and state of being persuaded of the voter base. (Also keep in mind that there will be small fluctuations due to unavoidable margin of error but those should cancel out over the long-term as the error can be in any direction).

Hence, fluctuating party-ID can function as a predictive variable even though it is not actually the causative variable. The fluctuations in this variable are probably a combination of sample fluctuations (normal and expected and happens with all variables within the margin of error 19 out of 20 times (95% confidence interval) and outside the margin of error 1 out of 20 times PLUS fluctuations in actual voter dispositions. While we cannot tease apart which part of the fluctuation comes from which part, it is safe to assume that statistical fluctuation will even itself out over multiple polls while the structural part that comes from voter excitement and engagement making people shift their self-perception and willingness to answer pollsters, can be useful and predictive even though it is, indeed, correlation not causation.


Occupy the Political: Politics of Self-Expression and the Occupy Movement

My colleague Daniel Kreiss and I have  paper titled “Occupying the Political: Occupy Wall Street, Collective Action, and the Rediscovery of Pragmatic Politics” which has been accepted to the journal Cultural Studies <=> Critical Methodologies.  In the paper, we compare some instances of strategic decision making in the civil rights movement with that of the Occupy movement. We argue that social movement participants have become so focused on self-expression as a mode of politics that strategic decision making and intervening (and changing) the broader political arena has taken an unfortunate back seat–and these developments are to the detriment of actual political change.

In other words, if movement participants concentrate this much on “being the change that they seek”, it may hinder their ability to bring about change.

If a movement is to be more than about creating a particular environment for people who have the resources, time and the possibility to occupy a public square for a longish period of time (in other words, for the 99% of working people), it needs to more strongly consider decision making and strategic action avenues that are more than about self-expression–and may even clash with it.

We wrote the paper to hopefully be part of such discussions we know are going on among many people who participated in the Occupy movement and are pondering the next step for politics of dissent and social change.

The full-text of the paper can be found here.

Title:  Occupying the Political: Occupy Wall Street, Collective Action, and the Rediscovery of Pragmatic Politics

Abstract:  In this paper we compare the institutional and strategic decision making structures of the civil rights movement with the Occupy movement with special emphasis on the role of self-expression as a political value versus strategic considerations. We argue that Occupy participants cast the values and form of the movement itself – how it operates and makes decisions – in terms that are synonymous with its very identity and survival. Occupy is the change that its members seek. There is both promise and peril in this approach. Occupy is finding it difficult to engage in institutional politics—which we argue is key to broad and durable societal transformations. We suggest that as Occupy goes home, and as it prepares to come back, it should renegotiate the tension between self-expression and strategic institutional action, and between movement itself as a goal and movement goals. In short, we argue that mistaking an anti-institutional style of participatory democracy and self-expression for both real democracy and radical capitalist critique undermines political power—and ultimately results in less progress towards participatory democracy as the movement becomes politically less relevant and less able to bring about societal change.


Feedback welcome!

#Egypt’s Muslim Brotherhood Teaches a Lesson in Elections in the Age of Twitter and Google Spreadsheets

Shortly after the announcement that Mohamed Morsi had won Egypt’s presidential elections, political scientist Marc Lynch asks a very good question:

Indeed, ex-president Hosni Mobarak had handily “won” the “elections” in 1987, 1993, 1999 and 2005, going to a mere 88% of the vote from a more proper 97% in 1987. The same apparatus that “won” Mobarak’s previous elections was in charge of most of the electoral machinery of the 2012 elections. It’s true that a major uprising|revolution|coup|spring [pick the term that least offends you] happened in 2011,  but attempts at election fixing post-uprising would certainly not be the first or the last time.

There may be many reasons for the apparent results, and it’s certainly possible that the Ancien Régime [or Felool as many Egyptians refer to them] just didn’t have it in them to fight this one.  Maybe they became late-believers in democracy (Though admittedly a weak hypothesis, as old-regime courts first disqualified the front-runners, then nullified the elected parliament, and finally greatly clipped the powers of the presidency in favor of the military council). Or, as some believe, maybe the military cut a deal with the Muslim Brotherhood.

Whatever else went into the apparent results, I’d like to suggest that the Muslim Brotherhood (or Ikhwan as Egyptians refer to them) made it harder for the election to be stolen because they combined a superior ground game with active and sophisticated online presence to control the narrative and force a level of transparency. (In other words, they forced it such that if the elections were going to be stolen, it was going to be “in-your-face” stolen which is a very different method with greatly different political implications than “under-the-rug” stolen).

Here’s how.

First, Muslim Brotherhood has a fairly active presence online, and especially in Twitter (although only their English feed is interactive at the moment). Throughout the election night, they kept tweeting out updated results:

I was curious about where they got their numbers so I asked them:

About 10 minutes later, I had a reply that they had people at polling stations who updated them regularly.  [Full disclosure: Due to my research activities, I do personally know at least one of the young woman who runs their English feed; in fact, most of my contacts with Muslim Brotherhood have been with young women as they happen to be quite active on their new media operations]. Other people I knew confirmed that MB had people on the ground at polling stations and were feeding results to the central headquarters as they became available.

This went on all night. As soon as results from all governorates (the official legal ruling structure of Egypt) were in, Muslim Brotherhood’s twitter feed, @Ikhanweb cheerily announced:


Plus, Muslim Brotherhood quickly put up a Google spreadsheet that had results from all governorates clearly showed an almost 4 percent victory for their candidade, Mohamed Morsi. I watched many journalists use this spreadsheet as a resource; in fact, Evan Chill quickly translated it to English, making it even more useful to a broader audience.

Meanwhile, Twitter feed for Shafiq’s official spokesperson was fairly quite. All the journalist –and pretty much everyone I know in Egypt– was up into the night and following the results, and the narrative was clearly emerging both on Twitter and on international media that MB had won the vote–but the question remained if they would win the election.

When Shafiq camp finally “woke up”, they had a relatively feeble response, made basic errors and invited ridicule rather than authority.


Egypt went to bed that morning with the feeling that MB had won the count, but with great uncertainty what the official results would be. (Keep in mind that this is the very first competitive election for presidency *ever* in Egypt.).

What happened next week was a true soap opera. It took a week for the Egyptian official election commission (PEC) to certify the results. The delay made many people fear that the vote would be rigged. Muslim Brotherhood took to Tahrir to demonstrate that they were not going to give up easily. Many speculated that MB and the SCAF, Egypt’s military council were in negotiations.

The final announcement of the results was almost comical if it had not been such a disservice to that country’s potential. After keeping the country waiting for a week, the head of the PEC showed up 40 minutes late and instead of announcing the results, gave an hour long rambling speech detailing how they dealt with various complaints while Egypt held its breath and world news organizations broadcast this live.

My Egyptian friends first reacted by making fun of this display of old guard authority, but were soon quite angry that this was the image that their country was broadcasting to the world–a country run by the old guard which was not about to give up its monopoly to speak as long as they wanted, no matter the occasion or the question. Many of us took this rambling speech with clear jabs at Morsi and the Muslim Brotherhood as a sign that the vote was going to be rigged–and that the winner to be announced was the establishment candidate, an ex-premier and a general.

And, yet. And, yet. When it came time for numbers to be added and presidents to be named, head of the PEC, Farouk Sultan, looked up from his papers and a bunch of numbers rolled of his tongue almost as an afterthought. And those numbers were within 0.01% 0.03% of the Muslim Brotherhood tally on that Google Spreadsheet.


In fact, revolutionaries and dissidents were quick to run the numbers and compare the official numbers of the PEC with MB numbers. (Heck, not only that, Moftasa Hussein had already applied the Benford’s test, which can be used to detect manual manipulation of numbers which should be more randomly distributed, to the unofficial numbers).

How influential was the fact that Ikhwan had so many verifiable numbers out in the open for the last week in convincing the old-regime apparatus to stick to the numbers as they were? We’ll likely never know. Perhaps it is best to assume that they would have done so anyway. However, it is hard to deny that had they been tempted to fix the numbers, that Google spreadsheet would stare at them at every turn. And the narrative laid by Ikhwan’s online and offline operation would be hard to surmount.

There are certainly huge problems with the electoral process in Egypt (here’s a bunch of them as outlined by Marc Lynch who has taken to calling the current system “Calvinball” ).  Egypt certainly has a ways to go, and hopefully for the better. Many will be watching to see whether Ikhwan can translate its new found position to charter Egypt in a more democratic and humane direction while respecting the rights of minorities and secular groups.

We have, however, just witnessed that they certainly know how to use on-the-ground electoral machinery with a sophisticated and 21st century approach to controlling the media narrative.


Does Facebook Cause Loneliness? Short answer, No. Why Are We Discussing this? Long Answer Below.

There is yet another article, this time in the Atlantic, asking the question “Does Facebook cause loneliness?” Like many articles on this topic, it ignores an enormous amount of data which –at a minimum- says, nope. (In fact, the main empirical article cited in the paper also says no). Research by many people (most importantly Keith Hampton) show again and again that Internet/Facebook users are less isolated than people who don’t use social media. Yes, there are complicated interaction effects but the simplest empirical answer to the simple form of the question is … no. By most standards of reasonable evidence, the answer is pretty much out there (even if ignored by most articles on the topic).

Why, then, is this question repeatedly invoked? I have a few thoughts on this. Here they are in a hastily written form (on my way to the airport!)

First, the findings I talk about refer to individual level effects. There are also societal effects to consider—and those are harder to measure. Maybe individual Facebook users are not less isolated than non-Facebook users, but maybe we have become more isolated as a society during the period that mediated-sociality has risen.

The answer to that question is yes, at least for closer ties. We are, on average, more isolated, at least in terms of strong ties. Three separate studies say so–and as we say in social sciences, once is a question, twice is a coincidence, thrice is a finding. (That is the General Social Survey with follow-up here, Pew Internet studies written up by Keith Hampton (with others) and a recent study by Matt Brashears).

I also have a paper (still) under review which shows –using the best dataset available—that Internet users fared better than non-Internet users during this period of increasing isolation. In other words, yes, we have less close friends than before, but Internet users are doing better at bucking this trend. I have a good deal of empirical probing of why this is so—I’ll try to write about it later. In short, I think this is because we are shifting from “ascribed ties” –people you inherit as close ties such as your family and your neighbors—to “achieved ties” –people you connect based on shared affinities and with whom you interact using multiple means of communication. It’s clear why Internet plays into this and fights off isolation. People who can use the Internet better to find and/or keep in touch with people with whom they share affinities with are more likely to be able to compensate for losing the neighborhood/family ties.

What data I’ve seen makes a strong case that social isolation is increased by factors like suburbanization, long-commutes, long work hours, decline of community and civic institutions, etc—not online sociality. As I argued, what data we do have argues that social media (and Facebook) works against this.

So, why so many articles like this which concentrate on the social media aspect (rather than things that appear to be a lot more directly related empirically to isolation)? Why does this question resonate so strongly?

I speculate that there are three reasons. Two are straightforward. Third is something I’m working through empirically and conceptually.

One, we are indeed more isolated, on average. That is true. And since increasing amounts of sociality are conducted –partially—online, it is a natural human tendency to create a narrative from two occurrences that follow each other. We are more online + we are more isolated = we are more isolated because we are more online. This particular conclusion is, in my read of all the available data, wrong but humans are narrative engines. We make stories whenever we see co-occurrences. (“I had a bad feeling in the morning and I broke a mirror before going out–then I had my car accident.” )

Two, face-to-face sociality *is* the bedrock of human communication. A baby few days old will respond differently to a figure in the shape of a human face compared with the same elements arranged randomly. There is something imitable about smiling with a friend. There’s just no replacement for hugging someone. That does not mean that people cannot have meaningful interactions that are not face-to-face, nor does it mean that online interaction is causing less face-to-face interaction. (As I said, data shows that people who interact online socially *on average* also interact more offline socially. It is a more-more setting. More social people are more social—online or offline.)

The third reason that I think such articles are still so popular is a phenomon that I’ve dubbed cyberasociality.

Cyberasociality is the inability or unwillingness of some people to relate toothers via social media as they do when physically-present. (Tufekci, 2010)

I’ve already published a bit about it (here) and I have an article I wrote about it that I will probably just put in my blog before it finishes going through the academic cycle. (Oh, what the heck, Internet. Here’s my unpublished paper on this. I presented these findings at last year’s American Sociological Association conference).

My argument is, briefly this: Just like we convert text (visual) into language in our head (which is all oral in the brain), we need to convert mediated-interaction to that visceral kind of sociality in our brain. And not everyone can do this equally well. And people who are cyberasocial are driving this discussion.

For example, reading is a hack of the brain. It is hijacking our visual systems and appending them to our oral/language centers (All human language is primarily oral. Writing is a recent, thin, barely-hanging there add-on). That is why you don’t need to train ordinary children to speak, but you need to train all children to read. Reading is something we use our cognitive capacity and neural plasticity to achieve. It is not an innate human ability.

Similarly, face-to-face sociality is a deeply-embedded feature. We react to faces and immediate presence in a very strong, very visceral sense. Online –and all mediated– sociality needs a similar hack into our sociality systems. We need to convert mediated-interaction to the same kind of unmediated interaction version we have to get evoke a similar reaction to “present in-person” sociality.

And I believe that this is easier for some people than others. Some are more disposed. There are also questions of technical skill, access, usability, etc. But I believe there is also an underlying personality variation as well.

My data says this is not a cohort effect. I’ve found strong and consistent distinctions among the college-age population. It’s simply not that the young are cybersocial and the old are not. It’s that different people are differently cybersocial. And for those who are cyberasocial, trying to describe online social interaction as “real” is like trying to describe colors on a oil-painting to someone who is color-blind indoors-and then to claim that there is a connection between the colors on the palette to colors of a sunny day.

They look, they squint, they think, they ponder and simply decide everyone else is crazy. There is a lot more to say but I’ll end with a quote from my (unpublished) paper. (My earlier published paper is here) :

I define the topic of my paper as explaining this:

Hardly another month goes by in which there isn’t a new article or book released on the
question of whether the Internet brings us together or separates us. Alternating between
lamentations by pundits on how social media tools are allegedly hollowing out our relationships
(Deresiewicz 2009; Mallaby 2006, Turkle, 2011), or by breathless reporting in newspapers about
how everything is now online, the debate refuses to die, and often seems unaffected by empirical
research on these topics.

And after going through the empirical findings, I speculate that “cyberasociality” is a real phenomenon, that it is not a cohort/age effect, and that for some people, online sociality is hard to comprehend because of deep reasons including being less disposed to converting mediated-sociality to broader sociality.

It remains a possibility that there are people for whom text is unable to evoke the same deep reaction embodied physically co-present interaction arouses. Such an inability, or an unwillingness, could be seen akin to another modern ailment, that of dyslexia (Wolf, 2007). The ability to convert alphabetical symbols to words, and then to seamlessly convert those words into meanings, is one of the more remarkable feats of the human brain and is mastered by most who are given persistent and competent instruction. However, for some segment of the population, this leap may remain unattainable and pose great difficulty even though the person in question may not suffer from any other disadvantage such as technological incompetence or inability or fear of using computers for instrumental purposes.

Whatever causes dyslexia, it would not have been detectable in a pre-literate population as among such people, words are always and only just sounds. In fact, linguists often caution against our tendency to equate words with letters and remind us that language is primarily aural and the transition to visual language is a late development. (Ong, 2004). Dyslexia emerges as a disadvantage only as a society incorporates the ease of use of the written word into the expected competencies into its portfolio, similarly, the increasing incorporation of online-sociality may expose a segment of the population that is similarly disadvantaged from being able to use these technologies as effectively as others.

Thus, conceptually, I propose a modern condition, named cyberasociality, which represents the possibility that some segment of the population remain unable or unwilling to relate to others via social media as they do when physically-present; and that this is not necessarily related to their general levels of sociality or to their competence with or use of computers or similar digital devices.

Finally, these findings may shed light on why the debate about the possibility of establishing deep bonds online refuses to die in spite of empirical findings that show that this is actually happening, at least for a segment of the population. It is possible that there are people who are deeply cyberasocial and are universalizing their subjective experience as the human norm, and thus are persistent in their claims that it is simply not possible to establish meaningful friendships using these technologies. Almost akin to a colorblind person insisting that all this talk of red and green must just reflect something superficial or made-up, and simply does not reflect a real division, the cyberasocial continue to write newspaper articles and even books lamenting the spread of these technologies. That is not to say that these technologies are not ever disruptive of sociality but that their effect may be much more complicated as some of the critics would claim.

Finally, I want to really, really emphasize that my argument is not a judgement or a statement of what’s inferior. I think this is part of normal human variation. I started looking into this as I was more curious about why this debate seemed to never die and this, in my opinion, is part of the story.

PS. Will add more links later! On my way to the airport now:-)

What Does Twitter Have to Do With Revolution? (Video of My Talk at Columbia U)

A talk I gave on March 5th at Columbia University about how and why social media plays an increasingly important role in social movements, especially in authoritarian countries:

The previous speaker I refer to is Bob Garfield of NPR whose talk had visual examples from a Krispy Kreme ad campaign. As I sat and listened to him, I thought that I better warn people that my talk will contain different kinds of visuals!

After a year of studying and talking about the Arab uprisings, I fear that my standards for “graphic content” have been dramatically altered. I often need to remind myself –and my audiences– of that fact–and that’s what I’m trying to do in the first minute of the talk…

Also, as background: More of my conceptual approach can be found here in this short article I had written for MIT’s tech review: New Media and the People-Powered Uprisings.

“We Love You Iranians” campaign on Facebook: War in the Age of Cat Videos

Last night, while I was flipping through Facebook looking for examples of the recent meme where Israeli citizens post the message: “Iranians. We will never bomb your country. We ::heart:: you” over their pictures,  I noticed a tweet from fellow academic Katy Pearce, who speaks Armenian, about how the word “pisi” means cat in Turkish, Farsi, Armenian and Azerbaijani. I responded that in Turkish, it was mostly used in baby talk, or to call out to cats: “come here pisi, pisi.” We got into a fun conversation about shared words in the region–and were soon joined in by others.

The word “kef” came up. Katy said it meant to “party” in Armenian.  Someone pointed out that it meant to “get high” in Uzbek. Another said that it refers to weed in Hebrew. We soon realized that it also had a similar connotation in Russian. I thought that it might all be related to the word “keyif” in Turkish which means pleasure in a sensuous sense.  But the word, it seemed, was everywhere and everywhere in the region. What gave?

A little bit of Internet sleuthing later, Wikipedia gave us our answer. In different incarnations, Wikipedia dutifully informed us, the word kef –kief, keef, kif, kef—refers to “resin glands of cannabis” which apparently contain higher concentration of psychoactive cannabinoids.  And the word, in all its incarnations, appears to have derived from the Arabic word “kayf” which connotes pleasure and well-being. A vice of pleasure, it seemed, had gone viral, at least linguistically.

Already sleepy, images floated in my head about how this phrase might have spread and take such strong root through this war-torn region. I had visions of an underground of devotees of hashish, smuggling their vice of choice among each other—would they continue their trade as wars broke out between the countries?  Would they speak each other’s languages beyond a few words? Would they sit around and get high together? Would a drug deal gone bad ignite more violence? I don’t mean to glamorize use of pyschoactive substances—I’m aware that one person’s pleasurable vice can be another’s addiction hell. Still, though, I was intrigued in the kind of human contact that must occurred to spread this word to so many corners—especially considering  that nowadays, ordinary Armenians and Turks –and for that reason Greeks, Russians and Arabs as well– often have very little day-to-day contact with each other.

And does contact make wars less likely? History of communication technology is littered with such optimistic predictions. Yet, it is hard not to be moved by how Iranians, too, have responded both privately and publicly by reaching out to Israelis. “We love you too,” apparently said one message from an Iranian responding to the campaign. “Iranian people, aside from the regime, have no hard feelings or animosity towards anybody, particularly Israelis.”

My thoughts were also with another past war yesterday was also the anniversary another major war, the Gallipoli campaign (“Canakkale” in Turkish) of World War I. (The Australians commemorate the battle as “Anzac Day” on April 25, while Turkey remembers it on March 18th.) About half a million people lost their lives in that fateful battle—many from as far away from Australia.  I never cease to shiver at the horror of a world that brought young man from as far away Australia to what was then the Ottoman Empire, half a world away–to kill and die among people they had never seen and never would again.

Hence, as it is probably the case on Anzac Day in Australia, there was a lot of news coverage in Turkish newspapers about the bravery and sacrifice of the soldiers –often peasant conscripts—who lost their lives in that battle.  Yet my inbox had a counter feeling. A more thoughtful friend sent me a song from Australia: “And the Band Played Waltzing Matilda.” I hadn’t heard it before. A haunting first-person indictment of the folly of war, the song is told through memories of a young man who lost his legs in Gallipoli, as “corpses piled higher” even around him. My friend said another friend had sent him the song. I thought about a network of people tuning out the official speeches that speak of a false glory of war, and instead share stories about its reality with each other.  What if we could do this before the soldiers got on the ships?

Would a world in which the reality of war is more visible also be less likely to fight wars? In his latest book, “The Better Angels of Our Nature: Why Violence Has Declined” linguist and Harvard professor Steven Pinker argues that level of war violence has gone done down dramatically in the 20th century. I’m only somewhat convinced. These days, most war deaths occur in the peripheries of battlefields—the collateral damage, the spiking infant mortality, malnutrition, disease. From Iraq to Afghanistan to the Democratic Republic of Congo, large numbers of people continue to die from wars, but fewer and fewer of them are soldiers on the battlefield.  Still, though, there is little doubt that the ratios of war deaths have gone down even as the numbers remain significant. Major wars used to kill tens of millions of people—and would still do if another one was fought. Does  globalization and increasing human contact decrease chances of wars?

From Wall Street Journal

People had high hopes that  radio, which started out as a two-way communication technology, would do just that–decrease wars by increasing human contact. Yet, Radio was shut down from its early interactive form right around World War I—and became, essentially a military technology. So perhaps history is not hopeful. Yet, there are reasons to think that human-to-human “diplomacy” is a more potent force than the past attempts.  Unlike early radio, the interactivity of Internet platforms is vastly more advanced and is used every day by billions of people. Plus, a kind of  “Globish” –a limited English– has indeed emerged as a shared language.  As the spread of Kony 2012 video shows, there is also tremendous interest in humanitarian fables–albeit in flawed an overly-simplified form as it was the case with that particular campaign.

Which brings me to my point: the greatest power the public sphere has is the act of granting of consent and the conferring of legitimacy to governments.  And that is where the Internet has been a game-changer. Often, it is erroneously thought that rulers rule through force–and dissidents have legitimacy on their side. In reality, it is almost always the other way around.  Consent and legitimacy among the population is the true the coin of all rule, including undemocratic and authoritarian ones.  Once the rulers lose legitimacy, they often also lose their control over force. Militaries refuse to go out of their barracks to save the regime, or turn against their old patrons. The police disappear from the streets. Elite divisions accelerate.  Ruling more through force than legitimacy requires draconian control over the public sphere. Hence, such regimes are thus distinctly vulnerable to disruption of new communication technologies like the Internet.

I would like to strongly note that I am not equating legitimacy among the population with democracy, or with liberalism, or with values that may be close to my own heart. Legitimacy can also be garnered through intolerance, extreme nationalism, and even, unfortunately, war-mongering. Hence, if one looks at cases where the spread of the Internet did not challenge authoritarian rule, one often sees a combination of different amount of two dynamics: a draconian attempt to keep control over the public sphere through the emergent challenge posed by the Internet *and* a good deal of legitimacy. Hence, from China to Russia to Iran, authoritarian regimes rule partly because they continue to have sizable amount of legitimacy among portions of their public (something Mubarak did not have as much in Egypt).

Thus, often, the strongest political act that a populace can do is to actively and publicly withdraw their consent and refuse to confer legitimacy on a particular regime—or, in this case, a war.  This is not to say that active withdrawing of consent guarantees to stop a war; but it certainly complicates efforts towards one. At a minimum, active displays of dissent can challenge a condition political scientists call “pluralistic ignorance”; i.e. when people privately hold beliefs that they believe are the minority but are in fact majority but hidden from view because people keep their preferences hidden. And Facebook and such platforms make it easy for people to easily reveal their true preferences—and to start rapid cascades. (Speed with which such things can spread is a very important dynamic as slow and fast diffusion do not create a little less or a little more of the same thing-they create very different consequences.) Hence, I am obviously not saying that Facebook postings are enough. However, if there was enough dissent within Israel towards a war with Iran, viral Facebook postings indicating such dissent would certainly be one active way of intervening in the public sphere and challenging the legitimacy of a pro-war drive.

What’s new and interesting about campaigns like the “We love Iran” is that they are easy enough for ordinary citizens to do. The fact that they are significant speaks  to failure of our conventional politics–in almost all developed nations, the institutional connections between citizens and levers of power have become weaker. The real critique should not be that the Internet is not enough to fix politics, but that most of politics has become so heavily dominated by insiders and the already-powerful.

I’ve already critiqued the derision towards so-called “slacktivism” –so common among gatekeepers– in my last post. As I said there, the concept of slacktivism misses the point of online symbolic action. Slacktivism “is not at all about ‘slacking activists’; rather it is about non-activists taking symbolic action—often in spheres traditionally engaged only by activists or professionals (governments, NGOs, international institutions.).” Rather than being pointless, such overt political displays –easily visible to one’s social networks; easy to share, to participate and to comment—provide a potential new path for ordinary people to have their voices heard in the public sphere. Symbolic action can certainly matter if the topic is important and the moment is right. And the Israel-Iran campaign could be such an example, depending on internal political dynamics is Israel and how widespread the dissent really is.

So while the outcome of all this is still uncertain,  one can certainly hope that it is harder to sustain a rhetoric of inevitable war against a country when, along with funny videos, pictures of pets, children and sunsets, their citizens are declaring their love for you on their status updates.

#Kony2012, Understanding Networked Symbolic Action & Why Slacktivism is Conceptually Misleading

By now, you probably know the story. A previously little-known group made a slick video with just the right mix of “the triumph of the human spirit” meets deep human tragedy about Ugandan warlord Joseph Kony. The video goes super-viral.  In just a few days, it is viewed more than 60 million times on just Youtube alone. Then comes a swift backlash and many people criticize the content of the video–ranging from its factual errors, to questions about the group’s finances, to the implicit “White Savior Complex”.  That is indeed an important discussion, one that has been covered in-depth elsewhere and not what I want to focus on here. (I wrote a bit about those topics in my short op-ed in the New York Times here).

Instead, in this post, I’d like to argue that evolution of Kony2012 has revealed how useless –and indeed harmful—the concept “slacktivism” has become to understanding networked symbolic action in the 21st century. I’d like to argue that people interested in social change need to step back and analyze the specifics of what is happening in its full complexity–without dismissing it, either due to their objections to the content or because they erroneously think it means nothing.  (On the other hand, if you get your kicks from erudite snarks about how kids in this generation have unruly hair, no manners, and no respect for its elders; well, enjoy. There is a huge market for that—just as huge, if not bigger, than the one for superficial hyping of social media as traditional gatekeepers often love to complain about how the new generations don’t appreciate Little House in the Prairie.)

My argument is this: the concept of slacktivism is not just naïve and condescending, it is misinformed and misleading.  What is called commonly called slacktivism is not at all about “slacking activists”; rather it is about non-activists taking symbolic action—often in spheres traditionally engaged only by activists or professionals (governments, NGOs, international institutions.). Since these so-called “slacktivists” were never activists to begin with, they are not in dereliction of their activist duties. On the contrary, they are acting, symbolically and in a small way, in a sphere that has traditionally been closed off to “the masses” in any meaningful fashion.

In other words, slacktivism should be seen as the encroachment of politics and civics into people’s everyday worlds which tend to be dominated by mundane concerns of day-to-day existence–or dominated by the consumerism transmitted through traditional media. It’s also a step in the unraveling of the professionalization of human rights and cause advocacy. [Credit: parts of this argument were developed in discussion with Alaa Abdal Fatah of Egypt and Sami Ben Gharbia of Tunisia].

So, not only are these people not slacking, they are acting symbolically in spheres that previously had higher barriers to entry. Symbolic action is not a magic wand–and its consequences depend on how it interacts with  other kinds of power, including institutional power.  Symbolic action and symbolic power, however, are not mere “epiphenomenon” of other kinds of power—as if they were a shadow, or an afterthought.

On the contrary, narrative and symbolic action are central forces in human societies. We are a highly-symbolic, group-oriented species and signaling our preferences –to others– is a key dimension of human action. “Public” is a meta-concept; it’s not just about what you know internally, but what you express and what others know that you believe and that you know that others know… ….  Hence, the public sphere is formed not just through people’s silently held beliefs, but through overt signaling of ideology and narratives-and this signalling increasingly takes place online.

As social psychology and related fields have long shown, and as any observant person knows –and like it or not– for the human animal, there are no pure facts; instead, there are narratives.  We act differently depending on our embedded narratives–even if we seemingly profess to the same facts as others. Humans accept, reject and make sense of facts within narratives. Just the very question of how we get to define what is a fact is a complex and socially-contingent one.

If humans came with a warning label, I’d like it to say: “WARNING: MAY APPEAR MORE RATIONAL AND LOGICAL THAN ACTUAL.”

Further, all human societies operate in a world of socially-constructed norms and ideals. And the changes to those ideals are immensely important.  If norms move, than often action also moves—not always in a straight line, and not always in a simple fashion because the world is complex and narrative power is but one kind of power among many. Still, action for change is always entangled with action for norm change, i.e. symbolic action.

Hence, there is no “activism” that does not have a strong symbolic side. Thus, today’s “meaningless click” is actually a form of symbolic action which may form the basis of tomorrow’s other kind of action.  Thus, signaling preferences –even seemingly obvious ones like being against child soldiers being abused—can be crucial.

Want some proof? Look at how many obvious –and not that hard– things we do not do globally. Children around the world die of the flimsiest causes.  Lack of so little action in that front is no doubt partially due to the fact that there is no strong symbolic infrastructure that makes those children important to us in a manner that is connected to capacity for impact. Clearly, the Kony2012 video is tapping into just that gap—and I cannot imagine a human-rights advocate who thinks that this is irrelevant.

It is well and worthy to discuss the interaction between symbolic power and other kinds of power—and how and when convergence among different kinds of power leads to change and when and how it remains confined or repressed. Still, the incorrect conceptualization encapsulated in the term “slacktivism” is making it harder to have exactly those conversations.

The power of the symbolic action to shape particular narratives is exactly why so many people felt the need to pushback against #stopkony—the video was effectively and powerfully laying down a narrative for a particular kind of action.

Indeed, go back in history, and you see these normative shifts, brought about by words and symbolic action, before or along with major social changes. History is full of such examples.  Harriet Beecher Stowe’s anti-slavery novel “Uncle Tom’s Cabin.” Anna Sewell’s animal rights story “Black Beauty.”  Or take a modern variation, like the clever video, “The Meatrix.”  These are examples of symbolic action which helps structure narratives within which further human action occurs.

And social media streams are a new and important dynamic in how those narratives are formed—and, importantly, who gets to have a say. I usually do not like to proclaim new developments as “good” or “bad”—they are often a complex interaction of both. However, contrast the swift pushback against the simplistic and dangerous narrative of #stopkony with the lead up to the Iraq War of 2003. It was clear to many people at the time that the narrative being built up in the rush to war in Iraq was erroneous, dangerous and in many ways, irresponsible.  However, opposition voices –while loud, organized and including many —were drowned out by the gatekeepers—big media, Sunday talk shows, political powers…

In contrast the swift backlash against Kony2012 was loud, organized and, most importantly, also able to command attention. In just one day, I saw more human-rights experts and African and Ugandan voices on mainstream media than I do in a month or three.  My social media stream was flooded by critical and in-depth discussion about the topic, often from Ugandans or topic experts. This is a key way in which Kony2012 differs from, say, “We are the World” campaign in the eighties in which Africans never got to be anything beyond silent victims. People can now talk back a lot more effectively. Indeed, the spread of Kony2012 is likely going to be remembered as one of the early examples how emergent networked global publics can connect amongst each other and focus their –and everyone else’s—attention in a manner that would have hard to imagine just ten years ago.

It also appears to me that this was, at least at first, spread most strongly by teenagers and young adults, at least at first. I obviously don’t have hard numbers at the moment –and hope we will at some point—but anyone with any experience in activism, organizing and social movements can immediately recognize that most lifelong dedicated activists have a “gateway” moment, often in their teens.  It would not be surprising if the intensity of the attention to this video –as well as the intensity of the backlash—did not become just such a moment for many future leaders. The kids are listening, maybe to a simplistic message, maybe to a misguided cause.  But some portion of them will keep looking, listening and learning. Such moments have long-terms consequences.

There is much more to analyze in this event in terms of content, politics of the message, the increasingly complicated interaction between our global institutions –such as the International Criminal Court—and increasingly networked global public. But here’s one thing. This is not unimportant. This isn’t about activists who are slacking. It’s symbolic action in a networked world, a complex and important topic for anyone interested in social change.


P.S. (Sorry for the linkless post: I am on a shaky Internet connection at the moment and will add more links later)



The Syrian Uprising will be Live-Streamed: Youtube & The Surveillance Revolution

I met this girl in a little village tucked away in the gorgeous but harsh highlands of Chiapas, Mexico:

After the Zapatistas burst into the scene in the 2000s 1990s with their uprising, there was a lot of discussion about their use of the Internet–and I was quite curious about that aspect*. I was a graduate student and it was quite easy to get to southern Mexico, especially if one took a cheap flight to Cancún and took the first bus out of that overbuilt heap of cement. There’s probably no harm in naming the particular village at this point, but the name hardly matters as it was a fairly typical Zapatista community, nestled between beautiful mountains and high enough to be in the clouds which look so delicate from afar but chill you to the bone when you are in their midst.Let’s just say it was this village:

I visited Chiapas multiple times over the years and I’ve always been careful about taking pictures of people as people’s attitudes to cameras vary from region to region, or even village to village.  They are often wary. On this occasion, though, it was this girl’s mother who insisted I take her picture. Puzzled, I tried to inquire why. Conversation was difficult as the women in these remote villages spoke very little Spanish, not to say anything about my linguistic inadequacies in Spanish. With the help of children who seemed to switch easily between Mayan languages of Tzoltil, Tzeltal, Tojolobal (as that particular area was at multilingual crossroads) and Spanish, she explained to me that she wanted for there to exist pictures of her children who grew before her eyes. It wasn’t just that there was no running water or electricity in this village. There were also no cameras. Time was reduced to its essence: ephemeral and passing. I did quickly ponder the ethics of the situation, and my own role in introducing new technologies, but quickly decided that I was in no position for any pretentious pondering of the “prime directive.” She wanted pictures of her children. I clicked on the shutter.

Fast forward just a little more than a decade, and we it seems we have cameras everywhere. Everywhere. Youtube’s “news director” Olivia Ma once told me that within about an hour of something of some importance happening anywhere in the world, Youtube has video of it uploaded. It is striking enough that Youtube has a news director. It’s stunning that videos of almost all events appear on Youtube within such a short time. In fact, correctly put, it would not be incorrect to say that Youtube is probably the biggest news site in the world–and that fact is often overlooked because there is also so much else on the site. Sometimes, it can take a while for a significant video to be discovered and no doubt that some never are beyond a small audience but there is almost always something, however shaky and grainy. This is nothing short of a revolution in surveillance capacity of citizenry.

One may wish that stoning death of Yazidi Kurdish young girl Du’a Khalil Aswad in 2007 was never discovered on Youtube, but that seems so trivial compared to wishing that she was never killed in such a cruel, brutal fashion. She was, though, for the alleged crime of seeing a boy of a different faith. She was murdered somewhere early in April 2007 and the video of her awful death started circulating widely later that  month. A few weeks after her killing, and a few months after the video was discovered and eventually made headlines around the world, a series of bombings shook Yazidi villages near Mosul, resulting in about 800 deaths and more than 1,500 injured—making it the single biggest episode of mass killing in an act of political violence since September 11, 2001. While the culprits were never discovered, most observers traced the events to the tensions that began with the video of her death.

The fact that the event was filmed and uploaded to the Internet is quite striking, too, considering the community. The Yazidis are a mostly Kurdish speaking religious group in the Middle East who keep to themselves as much as they can. The reasons for their protectiveness is lengthy and complicated but is related to the fact that a central figure in their faith, Melek Taus is accused of being identical to the Muslim figure of Satan. Having faced much prosecution, and also having a contentious faith in a contentious region, Yazidi society is predicated upon keeping outsiders out and practices strict endogamy—no marriage with outsiders.

Du’a Khalil, just 17 years old, crossed just that line with her alleged relationship with, and rumored conversion to Islam. For that, she was dragged to the street by a few dozen men who proceeded to beat her to death as she curled up on the ground, bleeding.  The shaky and grainy video, which I saw in bits over the space of a few days as I could not bear to watch in in a single sitting, shows at least *three*other people recording her stoning with cell phones.  It is quite stunning to think—not only are they killing her –this secretive, closed society which managed to survive for thousands of years by being so guarded and cautious— her killers felt like they should film this. Not one of them but at least four. And, more, upload it to the Internet.

Below, I’ve included the same stills from her death that Wikipedia has; the video is harrowing.


I’ve been thinking about that incident as I’m recently thinking a lot about what it means to have so much video of horrific death and destruction out of Syria. Indeed, as one journalist I spoke with put it, we are literally watching a war on civilians being live-streamed.  Despite all efforts by the Syrian regime, it is quite clear that this is not censorable or stoppable. The combination of cheap and small satellite modems and a fairly dispersed uprising has meant that there is truly no real way for the government to shut this down. The Syrian uprising will be live-streamed.

I have more questions than answers. What does it mean that everything –including  the most trivial but especially the non-trivial– has such a great chance of being available worldwide? Starting with the printing press, the threshold for the ability to publish has been getting lower, and the potential reach of publications has been getting bigger. We are now at the level of the person, publishing at the level of the world. The publishing revolution is almost complete.

Does this level of documentation make it more likely that the international community will be compelled to react to atrocities–which will likely come with higher and higher levels of visibility? Or will this, too, become just background noise, similar to famines or disease in Africa have become for most of the world (except the victims, of course)? Does the level of documentation and surveillance –and thus, evidence– make it harder to establish processes like the Truth and Reconciliation efforts in places ranging from South Africa to Guatemala? Will  this amount of documentation of atrocities make divisions even more likely and pernicious–as the ability to forgive often needs some level of forgetting? And the Internet, it seems, does not forget. Will this all make regime bureaucrats more likely to defect—as “I was just pushing paper and had no idea all this was going on” has become an even weaker defense? Or will they cling to power to the very end as much as they can, knowing their victims and survivors have much evidence as well as awful reminders of their crimes?

I don’t have the answers but I’m quite convinced that we’ve entered an irreversible point in terms of documentation of our lives, including death and destruction—not just baby pictures and trips, parties and graduations but also shelling of towns and killing of children. There is no going back. And tools matter. Just as wars with nuclear weapons are different than wars with bows and arrows, a world with cell-phone cameras in every other hand is different than a world which depended on traditional journalists and mass media gate-keepers for its news.

I want to emphasize that I am not making the argument that we, as humans, are drastically different in terms of our urge to document. In fact, I find such approaches to technology to be without support. Most of the time, the remarkable changes from technology come not because people suddenly start having new and unprecedented urges but rather because people continue to practice their mundane and human urges—but under drastically different conditions. The difference is not in human nature but in the socio-technical architecture. So, it seems, with our urge to capture out lies. The speed which we have taken this up clearly means we must have always had the itch to document, to surveill, and to display and share. But now we can scratch and scratch and scratch–and share on a global scale with a click.

I left my camera with that mother in Chiapas and have never gone back to that exact village to find out what has happened. It’s quite possible that not much has changed. Women are probably still painstakingly grinding the corn by hand, and children probably still go out at dawn to collect firewood from the dwindling surrounding forest for the stone tortilla ovens. There probably still isn’t a doctor within a reasonable walking distance of a day or so.

I am quite sure, however, that there would be at least a few cell phones with cameras in that village–as there are now in villages around the world, from Africa to rural India. And as more and more of those cell phones come with basic audio-visual capabilities, the surveillance revolution is here to stay. And I hope that little girl –probably a teenager now– is doing well, and I hope her mother has a few pictures of her as a toddler and as a young child to remind her of the days gone by. But I mostly hope that their future is even better.


* PS. The extraordinary thing about the Zapatistas and the Internet was not that the Zapatistas were that much different (or “postmodern” as commonly claimed)–in fact, they were a quite ordinary Latin American peasant uprisings in many respects. However, their insurrection took place in a world where the Internet was emerging as a powerful tool for political communication–and that did make a big difference.  That topic deserves a post of its own so I’ll leave it there.

*** PPS. The date typo confused a few people. I was not there in the early days of the uprising –followed it closely, but from a distance–. Rather, I traveled there later multiple times while the unrest was ongoing.

<Soundtrack for this post was Joni Mitchell, “The Circle Game.” >

Breaking Bread, Breaking Digital Dualism

New York Times author David Carr has a wonderful, mouth-watering piece about the joys of sitting to dinner with a group of friends and eating home-made bread baked from a recipe passed on from the host’s mother. Then he finishes the column with this point about face-to-face interaction:

All of which is a way of saying something that is probably obvious to others who are less digitally obsessed: you can follow someone on Twitter, friend them on Facebook, quote or be quoted by them in a newspaper article, but until you taste their bread, you don’t really know them.

I agree with most everything he says in the column because, besides the last paragraph, his column is a antidote to digital dualism–the idea that online and offline worlds are somehow separate entities, one “virtual” and the other “real.” But his column brings back digital dualism at the end–and does a disservice, in my opinion, to the rest of his points. He starts by explaining how the dinner group first met each other mostly online, then had this beautiful dinner together, and then shared the recipe over email–and ponders whether or not Google would ever find this recipe. (It seems that Carr asked the host, Clay Shirky, for the recipe and I have a feeling that it might soon end up where Google can find it.)

All these examples of how the online interacts with offline are clear example of why questions like “face-to-face or online friendship?” or “was online or offline more important in the Arab spring?” are not fruitful. The answer is yes. Because there is no “virtual” world separate from this world. As Nathan Jurgenson, who often writes about “digital dualism”, puts it the correct model to understand the Internet is not that the Internet is the “Matrix” and this world is “Zion” a la the movie Matrix. The world is one.

However, that one world is not the same one as the world before the Internet. What’s analytically and politically important is that the online is NOT the same as the offline, not that online and offline are somehow two separate planets. Bits and atoms have different properties and their current integration creates many novel configurations we have not yet adapted to as societies.

Bits are easy to copy while preserving their full organization, atoms are not (in other words, in the online world we have whatever Scotty in Star Trek used to beam people up by deconstituting them molecule by molecule and reassembling them someplace else. (Oops, if you are in an industry where your product is in bit form). Bits travel much easier than atoms, making bits much harder to censor and isolate (I’m looking at you, Mubarak). The architecture in the online world depends on the underlying code while the architecture of the offline world depends on laws of physics. Hence, online, we don’t have the same balance of privacy and visibility that come from the physical properties of space and time: that offline speech disappears after it is uttered; that, offline, we can usually see who is looking at us; offline walls, doors, locks and windows operate in a predictable manner. (That is why Facebook can be so jarring at times: it often ignores deeply ingrained cultural conventions based on laws of physics. It puts all your friends in the same room, by default–and its new timeline defies rules of flow of time as we knew it).

This lack of dualism applies to other technologies which separate the person from their words (which is the nexus of what Plato was vehemently objecting to when he decried the invention of writing and hence the separation of the human from their utterances. That critique is not invalid and remains applicable). When I was a child in Turkey in the pre-Internet era, I would have running conversations in my head with my favorite authors. These authors were quasi-people in my life because I had a void in my personal  world. It’s not that I was asocial, on the contrary. I had many  friends and I played in the streets and did all the kid things –and enjoyed them– but I also had a deep interest in topics that I was unable to find anyone around me to talk with in person (I suspect  this is not an uncommon experience). So, I’d go for walks and have conversations in my head. With authors. Whom I’d never met and probably would never meet.

And, somehow I just couldn’t manage this conversation as well with authors who I knew were dead. I was fully aware the odds were almost impossible –heck, I didn’t even speak English then– and I was a kid having these conversations walking by the sea by myself in a little sleepy town in Turkey. But there was always this tantalizing possibility that one day, I’d have that conversation with that person: books and my world were not dualistic, they were augmented. And I knew that the only true separation, the only true duality was death, and hence my deep disappointment whenever I learned a favorite author was not a contemporary. (A poignant  Turkish proverb asks: “Is there a village beyond death?” [“ölümden öte köy var mı?])

These days I meet people sometimes first online, sometimes first offline, but almost always have interactions that span both modalities. When a friend recently told me (online) that he was going on a pretty amazing trip, my first reaction was “how cool.” My second reaction was, “um, heck, does this mean I can’t email you with my random thoughts on stuff we’d been talking about?” Connectivity has become augmented and for me and the deeper divide has become not whether or not connectivity is online or offline, but whether there is some kind of connectivity or not.

There is certainly a difference between emailing someone and, say, sitting in a cafe by the Bosphorus; however, I am not able to categorize it merely as one is good/the other is bad. Each form has strengths and weaknesses depending on the topic, person, location, moment… Some things are better discussed over email. But sometimes you need to be able to hold out a hand. And as Carr mentions in reference to my work, interaction is one of the key mechanisms through which ties can strengthen or weaken–and certainly accessibility through online interaction is part of this mix. In fact, ubiquity of online platforms might increase the isolation of those who either through choice, disposition, or opportunity are not willing or able to be part of mediated, digital sociality and hence create a third level of “social” digital divide. (First level being basic access and second level being skill).

You might think my own experience is unique (and as a traveling academic I am certainly not typical), and Carr is also not typical. But recent survey findings reveal that ordinary people are also  increasingly establishing “migratory” friendships–In two separate studies, about 20 to 25% of respondents report friendships which begin online and migrate offline (Wang and Wellman 2010; Gennaro and Dutton 2007) so this is certainly not the exclusive domain of the digerati.

I am not at all claiming that this augmentation of bits and atoms does not have profound consequences; it does. For example, thanks to the Internet, we are more increasingly able to connect with people with whom we share affinities rather than people we happen to live next to (but it also works the other way around — when I moved to Chapel Hill, I chose to live where I live so I would be neighbors with  someone I knew previously and with whom I had mostly interacted online before). This is what sociologist Barry Wellman calls “networked individualism”: instead of being completely confined to historical “boxes” of family history and geography, we can open up to constructed ties of interests and affinities. This is a profound change and it is still playing out in the early stages. (Carr and others worry that this might lead to “filter bubbles” as we get our information mostly from chosen friends. Maybe, maybe not, as I reflect on this here.)

The fact of online and offline augment, rather than categorically oppose, does not change the fact that there is something deeply human and imitable about in-person interaction. Babies, even when a few days old, respond to a human smile differently than a non-smiling human, and distinguish between a representation of a face and a shape that has elements of a face but is not arranged like one. To this day, one of my greatest regrets is that I never managed to meet in person with one of the authors with whom I had these deep, personal but internal conversations, Edita Morris, and tell her she changed my life. She was alive when I read her book but died a few years after. By exposing me to the existential horror that can be brought about by scientific knowledge, something I had not considered before, her novels about the aftermath of the atom bombs started me away from the path of the child driven by scientific curiosity and one who wanted to be a scientist herself to who I am as an an adult–a person who wants to understand and help shape how science and technology interact with our world. If I had the Internet then, I could have at least emailed her.

Is the Social Web Less Surprising? The Internet of People and Social Flâneurism

As I read this essay on “The Death of the Cyberflâneur” by Evgeny Morozov who argues that the Internet lost its early quality of leisurely strolling and encountering the unexpected, I kept thinking about how this did not fit my experience. While there are many parts to Morozov’s essay –some of which I am not going to discuss here at all– I want to focus on the idea of “wandering around” the Web and encountering the unplanned in light of the emergence of the “social web”. For me, the social web has greatly increased exactly this quality of the Internet –encountering the unsearched and the unplanned– and I don’t believe this is because I am exceptional but rather it is because connectivity through people –the social web– yields more diverse and surprising encounters than mere connectivity through topics or information-the early Internet.

To go back to Morozov’s essay: there are two parts to his claim why the Internet is less about the aimless encounter: first is empirical, and the second is structural. Here’s the empirical argument:

Something similar has happened to the Internet. Transcending its original playful identity, it’s no longer a place for strolling — it’s a place for getting things done. Hardly anyone “surfs” the Web anymore. The popularity of the “app paradigm,” whereby dedicated mobile and tablet applications help us accomplish what we want without ever opening the browser or visiting the rest of the Internet, has made cyberflânerie less likely.

First let’s get some numbers out of the way. Is it true that “hardly anyone “surfs” the Web anymore”? In a recent (December 2011) and nationally representative survey, Pew Internet found that 74% of all online adults answered “yes” to the question if they do go online for “no particular reason except to have fun or to pass the time.” I don’t think this is blip. In my 2010 surveys with college students a whopping 77.9% said “very often” when asked how often they went online “for no reason at all”—and I’ve probed this question many times since 2005 and I have similar numbers for all the years.

I’ve asked Evgeny Morozov about his data and he pointed to a Microsoft advertisement survey—the survey had scant public information but actually shows a 4% increase in amount of surfing from 2007 to 2010; and surfing remains the second most popular activity after email. Morozov also referred to the famous Wired article which argued that the “The Web is Dead.” I’m sympathetic to the argument in the Wired if interpreted structurally: that the Web is being taken over by “walled gardens” like Facebook and single-purpose apps. Data-wise, though, the Wired piece does not speak to this argument of encountering the unexpected or wandering around (it’s comparing peer-to-peer and video traffic to “Web” traffic, for example). (Also, see this critique.)

So, let me consider structural argument which I find a lot more interesting. Here it is in a nutshell:

As the popular technology blogger Robert Scoble explained in a recent post defending frictionless sharing, “The new world is you just open up Facebook and everything you care about will be streaming down the screen.”

This is the very stance that is killing cyberflânerie: the whole point of the flâneur’s wanderings is that he does not know what he cares about. … Compared with Facebook’s highly deterministic universe, even Microsoft’s unimaginative slogan from the 1990s — “Where do you want to go today?” — sounds excitingly subversive. Who asks that silly question in the age of Facebook?

I have to say that at face value, this is an attractive argument, and one that has been echoed by others: the “filter bubble” or the “daily me”—and indeed raises deep questions about how to design for serendipity:to find not merely what we are looking for, but to wander and to stumble upon things a la the flâneur, to encounter the unexpected, to savor the surprise.

However, in my personal experience, platforms like Facebook and Twitter have greatly increased the amount of unplanned and interesting information I encounter exactly because they are person-to-person spaces rather than information-spaces. Online, I interact with people with whom I share at least one strong interest–but thanks to the social Internet, I am not exposed to just that limited topic but the much wider universe of what my friends share. And this is always much broader than the narrower affinity that first connected me with that particular person.

The trick to understanding this effect is to understand that people’s affinities don’t lump neatly, nor do they ever overlap completely. In other words, we are, fortunately, not clones of each other. Stereotypes work as broad generalizations at the aggregate, statistical level but break down spectacularly at the individual level: show me a stereotypical “X”, and I’ll show you traits he/she does not share with other stereotypical “X”s.

For example, I have strong interests in social impacts of technology, human rights and democracy and the Middle East–and often make friends based on that basis. However, I have other strong interests which do not fit into any of those categories—and, crucially, so do my friends. Social media thus exposes me to other interests of my friends, and exposes my friends to other interests of mine—and I am surprised everyday by the richness and diversity of these encounters. A friend I met because of shared interest in the emergence of the Internet and the Middle East may pleasantly introduce me to a music genre I dabbled in but never had the time and expertise to dive into. From another, I encounter the intricacies of representing light on metallic surfaces on oil paintings. Somebody shares an article on how bilingualism changes the brain.  Baby pictures, sunsets and cat videos, of course, also pass through my stream.  But along musings on nature parks in Latin America. There is a new restaurant in town—well, in a town I’ve never been to. It goes on and on and on in a great deal of richness, diversity and complexity.

In fact, encountering things I did not explicitly care about happens to me significantly more than the days I had to click-click-click my way around “cyberspace” because as much as I thought I wandered around, I could never wander around within the richness and depth of my encounters through the social web.

At this point, you might say that I have interesting friends and that is an anomaly. Perhaps, but that is not the point. The point is that I have friends who are not clones of me but have enough similarity with me that they introduce me to topics that I did not know I was interested in, but may plausibly be, if it just popped in front of me. This is likely true for most people for structural reasons.

Crucially, my online encounters introduce me to spaces that are both adjacent to (interests of people somewhat like me) but also far (not my interests) from me.  In the old days of seeking by information, I was more likely to encounter only information I was seeking (Duh) but not manage to get too far unless I truly started clicking randomly–but how do you do that? Search for “random”?  In the new era of connecting to people, I am exposed to a lot more because “people” are a lot broader than “categories of information” which are, by definition, narrow.

To put an empirical point on this, look at this striking finding from the latest February 2012 Pew survey.

It is commonly the case in people’s offline social networks that a friend of a friend is your friend, too. But on Facebook this is the exception, not the rule. A fully connected list of friends on Facebook would have a density of 1 (everyone knows everyone else). The average Facebook user in our sample had a friends list that is sparsely connected. As an example, if you were the average Facebook user from our sample with 245 friends, there are 29,890 possible friendship ties among those in your network. For the average user with 245 friends, 12% of the maximum 29,890 friendship linkages exist between friends.

For those of us familiar with research on this topic, this number, 12%, is very striking. Most of the time, it is fairly safe to assume that a person’s friends form “close triads”—people you know also know each other–and this creates a structure that looks like a triangle: everyone is connected to everyone else. What this 12%  number is saying is that on Facebook, your Facebook friends are almost never friends with each other, i.e. rather than triangles, they form “open triads”. People in closed triads tend to be more similar each other either due to origins of the connection  (living in the same environment) or due to convergent homophily (people exposing/converting each other through multiple encounters over time).


Finally, here’s the best empirical finding on this topic: Look at this striking study of 250 million people on Facebook [in a true experiment, no less] by Eytan Basky explained here:

Although we’re more likely to share information from our close friends, we still share stuff from our weak ties—and the links from those weak ties are the most novel links on the network. Those links from our weak ties, that is, are most likely to point to information that you would not have shared if you hadn’t seen it on Facebook. The links from your close ties, meanwhile, more likely contain information you would have seen elsewhere if a friend hadn’t posted it.

In other words, platforms like Facebook connect you with people who connect you with information you simply would not have encountered yourself—and these connections happen especially through your weaker ties who, structurally speaking, are more likely to be in open triads with your other friends, i.e not friends of your their friends.  I suspect this may be even truer for Twitter.

I don’t deny there is something to the argument that the shift to walled gardens comes with particular threats, but I disagree that the shift to the social Internet kills diversity, surprise, or richness. An Internet that is collection of sites which brings together, say, “people interested in model airplanes to talk about model airplanes” is going to be a lot less likely to expose us to the unexpected than the social Internet which connects us to people in their fuller richness.

It is not the Internet of things, or Internet of information, which keeps the Web brimming with the unexpected: it’s the Internet of people. Sometimes nothing is a more surprising and complex bundle of the unexpected as another human being.