Against studying the Internet

by Henry Farrell on April 19, 2011

I’ve been meaning to respond to this “very interesting Tom Slee post”:http://whimsley.typepad.com/whimsley/2011/03/blogs-and-bullets-breaking-down-social-media.html for weeks and weeks.

bq. Maybe we should stop talking about “information and communication technologies” or “the Internet” or “new and social media” as a single constellation of technologies that have key characteristics in common (distinctively participatory, or distinctively intrusive, for example), and that are sufficiently different from other parts of the world that they need to be talked about separately. The Internet is still pretty new, so we tend to look at it as a definable thing, but digital technologies have now become so multifaceted and so enmeshed in other facets of our lives that such a broad brush obscures more than it reveals.

I think that this is right, and have made at least vaguely similar arguments in the past, but disagree with Tom’s further suggestion that the best way to do this is through looking at “smaller scale structures” – e.g. drawing clearer distinctions between different kinds of social media, each of which works somewhat differently. The logic of this argument seems to me that people who now study and make claims about the Internet should be studying and making claims about e.g. “Internet-based social network platforms” instead. But it may not necessarily make much better sense to study “Internet-based social network platforms” than to study the Internet, because these platforms will be nearly as much a jumble of different causal effects as the broader Internet. One can easily imagine how the same kind of logic of debate will reproduce itself at still lower levels, with e.g. people who think that Facebook is teh awesome for social connectivity pointing to all the ways in which Facebook potentially brings people together, and others who think that Facebook is horrible pointing to all the ways in which it potentially turns people into isolated drones.

The better way to do what I think Tom wants to do is to try our best (this is admittedly hard to do consistently) to think about the universe in a quite different way. Rather than trying to drill down and down and down through ever more specific technologies until we finally discover the fundamental particle of electronically mediated social connectivity (or, more likely, give up from boredom or despair), we should start from a very different set of premisses.

Instead of wanting to study ‘the Internet’ or ‘Facebook’ or whatever, we should investigate the possible existence or relative strength of various posited _mechanisms_ which causally connect certain explanatory factors with certain kinds of interesting outcomes. Most technologies will potentially bundle a number of these mechanisms together – hence, the need to try to disentangle these mechanisms as much as is possible in specific instances. Sometimes the mechanisms bundled in a particular technology may work against each other, so that e.g. a particular technology may be associated with mechanisms which simultaneously undermine and strengthen e.g. increased political participation in different ways. The relative efficacy of these mechanisms (or, better, the circumstances under which they are likely to be more or less efficacious) should be the focus of investigation. Instead of asking ‘does Facebook help protests in authoritarian regimes?,’ one would ask questions such as ‘does social influence from peers make individuals more likely to participate in demonstrations?,’ ‘does widely spread information about protester deaths make individuals more or less likely to participate?,’ ‘does government-provided information make citizens less likely to participate in anti-regime protests?’ and so on.

This I think, helps to disentangle the specific causal questions from each other (one could imagine how Facebook could possibly be associated with each of these possible mechanisms, and with many others). It also breaks down (as Tom wants to break down) the artificial separation between ‘new media’ and ‘old media,’ by forcing the researcher to think about the _specific_ ways in which particular media may be associated with this or that causal mechanism. It is more plausible to me than to Tom that many ‘new’ media are likely to be more open to a wider number of participants than many more traditional media – but this is not a necessary prior – instead it’s a topic for empirical investigation.

I should make it clear that none of these arguments are original to me. Instead, they are an application (with minor updates) of claims made by Adam Przeworski and Henry Teune forty years ago. Przeworski and Teune famously (for political science values of ‘famously’) called for comparativists to ‘replace country names with variables.’ In other words, they wanted political scientists to stop maundering on about the differences arising from the ineffable national characters of Germany, France and Italy, and instead to formulate hypotheses about the relationship between variables that might be seen or not seen in different national cases (e.g. to construct a trivial and obvious example, to test the hypothesis that states with powerful left parties are more likely to have extensive welfare states). I’d like to see people who study the ‘Internet’ and ‘social media’ stop studying them, and instead start focusing on the role of causal mechanisms that might (or might not) be associated with specific technologies in explaining political outcomes, i.e. to start replacing technology names with mechanisms. This would, of course, require Real Research. But it seems to me more promising than the likely alternatives.

{ 31 comments }

1

jkd 04.19.11 at 4:53 pm

As someone who’s a newly-minted Ph.D. studying “the Internet” (or, you know, this), I agree almost entirely. It’s not much use to study “the Internet” or “social media,” the former of which is receding into the background architecture of all modern life, and the latter of which is a jumble of too many different kinds of things and behaviors to be of much use as a term. Rather what’s interesting is how people use new technologies in their interactions with one another, the state and society – as part of a constellation of pre-existing practices, and also as new emergent practices. Which is just another way of saying yes, replace country/technology names with variables.

I do agree almost but not quite entirely, because I think there is something special and genuinely unprecedented about the whole range of new communications technologies that, within short order, will make it at least theoretically possible for most people on Earth to talk to one another directly. That’s… I don’t even know what that is, but it’s damned fascinating, and a change of kind-not-degree in human identity and relations. How and what that change will be are, of course, the question. But in small ways, studying how new and emergent technologies are weaved into that web of human communications, perceptions and relations seems to me to be a very worthwhile endeavor.

2

dfreelon 04.19.11 at 5:22 pm

I wouldn’t be so quick to dismiss the roles that distinct technologies can play in political participation (if that is, in fact, what you’re doing). If, for example, we’re interested in the question ‘does social influence from peers make individuals more likely to participate in demonstrations?,’ we need to understand that the ways in which social influence is mediated may affect its efficacy. So Facebook may be a better platform for mobilization because people use it to connect with people they know, whereas Twitter may be better for spreading awareness internationally and for citizen journalism due to its stronger support for weak ties. And F2F communication may trump all of these in some circumstances but not others due to the immediacy of someone asking you to get out there and participate.

This is all really just to say that technology should not be completely effaced in 21st-century political studies (but nor should technology dominate). And I feel that this needs saying, because political science as a discipline is known for discounting the role of technology in politics.

3

Zeynep 04.19.11 at 5:34 pm

Henry, I am going to agree and disagree. In the last panel we were both on, I presented a brief talk titled “Faster is Different. Social Media, Social Change and Causal Mechanisms” to look into the Arab uprisings, specifically Tunisia, in light of social media. My argument was social media changes speed of information diffusion which makes it harder for governments to isolate protests under conditions of mass dissent. (On the contrary, under polarized situations, faster info diffusion often accelerates divisions, often to the benefit of the powerful. Compare Tunisia with Bahrain, for example). This effect, I argue, is compounded by the changes to connectivity so that one-to-one (or few-to-few; i.e. ordinary face-to-face interaction) and one-to-many (broadcast) is also joined by many-to-many connectivity.

So, yes, the causal mechanisms, rather than social media per se, are of most importance to me, as well.

That said, to get at those mechanisms, one really has to have an in-depth kind of understanding of Internet and social media usage. For example, one of the most common misunderstandings is that “cyberactivism” is meaningless unless it moves offline — as if there is a “cyber” out there separate from the world. Online and offline activities are fairly thoroughly-integrated for most people, especially in autocratic states where people get tortured for blog posts.

Or, take, another meaningless term: clicktivism. If one means that low-risk, low-effort signifiers of preference are meaningless that is, of course, an actually debatable proposition. A Facebook “like” for a cause in the U.S. is different than a “like” for “We are all Khalid Said” page in pre-revolutionary Egypt. In this sense, I am agreeing with you, a statement without the Internet, just referring to the level of risk, for example, is more clear. However, in order to correctly formulate such a statement in light of reality, one must understand actual practices of Internet use in different contexts and that does require studying the Internet.

Otherwise, one can blab about how the Internet is only about weak ties and derive grand theories from that, when, in fact, research shows that not to be true (it’s not an either/or at all; besides, tie-strength itself should be understood at a dynamic continuum subject to modification due to mediated digital interaction). But answering such questions requires the study of Internet and the ways in which it is integrated into people’s social practices.

4

William Timberman 04.19.11 at 5:40 pm

Late in 1993, a technofriend of mine called me, and demanded that I rush over to his house immediately. When I finally got there, he proudly demoed Mosaic for me on a 14.4K modem. As he breathlessly carried on about hyperlinks, and how they represented the future we’d all been waiting for, I watched the picture — an impressionist painting, as I recall — tick slowly down the screen.

Yes, I agreed, this looks like it, alright…but not just yet.

Early the next Spring, I had my first ADSL connection installed. The following month was consumed in dealings with phone company technicians who went on endlessly about bad pairs, x thousand meters to the central office, and bad DSLAM cards while poking about to no apparent purpose behind my walls. Then came a quick online course in HTML coding, after which I called another friend of mine, and together we launched a Magazine of Art and Literature on the Web.

One of the features of that venture was something called the Committees of Correspondence, which, given that it relied on e-mail correspondence that I sequenced and posted myself, turned out in practice to be something like the quill pen version of a blog.

Nothing much came of all this, of course, but it taught me two things: Like the proscenium arch in early motion pictures, the past is always embedded — at least initially — in any realized conception of the future. Also, If you build it, they will come.

5

Matt McIrvin 04.19.11 at 5:49 pm

North Americans and Britons tend to use LiveJournal (when they use it at all) as a semi-walled garden for carrying on discussions with their online social networks. Russians tend to use it more as a broadcast blogging platform, including many high-profile political blogs. There are exceptions, but that’s the trend. Anglophones on LJ had little clue about this until recent DDoS attacks aimed at Russian political bloggers brought it to light.

Is there anything about the nature of LJ, or of Anglospherians and Russians, that makes this outcome inevitable? I doubt it; Anglospherians just do political blogs in other ways, and LJ became the dominant Russian blogging platform by some path-dependent happenstance.

6

Henry 04.19.11 at 6:35 pm

Zeynep – yes. But what you are arguing here (unless I am mistaken) is not that we should be studying the Internet _as such_ but instead that we should have some understanding of specific technologies to be able to figure out how the causal mechanisms work in specific instances. I don’t think that this is a problem for my argument. Deen (dfreelon) – your argument could be interpreted in two ways, I think. One is as agreeing with Zeynep more or less, which is, I _think_ what you are saying. The other is as making a stronger claim – that there is something causally unique about certain technologies, which cannot be reduced down to mechanisms of the kind that I’d like to see as the baseline of social science explanation. This would be a stronger claim, obviously.

7

Don 04.19.11 at 6:41 pm

There’s a very useful discussion here about how to stop mistaking the Internet for something else.

8

Evelyn Messinger 04.19.11 at 7:30 pm

Thanks so much for this post. Looking at the Internet as part of a contextual web of politics, society and/or technology is obviously necessary, but overlooked in the understandable excitement over the truly ground-breaking extension of personal and public power afforded by new Internet technologies. As you say, it is time “to think about the specific ways in which particular media may be associated with this or that causal mechanism,” and your questions do force one to think in these broader terms.

I’ve been concerned about this because the exclusive analytical focus on social media has created a kind of blinkered field of analysis, where the public thinkers who should be helping us understand the whole picture of media and society have so narrowed their frame as to occlude what is obvious. I am as enamored of the wonders of the Internet as anyone, but my background in international television has forced me to notice this. So let me describe two of the most profound political developments of this era, which share a specific confluence of media that cries out to be better understood. These are the rise of the Tea Party in the US, and the revolutions in Tunisia and Egypt.

First the US: In 2009, some disgruntled Americans took to screaming at their elected representatives at Town Hall meetings; these rants were posted on YouTube. Fox News then plucked the ranters from obscurity, featuring many of them as interview subjects on many different Fox programs. (see http://www.youtube.com/watch?v=syWM7GXnJRA&feature=related) Then their YouTube hits went through the roof, and the rest is history. The history of the Tea Party.

Second example: A young Tunisian man set himself on fire. A few people staged a mini-protest, which they recorded on their cellphones. They took their video to Al Jazeera, which ran a story about the young man’s plight that was broadcast throughout the Arab world (see http://www.youtube.com/watch?v=jHw_auqod6Y). Existing Tunisian and Egyptian Facebook and Twitter protest sites subsequently lit up, ultimately resulting in revolution.

Al Jazeera’s role in the Middle East events was covered by the news, yet left out of analytical debates about whether these were or were not “Twitter revolutions.” The rise of the Tea Party has been mostly ignored by media analysts, perhaps because YouTube is an old story and Fox News is part of that medium long ago declared dead, broadcast television.

9

dfreelon 04.19.11 at 7:51 pm

Henry: yes, I’m more in agreement with Zeynep than your second interpretation. My concern was primarily with the absence of technology from your sample RQs, which may obscure how different digital platforms enable and constrain different political outcomes under different circumstances. Given the number of social scientists and popular commentators who claim that technology is irrelevant to politics, it’s important to clarify which side of the line one falls on. I can agree with your argument so long as it allows communication technologies to play some moderating role—of course, they may not always do so, but we should remain open to the possibility.

10

tomslee 04.20.11 at 1:19 am

Henry’s suggestion is more promising than mine, although “fundamental particle of electronically mediated social connectivity” sounds so much cooler than “mechanism”.

Internet-based activities are so visible and quantifiable they just demand to have hypotheses formulated and tested against them. But social networking tools find their heaviest use in environments where real-world social networks are also the densest (universities, for example), so yes researchers must understand “actual practices of Internet use” but somehow they must not ignore the other many-to-many communications tools that exist in those environments (posters, radio stations, bars and so on) that are less countable.

I have not heard or read Zeynep’s “Faster is Different”, but even “faster” is something that needs a case made for it. Posters, community radio, and other informal mechanisms work pretty quickly in socially dense environments too, though their influence is more difficult to track. As Henry says, it may be that “many ‘new’ media are likely to be more open to a wider number of participants than many more traditional media – but this is not a necessary prior – instead it’s a topic for empirical investigation.” When empirical investigation is so much easier for digital activities than for analogue, researchers must be doubly-careful not to assume the answer they are looking for.

11

novakant 04.20.11 at 9:09 am

12

Doug 04.20.11 at 11:01 am

One case that might be helpful for figuring out whether Now is in fact faster than Then is Poland in the time of Solidarity, either its founding in 1980-81 or in how it led to the round table and free elections in 1988-89.

13

Straightwood 04.20.11 at 3:35 pm

A more radical critique is that the conventions of academic analysis can’t effectively engage phenomena as turbulent and complex as (for want of a better term) the digital network revolution. When the change cycle is faster than the study/evaluate/publish cycle, then academia will often be explaining things that are moot or of dubious relevance.

It takes many years for a domain of university study to become recognized and established, and decades for a substantial body of scholarship to form in such a domain. But Internet/digiverse/Noosphere phenomena have life cycles measured in months. I believe that scholars will have to wait for the arrival of a stability plateau, or they will run the risk of producing largely irrelevant results.

14

chris 04.20.11 at 4:04 pm

I believe that scholars will have to wait for the arrival of a stability plateau, or they will run the risk of producing largely irrelevant results.

I don’t see why a historical result (that is, an analysis of a phenomenon that has ceased by the time the analysis is published) would necessarily be irrelevant. Particularly if similar phenomena recur, which is itself a question open to historical analysis.

15

Straightwood 04.20.11 at 5:05 pm

I don’t see why a historical result (that is, an analysis of a phenomenon that has ceased by the time the analysis is published) would necessarily be irrelevant.

A “result” assumes a settled observable state. One can declare something to be in such a state, but for rapidly evolving phenomena such declarations are of dubious value. That is why anyone attempting to write a history of Internet social media today is pursuing a fool’s errand.

16

Clay Shirky 04.20.11 at 8:48 pm

Unsurprisingly, Zeynep posts for me, but it’s a beautiful Wednesday afternoon and me with nothing to do except a few Looming And Very Important Deadlines, so I’ll post for me too.

Like Zeynep, my answer is ‘Agree in part, disagree in part.’ I agree that more talk about mechanisms (I like Slee’s language for this) will be valuable — the mechanisms I’m interested in are ways that media (however old or new) help citizens to synchronize their opinions, coordinate their actions, and then document those actions for wider dissemination.

Those mechanisms seem to me to be specific enough to allow for useful comparative research programs; one could ask “What was the difference in the ability of the protestors to document the state’s actions in Tianamen and Tahrir, and what difference did that difference make?” and get a pretty interesting answer, I think.

I also favor jettisoning most talk of “the internet”, because treating it as a thing seems to be obscuring the debate rather than helping it, by encouraging people to believe that, say, the internet and mobile phones are distinct networks, or that YouTube is an internet service whereas al-Jazeera is a television channel. We’ve long since passed the days where the internet was a distinguishable object from mobile phone networks, and as al-Jazeera’s incredible use of digital networks to both source and disseminate reporting demonstrates, we’re entering an era where the separation between different kinds of images is making less and less sense as well.

Even when we get rid of internet, though, as a lable past its Sell By date, I’d like to keep talking about things like social media and things like Facebook, for different reasons.

One can imagine replacing a phrase like social media with a list of more specific mechanisms. I’d replace it with something like “media that a) allows amateur access to public speech, and b) provides a platform for large-scale, asynchronous, and geographically distributed group conversation.” Now, if I was the kind of person who liked to talk about things A and B a lot, I’d blister my fingerprints off having to type that phrase over and over.

For me, talking about social media means asking “What difference does it make to synchronization of opinion, coordination of action, and documentation of results that amateurs can now speak in public in various media that allow for easy group communications?”, and I’ll argue that this kind of talk is not just permissible but useful, provided the phrase social media chunks a list of the kind of things we’d like to think about into an aceptable shorthand.

So that’s one reason to keep at least some of the putatively banned words around; we can make them earn their keep in descriptive compression. Contrariwise, when they don’t earn their keep in that way, as I think internet talk no longer does, we can stop using them.

The other reason, though, to keep at least some of these words around is that the people we are about care about them. Social science ain’t physics, since our atoms think about what it’s like to be an atom; when people are talking about overthrowing Assad, they are also thinking about what it means to be people talking about overthrowing Assad. People can give explanations of their actions, and they adjust those explanations to their actions and vice-versa, in continuous though naive and imperfect ways. (“Shape of…Anthony Giddens!”)

So when people in Tahrir Square hold up hand-lettered signs saying #jan25, or when Ghadaffi asks dissidents to take down their Facebook pages as a condition for reform, we can’t _not_ talk about Twitter or Facebook; they are not just sites of raw social capability, they are objects which serve as particular cultural references.

Even if we say “One mechanism by which Facebook affected the events in North Africa was by providing a generation gap, whereby youth wrongly but profitably believed that they were somehow fundamentally different from their parents’ generation, helping give them the courage to try something that had always failed in the past.” (This is an important part of Slee’s account, though I’ve compressed it badly here.)

In this case, the call for specifying mechanisms makes talking about Facebook _more_ important, because no Cairenes were holding signs reading “Thank you, Cicso” or “Thank you, WordPress”, so there was clearly something specific about Facebook that made it salient where other such cultural references were not. If what you are interested in is not one particular mechanism, of which Facebook might have been a part, but rather the list of all mechanisms where Facebook was a part, well, you are back to studying Facebook, but so long as you are giving and defending some sense of how you think it affected things, then why not?

I agree that we should both demand and indulge in less hand-waving about which entities in the set X affect outcomes; when a concept in X obscures more than it clarifies, out it goes, as with “the internet.” However, rather than throwing out all our old Xs and getting a bunch of new Przeworski- and Teune-approved Xs, I think we should _expand_ the range of X, based on the audience. We can, I think, talk about “information cascades, Facebook’s effect on” in one venue and “Facebook: it’s effect on information cascades” in another, without betraying the need to give and defned reasons in either.

17

Clay Shirky 04.20.11 at 8:52 pm

Err, that’s “people we _care_are about care about them.”

18

Straightwood 04.20.11 at 11:06 pm

The poverty of nomenclature for the digi-whatever-revolution is an important clue as to the intractability of analyzing this collection of phenomena. Zuckerberg himself has no idea of what the omega point for Facebook is. He was the first to stumble through the door of massive global social networking, and now wants to lead a triumphal procession – but to where?

Indeed, it can be argued that Facebook will become the AOL of social media, a launching point for a tremendous broadening of electronic social networking that will be left behind because of conflicts between its revenue model and the evolutionary forces shaping the Noosphere. (Just what is the logic of collecting rent on the fabric of human society?)

One can hear the cognitive gears clashing in our society, full of professionalized expertise, because there are no “experts” in digital social networking; it is still a rapidly exfoliating phenomenon whose historical impact and reach remain unknowable. The best c0mmentary we can expect in this domain is competent journalism. Scholarly analysis must await the emergence of stable forms.

19

Henry 04.21.11 at 9:41 pm

bq. For me, talking about social media means asking “What difference does it make to synchronization of opinion, coordination of action, and documentation of results that amateurs can now speak in public in various media that allow for easy group communications?”, and I’ll argue that this kind of talk is not just permissible but useful, provided the phrase social media chunks a list of the kind of things we’d like to think about into an aceptable shorthand.

This seems fine to me in principle. As noted, it is hard to stick to mechanisms talk all the time. If we can move to a world not of ‘replacing social media names with mechanisms’ but of ‘using social media names while always trying to remember that they are a convenient shorthand for a bunch of mechanisms that are really doing the causal work’ I’d be happy.

bq. Social science ain’t physics, since our atoms think about what it’s like to be an atom; when people are talking about overthrowing Assad, they are also thinking about what it means to be people talking about overthrowing Assad. People can give explanations of their actions, and they adjust those explanations to their actions and vice-versa, in continuous though naive and imperfect ways. (“Shape of…Anthony Giddens!”)

I’m not sure how much we disagree here on the fundamentals, but I think that even if social science isn’t physics, it is more susceptible to materialist explanation than you perhaps seem to be suggesting here. My guide is not Giddens, but Dan Sperber’s brilliant “Explaining Culture: A Naturalistic Approach” (see esp. chapter four on the epidemiology of beliefs). The belief that Facebook has certain causal qualities can itself be causally efficacious – it leads actors to do things that they otherwise would not have done. But this belief is itself a kind of a representation, which can spread to others, e.g. via social networks, perhaps changing in the process of dissemination, due to lossy transmission, interaction with pre-existing beliefs held by some individuals and not others etc. Such processes of intellectual contagion, in which roughly consonant beliefs spread across a population are themselves amenable to study via analysis of the mechanisms of diffusion, transformation etc that are involved.

Zeynep – on the speed thing, if I understand where your thinking is heading correctly, it is an argument about relative speed (some technologies allow e.g. protesters to react more quickly than governments). If that is right, one might plausibly expect the mechanism to apply to other contexts, and other forms of communication than Internet based ones, and this kind of comparison might be useful. If e.g. you end up with an agent-based model to capture this dynamic, it is likely to apply pari passu to other contexts too. So it is perhaps better to think of your argument and putative mechanism not as something about ‘social media’ but about ‘forms of communication with a certain set of asymmetric consequences for centralized and decentralized actors’ or somesuch.

20

LFC 04.21.11 at 11:01 pm

@8: Fox News is part of that medium long ago declared dead, broadcast television

Fox News is not broadcast television; it is cable/satellite television. Someone such as myself who has no cable or satellite subscription or service cannot watch Fox News on his (or her, as the case may be) TV – not that I want to. (Actually at the moment I can’t watch anything on my small flat-screen TV for lack of a sufficiently powerful functioning antenna, but that’s neither here nor there. It’s still possible, if not always easy, to receive digital broadcast channels in the U.S. without having any kind of cable or satellite service or subscription.)

Broadcast television continues to be regulated, however half-heartedly, by the FCC in ways that cable/satellite television is not. Unless Congress repealed the entire Communications Act of 1934 when I wasn’t looking.

21

Straightwood 04.22.11 at 1:55 pm

Broadcast television is a dead thing that has not yet fallen over. The steady increase of Internet bandwidth will make it possible to view high quality video, from thousands of sources, over computer screens, breaking the monopolies of the old commericial broadcast empires. There remains a question of how much political damage the perfection of broadcast propaganda can do before it is displaced by the diverse streams of Internet video, but the outcome is not in doubt. We are watching the final thrashings of the dying reptile of mass electronic broadcast media.

22

LFC 04.22.11 at 8:12 pm

We are watching the final thrashings of the dying reptile of mass electronic broadcast media.

Well this has been debated here before, but I think it is wrong. By this logic, TV should have displaced radio, screens should have displaced books, etc. None of which has happened. And not only do hard-copy newspapers (at least some) still exist, but if you go to, e.g., the Wash Post’s website you’ll see they have a tab called “today’s paper,” clicking on which brings one to an exact facsimile of the hard-copy print edition for that day. Makes sense, e.g. if you’ve read something in the hard copy that you want to link to electronically.

23

Straightwood 04.22.11 at 9:06 pm

Television will not vanish, but it will be displaced as the dominant information medium, just as it displaced radio. There are still hand lettered manuscripts being made for religious purposes, but that does not mean that printing has not displaced manual copying of books. Well-educated people now assemble their news from Internet sources; others will follow their lead. Diversity and precision are the insuperable advantages of interactive Internet media over broadcast, and these advantages will only grow over time. Even the humblest blog can embed a video stream. Not even the richest media baron can make his broadcasts interactive.

24

tomslee 04.22.11 at 9:46 pm

Even the humblest blog can embed a video stream. Not even the richest media baron can make his broadcasts interactive.

Reality competitions such as American Idol or X Factor are interactive. And there is a lot of publicity being given to phenomena such as tweeting during Glee. And Yahoo! has recently shown that 86% of people use mobile devices while watching TV.

Social media — Twitter at least — is beginning to look like a complement to TV, not a substitute. It is pushing viewers away from time-shifting TV programs as they want to be in on the conversation while the show is on. More at ReadWriteWeb.

25

Straightwood 04.22.11 at 10:38 pm

The writing on the wall for all mass media is that we are entering an era in which high status is increasingly associated with intensive usage of information and communication facilities. Conversely, low status is associated with passive viewing of mass broadcast media. Of course there will be attempts to hybridize conventional broadcast media, but you can’t make a train into an automobile. TV will move down-market and retain an audience that is mainly old, sick, poor, or otherwise feeble.

26

tomslee 04.23.11 at 2:30 am

Well if you say so. I’m off to watch TV myself.

27

Substance McGravitas 04.23.11 at 9:10 pm

28

Erhardt Graeff 04.23.11 at 10:54 pm

This has been an interesting article and discussion over the past week. I wanted to add the Web Ecology perspective drafted back in 2009. I think it incorporates a lot of the elements that have been discussed. It tries to be both comprehensive and reflective about limitations, but most importantly it has been useful to us in thinking about our research.

Reimagining Internet Studies: A Web Ecology Perspective
http://www.webecologyproject.org/2009/08/reimagining-internet-studies/

The key elements that we most often come back to regard the value of considering the microscopic and macroscopic views of social phenomena online and multiple layers of the internet. The internet enjoys a complex and dynamic cultural layer atop its many technological layers, and its possible to image that social behaviors might have atomic elements that are connected while being different in nature thus requiring different methodologies to study. For instance, a “like” on Facebook as Zeynep says could represent different things in different contexts, and yet also it might have an identical digital fingerprint to any other “like” in Facebook’s database. Web Ecology tries to consider both of those aspects simultaneously, as both represent valuable data.

We wrote the principles of Web Ecology as both a manifesto of what we thought was core to the way we considered how the web worked, and ought to be studied, but also as a kind of checklist for ourselves to make sure that we weren’t disregarding any valuable approaches to attack questions of online phenomena–since those are quickly lost when you focus on a single conception of what it means to “study the internet”.

29

Zeynep 04.24.11 at 4:44 pm

Greetings everyone from Istanbul! Fascinating conversation; let me try to add a bit more of my thoughts. (Although Clay said many things I would have wanted to add so rest of my remarks assume his contributions.)

Henry, you say:

‘..using social media names while always trying to remember that they are a convenient shorthand for a bunch of mechanisms that are really doing the causal work’ I’d be happy.

I think you would find that level of abstraction wanting because the functioning “those bunch of mechanisms” intrinsically rests upon how, say, Facebook or Twitter operates within the particular media ecology of a country/region/world. For example, here in Turkey, people use Facebook extensively (but tweeting from a phone is relatively rare.) Therefore, say, changes to how Facebook manages “issue pages” will have an impact on means of asynchronous political dissemination methods available to ordinary people. But to understand that, you have to be looking at Facebook and studying how it is used in a particular context, its affordances, it’s policies, etc.

So, Facebook is not a mere short hand for a causal mechanism because we don’t want to constantly type “asynchronous many-to-many connectivity available to digitally-conversant publics” but also a particular mechanism itself that is integrated into the dynamics of larger causal processes. The ecology of social media –and the word “social media” is indeed a shorthand for “asynchronous many-to-many connectivity …”–, includes Facebook *and* Twitter and many other platforms each of which fit separately and function somewhat differently into a larger systemic, the study of which always requires specifics.

Yes, ultimately, we are mostly interested in those causal processes, yes, those causal processes are always mutually-constitutive so that the arrows of change and causality run in multiple, messy, directions (what I call networked causality), but no, we cannot illuminate those processes without studying specifics which will eventually need to incorporate studying not the “Internet” –which is indeed too vague—but Facebook, Twitter, and other platforms, as appropriate.

Also, I’d argue that the issue with social science not being physics isn’t that we cannot incorporate materialist explanations into our work, but that the reflexive nature of the subjects of our field means that we cannot just import physics-y methods as is without losing a fundamental mechanism: the capacity the self-aware human to respond to what she knows other know and what others know she knows etc., and to also have concurrent meta-analyses of her own situation and her actions. I find physics-y and epidemiological-y methods to be illuminating to a degree, especially due to their greater capacity to take into account the networked nature social action, but I think we need to remain aware of the particular ways in which they really do fall short.

Also, on the question of causality: I wrote a blog post back when there was consternation among many people about the word “Twitter revolution”. It can be found here: http://technosociology.org/?p=263 . Basically, I use Aristotle’s levels of causality as a not-too-strict metaphor for distinguishing between different levels of causal analysis.

Finally, let me say a few words about the issue of speed. Henry says:

Zeynep – on the speed thing, if I understand where your thinking is heading correctly, it is an argument about relative speed (some technologies allow e.g. protesters to react more quickly than governments). If that is right, one might plausibly expect the mechanism to apply to other contexts, and other forms of communication than Internet based ones, and this kind of comparison might be useful. If e.g. you end up with an agent-based model to capture this dynamic, it is likely to apply pari passu to other contexts too.

That is not exactly what I am proposing, although, yes, one way to formalize what I am about to expand upon would be to use comparative agent-based models.

My main argument is that states are resource-constrained actors and they simply cannot crush widespread dissent and maintain a stable autocracy. (Yes, blood flowing in the streets can crush widespread dissent but no country can function semi-normally in that situation for long: civil wars lead to failed states). In response to this problem, Middle Eastern autocracies had perfected the art of the “whack-a-protest” which often followed a set script following an uprising or a protest: initial massive force, kicking out journalists, cordoning the area, jailing/killing leaders, relaxing the level of force and waiting it out – often followed by concessions some time *after* the protest had finally died out. That way, conditions often improve a little after protests to head off the next uprising but not because the protestors won but because they lost.

Of course, as I said, speed does not operate by itself but is integrated into the shape of the connectivity network. But speed itself is a factor. The whack-a-protest model fundamentally depends upon both the lack of many-to-many connectivity between ordinary people and also slow, bottle-necked information diffusion. By the time reports of Gafsa uprising, (Tunisia in 2008) became widespread, the protest was also half-crushed. In Sidi-Bouzid, the time between the beginning of the events to the spread of the news was near simultaneous.

Thus, speed of information diffusion doesn’t mean that protestors are quicker than the government, it also means that the protestors are *more* than the government. Again, states are resource-constrained actors. If an uprising can spread, a state cannot simultaneously cordon off and use overwhelming force in multiple places. (This is why one can create formal models, including agent-based models, to try to figure out comparative thresholds).

Without being cordoned off, and with the capacity to create conditions of mass uprisings, protestors can overwhelm state resources. In sum, all states can jail a few thousand people; no state can jail millions of people all at once.

Attention is also a mechanism affected by speed. Rapid diffusion and connectivity means that the whole country focuses its attention *all at once* to their collective displeasure with the existing system. A slow information cascade creates qualitatively different social dynamics compared with a fast information cascade. So, yes, in some sense, this is an information cascade but it is also more than that because of the reflexive nature of humans. It’s not just that a piece of information or a disease is traveling fast but there is also a “field” effect, something epidemiological models do not take into account. A field effect is when “something is in the air” – not just because your neighbor told you or you watched it in the news but it is everywhere. Countries have moods.

In other words, when everyone knows what everyone knows *almost simultaneously*, this is more than contagion or diffusion; network effects and field effects start operating simultaneously and strengthening one another. It’s a revolution.

30

Kevin Donovan 04.24.11 at 5:16 pm

Let’s not forget that the lawyers had this same debate more than a decade ago: http://en.wikipedia.org/wiki/Law_of_the_Horse

31

Straightwood 04.25.11 at 12:46 am

The dominant myopia regarding the Internet is that it will not eventually subsume existing social and political structures. The easy extrapolation of the RFC mechanism into the generation of law and social conventions seems to have eluded the commentariat. The biggest thing since Gutenberg has yet to be recognized widely as an epochal change that will reorder all aspects of human affairs.

When Internet-based political systems arise, all those currently dismissive of the Internet’s transformative potential will blandly acknowledge that “everybody knew” that the Internet would bring profound change. That’s how conventional wisdom handles reality recognition. What is heresy Friday afternoon becomes the standard doctrine Monday morning.

Comments on this entry are closed.