platforms and institutions

In the new edition of his book on the modern Left, which I review here, Roger Scruton writes,

Occasional lip service is paid to a future state of ‘emancipation’, ‘equality’ or ‘social justice’. But those terms are seldom lifted out of the realm of abstractions, or subjected to serious examination. They are not, as a rule, used to describe an imagined social order that their advocates are prepared to justify. Instead they are given a purely negative application. They are used to condemn every mediating institution, every imperfect association, every flawed attempt that human beings might have made, to live together without violence and with due respect for law.

Like Scruton and most other old-school conservatives, I believe that healthy mediating institutions are essential to a healthy society. And I think he is right in noting how relentlessly the Left attacks such institutions. But international capitalism does too, because every healthy mediating institution, by providing security and fellowship and belonging to its members, reduces its members’ dependence for their flourishing on what can be bought and sold. Neither the Left nor the Market want to see such institutions flourish, though their hostility sometimes stems from different agendas.

I’m usually allergic to generalizations in these matters, but let me risk a big generalization: I think what we have seen and will continue to see in our social order is the fragmentation of institutions and their effective replacement by platforms.

Let’s take education as an example: for much of American history people were educated in a wide range of (often highly eccentric) ways. This was generally perceived as a problem, and efforts at standardization kicked in, reaching their peak in the Sixties. Since then we have seen increasing fragmentation, with ordinary public schools, charter schools, magnet schools, various kinds of private schools, homeschooling, unschooling … but all of these work on the same platforms, that is, they rely on the same communications technologies, using either the open web or walled gardens like Facebook in order to promote interaction and accomplish goals (e.g., the completion of projects and other assignments, remedial tutoring, etc.). We will more and more be asking technological platforms to do the kind of unifying work that educational institutions can clearly no longer do, which, I believe, is asking platforms to do things that by their nature they’re unsuited to do.

They’re unsuited to do it becasuse platforms are unresponsive to their users, and unresponsive by design (design that emerges from their desire to be universal in scope). It is virtually impossible to contact anyone at Google or Facebook or Twitter or Instagram, and that is so that those platforms can train us to do what they want us to do, rather than be accountable to our desires and needs. A model of education tied to platforms rather than institutions may seem liberating at first — “I can learn everything I need to know at Khan Academy!” — but that sense of liberation will continue only insofar as users train themselves to ask the questions the platforms already know how to answer, and think the thoughts that the platforms are prepared to transmit.

Very few people will see any of this as problematic, and only those very few will look to work outside the shaping power of the dominant platforms. This means that such institution-building as they manage will have to happen on a small scale and within limited geographical areas. As far as I’m concerned that’s not the worst thing that could happen.

But the majority will accommodate themselves to the faceless inflexibility of platforms, and will become less and less capable of seeing the virtues of institutions, on any scale. One consequence of that accommodation, I believe, will be an increasing impatience with representative democracy, and an accompanying desire to replace political institutions with platform-based decision-making: referendums and plebiscites, conducted at as high a level as possible (national, or in the case of the EU, transnational). Which will bring, among other things, the exploitation of communities and natural resources by people who will never see or know anything about what they are exploiting. The scope of local action will therefore be diminished, and will come under increasing threat of what we might call, borrowing a phrase from Einstein, spooky action at a distance.

I for one don’t welcome our new algorithmic overlords.

the fool on the hill

Is this a stunt? Of course it’s a stunt, as it was when James Sturm did it a few years ago, though that was less Walden-esque, since Sturm has a family. Also, Thoreau’s life at Walden Pond was kind of a stunt too, since, despite the impression of absolute solitude he tried to give, he would regularly walk into town to visit friends and family.

Still, the fact that it’s a stunt doesn’t mean that it’s not a socially useful act. I think of those lines from Dylan’s “Property of Jesus”: “Go ahead and talk about him because he makes you doubt / Because he has denied himself the things that you can’t live without….” Of course I could live without the internet if I had to! Mmm-hmmm. But it’s not possible right now, not with the job I have … and it’s not like there are even any jobs out there that don’t require the use of the internet…. Mmm-hmmm. Please do go on.

It’s interesting to find out what we tell ourselves when confronted with the possibility that we could live Otherwise. And being forced into those consoling narratives is probably useful to us, in much the same way that it was spiritually and morally useful to kings to have jesters who reminded them of their mortality and fallibility.

In fact, maybe every town should hire someone to live in their midst without any connection to the internet. (In Tim Richardson’s wonderful book The Arcadian Friends he describes eighteenth-century aristocrats who, with sometimes comic results, hired hermits to live in the picturesque little hermitages they had built on their massive properties.) We would pass by them every day and perhaps be moved to genuine reflection on our choices, especially the choices that we constantly tell ourselves aren’t really choices at all.

And now, as we begin our day, as each of us prays to an image of his image of himself, let’s sing — softly now, softly —

Day after day
Alone on a hill
The man with no internet
Is keeping perfectly still
But nobody ever texts him
They can see that he’s just a fool
And he never goes on Twitter
But the fool on the hill
Sees the web going down
And the eyes in his head
See see us all spinning round…

open letter to Adam Roberts on the Protocols of the Elders of the Internet

This started as a reply to a comment Adam made on my previous post. But then it underwent gigantism.

Adam, I see these matters a little differently than you do — let’s see if I can find out why. I’ll start with cars. I’d say that the main thing that makes it possible for there to be an enormous variety of automobiles is the road. A road is an immensely powerful platform — in this case literally a platform — because it is so simple. Anything that walks, runs, or rolls can use it, which causes problems sometimes, as when a cow wanders onto the street; but for the most part that openness to multiple uses makes it an indispensable technology. Even if your British automobile has its steering wheel on the wrong side, you can still drive it in France or Germany.

By contrast, railway lines are rather less useful because of the problem called break of gauge, which has in the past forced people to get off one train at a national boundary and get on another one that fits the gauge of the tracks in that country. (And of course for a period there was no standardization of gauge even in England.)

Notice that the lack of standardization only becomes a problem when you get beyond your locality — but that’s precisely what mechanical transportation is for: to take us away from our homes. The technology’s power creates its problems, which new technologies must often be devised to fix.

All of these difficulties are dramatically magnified when we get to the the internet, which used to be called, in a significant nickname, the “information superhighway.” Here it seems to me that we have an ongoing struggle between the differentiation that arises from economic competition and the standardization that “platforms” are always looking for. PC makers want you to buy PCs, while Apple wants you to buy Macs, so they differentiate themselves from one another; but Microsoft just wants you to use Word and Excel and PowerPoint and so makes that software for both platforms. Though they’d prefer a world in which every computer ran Windows, they have enough of an interest in standardization to make cross-platform applications. And now they make web versions of their apps that you can use even on Linux.

They do that by employing protocols that were designed back in the era when there were few computers in the world but the ones that existed were made using a variety of architectures and a wide array of parts. When you email the Word file you created on your PC to a Mac user like me, you do so using those same protocols, and updated versions of the communications lines that were first laid out more than a century ago. So again: a tension, in purely technical terms, between variability and standardization.

This tension may be seen in other digital technologies as well. Board games all employ the same basic and highly flexible technological platform: paper and ink (with a bit of plastic or metal, perhaps, though those flourishes are unnecessary). But when it comes to video games, contra Roberts, everyone does not have an Xbox. My house is a PlayStation house. Which means that there are some games that we can’t play, and that’s the way the console makers like it: Microsoft wants us to choose Xbox and stick with it; Sony wants us to choose PlayStation and stick with it. However, the makers of video games don’t like seeing their markets artificially reduced in this way, and so, if they have the resources to do so, will make their games available on multiple platforms, and then will use those internet protocols mentioned above to enable players to play with and against each other.

But the most popular games will always be online ones, because they allow almost anyone who has a computer and an internet connection to play, and to interact with one another: like the makers of board games, they look for the broadest possible platform — but they also encourage us to look beyond the local, indeed to ignore locality when playing with others (players typically have no idea where their opponents and teammates are).

So digital technologies, like the mechanical transportation technologies mentioned earlier, are meant to transcend locality, to remove emplacement as a limitation on sociability. (You can typically only play board games with people who are in the same room with you — though it should be noted that there was a longstanding if not almost abandoned tradition of playing chess by mail.) But this can only be accomplished with the development of either (a) standardized platforms or (b) shared protocols designed to bridge the gaps created by platform variability. Thus Google wants to solve that problem you have sharing photos with your wife: you use the Android version of the Google Photos app, she uses the iOS version, and presto! Solution achieved.

So — still working this through, please bear with me — let’s look at Twitter in light of this analysis. Twitter may be understood as a platform-agnostic MMORPG which, like most other MMORPGs, relies on the standard set of internet protocols, and therefore exchanges data with everything else that uses those protocols. This means that while Roberts is once again wrong when he says that everyone is on Twitter — it has maybe 20% as many users as Facebook and 80% as many as Instagram — those larger platforms, and of course the great vastness of the open web, can be used to magnify the influence of anything tweeted. In that sense there are a great many people in the world who, despite not being on Twitter, are on Twitter. So while some people blame Trump’s skillful Twitter provocations for both his political success and the debasing of our political culture, and others place the blame on the fake-news-wholesaling of Facebook, in fact the two work together, along with Google’s algorithms. In these matters I’m a conspiracy theorist, and I blame the Protocols of the Elders of the Internet.

So, Adam, in your response you referred to “homogenization,” whereas I’ve been referring to “standardization.” At this point we have the standardization of practices without the homogenization of ideology — and that’s the source of all of our conflicts. The platforms that allow us to connect with like-minded people are equally open to people whose ideas we despise, and we have no reliable means of shutting them out; but the encoded, baked-in tendencies of Twitter as a platform are universally distributed, which means that whether you’re a SJW or an alt-righty, you’re probably going to respond to people you disagree by instantaneous minimalist sneering. (The tendencies of early print culture were rather different, but produced a similar degree of hostility, which I’ve discussed in this post.)

I think, though, that this conflict between standardization and homogenization could be a temporary state of affairs, at least for people who rely on the Protocols, and that means most of us. That is, while standardization does not inevitably produce homogeneity, it certainly nudges everyone strongly in that direction. There’s no way that public opinion in the U.S. about same-sex marriage could have changed so quickly without social media. TV certainly had a significant influence, but social media are collectively a powerful force-and-speed multiplier for opinion alteration. And if you feel good about that, then you might consider how social media have also nudged tens of millions of Americans towards profound fear of immigrants.

So in this environment, majority opinions and opinions that are held very strongly by sizable minorities are going to be the chief beneficiaries. And that could lead ultimately to significantly increased homogeneity of opinion, a homogeneity that you only stand a chance of avoiding if you minimize your exposure to the Protocols. And here, Adam, it seems to me that your novel New Model Army is disturbingly relevant.

For much of the novel, the soldiers who fight for Pantegral are independent, free agents. When they fight, they fight according to, yes, protocols established and enforced by the software they all use — but they can stop fighting when they want to, they come and go. Indeed, this is one of the chief appeals to them of the New Model Army: it doesn’t own them. Or doesn’t at first; or doesn’t seem to. In the end the protocols prove to be more coercively powerful (or should I say more intensely desirable?) than they had ever expected. And then we have homogeneity indeed — on a truly gigantic scale.

Caveat lector, is what I’m saying.

traffic patterns

Metafilter is in financial trouble, largely because it can’t stay in good favor with Google. Here’s more on the current internet:

Metafilter came from two or three internets ago, when a website’s core audience — people showing up there every day or every week, directly — was its main source of visitors. Google might bless a site with new visitors or take them away. Either way, it was still possible for a site’s fundamentals to be strong, independent of extremely large outside referrers. What’s so disconcerting now is that the new sources of readership, the apps and sites people check every day and which lead people to new posts and stories, make up a majority of total readership, and they’re utterly unpredictable (they’re also bigger, always bigger, every new internet is bigger). People still visit sites directly, but less. Sites still link to one another, but with diminishing results. A site that doesn’t care about Facebook will nonetheless come to depend on Facebook, and if Facebook changes how Newsfeed works, or how its app works, a large fraction of total traffic could appear or disappear very quickly.

The same variability afflicts a little blog like this one. For instance, here are the views for a series of posts from earlier this month:

  • 297
  • 904
  • 3810
  • 298
  • 822
  • 8498
  • 913

The variation depends on whether the posts are linked to by more popular venues, which can include blogs — I get traffic from Andrew Sullivan and Rod Dreher, for instance — but are more likely to be sites that people don’t even visit directly, but access via Facebook pages or Twitter accounts. It’s a weird world we’re writing and reading in. I liked some of the older internets better. And I wish Metafilter well.


A few days ago I read an interesting article in the New Republic on “trigger warnings” in college syllabi. The topic intrigued me, and I decided I wanted to write about it — but then I got busy. Way too much to do. And by the time I got a few minutes to think about what I wanted to say, so many people have written about it that it didn’t seem worth my time to add my two cents to an already enormous pile of pennies.

This relieved me greatly. And then I wondered why it did.

So I thought about it, and I came to the conclusion that I didn’t really want to write a post on this topic — not really, not in my heart of hearts — but felt some inchoate obligation to do so. It just seems like the sort of thing about which I ought to have an opinion that I ought to be able to state. But that’s silly. There’s no reason whatsoever for me to opine about this. And yet only a period of intense busyness kept me from rushing to my computer to commit opinionizing all over the internet.

I think there’s something to learn from this experience. For one thing, it enables me to see more clearly what we all know already: that when I see a topic being tossed around a lot on blogs and on Twitter, it’s easy to be swept along by that tide. I was looking the other day at the mute filters I have set up for my Twitter client, and I couldn’t help laughing at how many of them provided a record of those brief enthusiasms that take over Twitter for a day or two or three and then disappear forever. It took me a minute to remember who Todd Akin is. It took me even longer to figure out why I had added the word “tampon” to my mute list, but I finally remembered that time when Melissa Harris-Perry was wearing tampon earrings and everybody on Twitter had something to say about that. This is why some Twitter clients have mute filters that can be set for a limited time: I would imagine that three days would almost always be sufficient. Then the tide would have passed, and would be unlikely ever to return.

But I learned something else from this experience also: you can actually use the speed of the Internet to prevent you from wasting your time – or maybe I shouldn’t say wasting it, but rather using it in a less-than-ideal fashion. If you just wait 48 or 72 hours, someone you follow on Twitter will almost certainly either write or link to a post which makes the very argument that you would have made if you had been quick off the mark.

For me, these realizations – which might not be new to any of you – are helpful. They remind me to give a topic a chance to cycle through the Internet for a few days, so I can find who has written wisely about it and point others to that person; and, if there are things that haven’t been said that need to be said, I can address them from a more informed perspective and with a few days’ reflection under my belt. I can also practice the discipline — or maybe it’s a luxury rather than a discipline — of thinking longer thoughts about more challenging issues than are raised by than Melissa Harris-Perry’s earrings. Or even trigger warnings.

TV, still

George W. S. Trow’s “Within the Context of No Context” was a really famous essay at one time, and thought to be precisely diagnostic of a culture shaped and sustained by television, but no one talks about it much anymore because we feel that the internet is the thing now and TV (while still powerful) secondary and dependent on online conversation to sustain it.

But maybe just because of the internet, and before it, for a few years, personal computers, TV is more important than ever. In many countries people are watching more of it than ever — maybe it has new sources of impetus, new energies feeding its energies. The internet is the child of the PC which is the child of television. The PC took hold in large part because it had a monitor and we had already grown accustomed to receiving both education and entertainment via CRT. Perhaps until the internet of things reaches its full maturity we’re all still really watching TV, just with a keyboard attached. That is, maybe the interactive element (the keyboard, or for that matter voice recognition) isn’t as important as the monitor that occupies our eyes. In that case maybe Trow’s essay is diagnostic of a world that succeeded the one he wrote about.

A small thought experiment: What if the PC had been invented before the television? Or no one had thought to use a CRT as a device by which we might “monitor” — the noun has a verb in it — what the computer is doing? Maybe then using a computer would be more like writing a letter and receiving one in return, or using a Ouija board, or sending and receiving telegraphs at the Western Union office. If any of those things had happened, how would we understand our experiences with computers? And what would be our ruling metaphors for mind and thought?

“internet addiction” and other fictions

Imagine a child who’s on her computer all day long because she’s playing World of Warcraft. Imagine another who is talking with friends and looking at photos on Facebook. Imagine a third relentless in his pursuit of hi-def porn videos. And imagine a fourth — this will be hard, I know — who is learning how to code and has found a great many like-minded people in the coding subreddit. Taken together these kids make it clear why there’s no such thing as internet addiction.

“But they’re online all the time and we can’t get them to interact with real human beings!” their parents shout, in unison. “They’re addicted to screens!” To which the proper answer is, Well, some of them are interacting with real human beings, for better or worse, and — more important — none of them is addicted to screens. They’re addicted to certain experiences that they are getting access to via their computers. That is not at all the same thing.

Indeed, no one has ever been addicted to screens, or even addicted to the internet. Neither screens as such nor the internet as such have the power to enrapture people. Now, some might reply that this is a frivolous response, that such language is an easily-understood shorthand. But I would counter that it’s a highly misleading shorthand, because it teaches us to conflate extremely different experiences, to place them under a single umbrella, which is exactly where they don’t belong. Each of the four children I described above is moved by very different interests; and indeed, it would be easy to write more detailed stories for each of them that would show their behavior to be, in particular contexts, either more or less culpable or worrisome than my current bare-bones narration makes them sound.

So my suggestion would be this: whenever you hear that someone is suffering from “internet addiction,” just ask what, precisely, is it on the internet that he or she is addicted to. And get them to define addiction while you’re at it. And treat the whole exchange as the beginning of a conversation about a person, not a definitive diagnosis.

what and who

Returning to an earlier theme, and having read the comments on this follow-up post by Joe Carter, I’d like to note a couple of points many people fail to understand about online anonymity:

1) There’s an enormous difference between anonymity and pseudonymity: the person who posts or comments under a consistent pseudonym is assuming a level of responsibility for his or her words that the anonymous poster does not. Consider Yoni Appelbaum, who commented widely, but especially at The Atlantic, for a long time under the moniker “Cynic” — and thereby got himself a job blogging for The Atlantic. Everybody who read that site knew who Cynic was, could respond to him directly either in agreement or disagreement, could point out what what he said in one comment contradicted what he said in another, and so on. Any conversation with a completely anonymous poster is comparatively impoverished. Indeed, if you have ten anonymous comments you can’t know whether you’re dealing with one person or ten different people. Thus sock puppetry and the like are born.

2) People like to say that what matters is the quality of the ideas, not the person who utters them. But suppose the topic is something you don’t know much about — a subject requiring certain technical expertise — and you’re not sure how to assess the varying positions. In such a case it helps to know who is making the arguments. Consider the debate on James Fallows’s blog a few months back about the likely effects on human health of the radiation emitted by the TSA’s backscatter-radiation machines. Fallows can affirm that the writer he’s quoting is “a physics professor from a college in the East,” and while I’d like to know who he is and what college he’s employed by, even that much gives me reason to take the argument seriously. If the same argument had been presented anonymously in a comment thread, why should anyone take it seriously? why should anyone even read it? If I made the argument why should anyone pay attention?

Now, the fact that an unnamed physics professor made a set of claims did not settle the question — but knowing even a little bit about the physicist, and the qualifications of the person disagreeing with him, helps us to think more clearly about the issues raised. One of the really interesting questions raised is: What sort of scientist would be a reliable source about backscatter-radiation machines? Unless you think that on every conceivable subject one person’s opinion is as good as any other’s, or that eloquence alone counts, you need to think not just about what’s being said but also who’s saying it.

It’s not the only factor — in many cases it won’t be the most important factor — but it helps, because knowledge is good, and the more of it we have the better off we are. Even when we’re arguing online. Remember that it doesn’t count for much when Woody Allen himself (or his stand-in Alvy Singer) tells the obnoxious blowhard in the movie line that he doesn’t understand Marshall McLuhan; but when McLuhan himself shows up and weighs in. . . .

anonymity revisited

One of the most common defenses of anonymity on the internet is the one offered by Daniel Solove here:

For all its vices, anonymity has many virtues. With anonymity, people can be free to express unpopular ideas and be critical of people in power without risking retaliation or opprobrium. The anonymity in everyday life enables people to be free to do many worthwhile things without feeling inhibited.

I have no idea what that second sentence means — what “worthwhile things” are people “inhibited” from doing when their names are known? — but the previous sentence I get. And it’s true. But only in certain narrow circumstances. The problem is that over the years I have heard from many people who insist on anonymity in order to protect themselves from “reprisals” when in fact all they’re going to suffer is disagreement. And grownups ought to be able to deal with being disagreed with.

Moreover, every protest against injustice is far more meaningful when the person making it is willing to sign his or her name to it. As the literary theorist Mikhail Bakhtin pointed out long ago, in his early work Art and Answerability, to undersign a statement with one’s own name is a powerful act — an act of commitment, responsibility: one becomes “answerable” for it. This is a strong witness to others. Anonymous dissent, by contrast, is often empty to others because no one is answerable to it. Anonymous dissent requires numbers to have an effect. When many protest anonymously their position gains weight additively; the single anonymous protester comes off as a crank or a troll.

Anonymity on the internet may be desirable often but it is necessary only rarely, and surely in 98% of the cases in which it is invoked the conversation would be better when conducted by answerable individuals. Therefore anonymity ought to be hard to achieve. If “undersigning” is the online norm, then people whose hatreds and resentments are just casual will be less likely to trash conversational spaces. The people who genuinely need anonymity can create subterfuges when they need to; even when the use of real names is mandated, that demand is relatively easy to circumvent. On the internet, nobody has to know you’re a dog.