For many centuries philosophers, political scientists and lawyers have taken for granted that there is a reasonably clear distinction between the public and private spheres and the nature of “speech” in each space.
The private sphere—i.e., speaking with friends in your home—was a sphere of very limited audience and unlimited speech. You were free to say on- or off-the-record almost anything you wanted, but your audience was limited to the people you could pack in your living room.
This stood in stark contrast to the public sphere—speaking on a stage or to a mass audience through a broadcast—where your audience could be of almost unlimited size. But your speech was limited to the messages which would appeal to a diversity of people and you had to be OK with everyone knowing what you said.
In its first iteration, the largely anonymous internet of webpages looked like a simple, if powerful, extension of the public sphere. It allowed individuals to speak to more people than previous mediums did, on a far wider diversity of topics.
But, as the internet has become ever more personalized, it increasingly represents a strange hybrid of the public and private spheres. Through micro-targeting and customization, the internet now provides the opportunity for people to reach an unlimited audience with unlimited speech.
If there is, therefore, a conversation to be had about the impact of the internet on the election, it shouldn’t be about fake news and feed ranking (which I believe are red herrings). It should be about what it means that a public candidate can for the first time effectively talk to each individual voter privately in their own home and tell them exactly what they want to hear.
Clearly, this merging of the public and private spheres represents a new space for speech which is powerful, personalized and un-audited. We clearly don’t yet have good frameworks for managing this space. And, frankly, we need as a society to get a grip on managing it if we want democracy to thrive going forward.
The Technology Tools of Mass Private Speech
Mass private speech isn’t entirely new. Phone banking, where people dial a pre-determined list on a candidate’s behalf, is a crude, low-data form of a candidate talking privately with a string of individuals through a surrogate. Phone banking is expensive to run, though. And because intermediate people make phone calls on your behalf, the customization of your message can’t be very personalized.
Good old-fashioned mail and email have also been micro-targeting tools for a long time. The companies and movements that have the best data and are best at targeting can present themselves in massively different ways to different people.
Recently, there has been some attention paid to Facebook page posts, both free and sponsored, which can be limited with audience restrictions to only show specific messages to specific people. This is a powerful evolution from prior tools and there are good reasons for the product to exist. If you are a brand like Nike with a global voice you obviously want to be able to deliver different messages to people in the U.S. versus the U.K. And you want to deliver messages in French to people in France and in German to people in Germany. You also likely want to be able to target messages differently for men and women, and based on interests in different sports. This is how brands work when they buy ads in different cities or against different TV shows—it’s just far more effective with Facebook’s data.
Of course, once you open up the ability to customize whom you are speaking to, it is easy to go further with micro-targeting. Brands want a way of—and get a lot of value from—marrying data with purchase histories, web traffic, email directories, ad nauseum, etc. Of course, so do sophisticated politicians who know how to use the tools.
Customized Facebook ads and page posts might be easy to see as technology of mass private speech, but they are hardly the only tools at play. Most websites, which once showed more or less the same content to everyone, are now hyper-personalized based on your interaction history, interests and the tracking pixels all over the web. With rare exceptions like Wikipedia, what you see and what I see when I visit webpages is massively different—whether we are on SoundCloud or BuzzFeed.
Disappearing messages on Snapchat and Instagram only further exacerbate this trend. They allow messages to not only be targeted to specific people, but give the speaker the ability to make it difficult for the recipient to screenshot or share the message they are presenting. The privacy-oriented features that might have been originally intended to give teens the feeling of safety to speak more freely to their friends also have major payoffs to public figures who want more freedom in the extreme messages they are dispatching to individuals. These features allow speech to not only be personalized, but make it almost impossible for the speech to be audited by outsiders.
Next up likely will at some point be experiences with conversation agents and perhaps in things like VR. Your virtual reality experience, powered by intimate knowledge of who you are, will be very different from mine. When I talk to a chatbot, the suggestions it will make to me will increasingly diverge from what it says when you talk to the same chatbot.
The upshot is simple. Lots of data, and systems which can react properly to the interests, beliefs, and feelings of different people lead to a world where technology and brands tell us exactly what we want to hear in a way that can’t be tracked or audited. Tools intended originally to give individuals a more private space to share with friends are easily co-opted by public figures looking to speak in more extremes. And just as critically, we have absolutely no idea what the same voices are saying to everyone else.
Where Has the Public Sphere Gone?
It is always interesting when a candidate’s private speech to a small group of donors leaks to the public. It is a rare glimpse of the targeted message that candidates deliver to a hyper-targeted audience. I have been to events like this, and while I have never heard a “basket of deplorables” line from a candidate in private, I can tell you that they say different things in smaller groups (and are usually way more compelling to their targeted audience).
Take that concept, and now imagine that the candidate (or brand)—with good technology and data—can speak to each person individually in their home. They can say whatever they want with reasonable confidence that their words will not leave the room. In a world like this, why would they ever give public speeches or start a public discourse? It would be far better to have a customized and personal message for each person and then say nothing publicly.
I believe, at least in part, that is what we saw in this last election cycle. At a public level, in big speeches broadcast to everyone, neither candidate really said all that much. The debates were shocking mostly due to the fact that there were almost no really deep discussions of policy or viewpoints.
There are rumors that candidates were making bold and aggressive claims to small targeted populations away from the eye of the broad-based public via private, small groups, and one-on-ones, as facilitated by the internet. The challenge is, of course, that because of the nature of the speech it is hard to really know.
That said, it did feel like the goal of the public discussion became to say as little as possible to allow the candidates full breathing room to say in private exactly what they wanted.
The technorati, along with lots of media pundits, have been bemoaning the idea that we now live in “feed bubbles” and “echo chambers” where we only see the viewpoints of like-minded people and miss the broader narrative. To me, feed bubbles was the discussion to have five years ago. The far more complicated and insidious problem of today is personal, private and disappearing messaging that can be powerful but can’t be broadly traced or audited.
What to Do
Democracy, so the theory goes, only works with a vibrant public discourse. It requires multiple candidates putting forth to an electorate ideas and positions and competing for votes against one another in public.
If we are entering a world where elections are done via deeply informed, mass one-on-one conversations versus public sphere speeches, what can we do to preserve the discourse?
It would be both silly, and likely impossible, to somehow regulate the use of private speech or micro-targeting in elections. But what we could do is require that all the data and communication from all campaigns must be public record.
If all the data that campaigns use to craft messages for individuals, as well as all the things that candidates say to individuals, were forced to be public record, then at least watchdogs could figure out which people were being fed lies.
This will cause upheaval. It will mean that, at least for political candidates, there can no longer be such a thing as a private fundraiser dinner. It will also mean that anything they say to any individual citizen in private, and the data they collect on individuals, will have to be available to be shared with everyone.
That is a pretty big change from the way the world has worked to date. It basically would mean that candidates need to sacrifice any access to the private sphere they once had.
But, to me, this seems not only feasible but necessary. We cannot allow campaigns to be waged in private.
I believe that we will look back at this period and laugh about our distraction with sideshows like “fake news” and “feed bubbles.” The real story of this period is how the internet, and the world, is moving away from feeds and toward private messaging, groups, disappearing messages and closed spaces, and how successful brands (and candidates) are customizing their messages in an un-audited space specifically for those groups—not for re-shares in more public spaces.