II. The Epistemological problem
Part two of “Can Google and Facebook Save the Republic?”
We live in a fundamental epistemological conundrum called subjectivity. I see things one way; you see them another. Is the dress blue with black trim or white with gold trim. We have methodologies to try to contend with this. Logic is one, wherein we establish a set of rules by which self contradiction can be detected and false assertions winnowed out. All oranges are grown in California. This is an orange. It was grown in Florida. One or more of these statements must be false.
The scientific method is another — a means for comparison of perceptions that allows us to draw conclusions about the physical world.
Now, these are very helpful methods, but they are really very marginal in their day-to-day application. We make most of our judgments in some other way. A big chunk involve induction — I have a bunch of perceptions of something and infer from that repeated experience some characteristic of the object. The sun rises every day, so I expect the sun will always rise. All the oranges I’ve eaten are from California, so I infer that all oranges come from California.
The other major component of our judgment comes from authority. We receive information from a source — a newspaper, a parent, a friend — and accept it uncritically because we trust the source. A good example of my own: if someone asks me why I believe anthropogenic global warming is really happening, I explain my reasons as follows:
- I recognize that I do not have time to thoroughly review existing climate data, and…
- The vast majority of people who do so for a living have come to a consensus in favor of anthropogenic warming, and…
- Given #1, the most reliable thing I can do is trust #2.
- This is not a perfect solution, but it’s the best I can come up with.
Usually, the rebuttal to this argument is some flavor of “Well, I did have time to investigate climate data, and I discovered that…” Even letting alone that the subsequent narrative of discovery fails on its face to be compelling, these people don’t seem to realize they are merely trying to convince me to leave off belief in one authority — the combined efforts of climate scientists — and instead put my faith in some guy, maybe some guy I just met, which as a solution has a lot less going for it. (Look, I’m all in favor of arrogantly assuming I can learn anyone’s trade and excel at it, but come on, even superheroes have some limitations.)
So here’s the awkward reality: I grew up in a hotbed of libertarianism and required over four decades of life to become convinced that I might want to have my hair cut regularly. I am also devoted to Thomas Aquinas’ dictum that the argument from authority is the weakest form of argument. Nevertheless, I am telling you, for the most part in life, the weakest is all we’ve got.
For this reason, both highly educated coastal liberals and middle American blue collar workers must regularly appeal to a source — on climate change, on political malfeasance, on questions of morality. The selection of authorities depends on induction: which of these sources appears to give an accounting of life that corresponds to the broadest set of my experiences? (Remember that my internal ideation and emotions are every bit as much a part of my experience as events that happen in the world around me.) I trust the consortium of climate scientists because I know what it’s like to get a degree in college. I know what studying science is like. I know how math can help make judgments about data. I know what it’s like to develop a trade or skill and become adept at it. I know that it’s difficult to get 100 people in a field to agree about something. At least in my experience.
We do similar calculations when choosing what news sources to trust. This or that source says things I know, from experience, to be true or false, therefore I trust/mistrust the other things they say. Obviously, there is a feedback mechanism at work, too. I start to trust this source; it tells me things; I believe those things and incorporate them into a world view; I begin to mistrust sources that contradict this view.
In other words, we have been constructing our own filter bubbles from long before there were recommendation algorithms. Technology has simply made the bubble more efficient, automated, and of ever tighter radius. Add to technological change the advance of income inequality and our geographic self-sorting, and one can find oneself never bumping into an upsetting opinion. If a long run of such luck ends abruptly, one may feel surprised, upset, and even traumatized. So not only have we seen a growing inability of right and left to talk constructively, but even liberals and leftists cannot talk constructively, even right and alt-right cannot talk constructively. The “acceptable” portion of humanity is a thinner and thinner slice.
I don’t think it hysterical to say that we’re slouching towards civic catastrophe. We’re making it easier and easier to dehumanize larger and larger swaths of humanity. When you read the kind of racially charged commentary coming from alt-right Trump supporters or even the more mainstream commentaries arguing that there is no such thing as a good Trump supporter, you are watching people perform the mental exercises that prepare one for genocide or civil war — the public ritual of stripping the scapegoat of all human virtue. I don’t see a historical precedent for how that ends well.
So naturally we want to come up with ways to reverse these trends. We try to think of ways to push together these disparate tribes who can barely tolerate each other for the duration of a holiday meal, and thereby grow their empathy one for the other. We look at why housing has gotten so expensive in the areas where affluent white liberals congregate. We try to think of ways to broaden or shape the kind of information that each group ingests and considers.
But we all know that if an enterprise like Google or Facebook were to try to assert its authority — to suggest to individuals that it knows better what sources they should trust (ie, Google and Facebook) — many, if not most individuals would see this as an act of aggression. Even governmental institutions may take a sudden interested in how these private companies discharge such self-appointed authority. And we could end up with an information regime that serves no one’s interest.
Yet these companies already make a number of important decisions about what information we see. Thus far, with rare exceptions, they have restricted themselves to features and methods that stand clear of the taint of ideology. And I think that is the trick right now: what changes could be made to the Google and Facebook products to contribute to the health of civic discourse without the appearance or reality of partisanship.
Let’s look at the specific problems:
- Inaccuracy & propagation of error
- Outright fabrication/fraud
- The filter bubble
As noted in yesterday’s piece, I think #1 is the most difficult and perhaps least fruitful piece to bite off. There are some things that could be done. One could facilitate (as Google has already done) a specific tagging for fact checking articles and one could surface these items (along with corrections) to users who had previously engaged with the offending item. Surely, some people will turn such a feature off and be offended that you’re harassing them with your goddamned lamestream propaganda/corporatist news media bullshit. But some will find it useful and helpfully corrective.
The problem is the deeper issue of the fact policing itself. Most important fact checks require explanations of such length and complexity that they themselves invite critique and fact policing. It’s hard for me to see how Google or Facebook could stake out such a position without creating worse problems than the one they’re trying to solve.
Alternatively, one could simply restrict oneself to validating strict claims, and let the buyer beware. Yes, it’s true that 3% of services performed at Planned Parenthood are abortions, and here is the definition of services. Yes, it’s true that Assad has fought ISIS. Here is the extent of his battles against various opponent factions. This doesn’t remove misleading political and corporate propaganda, but it could make it easier for motivated individuals to understand the real context. And surely it’s better than either a) nothing, or b) merely becoming another layer in the onion of contention.
Number 2 is perhaps a subclass of #1, but it is distinct in an important respect. Something constructed of whole cloth — the New York Times raised the excellent example of an article circulating on Facebook that claimed Pope Francis had endorsed Donald Trump — is far easier to judge than a piece with some pretext of truthfulness. And from the stand point of public relations, no one should be affronted at the removal of an entirely concocted piece of writing, such as Pope endorses Trump, any more than they would be affronted by the removal of a business page on Facebook that ran a clearly fraudulent operation.
These cases are comparatively rare, but they constitute an important and damaging category, and one that can be easily dealt with.
The final case, the filter bubble, will live on no matter what we do. If nothing else, it is defined by your skull, by your cornea, by the various contours of your isolated physical self. Subjective mortals gonna be like that.
But what we might be able to do is push out the radius a bit. We used to have Walter Cronkite and the New York Times to tell us what important things happened in the world today, what I needed to hear about and care about in order to be an informed American. We don’t have authorities like that anymore, with that kind of reach, so we don’t have a polity unified by that kind of narrative coherence. This one’s on Salon.com, this one on John Oliver, this one on InfoWars, this one on Reddit, this one on The Jacobin, this one on Chapo.
And yet we still do have common topics that are critically important to every citizen of the country. We could expand these by exposing a broader swath of Americans to a common set of stories. And when I say “story,” I don’t mean a specific article. We could identify an important event and present coverage of that event from a very wide array of sources. This won’t stop me from hunkering down in InfoWars if I want, but showing me the Fox or National Review or local newspaper coverage of the event may help me to at least engage in the topic with someone else, and over time, it may keep my bubble from collapsing to quite so tight a space, or to one filled only with Bilderberg phantasmagoria and the crimes of Hillary Clinton.
So well and good, these things can be done and should be done: eliminate utter fabrications; add a resource to see strict fact violation/validations (with some explication); and positively present a common set of stories (from diverse sources).
Now, how do you do that stuff?
