The targeted precision really is what's different. Your candidate can be N different single-issue candidates in a way that's just not possible without precise targeting.
The "echo chambers" explanation, in particular, seems wrong. Echo chambers always existed.
> Your candidate can be N different single-issue candidates in a way that's just not possible without precise targeting.
Before near real-time national news media coverage and the ubiquity of portable recording devices and online media to distribute the recordings, that was normally the case, but the segmentation was geographic and by addressing selected audiences in closed events. Many political gaffes and scandals of the last couple of decades have been candidates getting caught with unexpected exposure of addresses to either specific geographical markets or closed-group events that were less palatable to general audiences. It won't be too long before political groups maintain social media personas with a variety of constructed backgrounds to capture and expose to wider audiences to which the message would be repugnant those ads that the other side targets narrowly.
Echo chambers always did, but I think the bubble-forming dynamic we have now is a bit different. It's not that I have my own views echoed back to me by the communities I'm in all the time, it's that so much of my exposure to opposing views is often coming from some of the dumbest and most unhinged people on Earth. It's the filtering for nutcases more than echoing back confirming views that seems to be the issue.
It's really easy to believe that certain perspectives are only held by insane people if the only people you see speaking for them happen to be insane. And even if they're not crazy, normal people aren't really going to be interested in communicating nuance or understanding if they're hastily firing off a missive during a toilet break.
There's a lot of reasons for this, but I think one of the big ones is that our intellectual institutions have given themselves over to internet troll logic, including many prestigious Op-Ed pages. In the past you might have encountered plenty of nuts, but you could also see similar enough versions of their ideas being advanced by people who could construct an argument that didn't rest entirely on motivated reasoning, and include enough nuance so as to not be monstrous. But at this point, many professional opinion havers aren't any better or more cogent than a regular Twitter troll. It's mostly just Frankfurtian bullshit all the way down.
>The rich have, Drezner writes, empowered a new kind of thinker—the “thought leader”—at the expense of the much-fretted-over “public intellectual.” Whereas public intellectuals like Noam Chomsky or Martha Nussbaum are skeptical and analytical, thought leaders like Thomas Friedman and Sheryl Sandberg “develop their own singular lens to explain the world, and then proselytize that worldview to anyone within earshot.” While public intellectuals traffic in complexity and criticism, thought leaders burst with the evangelist’s desire to “change the world.” Many readers, Drezner observes, prefer the “big ideas” of the latter to the complexity of the former. In a marketplace of ideas awash in plutocrat cash, it has become “increasingly profitable for thought leaders to hawk their wares to both billionaires and a broader public,” to become “superstars with their own brands, sharing a space previously reserved for moguls, celebrities, and athletes.”
> I think one of the big ones is that our intellectual institutions have given themselves over to internet troll logic, including many prestigious Op-Ed pages
Yes, absolutely. The prevalence of stupid "both sides" reasoning calls into being the contrarians and gives them a platform - in order to have someone say the world is round, you have to find a flat earth pundit as well and put them opposite each other so the public can enjoy the fight.
The kind of people who find themselves cited in mass shooter manifestoes. At least the Unabomber wrote his own deranged manifesto; these days people assemble them piecemeal from everyday racism and conspiracism.
Frequently cited by mass shooters: Melanie Phillips
Climate change denial correspondent for Times, Telegraph, Spectator: James Delingpole
Wrong about everything on purpose: Brendan O'Neill (and the rest of the ex-Living Marxism gang who pivoted from Marxism to Libertarianism without ever passing through sanity)
All over television until he had served his purpose and then invisible: Nigel Farage
And of course, reprimanded as a journo for lying about Europe too often; paid more for a weekly column than as a Minister: Boris Johnson
Echo chambers always existed, but now the gain is higher. Your comment reaches more people faster on Twitter, or Reddit, or Facebook than it would face-to-face.
> The "echo chambers" explanation, in particular, seems wrong. Echo chambers always existed.
The echo chambers haven't gotten worse at all. They've become easier to observe in action and quantify, because social media is leaving a giant digital trail that everyone can observe (whereas previously the average person didn't leave much of a partisan trail/record for observation or study or reporting). The perception that the echo chambers have gotten worse is merely the horror of seeing the already existing echo chambers in action so vividly now that everything is recorded and everyone has a bullhorn.
I think there is a case to be made that it has gotten worse. For example, before social media a lot of social interaction happened in public areas with a more diverse group of people.
With social media you can pick and choose who you interact with, and what content you consume. To make things worse, companies use algorithms that are more likely to show you things you like and already agree with to increase engagement in their platform.
Additionally from my perspective, it seems society is reverting back to a sort of tribalism, where people identify with a specific group and are more and more unwilling to find common ground with one another or to even have respect for differing viewpoints. Things are only seen as black and white, and anyone that doesn't agree is wrong, and must be silenced.
I've had older coworkers say the same thing. They said that when your social circle was usually your coworkers and neighbors, you had to make an effort to get along, and if you had any sort of extreme viewpoints, you had to remember that there were real life consequences to everything you said. Now you can hide your extreme viewpoints from people you interract with, and probably find groups online who share the same viewpoints and only interact with them, getting the false opinion that an extreme viewpoint is normal and socially acceptable.
> They said that when your social circle was usually your coworkers and neighbors, you had to make an effort to get along
People still have neighbors and coworkers.
> Now you can hide your extreme viewpoints from people you interract with, and probably find groups online who share the same viewpoints and only interact with them, getting the false opinion that an extreme viewpoint is normal and socially acceptable.
The choice to isolate yourself in a like-minded bubble was always available.
On the right: most church communities in small midwestern/southern towns will make social media bubbles look like veritable cornucopias of diversity. Or if you have to live in a larger metro, you can very easily find pockets of people who all attend the same church, work for the same few employers, live in the same zip-code, etc. At my first employer (small finance company in the rural midwest) I joined a church because it was the only way to fit in. I think it's fair to say that the majority of the private K12 schools in the USA and the majority of the home-schooling community are explicitly about isolating your family from the out-group.
On the left: same thing. Live in the city, in particular neighborhoods within the city, send your kids to the right montessori, attend a liberal mainline church (or no church), etc.
I don't think there are more people isolating themselves in bubbles. It's just way easier and far less painful to moan about social media than to point out that a huge fraction of our built world and social infrastructure has the effect of forming various types of bubbles.
Social bubbles make it hard to maintain a huge coalition, because disagreeing with any one part of the hive mind can make life in the bubble unbearable even if you are happy with 90% of the other stuff (e.g., if you're socially and fiscally conservative except that you're openly gay, then the rural midwest church is probably a bubble you'll leave). Precise targeted and personalized advertising doesn't have that attribute.
> it seems society is reverting back to a sort of tribalism
"Reverting" implies it ever left. Is it not possible that our media system, in the past, just enforced uniformity to effectively create one tribe of people who consumed it and excluded everyone else? As the environment gets more diverse, then the existing tribalism, and the conflicts it engenders, just become more evident.
One way it may have gotten worse is the perception that it’s not there. If I’m hanging out with friends and we all agree, it’s obvious we’re just a small group. When all the top votes posts on reddit or twitter agree with my leanings, it’s much easier to think that random tens of millions of internet users must be more representative. With some of the upvote dynamics, it probably won’t even be representative of their own whole userbase. It’s so easy to forget that if like 51% of users have a view which they comment and upvote, downvoting the rest, then all the top comments might end up agreeing and you’ll never see the views of the 49% unless you scroll a mile down. (Then you get another headline that causes maybe 2% change their minds or not click through and the 51:49 flips to 49:51 and looks like the whole community flipped around and is totally hypocritical.)
Crackpot movements used to require you have personal contact with the core members or have access to a newsletter. Now, impressionable people can just find them on FB, Reddit, or *chan.
In my opinion, surgically precise targeting was not the problem with Cambridge Analytica. The problems were:
1. Disregard for privacy, by scraping data from people who had not agreed to their terms. And, honestly, from most of the people who "agreed" to their terms.
2. Doubling down on the idea "a candidate should get votes", instead of "a voter should choose the best candidate." It's the classic problem of a measure (vote count) becoming less useful. This has always been the case in politics, but we should discourage it. Not sell it as a service.
I also remember hearing one of their goals was suppressing turnout among voters who likely supported the opponent, but I'm having trouble finding a good source. Even if Cambridge Analytica didn't focus on this, it's another case of the measure ruining its own intent. Voter suppression by official acts is often illegal, but it's not much better when done through legal messaging.
As powerful as Facebook is in the world of American communication, it can't solve those problems. Reducing political ads can help, but we're far from being able to wipe our hands and think we've accomplished much.
Surgical ads allow you to greatly reduce ad spend. If you know people who like A, B, C and are easily convinced of your worldview via data scrapping, and X, Y, Z are not, you can rile up a relatively small group of voters to shape the election outcome in battleground states.
I would say that social media generally is the biggest threat.
However, I find it hard to believe that ads are even remotely the real problem (though they're part of it). They're not nearly as effective at influencing people as friends/journalists/etc and the methods social media uses to reward and elevate angry, tribal content. An endless stream of confirmation bias and strawmen, mostly from other users. Ads are a small part of that river, more likely to be ignored/blocked and more likely to be understood as fiction vs other content. They do have an effect of course, but it seems to be greatly overstated.
Also remember the unprecedented $500M ad experiment Bloomberg ran to little effect. Would $50B even have done the trick? I would bet not.
I think you might want to brush up on more recent news, because by now Cambridge Analytica’s data has been shown to have been mostly useless at meaningful targeting. Even the New York Times reported that Cambridge Analytica’s impact on the election of Donald Trump as president was overrated.
“But a dozen Republican consultants and former Trump campaign aides, along with current and former Cambridge employees, say the company’s ability to exploit personality profiles — “our secret sauce,” Mr. Nix once called it — is exaggerated. Cambridge executives now concede that the company never used psychographics in the Trump campaign. The technology — prominently featured in the firm’s sales materials and in media reports that cast Cambridge as a master of the dark campaign arts — remains unproved, according to former employees and Republicans familiar with the firm’s work.”
But traditional media is incentivized to be biased against social media. So "even the NY Times" isn't particularly compelling. (Not to mention NYT is squarely center right, politically. "Left" of the GOP but not left in any meaningful sense.)
> NYT sits pretty much center left, but feel free to point towards a more comprehensive study.
It's sits center-left in that study because the study centers around the center of US media coverage, not any principled definition of political center; the NY Times viewpoint is, and has long been, neoliberal corporate capitalist, which is a center-right position (the same center-right position which was held by the dominant factions of both US political parties in the neoliberal consensus of the early 1990s, though the Republican Party has since shifted further Right, and the center-left faction has been gaining ground in the Democratic Party in the last several years.)
>It's sits center-left in that study because the study centers around the center of US media coverage, not any principled definition of political center; the NY Times viewpoint is, and has long been, neoliberal corporate capitalist
Your contention is that the center is really defined by the progressive/Marxist/"democratic socialist" minority? In that case, I guess anything to the right of that minority (i.e. everything) would be see as right-wing. Unfortunately, that's not the actual center. A better descriptor for those groups is 'far left'.
1. That the traditional media is incentivized to be biased against social media, and "even the NYTimes" came to a conclusion that advances a social media friendly narrative is exactly the point. If we are talking about the historical biases of the NYTimes, their reporting has consistently had a bent critical of the Citizens United decision (pertaining to paid political speech) and in this particular case, overstating the impact of Cambridge Analytica. That the NYTimes would then go on to conclude that CA's impact on the 2016 election was overrated is absolutely worthy of special remark.
2. Calling NYT "squarely center right and not left in any meaningful sense" is pretty odd. The left-right spectrum is created within the polity in question. It's of no use to an American to classify the American political spectrum from the locus of European politics, because Europeans have no say in American politics. It literally doesn't matter that Joe Biden is "to the right of" Corbyn, because the two will never run against each other in a political election, and British voters have no say in American elections. So while you're technically correct that, on a global scale, the NYT is "center right", that fact is not meaningful at all to their readers, the vast majority of whom are American voters participating in the American political spectrum.
3. Even if one were to concede your premise that it's somehow valuable to classify American politics through another political entity's spectrum of discourse, the left-right spectrum is an extremely lossy way of encoding both social and economic views. On social issues, I think you'd have a pretty difficult time arguing that the NYT is "squarely center right". The issues of election integrity, paid speech, and campaign finance are decidedly social in nature.
> Even if one were to concede your premise that it's somehow valuable to classify American politics through another political entity's spectrum of discourse, the left-right spectrum is an extremely lossy way of encoding both social and economic views.
Even breaking views down as "social" and "economic" encodes a lot of biases. In reality political views seem to align more based on a matrix of dimensions like "openness to novel experiences" or "preference for structure/stability." The system we use often surprises people with how often the Individualist Lefties (of the hippie variety) and Far Right align on both social and economic policy. It turns out their main differences are just about things like sexual mores and recreational drug use, but the general orientation on most other dynamics is the same. They just end up on different ends of the spectrum because of how factional alliances manifest in our political system.
> the left-right spectrum is an extremely lossy way of encoding both social and economic views.
It's not really in the US, where it's been studied and the main dimensions of political variability were a strong economic left-right axis and a weaker racial policy axis (Arend Lijphart, Patterns of Democracy), and since that study it's pretty clear that divisions on the race axis have aligned more tightly with those on the economic axis, making the US political spectrum even closer to a unidimensional one.
Just because you are far-left, doesn't mean institutions to the right of you are actually 'right-wing' or 'center-right'. NYT is in no way a 'center-right' publication.
The other thing that is notable about that kind of precise targetting is that nobody other than the recipients and the advertiser know these ads exist and what their contents were.
Do you inherently disagree that targeted information tuned for uptake via automated, ML driven feedback, is incapable of affecting a persons behavior? If so, do you disagree that information in general is capable of influencing people?
A lot of money is spent based on the assumption that they can—it's not a great situation even if the assumption were wrong. The worst form of government, except for the others.
Because voters are too stupid to tell fact from fiction, and need their betters to protect them from misinformation and make sure they are only exposed to approved viewpoints?
> Because voters are too stupid to tell fact from fiction
Education on critical thinking skills is a problem in the US, in my opinion. I don't think that needs to lead to 1984-esque "approved viewpoints", though.
Every single person is susceptible to manipulation. If you have enough data points you can manipulate someone. This has absolutely nothing to do with intelligence or “smarts.”
I would argue that there is actually an inverse relationship between how susceptible you are to manipulation and how susceptible you think you are.
The impact of Cambridge Analytica is so overstated that it crosses into hyperbole. They were a bunch of clowns with minimal impact that got magnified to a status of super-villain because you could use them to club Trump over the head with. And let's not lose sight of the big picture, namely before Trump accidentally won (yes, accidentally and barely), the Trump campaign was seen (correctly) as a total disorganized disaster and one of the biggest clown-shows in modern political campaigns ... and yet they were at the same time these nefarious puppet masters using modern high-tech technology to manipulate voters and ultimately skew the election.
They aren't the only ones doing it. I get what you are saying, but is your (more general) argument that CA was bad, or that precise, hidden information targeting based on a persons characteristics cannot be used in nefarious ways? Because if you aren't arguing the latter, I don't think its worth calling OP out on CA specifically.
> but is your (more general) argument that CA was bad, or that precise, hidden information targeting based on a persons characteristics cannot be used in nefarious ways?
The only argument I made was that the impact of CA was overblown for political reasons. I don't have a well-formed opinion on targetted political messaging, except maybe that it is a reality and we should just get used to it - perhaps having some regulations may make sense (the way lobbyists are regulated) so that the entire industry doesn't go underground.
>Because if you aren't arguing the latter, I don't think its worth calling OP out on CA specifically.
OP was making a political point - that CA was uniquely bad or an example of the problems of the industry because they were tied to Trump's joke of a campaign. If you actually read what they did, they were clearly clowns and shysters - as in, they siphoned consulting money from the campaign by promising the moon, and couldn't actually deliver. Ironically, their promises were taken at face value after Trump's election.
Exactly. Not just that, when the Obama used social media targeted advertising to its advantage it was fawned over in the media.
Cambridge Analytica was targeted purely because they were on the wrong political side, it's as simple as that. Does anyone really think they would have gotten any attention if they were working for Hillary?
Yes, I think they would have been screamed about from every Republican supporting media outlet there is. And I think they were targeted because of the underhanded way they went about collecting data, and for violating FB ToS.
But the real reason that I think they were targeted? I think they leaned into it as a submarine PR coverage tactic. "We're so effective that it's a national scandal" was a great marketing tactic in the (hopefully bygone) era when amorality had no consequences.
You cannot target people with such precision in traditional advertising.
I don’t think it’s unfair to say social media advertising and data collection is the biggest risk to democracy that we have ever faced.