Categories
podcast

“Opinion cascades” show some political party stances may be due to chance and initial conditions, with Dr. Michael Macy

In this episode of the podcast, I interview Dr. Michael Macy of Cornell University, whose research on “opinion cascades” show how some political group stances on issues can be rather arbitrary and due to initial conditions (a good summary of the study). Similar to how in many complex systems, slightly different initial conditions can lead to vastly different results later on, the early conditions in a country, including early opinion-holders and influencers, can influence a political party to be aligned with one or another stance on an issue. These early choices have a cascading effect, meaning that, for some issues, the political parties could hold reversed positions if things had gone a bit differently.

Transcript is below.

Links to this episode:

Topics discussed:

  • How was Michael Macy’s opinion cascades study set up? What kinds of issues did they ask participants about?
  • What political party stances might be due to fundamental ideological differences versus which ones may be more arbitrary and due to chance?
  • Could Democrat and Republican party stances on abortion, immigration, and other issues actually be reversed in a slightly different world?
  • How does this work relate to problematic political polarization?
  • Can political stances be influenced by simply wanting to be aligned against an opposing group’s stance?
  • Is there something inherent in humans that lead them to form contentious us-versus-them groups?

TRANSCRIPT

Welcome to the People Who Read People podcast. In this episode I interview Dr. Michael Macy about his work showing how some political party’s stances on issues can be rather arbitrary, and can be dependent on chance and initial conditions. 

You might remember a scene in the movie Jurassic Park where Jeff Goldbum’s character places drops of water on the hand of Laura Dern’s character. He does this to demonstrate how the unknown and unmeasurable state of things, like imperfections in Laura Dern’s hand or variations in the air, might lead a drop of water to travel one way or another. 

In a similar way, perhaps some stances on issues that are associated with political parties may also be arbitrary; perhaps they could easily be different, even entirely switched, if initial conditions had been different. For example, if political influencers in the early days of the United States had staked out their positions just a bit differently, others would have piled on and aligned with those positions, leading to very different political party stances. Researchers like Dr. Macy refer to such effects as “opinion cascades”; Similar to the water droplet in Jurassic Park taking different paths, our political opinions can take different paths dependent on early influencers.  

A little bit about Dr. Michael Macy: He is Goldwin Smith Professor of Arts and Sciences at Cornell and Director of the Social Dynamics Lab. His research has been published in leading journals, including Science, PNAS, Science Advances, Nature Human Behaviour, American Journal of Sociology, American Sociological Review, and Annual Review of Sociology.

To summarize the gist of Dr. Macy’s opinion cascades research, I’ll read from a 2019 article in the Cornell Chronicle: 

Social scientists have long wondered how political partisanship develops. In his 2002 book, “The Blank Slate,” cognitive psychologist Steven Pinker asked: “Why on Earth should people’s beliefs about sex predict their beliefs about the size of the military? What does religion have to do with taxes?”

Even more confusing is the tendency of U.S. political parties to radically shift platforms. Macy asks: “Why have the major political parties shifted positions on issues like free trade, balanced budgets, legalization of marijuana, same-sex marriage and trust in science? And how is it that voters on both sides often have contradictory positions on abortion rights and capital punishment?”

Macy’s team looked for answers by conducting an experiment in which they re-created the early days of opinion formation, to see how the cards might have fallen differently had early movers held different arbitrary opinions.

The researchers split more than 2,000 Democratic and Republican volunteers into 10 “parallel worlds,” each isolated from the others. Within each world, participants took turns filling out an online survey to indicate whether they agreed or disagreed with a series of unfamiliar political and cultural issues. In two of the 10 “worlds,” the survey was private, but in the other eight, whenever a partisan took a position on a given issue, all other participants in their “world” saw a real-time update of how each party was leaning.

The results showed how a handful of “early movers” can trigger a cascade in which later partisans pile on to their party’s newly emerging position, leading eventually to large political differences. The big surprise was that the party that supported the issue in one world was just as likely to oppose the issue in another world.

“Sometimes the same party’s early movers would go one way, and sometimes the other,” Macy said.

And in each world, participants followed these early movers – often in opposite directions.

“In one world, it was Democrats who favored using AI to spot online criminals, and in another world it was Republicans,” he said. “In one world, Democrats favored classic books, and in another world, Republicans favored the classics. In one world, Democrats were more optimistic about the future and in another world, it was Republicans.”

End quote 

Okay, here’s the interview with Dr. Michael Macy:

Zach Elwood: Okay, here’s the interview with Dr. Michael Macy of Cornell. Okay, thanks for coming on the show, Dr. Macy.

Dr. Michael Macy: Well, thanks for having me, Zach.

Zach: So your work points at a chaos theory view of political polarization situations that there’s a kind of arbitrary unpredictable aspect behind a lot of the views and stances that we tend to think are due to logical or coherent principles. Do you see your work as tying in to chaos theory initial condition types of views about the state of the world?

Dr. Michael: Yes. Cascades in fact illustrate the famous butterfly effect, which actually is more about complexity theory. Chaos theory is about sudden and unexpected system-wide transformations. Complexity theory is about the dynamics of interconnectivity of autonomous systems. Random perturbations can trigger a cascade that generates very highly non-random patterns, even though the origins are random. And these non-random patterns can in turn trigger the quintessentially human proclivity to invent post hoc explanations. Sometimes the faces in the clouds are just that, they are clouds. But one difference is that clouds quickly disappear, whereas cascades produced not only very strong non-random patterns, but they’re also very stable patterns. So it’s really no surprise that people assume that these patterns must reflect some non-random causal process and then we try to imagine what that might be. And what our research shows– and this is really the bottom line, this is really the most important conclusion from our research– it’s that we first have to rule out the possibility that the patterns might have happened purely by chance as the result of a cascade.

Zach: Your work seems to tie into a lot of kind of like Jon Haidt talks about motivated reasoning like we have these feelings or things that we’ve grown to think. And they’re not logically informed, they’re just things that have become emotional triggers for us or things that have, you know, we’ve become set in these ruts of ways of thinking.

Dr. Michael: Yeah, I think it seems to be part of our nature that we like to explain things. We don’t like to think that something is not explainable, and we sometimes invent deities to explain things when we can’t come up with anything better. And so I think it is somewhat dissatisfying or unsatisfying to learn that you can get really striking patterns that are very far away from random out of processes that are, in fact, random. And yeah, we don’t like to think that that’s the case. That’s even true in the scientific community. And so our paper in some ways is, as you say, it’s very much grounded in this idea and complexity theory that… It’s kind of a butterfly effect story.

Zach: Do you see this tying into sort of an existential need that humans have? We need certainty, we desire certainty, we want to feel certain about our self-concepts and about our ideas of the world and not want to believe that they’re just kind of arbitrary.

Dr. Michael: If I can just actually point out that what you just did is asked me to explain the need for explanation. [Zach laughs] I think that illustrates the point, we really do want to have explanations. And I have no business trying to explain that because this is not the area in which I do research, I’m sure there are people in cognitive psychology who have done research on that and can tell you a lot about it. Nevertheless, I will say two things. One is that you would like to hear an explanation and I would love to give you one. [Zach laughs] We just really do like to explain things. That seems to be part of our genetic wiring. But I’m going to resist the temptation.

Zach: Right. Well, that’s very respectable because I know that we all have the desire to weigh in on things, even when we don’t have a firm opinion on them, you know? And that’s a great point-

Dr. Michael: It could be a need for control. Right? If we think we explain things, then we think we can control them. So it might be a desire for control, it might be… You know, there are some theories in evolutionary theory that at some point, our genetic ancestors and our evolutionary ancestors went through a phase where the males who were more likely to attract a mate and survive and reproduce switched from being the alpha males, you know, the ones that were big, muscular and powerful people, to the ones who could spin a good story. And so maybe it’s just wired into our genes that there was some evolutionary advantage to being able to explain things. And I will certainly say that as a not-very-athletic academic, that’s an evolutionary story that I love to think.

Zach: [laughs] Right. Right.

Dr. Michael: It certainly was not true when I was in high school, I have to admit.

Zach: Right, it goes in phases. Yeah. Let’s see. Has there been similar research on this topic before you?

(We’ll now be hearing an ad. I don’t endorse these ads and I encourage you to remain skeptical of all ads.)

[Ad plays]

Dr. Michael: Yes. In fact, back in 2006 a former student, Matt Salganik along with Duncan Watts and Peter Dodds showed how cultural unpredictability could be explained as the consequence of cascades in which initial randomness is amplified into runaway cultural winners. You know, why does some music… Why does some song become such a runaway hit? We get these lopsided outcomes in which a particular song becomes a runaway hit. And when that happens, then yes, you start getting people racking their brains trying to come up with an intuitive explanation for why did this song become a hit. But in fact, what their research shows is that actually it could just be an accident amplified by a cascade. Some song has to win, and it could just be luck of the draw and then it runs away with it. It dominates the market, it gets to the Billboard and stays, and then we think it must be something about the song, when in fact it could just be chance.

Zach: Yeah. Your work makes me think of all the elements of chance in life, like business success. You know, there’s just so many situations where people create a business or create something and it succeeds and we tend to think, “Oh, that’s because we did something right.” Whereas it could just be completely arbitrary. Like I was talking to someone who almost had their company featured on Oprah, and it would have obviously been a life-changing thing and their company ended up failing. There’s all these elements of chance and all these things we do, but we tend to assign, “Oh, this is because we had a great idea or this is because we did something really right here.” Yeah, so your work just makes me think of all those.

Dr. Michael: Yeah, and our key contribution and theirs as well– Salganik’s group as well– is to show that it’s not just chance, it’s chance plus cascades. That cascades can take that butterfly flapping its wings, that chance-initial event, and cause it to run away to become a very far-from-random pattern and very far-from-random outcome. And that’s really the focus of our research, it’s that cascades can greatly amplify chance occurrences into highly non-random patterns that invite explanations, which could just be spurious.

Zach: I wanted to read a few of the issues that you surveyed people about in your research, and I think it’s interesting to hear a few of those. To reiterate how this worked for the audience, you would either ask people to give their opinions on which political party might have this stance at some point in the future, or in another version of this, they had a chance to win money by guessing which stance aligned with which political party? So before I read a few of those, is that a pretty good summary of the two things that you asked people?

Dr. Michael: Well, the key thing– and this is what we borrowed. In fact, we pretty much copied the research design used by the Salganik group who were studying music, but what we did is we changed music to politics. So we were looking at runaway political polarization where the two parties take entrenched positions on the opposite sides of an issue. And for sure, political scientists are going to look at that and political psychologists are going to look at that and come up with some explanation for why the Democrats are on this side and the Republicans are on that. But here’s the thing, like the Salganik study, we created these 10 parallel universes. If you’re in one universe, you do not know that there even are the nine other universes. So this is the only universe that you know. We’ve recruited the participants for the experiment and we’ve assigned them to these 10 parallel universes. And in eight of those 10 universes, you get to see what Democrats and Republicans who’ve gone before you have said on that issue. And then in two of them, you have to give your opinion on the issue without knowing what other Democrats or Republicans have said. That’s sort of our baseline or control condition.

So then what we found is that it was a coin toss. You take a particular issue, let’s say it’s Bitcoin regulation or it’s something about artificial intelligence or self-driving cars or school curriculum, those are some of the issues that we included in it. In one world or one of these parallel universes, the Democrats are on the pro side and the Republicans are on the con by a big margin. But then in the next world over, its flipped over and it’s the Republicans that support and the Democrats who oppose. But of course, we just live in our one universe and so we don’t realize that it could just as easily have been that the parties would have been on the opposite sides. Whether it’s tariffs and free trade issues, or internationalism, or even climate change, it could just as easily have been that the sides could have been switched.

Zach: Let me read a few of those issues. I’ll just read four out of the 20 you had, I think it’s interesting. One is, “The exchange of cryptocurrencies such as Bitcoin, Aetherium, and Litecoin should be banned in the United States.” Another one was, “Self-driving cars should be programmed to make life or death decisions that minimize total harm or death, even if that means sacrificing the driver’s own safety.” Another one was, “Artificial intelligence programs should be developed to serve as public defenders for those who cannot afford a human attorney.” And another one was, “Artificial intelligence software should be used to detect online blackmailing on email platforms.” That was just a few, and I’m actually going to read a few more of those at the very end, as I think it can be interesting for people to think about those and how they would answer those.

So some of the political division we see between US Democrats and Republicans makes ideological sense. And I’ve been reading Jon Haidt’s book, The Righteous Mind, and getting a better sense of how some of these stances on issues are due to underlying ways of seeing the world. For example, conservatism is usually defined by respect for tradition and authority, which would seem to help explain some of their stances on major issues, while liberals are more concerned with issues of compassion and have less respect for tradition and that can help explain a lot of their stances. So some ideological stances seem to make sense in this regard and seem kind of predetermined but then other issues might seem a bit arbitrary like they could go either way. Can you talk a little bit about how you see some political party stances maybe being a bit more predetermined in the way they align with fundamental worldviews while some might be on the more arbitrary and less predetermined end?

Dr. Michael: Well, that’s exactly why we pre-tested to make sure that the issues we were using were not already aligned. But again, in terms of conservatives being more concerned with preserving traditions. Here again, we have to be really careful not to see faces in the clouds. Conservatives overwhelmingly supported Trump on November 3rd. I mean, overwhelmingly! So let me ask you, is Trump preserving the traditional norms of presidential behavior? Is he conceding gracefully? Is he supporting the democratic process as a reasonable and fair way of selecting leaders? Those are the traditional norms of presidential behavior. Even norms of respectfulness and courtesy and deference; is he respecting those traditions, and do conservatives care?

Zach: Right, we’ve seen him shift parties a lot. You’ve got the civil rights issues, you’ve got the conservative economic stances which have obviously shifted under Trump to more protectionism. Like, you can have major shifts as you’re saying.

Dr. Michael: And in some ways, Biden is much more of a traditionalist than Trump. But you won’t find conservatives supporting Biden. And yes, your examples are great. Republicans used to be the party of Lincoln, and Democrats were the Dixiecrats. That’s very different today. Republicans used to be the ones that championed internationalism and opposition to Russia as a foreign adversary, and the Democrats were the isolationists. Democrats supported terrorists, while Republicans backed free trade. Oh, here’s another one. When I was in college, it was the Left who distrusted science, and now it’s the Right that distrust science. Liberals tend to be anti-vaxxers, but that could just as easily have emerged as a conservative position. Environmentalism was historically associated with the far Right –and still is, actually, in some places in Europe– but now it’s clearly seen as a liberal position. The same people who advocate for the sanctity of human life on abortion also often support capital punishment. People whose support law and order also oppose enforcement of government regulations to protect the environment against pollution that by the way kills more people than guns do. So good luck explaining issue positions based on a coherent worldview.

Zach: If I was gonna play devil’s advocate for some of these things, it would be conservatives who support Trump would say they’re offended by offenses to their traditions so much that they would support someone like Trump. I think if I was playing devil’s advocate, that is how they view things. They see so much assault in their mind on what their traditions are and how they define their traditions, that they’re willing to support someone like Trump. Now I’m just playing devil’s advocate but I’m trying to put it in that framework.

Dr. Michael: Well, that’s actually extremely useful. Your response is very useful because it shows that post hoc after the fact, we can reconcile what we observe with our favorite explanation. You can find a way to spin a pattern and to spin a worldview in order to reconcile the pattern with the worldview, no matter which side it’s on. So if it had been the other way around and Trump was the one supporting presidential norms, then we would focus on that. If he’s challenging the presidential norms, then we say, “Well, the conservatives supported him in challenging the presidential norms because they know he will support other traditions that we care about more. I think that’s exactly the tendency that we’re fascinated by; that we love to look at outcomes and then come up with the explanation after we’ve seen the outcome. Duncan Watts who worked with Salganik on that music study has an entire book on this. The title of the book is Everything Is Obvious: *Once You Know the Answer. That in some ways is our Bible for this study. That’s the thing that we’re taking on, is the obviousness of post hoc explanation. And what we’re doing is we’re showing that before you come up with the explanation, you first have to rule out the possibility that it’s random, that it could just as easily have been the other way.

Zach: Right, you make great points. You can easily imagine Trump… I mean, Trump obviously has been leading the way on many things and people just follow in his footsteps. And obviously, other political leaders you can point to have done similar things. But if we’re talking about Trump, you can imagine a world in which Trump was very pro-mask and then through some other means, things can easily get twisted and things would follow a completely different path just depending on some initial choice of Trump or whoever.

Dr. Michael: That’s a great example. Masks are really a good example for the point that we’re trying to make. That one really could have been the other way around and Republicans were the ones arguing masks. And when you think about it, there’s just really no reason ideologically why you should be for or against masks. Let’s face it, public health doesn’t have much to do with ideology and yet it has become a kind of bumper sticker for your ideology and your party affiliation. And I don’t know the answer on this, but I strongly suspect that there are very important tribal dynamics in which it’s not our ideas that shape our behavior, but maybe it’s the other way around that our tribal affiliations are the source of our ideas; that we adopt the positions on issues of our tribe, rather than joining the tribe because the tribe agrees with us on those issues.

Zach: I’ve been talking a lot about polarization on the podcast, and your word ‘bumper sticker’ or these things that shouldn’t be political just start to be associated with a certain group through some very chaotic initial condition process. And they shouldn’t really be politically associated but they become emblems or bumper stickers of a certain group. Another one I thought of was I know secular conservatives who have become very pro-life in the last few years and it’s almost like there’s been this increasing drift to the more extreme edges in things that previously people would have more variety on. On the left, too, I can think of examples but people have become more entrenched and more extreme in fitting the stereotypes of their groups, you know, getting all the bumper stickers associated with their group even though there shouldn’t necessarily have to be all that alignment. Yeah, maybe think of that, too.

Dr. Michael: And it’s really interesting the extent to which these bumper stickers are not just expressions of opinion, but they’re also expressions of belief, in fact, so that we actually can have these alternative factual realities in which our perceptions about what exists depend on our tribe and what our tribes tells us exists. Which gets reinforced in the echo chambers on social media and on broadcast media. So I think that in part what our research is showing is the way in which these tribal processes can become highly non-random, and yet the positions that are being taken are remarkably arbitrary. It could just as easily have gone the other way and the sides could have been switched.

Zach: Does your work show a dynamic where just by one side picking a side, that the other side instinctually finds themselves picking the other side in a kind of contrarian us-versus-them way, almost like people can’t seem to just agree; that there must be some sort of disagreement. I’m wondering if your work points at that.

Dr. Michael: Absolutely. That’s a great point, a great question. We have another paper, an earlier paper called Why Do Liberals Drink Lattes? And in that paper, we showed that what we call negative influence, the negative influence from the outgroup, from the other tribe is absolutely essential for polarization. It’s not enough that we agree with our tribe, it’s also important that we disagree with the outgroup. So you know, when we think of influence, we think, “Well, influence is my ability to convince you of something.” But there’s also a negative influence where I cause you to be the opposite of me, to differentiate from me. And we believe that’s extremely important and actually not appreciated sufficiently. So in our experiment, it was not enough just to say most people say this. We had to have, “Here’s what your party says, here’s what the other party says.” And it’s that party differentiation that drives the process. It’s not just agreement, its agreement with my tribe plus disagreement with the outgroup.

Zach: I’ve seen a lot of examples of this one. On the liberal side with Trump being so extreme and disliked by liberals, when he picks a side, it seems to drive everybody to take more extreme positions on the other side; whether that’s immigration, you can see liberal stances on immigration and policy stances change in the last few years very quickly. And it seems like there’s that kind of process playing out, which is kind of this natural process that seems to lead to large groups and nations being prone to increasing polarization.

Dr. Michael: There is actually some evidence that that happened on climate change. Back in the day when global warming was first becoming something that people knew about and talked about– just emerging onto the scene– there was no partisan division. Liberals and Republicans were equally likely to be concerned about it, even though they maybe didn’t know all that much about it at that point. And then it separated. And I think that that process of negative influence was key to that separation, otherwise we would have ended up with the two parties taking the same position. By the way, I should also add the caveat that most Americans regardless of party share a concern about global warming as an existential threat. So this is really much more the party elites being divided on that one more than their base.

Zach: Right. It’s the most extreme people that get the attention too that drive the conversation and drive the political elites too in some way. Can you imagine a world in which liberals are more pro-life than conservatives?

Dr. Michael: Absolutely. In fact, there are important currents within the Democratic Party that are pro-life. For example, traditionally, working-class Catholics were Democrats. That’s changed because now the working class is much more politically divided with a large chunk of non-college-educated White supporting Republicans. But in an earlier era, there were strong class divisions in the party, and working-class voters tended to be Democrats. Many of them also tended to be Catholic, and as Catholics they were pro-life.

Zach: It also made me think, too, liberals are usually the ones who have a lot of compassion for helpless things and so I can imagine a world in which, through some initial conditions, liberals were the ones saying, “Don’t kill these fetuses,” and conservatives were the ones saying, “Hey, I can do what I want.” You know, kind of fuck your feelings in a way. I can imagine these things just based on initial conditions, whether that be religion or even constitutional initial conditions where things went another way. But yeah, interesting.

Dr. Michael: Clearly, there are some issues in which ideology and worldview and moral foundations– to refer back to Jonathan Haidt’s work– that those things really do determine the outcome, and people don’t care what their tribe says. Or they’re in the tribe because of that issue so the tribe came second, it wasn’t the tribe that drove the process. So for sure, I think there are important issues where the ideology and the worldview come first. But what our work shows is that that’s not universal and that there are many issues on which the tribe comes first. And people are not adopting a position because that’s the position that’s consistent with a clearly held worldview where they can see the logical connection between the worldview and their position on the issue. Instead, they adopt a position on the issue because that’s the position that’s taken by their tribe.

Zach: Immigration seems like an issue where it seems pretty ideologically predetermined. That’s one that I can think of but yeah, I’m sure there’s many others that seem more predetermined.

Dr. Michael: I think immigration could have gone either way, although it’s tricky because immigration involves so many things. It involves who is my family, and where did I come from. But it also involves people’s sense of who are true Americans and who are the people who don’t belong here. It also involves racism, where the immigrants are not coming from Northern Europe but they’re coming from areas of the world– from Africa, from Puerto Rico, from Mexico. Race gets involved, so it’s very complicated with immigration. And clearly, we’re not saying that all issues are derived from your tribe, it’s very likely that a majority of the issues even are issues where people adopt a position independently of any social influence from their neighbors. They just figure it out themselves and they take a position based on their core values, their worldview, their ideology. But there are also issues that people are adopting on the basis of social influence from like-minded people and as you rightly pointed out, differentiating from the hated outgroup. And those issues, our point is that cascades can cause those alignments to have actually a random origin which could just as easily have gone the other way.

Zach: For the immigration issue, I was thinking recently imagining a situation in which most immigrants were voted Republican and in that case, I think you would quickly see a shift in stances there. Do you think if more people knew about your work there would be less animosity and more understanding of each other? Do you think understanding these factors helps lower the political tensions?

Dr. Michael: Well, unfortunately, [chuckles] I don’t think too many people are going to know about my work outside the scientific community. We’re mainly writing for the scientific community. Having said that, scientists are just as tribal as anyone else. And I am just as tribal as the next person. Tribalism is is not cognitive, it is emotional. And so people whose work is mostly intellectual like scientists are just as vulnerable to tribalism. So maybe the scientists who read our work, it might have caused them to think twice. I’m concerned about polarization, I’m concerned about tribalism, I would love to see our work contribute to turning down the temperature. But I have to be realistic here, that’s quite unlikely to happen. Our mission here is much more about making the case to the scientific community to be careful about seeing faces in the clouds. And before you invent an explanation, consider the possibility that it was a cascade in which the outcome could just as easily have been the reverse.

Zach: Right, that’s a great point. We all have so many biases and want to explain things, and trying to not do that and trying to withhold judgment is very important but very hard to do.

Dr. Michael: It might be useful, though, if people who do have a much broader audience than we have, such as yourself and pundits and news media, I think it would be useful if they pointed out to people how the sides have switched on issues. Because if people were aware that our party used to have the opposite position and their party used to have our position but it it changed over time, it might cause people to think twice and to be a little bit more careful about adopting a position that maybe they don’t even really understand, just because it symbolizes their group.

Zach: Yeah, it seems valuable. Even just thinking about thought experiments and imagination of these different worlds where things could go different ways kind of opens your mind up to thinking about how arbitrary some of your own beliefs might be. And I think that’s really valuable imagining. So in your paper, you write that “Social influence would likely have been much stronger had we conducted the experiment as a chatroom where participants might experience social pressure for ingroup conformity and outgroup hostility.” End quote. So this seemed to me to imply that you were of the opinion that using social media would enhance the effects you found. Is that accurate to say, and do you think we’re in a situation where social media is amplifying some of these tendencies?

Dr. Michael: Well, I think for sure it does amplify it. But the problem is not social media. Social media is just a platform. And there are other platforms. The social pressure is always there. It’s there, in fact, because we are tribal creatures. We are moral creatures. You know, people like to talk about the difference between morality and ethics. One way to think about it is ethics is about how we think we should behave, whereas morality is about how we think other people should behave. You cited Jonathan Haidt’s book which has the title, The Righteous Mind. Morality is about righteousness, but righteousness is not just about what makes us support our tribe or righteousness is not just about what makes us behave in a way that conforms to our moral principles, it is also what makes us pressure our fellow tribal members and our fellow tribe members to support the positions of our tribe and to oppose any tribe that’s seen as a rival. And so in a way, I think this is the great paradox of the human condition. We cannot live without one another, but we cannot live with one another either. We have a hard time living outside groups but once we’re in groups, those groups tend to engage in conflict with one another.

And one of the mechanisms that enable that to happen is the social pressure by which we impose our righteousness on others. And for sure, if you are a Democrat hanging out with Democrats and you say something that is Republican, you’re going to hear about it. And the same thing on the Republican side. So we tend to think the animosity and the sanctioning is always in the punishments are always directed between the tribes, but actually that is not the whole story. There is a lot of pressuring and punishment and sanctioning that goes on within your group to keep everyone in line. And I think for sure on both sides people feel that. They know that they’re going to hear about it if they violate the consensus in their tribe.

Zach: Yeah. And to your point, these are definitely natural tendencies and I would say I think that social media does amplify these in unique ways, though. We have Jaime Settle who I interviewed on the podcast recently talk about her work of how Facebook makes us realize the political views, or at least the perceived political and cultural views of everyone around us, which is something that’s entirely new. We didn’t use to have that visibility, and that kind of increases the us-versus-them feelings. And then you’ve also got a difference of speed and degree. Because you can see how an initial condition like Trump taking a stance on something shifts the conversation very rapidly throughout the whole nation, really, because it gets spread so quickly through cable TV also, but through all these very fast mechanisms through social media which the interactions help spread those ideas very quickly. And then those very quick reactions can have very quick actions on the other side and reactions on the other side. So even if it is a natural process that we’re dealing with, I kind of see all these modern communication tools as just amplifying and putting all this stuff on speed, you know? But yeah, I just want to mention that.

Dr. Michael: Well, you’re absolutely right. There are a couple of things to point out. One is the thing that you just mentioned that Facebook and social media allow you to see what your neighbors are thinking in ways that would be harder to detect outside your immediate circle. And that’s for sure. Another thing though that social media does is that it does allow for sanctioning. So there’s Twitter, especially, and Reddit are both notorious for the flaming and the attacking that goes on. And so there’s not only a lot of sanctioning and negativity flying around on social media between people who disagree, but also with in-groups so you will hear about it on social media. If you post something that deviates from the position of your ingroup, you will hear it from ingroup members. Social media allows us to, and has evolved norms that encourage us and open us to practices that would be considered discourteous and impolite in face-to-face communications. That sanctioning aspect of social media is also important, as well as just learning about the positions of your network neighbors.

Zach: This has been Dr. Michael Macy. Would you like to talk about anything else that you feel we’ve missed here? Or do you think that’s a good ending point?

Dr. Michael: There was one other study that we did on cascades that your listeners might be interested in hearing about. It was a cascade that we did based on a movie. Did you see the movie Pay It Forward?

Zach: I did.

Dr. Michael: We did a study on that. Yeah, Milena Tsvetkova and I did a study based on this movie, Pay It Forward. We replicated the dynamics in the lab by giving people the opportunity to be helped and to pay it forward. And what we found is that indeed as that movie suggests, that when you help people, it is not just the person whom you helped that benefits but also the people downstream who were helped because the person you helped became more likely to pay it forward. So we got these Pay It Forward cascades going in the lab. And that’s the good news. But now here comes the really bad news. The really bad news is that if you screw somebody, the same thing happens. You know, there’s a Smack It Forward cascade in which the person you harm becomes more likely to harm others. And worse yet, this negative effect is actually stronger than the positive one. So the next time you’re thinking about screwing somebody, remember you’re also hurting all the people downstream as well.

Zach: Right. There’s something about the negativity effect or something in psychology where negative things have more of an impact on you than positive things.

Dr. Michael: And this may have something to do with that or it may just be about norms of obligation that the helping behavior modifies, which is why you’d get the cascade– the Pay It Forward cascade– but I think it’s important for people to be aware that you’re affecting people downstream, whether you’re helping or hurting.

Zach: There’s that expression ‘hurt people hurt people’ and it seems like a lot of the ills in our society are due to people being abused in various ways and them taking it out on the world later.

Dr. Michael: And it may be useful for people, in general, to become more aware of the ways in which they’re living inside cascades, that everything they’re doing is triggering behaviors by others that in turn trigger behaviors by others. And so our behaviors are amplified through our interactions with others.

Zach: Yeah. And to take an extreme step further, I think free will is very unlikely. It’s hard for me to imagine the mechanisms behind having free will, so in some sense you can view the entire universe and your entire existence as one big cascade of things that came before you. And you can see it in that light that we’re all just part of a matter cascade or process cascade. Okay, getting too metaphysical there and maybe at the end. This has been Dr. Michael Macy. What’s the best way if people want to contact you?

Dr. Michael: My email at Cornell. If you Google me, you will get to my Google Scholar page, actually, and you’ll see other papers and contact information.

Zach: Okay, thanks so much for coming on. This has been great.

In the interview, I said I’d read some more of the questions from Dr. Macy’s Opinion Cascade study. But that seems a bit overkill right now so instead, I’ll just tell you how to find them. You can search for “Science mag, Michael Macy Opinion Cascades” and you’ll find his paper there. And the specific questions are in the supplementary materials linked at the bottom.