Part 2 of It’s Social Norms, Stupid
This is the second in a series of three articles. In part one, I made the case that the farmed animal movement’s dominant theory of change is neglecting the role of social norms in shaping people’s attachment to meat, resulting in incomplete strategies. In this section, I introduce the science of social networks, which has cast new light on how changes in norms spread. In part three, I’ll build on this science to discuss how social movements can unleash the power of social networks to change norms at scale, particularly for the animal freedom movement.
- Animal advocates are neglecting a major obstacle to changing the public’s habits of meat consumption. Social reinforcement from relationships with other people eating meat keeps people from changing.
- New research has shown how changes in social norms spread through networks.
- Complex behavior change spreads much differently from information. Behavior change does not spread fastest through virality and influencers. Instead, it spreads by building momentum in isolated social clusters on the edge of the network.
- Animal advocates should adopt new strategies focused on social groups rather than individuals. This approach can lay the foundations for a massive change in social norms.
Amongst animal advocates, there is a lot of excitement about a theory of change we can call the Taste, Price, Convenience (TPC) postulate. This theory holds that the moment animal-free foods taste the same or better, cost the same or less, and are as accessible as animal meat, people will choose to buy them instead. But it is incomplete. In fact, taste, price, and convenience represent just the tip of the iceberg when it comes to people’s food choices.
Consciously, people are motivated by strong cultural attachments to meat from animals, as well as values connected to naturalism. But the (often unconscious) influence of social reinforcement is even more important. Most people can’t imagine being the first one in their social group to change what they eat. As long as everyone else keeps eating animals, they will too. The most convincing arguments won’t matter.
On the surface, this seems like an impossible puzzle. If everyone eats animals, and everyone eats animals because everyone else eats animals, how can we ever get anybody to stop eating animals? Where do we start?
Nodes and Connections
It turns out that a simple shift in perspective holds the key. When we think about influencing the behavior of individuals, we run into a roadblock: their relationships with other people are stopping them from changing. But social reinforcement doesn't have to be our foe. In fact, it can be a powerful tool. The reason lies in the young science of social networks.
When you think of society, do you mostly think of people? For social network researchers, society is made up not of people, but of relationships. (In network parlance, these are nodes and connections). From this perspective, the attributes of individual nodes are much less important than the connections between them, and the way those connections are arranged.
On an individual level, humans are unique and full of quirks. But once we are arranged into a social network and norms take over, our individual traits stop having much effect on the behavior of the network. The larger and more complex the network gets, the more we start to look like legos: interchangeable. The behavior of the system is determined by how the blocks are connected to each other, much more than any traits of the blocks themselves.
Why It Matters
Once again, this may all seem a bit elementary. But the simple fact is that up to this point, the farmed animal movement has not been engaging with society according to this perspective. Efforts at influencing public behavior have focused on individuals, barely sparing a thought for the relationships between them. We have already established that the influence of relationships is one piece missing from the Taste, Price, Convenience postulate. Classic vegan outreach methods like leafleting share this flaw. Targeting individuals with information ignores the social forces that prevent them from changing their diet. This more than suffices to explain high recidivism rates among vegetarians and the inefficacy of leaflets.
Realizing this, many animal advocates have given up on winning over the public. Most of the organizations that used to focus on leafleting have redirected their attention to corporate welfare campaigns 1 To the extent that these campaigns (and protest activity more generally) have a goal of influencing the public, they represent a shotgun approach, reaching individuals at random without any attention to social clusters or a systematic plan for altering the public’s views (that I am aware of). (not to mention their budgets: such campaigns now guzzle 60% of total farmed animal funding from Effective Altruism-aligned grantmakers). While these campaigns may be worthwhile, it’s hard for me not to interpret this as a white flag. We have surrendered the war over public opinion for the more tangible battleground of corporate boardrooms.
The TPC postulate focuses on the single relationship between consumers and their food provider. It imagines that all the problems of animal-free meat adoption can be addressed there. That’s not how people work. We are social, and anything a tenth as fundamental and controversial as food is processed through our social selves. The price and chemistry of alternative meats may have an important role to play. But ultimately, the matter of animal freedom will be settled in the social arena.
I’m here to tell you that we don’t need to give up on social change for animals. In fact, the very social network science that explains why individual diet change advocacy was so ineffective offers a different and far more powerful solution.
The Myth of Virality
Social networks attracted the attention of researchers in the mid-20th century. Since the beginning, the principal metaphor used to describe how change spreads through these networks has been a virus. One person adopts a change, then introduces it to everyone they meet. Each connection presents an opportunity for the virus to spread. As long as each person spreads the change to more people (and sticks with it themselves) it’s only a matter of time before the change has spread throughout society.
For decades, this metaphor of viral spread was the accepted mechanism for social change. A social media post or hashtag, a new slang word or hand gesture for greeting, a diet or health philosophy, or a political idea would all spread like a virus through social connections. This theory led to three familiar ideas about how change happens, which in turn became common sense.
How Viruses Spread: The Old Model
- Target Influencers: If change spreads through connections, then the most connected people will be the most effective at spreading change. This leads to the influencer model, whereby the most effective individuals for spreading change are highly connected influencers who can each inform hundreds, thousands, or even millions of followers.
- Be Viral and Sticky: For an idea to spread fast, it has to be viral and sticky. Virality represents the likelihood that an idea will make the jump across any given social connection. Stickiness ensures that people will hold onto the change once they’ve tried it. Virality and stickiness are qualities of the change itself, not a product of the connections between people. If one idea spreads and another fails, it’s because the winner was catchier and had more staying power.
- Redundancy is Waste: If change spreads like a virus through social connections, then redundancy is waste. That is, each time someone announces a new idea to someone who has already heard about it, that’s wasted energy they could have spent spreading the change to someone new.
That last one is less widely known, but in fact, it describes the thesis of the most-cited paper in the entire field of sociology. That paper is Mark Granovetter’s “The Strength of Weak Ties” published in 1970. Granovetter's study began with the observation that not all connections are equal. Our connection to our friends and families is not the same as our connection to the dentist we see twice a year. Strong ties are the core of our social network, people we see many times a week. Weak ties are our casual acquaintances, our outer circle.
The key difference for Granovetter was that weak ties lack redundancy. Accordingly, information will spread faster through people who don’t know each other very well, because they’re much less likely to know the same people (compared to close friends, who are also close friends with each others’ close friends).
Distinguishing strong ties and weak ties was a brilliant insight. It propelled Granovetter to fame and helped establish the field of network science. Yet as it turns out, that distinction contained a seed that would grow to turn this entire paradigm on its head.
Dentists Don’t Spread Veganism
In the last couple decades, a new wave of research has radically overturned the understanding of social change that reigned when Granovetter published “The Strength of Weak Ties”. Social media and other digital technologies have enabled scientists to look inside these networks for the first time. One such scientist, Damon Centola, surveys this new era of research and its importance for changemakers in his recent book Change. Without hesitation, I would call this one of the two most important books published on social change in the last decade. (The other being Mark and Paul Engler’s This Is an Uprising, which I’ll discuss in part 3 of this series.) The rest of this post represents my best effort to summarize the relevance of this research to animal advocates in a few hundred words. That said, I implore any serious strategist to download and start reading Change immediately.
Like Granovetter, Centola’s work rests on a crucial distinction: simple contagions vs. complex contagions. It turns out that simple changes do indeed spread through social networks like a virus. If, for instance, you want to spread information about a breaking news event, the strategies above are effective. Highly connected influencers and other weak ties help information spread fast across different social groups. The defining quality of a simple contagion is that it doesn’t encounter resistance to its spread through the network. One or two exposures to the contagion is enough not only to change someone, but to make them a spreader. If you have a literal virus and you stay at home, it will just bounce around among your family or housemates until everyone is immune. But if you bring it to your dentist, she might spread it to dozens of other households before she has any idea she’s sick.
But the viral model breaks down with complex contagions. Spreading a behavior change is fundamentally different from spreading information or a virus. Changing one’s behavior comes at a cost: time, attention, and most importantly, social belonging. Most people won’t bear the cost of a behavior change just because their dentist is doing it. (Be honest, how often have you really been flossing lately2I’ve actually been flossing regularly for the first time in my life, but it had nothing to do with my dentist: two of my housemates floss twice a day, often in common areas. This is a perfect demonstration of the power of strong ties!?) The higher the cost, the more resistance the contagion will encounter. And the more resistance a change encounters, the more it must rely on a set of strategies that are almost exactly the opposite of the old paradigm described above.
The Snowball Strategy
Although this new model for changing social norms is a major departure from the influencer model, in its own way, it is also fairly intuitive. The starting principle is that the basis of a social norm is that most people are doing it. The bulk number of people doesn’t matter as much as the percentage of people who have adopted the behavior. And from the perspective of any individual in the network, that means the percentage of people they are connected with who have adopted it.
Let’s consider the network diagram above. Imagine it represents all the people who attend services at a neighborhood church. The core of the community is made up of people who regularly attend service and are involved in other programs the church puts on. These core individuals are highly interconnected and have strong ties with many other people in the community. At the very center of the network would be the pastor, who has strong ties with everyone in the core and many in the periphery.
If you wanted to spread a behavior change through this church community, where would you start? By now, you understand that it depends on what kind of change you want to spread, and how much resistance you expect. If resistance will be low, the logical place to start would be with the pastor.
Imagine the pastor sometimes has a hard time getting everyone to quiet down when she needs to make an announcement. Your idea is that when everyone is talking and one person raises their hand, everyone who sees them raises their hand and goes quiet, until the whole group has gotten quiet enough to hear an announcement. If you get the pastor on board, she can announce this new method at a few Sunday services. Soon, everyone will be notified of the change and it will become the new norm.
Reforming the Church
But what if you’re spreading a complex contagion? What if you want this church community to embrace the farmed animal movement, or even establish a norm of getting arrested protesting violence against animals? Even in a very liberal church, this will be no easy task. You’re expecting a lot of resistance.
Should you start with the pastor? It may be a moot point; she’s very unlikely to start advocating this from the pulpit if it’s significantly outside the social norms of the church. Like any community leader, part of her job is to reflect the values of the community. If she strays too far away from that, people will start to leave, or she’ll get fired. In fact, even if she did start calling for the community to change their values and norms to include civil disobedience for animals, it would likely just demonstrate the limits of her leadership position. People might ignore her.
When it comes to social norms, trusted influencers might have a bit more sway than the average community member. But this is more than counterbalanced by their highly connected position in the network. When it came to simple contagions, this high connectivity was an asset. But for spreading complex contagions, it becomes a weakness. Even if the pastor adopts the change, she’ll be surrounded by people who haven’t, and all those people will be surrounded by people who haven’t. All those other people represent countervailing influences: resistance to the new norm, in the form of reminders that the old familiar norm is still going strong. The pastor only makes up a small percentage of each person’s social network in the church. Despite her respected position, the pastor alone cannot declare a new social norm of such magnitude in this highly interconnected community. If she tries to, she’ll just invite a rebellion.
Power Lies on the Periphery
When it comes to spreading changes that face resistance, the potential for change lies on the network’s periphery. Looking at the diagram above, these are the green nodes that each have three or fewer connections. These people sporadically attend service and only have a few social connections in the church. The magic of the periphery is that change initiatives that begin here have a chance to build momentum before running up against the full force of the entrenched social norm and the countervailing influence of the highly interconnected network core.
Let’s say you have a finite amount of energy you’re able to invest in getting this church on board with animal freedom. You could spend all that energy trying to persuade the pastor. Even if you can convince her to take a big social risk, it may not even matter; she’s only one person.
Instead, you could identify someone on the periphery of the community. You find Sally, a hip 20-something who is already sympathetic to the movement. Sally has a small friend group of four. If you can help Sally convince one of her friends to get involved, suddenly, half of their friend group is on board with your campaign. Already, you can see how the dynamics here on the periphery are quite different than at the network core. For the other two members of this friend group, doing civil disobedience for animals will quickly become something they associate with their social life in the church; half of their friends are doing it. The power of social norms means they’ll be easier to convert, especially if Sally and her friend keep reminding them about it.
By now you can see why Centola calls this the Snowball Strategy. Looking at the diagram again, we can imagine the new change creeping across the network periphery, then encroaching on the core. Once 10 or 15% have converted to the new norm, its momentum is palpable and impossible to ignore. Now, if the pastor gets on board, she’ll have the backing of an organic movement. The tipping point is within reach.
Spreading Complex Contagions
Over the last two decades, social network scientists have found more and more evidence that complex social contagions spread most effectively through strong ties on the network periphery. This is just the opposite of simple contagions, which spread rapidly through influencers and weak ties. (Centola effectively surveys much of this evidence in Change, with a lengthy appendix on further reading.) This research has been made possible by social media platforms and other digital technologies. These recreate the dynamics of social networks while enabling interactions and connections to be recorded and measured. Taken together, this body of research arms social change strategists with a new playbook, an uncanny inversion of the strategies listed earlier for viral spread.
Overcoming Resistance to Change: The New Model
- Change spreads from the periphery, not the center. Spreading change is about identifying key points in the network. But while simple contagions spread fastest through highly connected influencers, spreading complex change (change that faces resistance) is about identifying clusters on the periphery and building momentum into a snowball. If you could choose ten people to magically convert to start spreading a complex contagion, Centola suggests the most effective choice would not be ten “influential” celebrities, but a group of ordinary friends in the same neighborhood.
- Change spreads through strong, redundant ties. The “strength of weak ties” discovered by Granovetter is that they aren’t redundant. Weak ties tend to bridge across different social clusters, allowing information to spread more rapidly without much repetition. But to overcome resistance, behavior change needs to spread through strong ties with as much redundancy as possible. To overcome the countervailing influence of dominant social norms, people must become surrounded by enough other people making the change, people they see as part of their social group. This means that an activist’s most important resource is their strong social ties! An activist is likely to have far more success building a snowball effect among their family and friends than handing out leaflets to perfect strangers they’ll never see again.
- Network dynamics, not virality and stickiness. Viral marketing is premised on the belief that if you can design a catchy enough contagion, it will catch on. But Centola’s evidence shows that the shape of the network is a far better predictor of the spread of a change than any quality of the change itself. Even a change that doesn’t seem “sticky” will stick if it starts by building momentum in the right part of the network. Animal advocates need to focus on who we are targeting, focusing not on individuals but on strategically located social clusters.
Norms, Networks, and…
Around this point, you might be thinking, “OK Aidan, I can see how this all could be applied to a church. But how can we apply this to a whole society? It all seems a bit abstract.” Well, that’s a great question. I wish I’d thought of it myself! Because it leads us directly to what I personally find to be the most exciting thing about all of this. Centola’s work on social networks provides the most compelling answer I’ve seen yet to a question that has confounded sociologists for decades: Why do social movements (and protest movements in particular) seem to be such an effective strategy for changing social norms? And perhaps more importantly, how can we design social movements that are most likely to succeed in doing so?
The next (and last) article in this series tackles precisely those questions.
In the meantime, here are some even simpler ways animal advocates and organizations could incorporate social networks into their current strategies:
- Target social groups: Any behavior change efforts should be aimed at specific socially connected groups, not just individuals. To begin with, this could mean groups of friends, or established communities like churches, sports teams, etc. For vegan groups on college campuses, this means your community-building efforts are probably much more impactful than any mass public outreach.
- Concentrate efforts locally: Behavior change efforts could have a greater impact by limiting their geographical scope to cities or neighborhoods where messages can be reinforced as they bounce back and forth between interconnected social groups. Organizations could choose just a few cities in which to concentrate their resources. We can go even further by choosing highly connected cities like these ones.
- Target social identities: Currently, animal advocates think about demographics like gender, age, and race to predict which individuals might be easiest to win over. But these demographics are too general to describe social connectedness. Social identities connected to people’s professions (e.g. lawyers, tech workers), hobbies (e.g. soccer fans, gamers), and religious activities are more likely to describe their social networks. These social identities are large and populous enough to span society, but specific enough to be highly connected. Organizations could choose one or more specific identities to focus on.
- To the extent that these campaigns (and protest activity more generally) have a goal of influencing the public, they represent a shotgun approach, reaching individuals at random without any attention to social clusters or a systematic plan for altering the public’s views (that I am aware of).
- I’ve actually been flossing regularly for the first time in my life, but it had nothing to do with my dentist: two of my housemates floss twice a day, often in common areas. This is a perfect demonstration of the power of strong ties!