At first blush it is tempting to just assign #MeToo and #NeverAgain to the growing pool of hashtagged social movements that happen to get their teeth into the media cycle for an extended period of time. They both benefit from being related, in one way or another, to the infinite-scroll train wreck that is the Trump presidency. And most importantly, both are at the spearpoint of what looks to be rapid and in many ways shocking social change.

But these two movements are interesting for another reason: They force the question of why this is happening. Or to put it more forcefully, why is this happening now? After all, in both cases the triggering events or circumstances are far from unprecedented. Harvey Weinstein had been threatening, harassing, and abusing women in Hollywood for a very long time. And in the case of the Parkland shootings, it is unfortunately far from the first time a student in the United States has gone into a school and massacred some students with a semi-automatic rifle.

So what is different this time? In these situations, it is common to fall back on a class of folk sociological explanations based on tipping-point models: The triggering event was the straw that broke the camel’s back; it was what caused the dam to finally burst; people finally said enough was enough. But these sorts of explanations are unhelpful because they just beg the question: What is different about this time?

In both #MeToo and #NeverAgain, a big part of what is going on is that there has been a sudden shift in certain social norms. In the case of #MeToo, the norm that compelled women to keep silent in the face of predatory male behaviour has rapidly dissolved. With Parkland, what we are seeing is the steady unraveling of the norm, enforced in some (largely Republican) communities, that demands unconditional support for the Second Amendment and especially the NRA-sponsored interpretation of it.

***

One of the more curious features of the Weinstein case (like that of Jian Ghomeshi before him) is that once it was all out in the open, hardly anyone who knew the man or had any connection to Hollywood was surprised. “Everyone knew” Weinstein was a sexual predator. His behaviour was an “open secret.” Seth Macfarlane joked about it, Courtney Love warned about it, the TV show Entourage did a whole story arc about a monster named “Harvey Weinfeld” that in retrospect was less a parody than an attempt at an intervention. How can this be?

In their new book new book The Elephant in the Brain, Robin Hanson and Kevin Simmler have a nice example showing the difference between something that “everyone knows” and something that is “common knowledge.” If you send an email invitation to a party for 50 people and put every invitee on the CC list, the existence of the party and the invitees is common knowledge. Everyone on the list knows about the party. They also know that everyone else on the list knows about the party, and they know that they know that they know, and so on. But if you send the invitation to 50 people but put all the invitees in the BCC field, then something curious happens. Everyone on the invitation list knows about the party. But they don’t know who else knows about the party. So when it comes to all 50 invitees, “everyone knows” about the party, but they don’t know that the others know. They don’t even know who else is invited! So this is what Hanson and Simmler call “closed knowledge.”

Using BCC to create a space of closed knowledge has another important effect: it creates a norm of secrecy around the party — where and when it is, who is coming, maybe even its very existence. After all, if you are invited by BCC and not CC, you can reasonably assume that the person doing the inviting kept the list secret for a reason. Hence, your default assumption should be that you don’t talk about the party to other people, and you certainly don’t publicly advertise the fact that you were invited.

This is a bit of a constructed example, but it shows how something can be widely known but also not talked about because of the relevant constraining norms. This is why, historically, gossip has been such an important subversive feature in regimes of control — it is one of the main vectors for the creation of a space of open information in a regime that wants information to be kept closed. And of course there was a lot of gossiping about Harvey Weinstein. But as anyone who has been in a situation where a norm of secrecy is being used to control information, gossip has its own dynamics, its own norms, surrounding who you gossip to and who you tell about what you discover. Which means that a large number of people can know something, and also know that certain other people know, without it becoming “common knowledge”. Until someone spills the beans publicly (like the kid who yelled out that the emperor had no clothes) and suddenly everyone says yes, of course, we all knew that all along.

So in the Weinstein case, what seems to have occurred is that lots of women knew he was up to no good, and they even gossiped about it, but they were constrained by a very powerful social norm that demanded women just accept the behaviour of predatory men in Hollywood. And they could never be sure who opposed the norm and who supported it, even amongst other women.

How and why did the norm suddenly change? The Weinstein case is complicated, and the full tick tock of how it all happened will make a fascinating book. But the gist of it is something like this: A handful of women finally decided to go on the record with their allegations against Weinstein. The publication of the allegations against him the New York Times and the New Yorker gave instant credibility to the accusations, which spurred a number of other women to go public with their experiences, which created a snowball effect that has rolled through not just Hollywood but almost every other male-dominated industry.

In his paper Unleashed, Cass Sunstein tells the story this way: A small number of “norm entrepreneurs” who opposed the prevailing norm that protected men like Weinstein decided to actively work to change them. This generated a “norm cascade” that, within a very short period of time, first undermined and then utterly dismantled the existing norm. But again, why now? As Sunstein argues, what determines whether a norm entrepreneur is successful or not depends a lot on the initial conditions: who they are, how they communicate, and who they are able to influence. It’s like starting a campfire: You need to start with something very light and flammable like paper or birch bark and some small pieces of very dry kindling before moving on to the big logs, some of which might be wet or rotten.

In a sufficiently large community there will be wide variation in the thresholds people have for deciding when to oppose a norm. If someone with a low tolerance for sexual harassment raises the issue with someone with a relatively higher threshold, they might get told to relax or calm down or keep quiet and not rock the boat. A higher status objector (like an Oscar nominee, or someone with deep Hollywood connections) might have better luck opposing a norm, but as we have seen, some of the highest status women in Hollywood felt powerless to oppose the norm that protected Weinstein. For many of them, it probably felt like trying to light a wet log with a match.

But when the NYT and New Yorker went to press, the allegations against Weinstein suddenly had two things going for them. First, they had the prestige of the publications behind them. And second, through social media, the accusers were able to reach everyone in the relevant community at the same time. Instead of whispering in secret, they could shout in public. Those with low thresholds for norm violation could respond with their own stories, which led others with slightly higher thresholds to see just how big the silent majority really was and to add their own voices to the mix, and so on. A spark became a roaring bonfire in a very short period of time. After years of acting with impunity, Harvey Weinstein found himself friendless and defenceless. Social media, especially the hashtagged free-for-all of Twitter, played an enormous role in this.

***

One of the most heartbreakingly cynical tweets that makes the rounds whenever there is a mass shooting in the United States came from the UK columnist Dan Hodges, who in 2015 tweeted:  “In retrospect Sandy Hook marked the end of the US gun control debate. Once America decided killing children was bearable, it was over.”

This gets to the heart of the conundrum: if the murder of 20 six and seven year olds in a school shooting didn’t spur Americans to rethink their approach to guns, what possibly could?

But here we are just over five years later, and the #NeverAgain movement seems to have momentum that never manifested in the griefstricken aftermath of Sandy Hook. In particular, the movement leaders such as Emma González have attained a level of popular support and public profile that suggests that actual social change on this issue might be possible. (Possible, but not probable — it’s not wise to get too ahead of things on this issue, given historical precedent).

One reason for this is the age of the kids. Unlike the children at Sandy Hook, the Parkland students are old enough to speak for themselves. And as Tyler Cowen argued in a recent column, they are in a bit of a sweet spot, age-wise. They are mature enough to have thought-out views, to make arguments, give speeches, and gain a following, but they are difficult for adults to attack without seeming cruel. And as he points out, there is a historical precedent: when adults allowed their children to participate in civil rights marches in Alabama, it made it harder for the authorities to justify violent crackdowns.

But there’s an additional difference which is that Sandy Hook is in Connecticut, a state that already has one of the strictest gun control regimes in the United States. Florida, in contrast, is one of the most gun-friendly states in the union. Paradoxically, that might make a gun-control agenda more palatable this time around.

The reason is that it is probably the case that many gun owners, even many who publicly support the NRA, would actually welcome increased gun control. They constitute if not a “silent majority” at least a silent but substantial constituency who would be open to some new restrictions or regulations. But they are constrained by two things. First, they live in a community where there is a strong norm in favour of defending an extremely generous interpretation of the second amendment. And second, they don’t know exactly how many of their friends and neighbours and colleagues feel the way they do. They may talk about it privately or in small groups, but they are in many ways like the women in Hollywood who put up with Harvey Weinstein for so long. What they believe might be widespread in their community, but it isn’t common knowledge.

In Parkland, the students are acting as norm entrepreneurs working against the existing pro-NRA norms. Through the mass media (which gives them status) and social media (which gives them reach), they are able to credibly communicate to the entire distribution of thresholds for opposition at the same time. And as children and as victims of the shooting, they are inherently sympathetic figures. As Cowen puts it, it is far easier for Republican gun owners to give a victory to students from their own community than to distant liberals who would interpret it as a moral victory for their worldview, which involves dismissing conservatives as simply bad people.

Again there’s a precedent for this. Sunstein gives the example of restaurant and hotel owners who actually lobbied in favour of the Civil Rights Act of 1964, that forced them to stop discriminating against blacks in service and hiring. The reason is that many of them wanted to stop discriminating, but felt they couldn’t do so because in their communities there were strong norms that demanded such discrimination. A restaurant that unilaterally started serving blacks would be seen as either greedy or soft on blacks. In this case, a law forcing them to do so was actually liberating.

***

What’s the moral from all of this? Social change can be unpredictable, in part because it is highly sensitive to initial conditions, but it isn’t inexplicable. But the examples of #MeToo and #NeverAgain should caution us against the idea that moral progress and social change go hand in hand. Remember, the reason why so many countries finally gave women the vote after the First World War was not because men suddenly discovered that women were rational, intelligent, competent people.

Social change is more often a consequence of things such as war, civil disruption, or technological development. We are only now beginning to get a handle on the power of social media to dissolve long-standing social norms, for better and for worse. But it is also useful to remember that social change is also a realm for entrepreneurship. As both #MeToo and #NeverAgain are demonstrating, it is still in the power of a few dedicated and well-positioned people to change the world.