A decade of failures in Science Communication.

Eleven years is a long life for a science blog. Southern Fried Science was born in 2008, when the main writers were all graduate students. Over the last decade the online landscape has changed. Science Communication changed with it, adapting and evolving to meet an ever-shifting ecosystem. Looking back on the last decade and thinking about the next, it’s becoming easier to see where we went wrong. It’s not quite as easy to determine what we need to correct the course.

This is not a scientific assessment, this is my own personal observations from the last decade of running Southern Fried Science, from teaching Social Media for Environmental Communications for the last 7 years, from working with Upwell, one of the most dynamic and visionary ocean NGOs ever conceived, from helping build and launch multiple online platforms, dozens of novel programs, and hundreds of outreach campaigns, and from spending a lot of time since November 2016 reflecting on what we’ve done wrong.

That Hideous Deficit

Do we really need another 200 words on how bad the deficit model is and why it needs to die?

Apparently, yes.

The basic premise: that science perception and policy is shaped by an information deficit and that if we just make good science education content and spread it, we can combat the spread of misinformation, people will learn, and everything will get better.

It doesn’t work. It never worked. And it ignores the reality that misinformation is manufactured for political and financial gain, with tremendous incentives and, often, far better funding than science outreach campaigns. But beyond that, multiple studies have shown that, when confronted with information that challenges their fundamental world view, people don’t throw out their worldview, they reject the science, creating a more entrenched and intractable audience.

Outsourcing Digital Infrastructure

Something happened over the last decade. Blogs died. Or they didn’t. Or they moved. Or we went to Twitter or Facebook or Podcasts. But something definitely changed, maybe?

When science communication exploded online, it was largely a DIY affair. We built our own blogs and hosted them and coded them and wrote those first posts in HTML. Platforms came along like Blogspot and WordPress that streamlined the nuts and bolts, making blog creation easier without sacrificing ownership over the content. Then networks formed, then social media exploded, and now we’re here.

What changed is not that blogs died, but that we increasingly outsourced our infrastructure to third parties, some of which had entirely different goals. Lacking any real way to fund an expanding blogosphere, we were forced to compromise, selling off parts of our creation in exchange for infrastructural support. Some of those platforms were great. Some of them turned out to be Epstein-backed ventures built to help to whitewash the reputation of a serial child-rape trafficker. Some of them were part of a catch-and-kill campaign from larger media outlets trying to squash the competition. Some were great, and then sold off to less great organizations who used all the great science writing they may or may not now own to lend legitimacy to pseudoscience. Some were, inexplicably, all of the above.

There are huge benefits to building a science outreach campaign on top of an existing platform. There is an inbuilt audience, the digital infrastructure is already in place, the support cost is often negligible, and the launch process is streamlined. But the trade off is enormous and often goes totally ignored.

As more and more outreach is outsourced to third-party platforms, more and more outreach gets tailored to the specifics metrics those platforms provide. So when a site like Facebook fraudulently represents its video metrics, metrics that major campaigns built their strategy around, it can destroy entire programs.

Our Howling Void

People like communities. Give humans space and the ability to network with a large enough population and we’ll form communities. This is generally regarded as a good thing.

We can talk endlessly about communication bubbles and how we’re not reaching the audiences we intend to and keep spinning around in circles and that’s pretty much the point.

Where this process fails is in how it breeds complacency in communication strategy. Over the last decade, we’ve seen a strong shift on outreach plans away from novel, weird, creative approaches built on limited, localized infrastructure and towards the now-standard methods of “We’ll do a blog/Twitter/Reddit AMA/Facebook/YouTube thing”. When done well, those can yield phenomenal results, but they can just as easily distort the impact of that outreach.

All these platforms provide baseline metrics, which help data hungry scientists satisfy their, and their funder’s craving for metrics, but they do little to actually measure the true impact of outreach efforts. And that can lead to a phenomenon where science communicators think their campaigns are doing really well when, in reality, their audience is not just smaller than expected, but comprised not of the people they need to reach, but of folks who were already strongly engaged.

When you build a strong network, it’s harder to break out of that network. And when online outreach gets prioritized over other kinds of outreach, in part because it’s easier to demonstrate impact through metrics, the outreach campaign has failed to serve its audience.

The War on False Balance

Fifteen years ago, the best and brightest in science communication were fighting against False Balance — the artificial elevation of fringe ideas in popular media in order to provide a counterpoint to consensus topics. The thesis was simple: if, for example, 97% of climate scientists acknowledged global warming, then having climate change deniers featured opposite experts created a false sense that there was more disagreement than there was. On the surface, that’s not wrong.

Even John Oliver got into the fight:

But what the warriors against false balance failed to account for was that the engines of the denial industry were far better funded, far more aggressive, and far more media focused than expert activists. For a climate scientist, their work is the goal. Media attention is a bonus but understanding the science of our changing climate is the priority.

For the denial industry, there is no work, their entire strategy is media saturation.

Pushing back against false balance didn’t lead to more representation of expert knowledge in popular media. Pushing back against false balance eradicated the expectation of balance. False Balance wasn’t keeping experts off the airwaves, it was the last fragile bulwark against a flood of uncontested bullshit. By pushing back, we gave mainstream media a license to promote the fringe, unchallenged. The War on False Balance didn’t give us accurate representation of science in the popular media, it gave a criticism-free platform to the anti-vaccination movement, climate change denial, and the Goop Lab.

What’s next?

So what can we learn from the biggest science communication failures of the last decade? We need to focus more on storytelling, connecting with the audience on their own terms, and less on deficit-thinking.

We need to think seriously about how to build our own platforms and tools for dissemination of expert knowledge. That kind of work requires tremendous labor, and so we must also think about how those platforms can be inclusive and provide space for others.

We need to be conscious of the fact the social media outreach has become the default and to always explore new avenues and audiences, both online and off. We need to break out of cycles that enrich institutions that we find unethical while recognizing the burden that places on individual communicators.

We need to ruthlessly pursue media representation to tip the scales away from politically-motivated nonsense. We need vision, but we also need coherent strategies and tactics to confront pseudoscience where it rises.

All of which is to say that Science Communication is a full-time job, but is rarely treated as such.

We’ve learned a lot about how to use this new digital ecosystem to share the world with, well, the world. According to the best available science, we’ve got about a decade left to get things right and avoid the worst possible outcomes from climate change.

Good luck.

4 comments

  1. Carolyn Thaler · February 12

    Great article!

  2. Ken Huang · February 12

    This a topic that I’ve recently became very passionate about. I would love to hear more about your thoughts on what needs to change in the current sci comm landscape. We have a lot to work towards and I hope that these discussions will help us pave a strategy forward.

  3. Pascal Lapointe · February 13

    Also, don’t forget that you have allies out there. Despite all you can think about the medias, there are science journalists who know their stuff, and who are as tired as you about the false balance and the deficit model things. Problem is, those science journalists are rare, because general medias do not think they are important, and science medias are poorer and smaller. But if scientists do realize that having allies is important, they could play their part in pushing medias to hire more science journalists —being more like the cultural communities, who take a stand so that their local paper do talk more about culture, or being more like all those niche markets who are supporting their specialized medias.

Comments are closed.