How the Internet is Broken: Big Questions and Bad Answers

I’ve always admired people who ask big questions. Who don’t just see the world with received wisdom, but interrogate it. The internet is helping people ask big questions by connecting them with others who question, who wonder. Is our water really safe to drink? Do kids from different ethnic and economic backgrounds really have a chance to succeed economically? Is our education system teaching people the right things? How is corporate control over high-end content production shaping our culture? How are changes in the ecosystem of journalism reshaping what we collectively believe? Asking big questions is a critical first step towards changing the world for the better.

What if, though, the internet is better at helping people to ask big questions than at getting them to correct answers?

I need for a moment to explain what I mean by a “correct” answer. Epistemology and sociology of science teach us that “truth” exists, but our access to it is indirect. What we all agree on is our best attempt at getting to truth.  This is as true for the most elite scientific circles as it is for day-to-day life. Scientists use the process of “peer review” to come to consensus on our best insights into truth, at the moment. And by critiquing and revising one another’s work, our insights continually improve. Knowledge (justified, true belief) is socially constructed.

My contention is that the internet is especially good at helping start this process of social construction of knowledge, but really bad at finishing it. Because it’s too easy to jump to incorrect conclusions.  Especially when you have a community of like-minded friends eager to support those conclusions.

An Example: Vaccination

Here are some important questions: Does the pharmaceutical industry have the best interests of people as its top priority? Is it possible that pressures to improve profits ever outweigh good medical judgment? Is a market-driven model for medical innovation really our best choice? Here are some more: Why are rates of autism rising? Why do children who begin to develop normally sometimes suddenly regress?

Asking those questions is important. Questions about autism are of life-changing significance for families with children on the spectrum, and the bigger questions about the design of our incentive system for medicine are critical for everyone. Groups of people talking together on the internet can help one another to see outside the frame they are trapped in, and ask big questions.  What’s harder is coming to correct conclusions.

Unfortunately, we don’t yet have satisfying answers to those big questions. We don’t know what causes autism (or even how many different medical phenomena make up what we call “autism.” The system that funds medical research has strengths and weaknesses, and the magic profit motive/market don’t always steer innovation in good ways.  For people who desperately need answers to these questions, it’s easy to find someone else who thinks they have an answer. And agreeing with their answer is easy and satisfying. Much easier than the harder truth: we don’t yet have good answers.

It’s easy to see how people might wonder if vaccines cause autism—many children regress developmentally right after receiving vaccinations. It’s  been conclusively proven that this is correlation, not causation. There is absolutely no doubt: vaccines do not cause autism. But it’s emotionally easier to for parents to believe they have a reason and someone to blame than the stark truth that their child’s autism is not yet understood.

Good Questions and Bad Answers

I see this pattern over and over again: people are asking good questions but jumping to believe in bad answers. My students and I have been studying people who believe in “chem trails,” the idea that the visible condensation trails behind airplanes are evidence of deliberate spraying for nefarious purposes (like climate control, or deliberate population reduction). Chem trails believers ask some great questions: Is our environment really safe? Are our systems for monitoring pollution accurate and honest? Are corporations being held accountable for environmental damage they cause? Asking these questions is important. Unfortunately, this particular community of people have grabbed onto something visible (condensation trails behind airplanes) as an explanation for the source of problems. It’s easier to blame something you can see than the harder truth that pollution (like lead in the water in Flint, Michigan) is typically invisible. It’s also easier to settle on a centralized explanation (the government is manipulating us deliberately) than a decentralized one. Changes in the environment are in fact part of a complex socio-technical system made up of human and non-human actors who have diverse motives and degrees of accountability. People leap to easy answers: choosing to believe there is a single, well-understood reason for a phenomenon when there are actually many poorly understood ones that interact in complex ways. Choosing to believe in a single nefarious actor rather than a set of complex, emergent phenomenon with diverse actors who are both well and poorly intentioned. Choosing to believe a provided answer that other people support rather than the harder truth that answers are currently unsatisfying and partial.

Solutions?

The fundamental solution to this problem is education. We need to teach school children to be better at interrogating evidence and deciding (for themselves) what to believe. We need to teach kids to be better critical consumers of information.

We also need to do a better job of teaching kids about the scientific method. A growing group of people now believe the earth is flat. To empirically “prove” the earth is flat, all you have to do is to conduct an experiment to show the curvature of the earth and do it badly.  We need to teach our children to observe the world accurately and debunk pseudo-science.

Finally, the design of the internet could be improved. We desperately need more meta-data. For any given piece of information online, it should be easily possible to ask: what supports this? What people will do is shaped by what is easy to do, so this meta-data needs to be incredibly easy to see. And there need to be multiple sources of meta-data to protect us against deliberate manipulation, so people can choose their gold standard.

The hardest part is how to make this happen. We need a change in the economic system underlying the internet that financially supports the creation of accurate information. We do not currently have a workable “marketplace” for ideas. We need to re-engineer the system so that helping people to find better answers is profitable, and good content spreads faster than bad.