"Wouldn't It Be Cool If..." Science is a Scourge and We Will Regret It

In obvious ways, this is a better time than ever if you want to learn about science. There’s an immense number of free resources available to you - Wikipedia, YouTube, podcasts, essays from magazines and newspapers, newsletters, MOOCs, even science-based video games. And yet I look at my YouTube feed and I despair, for (almost) all of the science content I see is hype-ridden, click-hungry, and thus distorted in focus and proportion if not flat-out factually wrong. It’s common for people to blame us for our own feeds - you decide what to click on! - but I do in fact try hard to tailor my experience towards responsible pop science and skeptical voices. As someone who simply lacks the math and underlying base of knowledge to evaluate most of the scientific claims I see or read, I have to be careful in that way, and I have tried to be. Yet as a person that YouTube identifies as science-consuming, I’m absolutely inundated with the most sensationalistic, overwrought, misleading, and attention-fixated videos you can imagine.

Here are some things that you’d likely believe if you got all of your science information from YouTube.

• That the Many Worlds Interpretation of quantum mechanics is plainly and simply true, that it is “settled science,” that we know that there are multiple dimensions that are created every time a quantum measurement occurs and that these dimensions are similar to ours in their basic composition and physical rules, and that therefore there are some/many/an infinite number of copies of each of us, running around in other dimensions, living busy little versions of our own lives

• That useful quantum computers exist now or will imminently exist, that they are generalist problem solvers that will replace conventional computers, that they can/will immediately enable super-intelligent AI, that they could prove/will prove/have proved the existence of alternate dimensions, that they will enable us to cross the boundary into those alternate dimensions, some how, some way

• That we know that a previous intelligent race lived on our planet hundreds of million of years ago and built civilizations before going extinct, the so-called “Silurian hypothesis”

• That we have already developed room-temperature superconductors that will revolutionize our electrical grids, transportation systems, and computers

• That warp drive or other faster-than-light engines are known to be theoretically possible based on current science with no major impediments to implementation beyond costs and a few minor technical details, that such technologies simply will work if we ever getting around to building them, and that the farther reaches of the galaxy will someday be accessible to us despite the vast distances involved

• That the so-called simulation hypothesis is an actually responsible, reputable, falsifiable scientific theory which can survive the usual standards of rigor that we apply to scientific theorizing, and is in fact very likely true

• That scientists have already accelerated particles faster than light/back in time, in defiance of the conventional understanding of relativity

• That the colonization of distant planets that are capable of sustaining life is a known possibility using realistic projections of current technology, that the impediments to this effort are fundamentally political and economic rather than scientific and technological, that we can confidently believe that we will make the journey out to the stars using either faster-than-light travel, some some of human stasis or hibernation system, or through multi-generational missions, that the stars are our future one way or another

• That there simply are other intelligent species in our galaxy, that the sheer weight of numbers tells us this has to be true, and that we have to constantly litigate and relitigate the (so-called?) Fermi Paradox at various levels of specificity in order to explain the seemingly-silent universe around us.

There’s more you could add to this list, I’m sure.

As stated, I lack anything resembling expertise in these domains, so I won’t pretend that I can personally reject any of these claims. But I would certainly suggest that the relevant and sober science regarding these questions generally concludes that these claims are not supportable according to the traditional standards of scientific evidence. Some are more credible, some are less, but they involve a degree of conjecture and handwaving away of core problems, and always in the same direction - towards hype, of the kind that’s useful in the attention economy but not great for public understanding of science. It’s not that there’s a ton of big channels out there that will baldly state all of this stuff as fact, although there’s some for any given claim. It’s that any provisos, qualifications, exceptions, or warnings are buried so far down in the mix, in this genre, that only the “wouldn’t it be cool” survives, not the “if.” Watch one science video about a topic from the list above, you might or might not leave with an appropriate level of skepticism. Watch a bunch of them, without a proper grounding in the relevant evidentiary standards, and you won’t.

I like this channel, Cool Worlds, which is a product of Columbia University and headed up by the astronomer David Kipping. And this video is a pretty good deductive argument against the simulation hypothesis, which has gotten very popular lately among Silicon Valley types and the Joe Rogan set. The problem is that a video like this inevitably implies that the simulation theory needs to be rebutted through an argument like the above. But we don’t need to rebut it. We don’t need to rebut the simulation hypothesis because the simulation hypothesis does not meet minimal evidentiary requirements as a scientific concept: there is no empirical evidence that we live in a simulation and, indeed, the hypothesis is entirely non-falsifiable. It therefore does not need rebutting according to the conventional standards of science as set by people like Russell, Popper, Quine, Kuhn, Lakatos, etc. You can contrast simulation theory with something like string theory, which has often been criticized for being currently non-falsifiable, but at least theoretically falsifiable; because those positing the simulation hypothesis can simply dismiss any specific objection to the hypothesis as being an expression of the simulation, the hypothesis can never be rejected. This is a constant and pernicious element of the current state of science media: even the more responsible voices and publications have to react to the discursive environment they find themselves in, which means they’re constantly swimming in the water of hype even if they’re doing their best to minimize sensationalism.

Right now, AI is of course the endless font of breathless hype and irresponsible science coverage. But the Red Queen Game that is quantum computing has perhaps been the paradigmatic case for the past decade or two. Quantum computing is an actual branch of research science that has seen actual progress, which is helpful. The advantage of quantum computers is based on the fact that, where a conventional computer uses transistors that have a binary state (good old 0 or 1), quantum weirdness allows for superpositions of states in between 0 and 1, which has benefits that I won’t pretend to understand. These computers have some very useful applications, particularly in cryptography; in fact, there are legitimate fears that widespread access to quantum computing will make encryption (and thus online security) impossible. So there is indeed a serious side to all of this. But quantum computing has also been subject to all manner of ridiculous claims that aren’t based on science at all, and which have a set of bad incentives common to science communication in the 21st century: researchers who need bold claims to secure the necessary grant funding, media entities that need hype to get clicks and ad revenue, and corporations that need to juice their stock price. None of those things point in the direction of careful and judicious evaluation of the scientific evidence, I’m afraid.

A perfect story to underline all of this is Google’s maybe kinda sorta claim that their quantum computer Willow proved the existence of alternate dimensions. The big story is the pursuit of quantum supremacy, which simply means being able to demonstrate that a quantum computer can perform given functions that conventional computers could not achieve in a reasonable timeframe. Google’s Willow seems to have achieved that, solving in a matter of moments what a conventional computer would need trillions of years (many multiples of the age of the known universe) to accomplish. Cool, cool. The conversation from New Scientist Weekly above suggests that the way Google is framing it is horseshit, but the actual improvement that Willow represents might very well be important and impressive, particularly when it comes to error rate, the usual bugaboo for quantum computing. IDK.

The trouble comes when you add the suggestion that got all the press, the notion that this feat provided evidence for the existence of many worlds - the notion that, in order to achieve the speed advantage described, the calculations must have been being performed simultaneously in many dimensions, whatever that might mean. This provoked the predictable spasm of irresponsible headlines and social media hype. It also provoked a wave of backlash and skepticism, for which I’m grateful; this Sabine Hossenfelder video is a good place to start. (It seems that Willow’s performance proved the many worlds interpretation… if you assume that the many worlds interpretation is correct.) The question, as usual, is what made a bigger impression, the initial wave of hype or the eventual ripples of doubt. I will again remind you that there was a period of time, about ten years ago, when the tech press was putting out pieces about how we’d soon all have industrial-grade 3D printers in our homes, making all of our durable goods.

None of this is surprising. Of course Google is suggesting that their quantum computer (maybe kinda sorta) proved the existence of other dimensions. Google has a stock price! And you don’t get to an almost 2.5 trillion dollar market cap by being sober and judicious in your marketing efforts. This situation is one of the best examples I can think of when it comes to useless tech hype for a very simple and direct reason: the head pf Google’s quantum AI scientist himself, Hartmut Neven, has stated plainly that the calculation performed by Willow has no practical use, which makes sense when you consider that it was specifically chosen to demonstrate the largest possible gap between conventional and quantum computing. Quantum computing in general is likely to have significant impact on very specific fields and applications, but given that it’s generally no better (and often far worse) at conventional computing tasks, there’s no good reason to believe you’ll be packing your Quantum Macbook on a long flight anytime soon. Is the kind of basic/foundational science this story represents important? Of course! Could this breakthrough be practically relevant in the nearish future? Certainly could be, in some limited sense. But the actual proximity to near-term usefulness matters, and absolutely every incentive of the current media economy militates against providing that kind of perspective. You would never know, from much of the coverage, that a quantum computer is very much not just a superpowered conventional computer. “This update to Google might eventually break HTTPS, but we don’t really know, ¯\_(ツ)_/¯” doesn’t get a lot of breathless headlines, I’m afraid.

And, if you’ll forgive me for talking way above my paygrade and for embracing a conservative scientific standard, to prove the existence of many worlds you’d have to produce empirical evidence of many worlds. Sorry.

When I was in grad school I met another PhD student, a biologist, who was particularly focused on science education. And she laid out the field’s dilemma in a way that I found quite clarifying: the more that you try and make science seem cool and fun and attractive to kids, the more you misrepresent what science actually is, which will inevitably result in some disillusionment in the future, whether that’s in high school or college or on entering the workplace. Actual science, particularly biology and chemistry and materials engineering and related, is much more like working in a factory than it is having sudden “Eureka!” moments, lit by flashes of lightning. Real life chemistry labs and biochem facilities and chip foundries and assorted are usually pretty sedate places, staffed by people who come in and laboriously build knowledge one painstaking attempt at a time. It’s largely a grind. Ask practicing researchers, they’ll tell you!

What’s more, as in any part of the labor market there’s more people seeking attractive jobs than there are openings. If you want to work in the field tracking wolves in Yellowstone for your job, or you want to be one of a small handful who can get your hands on the money and resources to actually build potentially-functional quantum computers, etc., you need to understand that there’s a ton of other people who want to do the same thing, so you need to go to a fancy school and distinguish yourself and get into a competitive lab where you’ll be exploited by your advisor for years…. There are certainly inspiring moments in everyday scientific research. But if you tell a bunch of 10-year-olds that studying science means doing cool experiments with pleasing visual results, and then they get to a later level and discover that they’ll be putting nearly-identical substances into centrifuge tubes and hitting the button over and over again for years of their life, they might feel like it’s a bait and switch. I’m afraid it’s not all making volcano eruptions out of backing soda and vinegar.

It all makes me think of the absurd Thomas Edison vs. Nikola Tesla furor of an earlier internet age, spurred in large measure by the endlessly annoying webcomic The Oatmeal. That comic argued, with comical passion, that Tesla was a true genius and that his supposed rival Thomas Edison was an evil fraud. This argument was repeatedly skewered, in large measure because there were some rather direct distortions of Edison’s record within it. The comic also ignored Tesla’s embrace of eugenics - actual eugenics, not the fake Twitter kind. But more important than any character issues was the simple fact that it’s the vision of Edison, not the vision of Tesla, that best fits most modern science. Setting aside the degree to which the caricatures of them are true, when we think of Tesla as a half-mad genius whose work was driven by flashes of profound inspiration and of Edison as a dogged but ordinary man slowly plugging away at problems with a systematic method, we must accept that the latter is what most science is actually like. Of course genius has periodically propelled science forward, and it will going on doing so into the future. But ordinary day-to-day scientific progress is the result of fundamentally boring procedures that have been developed over time to maximize the possibility of incremental progress. And it’s exactly that boring industrial science that “Wouldn’t it be cool if…” science communication has little time for. In the end, the triumph of flash over boring substance is not going to be good for the younger generation’s conception of progress or for science itself.

You can and should pursue a good number of skeptical sources even if you’re someone who likes the hype too. Skeptic Magazine still puts out good stuff, as does New Scientist. Hossenfelder is controversial for her stance on modern physics, but she’s a reliably and usefully skeptical voice. There’s also lots of great individual pieces out there, such as Kim Stanley Robinson’s remarkable (and remarkably frank) effort at popping the bubble of human colonization of other worlds. In the broader perspective, I certainly can’t tell any scientists how they should feel about any new piece of science information. But I would suggest a simple rubric for others who are, like me, entirely amateur consumers of science media: if you hear some science news, and there’s cooler and less cool explanations for it or potential consequences of it, side with the less cool until there’s strong evidence pushing you to do otherwise. Maybe very strong.