Nature of Science
<< Menu | Videos | Risk Management >>
Script
This video is called “Nature of Science” and is part of the expansion pack accompanying the orginial video “How It All Ends.”
This video will explore the nature of science a bit, looking at how it is unavoidably tentative and uncertain. The purpose is that we can then do a better job of putting into context the things we hear about the science of global climate change.
Let’s start out with my assumptions. If we’re talking about the meaning of life, then science can be informative, but is just one tool of many equally valid ones, like faith, love, and direct experience. But when we’re talking about trying to predict and manipulate the physical world, I think that science is our best bet. It’s certainly got by far the best success rate. As Carl Sagan observed, if you want to know when the next eclipse of the Sun will be, you might try magicians or mystics, but you’ll do much better with scientists. If you want to save your child from polio, you can pray or you can inoculate.
One more thing before we dive in: scientific thinking and critical thinking in my mind are essentially the same thing. So as I talk about how science goes about figuring out what to believe, underneath it all I am at the same time suggesting how we as individuals—as citizens—should go about deliberating issues.
It’s a well-established psychological phenomenon—and, in fact, is simply human—to start out with your beliefs, and then go looking for evidence to support them. The problem is, we tend to forget or simply not hear evidence that contradicts our beliefs. I mean, who wants to be shown that they’re wrong? Formally, that phenomenon is called “confirmation bias.” The devilish result is that if you’re not diligently aware of it, you could be served up a plate of equally balanced evidence, and come out convinced that yours is the viewpoint that was better supported by the evidence, because you gave greater weight to the evidence that agreed with what you already believed, and discounted—or simply didn’t hear—what contradicted it. So confirmation bias can serve to actually reinforce misconceptions in the face of evidence. That’s why it’s critical to be vigilant about it in your own thinking, and why you’ll hear me refer to it again and again.
In science (and in criticial thinking—like we should all be trying to do in the whole climate change debate), it’s the opposite. Instead of starting with beliefs and then looking for evidence, you start by looking at whatever the evidence is, and then use that to form beliefs. I think that’s pretty much what a chemistry professor of mine once meant when he was teaching us about climate change. He said “Get informed, and let it change you.” That’s sort of the nutshell of how a good scientist might go about advocating for something: he doesn’t tell you what to believe. He just reminds you: start with the evidence, and move to belief, instead of the other way around.
“That’s exactly what I’ve been saying!” I can hear the shouting in my head right now, from some online commentors who’ve latched on to my previous videos about climate change. “Why don’t we just go with the facts!??”
Hey, sounds good to me. Simple, right? Just go with the facts? The sticky part is determining what exactly are the facts. Here’s an example. I’ll give a series of increasingly complex statements, and you think about at what point we can no longer simply agree it’s a fact, and instead have to do some interpreting.
[Behind burning candle] I have a candle in front of me. Fact. The candle is burning. Fact. I’m sitting in a chair. Here you might ask for more evidence before we pronounce it fact, because you can’t see it, so how about if I showed you? Okay, with a little checking—fact.
The problem is, I’m not really sitting in the chair, because I’m not actually in contact. What’s really happening is my outer electrons are repelling the outer electrons of the chair strongly enough to make me hover imperceptively above it, like magnets that can push on each other without touching [DEMO]. Okay, so I’m tricky. “Don’t be such a dork,” you say. “Some things are just obvious.” Well [eat candle], one of my favorite quotes on the matter is from Buckminster Fuller. In fact, students have to walk underneath it to get into my room. He wrote “Everything you learned in school as ‘obvious’ becomes less and less obvious as you begin to study the universe.” [Open mouth with “Ahhh.”]
This becomes a central point. Because, while we may all agree that—for all intents and purposes—I am sitting in this chair, when we shout at each other about whether the globe is warming or not, it turns out both claims are subject to the same question: how are we to decide whether something is a fact or not? It’s not always as clear cut as we’d like.
This may seem like splitting hairs, but it becomes kind of important if you have a question about a complex system or a really important issue, like: gee, is that asteroid going to hit the Earth or barely miss? Is this case of bird flu a human-to-human transmission or not? Is the globe warming or not? Are we the ones doing it or not?
“Sophistry!” you cry. “We can just look at the evidence.” Well, problem is, evidence still needs to be interpreted, which can be done poorly or skillfully. You see webbed foot tracks in the hall, come across a shimmery green feather, and hear a quacking sound. You conclude there must have been a mallard duck who recently passed by. It’s obvious. But is it possible it’s actually a kind of duck you’ve never seen before, and had you been better trained as an ornithologist, you would have known that the green was slightly the wrong hue for a mallard, and the tracks a little too big? Interpreting evidence well takes skill, training, and experience.
You wouldn’t propose lowering prescription drug costs by hiring my high school chemistry students instead of people with Ph.D.s to research the drugs, would you? They both look at the same printouts from the same machines: who’s interpretation of the evidence are you going to trust? “Well then, let me do it myself,” you say. Um—go for it. But then don’t be expecting me to accept your drugs, which is the case with climate change, since it’s global, which means you’re not the only one affected by your decision. I’ll stick with the professionals, thanks.
Here’s an example I often give my students. I tell them that we’re going to get creamed in Friday night’s football game, because—have they heard?—the opposing side’s offensive line has an average weight of just over 300 pounds! That usually worries them, until I tell them that the linemen weigh 110, 103, 98, 97, and 1120 pounds. That leads to a discussion of the difference between the “mean” average and the “median” average, and gets them to question their faith a little bit in the reality so obviously implied by such simple numbers as “the average.” If something as simple as the average can be so tricky, how come we’re okay with Joe Schmoes like you and me doing armchair analysis of climate science—one of the most complex topics in human history—instead of leaving it up to the scientists?
Why does evidence need expertise to interpret it? Because things are almost always way more complicated than they seem.
I once cornered a Yale University particle physicist at a wedding reception, cuz even though I teach physics and chemistry, I’ve always got some questions myself, and no one around to answer them. Anyway, I asked him how big an electron really is. I’d been wanting to know for a while, so I was determined to get a solid answer. An hour and several diagram-covered napkins later, I finally got him to grudgingly assent to a single sentence answer that we’d negotiated like it was a UN treaty. The deeper you go, or the bigger the system (like climate), the less accessible the “evidence” is to easy interpretation. (Fair warning: if you’re an expert in some field of the physical sciences, you’ll probably want to avoid me at parties. . .)
Yeah, but not everything is as complex as the climate, you say. You’re right, some things are simpler, like 1+1=2. Here’s Bertrand Russell and Alfred North Whitehead’s proof of that outlandish mathematical statement.
This is part of why all science is inherently uncertain, and tentative. Because the world is tremendously complex. So how do we get any answers? Well, you delve as far into the complexity as you need to for your purposes, or as far as you can get with your measuring instruments, and then you make an explicit estimate of how close you think you probably got to the “true value,” acknowledging that you’ll never get there.
The goal, of course, is to make that uncertainty as small as possible. There’s a couple basic ways of doing that.
The first is to be very careful about what biases—or preconceived notions—the scientist brings to the table. The scientist Konrad Lorenz summed up that duty when he wrote: “It is a good morning exercise for a research scientist to discard a pet hypothesis every day before breakfast.” Why? Because if you aren’t aware of your preconceived notions, then you are susceptible to the trap of confirmation bias—starting with belief, and then looking for evidence, rather than the other way around.
This can be insidious, because you don’t realize you’re doing it, and as a result, you become more confident of your conclusions than the evidence really merits.
The author Douglas Adams put it perfectly when he observed that “assumptions are the things you don’t realize you have.” That’s what the candle thing was about.
Here’s a couple more examples. You’ll probably be on your guard now, but see if you can do more than just avoid being tricked. See if you can identify the assumptions you hold that allow me to mislead you.
[In front of board with 1+1=10 on it.] If I asked you to make some simple observations, you might say I’m sitting in a chair in front of a whiteboard, which has an incorrect equation on it, and my toy is missing a green ring. Well, we’ve already established that I’m not really sitting on my chair—I’m hovering imperceptibly above it. And what if I told you that’s not a whiteboard—it’s a showerboard from Home Depot? Or that I’m not missing a green piece, but a blue one [switch and reveal]. Or that this is “base two” math, and you approached it with the wrong assumption—that it was standard “base 10”—so that YOU were the one who was wrong, even as you pointed your finger at me? And I bet you thought this was a fancy hat. Well really, it’s folded newspaper. In each case, you make an unconscious assumption, which leads you from “the evidence,” to a totally incorrect conclusion. Not because you’re dumb, but because you didn’t have the appropriate training or experience to be qualified to interpret the evidence.
So in scientific or critical thinking, you take great pains to identify the assumptions you don’t realize you have, so that you can account for them, and not wind up with a wrong conclusion when you interpret the evidence.
Okay, okay, you’re saying. Let me try another one. I’m ready this time. This time you’ll need a pencil or pen, and a piece of paper. Hit the pause button while you go get one.
What I’m going to do is flash an image on the screen for just an instant. Your job is to reproduce it as accurately as you can on your paper. I’ll just flash it for an instant, and it’s not fair using the pause button. Ready? Here we go. [Flash “Paris in the the Spring.”]
Okay, now press the pause button again, and do your drawing. When you’re done drawing, play the video again.
[Playing with some toy.]
Okay, have you reproduced faithfully what you saw? Here it is again. See how you did. [reveal]
If you got it right, that means you’re thinking more like a scientist, trying to be deliberately conscious of your assumptions. Well done. Most people write “Paris in the Spring,” when it quite clearly says “Paris in the the Spring.” Why? Because the human brain is amazing. When it doesn’t have the opportunity to fully examine something (a picture, a sound, a social interaction, a political problem), it fills in the blanks using past experience.
That’s great, and really really useful, but the problem arises when we don’t realize we’re doing it, because it can cause mistakes. When I get disheartened with all of the really confident and totally incorrect stuff I hear from most people who are skeptical about climate change, I have to remind myself that their brains are just doing what they are supposed to: filling in gaps in a really complicated picture, using past experience. For instance, I often hear: “How arrogant to think that humans can change the planet—we’re so small.” Now that you’re more aware of how bias and preconceived notions influence conlusions, can you identify the past experience coming into play there? It’s probably that throughout human history, the weather and climate have always been acting on us, and never the other way around. So I guess it’s not surprising that people feel that way.
But it is disheartening, because I wish they would be a little bit more humble. To acknowledge that—hey—you might be wrong. Think of it this way: the only way to ever improve, is to admit that you might be wrong. Not one of us is infallible. That means that each of us—you and me included—is right now carrying around some beliefs that are mistaken. If we don’t acknowledge that we may have some, then we’ll never have a chance to get rid of them—to trade them in for more correct or more useful beliefs. That means you’ll never improve, and will die no more correct than you are right now. I don’t know about you, but the idea that I am right now as good as I will ever be is oppressive to me—as well as being flat out ridiculous. I mean, what are the chances that you know everything right now?
That’s one reason why I got frustrated during the online debate about my original video “The Most Terrifying Video You’ll Ever See.” When I reworked my argument in response to some holes that people had poked in it, a couple people essentially said “So why should we listen to you now, since you admit you were wrong before?” and sat back smugly, convinced they’d won the debate. To them, I lost credibility because I changed my argument in response to the critiques. That’s just crazy talk! In science and reasoning, admitting you’re wrong makes you more reliable, because in the future, people can trust that if you’re wrong, you’ll change. If you never admit you’re wrong, you lose credibility, because your claims of being right simply become unbelievable. No one is right all the time.
[At board] In fact, I would argue that it not only increases your credibility with others, but it increases your happiness to admit you’re wrong. Here’s what I mean. Let’s say you choose to belong to the group of people who never admit they’re wrong. In that group, there are two subgroups: those who actually never ever make a mistake, and those who sometimes make a mistake. If you’re always right, then hey presto—life is good. But if you are one of these people, bad things happen, cuz sometimes you’re going to be wrong, but not admit it. You get into nasty fights, you lose credibility with people, and you never learn anything new.
Now let’s say you choose to belong to the group of people who will admit to themselves and others when they’re wrong. Again, two subgroups: those who never make a mistake, and those who sometimes do. Again, being actually infallible is all giggles and joy. Here, when people admit it when they make a mistake, they can take that opportunity to fix it: they learn new things, they have less nasty conflicts, and people not only like them better, but respect their opinion more.
Wouldn’t you say there are probably precious few of these people in all of human history? What are the chances you or I are one of them? Since we’re almost certainly in one of these groups, don’t you think this is the better bet than this?
[Back at desk.] So how does this work in science? First, as I mentioned, scientists acknowledge that neither they nor their instruments are perfect, and so they always include an estimate of the error or uncertainty in any scientific statement. Second, scientists take great pains to identify and isolate their assumptions, trying to identify and eliminate errors that they may be making. Third—and this is terribly important—they put their work out there and ask for criticism, so that weak points can be identified and strengthened, and the uncertainty reduced. That’s why it’s so important to ask if the statement you’re hearing about climate change has been “peer-reviewed.” That’s the official process that science goes through to sift the solid, credible ideas from the sloppy science. Although it doesn’t always work, it is a bruising, messy, drawn-out process designed to only let the best, most robust ideas float to the top. If something has been peer-reviewed, generally that means that it’s methods are up to snuff, and the scientific community thinks its worth looking at. It’s getting close to “the best answer that science can give us.”
Keep in mind, it doesn’t always work. Sometimes a peer-reviewed scientific article is shown to have significant problems. Guess what happens then? The peer-reviewed journal that published the research admits it, and sometimes even formally retracts the article, apologizing in the process! Why? To increase their credibility!
Peer-review is the process science uses to get closer and closer to the truth, but it is critical to remember in this whole climate debate: science never claims to actually gets there. That’s the surprising thing: science—that most precise and anal of all human endeavors—is also the one to never claim to know the truth. Isn’t it ironic? Don’t you think?
Another dynamic of science that’s worth noting is that of establishing when one thing causes another, and when the two things are just correlated. Here’s what I mean. If you look at this chart, it is clear that as the number of pirates in the world has decreased, the average global temperature has increased. There’s the evidence, and no one disputes it. So, what’s the interpretation? That the lack of pirates causes global warming? That pirates combat global warming, and therefore we should start some pirate schools ASAP? This is an example of correlation, two things whose trends track each other. But science is careful to not yet say that one causes the other.
For a serious example, it turns out that left-handed women contract breast cancer at a higher rate than right-handed women. So does left-handedness cause cancer? How does science go about answering that question? Well, it’s complicated, but this much is useful for us lay people to know: if two things are correlated, but scientists can’t find a way to feasibly explain the mechanism by which one influences the other, then they are not considered cause and effect. That doesn’t mean science says “They aren’t cause and effect.” It means science says “We don’t have any reason to believe they are cause and effect,” but they are always open to future ideas and evidence.
Sometimes you hear the criticism: “Scientists can’t even predict the weather, so why should we listen to them about something even bigger, like the climate?” That’s a little bit like saying: mathematicians can’t even predict how this coin flip will turn out, so why should we listen to their predictions of how a million coin flips will turn out? Climate is about averages and overall trends, which are easier to predict than a particular occurrence. Also, the predictions are getting better and better over time. Remember: science never claims to have the exactly correct answer—that’s just a misconception—and an inappropriate demand—by the public and the media. But the self-critical nature of science means that it tends improve (which, incidentally, is why it’s so disturbing that the predictions of climate change have gotten more dire as time has gone on). But it never is done.
Sometimes you hear the criticism “There is no consensus among scientists about human-caused climate change.” News flash! There’s no consensus among scientists about anything! The inherent uncertainty of science means there will almost always be dissent on any scientific issue.
Pick the most well-known, well-established scientific law you can think of. The Law of Gravity, right? Guess what? There’s no consensus on it! We’ve got a satellite up there right now, Gravity Probe B, testing our current understanding of gravity. And you know what it’s looking for? I’ll give you a hint: remember the phenomenon of “confirmation bias?” Scientists are really careful to avoid that, so the probe isn’t so much looking for evidence to confirm our theory. It’s looking for evidence to contradict it! We’re actively trying to disprove perhaps the most widely accepted and beloved of all scientific theories. Why? Because we love it so much, we want to make it stronger. Looking really hard and conscientiously for contradictory evidence and failing to find it does more to increase our confidence than looking for supporting evidence and finding it.
Science is never certain. You know that classic Mentos and Diet Coke reaction? [show video] You want to know the scientific explanation for it? Here it is: no one knows!!! There’s lots of conjecture—it’s quite the hot topic in the chemistry education community. So you can find explanations, but the uncertainties associated with them are going to be very large. Why have we not studied it further to reduce the uncertainties? Because it’s not worth it. But the more important the issue is, the more research goes into it, the smaller the uncertainties become. But if you’re waiting for there to be no dissent at all, then you’ll wait forever, no matter what the scientific issue.
It’s sort of like with this whole climate debate, it’s easy to find websites giving all sorts of reasons to believe what I already believe. But that doesn’t increase my confidence. I want my argument to be rock solid, so I go looking for websites that contradict what I believe. And when I can’t find much that’s credible, that increases my confidence in my views. It’s like testing to see if something is watertight. You look for the leaks, and if you can’t find any, then your confidence increases. It feels great to have everyone tell you you’re right, but it’s a deceptive, complacent game. The way to really get confident is to go poking at the other side, saying “what’s your response to this? How would you contradict this?” which I’ve done quite a bit with my grid argument about global climate change, which is why there are so many bloody minutes of me talking on video as a result: the experience left me bruised and battered, but it left my argument that much stronger. In fact, as I film this, an early version of “How It All Ends” leaked onto Digg.com a couple days ago, and I am heartened the every single criticism I read there already is countered in my video scripts. It didn’t get that way by me talking to people who agree with me. In fact I handed my scripts to one of the best critical thinkers I know and said: “Please find the holes in this argument.” That’s why science is the most self-critical endeavor in the history of humanity—it knows that that is the most effective way to get better.
How about this objection: “Climate models are just models, just predictions about the future, which we can’t test until the future actually happens. We don’t know what’s really going to happen. So they’re just conjecture, and therefore useless.” My response is—ever ridden on a modern airliner? Cuz they’re all designed on the computer, modeled on the computer, tested in the computer model, then physical models, and finally computer models again, which is where the pilots learn to fly them. When the Boeing 777 was first flown, all the technicians, managers, and you can bet test pilots, were extremely confident that it would fly. Why? Because we’ve learned how to make good computer models, by tweaking them until their output matches what we see in the physical world.
Climate models on the computer, for instance, are calibrated with the observed climate of the past. If we feed a model the conditions in 1950 and it churns out predictions for the period 1950 to 2000 that closely match what actually happened, then that gives us confidence in the predictions it makes when we put in the conditions for 2000 and ask it about 2030.
It’s been proposed that the greatest knowledge is to know that you do not know. So when you hear pronouncements about how global climate change is bunk, or that we’re not the ones doing it, keep that in mind. Now that you understand a bit about the uncertain and tentative nature of science, ask yourself: how credible are pronouncements about a scientific issue, when they’re made with such certainty?
Along those lines, I was struck by how many people in the comments to my “Most Terrifying Video” made absolute statements of truth about the world. A ton of people flat out said “Humans are not causing global warming.” Other comments I got included:
“Humans are too small to have an effect on the climate.”
“Global warming is a ploy for the elites to grow the government and take away your freedoms.”
“It is true that the climate is changing, but there’s a lot of debate about whether we’re the ones causing it.”
“Taking action may make things worse.”
“Climate changes all the time.”
“We’re coming out of a cold cycle, so this is natural.”
“100 years of data is not enough to know 1000s of years of the past climate.”
“Personally, I don’t think global warming is as definitely man-caused as popular media make it out to be.”
“Personally, I think?!!??”
We’re talking the most complex science in the history of humankind. Chaos theory was discovered studying weather systems. “Personally, I think?!!??” Who the heck are you say what the physical truth is?
But then, I admit, I fell into a similar mistake of being absolute, claiming in that video that “the only choice” is column A. Who the heck are we to think we’ve got a lock on truth? Have you ever been completely sure of something, and then turned out to be wrong? Shouldn’t that temper our confidence the next time we feel that way? It should give you pause when the trained person is less certain of themselves than the untrained person. I was certainly humbled by the unexpected explosion in my classroom that I describe in the video “I Hope I’m Wrong.” I guess the bottom line lesson is here that we will probably do better for ourselves and for the whole with some humility.
Look, I don’t have the answers. And neither, probably, do you. But we, as a people, as a species, can probably come up with something that’s decent. Will it be right? Will it work? We can’t know for sure. Will it be better than nothing? Probably.
I forget where I read it—maybe it was even a bumper sticker LOL—but I recently came across a line that I think pretty well sums up the lesson in humility that scientific thinking teaches us. And I suspect it may help us make some headway in this whole discussion of what to do about climate change. It’s just this:
“Don’t believe everything you think.”
Comments
Youtube Comments: Part 1
Youtube Comments: Part 2
Youtube Comments: Part 3