Discover more from We Are Not Saved
On The Limitations of Science
There are lots of people out there condemning the debauchery of our modern world, and generally with more eloquence than I can muster. Additionally there are prophets, both ancient and modern who have already offered up rousing sermons and trenchant obser
If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:
There are lots of people out there condemning the debauchery of our modern world, and generally with more eloquence than I can muster. Additionally there are prophets, both ancient and modern who have already offered up rousing sermons and trenchant observations (one of which I took as the theme of this blog) and I would urge you to study the writings of those prophets before reading anything I write. So, if there’s better stuff out there why do I bother to blog? I believe there is a gap in the commentary. A hole in the discourse that I can fill. It doesn’t need to be filled. What I write is not critical to anyone’s salvation. I am not uncovering any lost principles of Christ’s gospel, nor am I speaking in a more timely manner than what you hear at the semiannual General Conferences. If that’s so, what niche do I fill? What unique insights do I provide?
If you read my very first post, then you’ll remember that I already touched on this. This blog will specifically focus on comparing the LDS Religion to the Religion of Progress and examining how the Religion of Progress has failed. The sacrament of the Religion of Progress is science. And it is appropriate that it be so. I myself am a believer in science. But like all sacraments, the sacrament of science can be partaken of unworthily. It can be misunderstood, and distorted. Just as partaking of the actual sacrament every week doesn’t immediately absolve you of all your sins if you’re not also actively exercising faith, repenting of those sins and seeking forgiveness; partaking in the sacrament of science doesn’t immediately make what you do and what you believe scientific, no matter how much you proclaim your love for it. Science has serious limitations, even if one is doing everything right, which most of the time they’re not. And many of the failures of the Religion of Progress comes when it ignores those limitations (or in the case of the last post, trades science for emotion.) Consequently, this post is all about examining those limitations.
Let’s start by examining the limits of science even if everything is done correctly. To begin with it’s really hard to do it correctly, and 90% of the time what passes for quality science are efforts which leave out a lot of the rigor necessary for truly conclusive results. This was not always the case at the beginning of the scientific revolution there was a lot low hanging fruit. Scientific results of surpassing clarity and rigor that could be obtained with only moderate effort (the gentleman scientist working nearly alone was a fixture of the time.) All that low-hanging fruit is gone, but people still expect science to come up with similarly ironclad results even though the window during which that was possible is long past. Also most of the really solid science involves physics, and the farther you get away from that, the less amenable things are to experimentation in general because there are too many variables.
Thus you’re left in a situation where if you want to do solid, incontrovertible science your best bet is to do more physics, and that’s going to cost billions of dollars, or you can use pieces of the scientific method and take a stab at the questions which remain after all the low-hanging fruit has been picked. I say pieces of the scientific method because, for example, there are all manner of subjects which can’t be subjected to an experiment with a control. This is a limitation in many fields, but one of the best examples is economics, particularly macroeconomics. You can’t create a copy of the world and have one world where the global economy stays on the gold standard and the control, a world where everyone moves to floating currency. You will still have economist who will tell you that one is better than the other, but this is based off bits of data they’ve gathered from a very messy environment. Not any kind of conclusive, replicable experiment.
Related the problem of creating a control group is the difficulty of isolating the variable you hope to study. Even if we were somehow able to create two versions of Earth, and create a control, how do we know that all the differences between 2016 gold-standard Earth and 2016 floating-currency Earth are due to the different currency systems and not other random fluctuations? Obviously this is already a fairly ridiculous example, but it illustrates the impossible hurdles necessary to even approach true experimentation on something like the economy.
Now you should not assume from this that I’m anti-science, far from it. I have a deep respect for science. And I think that, if anything, the world needs more science not less, but as part of that, we need, particularly if we’re piling up more science, to recognize the limitations of science, especially as it’s actually put into practice. Science isn’t conducted by perfectly objective robots, it’s conducted by scientists who have careers to think of, biases which blind them and limitations of time and money to contend with. All of which takes us to the next way that science can go wrong.
When I say the next way, there are literally hundreds of ways that scientific efforts can go wrong, but rather than try to focus on all of them we’re just going to look at something that has been in the news a lot lately, the replication crisis.
What’s interesting about the replication crisis is that it happened even in cases where it truly appeared that people were doing everything correctly. Trained scientists were conducting ground-breaking experiments, designed according to the best thinking in their field, the results were passed through a process of peer-review and then the results were published in a respected journal. Obviously this is not to say that there weren’t papers published where everything was not being done correctly, even some examples of outright fraud, but even if we exclude those there were still a lot of results which got published which later turned out to be impossible to reproduce. The biggest contributor to this appears to have been publication bias, or what is sometimes called the file-drawer effect because people only submit positive, exciting results and the rest get put in the file-drawer with all of the other experiments that didn’t show anything. This is a problem not only with the people doing the experiments but with the publications themselves, which are far more likely to publish positive results (or to be technical, statistically significant results) than a paper which didn’t have any results (or a null result). And as you’ve probably heard, for most scientists it’s publish or perish. Another factor which almost certainly contributed to the crisis..
You may think that a positive result is a positive result regardless of whether there were 100 other, negative results which got put in the file cabinet. The problem is that it’s not. If you take 100 coins and flip each of them 7 times you’ve got better than even odds that one of them will come up 7 heads in a row. You might then decide that that coin is unfair, and publish a paper, “On the Unfairness of the 1947 Nickel”, but in reality you just started with a big sample size. Doing 100 experiments works very similarly. (For a really in depth discussion including p-values and lots of statistics go here.) The problem of course becomes that people reading or citing your paper don’t know that you have 99 failed experiments which never saw the light of day they only know about the one successful experiment that actually got published.
Thus far I haven’t mentioned how often a study fails to be replicated, and you may think that it’s no big deal. A few here and there, but nothing to worry about. Well as it turns out in general less than half of studies can be reproduced and sometimes less than 15%! This would mean that six out of every seven studies put forth conclusions which later turned out to be untrue.
Once again it’s important to recognize that there is a continuum of scientific results. There’s not a 50% chance that the theory of gravity is wrong, or that protons don’t exist. But when it comes to the softer sciences (and they’re labeled that way for a reason) there is a better than even chance that their conclusions will turn out to be untrue.
Of course when the average person talks about scientific discoveries, ignoring for the moment whether the results can be reproduced, they’re generally not talking about what the scientist actually found. To a first approximation no one reads the actual scientific paper, and probably only 1 in 10,000 people even read the abstract. If you hear about a scientific result you’re hearing about it through the media, which further undermines the utility of science by distorting results in an effort to make them appear more interesting. In short when people think of science they think of gravity, but what they're actually getting is a Buzzfeed article written based on a press release from a conversation with a scientist who shelves most of his work, is desperate for tenure, describing a conclusion that is more than likely irreproducible. That’s like five layers of spin on top of a result that’s most likely false!
If the kind of “science” I’m talking about were framed as an amusing hobby and an article about bacon prolonging life was treated in the same fashion as a movie review then it wouldn’t be that big of a deal, but for many people science has taken the place of religion. And more than just religion, it has taken the place of deep thinking about the fundamental questions of life in general. People have replaced virtue with a sort of sloppy rationality which cloaks itself in science and is therefore considered progressive, but is really just the idea of doing whatever makes you feel good cloaked in a bunch of pseudo scientific babble. And decisions are being made which can cost people their lives.
As an example of this, I just finished the book Dreamland by Sam Quinones. It’s an in depth look at the opiate epidemic in America, and a stunning indictment of what passes for science these days. You’ve probably heard about the opiate epidemic, if not follow the link. The effects of the epidemic are so bad that as to be baffling and a whole host of factors combined to make the problem so terrible, but the misuse of science was one of the bigger factors, possibly the biggest. It’s not possible to go into a complete description of what happened (I highly recommend the book) but in essence using a combination of poor science and a morality devoid of any underpinning in religion or tradition, doctors decided that people could essentially have unlimited opiates, the best known of which is oxycontin. Exactly what I mean by doing whatever makes you feel good cloaked in pseudo scientific babble.
The first part, the misuse of science, hinged on placing far too much weight on a one paragraph letter published in the New England Journal of Medicine in 1980 which claimed that opiates only ended up causing addiction in 1% of people. Getting past the fact that the author never intended it to be used in the way it was, to base decades of pain management on one paragraph is staggeringly irresponsible. Even more irresponsible, when the pharmaceutical companies got around to trying to confirm the result they found the it didn’t hold up (to no one’s surprise) and they ended up burying and twisting the results they did get. The number of people that died of accidental overdoses directly or indirectly from this misuse of science is easily six figures, possibly seven, particularly since people are still dying. Of course in addition to the misuse of science there was the over reliance on science. I assume that on some level the pharmaceutical companies knew that they were not being scientific, but countless doctors, who were either naive or blinded by the gifts provided by the pharmaceutical company chose to at least to pretend that they were doing what they were doing because science backed them up.
I mentioned that one of the other factors was a morality devoid of any underpinning in religion or tradition. I’m not going to say that any religion specifically forbids overprescription of opiates, but most of them have some broad caution about drugs in general. And even if you want to set religion aside there is a strong traditional distaste for opium. And here is where the limits of science are most stark.
Frequently, people use science to declare any belief or practice or tradition or religion which is insufficiently scientific (which of course includes all religions, most traditions, and a majority of practices and beliefs over a few decades old) as nothing more than baseless superstitions. And while it was not labeled as such this is precisely what happened with opiates. All religions I’m aware of recognize that a certain amount of suffering is part of existence, but in 1980, doctors more or less decided it wasn’t. Sure they couched in the language of science with lots of caveats, but this is precisely the problem. The science turned out to be wrong and the caveats turned out to be insufficient barriers to abuse and somewhere north of 100,000 people died.
As I have repeatedly said, I’m not anti-science, but science without tradition, without morality, and without religion is prone to huge abuses. This blog will attempt to unite religion and science, but in doing so, religion is always going to hold primacy over science. And it’s not even necessarily because religion is backed by divine infallibility. Forget about that. Set that aside. While, I certainly believe that that’s the case, in these circumstances it doesn’t matter. The problem with science is that it hasn’t been around very long, and it assumes a sterile, rational world which bears no resemblance to the world we actually live in. Setting aside whether God exists, religion and tradition has been tested in the crucible of history. And have provided insights, particularly in the realm of morality that people ignore at their peril. Which will be the subject of my next blog post.