All the way back in 2011I wrote a post called "What is Science"? Here's the link: Sigma 5: What is Science?. In that post I took a hard look at the same question addressed in a new and very interesting book. It has a weird title: The Knowledge Machine. In it the author, Michael Strevens addresses the same question I took a hack at in that post.
I recently did a post that amounted to a book review of a new book by Bill Gates. The name of his book is How to Avoid a Climate Disaster, and the post can be found here: Sigma 5: How to Avoid a Climate Disaster. In that post I said that the Gates book was hard to summarize because Gates covered so much ground. That's not a problem with my present effort, taking a look at what Strevens has to say. His book summarizes nicely. But let me start with a few words about my 2011 post.
In it I spent some time looking at what Science is and what it isn't. And it definitely isn't the usual step 1, step 2 business that is the way the Scientific Method is usually presented. After delving into the subject at some length, I observed that the Scientific Method is actually a popularity contest.
We don't associate popularity contests with serious outcomes. People often see them as an opportunity to have a little harmless fun. A few years ago the British Government held a contests to name a research vessel. They were hoping to end up with something that had a little class and distinction. But the runaway winner was "Boaty McBoat Face", a result that shouldn't have come as a complete surprise.
Science and the Scientific Method are a serious business. But that doesn't make the basic method by which Science advances our understanding of the natural world any less of a popularity contest. Certain scientific theories become broadly accepted while others go out of fashion. One theory gains in popularity at the expense of the others. See -- popularity contest.
What makes the popularity contest work for Science is the scoring system. When evaluating which theories Scientists should get behind and which theories now seem lacking, Scientists ask one seemingly simple question: Which theory is more likely to be right? That's a more complicated question than it initially appears.
Strevens comes at the same question I did, why science works, and arrives at the same answer. He uses different terminology and he dives into the subject in a much more thorough way than I did in my blog post. Frankly, he knows much more about the subject than I do.
I found what he had to say quite illuminating. For instance, he asks a question that I hadn't even thought of. The Scientific Revolution started in about 1600 in Europe. Why did it start when and where it did?
The reason this book is so easily summarized is that the main points can be concisely listed. He even stops and does this for me so that I don't have to hunt around to find them. And that, for the most part, is what I am going to do in this post. But from a reader's perspective, that leaves out the parts that make the book so fascinating and such a good read.
Strevens dives into a lot of real historical events and a few fictional ones. I am going to leave almost all of that out of the post. But he is a good writer who has picked interesting and illuminating events and told their story well. So, while this post will provide you with the key takeaways, not following up by reading the book will cause you to miss a wonderful experience. Oh, well.
A subject like this needs a framing device. It is far easier for readers to follow his logic if there is some way to keep those readers oriented. He uses a popular framing device, a competition between two opposing individuals or factions. But he uses it well. That way he can shorthand a position by naming the champion of that faction. Enough! Let's get to it.
Why Science works is a mystery to most, but that doesn't mean people haven't taken a stab at the question before. For his first conflict he selects two answers that have been proposed and assigns a champion to stand for each. First, the obvious Choice. Karl Popper literally invented one of the answers. He then devoted the rest of his life to championing it.
Popper's key insight is that, while it is never really possible to prove a theory is true, it is always possible to prove that it is false. Popper opines that the way Science advances is by shooting theories down. Then, paraphrasing Sherlock Holmes, whatever you are left with must be the truth no matter how unlikely or unreasonable it is.
Choosing a position and a champion to put into opposition to Popper is harder. He chooses Thomas Kuhn, the author of the hugely influential book The Structure of Scientific Revolutions. When it came out in 1962, not only did Scientists sit up and take notice, but the general public did too. This is the book that introduced word "paradigm" and the phrase "paradigm shift" to Scientist and non-Scientist alike.
I read the book back in the day. The first part, where Kuhn describes what paradigms and paradigm shifts are, made complete sense to me. My objection was with the second half of the book. My recollection is that in it Kuhn said, in effect, "all paradigms are created equal". One paradigm gains favor over another for essentially cultural reasons. Scientifically, there is no reason to favor one over another.
Strevens has a different take on the second half of Kuhn's book. This is based in part on Kuhn's later writings and in part on a different interpretation of what Kuhn said in the second half of Structure. It turns out that for our purposes it doesn't matter who got it right.
What is important is that, if we accept Strevers' position on Kuhn and his thinking, then Kuhn makes a good choice of a person and a faction to put in opposition to Popper. So, for the purposes of this post I am just going to go along with the Strevens interpretation and say no more.
According to Kuhn, Scientists become so wedded to a paradigm that they literally can't let go of it even in the face of substantial evidence against it. In other words, they have a very unscientific relationship with paradigms. This sounds bad, but to Kuhn's way of thinking, it turns out to be a good thing.
Why? Because it makes good things happen. It causes these "true believer" scientists to push themselves and the the paradigm they believe in as hard as possible. This forces them to work hard, creatively, and tirelessly to extract the last bit of juice from the paradigm. This hard work and creativity also causes them discover all if its flaws, no matter how small.
The result is a vast trove of data that can be mined by a new generation of scientists. This new generation is not yet wedded to any particular paradigm. Thus, they are free to develop new and better paradigms. Science advances because each group plays its assigned role well.
The old generation, as a result of its complete devotion to a paradigm, puts in the hard and boring work necessary to generate lots of new data. All this new data gives the new generation the ammunition it needs to create "new and improved" paradigms.
Strevers calls this conflict between Popper's and Kuhn's approach "The Great Method Debate". By what method does Science succeed? Spoiler alert, it is not by way of the Popper method. Nor is it by way of the Kuhn way. It is a third way, which I call a popularity contest and which Strevers calls "The Knowledge Machine". Hence, the title of the book.
By this time Strevers has spent some time poking holes in the Popper Method. He has also poked holes in the Kuhn Method. But wait. There's more. He now spends time showing that it's not just a scientist here or there. It's not just a theory here or there. By objective measures, he argues, Science and Scientists never use the Scientific Method. In fact, he goes so far as to name an entire chapter "The Essential Subjectivity of Science".
He cites a study that followed a well respected bench scientist and his assistants around for a year. The study determined that (1) they don't do things the Popper way, (2) they don't do things the Kuhn way, and (3) they often abandon official scientific technique entirely. Harsh.
But how about this for an "out"? Maybe over a range of experiments over a period of time maybe all the bad behavior washes our and we end up being left with the good and true stuff. He demolishes that hope too. His point is that whatever makes Science work must be extremely powerful to overcome all of these obstacles.
His answer to the question of why Science works in spite of all this is what he calls "the iron rule". It is an "iron" rule because, unlike what has come before, Scientists don't violate it. They may do what they shouldn't do and not do what they should do, but they don't violate the iron rule.
So what is it? It's the scoring system in the popularity contest. When acting as a judge in a Science popularity contest the Scientist must :
- Strive to settle all arguments by empirical testing.
- To conduct an empirical test to decide between a pair of hypotheses, perform an experiment or measurement, one of the possible outcomes can be explained by one hypothesis . . . but not by the other.
Notice that no reference is made to who is doing the empirical testing, or what any religion or philosophy has to say on the subject. Instead of "pistols at dawn", it's "what measurement or experiment will convince all the interested parties that at least one of the hypotheses is wrong"? The conflict is, in effect, externalized and handed off to a neutral arbiter that all can respect.
The natural world doesn't have a dog in the fight. It just goes about doing what it does. If that's convenient, fine. If that's inconvenient, also fine. A well designed and well executed experiment sends the natural world a query. The results of the experiment are the answer. And its an answer that everyone can respect.
Disagreements are everywhere. Unfortunately, outside of the world of Science these disagreements too often generate ad hominem attacks. "X is a bad person so anything X says or does can be ignored". In Science, X may truly be a bad person. But that's not important. What's important is the degree to which he performed an experiment in a way that yields valuable and reliable data.
If it only matters whether X is a good or bad experimenter then the question of whether X is a good or bad person becomes just a distraction. Why would someone opt for an distraction rather than making a serious attempt to find out what the correct answer is?
When I see an ad hominem attack, and particularly when I see a group whose first impulse is to go with an ad hominem attack, that puts me on my guard Why go with an ad hominem attack at all? And especially why start off with an ad hominem attack? I am inevitably drawn to the conclusion that they are going the ad hominem route because they are sure that they are in the wrong.
Back to the book, which doesn't mention ad hominem attacks. But it does note that the iron rule is "an etiquette for argument, an agreement on how to disagree". Scientists get just as committed to their positions as others. But, because they try to hue to the iron rule, Scientists are much better at being able to "disagree without being disagreeable".
That is why ad hominem attacks are relatively rare in scientific disputes and common elsewhere. Politics, religion, even philosophy are diminished by the absence of the iron rule or some other method of discouraging ad hominem attacks.
Having established his key idea, the iron rule, Strevers switches to a historical approach. Where did it come from and how did it develop. He chooses to start with Sir Francis Bacon. Bacon did not come up with it, but he set the ball in motion.
Bacon is a name I was familiar with without quite knowing what his contribution to Science was. The author did a nice job of remedying that deficiency. Bacon is considered by a lot of people to be the first "Scientist", the first person to champion a process that resembles how Science is now done. He laid his thinking out in a book he wrote called The New Organon. (The tile is an homage to a book by Aristotle called Organon).
For explanatory purposes Strevens asks "how does heat flow work". he then introduces two fictional scientists, Romeo Montague and Juliet Capulet. They have different approaches. He then works through how each would proceed to apply Bacon's method.
Here's Bacon's method:
- Assemble all positive instances (supporting evidence) for the theory you are studying.
- Assemble all of the negative instances (disconfirming evidence).
- Assemble a list of all the things that might influence whatever you are studying. In the case of heat metals warm more slowly than air. So the type of material involved matters.
- Now, what can possibly explain all of the assembled facts? Certainly, whatever theories have been advanced qualify. So apply each theory to everything you have collected in steps 1-3 and see where you are at.
With luck, there will be only one theory left standing at the end. In the process of applying Bacon's Method the thinking of both Romeo and Juliet should draw closer, should converge. Even if their thinking doesn't converge completely, their thinking should be much closer at the end of the process than it was at the beginning. Strevens calls this "Baconian Convergence".
When using the methods traditionally championed by philosophers and religious people there is no reason for the thinking of these two people, or any two people, to converge. We have a modern term for this phenomenon: the "Zombie Argument".
Bacon's Method and Baconian Convergence were a big improvement on the previous state of things. Unfortunately, while Bacon's Method often works, it doesn't always work. No explanation may work. Several explanations may work. And, as a result, Baconian Convergence may not happen.
What Science has retained from Bacon's method is that requirement that all evidence must be examined. Only a theory which explains all of it is truly successful. The author then moves on to the Michaelson-Morley experiment. This is famous for it's negative result. Something was supposed to happen. Instead, nothing happened.
What was important about this is the difference between Einstein's Special Theory of Relativity, the theory that explained the Michaelson-Morley result, and the theory that failed, Newtonian mechanics. That difference was 0.00001 inch. That is a truly tiny difference. The obvious question is "who cares"?
Strevens' answer is: Scientists care. The apparatus Michaelson and Morley used was sensitive enough to accurately measure an amount that small. Einstein's Special Theory of Relativity also solved the "progression of Mercury" problem. I'm not going to go into what the problem was, except to note that the amount in dispute was another "who cares?" sized amount.
The author goes into all this so that he can he can name the process of chasing after tiny discrepancies "The Tychonic Principle". This principle is that "the secrets of the universe lie in minute structures, in nearly indiscernible details, in patterns that only the most sensitive, fragile, and expensive instruments can detect".
The Tychonic Principle has made headlines within the past few days. The headlines are a result of "the Muon problem". The "Standard Model" of particle physics predicts exactly how magnetic a Muon is supposed to be. A recent experiment found them to be more magnetic than they are supposed to be.
How much more magnetic? 2.5 parts per billion. The tale of how Scientists were able to measure such a tiny discrepancy is truly heroic. Needless to say, the instrument that made that measurement had to be extremely "sensitive, fragile, and expensive" to succeed.
All this may seem obvious. But it wasn't to Aristotle, or to anybody else until Europe in the 1600s. Before then people were capable of making exquisitely accurate measurements. They were capable of developing brilliant and elegant theories. What they were unable to do was to put it all together. They did not develop a method for properly rating which theories were right and which were wrong.
Strevers summarizes what he has so far had to say about the iron principle by saying it consists of:
- A notion of explanatory power on which all scientists agree.
- A distinction between public scientific argument and private scientific reasoning.
- A requirement of objectivity in scientific argument (as opposed to reasoning).
- A requirement that scientific argument appeal only to the outcomes of empirical tests (and not to philosophical coherence, theoretical beauty, and so on).
And I find that I have skipped over the "scientific argument" versus "scientific reasoning" business. This is confusing terminology that is unique to Strevens. But he is making an important distinction.
There is a channel for the formal communication of Scientific Results. Historically, this has been a Scientific Journal like the "Proceedings of the Royal Society", first published in 1695. The channel has expanded somewhat since then to include conference proceedings, video presentations, and the like.
Set against this are more informal communications channels. This may include popular books, TV appearances, musings with colleagues over a couple of beers, etc. There communications channels predate the formal one. Strevers uses the phrase "scientific argument" to apply to the formal and "scientific reasoning" to apply to the latter. This is confusing terminology. I am going to stick with the Strevers terms anyhow.
What's really important is not the terminology. Its that the iron rule applies only to scientific argument, the formal channel. Scientists are free to use arguments based on religion, philosophy, Tarot Cards, whatever, in their informal scientific reasoning utterances.
So why is the iron rule so important and consequential? Without it "there . . . is no timeless, a historical criteria for determining what will count as satisfactory to the understanding". Translation: there is no consistent rule for how to prove "this is how the world works".
And this results in the Kuhnian (in my interpretation of what he said) belief that "all paradigms are created equal". And that blocks the path to progress. For one thing it causes an inability to get to Baconian Convergence. The name for this is "explanatory relativism".
The author then moves on too Descartes, the "I think, therefore, I am" guy. Descartes followed Bacon. His treaties on the subject was modestly called The World. In it he opined that Force can only be transmitted by direct physical contact. His proof was solidly grounded in the philosophy and theology of the day.
When he tackled Gravity he was forced to invent an elaborate set of invisible rotating globes to explain how planetary orbits worked. Strevens does not discuss it, but Newton devoted a section of Principia to proving that no system of rotating globes would work. That was necessary in order to clear the ground for his ideas on the subject.
So was Newton just the next generation of explanatory relativism the way Descartes was. Strevens deals with the question by quoting what Newton wrote in the second edition of his Principia:
I have not . . . been able to deduce . . . the reason for these properties of gravity . . . . It is enough that gravity really exists and acts according to the laws that we have set forth . . .
No philosophy. No theology. Just the facts mam. This was a radical break from the past. Strevens labels this new kind of explanation a "shallow conception" of the concept of explanation. It does not try to answer "why". It is content to just stick with "what".
This move from a deep conception of the world to a shallow one is deeply unnerving. Put another way, "it's just not natural". And this "not natural" behavior of Science is the source of the uneasiness that causes so many people to reject it.
This change was necessary if Science was to be freed from explanatory relativism. As a result, Science is able to hang on to results from millennia ago. It does not need to abandon an old idea just because a new and improved version comes along.
Scientists study "Newtonian Mechanics" to this day even though it has been supplanted by "Quantum Mechanics" for close to a century. Old theories are seen as the foundation stones on which new theories are built, not something that must be completely rejected.
But Science is continuously forced to stick with "a scientific theory postulated some causal principles; what it explains is whatever can logically be derived from these principles" and no more. This plays out every day when it comes to Quantum Mechanics.
Everybody hates Quantum Mechanics including the Scientists who study it and practice it. It is just too weird. It's too unnatural. But it works. If you perform a calculation in the proper way you get the same answer as you would if you did the equivalent experiment. But why Quantum Mechanics works and what's going on down there? No one knows. No one even has a hint of a clue.
Put another way, "the modern scientific standard for explanation is as empirically demanding as it is philosophically lax".
Over the following pages Strevens goes on to attribute four "innovations" to the iron rule over subsequent pages. They are:
- Shallow Explanation. The only thing that matters are the results of experiments and observations. Science is not responsible for any deep explanations.
- The demand for objectivity. To the extent possible all subjective elements are to be removed. This is impossible to achieve in actuality, but it remains the goal.
- The distinction between "scientific argument" (the contents of Journal articles) and "scientific reasoning" (communications in unofficial channels) must be maintained.
- Only empirical testing counts. Show me your experimental results. If you don't have any, shut up.
He then moves on to the question of why Science in general, and the iron rule in particular, appeared when and where they did. He first demolishes several naïve suggestions.
Smart people have been present in all eras of history and in all parts of the world. It was not a lack of intelligence. Good experimenters were also distributed through time and space. He notes that Ancient Greece, the Muslim world, the Chinese, and many others, routinely threw up people capable of doing brilliant experimental work.
And there was no great advance in equipment or technique. He notes that people like Galileo and Brahe were able to able to do high precision work using just the "mark one eyeball" (although he doesn't phrase it that way). I think his argument about the next part is somewhat muddled. Here's my cleaned up version.
What kicked things off was Martin Luthor publishing his "95 theses" in 1517. That threw Europe into intellectual and religious turmoil. The reaction to the reaction was in full swing in 1620 when Bacon published The New Organon. The authorities quickly slapped him down. But their standing as arbiters of truth and morality was already weakening by this point.
Then Newton published the second edition to Principia in 1713. That was the first appearance of the "it is enough" quotation reproduced above. Europe was riven in the hundred years that separated Newton from Bacon and the 200 hundred years that separated Newton from Luther.
At the beginning of this two hundred year period Europe was intellectually and culturally one big happy family. Everybody shared the Roman Catholic religion. And intellectual discourse took place in Latin, the common language of all educated Europeans of the period.
At the end, people were identifying not as Europeans but as English or French or Swedish. They were having bloody wars over which of several religious sects was the "true" one. A single authority that everyone deferred to like the Catholic Church was a thing of the past.
This opened the door for a radical idea like the iron rule. It took this extended period of intellectual and political chaos to enable it to be seriously considered. Such a long period of intellectual, religious, and philosophical chaos had not occurred anywhere else or anywhen else.
But there couldn't be too much chaos. If society broke down completely then the resources that Science demands would also have been absent. At that time Science was the exclusive domain of "Gentlemen of Leisure". Such people existed, and existed in sufficient number throughout this entire period of chaos. They were critical to the necessary maintenance of continuity that permitted the iron rule to be developed and implemented.
Strevens also does not know what to do with the 300 year period from when the second edition of Principia was published and today. I have some thoughts about that too. As he notes, it was entirely possible that the iron rule would die out shortly after Newton published the famous quotation. But it didn't.
I think the reason was that even by the early 1700s the power the iron rule gave to Science was becoming harder and harder to ignore. And the case got even stronger as time passed. Britain, saw tremendous success. Europe and the U.S. saw tremendous success.
By the mid 1800s, a little more than 100 years later, we saw the world divided into the first world (Europe including the U.K, and the U.S.), the third world (most of the world), and a few countries like China somewhere in the middle. But all of the second world countries like China were in obvious decline relative to Europe.
As that three hundred year period has spooled out we have seen country after country adopt at least the Science and Technology component of the European model. Peter the Great, Tsar of Russia at the time of Newton, had tremendous Europe envy. Japan started westernizing in the mid 1800s. Even China eventually got around to it. China is now a Science powerhouse. Even second tier countries like Korea, Vietnam, and many more, have "gone over to the Science side".
In many cases, they haven't completely remodeled their culture and politics. It turns out that you don't need a democratic system of governance to be good at Science and the benefits it produces. In short Science has won.
Or has it? Strevens briefly mentions "Science Wars". But here too I think he gets it wrong. Science is under attack. The attackers are the ones who lost out those hundreds of years ago. The attackers are primarily interested in returning to "traditional" religious and cultural ways.
Science through the workings of the iron rule was able to replace a "deep" conception of truth with a "shallow" one. But it was from the deep part that these groups got their power. So, they argue that we should return to the deep conception.
A deep conception allows faith and philosophy to trump experimental results. At some level, its a power play. These groups want back the power that they had before the ascendency of Science. The fact that its a raw power play makes these groups more rather than less determined.
And they have been joined by a surprising ally, business. At first glance, business should be the ally of Science. But business wants to sell stuff to consumers. They don't want consumers to adopt a Science oriented "show me the data" approach because often the data is missing or worse. Instead they want to be able to use cultural or philosophical arguments in their ads.
We saw this first with the "Tobacco Wars". Cigarettes and other tobacco products are not healthy. They do not make you look more attractive to the opposite sex. They don't make you more popular or stylish or any of that other stuff. What they do is make you sick and maybe kill you. They also make tobacco companies a lot of money.
Tobacco companies found it to their advantage to attack Science in general and tobacco science in particular. They were successful in fighting a rear guard action for decades. Other industries noticed.
The Oil industry (pollution), the Gun industry (they kill people), and other industries, have followed the tobacco industry playbook with considerable success. These companies have a lot of money. They know how to use marketing techniques to get people to believe nonsense. They know how to manipulate elected officials.
On the other hand, Scientists are bad at all of this. They have trained themselves to pay no attention to culture or religion or philosophy. Science and Scientists look extremely powerful on paper. And they are if they can deploy the iron rule. But sadly they look extremely weak when it comes to fighting these real world battles. That's because the first thing their opponents do is pitch the iron rule over the side.
No comments:
Post a Comment