Is Scientific Skepticism Irrational?

I am revisiting David Stove‘s Popper and Beyond: Four Modern Irrationalists. There is something “off” about it, which is captured in a review at Amazon:

Stove’s primary target was the idea that there might be a problem about inductive inferences, one dating to Hume who was the first to notice it and try to solve it. His secondary target was Popper, whose solution to Hume’s problem was to develop a deductivist account of scientific rationality (critical rationalism, as an alternative to logical empiricism)–naively attempting to change the philosophy of science to address a problem which, in Stove’s opinion, doesn’t exist. His tertiary targets were “historicist” philosophers of science such as Kuhn and Feyerabend….

One does not learn the actual positions of any of these folks from Stove’s book, unfortunately, much less any of their actual errors or excesses. Stove’s own position seems to be a kind of “naive realism” about scientific change and progress: almost as if there were no questions worth asking on the subject. I’ve encountered secondary literature on Kuhn and Feyerabend before that utterly failed to understand them, but Stove doesn’t even make the attempt.

Here is Stove’s argument, reduced to its essence:

  • Scientific knowledge has progressed.
  • Some philosophers of science hold the view that scientific knowledge is provisional because what is believed to be true can always be falsified by new knowledge.
  • Saying that scientific knowledge is provisional is tantamount to saying that scientific knowledge has not progressed.
  • The philosophers of science who hold that scientific knowledge is provisional are therefore irrational because they effectively deny that scientific knowledge has progressed.

Stove assumes that which he seeks to prove. His reasoning is therefore circular. His book is a waste of ink, paper, and pixels (depending on the format in which it is published).

Stove, nevertheless (unwittingly) poses a question that demands an answer: If scientific knowledge is provisional (as it always is), is it possible to say that scientific knowledge has progressed; that is, more is known about the universe and its contents than was known in the past?

The provisional answer is “yes”. Human knowledge of the universe progresses, in general, but it is never certain knowledge and some of it is false knowledge (error). Witness the not-so-settled science of cosmology, which has been in flux for eons.

There is broad but not universal agreement that the universe (or at least the part of the universe which is observable to human beings) originated in a Big Bang. Expansion followed. But the rate of expansion of the universe and the cause(s) of that expansion remain beyond the ken of science. The knowledge that the universe is expanding — and expanding at an ever-increasing rate — is an advance on prior knowledge (or belief), which held that the universe is contracting or that it is neither contracting nor expanding. But the knowledge of an accelerating expansion must be provisional because new observations may yield a different description of the universe.

That example brings us to the essential dichotomy of science: observation vs. explanation. What is observed is observed with varying degrees of certainty. The variations depend on the limitations of our instruments, sensory organs, and brains (which may be conditioned to misperceive some phenomena). Where things often go awry is in explaining that which is perceived, especially if the perception is wrong.

A classic case of misperception is the once-dominant belief that the Sun circles Earth. It’s easy to see how that misperception arose. But having arisen, it led to erroneous explanations. One erroneous explanation was that the Sun is embedded in a “sky dome” that surrounds Earth at some distance and rotates around it.

A current case of misperception is the deliberately inculcated belief that the general rise in observed temperatures on Earth from the late 1970s to late 1990s is due almost entirely to an increase in the atmospheric concentration of CO2 caused by human activity. The dominance of that theory — which objective observers know to be incomplete and unsubstantiated — may well lead to the impoverishment of vast numbers of persons in North America and Western Europe because the leaders of those countries seem to be virtue-signaling CO2-reduction race to limit and eventually ban the use of fossil fuels.

If David Stove were still with us, he would probably say what I have just said about the current misperception, given his penchant for iconoclasm. But where would that leave his “naive realism” about scientific progress? He would have to reject it. In fact, knowing (as he undoubtedly did) of the erroneous belief in and explanation a geocentric universe, he should have rejected his “naive realism” about scientific progress before taking Popper et al. to task for their skepticism about the validity of new scientific knowledge.

Yes, scientific knowledge accrues. It accrues because knowledge (to a scientist) is ineluctably incomplete; there is always a deeper or more detailed explanation of phenomena to be found. The search for the deeper or more detailed explanation usually turns up new facts (or surmises) about physical existence.

But scientific knowledge actually accrues only when new “knowledge” is treated as provisional and tested rigorously. Even then, it may still prove to be wrong. That which isn’t disproved (falsified) adds to the store of (provisional) scientific knowledge. But, as Stove fails to acknowledge, much old “knowledge” hasn’t survived, and some current “knowledge” shouldn’t survive (e.g., the CO2-driven theory of “climate change”).

Here is the argument that Stove should have made:

  • Scientific knowledge has progressed on many fronts, but not to the exclusion of error.
  • Some philosophers of science hold the view that scientific knowledge is provisional because what is believed to be true can always be falsified by new knowledge.
  • Given the track record of science, those philosophers are correct to say that scientific knowledge is provisional.
  • It is possible for scientific knowledge to accrue, and to be provisional at the same time.

Think of all of the ink, paper, and pixels that could been saved if Stove had thought more carefully about science and issued a PowerPoint slide instead of a book.


Related post: Deduction, Induction, and Knowledge

A (Long) Footnote about Science

In “Deduction, Induction, and Knowledge” I make a case that knowledge (as opposed to belief) can only be inductive, that is, limited to specific facts about particular phenomena. It’s true that a hypothesis or theory about a general pattern of relationships (e.g., the general theory of relativity) can be useful, and even necessary. As I say at the end of “Deduction…”, the fact that a general theory can’t be proven

doesn’t — and shouldn’t — stand in the way of acting as if we possess general knowledge. We must act as if we possess general knowledge. To do otherwise would result in stasis, or analysis-paralysis.

Which doesn’t mean that a general theory should be accepted just because it seems plausible. Some general theories — such as global climate models (or GCMs) are easily falsified. They persist only because pseudo-scientists and true believers refuse to abandon them. (There is no such thing as “settled science”.)

Neil Lock, writing at Watts Up With That?, offers this perspective on inductive vs. deductive thinking:

Bottom up thinking is like the way we build a house. Starting from the ground, we work upwards, using what we’ve done already as support for what we’re working on at the moment. Top down thinking, on the other hand, starts out from an idea that is a given. It then works downwards, seeking evidence for the idea, or to add detail to it, or to put it into practice….

The bottom up thinker seeks to build, using his senses and his mind, a picture of the reality of which he is a part. He examines, critically, the evidence of his senses. He assembles this evidence into percepts, things he perceives as true. Then he pulls them together and generalizes them into concepts. He uses logic and reason to seek understanding, and he often stops to check that he is still on the right lines. And if he finds he has made an error, he tries to correct it.

The top down thinker, on the other hand, has far less concern for logic or reason, or for correcting errors. He tends to accept new ideas only if they fit his pre-existing beliefs. And so, he finds it hard to go beyond the limitations of what he already knows or believes. [“‘Bottom Up’ versus ‘Top Down’ Thinking — On Just about Everything“, October 22, 2017]

(I urge you to read the whole thing, in which Lock applies the top down-bottom up dichotomy to a broad range of issues.)

Lock overstates the distinction between the two modes of thought. A lot of “bottom up” thinkers derive general hypotheses from their observations about particular events. But — and this is a big “but” — they are also amenable to revising their hypotheses when they encounter facts that contradict them. The best scientists are bottom-up and top-down thinkers whose beliefs are based on bottom-up thinking.

General hypotheses are indispensable guides to “everyday” living. Some of them (e.g., fire burns, gravity causes objects to fall) are such reliable guides that it’s foolish to assume their falsity. Nor does it take much research to learn, for example, that there are areas within a big city where violent crime is rampant. A prudent person — even a “liberal” one — will therefore avoid those areas.

There are also general patterns — now politically incorrect to mention — with respect to differences in physical, psychological, and intellectual traits and abilities between men and women and among races. (See this, this, and this, for example.) These patterns explain disparities in achievement, but they are ignored by true believers who would wish away the underlying causes and penalize those who are more able (in a relevant dimension) for the sake of ersatz equality. The point is that a good many people — perhaps most people — studiously ignore facts of some kind in order to preserve their cherished beliefs about themselves and the world around them.

Which brings me back to science and scientists. Scientists, for the most part, are human beings with a particular aptitude for pattern-seeking and the manipulation of abstract ideas. They can easily get lost in such pursuits and fail to notice that their abstractions have taken them a long way from reality (e.g., Einstein’s special theory of relativity).

This is certainly the case in physics, where scientists admit that the standard model of sub-atomic physics “proves” that the universe shouldn’t exist. (See Andrew Griffin, “The Universe Shouldn’t Exist, Scientists Say after Finding Bizarre Behaviour of Anti-Matter“, The Independent, October 23, 2017.) It is most certainly the case in climatology, where many pseudo-scientists have deployed hopelessly flawed models in the service of policies that would unnecessarily cripple the economy of the United States.

As I say here,

scientists are human and fallible. It is in the best tradition of science to distrust their claims and to dismiss their non-scientific utterances.

Non-scientific utterances are not only those which have nothing to do with a scientist’s field of specialization, but also include those that are based on theories which derive from preconceptions more than facts. It is scientific to admit lack of certainty. It is unscientific — anti-scientific, really — to proclaim certainty about something that is so little understood the origin of the universe or Earth’s climate.


Related posts:
Hemibel Thinking
The Limits of Science
The Thing about Science
Science in Politics, Politics in Science
Global Warming and the Liberal Agenda
Debunking “Scientific Objectivity”
Pseudo-Science in the Service of Political Correctness
Science’s Anti-Scientific Bent
“Warmism”: The Myth of Anthropogenic Global Warming
Modeling Is Not Science
Demystifying Science
Analysis for Government Decision-Making: Hemi-Science, Hemi-Demi-Science, and Sophistry
Pinker Commits Scientism
AGW: The Death Knell
The Limits of Science (II)
The Pretence of Knowledge
“The Science Is Settled”
The Limits of Science, Illustrated by Scientists
Rationalism, Empiricism, and Scientific Knowledge
AGW in Austin?
The “Marketplace” of Ideas
Revisiting the “Marketplace” of Ideas
The Technocratic Illusion
The Precautionary Principle and Pascal’s Wager
AGW in Austin? (II)
Is Science Self-Correcting?
“Science” vs. Science: The Case of Evolution, Race, and Intelligence
Modeling Revisited
Bayesian Irrationality
Mettenheim on Einstein’s Relativity
The Fragility of Knowledge
Global-Warming Hype
Pattern-Seeking
Hurricane Hysteria
Deduction, Induction, and Knowledge
Much Ado about the Unknown and Unkownable

Deduction, Induction, and Knowledge

Syllogism:

All Greek males are bald.

Herodotus is a Greek male.

Therefore, Herodotus is bald.

The conclusion is false because Herodotus isn’t bald, at least not as he is portrayed.

Moreover, the conclusion depends on a premise — all Greeks are bald — which can’t be known with certainty. The disproof of the premise by a single observation exemplifies the HumeanPopperian view of the scientific method. A scientific proposition is one that can be falsified  — contradicted by observed facts. If a proposition isn’t amenable to falsification, it is non-scientific.

In the Humean-Popperian view, a general statement such as “all Greek males are bald” can never be proven. (The next Greek male to come into view may have a full head of hair.) In this view, knowledge consists only of the accretion of discrete facts. General statements are merely provisional inferences based on what has been observed, and cannot be taken as definitive statements about what has not been observed.

Is there a way to prove a general statement about a class of things by showing that there is something about such things which necessitates the truth of a general statement about them? That approach begs the question. The “something about such things” can be discovered only by observation of a finite number of such things. The unobserved things are still lurking out of view, and any of them might not possess the “something” that is characteristic of the observed things.

All general statements about things, their characteristics, and their relationshships are therefore provisional. This inescapable truth has been dressed up in the guise of inductive probability, which is a fancy way of saying the same thing.

Not all is lost, however. If it weren’t for provisional knowledge about such things as heat and gravity, many more human beings would succumb to the allure of flames and cliffs, and man would never have stood on the Moon. If it weren’t for provisional knowledge about the relationship between matter and energy, nuclear power and nuclear weapons wouldn’t exist. And on and on.

The Humean-Popperian view is properly cautionary, but it doesn’t — and shouldn’t — stand in the way of acting as if we possess general knowledge. We must act as if we possess general knowledge. To do otherwise would result in stasis, or analysis-paralysis.