Karl Popper, Falsifiability, and Evolution

I’ve seen a lot of talk from creationists about evolution, in particular, that species evolution (at least macro-evolution) is not “falsifiable”, therefore, not scientific.

The idea of falsifiable is very loaded, and the baggage creationists are smuggling by using terms like these escaped my attention the first few times I noticed the theme.

My first thought when I saw reference to falsifiable was that they were committing an error in terms of how they were discerning between induction and deduction; often, science is an inductive process, where we collect data, then make assumptions based on trends we see in this data.  Of course, we are prone to errors when relying entirely on induction, but science manages this by maintaining a tentative spirit.  If all you ever see are gray squirrels, the statement “all squirrels are gray” is true until you see a black or white one.

The next thought I had about falsifiability was in terms of the null hypothesis.  In general, the null hypothesis can never be shown to be true, and this is a tenet of science.  We only reject or fail to reject the null hypothesis.  For instance, if the null hypothesis is that a pharmaceutical drug has the same effect as placebo, we might arrange a strategy to test the assertion.  If there is a significant difference between the pharmaceutical drug and the placebo (or alternatively, no intervention at all), then we fail to reject the null hypothesis, and we give it a probability value.

But I was still missing something, in terms of the bridge between evolution and falsifiability.  The missing link (pun sort of intended) was in a statement Duane Gish (a prominent young earth creationist) first made in 1981:

There were no human witnesses to the origin of the Universe, the origin of life or the origin of a single living thing.  These were unique, unrepeatable events of the past that cannot be observed in nature or repeated in the laboratory. Thus neither creation nor evolution qualifies as a scientific theory and each is equally religious. As the scientific philosopher Sir Karl Popper has stated, evolution is not a testable scientific theory but a metaphysical research program.

It is interesting Gish invoked Karl Popper in his appeal to authority, considering that Popper had no trouble accepting evolution as fact.  Nevertheless, Gish’s statement made me curious about how he interpreted Popper, so I thought I’d isolate Popper’s position on falsifiability (for which Popper was an advocate, in terms of defining how scientific something is).  These are the primary points Popper makes about structuring science in terms of fasifiability;

1. It is easy to obtain confirmations, or verifications, for nearly every theory-if we look for confirmations.

2. Confirmations should count only if they are the result of risky predictions; that is to say,if, unenlightened by the theory in question, we should have expected an event which was incompatible with the theory-an event which would have refuted the theory.

3. Every “good” scientific theory is a prohibition: it forbids certain things to happen. The more a theory forbids, the better it is.

4. A theory which is not refutable by any conceivable event is nonscientific. Irrefutability is not a virtue of theory (as people often think) but a vice.

5. Every genuine test of a theory is an attempt to falsify it, or to refute it. Testability is falsifiability; but there are degrees of testability; some theories are more testable, more exposed to refutation, than others; they take, as it were, greater risks.

6. Confirming evidence should not count except when it is the result of a genuine test of the theory; and this means that it can be presented as a serious but unsuccessful attempt to falsify the theory. (I now speak in such cases of “corroborating evidence.”)

7. Some genuinely testable theories, when found to be false, are still upheld by their admirers-for example by introducing ad hoc some auxiliary assumption, or by reinterpreting theory ad hoc in such a way that it escapes refutation. Such a procedure is always possible, but it rescues the theory from refutation only at the price of destroying, or at least lowering, its scientific status. (I later described such a rescuing operation as a “conventionalist twist” or a “conventionalist stratagem.”)

 I’m going to approach each of these points individually to see if I can make a case that evolution doesn’t violate Popper’s positions.  By the way, these refutations are not implications that I think they’re all correct or all-comprehensive of good science, but the point I’m trying to make is that I don’t think evolution as a theory violates Popper’s premise:

 #1.  It is easy to obtain confirmations, or verifications, for nearly every theory-if we look for confirmations.

Popper is loose with the term “theory” here.  In science, a theory is a well-tested hypothesis that is demonstrably more correct than any alternative.  It simultaneously provides predictive capacity, given certain pre-conditions.  Popper was framing this in terms of social science’s looseness, in particular, referring to Freud and Marx.

In any event, Popper is referring to confirmation bias.  I think an equivalent of point #1 Popper made was stated by Richard Feynman when he said “the easiest person to fool is yourself.”  Confirmation bias is a real thing, and it’s important to recognize when we’re doing it.  It’s easy to find confirmation for one’s own belief, and Popper made this point by talking about Marx and Freud.  Popper didn’t like how imprecise social sciences were, and that was because social science makes it easy to link phenomena with social theories.

#2.  Confirmations should count only if they are the result of risky predictions; that is to say,if, unenlightened by the theory in question, we should have expected an event which was incompatible with the theory-an event which would have refuted the theory.

When we propose that we expect a fossil species to be found in a particular strata of soil, or that there must have been a species that existed that had certain attributes, even if we don’t yet have evidence that the species existed, I would argue that’s a risky prediction.  The fact that we’ve verified this, over and over again, with species like Tiktaalik and mammal-like reptiles such as Eothyris, Edaphosaurus, and Estemmenosuchus, and the fact that we consistently find fossils where we would expect them to be, adds a lot of Popperian weight to the matter.

Again, I think Popper was asserting this claim with regards to social sciences.  For instance, it’s not risky to say that a wide-scale famine is caused by the underlying machinations of capitalism, because, simultaneously, there may have been multiple factors contributing to the famine unrelated to capitalism AND because it’s almost impossible to build a model that demonstrates the relationships in a linear fashion.  In other words, any number of things could have caused the famine.  Why didn’t famine occur in other capitalist societies?  If capitalism causes famine, then we should see famine whenever we see capitalism.  Likewise, we might expect to NOT see famine in non-capitalist societies.  There’s no solid model to be built here.

Contrast that with a claim like:  if you start digging in a specific area at a specific depth, you will probably find Neanderthal bones – we know this because we have a good idea of when they lived, the age of the soil at various depths, and we know roughly what geographic regions they lived.

 The Neanderthal claim is a risky prediction and not at all guaranteed to be true; however, one of the things science gives us is increased predictability in a world that is imprecise.  It is true that, even in this context, there is some fuzziness, with concerns about causation versus correlation, but coupled with mountains of other science supporting evolution, it’s pretty damning for creationists that we are able to predict with any success at all, let alone the tremendous success we actually have,  where specific fossils can be found.  In “Your Inner Fish” (episode 1), Neil Shubin talks about trying to guess where a particular fossil species would be found.  He opened a geography book that showed where particular geographic formations can be found, and he knew right away where he needed to go to find them.  He used supporting sciences to predict something that he would have never been able to predict otherwise.

 #3.  Every “good” scientific theory is a prohibition: it forbids certain things to happen. The more a theory forbids, the better it is.

Once again, Popper was referring to social sciences such as economics and psychology.  If the Oedipus complex should prevent something from happening, then it should never happen; the moment that phenomenon does happen, the theory becomes less sound.

 Of course, given the messy world, there are problems with this position.  That’s why we have quantum theory, relativity, and other sciences that caused problems in our classical models.  The universe is nuanced, and there is no universal code or equation.

But evolution certainly forbids things this to the dismay of young Earth Creationists.  Evolutionary theory forbids a young Earth, Noah’s worldwide flood, and Adam and Eve.  Geology plus radiometric dating does more to collapse the young Earth hypothesis than evolution does, but I think a lot of young Earth creationists tend to lump these sciences together as a straw man tactic – it’s much easier to argue against one generic science than it is to argue against biology, chemistry, physics, math, and geology.  Making reference to the evolution boogey man is more effective when you load the term.

 #4. A theory which is not refutable by any conceivable event is nonscientific. Irrefutability is not a virtue of theory (as people often think) but a vice.

I think this is the key premise in the young Earth Creationists’ argument.  In other words, a theory is scientific if it makes clear predictions that can be unambiguously falsified.  Popper here was making the case against Freud, Marx, and other social scientists, and contrasting them with Einstein, making reference to relativity and the famous solar eclipse test that Einstein hypothesized should demonstrate bending of light that would either validate or invalidate his theory.  Clearly, Einstein made a risky prediction (see #2), and falsifiability here meant that the underlying element of the position could be proved true or false, given an event.  If Einstein could not come up with a natural phenomenon that could test his theory (eg a solar eclipse), then that would have rendered relativity non-scientific.

In the same vein, a critique could be lobbed at social sciences because of their impreciseness.  Where Einstein’s theory implied specific and measurable bending of light when looking at stars during an eclipse, claims in Marxism or Freudism were less precise and open to interpretation.  For instance, if Marxism claims that laissez faire capitalism makes people less free, that claim is true until a society is found that simultaneously is laissez faire and its people are free.  However, perfect vacuum-type implementations are rare with humanity, and the wiggle room available in social sciences allows people to engage in motivated reasoning to link events with underlying theories.  The missing piece is being able to describe why it is that a phenomenon was caused by the social theory put forward, AND why that phenomenon couldn’t have happened because of some other reason, perhaps cyclical patterns of human behavior.

There’s a problem when we mix-and-match terms and theories, because not all theories are derived in a similar way.  For instance, we see examples all the time where gravity is shown to be true (throw a baseball into the air, and see if it ever falls down); however, we can imagine situations where gravity (and its constant behaviors) are less consistent, for instance, in outer space, or in a situation where drag and lift are affected, such as when we drop a feather.  This is well explained by gravitational theory and classical physics, but it demonstrates that there is not a universal constant that explains everything all the time.

How we might falsify evolution is to find examples where evolution asserts something that could not be possible given its position, such as a fossil of a human being found in the same soil strata as a dinosaur.  According to evolution, this shouldn’t (and doesn’t) happen.  If it did happen, the theoretical underpinnings would be challenged, particularly if it could be demonstrated that the dinosaur bone appeared to be the same age as the human bone.  It would be tremendous if we actually managed to consistently find evidence that challenged evolution; but we don’t.  The only evidence we ever find reinforces evolution as an explanation for species diversity.

#5. Every genuine test of a theory is an attempt to falsify it, or to refute it. Testability is falsifiability; but there are degrees of testability; some theories are more testable, more exposed to refutation, than others; they take, as it were, greater risks.

If Duane Gish had bothered to consider this part of Popper’s position, he might not have invoked Popper at all, especially with the position that some theories are more testable than others, and science is messier than philosophy in that it is constrained to limitations in the real world when trying to demonstrate the correctness of a position or theory.

Coming back to why creationists have latched on to this notion of falsifiability, I think it’s because it allows them to rely on the fact that we weren’t  present to observe large evolutionary shifts, and therefore, we just can’t prove it.  I would encourage any such person to please abstain from ever serving on a jury for a murder trial, because forensic science specifically deals with reconstructing events in the absence of observation.  Many tools are used in this process, such as analysis of bacteria growth, insect growth, blood splatter, body temperature, body decomposition, etc.  All of these testing mechanisms are admissible in court because the science behind them is sound and they have been shown to be helpful in successfully predicting answering questions about events we did not observe.

#6. Confirming evidence should not count except when it is the result of a genuine test of the theory; and this means that it can be presented as a serious but unsuccessful attempt to falsify the theory. (I now speak in such cases of “corroborating evidence.”)

I disagree with Popper here, as do many scientists.  Sometimes, all we have is induction.  I think an alternative or post-script to this comes with science’s position of tentativeness.  All squirrels are gray until we find one that’s not.  We don’t always have the tools or evidence to test a claim.  For instance, we can’t travel back in time to observe transitional evolutionary patterns.  Does this mean we should bury our heads in the sand, and claim we can never prove evolution?  Young Earth Creationists salivate at this notion, but they’re dead wrong, and their position is harmful to anyone who falls under its spell.

Genetics alone proves evolution.  Shared genes, genes that are “turned off”, divergence, allele frequency, and other clues in the genetic code all  demonstrate slow and steady genetic change over time.  These all support evolution.  But we knew about evolution before we understood genetics – depending on how you define “understanding genetics”, maybe 100 years earlier.  Certainly, the fossil record coupled with radiometric dating proves evolution.  In other words, the fossil evidence is so compelling, we don’t need genetics to demonstrate evolution, because we see transitional figures in the fossil record, and we have a very reliable way of determining their age.

But Darwin had neither genetics nor a robust fossil record to refer to when summarizing his ideas on evolution; rather, it was clear to him, based on his observations of various fauna on the Galapagos, there is an undeniable similarity between species, and that similarity could not be chocked up to chance.  It was brought on by natural pressures and exacerbated by niches and isolation.

To test evolution is a difficult task because people take evolution to be such a broad term.

If you want to test the notion of whether there was a time humans did not exist on the earth, but other animals did, that’s easy to test.  If we fail to find evidence of humans existing, while we simultaneously find evidence that other species did exist, that is supportive of the claim that humans did not always exist in their current form.  We can leverage radiometric dating and look for fossil evidence to build our test for this claim.

If you want to test the notion that, prior to humans existing on the earth, there were quasi-human species that existed, that’s easy to test, too.  It adds an extra layer of fun to the matter when the remains of humanoids (Neanderthals, Denisovans) still contain genetic information that can be compared to modern humans.

Here we can use deduction from our inductive data collection to frame a hypothesis:

1.  Humans did not always exist on Earth
2.  Prior to human existence, there existed non-human humanoid species
3.  The humanoid species were genetically distinguishable from humans
4.  There is evidence that we share a small percentage of our genome with the humanoid species
5.  Therefore, a natural mechanism exists where genetically dissimilar species can reproduce fertile offspring

#7. Some genuinely testable theories, when found to be false, are still upheld by their admirers-for example by introducing ad hoc some auxiliary assumption, or by reinterpreting theory ad hoc in such a way that it escapes refutation.

Clearly, Popper had Marx and Freud in mind when asserting this.

Young Earth creationists would make the claim that evolution has been proven to be false.  Others would claim that it couldn’t be made false because it doesn’t fit their preconceived notion of falsifiable.  The irony here is that their preferred explanation (God) is the ultimate example of something that is untestable and unfalsifiable.  All the evidence in the world would either prove or disprove evolution, but all the evidence in the world would almost certainly not prove God.

Karl Popper’s framework is not without critics, and those critics are probably less concerned about proving evolution than they are with the concern that Popper is overly rigid in light of a complex and messy world.  Karl Popper was a philosopher, not a scientist, and philosophy sometimes has luxuries that science does not.  I think science and philosophy ought to work hand-in-hand in a perfect world, but obviously we miss that part sometimes.

Nevertheless, the fact that science manages to improve predictability via hypothesizing and fine-tuning (ie the scientific method) is pretty good evidence that science is the best method we have for differentiating fact from fiction.  If a theory allows us to make predictions that are consistently correct, then we either have the correct theory, or were very lucky, or we identified inputs (parameters) that are very correlated to the actually correct parameters (for instance, if we say that ice cream sales have been high during the summer, therefore we expect to see more shark attacks, then we are wrong with the input parameters we specified, but we did find an input (ice cream sales) that is correlated to the correct input (air and water temperature PLUS humans in the water).

Popper played an important role in science, and his contributions served to improve the standard for what science should be.  His framework is indeed rigid, and I don’t think that something becomes pseudoscientific if it tentatively relies on induction.  In any event, the claims young Earth creationists put forward, in terms of evolution and unfalsifiability are dead wrong on a number of levels, and their failings are not isolated to inductive evidence.

Advertisements

Author: Tim...Stepping Out

Tim Stepping Out

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s