No true Scientist commits fallacies

Every single time the weather gets colder someone tells a joke about global warming. Every single time someone does that a logical fairy dies. I don’t know how extinction works with imaginary beings, but I believe their population is pretty thin right now, with all the anti-vaxxers, flat-earthers, climate change deniers and other “it’s just a theory” guys and gals.

Look what you have done!

But…why does a logical fairy die? Shouldn’t it be an environmental fairy? Well… those have been gone for a long time, and besides that, the whole argument is a fallacy, i.e. it sounds right, but has a big logical flaw. In this case it’s a kind of suppressing evidence fallacy or cherry picking. Cherry picking happens when someone selectively picks one or a few data points that fit the fallacious presumption in order to convince a naïve public. Accordingly, when your foolish friend sees you wearing a jacket, trembling in the cold, and says “what about global warming, huh?!”, or when a US senator tosses a snowball down the senate floor (it has happened, google it) they’re both committing a fallacy.

Do you see what he did there?

A fallacy is an invalid, faulty, misleading or just absurd argument used intentionally or unintentionally. There are fallacies with inner logic errors, like an appeal to probability, or the gambler’s fallacy (independent events are, well, independent, so if you toss a coin four times and gets three heads in a row you won’t have a higher chance of getting tails in the fourth), or informal fallacies, like the slippery slope (when one establishes a series of catastrophic events starting with an argument and culminating in a major disaster, e.g. “if we open our doors to immigrants they won’t incorporate our traditions and will preach their culture and religion to our kids, and then in a few years our traditions will vanish”).

Seriously?!

The use of fallacies is common in everyday speech, like in social media arguments and informal talks, and unfortunately also in formal media outlets, political hearings and rallies and even in some scientific discussions. Yep, scientists, like every other human being, eventually commit fallacies in the middle of their arguments, sometimes unknowingly but sometimes intentionally. The reasons are many; at times the scientist just makes an honest mistake, but at other times they actively protect the interests of corporations and other groups, like the Robert A. Kehoe and Ethyl Corporation case (see more in the Cosmos: A Spacetime Odyssey episode 7).

The thing about identifying, understanding (sometimes pointing out) and countering fallacies is that they are, by their own nature, a tool in every jerk’s rhetoric toolbox, used to mislead or manipulate public opinion or to try and undermine whole groups and their members. Even though fallacies can be present in any kind of conversation, debate or interview, not only scientific ones, I believe it is our job as researchers to inform the general public about science and its logic, hence it is also our job to educate the public on this kind of fallacious rhetoric.

Everywhere, everyplace, everytime

The first thing to do is identify the fallacy. There is a great website here that details 24 types of fallacy with examples and also has a big nice poster about it. We have already talked a little about three of them (cherry picking, the gambler’s fallacy and the slippery slope), but let’s look at some other common types.
The strawman fallacy is really common in both scientific and non-scientific environments, and consists of a misrepresentation of a theory, concept, idea, etc, in order to try to debunk the whole theory.

Dude, it’s just pizza.

Another really common and eventually innocent fallacy is the false correlation, where someone argues that if two or more thing happens simultaneously or at the same rate they are necessarily connected. This happens a lot since we normally assume that correlation is a symptom of causation, which it isn’t. To see how correlations can be spurious a website named Spurious Correlations gathered around 30,000 of them. My favorite one is about Nicolas Cage films and cases of drowning in pools. This kind of fallacy is regularly used by a majority of frightened parents and by some unreliable doctors and policy makers supporting the anti-vaxxer movement.

Well, we can never be sure…

There are many other examples you can find in rhetoric books, the internet, your philosophy or logic classes and so on, and I strongly encourage you to read more about them, learn about these kinds of argumentative strategy, spread the word about the differences between reliable and fallacious arguments and crush those trolls and science deniers on Facebook.

P.S. Remember: this is never about convincing the troll/denier, in fact it’s about shedding some light on a misunderstood subject so everyone interested in it can grasp it correctly (but go crush those morons nonetheless).

The ducks of science

Living in the XXI century with our western culture and stage of technology, environmental and political issues, degree of education and concern about the future of humanity, we as human beings must have pretty clear in mind the principles, assumptions and precepts of science so we can make better decisions, take care of ourselves and the world we live in. The actual conditions where Brazil and other countries are in terms of basic and scientific education show us the great gap in understanding how science works and allow all the knowledge and technology that our world has formed.

First thing to know is that science doesn’t prove anything! We never achieve the truth. This is not pessimism, it is just to assume that we are beings that have understanding limitations (including sensorial and technology issues) towards the complexity of natural world. It is impossible to know about all variables and conditions that exists allowing patterns and process to be as they are. So what occurs is: if there are many strong evidences that support a hypothesis or a theory and nothing against, we say that this hypothesis or theory is well corroborated and the best available…and that is all!              

It is extremely important to notice which hypothesis is scientific and which is not. According to Karl Popper, a hypothesis must be falsifiable to be part of science. The more a theory states, the more opportunity there is to show that the world does not actually behave in the way it is. A good theory makes broad statements about the world, and as a consequence is highly falsifiable and resists every time it is tested. A vague theory is not falsifiable. If it does not make it clear exactly what it is claiming, then testing it can always be interpreted as consistent with the results of those tests. Hence the need to be clearly affirmative and precise.

It is only by excluding a set of logically possible propositions of observation that a law or theory is informative. Affirmations such as: (1) All points in a Euclidean circle are equidistant from the center or (2) I am very lucky person, tell us nothing  about the world. A law or scientific theory should give some information about how the world actually behaves, thus eliminating the ways in which it could possibly behave otherwise. The law “all planets move in ellipses around the Sun” is scientific because it states that planets actually move in ellipses and eliminate orbits that are square or oval. Just because the law makes decisive statements about planetary orbits, it has informative content and is falsifiable.

Observation is guided by theory and presupposes it. We only see what we already know. It is impossible to observe natural phenomena with nothing in mind and without all the knowledge that has already been taught to us. So we know theories and while making observations about the natural world, we question and raise hypotheses about it. However it is impossible to arrive at universal laws and theories assuming that theories can be established as true or probably true in the light of observational evidence. When somebody hypothetically says “In my life I only saw black ducks (because I only saw Daffy Duck cartoon in my whole life)”, we can’t deduce from this that “all ducks are black”. On the other hand, it is possible to make logical deductions starting from singular propositions of observation as premises, to arrive at the falsifiability of laws and universal theories. That is: you don’t confirm when your observation agree with the hypothesis or theory, you just don’t exclude until future observations and/or experiments fall into disagreement. So the premise “all ducks are black” is easily and actively refused with the encounter of ducks of other colors (Donald duck for example). So the acceptance of theory is always tentative, but rejection of theory can be decisive.  

So basically science begins with problems associated with explaining some aspects of the world or universe. Falsifiable hypotheses are proposed by scientists as solutions to the problem. The conjectured hypotheses are then criticized and tested. Some will be quickly deleted. Others may prove to be more successful. These should be subjected to further criticism and testing. When a hypothesis that has passed a wide range of rigorous tests successfully is eventually falsified, a new problem arises. This new problem calls for the invention of new hypotheses, followed by criticism and renewed testing. And so, the process continues indefinitely. science progresses by trial and error, by conjectures and refutations.

Bulletproof ideas

But life is not a bed of roses! We cannot forget that a scientific theory consists of a complex of universal statements rather than an isolated statement like “all ducks are black.” Very often a theory can be modified to “protect it” in case of a falsificationism threat. This “theory improvement” by adding new content and information can be very inefficient, even though it could be considered acceptable by “falsificationists” at a first sight. These later modifications  added to an existent theory are called ad hoc modifications – if it doesn’t have testable consequences. For example, considering a generalized theory where is said that “(all) water is good for health” and then, looking at a village where people drank the water getting sick and dying, we clearly can see that this premise is false. However, in attempting of “save” his work the researcher responsible for that theory rewrite the phrase saying that  “(all) water is good for health, except that water of that certain village”…well, this can’t be done. It’s a explicit example of an ad hoc modification, where we can see that this changed part of the theory can’t be tested separated of the original one. The consumption of water in the whole world by any human being becomes a test of the original theory as the tests of the modified theory are restricted to that village. Thus, the modified theory is less falsifiable than the original one and we don’t have to be a very careful falsificationist to see that. Of course in the real world this ad hoc modifications are way more complex, so we have to be careful with this additions to theories and be able to discern the valid ones

We must have in mind that the propositions of observation depend on theory and that they are fallible. So the problem is: falsifiability is based on the assumption that propositions of observation are perfectly safe. So if a theory or part of it collides with some proposition of observation, nothing in the logic of the situation requires that it should always be the theory to be rejected. An observation proposition can be rejected and the theory with which it is shocked to be retained. Consequently, direct, conclusive falsifications of theories are not achievable and not according to this is being part of the so-called naive falsificationism.

Historically it is possible to find observational propositions that were generally accepted at the time and were considered inconsistent with theories, which were not rejected. If so, classical scientific theories widely accepted today would never have been developed. Important concepts in the history of physics such as force and inertia did not arise as a result of careful observation and experimentation. Neither they arose through the falsification of audacious conjectures and continuous substitutions. Initial formulations of new theories, involving new conceptions incompletely formulated, were maintained with perseverance and developed in spite of apparent falsifications.

Literature:

CHALMERS, Alan F. O Que é Ciência Afinal? São Paulo: Brasiliense, 1995.

by Stephanie Sampronha and Bianca Cegolin