How do we know what is true?

The ultimate goal of the scientific method is to falsify the closest approximations to reality. There are no absolute truths in science because all theories are liable to perpetual refinement and revision.

In science, plausibility is established not by proving things true but in fact, but by demonstrating that they have not yet been shown to be false. Tentative statements of empirical reality, or hypotheses, are constructed from observations of natural phenomena. The observations are derived from experiential information obtained by the senses. The hypothesis is then tested using carefully-controlled experiments and tests that aim to falsify it. If it continues to survive a battery of falsification tests for a prolonged period of time, then it becomes a theory, which is a conceptual framework that relates facts together into an explanatory structure. Theories can never be proven true but they can only be shown to be false. Acceptable theories are ones that are temporarily not wrong.

Thus, the way that the scientific method claims epistemic authority on anything is by recognizing its own fallibility. A hypothesis must be falsifiable, otherwise it is not an acceptable hypothesis.  Repeated failure to falsify a hypothesis does not confirm its absolute truth, whereas just one falsification alone can be enough grounds for rejecting it. Therefore, we cannot be certain about what is true, but we can, however, be certain about what is wrong. There is a possibility of disconfirmation which must always be acknowledged.

Theories can replace preexisting theories if they have more explanatory power. Einstein’s theory of gravity replaced Newton’s theory not because Newton was wrong but because his model was incomplete in accounting for some of the anomalies observed in planetary motion (the problem of Mercury’s perihelion). Einstein’s theory accounted not just for things that the Newtonian model explained but also for the new observations that were found to be incompatible with Newton’s theory. As previously discussed, what theories fundamentally are is that they are conceptual superstructures that describe facts. When there is a better way to relate those facts together, a new theory emerges.

Creationists always like to point to evolution as “only a theory” as if that’s a bad thing. But, in science, a theory is a good thing. In fact, catapulting a hypothesis into the status of a theory means acknowledging its good epistemic value and its high empirical content. In addition to that, creationists never present a side or attempt to support it with evidence. Their only evidence is to attack evolution, which somehow lends credibility to their beliefs that, to begin with, are not really grounded in any testable hypothesis. Creationism is a belief that has no explanatory power and which cannot and does not generate predictions that can be tested against experiment. You cannot deduce any facts from it. And, most importantly, it is a firmly held conviction that leaves no room for falsification. To be a creationist means to have already decided what is true and to be blinded by your preconceived prejudice that anything contradicting your beliefs must be inherently false. If you set up an experiment, you would impose a prejudiced idea of what the results should be like and you would twist the scientific evidence to fit your bias.

But, science never runs on assertions. Science is a discipline that recognises and celebrates uncertainty. This is crucial because it provides opportunities for ongoing refinement and for new ideas to be hatched up.  To cling to a belief and will it into existence means to miss out on any new insights and ideas that have been clouded out by  your preconceptions. Facts about the natural world should be presented simply as the evidence of reality presents them to our logical faculties. They should not be presented as our assumptions want them to be. Our logic should therefore always derive from and be continually morphed by the evidence of reality because nature is so much more imaginative. To hold a firm conviction prevents that.

“It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.” – Sherlock Holmes


Featured image credit: Hidden London


8 thoughts on “How do we know what is true?

  1. It’s rather unfortunate that you believe this. Especially from a school with such an excellent History and Philosophy of Science program. While falsificationism was for awhile a popular theory in philosophy of science, it was showing its age by the 60s and by the 70s and 80s was very close to six feet under.

    Even Karl Popper, the developer of falsificationism, believed differently than you that scientists do in fact start with some theory in mind. Otherwise they would have no way to decide the relevant facts, because the same ‘relevant facts’ aren’t obvious to everybody. For Aristotle for instance, the relevant facts about motion were whether the object was moving in a natural motion or if it were constrained by some force. As it appeared obvious to him that different things moved in different ways in the absence of force (fire rises, earth falls), the natural motion of a substance varied depending on its constituent elements. In psychology, behaviourists thought that it only really made sense to look at outward signs, seeing how things acted in order to piece together a coherent psychology. They were very successful in this, although of course they were eventually augmented with cognitive psychology. More recent examples are presented by Thomas Kuhn in “The Structural of Scientific Revolutions”.

    As for falsificationism, it does not square with how science actually progresses. If Newton had decided to discard his theory upon falsification, he should have been discarding his theory at any moment of his life. It took something like a century to explain the orbit of the moon, and it took Einstein to explain the orbit of Mercury. If it had been up to a naive falsificationist, we would not have Newton’s theories despite all their explanatory power, because according to the falsificationist, explanatory power is useless. Psychoanalysis has explanatory power! Marxism has explanatory power! Yet Marxist ‘science of history’ does not qualify as Popperian science. Creationists can explain, or rather explain away, lots of things. You put it as ‘Einstein’s theories had more explanatory power’ but more accurately, Einstein’s theories weren’t falsified by the evidence, unlike Newton’s. This indicates just how insanely absolutist falsificationism is – while we threw out Newton’s theories when Einstein gave us a better alternative, the falsificationist would have had us throw them out from the beginning.

    Other examples are everywhere, the particle-wave duality used in quantum physics is a contradiction in terms, not merely in a mind-blowing way but in such a way that the notion of an overlap between the two is logically incoherent. That is a great instance of instant cause for falsification and throwing out the theory, however it continues to make good predictions so until something comes along that better suits the situation that’s what we use.

    Even latter day falsificationists such as Lakatos made room for dogmatism in science. The prevalent mode of explanation was a hard core of a theory which was dogmatically defended (for Newton, the three laws of gravity and universal law of gravitation) and a ‘protective belt’ of potential implications of the theory (the explanation of the moon’s orbit, etc) which could be cut out at will. While falsificationism is enough reason to weed out hypotheses in the protective belt, it is not enough to remove the hard core because of how successful the Newtonian hardcore was at explaining mechanics. Only if the protective belt dwindled is it time to reconsider the theory.

    You can demean creationists, say they’re not doing science properly, say they’re propagandists, or simply flat out wrong. However don’t do it by sounding like someone with an ignorance of philosophy and history of science comparable to creationist ignorance of science proper. And as much as we would all love it if such a beautiful theory of science worked, please consider the realities and that such absolute stances on science would in fact hamper scientific progress as we throw out all of our best theories on small falsifications.

    Liked by 1 person

    1. > Well, you could argue for dogmatism in science but that is not how the scientific method itself is set to work.

      The scientific method is an abstraction, a model for scientific conduct.

      The issue is that a) it does not appear to be an accurate model (ironically the model of scientific conduct based on falsificationism is falsified by history) as one may observe in the fact that Newton’s theories weren’t thrown out, b) its prescriptions if taken to include naive falsificationism would in fact leave science worse off (given that it suggests the expulsion of Quantum theory for being contradictory as explained before, the expulsion of Relativity for not working at the Quantum scale, and String Theory for being unfalsifiable without colliders with energy requirements resembling that of the sun), and c) there doesn’t seem to be any overarching things that all scientific conduct have in common that could be distilled to a method, e.g. how much does a evolutionary theorist such as Massimo Pigliucci who spends a lot of his time offering theoretical extensions of evolutionary theory have in common with a Phd student watching neutrinos move through a detector inside a mineshaft?

      The example of statistical tests is fair, but considering statistical analysis as representative of scientific method overall is a huge mistake. Do you think that the people finding the percolation thresholds of high-dimension fitness landscapes are using p-values to determine whether their mathematics checks out? What about case studies? There’s no test for the statistical significance of a single data point. Or what about scientists selecting something for its explanatory power rather than successful new predictions (as in the case of Einstein’s correct prediction of the perihelion of Mercury, which is what got scientist’s attention despite falsificationist swooning over the Eddington expedition).


  2. This is just a rehash ad nauseam of Popper’s falsificationism. Unfortunately, it looks good on paper to some but doesn’t really stack up against how science really operates. In the real world sometimes scientific theories are held onto despite apparent falsifications and these have been “proved” correct at a later date. For example, people persisted with the heliocentric theory when there was a plethora of apparent falsifications of it from those who were wedded to the geocentric theory. Falsification does not necessarily mean that a theory is incorrect and is not necessarily a reason to abandon a theory. Other factors operate in the dynamics of theories.


  3. I am not a religious person and have no stake in Creationism, however your criticism seems hypocritical in that Evolution is a faith based hypothesis as well. In what way is Evolution falsifiable? What specific prediction does it make and what experiment has or can be devised to falsify it? I suspect that Evolution is much closer to the truth than Creationism is but please remove the spec in your eye before criticizing others.


  4. To state “there are no absolute truths” requires the establishment of a truth in consciousness to make such an assertion and then to hold on to that axiom.

    The reality is that there have/has to be an establishment of truth to claim that there isn’t/aren’t any absolute truths.


  5. The God Theory: Universes, Zero-point Fields, and What’s Behind It All by Bernard Haisch (Author)

    On the one hand, we have traditional science, based on the premises of materialism, reductionism, and randomness, with a belief that reality consists solely of matter and energy, that everything can be measured in the laboratory or observed by a telescope. If it can’t, it doesn’t exist. On the other hand, we have traditional religious dogma concerning God that fails to take into account evolution, a 4.6 billion-year-old Earth, and the conflicting claims of the world’s religions. In The God Theory, Bernard Haisch discards both these worldviews and proposes a theory that provides purpose for our lives while at the same time is completely consistent with everything we have discovered about the universe and life on Earth. To wit, Newton was right — there is a God — and wrong — this is not merely a material world. Haisch proposes that science will explain God and God will explain science. Consciousness is not a mere epiphenomenon of the brain; it is our connection to God, the source of all consciousness. Ultimately it is consciousness that creates matter and not vice versa. New discoveries in physics point to a background sea of quantum light underlying the universe. The God Theory offers a worldview that incorporates cutting-edge science and ancient mystical knowledge. This is nothing less than a revolution in our understanding.


  6. I don’t understand why the book advertisement makes it through moderation but my comment explaining how the history of science contradicts your analysis of science is stuck in limbo?


Leave a Reply to cwhiatt Cancel reply

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

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

Facebook photo

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

Connecting to %s