Adonis Diaries

Posts Tagged ‘Imre Lakatos



Scientific theories must be Falsifiable? Best indicator to sort out the pseudo-sciences

If certain forbidden conditions arise, then the scientific theory can be said to be falsifiable. (Karl Popper).

If a theory cannot logically be refuted, which means it says nothing about reality, then it is vacuous, absurd. This theory does not speak about reality and it can be classified as a pseudo-science, like psycho-analysis and Marxist theories. Why?

Because the discovered things are observed Only after the event or session.

Many significant mathematical and logic theorems cannot be refuted and are fundamentally not directly related to reality around us: They are of abstract concerns.

Scientific predications should be refuted.

Scientific theory starts with bold conjectures and should stand up to rigorous tests that are meant to refute the theory.

Since inductive reasoning cannot be justified, then observations are not meant to support the “truth” of the conjectures but to try to refute them.

Basically, no matter how often the theory survives the tests, it still remains a conjecture, until it fails and should be rejected by the real scientists.

Many scientists tend to cling to the failed theory by trying to introduce ad hoc modifications with the purpose of protecting their niche of specialization that their livelihood is based upon.

It is alright to add modifications in the conjecture, as long as they are falsifiable and withstand the tests of falsification.

For a century now, the interpretation of data analysis of scientific experiments sets a margin of error at 5%, by consensus, for disproving a hypothesis.

The game is to state the opposite of the conjecture (hypothesis) that we strongly believe in so that this opposite hypothesis fails “significantly” and claim that the conjecture is still valid.

Obviously, if the consequences of the hypothesis are dangerous, such as in the medical field and treatment of diseases or safety behaviors, it is logically tantamount to decrease the error margin to below 5%.

Apparently, researchers are wary of adopting a lower level of 5% in order to resume their research on the same conjecture.

In many cases, scientists will do their best to find fault in the experiment procedure and processes, and the variables that they failed to control… just to maintain the well-trotted theory.

This is a very common behaviour, given that researchers plays within the accepted consensual paradigm (framework of investigation), in order to easily get grants for “further research” work.

It good to ponder on the case of the philosopher Imre Lakatos.

An astronomer is not willing to abandon the Newton’s theory on planetary movement. Each time his computation does not match the theory, he introduces additional conjectures such as other influencing factors (an invisible twin body, forces, magnetic clouds fields…) and asks for grants to send satellites, sophisticated advanced telescope and equipment to double check his conjecture and validate the original theory he is clinking at.

There are more complicated issues than refuting the conjecture.

We need to consider the initial conditions or premises and verify that they are compatible with the theory.

We need not to just hang on the current paradigm or framework in our investigation and get out of the comfort zone when our scientific credentials are well established. Otherwise, who is to take on the task of moving on and allowing science to progress in the right direction?

That a scientific theory must have a predictive power should not necessarily be a basis to abandon a theory when a couple predictions fail to materialize.

However, theories that have sustained frequent tests should not be merely considered as conjectures, but a phenomenon that is aching to a “truth”, a firm law. Otherwise, science will keep circling in the vicious process of trying to refute a well-tested and sound theory.

Note 1: Read “How to think like a Bat” by Peter Cave 

Note 2: I conjecture that psycho-analysis can be transformed into a predictive scientific theory. The cases have accumulated and can be sorted out to discover the initial conditions or factors that affect a particular emotional disturbance.

Actually, experimented psychologists start with a conjecture after a few sessions with the patient and turn around their hypothesis for confirmation. And this is a human behavioural problem even if scientists are trained to avoid the confirmation fallacy in their work.




March 2023

Blog Stats

  • 1,518,977 hits

Enter your email address to subscribe to this blog and receive notifications of new posts by

Join 764 other subscribers
%d bloggers like this: