The Scientific Method and the Myth of the Great Thinkers.

The following is part of Chapter IV-1 of "A Scientific Model Of The Brain: From Instinct to Reason, How Does The Mind Work".   

Since we can only reasonably trust a theory if its predictions are confirmed, if a theory does not make predictions, - regardless of how intelligent its author is believed to be - the theory has 0.0 (Zero) scientific value.

1. The answer according to which a concept is formed when our brain's intelligent agent has an aha moment and becomes aware of it, not only is not scientific, but is nothing but pure magical thinking and, therefore, most absolutely useless.

Scholars in Human Studies adamantly reject neuroscientists' observation, according to which humans basically operate on autopilot. For some truly lamentable reasons, since the rise of monotheistic religions and still to this day scholars in Human Studies have always insisted to lecture scientists in the Life Sciences on how the world works. How could human beings possibly operate on autopilot?; that cannot possibly be correct, since humans' self-awareness is proof that there is indeed some pilot driving our choices and decisions. Now, if we assume and rely on the existence of a intelligent agent, it becomes much easier to explain even the most complicated and momentous phenomena. What led to the invention of farming?: well, someone somehow reasoned it out and everybody followed afterwards. Why did humans start building states?: someone's light bulb went off, obviously. How was writing invented?: Someone had such an aha moment, dah! If the vehicle turn this way and then that way; it is obviously because the pilot simply made those choices. 

Now, if such line of thinking is correct, then you would be most absolutely justified in cursing at me for having been tormenting you for hours on end with this convoluted theory of how consciousness emerged and the autonomous vehicle little by little became able of driving around without any physical pilot's intervention. Certainly, why the heck did I not just sum it all up by saying, that there is a intelligent, decision-making agent in each and all of us, and one day it simply, magically woke up, rose up and began walking. As a matter of fact, life would so be much easier for neuroscientists, as they could then conveniently shift the burden of proof to neurobiologists: namely, since neuroscientist cannot find neural correlates of any intelligent agent, it is left up to neurobiologist to explain how individual neurons dwell on problems until they finally take a decision.

2. The fact that most inventions evolved in fits and starts constitutes evidence, that they were not the result of a series of aha moments; but the pattern can be better explained by a trial-and-error model (of the kind followed by a beginner chess player).

THe problem, however, is that the data simply nowhere provides any support for the physical existence of any intelligent agent anywhere. Indeed, contrary to what until now has always been believed, modern scholars in Human Studies point out , that prehistoric humans did not wake up one day and decided to start farming. Rather, for a period of a few thousand years, they were experimenting with agriculture. Yet, how could anybody experiment with 'agriculture', if one does not know 'agriculture' in the first place? Yes, our preconceptions are so pervasive that we constantly mould our reasonings to conform to their designs. Yet, all the evidence speaks against such proliferation of aha moments:

If all had been a matter of someone's light bulb going off, then the process of gradually developing and improving techniques for the production of food had obviously not lasted thousands of years. Similarly, contrary to what ancient historians narrated, today there is no doubt, that the Roman Republic did not come into being fully formed, as a result of any prodigious aha moment. Rather, it slowly unfolded in fits and starts over a period of hundreds of years. An even stronger case can be made on one of humanity's most extraordinary and remarkable enterprises: to the extent that we are still working on getting it right (and actually making quite a bit of a muck at the present time), there is no way anybody can reasonably sustain, that state formation was someone's glorious idea.

3. Investigating how things work yields better results than asking what things are. 

With the development of writing, human knowledge flourish for hundreds of years, (arguably) ultimately reaching a climax with the rise of Greek philosophy. However, the end of the Classic era, gave way to a long period of stagnation, which lasted well until the middle of the second millenia, when the scientific method was developed. Incredibly enough, the whole problem resided in one single word, and it all got resolved as soon as it was replaced for another. Indeed, the critical breakthrough introduced by the scientific method consisted in the wording of our questions: all until then humans have started all their questions with "what"; empirical research will instead inquire "how" things work. This little detail fixed and the world would never again be the same.       

It is certainly natural to start our questions with "what". In fact, as little children we cannot stop asking "what is this?" and "what is that?". Clearly, for obvious reasons, we are eager to know, and asking for intriguing pieces of information definitely appears as the fastest path to learning and wisdom. After all, there are good reasons to think, that some loving and caring adult will know the answer, and be happy to provide an explanation. However, it is not so effective to investigate on our own "what" is the nature or meaning of something. If there is only Nature to inquire, it is only foolish to expect Nature to ever open its mouth and provide an explanation. If only intuitively, it is easy to see, how we will be more likely to obtain a greater benefit, if we investigate instead "how" stuff works. It all makes sense, so long humans believed in gods, they persisted asking what is the nature of things. However, as the gods kept failing to provide any kind of useful answer, certain humans grew increasingly frustrated and began to look for explanations on their own.   

Moreover, while, as little children, we may have started our learning process by asking: "what is this?" and "what is that?", it is not long before we go on wondering: "why this?" and "why that?". It does help to ask 'what', so that we can identify and demarcate the thing we are intrigued by (For only the principle of divide and conquer, if we are able to partition the world in meaningful things, it will be much easier to make sense of our environment). However, once we know 'what' is what we are dealing with, it is time to begin investigating how such fascinating thing works. We want to know how it works, so that we can then use it in our benefit or, at least, anticipate its behavior. Hence, we need to inquire 'why', so that we can understand the causes and reasons behind the thing's behavior. In other words, as we investigate why stuff happens, we learn how things work. 

Now, if we want to find out why stuff happens and how things work, there are obviously two ways to go about it: either we ask someone else, or we explore the question ourselves. In other words, there are two basic forms of learning: either we listen to someone else's teachings, or we investigate the matter by ourselves. In a way, the first option is represented by the scholastic method and the latter is represented by the scientific method, where the key distinction between the two is that the scientific method does not depend on a human teacher, but the observed data plays the role of the teacher. Hence, at least to some extent, it is possible to establish some parallelism with AI's supervised learning and unsupervised learning.

Obviously, the scholastic (supervised) learning method allows for much faster learning, but - to the extent that the information is administered by a human teacher (the scholastic method's 'Authorities') - the acquired knowledge is likewise much more subjective and unreliable. On the other hand, - since the scientific method requires the learner to investigate the phenomenon in question on his or her own - through a careful observation of the data -; the scientific-method learning process evolves at a far slower pace and demands a much bigger effort; but - on the positive side - the knowledge discovered is significantly more reliable. Moreover, if the scholastic (supervised) method depends on some Authority on the subject to lecture us on how things work; it is worth considering how did such Authority acquire his or her knowledge: did our pandit learned it in turn from someone else, or did he or she investigate it on his or her own? Indeed, sooner or later someone will have to roll up his or her sleeves (that is, unless our Authorities, concerned of ever losing the sway they exert on the Children, are able to convince us, it is hopeless to think the scientific method could be applied to try to improve the revealed truth, they have already lectured us on). Long story short, not only is the scientific (unsupervised) approach the only known method to yield genuine new knowledge on how things work; but - crucially enough - it is objective through and through (since the learning process is not guided by a human teacher, but it is driven by the data generated by the phenomenon under investigation).

4. The scientific method seeks to capture in a simple model the essence of how the investigated system works, so that testable predictions on the system's behavior can be made. 

It is certainly not a coincidence that the scientific method follows the same approach as the brain: it basically consists in the elaboration of a model, where we will try to capture the essence of the system or phenomenon we want to understand. In other words, the model is a simplified version of the reality; in that the minor details are left aside and only the important stuff is kept. THe whole point is (in statistical learning terms) to discover and extract the basic structure of the data - so that we can identify the data-generating model - or (in colloquial terms) what are the actual causes and reasons for the phenomena we observe. In short, how does the thing basically work? 

From our experience with the system under investigation we should be able to make an educated guess as of the principles that govern it, and are therefore responsible for producing the data we observe (i.e. the data-generating model). If the theorized principles lead the model to reproduce the system's behavior, we will be justified in growing confident - but (importantly) by no means certain - of their accuracy; otherwise, we will have to re-formulate and implement them again in a new model. For example, from his observations of the environment, Newton postulated that, whenever a force 'F' is applied to a body of mass 'm', said body would accelerate at a rate of 'a'. This formulation in itself represents a very specific and unambiguous prediction on how bodies move, which (importantly enough) can then be unequivocally tested. For that purpose, a model is build implementing such law. Then, if the model reproduces the behavior we observed in Nature (e.g. any body subject to gravity falls at the acceleration rate predicted by the law), we can grow confident, that the postulated law applies, indeed. If, however, the model's behavior does not mimic the observed data, then we will have to go back to the drawing table.  

5. The brain's non-logical probabilistic modeling yields more information on how the real world works than purely drawing logical inferences.

Now, crucially, contrary to almost unanimous belief, 'scientific' reasoning whatever that actually means) is not about thinking logically (you know, it is always only me who thinks logically and whoever does not agree with me is just being irrational). To begin with, learning in the brain does not follow the strict laws of Formal Logic. According to logic theory, it would not make any sense to say, that just because B follows A (A->B) it can be assume that A likewise follows B (B->A). Yet, the kind of associations learned by the brain does make such assumption. For instance, if we are able to establish that being intelligent leads to making a lot of money (Intelligence -> Money), eventually, unavoidably, we would start assuming that, if someone has a lot of money, in all likelihood, it is because that person is very smart (Money -> Intelligence). The key distinction is that, whereas logic functions in black-and-white terms, and therefore only understands true-or-false propositions; the brain is better suited to deal with all the uncertainty in the real world, because it operates with probabilities. Consequently, as illogical as it may sound, fact of the matter is our brain's 'illogical' probabilistic reasoning is actually better suited to make out how the real world works. As a matter of fact, if we are ready to accept, that, in the real world of ours, intelligence leads to wealth; then it is not really far out to believe, that whoever has a lot of money, in all likelihood, is very smart. In other words, what the brain actually learns is to estimate the probability that a wealthy person may be very smart as well, which in fact is much more informative than what pure logic will ever be able to tell us. Indeed, if - after having met 9 wealthy people, who exhibited "great intelligence" (so to speak) - we come across a wealthy moron; we would have to altogether drop our pure-logic law, according to which all wealthy people is "smart". In contrast, the brain would simply adjust from 1 to 0.9 the probability, that a wealthy person is "very smart".  

6. Whereas philosophizing about things only produces respectable opinions; the scientific method - not unlike the brain - yields testable predictions, that we can then use to optimize our interaction with the world around us. 

Considering the benefits of the brain's probabilistic reasoning - to the extent that scientific research follows the brain's model - it becomes easy to appreciate the beauty of the scientific method. Indeed, the key advantage of asking questions correctly, is that it yields useful information; namely, if we learn how things work, we will be better able to optimize our interaction with those things, in order to achieve our goals. In stark contrast, if we only philosophize about "what" is the nature of things and the meaning of stuff, all what we will get is nothing but opinions. Do not get me wrong, I have also been taught the propriety of always valuing to bits and expressing the most vivid interest for anybody's points of view; but just because all opinions are always respectable, it does not mean they are all coherent or informative. 

7. Since it is strictly impossible to prove the inexistence of something, the mere hypothesis of the existence of a 'thing' does not add anything to our knowledge, but only leads to a meaningless sophistry contest.

Indeed, in the abstract world of theory, departing from a set of axioms, it is possible to prove by logical inference the validity of a certain proposition. . However, in real life, by pure logic, it is strictly impossible to prove that something will always be true, as well as it is strictly impossible to disprove that something will never be correct. Clearly, just because we have not yet found a counterexample, it does not mean that one day one will appear. Likewise, just because we have not yet found any case, where the argument in question is actually correct, it does not mean that one day one will appear. Now, since it is not possible to prove anyone absolutely right or absolutely wrong, all opinions are respectable. Then, since we all shape our arguments based on what fits our interests best, the philosophical debate degenerates into a meaningless sophistry contest, where reason and logical thinking are set aside, and personal loyalties come instead to drive the conversation: if you are with me, you will agree with and support my arguments, and, if you are with them, you will agree with and support their arguments.

8. A definition of 'what' a thing is will only be of any use, to the extent that it provides some indication of how the thing works.

One excellent example of the futility of starting our queries with "what" is the perennial question on "what" is human nature. Needless to say, intellectuals have been arguing this issue for ages, some people think human nature is good and some other think human nature is bad, but we have never been able to reach any sort of convincing conclusion. Certainly, both sides have basis to make some really good cases, as human beings are obviously capable of the most wonderful, as well as the most abominable acts and conducts. Now, tellingly enough, those who most benefit from the System will typically credit it for everything good about humanity, as much as they will blame on evil human nature everything wrong about us; whereas those who hold the short en of the System's stick (or simply nurse a grudge against it) will conversely portray humans as innocent, loving and beautiful creatures brutalized by the ruthless, evil System (Notwithstanding, however, nowadays that Media has popularize the upper class' way of thinking and elite values and culture has permeated even the deepest levels of the society, we are all at last slowly getting convinced that human nature is definitely evil).

In fact, in the humanities and social 'sciences' scholars still mostly focus their efforts on finding the best definitions for the systems or phenomena they study. However, if it is so extremely difficult to define concepts such as civilization, state, power, love, intelligence or consciousness is because we do not have any clear idea as of how any of those 'things' work. Then, everybody proposes his or her definition. Clearly, we all want to be the genius, who gets it right and goes down in History as the one, who coined the term by which everybody knows the 'thing'. In actuality, all those definitions represent unconscious attempts to capture the essence of how the system or phenomenon in question works. For instance, what is a tumor? Well, the critical piece of information about tumors is that there are two basic types: benign tumors and malignant tumors, where the most significant distinction between these two types resides in "how" they "work"; namely, benign tumors are unlikely to lead to the patient's death, whereas the odds of a fatal outcome is significantly higher in cases of malignant tumors. Now, on the other hand, considering the remarks above on our brain's 'illogical', probabilistic reasoning, is it correct to define the thing we call 'Reason', as "the power of the mind to think, understand and form judgements, by a process of logic"? or does the Dictionary's definition reflect our misunderstanding of how the thing really works? 

Or, perhaps, we may need to consider a third, rather unsettling, possibility. Indeed, as Civilization has grown more and more invasive and the System has tighten up its grip over our minds, it has become often the case that the name given to a thing is chosen based on how they want us to perceive said thing. As mentioned before, the ideal definition for a given thing should explain how the thing works; hence, there is no more effective name for a thing than one that includes its defenition. Then, if they want us to love it, the name will define the thing works wonders; whereas if they want us to hate it, the name will explain how evil the thing is. For instance, for those who benefit the most of the system, there is no better name to give to the society's political system than "Democracy". On the other hand, anybody who they want us to perceive as our enemies will quickly receive the name of "terrorist". Evidently, the best evidence of the inexistence of any kind of logic inference engine in our brain is all the grotesquely illogical myths - still to this day - we continue falling for hook, line and sinker. For instance, - as the brightest light bulbs among us keep repeating - our so-called Democracy is not perfect, but it is better than nothing. Yet, is a bit of something bad better than nothing at all?, is a little piece of false information better than no information at all?, is a fake better than no fake at all? A good question here to ask is why is the educated elite so determined to have us believe our fake democratic system is an actual democracy? Be it as it may, forbodingly enough, as it turns out, all what it takes for the people to believe we live under a democratic system is the educated elite give the name of Democracy to the evil monster. Definitely, if the most knowledgeable in the society assert, that in order for the government to look after everybody's interests (i.e. a democracy), all what is needed is separation of powers, Public Opinion in the hands of the wealthy educated elite and elections of the president and the Parlament members by universal suffrage, who are you to call the magical recipe into question?   

9. Neither the countless number of positive examples, the flaws of opposing theories, the authors' reputations nor the number of followers, - unless a definition specifies how the thing works - there is no principled, objective method to determine the correctness of a definition.

Evidently, given that - unless we know how the thing works - there is no principled, objective mechanism to assess the correctness of a definition, eventually, in the battle for eternal glory, any kind of tactic is fair game and any argument can - and often will - be used. Probably the most obvious approach to boast one's theory' correctness consists in citing countless supporting examples. However, contrary to what may be intuitive, just because there are infinite examples consistent with a theory, it is still not possible to firmly conclude the theory will necessarily always be true. On the other hand, needless to say, another frequent, quite effective tactic is to attack someone else's definition. In this wild world of ours, if the winner is going to take it all, one may as well resort on misrepresenting the rival's definition. Clearly, it is much easier to point out flaws in an account, after it has been conveniently manipulated to state whatever one needs it to state. Worst of all, since we choose our loyalties based on, who we perceive is stronger, wiser or, all in all, is going to keep us better fed; at the end of the day, the adherence and following a given theory will be able to obtain will often be determined by the author's reputation, rather than the correctness and accuracy of the theory itself. Thus, as we shall see next, we should be cautious in taking as absolute truth the dominant expert opinion on a given subject at a certain time.

This article is part of Chapter IV-1 of "A Scientific Model Of The Brain: From Instinct to Reason, How Does The Mind Work". 

If it makes sense to you, please share it with your network of friends. It will mean a lot.