In the previous article we touched upon how nothing can ever be “proven” to be true. This is because no matter how many examples of a particular observation we corroborate, (a process known as inductive reasoning), it only takes one example to the contrary to prove that our theory was false.
For example, our understanding of evolution as “True” is based on the fact that for every single fossil uncovered and dated in the history of the practice of uncovering and dating fossils, every single one has fit within an order which implies that the theory of evolution is an accurate representation of reality. It would only take a single fossil dated to the wrong place in the timeline to prove evolution to be false, which is why evolution is regarded to be a “Theory” rather than a “Fact”.
The same is true of gravity, all we need to do is find a single location where the rules of gravity do not hold true in order to expose our entire theory of gravity as being an inaccurate representation of reality.
If this seems hard to digest, consider that a single discrepancy in found during observation of the orbit of Mercury led to the dismissal of Newton’s universal law of gravitation, which up until that point had been believed to be absolute. This discovery that Newton’s “universal law” was in fact not an accurate representation of reality eventually led to the development of Einstein’s theory of relativity. It is in this constant flux of dismissing our old beliefs as new evidence reveals itself that the sum of human knowledge moves forward.
So as it seems that facts are ever elusive, what is it we are really trying to do, for instance, when we endeavour to learn? It was thought for a long time that the quest for knowledge was synonymous with the quest for facts, for details about the world and universe as it exists, independent of human experience. However, as we discussed in Part 1, our mind’s ability to grasp information is tragically lacking, and so our naive belief that we dabble in the discovery of the truth of our world must be put aside, and we must ask ourselves the meta-question of epistemology. What are we trying to achieve when we set out to know?
The Why of Science
If knowing is truly so elusive, why endeavour to know? And once we are satisfied that a thing can be known, what does it really mean to know it?
Similarly to the case of ethics, our quest for knowledge is grounded foremost in utility: abstractions of complex, nigh insensory phenomena such as spending habits and relationships between disease and behaviour are, for the most part, very useful things to have. However, a prime difficulty arises when we set out to drive the nail so slightly further than abstraction, and we stumble upon the impossibility of deriving causation.
Just as it is impossible to truly know something, (we can only hold confidence in a thing with respect to avaliable evidence), the link between two things, two phenomena, is also an elusive non-entity, which for exactly the same reason as all other knowledge, is impossible to conclude.
This impossibility of establishing causation marked a fundamental problem with the common pursuit of knowledge, which was formally understood to be the search for the metaphysical truth of the universe, (known then as “Natural Philosophy”), until the 1830s, when Philosopher August Comte challenged the notion that our common idea of science was to be in pursuit of metaphysical truth. Comte’s way of thinking is known as Scientific Positivism, and espouses that observations merely allow us to form predictions, and consistently confirmable predictions allow us to generate laws, or theories, such as the theories of evolution and gravity. Comte argued that since causation is unobservable, and thus conceptual rather than physical, it is epistemologically irrelevant and thus can only be ignored when considered with respect to the accumulation of collective human knowledge.
To elaborate, whilst humans do tend to believe that they bare witness to causal relationships each day, all we really experience are sequences of sensory inputs – inputs which may as well be coincidental. This is because the “cause” itself is simply conceptual. Although we habitually infer cause in the abstract, we never actually experience it; it is not “there” to be experienced. Instead, cause is a structural tool borne of our ability to abstract rather than a fundamental part of reality.
Theories, also known as natural laws, are things that are empirically justified to be consistent regularities. For example, the natural law of gravity explains why the ball falls when you drop it. However, to confuse the abstraction of gravity with some sort of ethereal cause and effect mechanism existing in the world in a sort of meta-physical way would be a mistake. The theory of gravity may explain the ball falling, but it does not cause it. To claim this is to mistake a theory for reality, (or, in fancier words, to mistake an epistemological entity for an ontological one).
The letting go of and the falling of the ball are both separate events which occur in isolation, and our theory of gravity is an abstraction, (a practical tool borne of our minds) that allows us to take advantage of a seemingly consistent sequence of events which so far seem to always follow one another. Whilst mistaking the explanation of gravity for the cause of the ball falling may not seem like a large or particularly consequential mistake, we shall see later in this series how conflating abstractions and reality in this way (formally known as a category mistake), can lead to disastrous consequences.
Humorously, one of the most common explanatory words of our language, “because”, is rendered philosophically moot in the face of Comte’s wave of Scientific Positivism. But what are we to make of a science which dismisses causation? One would intuitively believe causation to be all too a fundamental mechanism for it to be entirely removed from the conversation of knowledge, but yet that intuition is merely based on a misunderstanding of what our accumulation of knowledge really stands for.
The Mass Corroboration of Correlation
When we as responsible seekers of knowledge endeavour to add to our collection of uncovered mysteries, what we’re really doing is observing coincidences. We’re observing instances (otherwise known as phenomena) which seemingly only occur in close relationship with one another. When we research we are not seeking causation, only as wide a set of as tight and relevant correlations as possible.
You may be familiar with the old phrase “correlation does not equal causation”, and that is principle to the fact that *nothing* equals causation. All we can do is accept that when the correlated incidence occurs, it will coincide with the incidence we are expecting it to. This attitude to, or rather, this dismissal of, the concept of causation is the primary reason why therapies such as acupuncture are continue to see use despite not being well understood casually. We know that when such therapy is practiced it tends to be followed by a positive outcome, that much is clear even in the absence of a compelling explanation.
I am aware that a few of you may be scratching your heads at the idea of a causeless world, but the attribution of cause in daily life is better understood as a social mechanism than an accurate mapping of reality. It is useful to be able to attribute cause and effect across a spectrum of subjects during daily life and for the most part, just as the case was with the phenomena of knowledge discussed in part 1, a carefree approach to the subject mixed with a bit of luck is all most of us really need to get through our lives, that is until we begin tackling broader subjects.
In order to better conceptualize the causelessness of reality, I invite you to consider the following. If I were to hold a ball around waist height and then suddenly drop it, would you consider my letting go of the ball to be the reason that it falls? Perhaps my letting go and the falling of the ball were simply two separate events which, upon repeated observation, seem always to occur together. This distinction may seem petty in such a benign context, and so I would like you to then consider a negative correlational relationship between watching television and achieving lower grades in school, (that it to say, a dataset that implies that students who watched more television achieved lower grades). Whilst the astute of you may already be considering that the achievement of lower grades is likely to be due to something other than watching TV, I would like you to instead ignore that urge and instead consider: despite the potential existence of a more satisfactory casual explanation for the correlation, can one regardless use the observation of a particular student watching a lot of TV to predict with some competence of accuracy that his grades in school will be quite low? The answer is yes. Although the TV watching may not be directly or even implicitly causal for the bad grades, it is possible that both the watching of TV and the achievement of bad grades are linked in such a way that we can, with some degree of confidence, use the occurrence of each to predict the occurrence of the other in the case that more compelling evidence is unavailable. A satisfactory casual explanation is actually unnecessary for this process.
So you see, even without establishing causation, a consistent observation entails a fair amount of confidence with regards to future prediction. This statement does, however, loop us back around to the problem of induction – the problem of the chicken who thought the farmer was going to feed him every day and was eventually served for dinner. Still, the chicken was correct in his assumption almost 100% of the time, and in the absence of corroborative evidence, or evidence to the contrary, that is really the best that any of us can hope for.
A Summary of The Discussion So Far
Causality, just like all other forms of knowledge, can never be proven – we can only improve the confidence we have in our explanations. We do this by corroborating sets of correlations and using them to imply definite coincidence across a timeliness, wherein the occurrence of a single phenomena is consistently followed by the occurrence of a specific other phenomena.
Because all knowledge is just a single verifiable counter-example away from being disproven, we sometimes see otherwise thought to be definitive theories of metaphysics become obsolete in the face of surprising new discoveries.
Despite the tendency to form theories about the nature of the universe, (theories of metaphysical truth), from our scientific observations, the purpose of science is not the pursuit of these “truths”, but of the pursuit of bringing together evidence which allows us to make consistently reliable predictions about the relationships between phenomena. This way of thinking is known formally as Scientific Positivism.
I hope you have found this discussion on Epistemology at least somewhat enlightening. During the read so far you may have noticed that I have several times made a distinction between conceptual and physical phenomena. The next article in this series will go deeper into the distinction between these two terms, and is, I believe, one of the most fundamental topics in philosophy that one must grasp if one is to move on to more complex topics such as politics or ethics. See you there.
