Knowledge in the "hard sciences" seems to depend entirely on attested facts upon which a theory is built, the goal of which is to imagine the most plausible none-attested facts (which are in truth no facts). A theory is then a construct which augments - through a mouvement, slow and hesitating - a range of

  • attested facts (how to observe them, where to look for them, etc.)
  • none-attested facts which are most likely to fit both the attested facts (external, both old and new) and the internal orientation of the addable none-attested facts.

My language and way of presenting the subject is entirely personal, I have no education on the subject, so forgive in advance the presentation. I do however think the idea is conveyed.

Question: why the need for Theory - outside from the sentiment of coherence or the occasional assistance in direction of research ?

Explanation of the question: It seems to me that we could be at a point where we could start collecting attested facts (granted it will be through the prism of some model, a model is however less intrusive than a theory) without trying to overly extend our epistemology to what is not attested. It is as if we picked-up the "habit" of building theories because they were justifiable in a given part of our history (because of how few attested facts we were able to gather), the habit today might be costly, since we could know more without having to presume. This does not mean we should try and make a whole out of the fragments, but that would and should come naturally once enough attested facts emerge a viable grand scale image (a couple [set of attested facts, set of models] is intuitive). It's forcing such an image which seems to be problematic in my opinion.


1 Answer 1


I can think of two reasons:

  1. Theory allows for prediction. By building a theory based on observations (attested facts) we can make predictions about how other systems might behave. I'm an organic chemist, so let's use the synthesis of organic compounds as an example.

Many organic compounds exist in nature. These include things like sugars, amino acids, proteins, DNA, hormones, neurotransmitters, fats and many more categories of compounds. Without theories about the chemical nature of these compounds (which is not directly observable - my next point) it would be impossible to synthesize not only naturally occurring compounds, but also to imagine and synthesize new, non-natural compounds (such as medications).

  1. Theory allows for explanation of non-intuitive observations. Many natural phenomenon exhibit properties that cannot be intuitively understood by humans. The chemical nature of matter is one example. The physics revolution of the early twentieth century provides two more striking illustrations of this. Newtonian mechanics had provided adequate explanations for much of the behaviour by moving objects on earth. However, the interaction of light with matter could not be properly explained using Newtonian deterministic mathematics (I should point out that Newtonian mechanics are a theory as well, one that allowed Newton to do things like correctly predict the orbit of planets, or allow us to calculate the flight path of rocket). One key illustration of the problem was the Ultraviolet Catastrophe (as an aside, that is the best name for a scientific problem ever). The theoretical frameworks of quantum mechanics and relativity were required to explain the behaviour of either very small particles moving close to the speed of light (quantum) or very large bodies and distances interacting with light (relativity). And yes, this is a massive oversimplification.

This then ties back to my first point (prediction). For example, GPS satellites use relativity to correct for differences in their velocities relative to the surface of the earth.Source Without the mathematical framework of relativity, there would be a margin of error of about 11 km per day in GPS locations. Source

Ultimately, your argument rests on the idea that our senses are sufficient for observing and describing all natural phenomena. Theories allow us to make predictions about how systems will behave, and to explain our observations.

  • Hi ! In fact, some theories (the few you've mentionned included) do a fine job of prediction, other theories seem to rely more on the none-attested than the attested. Take quantum theories, take string theories. Leaving aside the fact that much of our knowledge is none-factual in any case, the extent of how far the none-factual has been driven is dazzling. This begs one of two conclusions: either all our knowledge is theoritical at base (we never know the truth, but only a controlled span of phenomenology) or we can know some truths and we accept polluting it for prediction.
    – Gloserio
    Commented Jun 5, 2019 at 13:16
  • +1, Thanks for the reply btw, and sure enough: Ultraviolet catastrophe is bad jokking or jokking badly xD
    – Gloserio
    Commented Jun 5, 2019 at 13:17
  • @Gloserio, I disagree about quantum mechanics. Much of what lay people think it predicts have nothing to do with the actual theory, and it has allowed important developments (predictions) [What Has Quantum Mechanics Ever Done for Us}(forbes.com/sites/chadorzel/2015/08/13/…). A better example on your side might be string theory. Commented Jun 5, 2019 at 13:28
  • @Gloserio, the ultraviolet catastrophe is so-called because of the experimental results were catastrophic for the theories about light being used at the time. Commented Jun 5, 2019 at 13:29
  • Lauteman: I am not denying that quantum theory has superb predictive potential, but the whole construct is far too convoluted (both mathematically and philosophically), which I guess is tolerable as long as the theory works. Regarding ultraviolet catastrophe, it was just an attempt for a joke, I looked it up when reading you.
    – Gloserio
    Commented Jun 6, 2019 at 7:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .