How Do I Justify Saying That There Are No First Principles Available To Us When We Try To Generalize Measurement To Include Its Uncertainty?

?

First, what is a First Principle? Google quotes Aristotle: “a first principle is the first thing from which a thing is known”. Where these principles already exist, scientists, learners and teachers have some huge advantages compared to areas where First Principles still elude us.

?

In contrast, I still contend that there is no First Principle in our vicinity that can guide us when we try to explore, improve or maybe even eliminate the Measurement Uncertainty that we always find somewhere within any measurement. This First Principle vacuum goes a long way to explain the weird messiness of this fascinating field of Metrology.

?

To explain myself, I will compare the First Principle situations faced by great Scientists like Isaac Newton or Albert Einstein with those faced by measurement people today.

?

Isaac Newton had a huge advantage over modern Metrologists when he started the work to discover a Law Of Gravitation. For one thing, his field was wide open. Wide open in the sense that no one had discovered a First Principle ahead of him. There was little if anything for him to start with or deduce from. Newtons work lead him to craft a First Principle in the form of a beautiful mathematical formula to describe the way that matter interacts with itself to produce the effect that we call “gravity”.

?

Please note that First Principles don’t get their name because we think that they are correct for all time, but from the fundamental level at which they attack a question. That is why it’s not particularly useful to say that Newton was wrong, and Einstein was right. Einstein’s work contained Newtonian mechanics and explained it within a deeper framework. Perhaps someone living today will contain Einstein in the same way that Einstein contained Newton. If they are already alive, the chances are that they are on their phone this very minute!

?

What about Metrology on the other hand?

?

Since Newton didn’t specify a limit to his equation, we can take his formula to apply across the Universe, instantaneously, and to all matter. In fundamental contrast, Measurement, as far as we know so far, is a strictly Human activity. I freely admit that the only even vaguely interesting thing about Measurement as an activity is the study and characterization of the uncertainty that must always attach itself to all measurements. For me, we should consign the rest of it to a herd of robots, ASAP.

?

We only have to take the shortest glance at the history of Uncertainty to see its “ad hoc” character. Start your search anywhere along a line extending several centuries into the past and you the same thing over and over again: people guessing first and then later going back and creating tools because there was nothing useful already there. Oh, and arguing a ton the whole time! This is a classic picture of an area that has, as yet, no First Principle to guide it.

?

Here's more concrete proof: The ISO GUM (Guide to the Estimation of Measurement Uncertainty), published in 1993. If, in 1993, there were already a First Principle covering Measurement Uncertainty, like Newton covered gravity, the GUM would not have ever been necessary, in the first place. You either follow a First Principle, or you don’t, but you shouldn’t ever need to “interpret” it, like the GUM does, because if you can, you aren’t quite down at the base yet.

?

And more proof: criticism that the GUM has changed from “100% Frequentist to 100% Bayesian” just shows there has never been a First Principle to solidly anchor Uncertainty analysis and prevent any sloppy migration between two quite different analytical viewpoints. You cannot drift in the presence of a First Principle, you have to follow it, or else you have to yank it out by the roots!

?

?

Why Have I posted this piece?

?

This horse seems to be in a temporary state of submission so I will desist. I intend what I have written here to serve for as a navigation point, especially for newcomers! ?I want you to understand why it seems to you that no one present, including all of your mentors(should you be lucky enough to locate them), seem to have a whole lot of strong anchor points to start from when they try to explain why we do what we do. ?When I refer to “those present” I also include your measurement customers, the people for whom you calibrate, measure or adjust “stuff”. You have probably already seen that most of them struggle right along with your measurement mentors.

?

?

Measurement Uncertainty: At the moment, what we can be the surest about is that all measurements contain a measurement uncertainty element of “unknown magnitude and sign” because we have no perfect measuring instruments with which to render Uncertainty-free Measurements. That just doesn’t sound quite like a First Principle to me. Actually, it sounds a lot more like an excuse.

?

Thank you for sticking with me once more!