Constructive and principle theories
The distinction between constructive and principle-based theories, according to Einstein, was
Constructive theories were assembled from existing toolkits of components, like Lego or Meccano, to solve problems. The constructive approach then let us adapt and reshape our tools to produce an arbitrarily-close emulation of the system that we wanted to model, and if the system turned out not to behave as we expected, then we had freedom to modify or change our tools to suit. This adaptability made the constructive approach ideal for engineering problems.
Principle theories were more purist. Instead of working out how to model the existing known behaviour using existing tools, the principle approach tried to extract the most fundamental rules possible for how the phenomena ought to behave, and then forced us to select tools that obeyed those principles. Principle theories could let us predict surprising new effects that have never been seen before, and their results were less open to negotiation - a result HAD to be correct, or else the principle and the ensuing theory were simply wrong. Principle theories have greater scientific credibility because of their minimalism, their dependence on a clear set of falsifiable rules and the increased scientific falsifiability of the resulting system.
The difficulty of the principle-based approach is that while almost any problem can be attacked using the constructive approach, new principles do not come along every day ... ad when they do, it is not always immediately obvious how we should best attempt to implement them. at their worst, principle-based theories can be high-minded but lacking in mathemtical machinery or experimentally testable predictions (e.g. Mach's principle before general relativity), and constructional theories can be "mere curve-fitting", where theorists constantly adjust that theory to agree with whatever new experimental data has been discovered, without having any deeper insights as to the mechanisms used by Nature (as opposed to the mechanisms of their mathematical tools).
We can distinguish various kinds of theories in physics. Most of them are constructive. They attempt to build up a picture of the more complex phenomena out of the materials of a relatively simple formal scheme from which they start out. Thus the kinetic theory of gases seeks to reduce mechanical, thermal, and diffusional processes to movements of molecules – i.e., to build them up out of the hypothesis of molecular motion. When we say that we have succeeded in understanding a group of natural processes, we invariably mean that a constructive theory has been found which covers the processes in question.
Along with this most important class of theories there exists a second, which I will call "principle-theories." These employ the analytic, not the synthetic, method. The elements which form their basis and starting-point are not hypothetically constructed but empirically discovered ones, general characteristics of natural processes, principles that give rise to mathematically formulated criteria which the separate processes or the theoretical representations of them have to satisfy, from the universally experienced fact that perpetual motion is impossible.
The advantages of the constructed theory are completeness, adaptability and clearness, those of the principle theory are logical perfection and security of the foundations.
The theory of relativity belongs to the latter class.
- Einstein's original 1916 general theory used a mixture of principle-based and constructive approaches – it used the general principle of relativity, but then also co-opted existing tools such as special relativity in order to construct its implementation of the general principle. Since SR was adopted for speed and convenience without proper checks as to whether it was compatible with the GPoR, this element of the 1916 theory was actually constructive. GR1916 was highly falsifiable, as was demonstrated in 1960 when it was falsified logically ("without a shot being fired"), without the need of experimentation.
- The 1960 reinvention of general relativity was a response to our realisation that the GPoR and SR were incompatible. Rather than start from scratch (the "total rewrite" option), we chose to retain the structure of GR1916, but downgrade the GPoR from a principle to a guideline. as a result, GR1960 was no longer strictly a principle theory, as it allowed its founding principle, the GPoR, to be overridden by the existing implementational behaviour – having decided that the theory didn't match its specifications, we decided that the theory was right but the specifications had been over-strict, and relaxed them. The resulting 1960 theory was now a constructive theory - easier to modify and change, but less falsifiable (presumably since there was no obvious way to reintroduce strict falsifiability without discarding SR, or the theory then immediately being once again logically falsified).
- Advanced General Relativity (AGR) reinstates general relativity as a fully principle-based theory (using the GPoR), the associated cost being the downgrading of special relativity from an exact physical limit to an approximation.