It was nothing to find a glaring contradiction in the thought of a major philosopher. In fact, it's the easiest thing in the world. As I explained to my thesis adviser - who knew Popper and was impressed with my thesis - the goal was to move from this reassessment of the structure and nature of Popper's thought to a consideration of Popper and Kuhn, as well as Paul Feyerabend and Imre Lakatos, on what science is.
In The Copernican Revolution, one long-standing issue with which Kuhn deals is what, precisely, was at stake as astronomy moved from pre-Copernican to Copernican views of the heavens. One thing most assuredly was not even a question, nor would have been understood to be such, was whether or not the Ptolemaic cosmology offered a more accurate representation of the Universe. It was a particular conceit, particularly in the 18th century, to consider Copernicus' De Revolutionibus nothing less than a complete shift not only in astronomy, but in the way we human beings understand the universe and our place in it. Actually reading Copernicus, however, should disabuse anyone of such an idea.
Since Ptolemy, astronomy had been a thoroughly quantitative science; by the early 16th century, it had developed to the point that the equations used to describe the movement of the planets against the background of "fixed stars" might not have been recognizable, yet was clearly within the parameters set by Ptolemy. The increasing complexity of the equations involved was a result of attempts to account for observed variations in the purported regular motion. Of acute concern was what is known as the retrogression of Mars. As Mars moves across the sky, at irregular intervals, it appears to move backward against the fixed stars. Other issues, from slight variations in the periodicity of Mercury to similar, yet less frequent and small, retrogressions of Jupiter and Saturn, had led to the creation of what were called "cycles" and then, added to these, "epicycles", sometimes several piled upon one another as repeated observations invalidated previous attempts to resolve observed irregularities.
In their simplest terms, the cycles and epicycles were mathematical representations of variations on the assumed circular motion of the planets within their "spheres".
Rooted in a Platonized Pythagorean dedication to the "perfection" of the heavenly realm, this insistence on understanding the movement of the planets within "spheres" that traced perfect circles led astronomers to include calculations in which the spheres appeared to circle round particular foci, then the center of these alternate movements also became foci for further "epicycles". Copernicus offered a couple simple suggestions. The mounting number and complexity of cycles and epicycles was clearly far too complex to make the construction of tables of planetary motion - needed for everything from calculating the planting and harvesting of crops through regularizing calendrical conventions to calculating certain holidays - so Copernicus suggested shifting the earth from the center to a position in orbit around the new center, the sun. In so doing, dozens of cycles and epicycles dropped away, and Copernicus offered the promise of more precise calculation in the future.
At no time had Copernicus suggested that the Sun was the actual physical center of the solar system or universe. Indeed, had such a notion been ascribed to him (the work was published posthumously, in large part because it took him so long to redo the new calculations), it would have been unintelligible to him. Astronomy had nothing to do with the structure of the heavens. It was a way to calculate the movement of pinpoints of light human beings used to determine the passing of the seasons and years; the cosmos was as God created it, with the heavenly realm a place of perfection, with God - not this or that heavenly sphere - at the center.
It took over a century for the full implications of Copernicus humble suggestion for clearing away the rubble from mathematical astronomy to be felt.
Along the way there was the not unimportant fact that the promise of better predictive results from streamlining away dozens of those epicycles didn't pan out. In fact, reverting to earlier tables, with the earth at the center and all those circles around circles around circles, rendered more accurate observable results, at least in the case of Jupiter and Saturn, than Copernicus' innovation. Furthermore, most astronomers were wary of the innovation, not because it violated some theological or philosophical conviction, but for the far more mundane reason that people don't like innovation.
At no time during the century and a half of the working out of Copernicus' theory did anyone consider that the developing theory of the structure of the heavenly spheres (as they continued to be called) prove either that the previous view was "wrong" in that it did not correspond to the way the universe was "really"; nor did they believe the new discoveries demonstrated that the developing science of astronomy now had a more accurate "picture" of the universe. This popular notion - that the Copernican Revolution succeeded because it offered a more accurate representation of the "real" structure and shape of the solar system - is mistaken for one simple reason: no one actually involved in the controversies of the century and a half from Copernicus afterward would have understood the issues in those terms.
The broader implications, for Kuhn's narrative, should be clear enough. Elaborating on them in The Structure of Scientific Revolutions, Kuhn makes the exact same point Popper had made: Science is not about representing to us the way the world "is". Rather, science is the construction of mathematical and verbal models, which themselves exist within interlocking, contingent frames of reference (here, Kuhn fell back upon the idea from psychology of Gestalt) for which there exists no outside arbitrator to decide how accurately they represent "the world". Language, it seems, even the mathematical sciences, do not "represent" anything; science uses the tools available, including language, broader social understandings, and the various rules of social interaction and discourse, to make clear, in terms that are always contingent and sometimes unintelligible across the barriers of language, society, and time, how we understand the world.
This description of science is little different from Popper's. The difference between the two men - and Popper's essay in Criticism and The Growth of Knowledge is one of the most condescending pieces of writing from one scholar to another I've read - was Popper's assumption that the rules of logic and mathematical rationality were constants that were the bones upon which the contingent structures of science hung. Kuhn, on the other hand, while certainly not disparaging rationality, made the point over and over again, that this was never a matter under dispute. His was not an attack on science or human reason. It was, rather, a description of the scientific enterprise that made clear something in which both men believed: Science does not represent "the world".
There is no way for human beings to represent with anything like accuracy or even approximation "the world" precisely because "the world" is an open system; science, limited by the fact that it is a human construct with all sorts of factors both within language itself as well as extraneous factors, clarifies particular matters of interpreting and understanding the world without ever making the category error that it, somehow, "describes" the "world".
I offered this as the topic I wanted to explore for my dissertation, and my thesis adviser told me to go for it. The only reason I didn't entered the world on July 10, 1997. The happiest decision I ever made was turning my back of the above because I had this beautiful tiny bundle of new person for which to care.