From the very beginning of the first appearance of intelligent being on earth, the struggle to understand the world has never stopped. The world is changing, due to our influence on it, as well as our ideas about it. To perceive the world from different perspectives is just like gazing at the life of the same blossom of four seasons, blooming and fading, as we understand the same very flower with different respect. To understand the theory change in science, it is yet essential to understand what theories are made of, or what theories are represented by. Here in this essay, I‟ll introduce different aspects of understanding concerning theories, mainly based on Kuhn‟s theory, as well as some following philosophical ideas, and the revolution of human conception of the world and beings.

Kuhn, a very respectable figure in the philosophy of science, has given an idea of symbolizing scientific theories by paradigms, a terminology typically of Kuhnian style. According to Kuhn, a paradigm is a package of claims about the world, methods for gathering and analyzing data, and habits of scientific thought and action. A literal understanding of this should imply a story like this: the very first day an intelligent human being (well, a human being should be intelligent, otherwise it‟s not a human being, carry on!) started to consider the environment around him, and began to investigate and invent ideas due to any motive whatsoever, knowledge about the world started to accumulate, until a significant amount of them was found by a certain group of people, with organized thoughts about the relations around and in between them. A pre-paradigm (a terminology quoted from Kuhn, describing a certain stage at which a naively organized group of knowledge, claims etc. starts to form a set of principles, which set a basis for successive paradigms) is formed, which is neither well-organized nor effective in explaining observed and potentially observed reality. Such story went on and on until the group of human beings were able to work out the inner relationship between the claims, and were able to explain old and new claims using the fundamental ones together with the habit and logic of thoughts. A paradigm was formed. However, a paradigm cannot be formed by any single idea, accumulated together with surrounding ideas. The outburst of a paradigm, according to Godfrey-Smith, is triggered by a certain influential achievement, together with surroundings and followers, e.g. Newtonian mechanics.

Kuhn categorized paradigm into two classes: the broad and the narrow. In the broad sense, a paradigm is the group of observations, principles and claims about the entire ecology of existence, whilst in the narrow sense, a single significant discovery can qualify to be a paradigm, e.g. Copernicus‟s claim about the solar system. And once a paradigm is formed, it coordinates the work of individuals into an efficient collective enterprise. Thus, a paradigm requires the agreement on basic fundamental claims about the world, otherwise there‟ll be a forever ongoing fight over the basics, and people will never settle down to reach an agreement over what‟s „legal and permitted‟ in the reign of science. However, such an agreement restricted the discussion of paradigms to the narrow sense, since if the broad is changed, the paradigm must change too, since the broad is about the entire human understanding of the entire world, regardless of the division of classical mechanics, relativity or quantum mechanics. Coming back to the main road, when such agreement is formed, the paradigm enters into the era of normal science, which is work inspired by a striking achievement that provides a basis for further work, in which scientists agree on the fundamentals, and extend and refine the paradigms.

Then, how do progresses occur in science? It is with great sense of certainty to say that every discarded , even some not-yet-discarded, scientific theory has been proved to have anomalies in them. Therefore, anomalies should be something of great interest in our investigation of scientific progress. It is almost certainly the case that when normal science goes through time, anomalies occur. And when such anomalies occur, scientist normally don‟t give up their paradigms or theories right away, as a metaphor quoted from Godfrey-Smith: only a poor workman blames his tool. It is with great probability that scientists begin to investigate the anomaly to see whether the paradigm is compatible with the anomaly, which means the paradigm is able to explain the anomaly as a special case. And it is with great significance to do so, since, according to Kuhn, if science does treat these constantly arising anomalies as refutations, and if scientists dropped their paradigms every time a problem arose, there‟ll be no further development of paradigms or normal science, but only the chaos of different claims about fundamentals. And it is the balance between being too resistant to anomalies and not being resistant enough that makes science work and progress.

However, it is with huge possibility that sometimes anomalies accumulate to a certain degree, which, employing Kuhn‟s terminology, reaches a stage of critical mass. Here and then a crisis to a paradigm has arisen. However the point here is to what extent the mass has to accumulate to qualify as a crisis? 1000? 10000? And since human beings has the nature of being lazy, if nothing else comes up, people may as well stay with the old paradigm and treat the 10000000000 (God knows how many) anomalies as special cases. Thus it seems that only the critical mass of anomalies cannot give rise to the revolution of paradigms. And the point was discovered by Kuhn, who has added another point to the condition for scientific revolution, which is the appearance of a rival paradigm. The two factors constitute the way in which revolutions occur.

Here I have to bring the human nature of laziness again, together with the nature of the favor of reminiscence. If a rival paradigm does occur, why to people favor the new to the original, there must be some reason for people to choose the new one, apart from being fashionable (which may as well be a „good‟ reason for some, who I‟m not going to argue with). And Kuhn himself has interestingly enough posed the incommensurability theory, which well defies such possibility of comparing two paradigms in the same field. According to Kuhn, there’re two reasons why different paradigms are incommensurable. The first one is that people in different paradigms are unable to communicate to each other to compare, since the key terms are used in different ways, or in other words, they speak different “languages”. This claim should no concern us too much since what we‟re looking for are the observables referred to in reality, rather than differences in terminology. Therefore if scientists in different paradigms are able to describe and refer to real existence and observables, then we have no problem. Just like people speaking different natural languages can understand each other, people in different paradigms can communicate with each other once the “translation” is done. The second one is that people in different paradigms use different standards or parameters to measure their paradigms. Such different standards can be different concepts of justification, for instance, a single peer of a person‟s head can be the justification of his crime in one paradigm, while in another paradigm it cannot be; and different conception of causation and so on. The clearest example being the wave-particle fight, whilst in Newtonian mechanics, the universe is understood as being constituted by particles and positions whilst in quantum mechanics, the universe is understood as being wave-distributed matters. Therefore the second should really deserve our concern.

As a first-aid, here I quote a short paragraph from Godfrey-Smith‟s Theory and Reality, which is a description of an idea of Kuhn:

“Science has a special kind of efficiency, and this efficiency results in a real form of progress across revolutions. The progress is measured in problem-solving power; the number and precision of solutions to problems in a scientific field tend to grow over time.

“If our later paradigms have more overall problem-solving power then earlier ones, then it seems that we are entitled to regard the later ones as superior.”

Here the problem solving power refers to the precision and number of problems a certain paradigm can settle. It is obvious enough that people should favor the paradigm which settles 1000 questions to the one which settles 999 questions, and the one which answers the question to a further degree to the one to a less further degree. Hence the solution to the previous problem seems to be found in this. Suppose we prefer quantum mechanics to classical mechanics as to explain the world, we believe then the quantum mechanics theories have more problem-solving power. And here by applying the problem-solving power, the standard problem can be well avoided, since what we are looking for is not the power of a problem dealing with problems in a SAME method, but rather the power of solving the problem using whatever methods available.
At the end of this essay, I would like to spend a short time on two other philosophers concerning the same problem, influenced by Kuhn. The first one is Lakatos, who found Kuhnian paradigm destructive prone to chaos.

With Analogous to Kuhn‟s „paradigm‟ paradigm, Lakatos used the terminology of Research Program. One and not the only difference is that in Lakatos‟s Research Program, there‟s usually more than one program per field at any given time. And according to Lakatos, it is the competition between research programs that we actually find in science, which is also essential to rationality and progress. A research program is constituted by two parts: the hard core, which is the basic idea, with analogy to the fundamentals in Kuhnian paradigms; and the protective belt, which are less fundamental ideas to apply the hard core to actual phenomena. And the progress in science is the expansion of protective belt, the vivid picture of which is Indiana Jones whipping around himself to mark his own region.

Another similar approach to scientific revolution is Laudan‟s Research Tradition, which is considered superior then Lakatos‟s Research Program. In Laudan‟s theory, acceptance of a theory and pursuit of it are differentiated. To accept something is to treat it as true, which is determined by the problem-solving power at a very moment; and to pursue something involves deciding to work with an idea, which is determined by the rate of change of problem-solving power, or d(problem-solving power)/dt. Therefore one may pursue something without accepting it, but merely suppose that it might be true.

However, I don‟t see much difference the latter two theories have make, either in terms of problem-solving power to the revolution problem or substantial change in concepts about scientific revolution. And Kuhn‟s „paradigm‟ paradigm does supply a good explanation of revolution.