INTRODUCTION

Teleology and Teleonomy items are widely used in science and philosophy. If a scientific theory implies teleological factors, to some extent, it would be considered as unmodern. It is difficult for specific scientific practice to eliminate teleology or reduce it to a non-teleological explanation. The dilemma of teleology lies in the fact that it is asymmetric on the uniformity of interpretive principles with respect to non-teleology. Non-teleology keeps monism of causation on each step of interpretation process, however, teleology of scientific theory goes in a different direction, that is, teleology partially yields to monism and would not shift it to pluralism until the key step comes up. Occam's Razor opposes the theories committing infinite regress of entities, and denies the explanations yielding to alternative interpretive principles. The teleological form of explanation proposed by Nagel is generally accepted, from which any further analysis comes to a conclusion that teleology implies more imformation than that of non-teleology. The book mainly takes as Nagel's form of teleology, and extends its concept to a broader notion by the boundary of Mayr's types of teleology. The extended one is an explanation of generalized teleology. It makes sense in dealing with large amounts of teleological factors in modern science historically or ideologically. The theoretical tool comes from the premise of Nagel's model of reduction, that is, the properties in explanan and explanandum are homogeneous. More strictly, explanatory reduction lies on the conceivability of homogeneity of explanan and explanandum. On the contrary, emergent explanation is realized.

The key characteristic of emergence is the novelty of hierarchical properties. Whether teleology could be further explained by emergentism depends on the novelty of teleonomies on each hierarchy. The book will analyse Anderson's theory to show the condition of novelty of hierarchical properties, that is, whether the key characteristics of properties named by the same item are suddenly truncated or separated in some middle level. Besides, according to S-R model, if a large number of repeatable facts show that two types of events occur continuous enough spatio-temporally,it can be explained that there exists causation between them. The scientific evidence is overwhelming that emergence is usually accompanied by the breaking of symmetric. Teleonomy also goes hand in hand with the interchange of symmetric and asymmetric between its level and the adjacent lower one. It can be concluded that there surely exists causation between teleonomy and emergence. Moreover, emergence interpretation of quantum measurements provides explanations of influences on non-teleological sub-systems from a teleological one. Finally, to be a reasonable explanation, emergence thoery itself still faces its own problems. Kim holds that emergence is denied by three problems, that is, overdetermination, causal difference of multiple realizers, and individuation of levels. The book gives a response to Kim in the three aspects.

Teleology commits backward causation which is the influence on behavior in past from state in future. The explanatory foundation comes from time asymmetric. How is emergent explanation of the generalized teleology possible? The key to the argument lies in whether that foundation is reductive. Nagel shows three types of homogeneity in explanatory reduction, the third one of which could be named the type of statistical stipulation. It depends on the multiple realizability and emergent explanation. The book will propose the concept of granularity which means the degree of coarse-gaining. It is exemplified by the method of history of science that it leads to hidden wrong conclusions if it is lack of discussion on granularity difference between premise and reasoning process in thought experiments when it involves levels. That is to say, the equivalence of conceivability and consistency by logic or possibility by laws of nature is not universal. It hardly denies the existence of emergence in a system, nor proves the homogeneity of properties in explanan and explanandum. The concept of granularity fits on the respond to Nagel's third type of explanatory reduction, that is, the multiple realizability and hierarchy-reciprocating leads to emergent explanation. Multiple-realizability is considered as one of the judgement conditions on emergence. Hierarchy-reciprocating shows the non-monotonicity of leves in defining concepts determined by hierarchy, that is, the concept involes the events or properties in a lower level and implies some constraint from an upper one. Essentially, it yields to downward causation epistemologically. Penrose proposes a new notion of entropy. On account of multiple realizability and hierarchy-reciprocating, Penrose entropy is a tricky one to be identified as explanatory reduction.It inevitably needs interpretation from emergence theory.

In a word, by the means of conceptual analysis, thought experiment and history of science, we try to propose a model of explanation by emergence theory, in order to illustrate teleology and teleonomy and to solve the dilemma of teleological explanation. On this basis, people's misconception on explanatory reduction will be reversed. It will be revealed to everyone that granularity is able to be the boundary condition of thoughts and it determines why emergent explanation of generalized teleology is possible.

KEY WORDS: Explanatory Reduction, Multiple-realizability, Coarse-gaining, Backward Causation, Conceivability, Thought Experiment, Quantum Measurement, Thermodynamic Entropy