One of the interesting aspects of Elias’s work is the way he characterises the various sciences collectively as a ‘model of models’ that map on to the ‘great evolution’ and what this means for sociology as an autonomous discipline, in fact one of the sciences within the model of models, and what it says about the nature of sociological knowledge and implies for sociological research methods. By the ‘great evolution’ we are referring to the way that different material levels of integration emerge over planetary time such that each new level is dependent on the preconditions of the earlier level but, due to characteristics of the emergent level, each new level cannot be understood or explained purely in the terms of the level from which it emerged. So for example, the level of integration we call ‘life’ was necessarily preceded by the the physical and chemical levels but, once life emerged on earth, it cannot be explained (modelled scientifically) purely in terms of the features of physics and chemistry. What is more, once a new level has emerged it reacts back upon and alters the nature of its ‘host’ levels. Examples of this are the oxygenation of the atmosphere and the development of the pedosphere (basically the soil that the later forms of life depend upon), both of which are the results of the emergence of life. Another example is that of climatic change due to the emergence of life (the biosphere) long before the emergence of humankind and the resulting anthroposphere (that part, in fact most, of the biosphere that has been affected by the emergence and development of human groups and societies). What this implies is that, to some extent at least, a full understanding and explanation of an earlier level of integration requires some reference to the effects produced by the subsequent levels of emergence that have both changed the dynamics of and processes within that level. An example of this is the way that practically every aspect of the biosphere is now affected by and has been changed by its incorporation within the anthroposphere. This has had the effect of changing the context of the processes of evolution that the emergence of humankind was predated by and was dependent upon. Another example would be the changes in human biology and psychological functioning that have come about as the result of what Elias called ‘symbol emancipation’ and the development of culture. What is of crucial importance here is the fact that culture as a symbolic system becomes to a certain extent decoupled from material reality and takes on autonomous characteristics and possibilities and that it is external with respect to individuals. Cultural maps of the external environment and the ‘recipes’ for behaviour are very different to the internal ‘maps’ of the external environment that inform the somatic and instinctual behaviours that are internally constituted through blind evolutionary processes of adaptation. It is argued that at the very least an understanding of phenomenon at any one level of integration would need to take into the account of the two bracketing levels, those above and below the level of the phenomenon. Of course the number of levels of emergence and integration, what our scientific demarcation rules for establishing boundaries and how we label these is a significant problem. I understand that a living cell has within it about 20 levels. A human individual is biological, chemical and physical as well as social and cultural. As a sociologist the levels that need accounting for will depend upon the specific level in question. Much sociology does this already, at least implicitly, for instance C Wright Mills distinction between personal troubles and public issues and his claim that individual experience has to be related to both social institutions and their location in historical processes. This is also sometimes quite explicit in the methodological claims that society and social processes cannot be explained purely in terms of individuals’ conscious motivations and experience or their own understanding of their actions as the historical process and structural level of society are as much the product of unknown and unanticipated consequences of behaviour with a scope in time and place far beyond the subjective experience of actors. On the other hand the emergence, reproduction and development of society, social institutions and structures, cannot be explained without reference to the grounds of individuals’ behaviours and understandings.
One thing is made quite clear however, by the ‘three levels ‘approach to sociological understanding. A sociologist does not have to study physics in order to do sociology. Equally, a scientific understanding of society cannot be based upon or extrapolated from a knowledge of the physical and chemical levels of integration. In addition, different levels of integration will have different scientific theories and models. Mechanistic and mathematized models are inadequate for studying and representing society and social processes because of the nature of the reality being dealt with. The limits of quantitative and statistical models of society are not due to inadequate mathematical and statistical knowledge but because there is a mismatch between the representational and modeling capacity of mathematics and the ontology of the exponentially more complex levels of integration that are the the psychological and social. It seems that what we might call the representational or theoretical language in which a level of integration is described and modelled must be adequate to that level of integration. Even if it makes sense to say that mathematics is the appropriate language of the physical levels it doesn’t follow that it is for other levels. No doubt mathematization and quantification have a methodological part to play in sociology, but this is within limits and is only approprate for phenomneon that are amenable to mathmatical description. For instance some structural and patterned aspects of social processes can be represented to some extent mathematically, albeit only descriptively. But in the end we are reliant on words (as indeed are, in the final analysis, physicists).