Stratification, Emergence and Reduction
The idea of reality as layered, or stratified, can be taken further. Once a science has discovered mechanisms which constitute the level, or aspect of reality, which it has as its subject-matter, it is then possible to ask what deeper-level mechanisms are responsible for those and so on.
So, for example, physiological mechanisms explain many of the characteristic activities of animals and plants - metabolism, reproduction, respiration and so on. But these physiological mechanisms can themselves be explained in terms of the chemistry of the complex organic molecules of which living organisms are composed. There is no obvious stopping point to this process of probing evermore deeply into the microstructure of the natural world. To the extent that the different scientific disciplines are each concerned with a particular level of reality, they, too, can be ordered into a hierarchy of levels. So, we might put the sciences into an ordering something like this:social sciences
psychology
physiology/anatomy
organic chemistry/biological chemistry
physical chemistry physics
This way of ordering the sciences could be justified in terms of the way the mechanisms characteristic of each level are explicable in terms of those of the next one below it. This corresponds to a view of science as explaining wholes in terms of the parts of which they are composed. It will be noticed that several sciences don’t seem to fit into this hierarchy at all. These include ecology, geography, meteorology, oceanography, palaeontology and so on, and we will consider them in the next section.
Although the general idea of sciences uncovering layers of a stratified reality is quite widely shared, there remain many issues which divide critical realists among themselves, and also divide them from other realist approaches. There is space here to deal with only two of these issues.
One has to do with the way we think about the relationship between levels. If the mechanisms of one level are held to explain those at a higher level, then it might seem that as soon as the lower-level science has been established it can replace the higher-level one: that physiology becomes redundant when biochemistry is developed, for example. This, in some versions, leads to the view that ultimately all sciences will be reduced to the basic laws of physics (see the sections on theoretical entities and the hypothetico-deductive account of theories in Chapter 3). Much more influential versions are concerned with the explanation of human mental and social life in terms of physiology, genetics or natural selection.Such ‘reductionist’ interpretations of the layering of reality are opposed by critical realism. There are three main reasons why reductionism does not work. First, the lower- level science explains (at best) only the constitution of the mechanisms at the higher level. It does not explain when or with what effects the powers established with those mechanisms will be exercised. So, for example, having the right anatomy and physiology (hearing, vocal organs, appropriately constructed cerebral cortex and so on) may explain why humans have the power of speech (which, for example, closely related primates do not), but it does not explain when and how any particular human will learn to speak, which language she will learn or what she will say.
Second, once higher-level mechanisms are formed, their activities have effects on lower-level ones. For example, an emotional trauma can have effects on the interactions of the central nervous and endocrine systems, so altering the rate of the chemical reactions involved in respiration, the rate of blood supply to muscles and triggering anaerobic chemical reactions at the level of cells and tissues. So, causality can flow down the hierarchy as well as up it. It follows (on the critical realist equation of ‘being real’ with ‘having effects’) that the mechanisms constituted at each level have their own specific reality.
It also follows that the sciences of the lower-level mechanisms can contribute to explaining, but never completely explain, the behaviour of the higher-level mechanisms.Finally, the association between levels and particular sciences is in part explained by the way entities at higher levels have properties and powers not predictable in advance on the basis of properties of lower-level entities. The macromolecules we call genes play a part in explaining the characteristics of the living organisms in whose cells they reside, but unless we had first encountered living organisms (including ourselves) we would have no idea what properties of living things were explicable in terms of them. Again, it is reasonable to believe that human consciousness depends on a certain level of complexity in the organization and functioning of the brain and central nervous system, but ‘consciousness’ could not have been predicted from even the most sophisticated neurophysiology unless we first knew what it was to be conscious. Indeed, since all knowledge depends on the activities of conscious beings, this higher level has to be presupposed in the discovery of the lower and more fundamental levels. As we saw in Chapter 3, the formation of qualitatively new properties or powers at each level of organization is termed ‘emergence’, and critical realists are often committed to some form of ‘emergent powers materialism’.
The second area of controversy among realists concerning the relationships between different levels focuses on the higher levels. In particular, it concerns the relationship between the human biological, the psychological and the social, and so we will return to it in a later section of this chapter. For now, it should be noted that the hierarchy of sciences listed above puts the social at the top, the psychological below it and the biological below them. This reflects the common-sense view that bodily organization and functioning is the basis of psychological mechanisms, and that society is formed by the conscious actions of individual people.
However, it can be argued against this that at least some psychological mechanisms (for example, whatever mechanisms enable us to form sentences according to the grammatical rules of our own language) can only be explained in terms of social processes (in this case, the learning of our own language). This is the sort of argument that a follower of Durkheim or Saussure might advance. On that basis, psychology should be above sociology in the hierarchy of the sciences, since we explain psychological processes in terms of social ones. Society is not created by the conscious decisions of individual people, but pre-exists them and moulds their mental life.
Equally, however, it seems clear that unless human beings had certain innate psychological capacities and dispositions (for example, to learn some language) it would not be possible for society to shape them in the way it does. Considerations like this point to a branching hierarchy, in which psychology and the social sciences would figure at the same level, with mutual interdependence of the mechanisms described by each. This would be consistent with contemporary views of human evolution, according to which modern humans and their distinctive patterns of sociability evolved together (for contrasting views on this issue, see New 1994, 1996: ch. 2; Collier 1994a: ch. 4).
However, a view of human social life as involving only human individuals, and involving them only in virtue of their mental activities, would be drastically impoverished. Central features of human societies, the way they organize production of food, clothing and shelter, the relationships through which they regulate sexuality, reproduction and child-rearing, and so on, are all unintelligible unless we understand humans as embodied beings, with organic-functional needs and vulnerabilities. So, we are essentially social beings through our embodiment, not just our mental lives. But in noting this we are also committed to an understanding of the relationships which constitute society as binding embodied humans to other living and non-living beings: to physical spaces, raw materials, tools and machines, domesticated and wild animals and plants, agricultural and semi-natural ecosystems, buildings, highways and so on. All of these things and relationships are produced, reproduced or transformed as elements in the overall metabolism of society. It follows that society cannot reasonably be represented as a single level in the hierarchy. Rather, it is a heterogeneous complex of mechanisms and processes constituted by the combination of mechanisms drawn from several of the other levels: psychological, physiological/anatomical, ecological, chemical and so on. We will return to this view of the nature of social life later in this chapter.