Emergence and Reduction in Morphogenetic Theories

Mariano Artigas
Lima, 1989
Published in E. Agazzi y A. Cordero (editors), "Philosophy and the Origin and Evolution of the Universe" (Dordrecht-Boston-London: Kluwer Academic Publishers, 1991), pp. 253-262.

The origin of the universe and mankind are the two limiting cases within the evolutionary worldview, whose main task consists in providing morphogenetic theories that may explain how new levels emerge out of other more basic ones. In this context, problems about emergence and reduction occupy a central place.

The following reflections refer, first of all, to some difficulties that are found in the already classical analysis of reductionism, and suggest that the problem of reduction may be properly replaced by the analysis of the relations between levels. These reflections are afterwards applied to the examination of some morphogenetic theories, and finally to the problem of ontological emergence, including an evaluation of some ideas about the origin of the universe and mankind.

1. The possibilities of reduction

Nowadays there is a general agreement about the limited value of classical analysis of the problem of reductionism. On the one hand, this analysis depends too much on the deductive model of scientific explanation, which has been improved by emphasizing the central role of conceptual aspects in science. Furthermore, this model does not fit well with actual scientific practice.

Reductionism occupied a preeminent place in the neopositivist program, which aimed at unifying science through the reduction of its different branches to a physicalist language. Although this ideal was soon given up, reduction continued to be treated as logical deduction of laws or theories of the reduced or secondary science out of those of the reducing or primary one. In this line of thought a distinction was made between two types of reduction, namely that called homogeneous or unproblematic, when all terms of the secondary science were already present in the primary one, and other called discontinuous or problematic, when there was at least one new term which was found only in the secondary science. In this second case, the bridge to relate both sciences could be established by means of a logical connection between the terms, or by postulating coordinating definitions, or through factual hypotheses (Nagel [1961], pp. 336-397).

Afterwards some precise arguments were introduced in the discussions, that pointed towards an increasing scepticism about the possibilities of deductive reductions. With respect to homogeneous theories, it was pointed out that, as a matter of fact, reductions by strict logical derivation seldom occurred or that they were even impossible in practice, due to the fact that, when one or several laws are explained by a new theory, the meaning of some basic terms usually change. It was admitted that a weak or instrumentalist reduction was possible, but this kind of reduction was a merely partial and approximate coincidence of results. A strong reduction between heterogeneous theories would also be possible, so that the reduced theory could be retained and even better corroborated; this would happen when some correlational laws are empirically established, or when two classes of entities in both theories are identified. The conclusion was that it would be convenient to substitute the very term reduction by that of quasi-reduction or partial explanation, to stress the difficulties of a full reductionism (Sklar [1967]; Friedman [1982]).

That those points of view are generally accepted could be ascertained at the XIIIth International Conference on the Unity of Science, whose central topic was the problem of reduction and emergence in the main scientific disciplines. There it was stressed that, if actual scientific practice is properly taken into account, what reduction really means is establishing partial connections between different epistemological levels, and it was pointed out at the same time that those connections may adopt very different modalities (Radnitzky [1988]).

What all this means is that it may be safely be said that derivational reductionism corresponds to a philosophical ideal that neither fits well with actual scientific progress nor is practically feasible. Theoretic constructs are formulated with the purpose of solving particular problems in specific scientific areas, and are constructed according to the conceptual and instrumental resources that are available in every moment in time; it seems therefore convenient to replace the traditional problem of reduction between theories by that of establishing relations between fields or areas of investigation, so that epistemological analysis should be focused on concrete relations, that admit a great variety, between problems and solutions that overlap in the development of actual investigation (Darden - Maull [1977]; Artigas [1989], pp. 5-110).

2. Morphogenetic theories

If it is assumed that a morphonegetic theory is one that relates different levels, its possibility will depend on the existence of ordered levels. From a global and quasi-intuitive point of view, these levels are those of physics, chemistry, biology and human sciences, and the problem consists in relating the superior and inferior levels. This problem may be considered from a dynamic point of view, if genetic relations about the origins are considered, or otherwise from a static one; both perspectives illuminate and complement each other.

Difficulties already arise when we intend to establish hierarchical links between the first two levels, those of physics and chemistry. We need to define from the beginning our ideas about the structure of theories and the relations between them. Moreover, every discipline and even every theory is centered on specific classes of problems and use its own intellectual and instrumental resources. Progress towards unification does not eliminate totally such differences, that arise from the limits of our knowledge and from the respective need of adopting partial points of views according to the possibilities of our knowledge. All that may be summed up by saying that there are different epistemological levels (Kanitscheider [1988]), and even a plurality of ontologies that are implied by the different theories and usually are not replaced when some deeper theories are formulated (Rohrlich [1988]). Those reflections are already important when applied to the analysis of the very fist level of physical theories, and even more when we study the relations between physics and chemistry (Primas [1988]).

Much greater are the difficulties when we consider the relations between the physico-chemical and the biological level. Important results have been obtained, such as the knowledge of the structure of DNA and its connections with genetics, the explanation of some functions of hemoglobin through the knowledge of the sequence of aminoacids which determines the structures of the molecule, and the possibilities opened by the techniques of recombinant DNA for decoding the composition of genes and the structure of proteins.

These achievements point out towards the reduction of biology to physics and chemistry. Doubtless they reveal that physical and chemical laws can be applied to a great extent in the realm of living entities. As mechanics showed that celestial bodies obey the same physical laws that are valid for terrestrial bodies, so biochemistry and molecular biology show that living entities obey the same physics and chemistry that other bodies obey. Nevertheless, this does not mean that biology has been reduced to physical science. When we study biological problems we need to use specific concepts and techniques. Explanations expressed through physical sciences neither eliminate nor make useless the epistemological levels of biology (Kitcher [1982] y [1984]; Rosenberg [1985], pp. 69-120).

To exemplify this we may refer to four types of relations between the level of biology and that of physical sciences: the part-whole relations, such as the proof that genetic unities are a part of chromosomes; the explanation of the physical nature of an entity or a process: for instance, the biochemical explanation of the represors postulated in the operon theory; the relations between the structures of entities and processes and the functions adscribed to them: in that way the knowledge provided by chemical physics about the structure of molecules provides an understanding of their biochemical functions; and the relation cause-effect that exists, for instance, in the explanations provided by the theory of allosteric regulation(Darden - Maull [1977]). None of these relations is properly equivalent to a reduction.

The preceeding reflections refer to relations between theories that belong to the same level or to adjacent levels. Greater even are the difficulties when the results of one level are used to explain problems that belong to more distant levels. In this context reference may be made to some theories that usually are labelled as morphogenetic in a special sense, namely non-linear thermodynamics, synergetics, catastrophe theory and theories about chaos.

Non-linear thermodynamics shows that in principle biological processes may be compatible with the second law of thermodynamics: it suggests that in open systems, far from equilibrium, biological structures can arise through amplification of fluctuations that lead to a new state in which dissipative structures would be maintained. Doubtless, this theory is interesting from the morphogenetic point of view, because it points out that a state characterized by a certain degree of organization may be generated out of a less structured state. Nevertheless it can be said that in that case too there is neither logical reduction nor elimination of properly biological levels (Friedman [1982], pp. 28-39).

Synergetics, catastrophe theory and theories about chaos may be considered too as morphogenetic theories, in that they provide explanations about the genesis of new structures; moreover they intend to relate levels that are not only different, but even in many cases are far away from each other. This is why it is very difficult to establish in a rigorous way the general validity of the models that they provide. From the morphogenetic point of view, the main interest of those theories is an heuristic one, as they suggest the existence of similarities in patterns that may be found in different levels.

Finally, morphogenetic theories in a proper sense are those that refer to evolutionary processes. They have a very special epistemological status. By the one hand, there are arguments that support the existence of such evolutionary processes. But, on the other hand, it is difficult to establish with certainty their concrete mechanisms. This is due not only to the limits of our knowledge, but also to the unique character of the processes that must be described. In these circunstances it is evident that differences, even very deep ones, may arise between various explanations. In this context, the realization of those difficulties becomes a guarantee of progress; in fact, whenever partial explanations are considered as if they were complete, this tends to prevent the discovery of better explanations.

3. Ontological emergence

The partial character of epistemological reductions implies that there are limitations in the possibilities of establishing ontological reductionism. The complete explanation of upper ontological levels resorting only to more basic ones cannot be justified from the scientific point of view. Nevertheless, the quest for the unity of science stimulates searching for relations between the different levels and, in this respect, partial methodological reductionism finds its proper justification. An analogous effect may be attributed to the conviction about the unity of nature, which may be considered as one of the foundations of the scientific enterprise.

Of course, the first condition for the formulation of morphogenetic theories is the existence of the levels that are to be related. In this respect it may be pointed out the basic shortcoming of any physical explanation about the alleged self-creation of the universe (Atkins [1981]; Davies [1983]; Smith [1988]). The problem in this case is not only that quantum gravity is still in its beginnings, nor that there are difficulties in admitting the uncaused character of quantum fluctuations; the main problem is that, if such an explanation were possible, physical concepts should have a significance that transcends the possibilities of experimental method. In fact, physical concepts would not relate two epistemological or ontological levels: they should connect one real level with a non-existing one (Craig [1986]; Artigas [1987]; Carroll [1988]). Moreover, physical method does not allow one to establish that a concrete event has been the absolute beginning of the universe, even in the case that it really would have been [Jaki [1982], p. 260). The problem of the absolute origin of the universe must be formulated within a metaphysical context.

If the problem of the absolute origin of the universe surpasses the possibilities of the experimental method due to the inexistence of the basic level where this method must be grounded, a similar conclussion may be obtained with respect to the complete explanation of man. In this case, the existence of a specific human level is a necessary precondition for the very existence of science. In other words: experimental science cannot deny the peculiarity of man without denying, at the same time, the possibility of its own existence. The ability to formulate questions about the validity of knowledge is a prerequisite if critical argumentation is to have any meaning at all, and this ability can only exist in the context of a subjectivity that, even though is bounded by physical conditions, cannot be reduced to them.

Ontological reductionism, when formulated in the form of a scientistic naturalism or materialist monism, must resort to arguments of the type nothing but. Those arguments, however, lead to blind alleys. If one states that the universe or man is nothing but matter, he is obliged to determine what is to be understood as matter, and this is a task full of difficulties. In fact, sheer matter does not exist, because any material entity is organized and informed. Moreover, naturalism has no scientific or epistemological grounds, unless one denies the reality of the objects that cannot be studied by means of the methods of experimental science; however, this denial leads to a contradiction or vicious circle, and besides it prevents the study of some epistemological and ontological assumptions whose existence is a necessary condition for the possibility and validity of experimental science.

The evolutionary worldview is sometimes used to argue for a substantive monism, that is the thesis that all objects are in the last analysis different forms and manifestations of some original entities (Rohrlich [1988], pp. 297-299). This thesis, however, is a philosophical one, and cannot be justified by means of scientific or epistemological arguments.

Some problems posed by the evolutionary worldview can be classified under the problem of emergence. Of course, emergence can be partly understood by means of the mechanisms that underlie it. However, a deeper understanding transcends the explanations that are characteristic of the experimental method. To recognize this point one has only to consider that any scientific explanation should refer to the existence of laws that are found through all natural levels. Nature considered in its classical sense as an internal principle of activity is a basic prerequisite of science. Experimental science allows us to obtain an increasingly deeper knowledge and control of natural structures and processes; nevertheless, the very existence of natural activity poses problems whose philosophical aspects are not exhausted by scientific explanations.

References

Artigas, M. [1987], "Física y creación: el origen del universo", Scripta Theologica, 19, pp. 347-373.

Artigas, M. [1989], Filosofía de la ciencia experimental, Eunsa, Pamplona.

Atkins, P.W. [1981], The Creation, Freeman, New York.

Carroll, W.E. [1988], "Big Bang Cosmology, Quantum Tunneling from Nothing, and Creation", Laval théologique et philosophique, 44, pp. 59-75.

Craig, W.L. [1986]. "God, Creation and Mr Davies", The British Journal for the Philosophy of Science, 37, pp. 163-175.

Darden, L. - Maull, N. [1977], "Interfield Theories", Philosophy of Science, 44, pp. 43-64.

Davies, P. [1983], God and the New Physics, Dent, London.

Friedman, K. [1982], "Is Intertheoretic Reduction Feasible?", The British Journal for the Philosophy of Science, 33, pp. 17-40.

Jaki, S.L. [1982], "From Scientific Cosmology to a Created Universe", The Irish Astronomical Journal, 15, p. 260.

Kanitscheider, B. [1988], "Reduction and Emergence in the Unified Theories of Physics", in: Radnitzky, G., ed., Centripetal Forces in the Sciences, pp. 89-110.

Kitcher, P. [1982], "Genes", The British Journal for the Philosophy of Science, 33, pp. 337-359.

Kitcher, P. [1984], "1953 and All That. A Tale of Two Sciences", The Philosophical Review, 93, pp. 335-373.

Nagel, E. [1961], The Structure of Science, Routledge & Kegan Paul, London.

Primas, H. [1988], "Can We Reduce Chemistry to Physics?", in: Radnitzky, G., ed., Centripetal Forces in the Sciences, pp. 119-133 y 137-142.

Radnitzky, G., ed. [1988], Centripetal Forces in the Sciences, volume 2, Paragon House, New York.

Rohrlich, F. [1988], "Pluralistic Ontology and Theory Reduction in the Physical Sciences", The British Journal for the Philosophy of Science, 39, pp. 295-312.

Rosenberg, A. [1985], The Structure of Biological Science, Cambridge University Press, Cambridge.

Sklar, L. [1967], "Types of Inter-Theoretic Reduction", The British Journal for the Philosophy of Science, 18, pp. 109-124.

Smith, Q. [1988], "The Uncaused Beginning of the Universe", Philosophy of Science, 55, pp. 39-57.