The ontological autonomy of the chemical world

Olimpia Lombardi a - Mart�n Labarca b

a CONICET – Universidad Nacional de Quilmes Universidad Aut�noma de Madrid

b Universidad Nacional de Quilmes – Universidad Nacional de San Mart�n



In the problem of the relationship between chemistry and physics, many authors take for granted the ontological reduction of the chemical world to the world of physics.  The autonomy of chemistry is usually defended on the basis of the failure of epistemological reduction: not all chemical concepts and laws can be derived from the theoretical framework of physics.  The main aim of this paper is to argue that this line of argumentation is not strong enough for eliminate the idea of a hierarchical dependence of chemistry with respect to physics.  The rejection of the secondary position of chemistry and the defense of the legitimacy of the philosophy of chemistry require a radically different philosophical perspective that denies not only epistemological reduction but also ontological reduction.  Only on the basis of a philosophically grounded ontological pluralism it is possible to accept the ontological autonomy of the chemical world and, with this, to reverse the traditional idea of the ‘superiority’ of physics in the context of natural sciences.


The ontological autonomy of the chemical world

Olimpia Lombardi a - Mart�n Labarca b

a CONICET – Universidad Nacional de Quilmes – Universidad Aut�noma de Madrid

b Universidad Nacional de Quilmes – Universidad Nacional de San Mart�n


    1.-  Introduction

At present, several disciplines devoted to the philosophical problems of particular sciences have branched off from the philosophy of science: in the literature we find philosophy of physics, philosophy of biology, philosophy of mathematics, etc.  However, the philosophy of chemistry is usually absent in this list.  In fact, there are very few journals specialized in the philosophical issues derived from chemistry, and philosophers of chemistry are generally confined to the sections corresponding to the philosophy of physical sciences in academic meetings.  This situation, clearly described by Scerri and McIntyre (1997), is particularly surprising in the light of the long history of chemistry as a scientific discipline and its relevant position in the contemporary context of natural sciences.  Why chemistry is ignored by philosophers?

    As many authors agree, this situation is due to the traditional assumption about the relationship between chemistry and physics: in spite of the fact that chemistry has historically evolved independently of physics, since the impressive success of quantum mechanics, it has been widely accepted that chemistry can be completely reduced to physics.  In other words, chemistry has been conceived as a branch of physics to the extent that it deals with complex systems or particular processes which, nevertheless, could ‘in principle’ be described and explained by means of quantum theory.  A famous statement of Dirac has become a commonplace in this line of thought: “The underlying physical laws necessary for the mathematical theory of a large part of physics and the whole of chemistry [are] completely known from quantum mechanics” (Dirac, 1929, p.714).  Dirac’s views rapidly spread out, being propagated by many authors from physics and philosophy (cf. van Brakel, 2000a, Chapter 5).  Of course, this traditional assumption not only deprives the philosophy of chemistry of legitimacy as a field of philosophical inquiry, but also counts against the autonomy of chemistry as a scientific discipline: whereas physics turns out to be a ‘fundamental’ science that describes reality in its deepest aspects, chemistry is conceived as a mere ‘phenomenological’ science, that only describes phenomena as they appear to us.  This supposed difference between both disciplines agrees with the traditional hierarchy of the natural sciences, rooted in the positivistic thought of the end of 19th century.  Due to its fundamental character, physics is at the top of the hierarchy, whereas chemistry is relegated to an inferior position, to the extent that it can be derived from fundamental physical laws.

    In recent years some authors have begun to challenge these traditional assumptions in order to recover the autonomy of chemistry and, as a consequence, the legitimacy of the philosophy of chemistry.  In general, the arguments are based on the discussion of the philosophical notion of reduction.  The autonomy of chemistry is usually defended on the basis of the failure of epistemological reduction: not all chemical concepts and laws can be derived from physics.  In particular, many relevant chemical notions, as chemical bond, chirality, molecular shape or orbital among others, are not amenable to rigorous quantum mechanical treatment.  Nevertheless, in the case of the relationship between chemistry and physics, ontological reduction is taken for granted: the ontological dependence of the chemical world on the physical world is usually presented as a thesis which does not need to be discussed.  This new perspective guarantees the methodological autonomy of chemistry with respect to physics, since it provides chemistry with a realm of specific concepts that have no place in fundamental physics.  However, a new question arises: Is the mere failure of epistemological reduction sufficient for rejecting the secondary position of chemistry with respect to physics?  When the ontological reduction of the chemical world is not questioned, the fundamental physical entities become the only ‘real’ entities and all the chemical concepts non derivable from quantum mechanics lose their referring character.  In other words, the physical reducing realm has ontological priority; chemical descriptions only refer to appearances or, at most, to secondary entities endowed with a merely derived existence.  As a consequence, the metaphysical picture supplied by ontological reduction leads to a secondary position of chemistry with respect to physics: whereas physical descriptions refer to the true and ultimate ontology, chemistry only describes apparent phenomena or, at best, an ontologically inferior level of reality.  Of course, this subordinate position of chemistry does not affect the everyday activity of chemists, since the methodological autonomy of their work is guaranteed by the failure of the epistemological reduction of chemistry.  But the ontological relationship between the chemical and the physical worlds is still a relevant philosophical question which has direct consequences with respect to the legitimacy of the philosophy of chemistry.  In fact, if chemical entities are no more than secondary entities ontologically derived from underlying physical entities, the philosophical problems of chemistry, when considered in depth, are problems belonging to the philosophy of physics.

    More recently some attempts have been directed at reversing the secondary character of chemistry by appealing to symmetrical relations between the discourses of chemistry and physics (van Brakel, 2000a, b) or to autonomous though related levels of reality (Scerri, 2000a, b).  In this paper we will argue that the rejection of the traditional hierarchy of natural sciences, where physics occupies the top position, and the defense of chemistry as a legitimate field of philosophical inquiry require to abandon not only epistemological reduction, but also ontological reduction.  Our starting point will be Putnam’s internalist realism, according to which ontology only arises as a synthesis between the noumenal reality and a conceptual scheme.  This philosophical position will lead us to an ontological pluralism which supplies the necessary foundations for supporting the ontological autonomy of the chemical world.  Once it is accepted that chemical concepts and laws refer to entities and regularities of a non-derived, non-secondary ontology, the traditional idea of the ‘superiority’ of physics with respect to chemistry can be abandoned and the philosophy of chemistry becomes an area of study in its own right.

    2.-  Ontological and epistemological reduction

When the problem at issue is reduction, the first step is to distinguish between ontological reduction and epistemological reduction.  Ontological reduction refers to the ontological dependence of the entities, properties and regularities of a stratum of reality upon the entities, properties and regularities of another stratum considered as ontologically fundamental.  Therefore, ontological reductionism is a metaphysical thesis that postulates the ontological priority of a certain level of reality to which all the other levels directly or indirectly reduce.  Epistemological reduction is concerned with the relationship between scientific theories: a theory can be reduced to another when it can be deduced from the latter.  Then, epistemological reductionism is an epistemological thesis according to which science can be or should be unified by deducing all scientific theories from a privileged one.

    Under the influence of the linguistic turn of 20th century, the traditional philosophy of science has been mainly interested in epistemological reduction.  The classical locus of the analysis of epistemological reduction is the book of Ernest Nagel (1961), The Structure of Science, where reduction is conceived as a logical relationship between theories: the reduced theory is a logical consequence of the reducing theory plus some definitions that connect the terms of the reduced theory with the terms of the reducing one.  The paradigmatic case of intertheoretic reduction is taken to be the reduction of thermodynamics to statistical mechanics: “the classical theory of heat has been reduced to statistical mechanics” (Kemeney and Oppenheim, 1956, p.7).  This assumption is what leads Nagel to consider that the term ‘temperature’ of thermodynamics can be reduced by definition to terms of statistical mechanics: temperature is identified with the mean value of the kinetic energy of the molecules of a gas (Nagel, 1961, Chap.11).

    In the last years, the applicability of Nagel’s model of intertheoretic reduction has been severely criticized.  Many authors have emphasized that there are no non-trivial and scientifically well-founded examples of epistemological reduction in the sense of Nagel (cf., e.g., Primas, 1983; Scerri and McIntyre, 1997).  In fact, the relationship between different theories is always rather more subtle than the traditional deductive view supposes: intertheoretic links usually involve limiting procedures, coarse-grainings, approximations and other mathematical techniques that are far more complex than the simple logical deduction proposed by Nagel (cf. Rohrlich, 1988, 1990).  Furthermore, recent works have shown that the analysis of the formal and conceptual relationships between theories gives rise to new fields of inquiry: there are complex phenomena that inhabit the asymptotic borderline between theories related by singular limits (Batterman, 2002).

    Summing up, on the basis of the distinction between ontological and epistemological questions regarding reduction, contemporary philosophers of science agree on rejecting the relevance of Nagel’s deductive model to the explanation of the links between scientific theories, even those belonging to a particular science as physics.  As we will see, this discussion reappears when the problem consists in elucidating the peculiar relationship between chemistry and physics.

    3.-  The usual defense of chemistry

The usual line of argumentation proposed by philosophers of chemistry to defend the autonomy of chemistry and the legitimacy of their own field of philosophical inquiry emphasizes the failure of the epistemological reduction of chemistry to physics.  Although the particular arguments differ from case to case, authors generally agree on considering that chemical concepts and descriptions cannot be derived from the concepts and the laws of physics as the epistemological reductionist supposes.  For instance, Vemulapalli and Byerly (1999) claim that epistemological reduction fails even in relatively simple cases: in general, the properties of a chemical system cannot be explained in terms of the properties of its physical micro-components; and even when the properties of a chemical macro-system can be derived from those micro-components, this requires additional assumptions related with macroscopic phenomena.  One of the situations considered by the authors is equilibrium in non-ideal multicomponent systems: although there exists a method for relating the properties of a system to the activities of its components, the numerical values of the individual activities must be derived empirically from experiments on the system or theoretically from postulated intermolecular forces or from other ad hoc hypotheses coming from the outside of the main body of the theory; in either case, they cannot be deduced from theories involving only the micro-components of the system.  Analogously, in the case of non-ideal systems in statistical thermodynamics, the equations of state used to estimate the energy of interaction between molecules cannot be deduced from any fundamental theory.  On the basis of these and other examples, Vemulapalli and Byerly (1999, p.37) conclude that: “Epistemological reduction fails radically since attempts to derive specific chemical explanations from fundamental physics [...] only succeed in deriving chemical results from assuming chemical data”.  In the same sense, Benfey (2000, p.198) asserts that: “There is an intrinsic limit in what physics can do in predicting the phenomena of chemistry. [....]  Only chemical data can indicate which quantum mechanical approximation is valid”.

    In the same line of thought, Scerri and McIntyre (1997) distinguish between ‘quantitative reduction’ and ‘conceptual reduction’.  Quantitative reduction refers to the calculation of chemical properties from physical theories, in particular, quantum mechanics.  This kind of reduction requires approximation techniques that can only be justified on a post hoc basis, that is, on the basis of the experimentally observed data which one is trying to calculate.  On the other hand, conceptual reduction is concerned with to the definition of chemical concepts in terms of physical concepts.  According to the authors, this form of reduction is impossible due to the very nature of the chemical concepts themselves: the concepts of composition, bonding or molecular structure cannot be expressed except at the chemical level.  As the result of the failure of both kinds of reduction, we should “eschew the epistemological reduction of chemistry to physics” (Scerri and McIntyre, 1997, p.220).

    Van Brakel (1997) addresses the traditionally alleged reduction of thermodynamics to statistical mechanics from a similar perspective.  He correctly points out that, in general, temperature cannot be defined as the mean molecular kinetic energy: this is true for perfect gases compounded of idealized ‘billiard-ball’ molecules in random motion, but not for solids, plasmas or a vacuum.  According to Van Brakel, all the problems for reduction seem to be related to the macroscopic notion of equilibrium, the central notion of thermodynamics: for instance, the macroscopic concept of temperature only makes sense for systems in equilibrium; but microscopically there is no such thing as equilibrium.

    Although at present there exists a widespread agreement between philosophers of chemistry about the failure of the epistemological reduction of chemistry to physics, almost nobody casts doubt on ontological reduction: chemical entities are, when analyzed in depth, no more than physical entities.  For instance, Vemulapalli and Byerly (1999) adopt a physicalist position according to which chemistry remains ontologically dependent on fundamental physics in spite of the fact that the properties of a chemical system cannot be effectively derived from physical properties: “Ontological reduction, in the sense of showing the dependence of all phenomena on constituent physical processes, has been a highly successful research program” (Vemulapalli and Byerly , 1999, p.18).  For the authors, emergence should be interpreted only in an epistemological sense: physicalism avoids the emergence of ‘dubious’ entities with no ontological support.  From a similar perspective, Scerri and McIntyre (1997, p.215) believe “the ontological dependence of chemistry on physics to be almost a foregone conclusion”; for them, the problem of reduction which need to be solved to preserve the autonomy of chemistry is an epistemological and not an ontological question.  In turn, Luisi (2002) replaces the distinction between ontological and epistemological reduction by the distinction between in principle and practical or effective deductibility: chemical properties are in principle deducible from physical properties; however, they cannot be effectively derived from the features of the physical level “because of technical difficulties, such as the lack of computational power or the insufficient progress of our skills” (Luisi, 2002, p.192).  In other words, although the properties of a chemical system are not deducible a posteriori from the properties of its physical components, a priori predictability cannot be denied without reintroducing a kind of unacceptable vitalism into science.  In a similar sense, Benfey (2000) recalls the second part of Dirac’s traditionally quoted assertion: “... the difficulty [for deriving chemistry from physics] is only that the exact application of these laws [of quantum mechanics] leads to equations much too complicated to be soluble” (Dirac, 1929, p.714).  Following Dirac’s claim, Benfey points out the intrinsic limit to the effective deduction of chemistry from physics: for instance, the mathematical computation needed to describe the electron density profile of a structure as simple as that of benzene would require a number of entities greater than the number of the fundamental particles in our universe.

    All these considerations show that there seems to exist a broad consensus among philosophers of chemistry about the relationship between chemistry and physics: epistemological reduction must be rejected, but ontological reduction cannot be denied; chemical properties are emergent only in an epistemological sense, since to accept ontological emergence amounts to populate the ontology with dubious metaphysical entities whose existence is not guaranteed by science.

    4.-  The ontological status of the chemical world

As we have seen, the discussions about the alleged reduction of chemistry to physics follow the traditional philosophy of science of 20th century in focusing on epistemological issues: the conclusion is that not all the chemical concepts and laws can be deduced from the concepts and laws of physics.  However, ontological questions usually do not go further than unquestioningly accepting ontological reduction.  As van Brakel says, “the question of reduction is tied to that of autonomy”; however, the issues addressed in the philosophy of chemistry generally “have little bearing on discussions about the ontological autonomy of chemistry” (van Brakel, 2000a, p.121. Italics in the original).  Our aim is to reverse this situation by analyzing ontological problems, that is, questions focused not on the epistemological reducibility of concepts and laws but on the ontological dependence of chemical entities and regularities on the entities and regularities of the physical world.

    Somebody could ask about the need of addressing ontological issues in the context of the relationship between chemistry and physics: Why are they relevant? What are we missing if we do not discuss them?  In fact, the failure of epistemological reduction is sufficient to guarantee the methodological autonomy of chemistry with respect to physics.  To the extent that there are specific chemical concepts that are not amenable to rigorous quantum mechanical treatment, chemists can continue with their work with no worry, since physicists will not replace them in the laboratories.  As Benfey (2000, p.199) says, “we will continue to carry out the work of chemists because no non-chemists are equipped to do it”.  However, when we ignore ontological issues, we miss an important philosophical question: Why is chemistry a ‘secondary’ science?  The answer to this question strongly depends on the assumption of ontological reduction: if the physical reducing realm has ontological priority on the chemical reduced world, the chemical concepts that are non-reducible to quantum mechanics refer to apparent or secondary entities endowed with a derived ontological status.  For instance, molecular shape turns out to be only a “powerful and illuminating metaphor” (Woolley, 1982, p.4).  Under this assumption, whereas physics describes the deepest and fundamental structure of reality, chemistry is a secondary science which studies ‘metaphorical’ entities not really existent.  Of course, this fact does not affect the work of chemists, but it has important philosophical consequences: the philosophy of chemistry, deprived of specific metaphysical problems concerned with real chemical entities, would be confined to methodological discussions about chemists’ activity.

    In recent years, two authors have directed their attention to ontological questions related to the referring character of chemical descriptions.  One of them is van Brakel (2000a, b), who discusses the reference of the physical and the chemical discourses under the paradigm of the mirror of nature: “Each mirror gives a different autonomous picture of (part of) the world, but one mirror the ideal physical one mirrors reality as it is (ontologically speaking).  All other mirrors [] picture mere appearances, without cosmic significance” (van Brakel, 2000b, p.169. Italics in the original).  According to the author, this paradigm should be abandoned by denying the asymmetric relationship between chemistry and physics: “The same event can have a chemical and a physical description, [] but no privileged description exists” (van Brakel, 2000b, p.171).  In fact, if quantum mechanics would turn out to be wrong, it would not affect the chemical knowledge about molecular shape, bonding or chirality.  As a consequence, the relationships between chemistry and quantum mechanics “are best seen as symmetrical relations” (van Brakel, 2000b, p.177).  Furthermore, van Brakel stresses the fact that the failure of epistemological reduction implies nothing about the interpretation of chemical concepts and their relation to what is ‘real’.  If there is no privileged description, chemical and quantum-mechanical concepts may be both ‘powerful and illuminating metaphors’.  On this basis, van Brakel concludes that “we could be tolerant enough to leave equal ontological room for manifest water, water in terms of the thermodynamic theory of substances, the molecular structure of water (‘constructed’ out of spectroscopic measurements), the ‘proper’ quantum mechanical equations for an isolated water molecule, and experiments with isolated water molecules which, depending on the measurement technique, show more of less of the ‘classical’ molecular structure” (van Brakel, 2000b, pp.147-148. Italics in the original).

    An even more interesting case of this new interest in ontological issues is the position of Scerri about the interpretation of the concept of orbital.  Scerri admits that, under the assumption of ontological reduction, terms as ‘orbital’ or ‘molecular shape’ “philosophically speaking, are nonreferring terms” (Scerri, 2000b, p.51).  But chemists are often realists in the sense that they believe in orbitals as though they were real and concrete entities; chemists and chemical educators show a great reluctance to abandon that realistic interpretation in spite of the theoretical pronouncements of philosophers of science, who claim that orbitals do no correspond to anything in the real world.  In his first works about this subject, Scerri (1991) advocated for the rejection of the realistic interpretation of orbitals because it is not possible to obtain exact solutions of the Schr�dinger equation in many-electron systems, and this fact represents a clear example of the failure of the epistemological reduction of the orbital concept.  However, later he has revised his original views, claiming that the chemists’ realism may not be such a bad thing.  In fact, in more recent articles, Scerri (2000a, b) proposes an intermediate position between realism and ontological reductionism, which leads to the autonomy of chemistry as the result of a form of liberation of the ‘physics imperialism’.  According to this new view, the interpretation of scientific terms is theory-contextual: “Not only is the question of the realistic as opposed to the anti-realistic interpretation of the orbital concept contextual with regards to whether one considers chemistry or theoretical physics, but even within chemistry it emerges that practitioners in different sub-fields generally adopt opposite interpretations” (Scerri, 2000a, p.421).  In other words, there is no single ontology to which all the scientific knowledge refers.  On the contrary, each science, and even each theory, operates on its own ontological level, where the entities and regularities referred to by the theory may be legitimately considered as real: there is no contradiction in conceiving orbitals as existent entities at the chemical level but not real in the quantum mechanical world.  It is in this sense that Scerri argues for the view of “autonomous though related levels of reality” (Scerri, 2000a, p.412), in terms of which the autonomy of the secondary sciences can be consistently defended.

    It is quite clear that, behind the positions of van Brakel and Scerri we can identify an implicit rejection of the ontological reduction of chemistry to physics.  Both views seem to tend towards an ontological pluralism which would permit chemical and physical entities to be existent with the same degree of reality.  In the next sections we will argue that Putnam’s internalist realism provides a fruitful philosophical framework for addressing the problem of the relationship between the physical and the chemical worlds.  The internalist perspective leads to an ontological pluralism that preserves the autonomy of the chemical world but avoids the intrusion of phantomlike entities or properties lacking any scientific support.

    5.-  Internalist realism

Putnam begins his book Reason, Truth and History (1981) by claiming that the acceptance of the objective-subjective dichotomy leads to the unavoidable choice between two mutually exclusive alternatives:

  • the adoption of the ‘copy’ theory of truth, which presupposes the existence of an objective world, completely independent of human mind.  Such a world admits the ‘One True Theory’ as the only description.
  • the rejection of the ‘copy’ theory of truth, which leads to a relativistic perspective: systems of thought, ideologies and even scientific theories are considered as subjective.

Putnam’s starting point is the rejection of the traditional objective-subjective dichotomy.  He tries to find a middle way between metaphysical realism and radical relativism, and usually calls his position ‘internalism’ or ‘internalist realism’, as contrasted with externalism or metaphysical realism also referred to as “God's Eye point of view” (Putnam, 1981, p.49).

    According to externalism, the world exists independently of our knowledge and theories, and consists of some fixed totality of mind-independent objects.  Then, there is only one true and complete description of ‘the way the world is’, whose truth involves some sort of correspondence between words and objects and between sentences and facts.  By supposing that reference is a relationship between the words of the language and the external, independent ontology, externalism needs a non-human point of view God's Eye to fix the reference of the language and, with this, the truth-value of its sentences.

    As P�rez Ransanz (1999) argues, the key for understanding the disagreements between externalism and internalism is the concept of object.  For the internalist, “’objects’ do not exist independently of conceptual schemes.  We cut up the world into objects when we introduce one or another scheme of description” (Putnam, 1981, p.52).  The question ‘What objects does the world consist of?’ only makes sense to be asked within a theory or description.  In other words, objects depend on conceptual schemes in a strong sense, which includes existence.  This means that conceptual schemes are not mere intermediary elements between subjects and objects; rather, they play an essential role in the constitution of objects.  Therefore, even though there is a reality independent of the subject, ontology only arises from a conceptual scheme.  Objects resulting from the synthesis between each conceptual scheme and the independent reality are the only objects.  Objectivity does not mean independence from the subject, but results from our conceptual schemes applied to reality.  This is an objectivity for us, but it is the only possible objectivity when we have given up the idea of God's point of view.  In other words, from this perspective rooted in Kantian philosophy, the reality independent of the subject is a noumenal realm, which is not object of our knowledge.  It is nonsense to think about an independent ontology, since there are no entities in the noumenal reality: as in Kant’s philosophy, the noumena is ineffable, it can only be conceived as a limit of thought.  This means that the ontology resulting from the synthesis between a conceptual scheme and the noumenal realm cannot be interpreted as an epistemologized ontology resulting from our contingent means of access to reality, in the same sense as Kantian system is not an epistemology but a broad philosophical framework that establishes the necessary conditions for knowledge and, therefore, for any meaningful scientific discourse.

    To assume that different conceptual schemes define different ontologies leads to the thesis of ontological pluralism, a central feature of Putnam's internalism.  Ontological pluralism points in the same direction as the underdetermination of theories by evidence, according to which there can exist incompatible theories that explain the same set of phenomena: we may have two empirically indistinguishable scientific theories that refer to completely different ontologies.  Both theses have been used against scientific externalism, which assumes that science converges to the true description of the real ontology the One True Theory.  But although Putnam’s internalism has been widely discussed in the context of the problem of theory change, in general it has not been applied in a synchronic sense.  Nevertheless, when we realize that different theories or even different disciplines are accepted at the same historical time and in the context of a single paradigm, we must also admit that different ontologies may coexist since each one of them is constituted by its corresponding theory.  And to the extent that the different accepted theories cannot be linked by epistemological reduction, there is no reason to believe that one ontology has metaphysical priority over the others.  Since the privileged viewpoint of God’s Eye does not exist, there is not a single ‘true’ ontology: all ontologies have the same metaphysical status because all of them are constituted by equally objective descriptions.  Once again, they are not mere ‘epistemologized’ ontologies as opposed to the ‘real’ ontology: when there is no single ‘real’ ontology, the very expression ‘epistemologized ontology’ loses any content.  As a consequence, the so-called ‘fundamental’ theories cannot be conceived yet as describing, even in an approximate way, reality as it is in itself, since the noumenal reality does not constitute an ontology to be described by science.

    In previous works, one of us has applied this philosophical framework to traditional problems of the philosophy of physics.  For instance, the internalist perspective has been used to deal with the question of the compatibility between deterministic and indeterministic descriptions in highly unstable systems (Lombardi, 2002), and to explain the objectivity of the classical description that arises as the result of the classical limit of quantum mechanics (Castagnino and Lombardi, 2004).  But it seems quite clear that similar arguments can be transferred to the problem of the relationships between chemistry and physics.  As we have seen, philosophers of chemistry reject epistemological reduction but accept the ontological reduction of chemistry to physics.  This position is implicitly based on an externalist or metaphysical realism according to which there exists a single ‘true’ ontology: all the entities, properties and regularities not belonging to that ontology have a ‘secondary’ existence, which depends on and, then, can be ontologically reduced to the fundamental entities, properties and regularities of the only real ontology.  Some authors explicitly express this position; for instance, Ramsey (1997, p.239) suggests the need of rethinking the notion of level because “[U]ltimately, there is only one level of reality”.  Since the ontological priority of the world of physics is taken for granted, the entities and properties of chemistry have only a derived existence.  This assumption amounts to accept that, whereas physics studies the only objective ontology that is, reality as it is in itself, chemistry merely describes a metaphorical or, at best, an ontologically derived realm.  But this traditional position can be abandoned from a consistent internalist perspective.  When the idea of God’s point of view is given up, it is not yet possible to conceive the description of reality in itself: even quantum theory corresponds to a particular conceptual scheme that constitutes the quantum ontology.  On the other hand, chemistry involves its own conceptual framework and, as a consequence, refers to its own ontology.  If chemical concepts could be epistemologically reduced to quantum concepts, there would be a good reason to believe in the ontological reduction of the chemical world to the ontology of quantum mechanics.  But when the epistemological irreducibility of chemistry is accepted, there is no argument other than metaphysical realism for postulating ontological reduction.  Summing up, from the internalist viewpoint, chemical concepts like composition, chemical bonding, molecular shape and orbital refer to entities belonging to the chemical ontology, which only depends on the theory that constitutes it and does not derive from an ontologically more fundamental level of reality.  In this way, we can fulfill van Brakel’s desideratum of leaving equal ontological room for water in all its scientific meanings.  And we can also give philosophical support to Scerri’s autonomous though related levels of reality, in some of which orbitals exist.

    It is worth stressing that the philosophical framework supplied by internalism can be applied not only to chemistry, but also and by means of analogous arguments, to all the other secondary sciences, as biology, psychology, etc.  Furthermore, in a recent paper internalist realism has been proposed as a non-dualistic response to the traditional mind-matter problem (Lombardi, 2002).  However, strictly speaking, ontologies are theory-dependent and, therefore, different ontologies may coexist in the context of a single scientific discipline.  For instance, when this philosophical approach was used to explain the objectivity of the classical description arising from the classical limit of quantum mechanics (Castagnino and Lombardi, 2004), the problem consisted in the relationship between the classical ontology and the quantum ontology, both described by physical theories.  As van Brakel correctly points out, if the relationship between chemistry and physics is considered, “more than two discourses, not only ‘chemical’ and ‘physical’, need to be distinguished: there is manifest water, physico-chemical, chemical and biochemical macroproperties of water, there is (chemical) molecular microdiscourse and (physical) quantum mechanical discourse” (van Brakel, 2000a, p.122).  Internalist realism can account for the objective realms referred to by all these discourses corresponding to the different theories of chemistry and physics.

    6.-  Objective relationships between ontologies

Up to this point we have emphasized the fact that internalist realism allows us to conceive a diversified phenomenal reality, where different theories cut their own ontologies out of the same noumenal substratum.  But this does not mean that the coexisting ontologies are completely disconnected in the sense that they are not related to each other in any way.  The question is how to conceive the relationships between these coexisting ontologies.

    As we have seen, from the traditional philosophical viewpoint reduction is a logical relationship; the links between theories are logical connections, analytical definitions or contingent conditions.  This implies that the reduced theory adds no nomological content to the reducing theory.  This view is perfectly consistent with the assumption of a single objective description of reality, which is, of course, the description given by the reducing theory.  Therefore, emergent properties introduced by the reduced theory are not really properties.  In other words, emergent predicates have no ontological reference; they are only economical ways of saying what we could have said, in principle, without them.

    This traditional concept of reduction is strongly rejected by Primas (1998) by arguing that reduction is not a logical relationship.  On the basis of several examples coming from physics and chemistry, Primas shows the need of identifying with precision all the auxiliary conditions required for one theory to be rigorously and completely deduced from another theory.  These auxiliary conditions are the context that defines the validity domain of the second theory.  In a mathematically formalized theory, a context can be introduced by choosing a new coarser contextual topology, compatible with the finer topology of the more basic theory.  In such topologies, certain asymptotic limits give rise to emergent properties, unavailable in the finer description: “In this mathematically precise sense, one can speak of the emergence of novelty in descriptions of a higher level” (Primas, 1998, p.86).  Therefore, the specification of the context is at least as important for intertheoretic connection as the laws of the finer theory.  Primas introduces the notion of contextual ontologies as the reference of context-dependent theories: contextual objects and emergent properties are the elements of such higher level ontologies.

    Atmanspacher and Kronz (1998) discuss Primas’ ideas as an example of their notion of relative onticity an extension of Quine's ontological relativity, according to which the contextual dependence between two levels of description can be shifted to cover the entire hierarchy of complex systems.  This means that a theory derived from a more basic one may become a reducing theory with respect to a higher-level theory if the new context is properly specified.  From this point of view, the many contextual ontologies with their corresponding emergent properties may be hierarchically organized in growing levels of complexity: “The problem is, then, to use the proper level of description for a given context, and to develop and explore well-defined relations between different levels" (Atmanspacher, 2002, p.52).

    These ideas are very appealing from the internalist viewpoint because they both seem to advocate for the non-uniqueness of ontology and for the view that objects and properties are always relative to a theory or conceptual scheme.  However, there is a relevant difference between them and the internalist approach.  The notions of contextual ontology Primas and relative onticity Atmanspacher and Kronz are based on the distinction between ontic and epistemic descriptions, first introduced by Scheibe (1973) in the framework of quantum mechanics.  Ontic descriptions refer to things ‘as they really are’, independently of any epistemic context; epistemic descriptions refer to our partial knowledge of the properties of a system.  Context-dependent theories are the epistemic descriptions where emergent properties arise (for more details, cf. Atmanspacher, 2002).  From an internalist perspective, on the contrary, there is no way of describing things ‘as they really are’: we have always access to reality through our conceptual schemes; there is no noumenal ontology.  If we have given up God’s point of view, there is no privileged perspective, and all descriptions have the same degree of objectivity all ontologies are equally real.  Therefore, we cannot accept that the so-called fundamental theories describe reality as it is in itself that is, the noumenal reality since these theories are also conceptual schemes through which relative ontologies arise. 

    But, then, how to conceive the relationships between theory-relative ontologies and, correspondingly, between different theories from the internalist perspective?  Primas is right when he rejects the traditional idea of logical reduction, but this is not sufficient for explaining the links between theories leading to different ontologies.  Such an explanation requires to make a decision about the status of the auxiliary conditions necessary for the effective derivation of one of the theories from the other.  Primas and Atmanspacher choose to suppose that the context required for intertheoretic connection consists of purely conventional conditions.  It is in this sense that Atmanspacher speaks of “contexts or contingent conditions” (Atmanspacher, 2002, p.52).  Then, the context is what allows us to retain what we consider as relevant and to disregard the elements considered as non-essential for our purposes: “The context of an operational description is given by the distinction between what we consider as relevant and what as irrelevant for a particular experiment or observation” (Primas, 1998, p.83).  In other words, to pass from a certain level of description to another involves an abstraction; that is, we neglect some irrelevant aspects and “ontologize” contextual objects and properties “as if they were actually existing” (Primas, 1998, p.96. Italics in the original).  Therefore, the epistemic description refers to our partial knowledge of the ontology described by the ontic theory.  Although useful, this partial knowledge only reflects some aspects of the basic ontology.

    The question is: can the ontic-epistemic distinction account for the objectivity of emergent properties?  It seems quite clear that, to the extent that emergent properties belong to the epistemic level, they are mere constructions based on our partial knowledge of the real properties of a system and, therefore, they are nothing else than useful but dispensable descriptive tools: emergence is not ontological, but just a matter of description.  This position agrees with Vemulapalli and Byerly’s view (1999), according to which emergence must be understood only in an epistemological sense.  In other words, the ontic-epistemic distinction cannot justify that emergent properties have the same ontological status as the so-called ‘fundamental’ properties, since it hides a position that is not far from traditional metaphysical externalist realism with its adoption of an only ‘true’ fundamental ontology.

    In order to defend the autonomy of relative ontologies and, as a consequence, the objectivity of emergent properties, it is necessary to stress that the elements required for interconnecting different theories are neither contingent nor conventional factors, but have a nomological content.  This implies laying aside the traditional epistemological view according to which the nomological content of science is concentrated at the laws of the theories, whereas contextual and auxiliary conditions represent the contingent aspects of each situation (cf. Wilson, 1989).  From this new perspective, the links between theories do not play a role as passive as the traditional view suggests but, rather, they express an essential part of our scientific knowledge.  In other words, the context required for intertheoretic connection is not a mere convenient tool to describe some aspects of a single ontology in response to our practical purposes or to our perceptual or technological limitations, as Primas and Atmanspacher suggest; on the contrary, such a context has a nomological character.  The idea can be expressed in an intuitive way by saying that, besides the ‘horizontal’ laws of theories, there are also ‘vertical’ intertheoretic laws connecting different theories, and which refer to the objective relationships between relative but objective ontologies.  From this philosophical viewpoint, we live in a diversified phenomenal reality, organized in multiple ontological levels, each one of which is ontologically and, therefore, non-trivially interconnected with the other levels (see figure).  Furthermore, we can explain the ontological emergence of novelty in that diversified reality: emergent properties are not mere constructions reducible to a single ontology; they are objective properties with the same ontological status as fundamental properties and reflect the diversified structure of our reality.


Diversified Phenomenal






                  Theory 1

                  Theory 2



Ontology 1

Ontology 2






    This non-reductive picture of a stratified reality is the philosophical framework that allows us to defend the ontological autonomy of the chemical world.  Since each conceptual scheme constitutes a relative ontology when it cuts its own entities and properties out of the same noumenal substratum, the ontology of chemistry is as theory-dependent as quantum ontology: both have the same degree of objectivity to the extent that the privileged viewpoint of God’s Eye does not exist.  The traditional acceptance of the ontological dependence of the world of chemistry upon the world of physics relies on metaphysical realism and its fruitless seeking for the only ‘real’ ontology, as well as on the unjustified belief that quantum mechanics describes, at least approximately, reality in itself.  Once we admit that the reference of each theory is partially constituted by the theory itself and that this is the only meaningful notion of ontology, the illusion of the ultimate discovery of an independent reality vanishes.  As a consequence, chemical ontology does not depend on a more fundamental ontology, but only on the conceptual framework that constitutes it.  The fact that the chemical world does not ontologically reduces to the world of physics does not mean that both ontologies are completely isolated from each other; on the contrary, they are interconnected by non-reductive, nomological links which permit the existence of objective relationships between both worlds but preserve the ontological autonomy of each one of them.  On this basis, chemical emergent properties like bonding, molecular shape or chirality do not need yet to be referred to physical properties in order to acquire ontological legitimacy: they are properties belonging to the chemical world, and their objectivity does not depend on the possibility of being ontologically reduced to the supposedly more basic properties of physics.  Only by accepting the autonomy of the chemical world, the well-known causal efficiency of chemical properties can be explained without turning them into illusory features of metaphorical concepts.

    7.-  Conclusions

The question of the relationship between chemistry and physics is a philosophical problem and, as such, it must be addressed by means philosophical arguments.  In this paper we have argued that, although the failure of the epistemological reduction of chemistry to physics can preserve the autonomy of the chemists’ everyday activity, it is not sufficient for eliminating the traditional conception of chemistry as a secondary, merely phenomenological discipline: only the rejection of ontological reduction can reverse such a traditional assumption.

    The main aim of this paper has been to present a philosophical framework, inspired by Putnam’s internalist realism, that permits the coexistence of different but equally objective theory-dependent ontologies interconnected by nomological, non-reductive relationships.  On the basis of this framework, the ontological autonomy of the chemical world can be justified on strong philosophical grounds.  In turn, this autonomy is what places chemistry in the same hierarchical position as physics within the context of natural sciences.  When chemistry begins to be conceived not as a secondary field devoted to study secondary and derived entities, but as a scientific discipline referred to an ontologically autonomous realm, it can also acquire legitimacy as an area of philosophical concern: chemical problems deserve philosophical attention in their own right since they cannot be reduced to the issues traditionally discussed in the philosophy of physics. 


We are grateful to two anonymous referees for their very constructive comments.  This paper was supported by grants of CONICET, of the Universidad Nacional de Quilmes and of the Secretar�a de Estado de Educaci�n y Universidades de Espa�a.




    Atmanspacher, H. (2002), “Determinism is Ontic, Determinability is Epistemic”, in H. Atmanspacher and R. Bishop (eds.), Between Chance and Choice: Interdisciplinary Perspectives on Determinism, Imprint-Academic, Thorverton, England, pp.49-74.

    Atmanspacher, H. and Kronz, F. (1998), “Many Realisms”, in G. Farre and T. Oksala (eds.), Acta Polytechnica Scandinavica, 91, pp.31-43.

    Batterman, R. (2002), The Devil in the Details. Asymptotic Reasoning in Explanation, Reduction, and Emergence, Oxford University Press, Oxford.

    Benfey, T. (2000), “Reflections on the Philosophy of Chemistry and a Rallying Call for Our Discipline”, Foundations of Chemistry, 2, pp.195-205.

    Castagnino, M. and Lombardi, O. (2004), “Self-Induced Decoherence: A New Approach” Studies in History and Philosophy of Modern Physics, 35, pp.73-107.

    Dirac, P. A. M. (1929), “Quantum Mechanics of Many-Electron Systems”, Proceedings of the Royal Society, A338, pp.714-733.

    Kemeney, J. G. and Oppenheim, P. (1956), “On Reduction”, Philosophical Studies, 7, pp.6-19.

    Lombardi, O. (2002), “Determinism, Internalism and Objectivity”, in H. Atmanspacher and R. Bishop (eds.), Between Chance and Choice: Interdisciplinary Perspectives on Determinism, Imprint-Academic, Thorverton, England, pp.75-87.

    Luisi, P. L. (2002), “Emergence in Chemistry: Chemistry as the Embodiment of Emergence”, Foundations of Chemistry, 4, pp.183-200.

    Nagel, E. (1961), The Structure of Science, Harcourt, Brace & World, New York.

    P�rez Ransanz, A. R. (1999), Kuhn y el Cambio Cient�fico, Fondo de Cultura Economica, Mexico.

    Primas, H. (1983), Chemistry, Quantum Mechanics and Reductionism, Springer, Berlin.

    Primas, H. (1998), “Emergence in Exact Natural Sciences”, in G. Farre and T. Oksala (eds.), Acta Polytechnica Scandinavica, 91, pp.83-98.

    Putnam, H. (1981), Reason, Truth and History, Cambridge University Press, Cambridge.

    Ramsey, J (1997), “Molecular Shape, Reduction, Explanation and Approximate Concepts”, Synthese, 111, pp.233-251.

    Rohrlich, F. (1988), “Pluralistic Ontology and Theory Reduction in the Physical Sciences”, British Journal for the Philosophy of Science, 39, pp.295-312.

    Rohrlich, F. (1990), “There is Good Physics in Theory Reduction”, Foundations of Physics, 20, pp.1399-1412.

    Scerri, E. (1991), “Electronic Configurations, Quantum Mechanics and Reduction”, British Journal for the Philosophy of Science, 42, pp.309-325.

    Scerri, E. (2000a), “The Failure of Reduction and How to Resist Disunity of the Sciences in the Context of Chemical Education”, Science & Education, 9, pp.405-425.

    Scerri, E. (2000b), “Realism, Reduction and the ‘Intermediate Position’”, in Bhushan, N. and Rosenfeld, S. (eds.), Of Minds and Molecules. New Philosophical Perspectives on Chemistry, Oxford University Press, New York, pp.51-72.

    Scerri, E. and McIntyre, L. (1997), “The Case for the Philosophy of Chemistry”, Synthese, 111, pp.213-232.

    Scheibe, E (1973), The Logical Analysis of Quantum Mechanics, Pergamon Press, Oxford.

    van Brakel, J. (1997), “Chemistry as the Science of the Transformation of Substances”, Synthese, 111, pp.253-282.

    van Brakel, J. (2000a), Philosophy of Chemistry. Between the Manifest and the Scientific Image, Leuven University Press, Leuven.

    van Brakel, J. (2000b), “The Nature of Chemical Substances”, in Bhushan, N. and Rosenfeld, S. (eds.), Of Minds and Molecules. New Philosophical Perspectives on Chemistry, Oxford University Press, New York, pp.162-184.

    Vemulapalli, G. K. and Byerly, H. (1999), “Remnants of Reductionism”, Foundations of Chemistry, 1, pp.17-41.

    Wilson, M. (1989), “John Earman’s: A Primer on Determinism”, Philosophy of Science, 56, pp.502-532.

    Woolley, R. (1982), “Natural Optical Activity and the Molecular Hypothesis”, Structure and Bonding, 52, pp.1-35. 



Recent Documents:

Set Home | Add to Favorites

All Rights Reserved Powered by Free Document Search and Download

Copyright © 2011
This site does not host pdf,doc,ppt,xls,rtf,txt files all document are the property of their respective owners. complaint#downhi.com