There is an obviousness to the idea that the basic ontology of modern science is a material ontology, this raises the question of what a material ontology is, what it implies, and in what way it either is or isn’t a viable ontology for the sciences.
In looking at variations of material ontology the following requirements of a physical ontology are noted as being particularly relevant to modern science:
Part-Whole: One piece of matter can be part of another.
Additivity of Mass: The mass of the union of disjoint pieces of matter is the sum of their individual masses.
Rigid solid objects: A solid object moves continuously and maintains a rigid shape. Two objects do not overlap.
Continuous motion of fluids: The shape of a body of fluid deforms continuously. Chemical reactions. One combination of chemicals can transform into a different combination of chemicals. Chemical reactions obey the law of definite proportion. Continuity at chemical reactions: In a chemical reaction, the products appear at exactly the same time and place as the reactants disappear.
Gas equilibrium: A gas in a stationary or slowly moving closed container quickly attains a state of equilibrium.
Gas laws: A body of gas in equilibrium in a closed container obeys the ideal gas law and the law of partial pressure.
Liquid in a cup: A body of liquid remains at rest at the bottomof a motionless cup. It can be carried without spilling if the cup is moved smoothly and slowly.
Availability of oxygen during combustion: If fuel is burned in air then (ordinarily) oxygen remains available at the point of combustion so that the process can continue.
Passivized surface of metals: In passivization, the surface atoms of a bar of pure metal combine with atmospheric oxygen to form the metallic oxide. The oxide is chemically inert and adheres to the metal, so the interior of the bar remains the pure metal. The problem is to represent a surface that is chemically different from the interior.
Leaving aside for the moment the question of how a purely material ontology can account for some of the questions raised in quantum mechanics, these appear to be the basic requirements necessary to achieve a material ontology that suffices for the sciences.
The problem with this set of requirements (and other sets that could be constructed in a similar fashion so as to support the major assumptions of various sciences) is that a number of the requirements appear to common sense ontology to be contradictory. In order to allow for these contradictions multiple simultaneous material ontologies are proposed:
Determining the true ontology of matter is a question for physicists; determining what ontologies are implicit in the way people think and talk is a question for cognitive scientists and linguists; determining the history of views of the true ontology is a question for historians of science. For those of us who are researching knowledge representations for intelligent systems, the question is, what ontology can be most easily and effectively used as the basis of a knowledge base that will be used for intelligent tasks? In the long run the answer almost certainly involves using multiple, mutually inconsistent ontologies, with ameta-level structure to choose between them or combine them when necessary. However, at present, such a system would be very hard to design or validate. If we wish to address the representation of matter without taking on the difficulties of inconsistency resolution, the question becomes, what single model is most effective as a basis for reasoning?
Ontologies and Representations of Matter, Ernest Davis
This appears fairly straightforward, each area of the sciences uses an ontology that is appropriate to their subject matter. However the very phrasing betrays the solution as begging the question. If only physics has “true” ontology as its basis, how precisely does the matter of other sciences render such a true ontology as at the very least inconvenient. What is the nature of “true” that is being invoked, where it applies to one area of experience and no others? As material ontologies that are dependent on a specific subject matter for their validity, is there even a potential for a “true” ontology, or is the ontology of physics simple another convenience that is based on its subject matter? Finally, if a material ontology must determine the meaning of “matter” in each case, how can it in turn be determined by the matter that the science has for its object of study?
The common sense answer is that we are simply playing word games with fundamentally different meanings of the word matter, however the word does not etymologically have multiple origins that might account for its use in these various ways. Another oddity that might have been noticed is that a given science has a subject matter that is simultaneously its object, and precisely a material object. How is it that scientists, of all people, put up with fundamental contradictions at the bases of their own specialties?
The pragmatic scientist might respond that these are all very interesting questions, no doubt, but have nothing to do with the reality of research as a practical concern. The difficulty with accepting this response is implicit in the very project Davis is undertaking. The various sciences, together, do not claim to be merely vehicles for useful information, pragmatic concerns equivalent to carpentry or software development. The claim of the science is intrinsically a truth-claim, that their results are knowledge, not merely information, and indeed are the arbiters of the validity of any knowledge obtained from other fields. Davis’ question: “ what ontology can be most easily and effectively used as the basis of a knowledge base that will be used for intelligent tasks?” is precisely the question of a universally valid ontology that can be used to interpret information of whatever type into knowledge, which simultaneously implies that a judgement of the truth of the information is part of the interpretation.
If the truth-claim of the sciences is to be upheld, the apparent contradictions in the various ontologies in practical use within the sciences have to be resolved. This however requires looking at how the apparent contradictions arose in the first place, on what assumptions they were based and whether those assumptions retain their own validity, and what implications for a clear, universally valid ontology they have.
The apparent contradictions surround the terms matter, material, subject and object. Obviously the first terms are crucial if the ontology is to be a material ontology, and the word matter does have a number of intrinsic meanings. Since those meanings originated out of an original univocity, it would be simplistic to write them off as circumstantial artifacts without inquiring into why that word in particular was seen as appropriate for what seem initially to be multiple unrelated things. The question of the apparent confusion in the terms subject and object, and what that implies for the corresponding notions of subjectivity and objectivity are also obviously relevant to the development of an ontology that can support the universally valid truth-claim desired by Davis, and required if the truth-claim of the sciences is to be in any way upheld. If the “knowledge” of the various sciences is based on interpretations that themselves do not agree, the knowledge-claim of the sciences devolves to a claim to pragmatically useful interpretations of information that was itself gathered via research based on pragmatic concerns.
Before we ask the question “what is matter?” and then go on to the question of a possible material ontology, we need to deal with the question of what the word itself is intended to mean in a pre-ontological sense.
The original usage of the word, oddly, was synonymous with ‘subject’ in its sense of the topic, i.e. the ‘matter at hand’, the way we still refer in grammar to the subject of a sentence. Thus in its original meaning the phrase ‘subject matter’ is redundant. The verbal sense of “to matter”, i.e. to be of concern, arose almost simultaneously. Yet it took more than two centuries before the Greek word “hyle”, which was used to refer to something physical, became another meaning of the word, through its synonymity with subject, and therefore a translation of the Latin equivalent, “substantia”. Prior to this usage by translators from Greek and Latin texts, then, the basic pre-ontological intent of the term is something akin to ‘topic of concern’.
The apparent confusion between subject, as the topic at hand, and subject as the knower of the topic, and the subsequent substitution in various instances by the word object, arose in the Cartesian Meditations. The intent of the Cartesian Meditations was precisely to obtain a certainty of the validity of scientific endeavor, and in particular to obtain a certainty that would be founded on the certainty of mathematics. That one of the crucial terms in operation here changed its meaning directly as a result of an attempt to secure the knowledge-claim of science is definitely relevant to our own inquiry. Lastly, the term object originated from the Latin objectum, which meant to oppose, to put into opposition, much as we still use the term as a verb, “to object to (something”, or the later nominal form objection. The outcome of the Cartesian Meditations is that anything real (res in Latin) is known by being placed in opposition to the subject, which is now intending the “I” of the knower. The fundamental properties of any real, according to Descartes, are spatial and temporal extension, and therefore measurable. Over time measurability in general became seen as the fundamental determination of the real. It should be noted that real, in Descartes sense, does not imply actual. A unicorn is real, in the sense that has the necessary properties of a res. It’s lack of being actual is not relevant to its definition.
Since the object is determined as an object by being placed into relation with the subject (a relation specifically of opposition), the object and its objectivity are intrinsically dependent on the subject and its subjectivity. Since this relation determines the known as known, knowledge is also by definition relative to the subject as the I-Subject. It is in this basic ontological stance that the importance of the repeatable experiment becomes comprehensible as a feature of the truth-claim of the sciences, not to mention the hyper focus on measurable aspects within the sciences that deal largely with topics that are for the most part not easily subject to measurement. In the repetition of a given experiment by different I-Subjects, dismissing a given result as ‘merely’ subjective, i.e. only related to one I-Subject, is partially overcome. However no matter how many times an experiment is repeated, it remains possible that the next repetition will fail in some way, in this admission the relativity of scientific knowledge is maintained within science’s own definitions. That experiments are rarely repeated precisely in actual research, specifically because the exactly repeated experiment is not particularly useful other than to guarantee a level of ontological validity, demonstrates that the repeatable experiment as crucial to modern scientific method (in fact part of the definition of modern science as modern) is an ontological, not a pragmatic, requirement. Only the measurable being seen as real turns out not to be a validated proposition but a necessary assumption in order to put modern science on a mathematical basis, the latter being intended prior to any questioning commenced.
Beyond the undemonstrated assumption that certainty would be mathematical in nature, there are two symmetrical problems with the methodology of the Cartesian Meditations. The meditations begin with a promise to ‘doubt everything’, in order that the certainty of knowledge can be based on something that is intrinsically indubitable. The outcome is that the “I” is the only certain subject, expressed in the famous “cogito ergo sum”. We can only posit the I as a subject, though, in terms of its being the experiencer of representational experience, which places representational experience as an a priori not admitted by Descartes. The symmetrical problem is that the nature of the I is undetermined, precisely because it can only be determined as the negative of any particular positive experience. In and of itself it can never properly become an object to itself, and thus can never experience itself precisely as its subject. A second pair of problems arise from the representational nature of the subject’s experience: for it to be represented it must have been previously presented in some fashion, specifically to the originator of the representation; the identity of the I-Subject with the Self has been necessarily assumed but not in any way demonstrated, and given that the originator of the representation (and thereby the experiencer of the presentation) appears to be by definition the Self, that identity falls apart.
The various issues that this led to comprise the history of epistemology from Descartes on. At the extreme of its critical impasse in Kant, the result is that nothing can be truly known, and the knower is itself illusory. The paradoxical nature of this statement, and the natural objection that as paradox it can only be based on invalid assumptions, is still not recognized by the latest claims in cognitive science, which beginning with the same basic ontological assumptions produce the same paradoxical conclusion as Kant.
Returning to the pre-ontological meaning of matter, if the ‘physicalist’ aspect of the material is not primary but a derivative meaning, the primary meaning being that of a ‘topic of concern’, or ‘relevant subject’, we have to inquire into how ‘being of concern’ or ‘being relevant’ could enter into a clarified material ontology:
Previous letting something be relevant to … with … is grounded in an understanding of something like letting things be relevant, and such things as the in-which and with-which of relevance. These things and what underlies them, such as the what-for to which relevance is related, the for-the-sake-of-which from which every what-for is ultimately derived, all of these must be previously disclosed in a certain intelligibility. And what is that in which Da-sein understands itself pre-ontologically as being-in-the-world? In understanding a context of relations, Da-sein has been referred to an in-order-to in terms of an explicitly or in-explicitly grasped potentiality-of-its-being for the sake of which it is, which can be authentic or inauthentic. This prefigures a what-for as the possible letting something be relevant which structurally allows for relevance to something else. Da-sein is always in each case already referred in terms of a for-the-sake-of-which to the with-what of relevance. This means that, insofar as it is, it always already lets beings be encountered as things at hand.
Haidigger, Martin (2012-07-30). Being and Time – Sein und Zeit (Kindle Locations 1561-1568). . Kindle Edition.
If it is relevance that lets beings be encountered as at hand in the first place, then it is only as already relevant that a being can be so encountered, and only as relevant that it can become an ‘object for science’. It is this relevance, that something ‘matters’, that first enables science to get hold of its object in an ontologically valid manner. Simultaneously, the concern for things of relevance as the fundamental meaning of human being allows the human being to be scientific.
It may seem odd at first to speak of a material ontology that is not intrinsically physicalist, but instead derives from the verbal sense of matter – “to matter”, as well as the original nominal sense – “what’s the matter?”. However since the physicalist interpretation fails the test of scientific evidence itself, particularly in the area of post-modern physics, where as Heisenberg put it “The ontology of materialism rested upon the illusion that the kind of existence, the direct ‘actuality’ of the world around us, can be extrapolated into the atomic range. This extrapolation, however, is impossible… atoms are not things.”
Lonergan’s ontology of the thing as specifically what is of concern in a particular area of experience provides a means of understanding how a universally valid ontology can be pragmatically modified in various areas of endeavor via complementarity of the differences in various sub-ontologies without thereby invalidating the universality of the basic ontology.
the notion of a thing is grounded in an insight that grasps, not relations between data, but a unity, identity, whole in data; and this unity is grasped, not by considering data from any abstractive viewpoint, but by taking them in their concrete individuality and in the totality of their aspects. For if the reader will turn his mind to any object he names a thing, he will find that object to be a unity to which belongs every aspect of every datum within the unity. Thus, the dog Fido is a unity, and to Fido is ascribed a totality of data whether of color or shape, sound or odor, feeling or movement. Moreover, from this grasp of unity in a concrete totality of data there follow the various characteristics of things.
Lonergan, Bernard; Crowe, Frederick E.; Doran, Robert M. (2012-05-23). Insight: A Study of Human Understanding, Volume 3: 003 (Collected Works of Bernard Lonergan) (Kindle Locations 5728-5733). University of Toronto Press. Kindle Edition.
This unity and identity in a given set of data taken in its totality assumes the data taken is a priori grasped as a unity ontologically, and is thus relevant to the person making the determination. As a result the “things” of the physicist, while they relate to the same unity, will necessarily grasp different aspects present in that given unity than will the “things” of the historian or psychologist.
When the various areas of experience refer to things, they most often mean what Lonergan refers to as ‘bodies out there now real’, and of course every area of experience wants ‘their’ things to be real in the sense that an animal would understand something as real, external and present. While we remain animals and therefore have a biological pattern of experience, we spend most of our time in other patterns, such as common sense, scientific, dramatic, artistic etc. The ‘things’ of those patterns often do not map to anything like ‘bodies out there now real’. As Lonergan puts it:
Animals have no epistemological problems. Neither do scientists, as long as they stick to their task of observing, forming hypotheses, and verifying. The perennial source of nonsense is that, after the scientist has verified his hypothesis, he is likely to go a little further and tell the layman what, approximately, scientific reality looks like! Already we have attacked the unverifiable image; but now we can see the origin of the strange urge to foist upon mankind unverifiable images. For both the scientist and the layman, besides being intelligent and reasonable, also are animals. To them as animals, a verified hypothesis is just a jumble of words or symbols. What they want is an elementary knowing of the ‘really real,’ if not through sense, at least by imagination.
Lonergan, Bernard; Crowe, Frederick E.; Doran, Robert M. (2012-05-23). Insight: A Study of Human Understanding, Volume 3: 003 (Collected Works of Bernard Lonergan) (Kindle Locations 5867-5871). University of Toronto Press. Kindle Edition.
The current ‘belief’ in science tends to support the notion that scientific ‘things’ are the ‘true’ ‘out there now real’ bodies, ironically precisely at the time that many of the images of science are unverifiable and thus have the least claim to validity of any of the ‘things’ of the various patterns of experience. Scientific ‘things’ require observable verification, precisely because there is no other validation available: by contrast common sense ‘things’ are validated by our ability to get around in the world, biological ‘things’ by our ability to survive and replicate, dramatic things by political efficacy, etc.. Since they have a verifiability not dependent on physical observation, they also require no physicalist ontology. Outside the biological pattern of experience there is no necessity for ‘things’ to be physically observable, because the data in which the unity is found does not have to be a physical set of data.
The worst excesses of scientific ideology are exhibited precisely in the sciences that do not deal with physical sets of data. The attempt to understand the mind, for instance, via neuropsychology or cognitive science is not simply an unwarranted reductionism, it misapplies physical concepts to a set of data that cannot be mapped to physical entities. Certainly the mind has its basis in the neurological system, but the unwarranted reduction is to equate thinking with neuronal firing. As an emergent set of data, our experience of thinking has nothing to do with the experience of neuronal firing on the neuronal level. What is a logical, meaningful series of thoughts for the mind is by definition incomprehensible at the neuronal level, and appears random. While neuronal ‘structures’ that map vaguely to areas of experience can be isolated in individuals, the inability of neuroscientists to find the same mapping in any other individual implies that they are persistent artifacts of the mind, not causal explanations of thinking. What differentiates a science from a pseudo science is the validity of its access to its objects of inquiry, and as a result neuropsychology and cognitive science, by their physicalist assumptions, are intrinsically pseudo sciences, since they have in the first instance no valid access ontologically to the psyche, in the latter instance to cognition, neither of which can be represented by a set of physical data nor can be causally mapped to any set of physical data. Any psychology has to start with observation and experience of the psyche, there is no other valid basis on which to predicate any psychological hypothesis. Likewise any cognitive science has to start with observation and experience of thinking, there is no other valid basis on which to predict any cognitive hypothesis. Every attempt to base either on physical sets of data is already a failure before it begins, because there is no validity to the initial assumptions, i.e. the ‘things’ of neurology have nothing to do with the ‘things’ of psychology or cognition, and the result is parallel with the physicist trying to foist unverifiable images on the layman as more ‘real’ than the ‘things’ of the layman’s various patterns of experience.
The question raised as to the knowledge-claim or truth-claim of the sciences can now be assessed. The ontological validity apparently secured by the repeatable experiment is only a relative validity, and one with serious limitations, since by definition any two scientists in a given field are very similar as entities, relative to all the other entities in observable reality. As well, the ‘things’ of the sciences have no intrinsic validity outside the scientific pattern of experience unless they can be verified observationally, which is often not possible when dealing with sciences that attempt to measure phenomena at vastly different scales than the scale at which we observe reality. The claim to a ‘certain’ base for knowledge in the sciences thus turns out to be a bogus claim. Going back to the Cartesian Meditations the reason that the bogus claim was initially believed, then its foundation forgotten, is that the missing ‘absolute’ perspective that can never be duplicated by any finite number of repeated experiments is provided by the Christian god. As a theological posit, and by theology’s own definition an unverifiable posit, the ontological basis of science’s truth-claim is not itself scientific but religious, nor is it based on evidence but on belief, which makes the atheism of many scientists an intrinsic and fundamental Self-contradiction if they want to maintain any truth-claim beyond practical usefulness. That this reduces the ‘things’ of science to the same ontological status as the ‘things’ of carpentry, software development or political activism is an unavoidable conclusion, and the truth-claim or knowledge-claim of science is untenable.
Simultaneously, for Davis’ project, the assumption that a universally valid ontology with which to interpret data is possible is fundamentally flawed. There is no ‘true’ ontology supplemented by the pragmatic ontologies of various disciplines, instead there are only ontologies of relevance to various disciplines insofar as they study and represent different patterns of experience. Each of these ontologies has certain basic assumptions that are unquestionable within that ontology, and therefore within that discipline, and any attempt at an intelligent interpreting system has to be able to determine from context what the guiding assumptions are, and therefore what interpretation is relevant in a given context, as well as how to perform that interpretation. Creating a single, mush-mashed ontology simply doesn’t work. Each ontology of relevance has to itself be clarified and validated for the system to be able to adequately interpret complex data in multiple contexts. As well, how a unity is perceived in a set of data in a given context – a contextual mereology, is a further basic requirement.