Table of Contents

The Philosophy of Physical Realism is a 1966 text by the American philosopher Roy Wood Sellars.


Roy Wood Sellars in his work The Philosophy of Physical Realism took up the cause of physical realism at a time when rapid advances in the sciences were rapidly changing the scientific image of the world. This was a time when the impressions of relativistic physics, quantum mechanics, and Darwinism were still very fresh. I myself can only imagine the challenge of approaching scientific problems philosophically during the time in which Einstein was actively engaged in his work.

In chapter 8 of Roy Wood Sellars’ attempt at a denial of universals appears to be a breaking point where Roy Wood Sellars’ prose begins to stall on the uncertain nature of our knowledge of physical reality. This particular chapter begins with the observation that the philosophical lexicon requires revision rather frequently, and Roy Wood Sellars’ belief that only individual things may be real by way of their relations. He makes the stand that generic universals cannot be real in the way that a particular thing is real.

Roy Wood Sellars on the hundredth page of Physical Realism entertains the question of whether the ultra-microscopic perceptible. After considering the careless way that perception is typically used at the common sense level, the definition of something as perceptible is set by Roy Wood Sellars to refer to the capability “of reporting itself directly or indirectly, in the sensory field.”(1) The indirect report may include phenomena that though readily inaccessible to the unaided senses may accessible report of varying quality and certainty though tools that allow for a mediated report, such as a microscope, cloud chamber, or magnetic spectrometer. For Roy Wood’s data must come from a phenomenon in a sensory arena in order to be able to feed the theoretical which guides experimentation and verification. From here perceptibility becomes a distinct property of anything we may denote as physical, for our ability to denote the actual physicality of anything hinges most decidedly on its having been reported through causal means to our senses.

With respect to the ultra-microscopic, those things which have merely been speculated or conceived to exist can’t yet be given the same accord as those things that have been found and reported. Ultra-microscopic particles like a Higgs Boson which only exist to fulfill a condition in a theoretical system can’t be held on the same level as an observed particle like electron, protons, or quarks. The perceptible can even challenge the theoretical with things observed that don’t necessarily fit into the standard model such as the electron’s larger relatives the muon and the tau and they are known to exist an often appear in searches for other things.(2) Even the observed and demonstrated ultra-microscopic phenomena in the world though are eclipsed in immediate importance in the lives of most people by the objects in which they interact on a perceptible level, at least in terms of how a person takes in, processes, and interacts with the world. Even the researcher dealing with the submicroscopic observes interacts with the ultra-microscopic using tools which appeal to the researcher’s perceptual resolution.

In what may function as a form of substitute for concrete universals Roy Wood Sellars as well as an argument against concrete designation as an absolute certainty he offers the levels of knowing. The levels of knowing never feature a level which would represent a complete knowledge of a thing. My knowledge of the desk in front of me is something I have definite knowledge of though I do not have a kind of knowledge of the arrangements and relations of its constituent particles down to the level of quarks, leptons, and beyond. I have a certain kind of knowledge of its assembly down to the level that it uses self threading screws to retain some metal parts in their relation to other parts, but I only have this knowledge from assembling the desk. For all practical purposes had I not assembled the desk this knowledge would be meaningless, and for practical purposes is meaningless on most days I use this desk. It is merely sufficient for me to know that this desk is capable of serving the role in which I require it, though this may actually be too little information. I may need to be able to distinguish it from other desks, locate it in relation to other objects, conceive of it in terms of the storage capacity of its drawers while seeking to stock it with supplies, or have knowledge of it or its relations in many different situations or contexts that I may or may not expect or possess information about it without even actively considering myself to be in possession of that particular bit of knowledge.

Much as I know this desk has a purpose it is capable of serving, and that I may report this observation the physicist reports on the ultra-microscopic. That I lack the perceptive abilities and technology to myself expand my sensory capabilities in order to search through my desk for muons that may be hiding within it, my account of the existence of this desk and the account of the physicist reporting the existence of muons are both reports of existing things which are physical and perceptible. While the size of the selected objects varies in both cases there is very much a reporting object in the physical world. I may consider the possibility that one of the carbon to hydrogen bonds in the cellulose fibers of the desk’s particle board top may actually involve a muon instead of an electron. This suspicion may be unlikely but it may also as well be unfalsifiable. Knowledge of the behavior of muons may not entirely rule out the possibility that they may take on a role replacing electrons in their atomic orbits or in bonds between atoms. The nature of this suspicion seems to be one of unimmediately demonstrable trivia as it is too unimportant to investigate, at least in the question of whether the possible phenomena is occurring in the top of my desk. That this three way synthesis can occur between an observed ultramicroscopic particle, a speculation of its possible behavior based on a related particle, and an object immediately perceptible to me in the form of my desk creating this bit of speculative trivia demonstrates that that we can positively construct somewhat plausible questions into which no immediate inquiry may be fruitfully taken leaving doubt about an aspect of an object.

This sort of relationship of knowledge to a physically real world to be a great strength for Roy Wood Sellars through this point, as it avoids the logical problem of deductive completeness in handling our knowledge of objects in the real world. The problem of completeness is most greatly exemplified in Whitehead and Russell’s Principia which inspired Gödel to pose his two theorems of incompleteness against axiomatic systems in the mathematics. To reach an absolute true designation for a given schema of the sort implied by the idea of the schema having a universal quality is additionally tied to the idea that the construction of a system, and much of the history of western metaphysics is in the construction of systems. Considering the rejection of universals it may make sense to consider objections to systems which support concrete universals logically and practically.

The logical objection contemporary to The Philosophy of Physical Realism can be found instantiated in the work done by Kurt Gödel in mathematics a year before the work’s publication in which the ability for a formal system to cover all cases in a deductively complete manner is found to be flawed at the system’s axioms which themselves are taen to instantiate a form of universal. A formal system is complete if and only if every logically valid formula expressed in the formal language of the system is a theorem in the system, but to be deductively complete a formal system must for every formula in the system have either that formula or its negation as a theorem in the system. The problem Gödel highlights which affects the Principia along with Zermelo- Fraenkelian set theory shows that not all problems expressible in terms of these systems are open to proof within these systems deductively, and that this problem “is not due to the nature of these systems” while additionally not being limited to these systems.(3) Gödel’s demonstration shows that no recursive system can reach the deductive level of completeness. Previously in geometry the axiomatic method of Euclid was failed by Gauss, Lobachevsky, and others however Gödel’s work ended the last effort to revive an axiomatic geometry as last seriously attempted by Hilbert.

This isn’t at all to say that arithmetic calculation use, or even that claims unprovable in an arithmetic system as highlighted by Gödel aren’t provable in other systems. It does raise questions though as to the possibility of truly universal standards for known concepts. A universal as a standard relies on a recursive relation to itself as universal. A problem arises in taking from a report of an existing object, qualitative information and producing a judgment that a quality of that object may be shared with another reported object. The problem is in whether a difference exists between a report of existence and a report of a quality. Consider that ‘there exists a pill’ and ‘there exists a red pill’ are making two different sorts of claims. To more readily contrast the statements let them be restated ‘there exists a pill’ and ‘there exists a pill with a red appearance.’ The first makes a simple existential claim where there is an object which has been selected and reports its existence, while the second adds to the existential claim a qualitative statement about the manner in which the object appears in its report. The quality of redness as a property belonging to a reporting object opens problems of precision beyond the simple existential claim.

Red is a color, perceived when certain wavelengths of light stimulate certain receptors on the retina. There is a limit inherent in the structure of the human retina as to ability to distinguish between slightly different wavelengths of light. For claims of appearance of an object, there is a level of approximation to be considered in how qualitative information is reported to the senses, which is inherently more complex conceptually than the simpler existential claim that there exists a reporting object. Consider the example of a pile of straw in that while it is a simple claim to select the pile as existing, qualitative claims about the pile are going to acquire either a great level of complexity or be filter through a process of approximation in describing the pile.

If there is a point at which efforts to systematize that mathematics of arithmetic fail in either completeness or consistency at any level as demonstrated by Gödel in his first work on undecidable propositions. Indeed if logical completion and consistency of arithmetic is beyond axiomatic proof, then it seems in the interaction between language and the world it describes a naïve or simple account of things at hand when we select and describe and object have still less hope considering the variety of existing things compared to the cleanliness of the number line. It seems as though any collection of rules in a system must be open to frequent revision in order to adapt to the changes inherent in the inquiry which expands our knowing. While skepticism may seem attractive when deductive completeness is a standard failed by so many attempts at creating logical systems, it is important to hold on to the Cartesian that I think therefore I am while maintaining the caveat that I am, I am inundated with reports.

With the logical problem of completeness considered as a structural flaw of systems inviting universals, it becomes relevant to consider the philosophical position of tropism against practical formulations of universals. Considering particular things and the predicates we attach to particular things, something about the predicate we attach to a particular thing makes that predicate’s relationship to that particular thing unique. Kierkegaard’s anxiety is somehow different from the anxiety we can predicate as belonging to other particular instances. The striving associated with Malcolm X is a different determination that that associated with Lance Armstrong.

The consequences of accepting this sort of tropism may be carried on to physical properties ascribed to particular objects as well. The black of a particular eight ball is different in some fundamental way from the black of even an apparently identical eight ball. As ridiculous as this might seem at first glance, consider that on some level a particular eight ball is a collection of particles and their relations and progressing downward to the smallest known constituent particles of the eight ball at a certain point the ability to observe the positions and relations of particles at a quantum level meets with uncertainty. The principle of identity distinguishes each from the other. When properties are by all measures indistinguishable a property may be predicated to one in terms of its identity.

When grounding an ontology deep in physical reality as Roy Wood Sellars is attempting to do even with the two apparently identical eight balls it is unreasonable to be able to attach the same descriptive predicate to both eight balls on a universal scale. In normal language it is reasonable to say that the two eight balls are a matching color black, but the normal language statement’s ability to have been synthesized does not imply that there is a universal color black being predicated to the being of both billiard balls. The sort of tropism outlined in the last two paragraphs seems to be supported in the program of Roy Wood Sellars levels of knowing, and this kind of tropism seems to be at the root of his adoption of concept over universal with the loaded metaphysical baggage attached to the idea of universals that makes them prone to critique through this kind of tropism. It would be unreasonable to attempt an effort at a common language where ontological concerns would override our abilities to speak discursively about our surroundings to the limit in which our sensory abilities may no longer reliably provide further useful information. That there can be a color of this billiard ball attached to the condition of this billiard ball as being distinct from that billiard ball with the color of each predicated distinctively from the other on their attachment to one billiard ball over the other at least contingently. It is not a practical distinction even if the colors are identical or even if the colors are necessarily distinct in some minimally distinguishable way as it may be practical to consider two billiard balls red even if one is a noticeably more faded. It is a contingent distinction that if nothing else this is how color is manifest on one billiard ball and not the other billiard ball.

When hiking in pristine wilderness we need not chemically and physically assay every natural object that catches our fancy in order to speak about it on the most intimate physical level to be able to regard it as physically real. Observe a forest without attentively noticing one particular leaf. Similarly we can speak of things encountered throughout everyday life without breaking them down into the smallest bits understandable by the science of the time. Roy Wood Sellars uses concepts in lieu of the more ontologically and metaphysically demanding universals in order to allow for persons to develop meaning for internal and shared linguistic use.

The development of a concept as told by Roy Wood Sellars is the building of a particular content through a sort of accretion of data to develop a meaning in concept. To develop a concept of dog through the gathering of bits and reports to refine an idea of dog as distinct from things not dog with further refinement into ideas of different types of dogs and different particular dogs. For a sufficiently young child their concept of dog may be limited to a particular dog expanding over time to a wider variety of dogs with characteristic and the degrees to which they are considered constitutive to what a dog is varying expanding and sometimes shrinking or disappearing over time. There can be gray areas including wolves, foxes, and coyotes and the degrees to which each may be dog or dog like. Roy Wood Sellars dismisses the task of psychologically investigating meaning, but he does seek to give a functional account of concepts.(4) In clarifying what he is seeking to explore about a concept he accedes that what he is, “interested in is the characterizing of the object and not my idea as an event.”(5) He shortly clarifies that idea by putting forth, “Another way of putting my thesis is that concepts are always implicit judgings or propositional functions.”(6) It is a propositional “this seems to be” rather than an absolute “this in essence is.”

The problem of universals and of their partner in metaphysical crime essences are the philosophical baggage he is stuck working against if really want to avoid the problem of idealism he made so clear his revulsion of early on. He traces the history of idealism as crucial phase in Western philosophy which was in ways necessary in the light of science’s erosion of the fundamentals assumption of the world, when beginning with Descartes dualism the faculty of knowledge had to be turned to analyze itself. Berkeley’s dogmatic idealism is pointed out as a particularly necessary obstacle which must be conquered for a return to realism after such a sound attack on what amounted to the fundamental assumptions that have followed realist metaphysics since Aristotle, with the objective idealism beginning with Kant serving as an episode of calm before the idealist systems would eventually erode.

Idealism though is not Roy Wood Sellars’ greatest obstacle in the way that the naïve realist tradition that follows from the tradition of Aristotle is. While it wouldn’t necessarily be unfair to call much of Aristotle’s scientific work proto-scientific, such scrutiny rarely has come up against his logic and been taken seriously. The idea that subjects to which we attach predicates are in some way a primary oussia or mode of being, throws a big blanket over the potential for meaningful empirical progress in the science. That the chair I’m sitting on is in its more primary mode of being is a chair chairing neglects so much of the fact of the chair’s physical reality and the consequences of such reality. Through his detachment through levels of knowing of the report from the object and the detachment of language from a foundation in logic of an absolute, Roy Wood Sellars manages to create the framework of an ontology which through embracing some uncertainty allows for a physical realism which avoids the logical bugbear of deductive completeness. In The Philosophy of Physical Realism Roy Wood Sellars accomplishes this at a time in history in which the body of scientific knowledge and the speed at which it was growing was rapidly increasing.


  1. Sellars, Roy Wood. The Philosophy of Physical Realism, p.101
  2. Greene, Brian. The Elegant Universe. p. 8
  3. Gödel, Kurt. On formally undecidable propositions of Principia Mathematica and related systems I, trans Martin Hirtzel (Boulder: University of Colorado, 2000) p. 2
  4. Sellars, Roy Wood. The Philosophy of Physical Realism, p.164
  5. Ibid. 164
  6. Ibid. 165

Works Referenced

Kant, Immanuel. Critique of Pure Reason. trans. Smith, Norman Kemp. (New York: Palgrave Macmillian, 2007), p. 9 Green, Brian. The Elegant Universe. (New York: Norton, 1999) Gödel, Kurt. On formally undecidable propositions of Principia Mathematica and related systems I, trans. Martin Hirtzel (Boulder: University of Colorado, 2000) Sellars, Roy Wood. The Philosophy of Physical Realism. New York: Russell & Russell, 1966.

Book Review | Philosophy

QR Code
QR Code philosophy_of_physical_realism (generated for current page)