The Existence of Mind-Independent Physical Objects

4. Realism as Empirically Progressive

Book cover: The Problems of Philosophy by Bertrand Russell

In this final section, I want now to go on to defend the realist's thesis b.; that is, the thesis that the physical objects described in our common-sense and scientific theories do in fact exist. Here, I shall confront the eliminative idealists, such as Collier [1713], who agree that thesis a. is analytically true, but who deny thesis b. It may seem that the issue between the idealists and realists is empirically undecidable, for no matter how long our run of experiences is, a description of their sequence and content is logically compatible with there being no physical objects. That is, their sequence and content may be just as if there were mind-independent physical objects when in fact there are none. The eliminative idealist will conclude at this point that no experiential evidence could possibly decide the matter.

This possibility seems to leave the way open for an eliminative idealist interpretation of our experiences in terms of an underlying and ultimate mental reality. I will argue below that the issue is empirically decidable (not of course to the extent that we are assured of logical certainty) and that we can decide the issue in the same way that we can decide between other competing empirical theories. It is for this reason, that the dispute is, in the final analysis, decidable on empirical grounds, that I have, by and large, left aside the conceptual issues raised by realists and idealists, although I think these very important. (I have touched on one such issue in the above discussion on phenomenalism.)

What positive reasons do we have for believing that physical objects exist? I shall begin my answer by pointing out that there are certain basic epistemological principles that I think we all use, albeit usually imperfectly and unconsciously, in evaluating alternative explanatory theories. Two of these principles are relatively uncontroversial. We do not accept a statement as evidence for a theory unless:

  1. it is true, and
  2. it is logically entailed, or its probability is logically entailed, by the theory in conjunction with auxiliary hypotheses, but not by the auxiliary hypotheses alone.

It is important to note here that explanatory theories do not entail evidence statements in isolation. For the logical entailment to proceed, auxiliary hypotheses are required. These auxiliary hypotheses include theories of observation, statements of initial conditions, and so on.

For an epistemology to earn the title of being an 'objectivist' theory of knowledge, two further requirements need to be satisfied.[11] These requirements hinge on the notion of theory-evidence independence. For an explanatory theory to be objectively validated, it must satisfy this stronger requirement that the evidence-statement in support of it is somehow 'independent' of the theory it is supporting. I state this requirement as follows.

An evidence-statement is independent evidence for a theory if:

  1. c.
    the epistemic status of the evidence-statement is determinable independently of the assumption that the theory is true, and
  2. d.
    the evidence-statement is independent of the data used to construct and modify the theory.[12]

What this means is that for an explanatory theory to be rationally justified, it must, in conjunction with auxiliary hypotheses, allow for the successful prediction of some hitherto unknown fact (novel prediction) or the successful novel derivation of some hitherto known fact not used in the construction and revision of the theory (postdiction). Furthermore, explanatory theories are not evaluated in isolation. They are evaluated against the relative success of a background theory or other rival theories. So, for a theory to be vindicated, there must be no rival theory with greater or nearly equal independent evidence in its favour. On this scheme, rival explanatory theories are judged in their historical setting by how well they anticipate novel facts and provide novel explanations of already familiar facts; that is, by how well they lead to the growth of our understanding. A successful theory, then, is not made obsolete until an even more successful theory appears.

It is my contention that realism satisfies these criteria of objective theory choice while idealism does not.[13] Consider firstly how we come to believe that there is a world of mind-independent physical objects. For the answer to this, we must rely on the researches of cognitive and developmental psychologists. Up to a few months of age, the child's conception of reality is 'phenomenalistic' in that she considers the objects of her present sensory field to be wholly dependent on her motor activity. At a few months of age, the child forms the conception that the objects in her sensory field exist independently of her experience of them. The child thus begins to retrieve objects hidden behind screens and to search actively for vanished objects at places with which she had not associated the object on previous occasions. (Prior to this stage of development, the child had simply ignored an object that had vanished from her sensory field or had simply repeated a motor response that had yielded the object on previous occasions.)

At this stage of development, the child has formulated a realist theory and, in conjunction with a number of auxiliary hypotheses, such as a naïve observation theory, a primitive conservation principle, and so on, she is able to anticipate the results of a novel activity that she was not able to anticipate whilst relying on her former 'phenomenalistic' theory.[14] I am not suggesting that the young child at this stage is able to verbalize her theories, nor even that she is able to deliberate consciously over the alternatives. Nonetheless, I think it true to say that the child has made a pre-linguistic conceptual advance.[15] Now, if we take the insights of cognitive and developmental psychology (whether we interpret the science of psychology realistically or within some idealist framework) as indicating our own cognitive psychological development, then we have an important argument for realism. We can conclude that realism has had evidential support in our own lives.

Furthermore, realism is confirmed many times over in our present day-to-day circumstances. For example, on the basis of our set of complicated and interconnected physical beliefs, we are able to anticipate what we will find at the end of a plane journey to a new destination. Similarly, when we are called upon to repair some piece of machinery or equipment, such as an internal combustion engine or a short-wave receiver, we are able, if we are proficient, on the basis of our theoretical physical knowledge of the machine and its construction, and of the symptoms it displays, to predict which component or components will be found to be at fault, even if we had never experienced just these symptoms and just these faulty components before.

These arguments for realism are convincing enough in the private domain. However, once we accept that there are other private individuals with psychological histories similar enough in structure and quality to ours to enable interpersonal communication to take place and, consequently, a body of public knowledge to develop, the arguments for realism are even more compelling. Most idealists will thus accept the existence of a scientific community (we have already made use of this assumption in the above discussion of the cognitive development of very young children) and the long historical development of the sciences. However, the idealist will interpret these states of affairs idealistically. A review of the history of science reveals a succession of realistically interpreted scientific theories in a variety of scientific domains that have yielded an abundance of successful predictions and novel derivations of previously known facts. Each new successful theory constitutes empirical growth in comparison with its defeated rival through divulging more of the world to us and greater intricacy in its inner workings. This progressive nature of the realist program grants us overwhelming and increasing independent evidence in its favour.

Book cover: Perceptual Knowledge by Georges Dicker

Consider just one such theory viewed from an initial idealist standpoint; Newtonian dynamics. From such a viewpoint, Newton is regarded primarily as a mind or an aspect of the Absolute (or whatever mental entity the idealist proposes) who postulated his realistically interpreted Three Laws of Motion and his Universal Law of Gravitation. (I have already argued in the previous section that qua physical theory his postulates cannot be interpreted non-realistically.) Newton's theory, in conjunction with the realist interpretation of auxiliary hypotheses (such as a theory of observation, a statement of initial conditions, and so on), allowed for the prediction of certain novel perceptual experiences; experiences that would not have been expected had it not been for Newton's postulates. These novel perceptual experiences included those experiences that we would describe as perceiving the existence and position of the hitherto unknown planet, Neptune, and perceiving the return of Halley's Comet in 1759. Furthermore, his theoretical system provided a novel derivation of a description of those experiences we call perceiving the progress of the moon's apogee.

My point here is that the idealist can describe these novel perceptual experiences in the idealist's own schema, just as he can describe the events and states of affairs preceding the successful predictions and postdictions. However, once we accept an objectivist framework for evaluating rival explanatory theories, Newton's predictive and postdictive success constituted independent evidence for Newtonian mechanics. That is, they provided confirming instances of Newton's theory that bodies possessing the properties related in his axioms exist independently of minds, in absolute space and time. (It is important to note that the auxiliary hypotheses are also confirmed at the same time.)

But now, the same could be said for the confirmations of Einstein's Special and General Theories of Relativity, the Rutherford–Bohr atomic theory, the Kinetic Theory of Gases, the synthetic theory of evolution and the modern neurophysiological theory of perception, and so on and so on. The upshot here is that even if we start out as idealists, the empirical growth of scientific theories consequent to their realist interpretation impels us to adopt realism. Although the truth of our current realistically interpreted physical theories cannot be logically guaranteed (and here they share a feature with all idealist theories), each was able to derive striking and novel evidence in its favour, marking an advance over its defeated rivals.

It seems to me that if idealism is to gain our rational assent, idealists must provide us with a version of their theory that has independent evidence in its favour. Idealism, so far, has proved heuristically sterile. This may be an unavoidable feature of its conceptual commitments. Consider once again our infant on the verge of adopting a new conceptual framework and imagine her, for a moment, adopting the theory that Berkeley's God causes her perceptual experiences, or that her experiences are an aspect of the Absolute. I find it very difficult to see how she could come to anticipate novel experiences in the way that a child operating within a realist framework is able to do. Such posits are far too non-specific to limit the range of possible future experiences. An idealist infant, it seems, would simply have to content herself with waiting patiently to see what God did next, or what mode of existence the Absolute would take on in the next instance.

I think that this non-specificity has been one of the main reasons that idealism has been parasitic on realist advances. Witness how phenomenalism (though not necessarily idealistic) has been able to offer semantic reconstructions of our theoretical concepts only after the requisite theoretical advance had already been made on realist assumptions. For idealism to become a credible challenge to realism, its proponents need to generate a progressive research program with a theoretical underpinning that is able to predict or postdict some novel phenomena.


  1. [11] I am presuming an objectivist epistemological framework to help decide the ontological view at stake here. Subjectivist and relativist epistemologies can't do the heavy lifting required as they quickly collapse to an uninteresting statement of the author's unsupported personal view or that of the author's social group.
  2. [12] I have explained these principles in greater detail and have tried to justify them by showing how they satisfy the demands of a general objectivist epistemology in my Allan [2016].
  3. [13] The situation is much more complicated than the simple picture presented here. In actual fact, 'idealism' and 'realism' constitute the conceptual cores of competing research programmes in which the realist and idealist theories presented have undergone continual revision and refinement. However, these details will not affect the argument that I want to present here and so need not detain us.
  4. [14] For more examples of anticipations of novel facts at this stage of cognitive development, see Gardner [1978: 71f, 76f], Baillargeon [1986] and Spelke [1990].
  5. [15] The picture that I have presented here is very much simplified. For example, the child at this stage has no idea of a private-public distinction and her early beliefs are largely animistic. See, for example, Gardner [1978: part 1, §2], Manis [1973: ch. 3], Mussen [1973: ch. 3], Piaget [1970, 1973] and Vernon [1971: ch. 2].

Copyright © 2016

You will be interested in

Share This

  • twitter
  • facebook
  • linkedin
  • googleplus
  • gmail
  • delicious
  • reddit
  • digg
  • newsvine
  • posterous
  • friendfeed
  • googlebookmarks
  • yahoobookmarks
  • yahoobuzz
  • orkut
  • stumbleupon
  • diigo
  • mixx
  • technorati
  • netvibes
  • myspace
  • slashdot
  • blogger
  • tumblr
  • email
Short URL:


PDF Download The Existence of Mind-Independent Physical Objects

Download this essay