Towards an Objective Theory of Rationality

5. Metaphysics and Epistemology

Book cover: An Idealist View of Life by  S. Radhakrishnan

The specific details of an adequate theory of rationality are dependent on our metaphysical presuppositions; more particularly, on the ontological and cosmological commitments we make. How we infer truths about our internal states and the external world, for example, depends on our notions of phenomenal experience, mind and matter and how we think they interact. Our choice of metaphysics is, therefore, inextricably bound up with our theory of rationality, and vice versa. What ontology-cosmology should an objectivist assent to? The two great traditions, idealism and realism, have had a long and tortuous history dating back more than two thousand years. There have been many enlightening and interesting developments resulting from the interaction between the protagonists for each side. I will not discuss these here.[13] However, in this section, I do want to explore what I think is a new turn in the argument provided by the type of theory of rationality developed by Lakatos and his school and defended here.

I shall characterize idealism and realism as two broad streams of thought offering contrary theses; realism on the one hand asserting the existence of mind-independent entities, knowledge of whose properties is afforded access by sense-experience, and idealism on the other denying the existence of such mind-independent entities and the consummate role of sense-experience. Realist and idealist theories have had to contend with a number of serious problems, especially those of logical coherence and meaning. Whereas realism has progressed, though, idealism has faltered. Idealists had sought to provide an idealist analysis of natural languages that would eliminate realist assumptions. (Phenomenalists became engaged in their own programme of elimination.) They have been largely unsuccessful, for natural languages appear to be inherently realist.

There is a second serious impediment to accepting an idealist framework. Recent developments in neurophysiology and psychology indicate that the act of perceiving physical objects is not simply a matter of conveniently grouping discrete units in our visual field into particular packages. Our perception of solidity, depth and contour are not arrangements of the elements of our visual field, but the result of complex neural processing in the brain. In perceiving three-dimensional solid objects, there occur not two discrete acts, one of the awareness of the discrete elements in the visual field and the other of cognitive interpretation, but the one indissoluble act of seeing a three-dimensional object. An example of this is our perception of depth in the reversible perspective drawings of cubes, staircases and other objects given in textbooks on perception.[14]

This act of seeing solid objects in three-dimensional space is only partly learned. The form of neural processing of visual information that goes on in our brains is also partly genetically determined. Our brains are hard-wired for us to perceive depth and contour the way we do.[15] The Ames illusion[16] is a classic demonstration of how our brains are hard-wired to see a trapezoidal-shaped room as symmetrical when the usual visual cues are hidden. Another well-known example is the rotating mask illusion.[17] A convex mask of a human face is rotated. As the inner-side of the mask comes into view, we see the mask as convex. This is because we have an inbuilt predisposition to believe that noses always point outwards on human faces. Because of this inextricable and unconscious fusion of our reception of sensory data and its interpretation, it is little wonder that idealists have been unable to construct a non-realist interpretation of natural languages.

This illuminates a third major obstacle for idealist theories. Realist assumptions are necessary for the living of our everyday lives. When a mother picks up her child, for example, she does not infer the expected position and look of her hands and her child's body as bundles of actual and possible sense-experiences. To calculate such inferences given the variability and uniqueness of much of our sensory world is a psychologically impossible task. Similarly, in everyday life, the idealist finds it impossible to regard other minds as simply bundles of actual and dispositional behaviours. Just as the idealists' gestalt sees physical bodies imbued with minds (that is, they see living human forms as persons and not as bundles of actual and dispositional behaviours), our gestalt sees external objects located and moving in three-dimensional space, with no conscious inference from discrete sense-impressions. Modern neurophysiology and psychology are revealing for us how this is so.

Now, an idealist may concede these points, claiming, however, that the presuppositions of a realistically interpreted natural language are no argument for the veracity of such unconscious assumptions. He may explain post hoc the latest developments in psychology and neurophysiology. In the final analysis, he may admit that it is useful to act as a realist would act, while consistently adhering to his idealism. I have little doubt that a coherent version of idealism can be developed and modified post hoc to account for any current and future developments in science and any possible configuration of our perceptual experiences. That such an idealist theory could explain our experiences is not strong evidence for its veracity. To earn our rational assent, according to Criterion 5, the idealist theory must have evidence in its favour independent of its construction and modification.

Book cover: The Invention of Science: A New History of the Scientific Revolution by David Wootton

Realism, on the other hand, has overwhelming independent evidence in its support. The realist interpretation of physical and neurophysiological theories has led to the derivation and confirmation of novel phenomena; that is, phenomena that had played no part in the original construction and modification of the theory. Those theories that had independent confirmation, thereby satisfying my criteria for rational appraisal, confirm in a spectacular way the realist ontology-cosmology.

Consider, for example, Newton's Three Laws of Motion and his Universal Law of Gravitation. These two theories, interpreted realistically (in conjunction with a realist interpretation of the auxiliary hypotheses), were independently confirmed by the successful predictions of the existence and position of Neptune, the date of the return of Halley's comet and the novel derivation of the measure of the progress of the moon's apogee. These spectacular successes of Newtonian mechanics also simultaneously supported its ontological and cosmological underpinnings; the idea that bodies possessing the properties related in Newton's axioms exist independently of minds, in absolute space and time.

The same can 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. These examples are but a very small fraction of the empirical growth in our understanding of the universe prompted by advances in scientific research. My key point here is that each of these theories, realistically interpreted, made bold novel predictions that were subsequently stunningly confirmed by independent observers. Conversely, idealism has had little or no empirical confirmations that satisfy my criteria for objective theory appraisal. It is only able to explain the realist's anticipation of novel phenomena and his novel derivation of statements about well-known phenomena post hoc.

Taking a step back to fundamentals, we can say that the data for which realism and idealism seek an explanation are those private phenomenological experiences that we refer to as 'sense-experiences'.[18] These 'sense-experiences' for the idealist and the realist, equally, are epistemically primitive in the sense that our own private phenomenological experiences are what each of us have most directly. (Even though they are the most direct, they are not necessarily infallibly known.) The upshot here is that realistically interpreted ontology-cosmologies lead to the successful prediction of new and unexpected, and the novel derivation of previously well-known, sense-experiences. Idealism has been, in comparison, largely unsuccessful in the anticipation and novel derivation of such experiences.

The same type of argument may be successfully adduced against the instrumentalist interpretation of explanatory theories in science. It can also be used to elucidate how each of us as children comes to a realist interpretation of our phenomenal world.[19] The argument against idealism and in support of a belief in mind-independent entities is a substantial one. I shall now go on to further develop the theory of rationality, proposed above, within such a realist framework.


  1. [13] See my Allan [2016a] for a critical discussion of realism and idealism.
  2. [14] For examples of reversible/bistable images, see Thomson [2009a, 2009b], Kogo [2012] and Nordhjem [2012].
  3. [15] See Vernon [1971: ch. 8] for a general discussion. For details on the genetic component contributing to our perception of depth and contour, see the important report by Julesz [1964]. For experimental reports on neural processing in the visual cortex in higher vertebrates, see Hubel and Wiesel [1962, 1968].
  4. [16] For a visual demonstration, see Scientific American [2012].
  5. [17] For a visual demonstration, see eChalk Education [2012].
  6. [18] The contents of our subjective experiences of observed objects have been variously termed 'sense-impressions', 'sense-data' and 'qualia' by philosophers.
  7. [19] For an exploration of Piaget's and other developmental psychologists' research on how children develop the concept of an external physical world, see my Allan [2016a].

Copyright © 2016

You will be interested in
Book cover: Epistemology: A Contemporary Introduction to the Theory of Knowledge by Robert Audi
Book cover: The Logic of Scientific Discovery by Karl Popper
Book cover: Progress and Its Problems: Towards a Theory of Scientific Growth by Larry Laudan
Book cover: Fads and Fallacies in the Name of Science by Martin Gardner
Book cover: An Inquiry into Meaning and Truth by Bertrand Russell
Book cover: Philosophy of Science: An Historical Anthology by Timothy McGrew, Marc Alspector-Kelly and Fritz Allhoff

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: