Critical Realism

views updated Jun 08 2018


Critical Realism is the title of a book by Roy W. Sellars published in 1916. The name was adopted by a group of philosophers who shared many of his views on the theory of knowledge. Essays in Critical Realism: A Cooperative Study of the Problem of Knowledge by Durant Drake, A. O. Lovejoy, J. B. Pratt, A. K. Rogers, George Santayana, C. A. Strong, and Sellars was published in 1920.


Much of the epistemological debate since the seventeenth century stems from the matter-mind dualism of René Descartes, who argued that what we know first and most surely is not a physical world but the existence of our own minds, and of John Locke, who argued that we are immediately acquainted only with our own ideas. Starting from these assumptions, how can one know a physical world external to the mind, if, indeed, such a world exists at all? Critical Realism is a chapter in this long debate. Some philosophers, finding it impossible to bridge the gap from a mental world to a material reality that transcends it, turned to some form of subjectivism or idealism; at the beginning of the twentieth century the dominant philosophy in Britain was the Neo-Hegelian idealism of F. H. Bradley and Bernard Bosanquet, and in America it was the voluntarism of Josiah Royce, the personalism of George H. Howison and Borden Parker Bowne, or the pragmatism of William James. But idealism, uncongenial to common sense and to ordinary interpretations of physical science, was followed by a reaction. Scientific knowledge seemed to support philosophical realism rather than idealism.

Shortly before the emergence of Critical Realism a group of philosophers, calling their view the New Realism, argued that even if it is true that whenever something is being perceived, it is an object for a mind, it does not follow that it has no existence except by being perceived. Hence, the idealist commits a fallacy if he concludes that the whole world is nothing but ideas from the truism that when something is known, it is an object for a mind. The American new realists, thenand here they could claim the support of such important British thinkers as G. E. Moore and Bertrand Russellmaintained that elements in perception can at the same time be elements in the physical world. Things do not cause ideas in us, as Locke would have said, so that we first know only ideas and then try to infer from them the nature of the real world which is never directly perceived. Rather, knowing is more akin to selecting, or throwing a light upon, aspects or parts of a world already there to be selected or illuminated by the light of consciousness.

The Critical Realist Position

The critical realist agrees with the new realist in holding that there is an objective physical world; their disagreement is chiefly on the question of the relation of the datum of knowledge to its object. Physical things, or parts of them, cannot be directly presented to us in perception. Considering the great variety of what is perceivedthe double image, the partially submerged bent stick, the toe that is felt after the leg has been amputatedunder various conditions by both normal persons and those who are, for example, inebriated or color-blind, are we to say that the real world actually contains all that is disclosed in all these circumstances? And is there no such thing as error? The trouble is that the "direct" realist, by identifying the immediate data of knowledge with elements of the physical world, is trying to account for the universe with an insufficient number of categories or kinds of entities. The knower, whether he is conceived as an organism and a part of nature or as a mind, does not "take in" the physical world. According to Santayana, the datum is an essence, a Platonic universal, which has an identity by being just the character it is, whether it characterizes one or many things in nature or characterizes no existent whatsoever. The datum, the immediately intuited evidence of reality, cannot be numerically identical with any part of that reality.

It is on this epistemological point that the critical realist opposes both the idealist and the direct realist. Whatever exists and whatever its character may be, no datum, or essence, given in experience exists, at least not in the sense in which we say that the objects of perception exist. As Santayana says in Scepticism and Animal Faith, "Existence is never given." When the astronomer talks about the moon, he does not mean by "the moon" the yellow disk image that may come to your mind; no doubt, a different image will come to the mind of your companion. If both of you understand the astronomer, "the moon" will mean to both of you, and to the astronomer, the same object to which your thoughts or perceptions are referrednamely, the distant satellite of Earth to which you ascribe certain physical properties. The words or images are the symbols of a meaning, but the essence of the word or image is not, in general, the essence of its meaning. The essence of the meaning is intended to be, but in cases of error will not be, the essence of the actual moon in the sky. This distinction, perhaps difference, between the nature of an image or sense datum and the nature of the object known by means of it is still more obvious when we consider feelings instead of visual images. When sympathizing with a person who has a toothache, we do not say, "I feel the way you feel"; we say, "I know how you feel." Knowing about another person's toothache is not having a toothache.

In perception, as distinguished from thought or conception, there is a tendency to identify image with meaning, so that an effort of analysis is required to separate image, meaning, and object. Paradoxically, the meaning is often psychologically prior to the image. For example, we may perceive a penny as round and as "out there" before noting that in the given perspective it presents an elliptical image. We can then analyze the situation into the image (elliptical), the meaning (round), and the belief that a round object was out there. Error is possible because there may be no object having the same essence as that contemplated in the meaning we have given to the elliptical image that was presented to us. A resolute skeptic who doubts all existence cannot be proved to be mistaken, but if he is consistent, he should be as skeptical of the existence of other minds and even of his own living self as he is of a physical world. Since the idea of change is no guarantee of actual change or process, he should arrive at an inarticulate solipsism of the present moment.

Yet there is no doubt that philosophers as well as laypeople normally believe in the existence of themselves and other minds and ascribe at least some of the characters they intuit to things that exist in space at present or past times. In memory and in the belief in history, the referent of present thoughts is a world of things and events believed to be existentially real and independent of any intuitions, present or past. An actual past or future is not given in any datum, but when one speaks, as David Hume did, of having or of being a succession of perceptions, one posits the existence of a temporal series of events and thereby instantiates in existence one or more essences. To ascribe existence to an essence as such would be a logical or categorial error; it would equally be a logical error to assert that an essence had been intuited by some mind or that some event or perception had occurred and at the same time deny that there is any factual temporal existence. The ontological status proper to essences is timeless subsistence. Actual intuitions come to exist on particular occasions, but knowledge of what they mean, says Santayana, "involves a leap of faith and action from the symbol actually given in sense or thought, to some ulterior existing object."

In Essays in Critical Realism Santayana argued that a child reaching for the moon is in quest of an object deployed in a physical world along with the outstretched arm and other bodies. If the moon did not transcend experience, if what is experienced were itself the object striven for, it would already be attained, and there would be no biological need to employ the presently intuited essence as a symbol for an existence still to be reached. There would be no knowledge about anything nor any need for it. If there is any validity in our scientific and commonsense beliefs, our intuitions are engendered in a biological organism by a natural environment. Matter in flux embodies now one essence, now another, and the set of propositions that describes all that exists at all times constitutes the realm of truth. Truth is therefore that part of the realm of essence that happens to characterize existence, and to have knowledge is to believe what is true.

But believing a proposition does not guarantee its truth, beyond the truth of the fact that it is believed. The terms of our beliefs are, in general, symbolic rather than literal representations of nature. Does it make any difference, then, if we clothe nature with intuited essences that are more fanciful than true as long as they are signals for successful action? If a pragmatist at this point suggests that truth means no more than the verification in later experience of the anticipated result of action guided by the earlier experience, the realist cannot agree. The pragmatist does agree with the realist that in knowing there is a reference beyond the immediate having of perceptions, but for the pragmatist the consummation of knowing, the successful working of an idea, does not go beyond experience; the referent of an idea is another experience. This avoids the problems of a mind-matter dualism and avoids the unanswerable question: How can we know when our ideas correctly represent external things? But the realist sees the pragmatist position as a reversion to idealism and subjectivism and will have none of it. If the pragmatist, to escape idealism, speaks in naturalistic terms, he admits all that the realist asks for. Lovejoy quotes William James: "Practically our minds meet in a world of objects which they share in common" (Essays in Radical Empiricism, New York, 1912, p. 79) to show that the practical man, going about his business of solving problems, must assume the existence of an external world; it is important that he discover what its properties were antecedent to and independent of the inquiry in which that discovery is made. Phenomenalism and positivism, sharing with pragmatism the view that the referent of all that can be meaningfully said about real existence must be, in principle, capable of being found in direct experience, are likewise rejected by the critical realist.

How, then, is knowledge of an external reality possible? The critical realist maintains that Locke erred in taking his own ideas to be the objects of knowledge. Knowledge, Locke said, is nothing but the perception of the agreement or disagreement of our ideas. When he comes to a discussion of "knowledge of real existence," however, he is forced to abandon his own definition, and true knowledge becomes the correspondence between ideas and external things. The critical realist argues that Locke should have recognized that when ideas are used in knowing, as distinguished from being merely entertained or had as an experience, there is always reference to an object other than themselves. But merely insisting that data have a referent beyond themselves does not tell us why we should believe one interpretation of them to be a truer description of the facts than any other interpretation. In his more skeptical mood Santayana tells us that knowledge is only faith mediated by symbols, yet in The Realm of Matter he sets forth what he takes to be the "indispensable properties of substance." Presumably, he means literally true properties. Substance has parts external to one another and, being in flux and unequally distributed, constitutes a spatial and temporal field of action. These are very nearly those primary qualities that Locke had said resemble the ideas the mind has of them, and if the critical realist seems to have a better case for his position than Locke had, it is chiefly because the sciences have supplied us with a detailed account of the mechanism of perception.

The scientist finds by actual experiment that the date of emission of light from the star, the distorting intervening media between the star and the observer, and the physiological peculiarities of the observer's body all condition what turns up at what time in the experience of the observer. But this scientific account cannot be used by the critical realist to support his position without begging the question. What is proved is that whenever something is found in our world, we can also find something else related to it; scientific knowledge consists of finding what is related to what. This supports the critical realist's thesis that experience depends upon a reality outside all possible experience only if it is assumed from the outset that the experimental data used by the astronomer and by the physiologist are experienced effects of a physical star and a physical organism. The scientist could interpret his explanatory theories on idealist or pragmatist, instead of on critical realist, assumptions. Hence, it is not what the scientist finds, but the epistemology he happens to assume, that supports critical realism. The best that can be said for this realist assumption is that it may be the most economical way to predict and control our experiences and that it may even be the truth about reality.

Differences among Critical Realists

Some of the critical realists, including Sellars, believe that their position is not best interpreted, even by some within their own camp, when a curtain of essences, ideas, or sense data is drawn between the perceiver and the objects he wants to know by means of such data. For in that case, as in Locke's representative perception theory, the essences or ideas are themselves the only possible objects of knowledge. Sellars would escape this difficulty by what he believes to be a more adequate account of perceiving. When a biological organism has sensationsthat is, is affected by an object in the environment with which it must come to termsthe sensation functions as information about the object that caused it. Perception is a response; it is an act of taking the sensation as the appearance of the external object. It is not the sensation or the sense datum that appears; it is by means of the sensation that the object appears. A sophisticated analyst might make the qualities of the sensation the object of his study, but then he is no longer using them to decipher things.

Sellars finds an ally in the English philosopher Gilbert Ryle, who follows common sense in the belief that we perceive trees and hands, not sense data. Here it would seem that Sellars has left the critical realists to join the direct realists, but he would insist that he is not taking either a direct presentational view or a Lockean representational view. The mediating role of sensation, which determines how the object will look, is not to be ignored. We look with our sensations but not at them.

In addition to some differences about the role of essences and of sense data, the critical realists are not all in accord on questions of metaphysics. Sellars and Santayana could be called metaphysical monists because for them only one kind of substancematterexists. The psyche of which Santayana speaks is the conscious material organism. Sellars thinks of the so-called mental functions not as being carried on by a substantial mind but as ways in which biological organisms, after a long evolutionary development, have learned to respond to stimuli.

Lovejoy, on the other hand, maintains that only a psychophysical dualism is a tenable corollary of an epistemological dualism; only a mind could have sensations and thoughts and intend or mean objects by them.

There has, then, been considerable divergence in the views of thinkers who were, and many who still are, called critical realists. Some have drawn closer to the positions of the direct realists in America or in Britain, and it may be that the label will cease to characterize a definite epistemology.


Boyd, Richard. "Scientific Realism and Naturalistic Epistemology." In PSA 1980: Proceedings of the 1980 Biennial Meeting of the Philosophy of Science Association. Vol. 2, edited by Peter D. Asquith and Ronald N. Giere, 613662. East Lansing, MI: Philosophy of Science Association, 1982.

Chisholm, Roderick. Perceiving: A Philosophical Study. Ithaca, NY: Cornell University Press, 1957.

Churchland, Paul M. Scientific Realism and the Plasticity of Mind. Cambridge, U.K.: Cambridge University Press, 1979.

Drake, Durant. Mind and Its Place in Nature. New York: Macmillan, 1925.

Drake, Durant et al. Essays in Critical Realism: A Cooperative Study of the Problem of Knowledge. London, 1920.

Dretske, Fred. Seeing and Knowing. London: Routledge and Kegan Paul, 1969.

Goldman, Alan. Empirical Knowledge. Berkeley: University of California Press, 1989.

Haack, Susan. Evidence and Inquiry: Towards Reconstruction in Epistemology. Oxford: Blackwell, 1993.

Hanley, Richard. "Much Ado about Nothing: Critical Realism Examined." Philosophical Studies 115 (2) (2003): 123147.

Kulp, Christopher, ed. Realism/Antirealism and Epistemology. Lanham, MD: Rowman & Littlefield, 1997.

Lovejoy, A. O. The Revolt against Dualism. New York: Norton, 1930.

Moser, Paul K. Knowledge and Evidence. New York: Cambridge Unversity Press, 1989.

Murphy, A. E. "The Fruits of Critical Realism." Journal of Philosophy 34 (1937): 281292.

Pratt, J. B. Personal Realism. New York: Macmillan, 1937.

Quinton, Anthony. The Nature of Things. London: Routledge and Kegan Paul, 1977.

Santayana, George. Realms of Being. 4 vols. New York: Scribners, 19271940.

Santayana, George. Scepticism and Animal Faith. New York: Scribners, 1923.

Schilpp, Paul A., ed. The Philosophy of George Santayana. Evanston, IL: Northwestern University, 1940.

Sellars, Roy W. "American Critical Realism and British Theories of Sense Perception." Methodos 14 (5556) (1962): 61108.

Sellars, Roy W. Critical Realism. Chicago: Rand, McNally 1916.

Sellars, Roy W. The Philosophy of Physical Realism. New York: Macmillan, 1932.

Sellars, Wilfrid. Science, Perception, and Reality. Atascadero, CA: Ridgeview Press, 1991.

A. G. Ramsperger (1967)

Bibliography updated by Benjamin Fiedor (2005)

Critical Realism

views updated May 23 2018

Critical Realism

Critical realism is a philosophical view of knowledge. On the one hand it holds that it is possible to acquire knowledge about the external world as it really is, independently of the human mind or subjectivity. That is why it is called realism. On the other hand it rejects the view of naïve realism that the external world is as it is perceived. Recognizing that perception is a function of, and thus fundamentally marked by, the human mind, it holds that one can only acquire knowledge of the external world by critical reflection on perception and its world. That is why it is called critical.


Critical realism arose in German philosophy in the late nineteenth and early twentieth centuries as a reaction to idealistic and phenomenalist types of philosophy. German critical realists took account of Immanuel Kant's (17241804) view of the subjectivity of knowledge but denied that this precludes access to "things-in-itself." In American philosophy, critical realism designates a movement initiated by Roy Wood Sellars (18801973) in 1916. It purported to integrate insights of both idealism and new realism, which was a naïve realist reaction to idealism. Through the work of Wilfrid Sellars (19121989), Roy Wood Sellars's son, critical realism influenced scientific realism, which arose in the 1950s in opposition to positivistic phenomenalism. Scientific realism basically claims that mature scientific theories are approximately true (in the sense of corresponding to the external world) and that their postulated central entities really exist.

The term critical realism was introduced into the dialogue between science and theology in 1966 by Ian Barbour. Barbour used the term to cover both scientific realism and a theological realism that takes seriously the cognitive claims of religion, that is, religion's claims to convey knowledge of a mind-independent divine reality. Subsequently Barbour pointed to the cognitive role of metaphors, models, and paradigms in scientific as well as religious language. His ideas were later assimilated and elaborated by Arthur Peacocke, John Polkinghorne, J. Wentzel van Huyssteen, and others. Actually, critical realism has been the dominant epistemology in the dialogue between science and theology for several decades. However, since the 1990s the transfer of critical realism from science to theology has increasingly been disputed, mainly on the ground that it does not, or does not sufficiently, do justice to the specific nature of theology.


On closer inspection, critical realism as a view of scientific and theological knowledge comprises three theses:

  1. Metaphysical realism, which holds that there exists a mind-independent reality. In scientific realism this reality is the material world; in theological realism this reality is the material world and also, primarily, God.
  2. Semantic realism, which holds that science and theology contain propositions, that is statements capable of being true or false in the sense of correspondence to the reality to which they refer. In scientific realism the focus is on propositions about unobservable entities; in theological realism the focus is on propositions about God.
  3. Epistemic realism, which holds that it is possible to put forward propositions that are approximately true, that some propositions actually are approximately true, and that belief in their approximate truth can be justified. In scientific realism this applies primarily to theories and theoretical propositions about unobservable entities; in theology it applies to propositions and theories about God.

The first thesis distinguishes critical realism from idealism and positivism, but also from Hilary Putnam's (b. 1926) "internal realism," which defines reality as a function of human conceptualization of the world. The second thesis distinguishes scientific realism from an instrumentalism that regards statements about unobservable entities as useful fictions without propositional content. Similarly, it distinguishes theological critical realism from the Wittgenstein-inspired view of religious language as mere expression or recommendation of a way of life. The third thesis distinguishes critical realism from a skepticism that affirms the first and second theses but denies that it is possible to acquire justified approximate knowledge of a mind-independent reality. On the other hand, the qualification "approximate" entails a dissociation from the naïve realist claim that reality is as it is perceived.


The main arguments in favour of scientific realism are:

  1. The fact that observation and experiments again and again compel scientists to change their prior ideas points to a substantive external input into science.
  2. The predictive success of mature theories can only (or at least best) be explained by the view that the processes, structures, and entities postulated by those theories approximate reality.
  3. The effectiveness of science-based technology can only (best) be explained by the view that mature scientific theories match nature to a substantive degree.

The main arguments against scientific realism are:

  1. Scientific theories are underdetermined by the empirical data; that is, the same data permit different theories that explain them. Therefore, empirical success is not a sufficient reason to assume that a theory is true.
  2. The history of science abounds with once empirically succesful theories that are now abandoned (e.g., a whole cluster of nineteenth-century theories assuming the existence of ether as a central entity). Therefore, empirical success is not a sufficient reason to assume that a theory is true.
  3. Scientific realism claims that those theories that offer the best explanation of the data are (approximately) true. This claim is thought to be supported by the argument that realism is the best explanation of the predictive success of science. However, this argument is viciously circular because it employs the kind of reasoning the validity of which it has to vindicate.

The main arguments in favour of transferring critical realism from science to theology are:

  1. Like science, theology makes cognitive claims.
  2. Science seeks to explain sense-experience with reference to the natural world, just as theology seeks, or should seek, to explain religious experience with reference to a divine reality.
  3. Both science and theology employ metaphors and models as approximative descriptions of an external reality.

The main arguments against transferring critical realism from science to theology are:

  1. Religious language has an expressive or recommending function, rather than a cognitive one. Therefore, theology should not be concerned with an external divine reality.
  2. Theology concerns itself with God, who is wholly different from the natural world, which is the subject matter of science.
  3. Theology cannot refer to a similar predictive success as science. Therefore, theology lacks a counterpart of the principal reason for a realistic view of science.

In evaluating a critical realist view of science and theology it may be useful to realize that the discussion of scientific realism has focused on scientific theories, especially on unobservable theoretical entities. One should not forget, however, that science is more than theories. It comprises also a wealth of observation statements and statements of primary relations, such as the statement that the specific gravity of lead is approximately 11.4. Although such statements are not theory-free, they will often have a realist plausibility that will even be acknowledged by most instrumentalists. As a consequence, a realist understanding of large parts of science seems to be a plausible option. However, scientific realism can hardly be a global view of science. Realistic plausibility has in principle to be established for each proposition and theory in particular. It would seem that this specification lessens the force of those arguments against scientific realism that aim at a global view.

As for the plausibility of transferring critical realism from science to theology, it should be realized that there are great differences between theology and science. As a reflection on religion, theology is primarily concerned with the question of the meaning of life, which implies that theology, unlike science, has an existential dimension. This does not, however, alter the fact that theological statements, insofar as they are propositions about God, make cognitive claims. Hence, critical realism is at least a logically possible view of theological propositions. But since God is not accessible to sense experience and experimental control, critical realism can hardly have the same rational plausibility for theology as for science. It would seem that a critical realist view of theology, or rather of particular theological propositions about God, is only a viable option within the context of faith.

See also Coherentism; Epistemology; Kant, Immanuel; Realism


barbour, ian g. issues in science and religion (1966). new york: harper and row, 1971.

barbour, ian g. myths, models, and paradigms: a comparative study in science and religion. new york: harper and row, 1974.

leplin, jarrett, ed. scientific realism. berkeley: university of california press, 1984

mcmullin, ernan. "enlarging the known world." in physics and our view of the world, ed. jan hilgevoord. cambridge, uk: cambridge university press, 1994

peacocke, arthur. intimations of reality: critical realism in science and religion. notre dame, ind.: university of notre dame press, 1984.

polkinghorne, john. one world: the interaction of science and theology. london: spck, 1986

psillos, stathis. scientific realism: how science tracks the truth. london: routledge, 1999

van huyssteen, j. wentzel. theology and the justification of faith: constructing theories in systematic theology. grand rapids, mich.: eerdmans, 1989

van kooten niekerk, kees. "a critical realist perspective on the dialogue between theology and science." in rethinking theology and science: six models for the current dialogue, eds. niels henrik gregersen and j. wentzel van huyssteen. grand rapids, mich.: eerdmans, 1998.

kees van kooten niekerk