Hypothesis and Realism: A Comment on Sutton

David Kenneth Johnson

Introduction

Robert Sutton's recent contribution to these pages hopes to consign fallibilistic (or hypothetical) realism to the dustbin of "irrelevant Western ideas" (Sutton, p. 40). Concerned about the logic of combining realism (the epistemological-metaphysical view that we can know objective reality) with fallibilism (the view that many, if not most, of our views could be false) he writes:

...continuing to support realism while, at the same time, claiming that it does not deliver the goods (true and reliable knowledge of an objective reality) sounds like a religious claim of faith or philosophical mythology much more than a useful idea (p. 41).

It seems clear, however, that fallibilistic realism can usefully accommodate both non-mystical, perfectly straightforward talk of objective truths and the possibility of (even massive) perceptual or conceptual error. Realism is speculative, hence fallible, as it provides hypotheses (or maps and plans) for locating ourselves in our natural and social surround. Realism is objective, as it confirms and extends our understanding of a world not entirely of our making.

Of course, the various parties to this dispute will offer very different accounts of the success and accuracy of these plans and maps, choosing to emphasize either the realistic discovery of a world that contains and explains human experience or, on the antirealist (but still pragmatic) side, the active imposition (Putnam), interpretation (Rorty, Gadamer), or construction (Carnap, von Glasersfeld) of our many "experiential worlds." Success, these theorists hold, cannot serve as evidence for the representational accuracy of our maps, since truth and meaning are either "internal" to the map or "there is no such thing as the world for anything to conform or fail to conform to" (Goodman 1972 p. 31; see also Maturana).

I assume that a truly pragmatic epistemology or ontology will owe nothing to these internalist varieties. Contemporary realism re-establishes its traditional connection to a generally pragmatic outlook in this way: All reflection begins with and receives its confirmation from practice, though the success of our plans and maps speaks to more than the bare fact that they allow us to "cope." Indeed, as commonsense would rightly have it:

Errant speculation is quickly exposed: The instrumental relations specified by a plan do not obtain; or we are frustrated, wet, and hungry when the map for applying it misdirects us. Behavior is successful only as our maps and plans look beyond personal concerns to things in themselves. It is the truth of these maps and plans, and not only our need and convictions, whether personally or socially founded, that explains the success of our thought-directed behavior (Weissman 1989b, pp. 28-9).

Two related antirealist concerns emerge. First, on the internalist assumption that language- or concept-users create, at least in part, the objects and relations of their maps and plans as they chose one or another conceptual scheme, the notion of objectivity no longer requires reference to "things in themselves." As Sutton notes, antirealists typically assume a (pragmatic, coherentist) account of objectivity or truth that purportedly avoids all talk of mind-independent objects (or things in themselves). It follows, second, that the mind- or subject- or language-dependency claimed for the antirealist's objects and relations is preferable to realism's "impossible epistemological perspective" which claims to identify the "way things really [objectively] are" (Sutton, p. 41).

These charges will prove effective against only a naive version of realism. Accordingly, my defense of hypothetical realism begins with the admission that there can be no unmediated or direct access to the world.

Realism and Hypothesis: Four Guiding Principles

The hypothetical realism I defend sets out from four basic principles. It would require several books to defend and explain adequately the realistic use of hypothesis. (In fact, I think David Weissman has done just that.) I will only briefly describe these principles here, characterizing them as assumptions, and refer the reader to more detailed treatments elsewhere. My more limited goal is to defend realism against Sutton's charge that it cannot "deliver the goods"; that is, provide true and reliable knowledge of an objective reality while remaining committed to fallibilism.

(1) Most generally, I assume that philosophy's fundamental role is a theoretical one. More exactly, philosophy is a kind of narrative: We philosophize as we recommend and defend a conceptual network designed to tell a coherent story about the world. While knowledge of that world is our goal, inquiry to justify our beliefs remains our instrument, as these stories represent the extensions of our first successful accommodations to reality. When do these stories begin? I think it likely that infants are capable of telling themselves private stories of the most Spartan kind, having their source in pre-linguistic and genetically programmed sorting behavior (see Russman 1987, p. 12; Weissman 1989b, p. 198). There will be no identifiable point at which those stories become scientific or philosophical; rather, our accounts evolve in complexity as we move from our first behavior-directing plans through science to the speculations of philosophy. At the furthest reaches of thought's generality, these are the stories of metaphysics, where we have extended, by the methods of abstraction and variation, our information about our place in nature, its form, and conditions. But, even here, theory's bond to our original maps and plans is undeniable, as these stories have their more remote source in our first, successful inferences and generalizations about the world and our place there:

Science and metaphysics are overlapping orders of reflection coming after the maps and plans of ordinary practice. Indeed, much of the evidence relevant to our metaphysical hypotheses will have been available at the moment of our first encounters with the world... [F]or we have no detailed, a priori knowledge of the world. We can only speculate about its character, using whatever hypotheses we have for finding our way, here in the middle of things. This tentative, speculative attitude infuses all of our reality-testing behaviors (Weissman 1989b, pp. 118-131).

One hypothesis (or map, to continue Weissman's metaphor) suggests that a particular rose is red; another that Sutton is a philosopher. We often have good reason to say (and very little reason not to say) that these, and infinitely many other stories, provide true and reliable knowledge of a mind-independent, hence objective, reality.

(2) Despite the apparent diversity of philosophical methods for investigating the nature of things, I would suggest, along with Engels and Weissman, that our choice of methods reduces to two: We inspect the things set before us when they have been given to or "constructed" by our minds, so that every difference creditable to the world has its source in thought or language; or we use thoughts, words and percepts as signs, creating thereby the plans and hypotheses representing possible states of affairs and locating us in a world we have not made (Weissman 1989b, p. 1). Realism chooses to exceed the (immediate) data: Fearing solipsism or the delusions of a self-certifying method, we prefer the uncertainty (fallibility) of hypothetical realism to the certain, though restricted (because anti-metaphysical) scope of its rival. Exceeding the data, we explain it: "Sutton is a philosopher" is true if there exists, objectively, a philosopher named Sutton.

(3) From the assumption that humans are natural creatures and a product of evolution, it follows that mind-as essentially the activities of a physical system-ought to be located, naturalistically, among the things we can know and discover. It follows, too, that knowing must be some kind of natural relation that sentient creatures bear to their world, so that

...knowing must be something that [humans have] been doing all along....and that has adapted [them] to that world, by contrast with which not knowing, being ignorant, is something objectively different and less advantageous (Millikan 1984, p. 7).

Millikan's naturalism rightly suggests that realist metaphysics is relatively late in the order of inquiry. In particular, the mind- or language-independence of the world does not emerge as a discovery of metaphysics:

This is the inference of everyone frustrated when acting on a plan, or damaged after tripping in the dark. The world does not work as we supposed, or it does work as we expected but not as we can manage, or we had no idea the furniture had been moved. Realist metaphysics comes after the time when reality and our need for accommodating to it are already acknowledged. Its theories confirm and complete an understanding that does not wait for metaphysics to direct it (Weissman 1989b, p. 203-4).

This is apparently the impulse behind one of Michael Devitt's "naturalistic maxims": settle the realism issue before any epistemic or semantic issue (Devitt 1984, pp. 3-4). Refusing to follow any "linguistic turn" in philosophy that claims to have dispensed with extra-linguistic reality (Rorty, Dummett), we say that these semantic theories (like all scientific theories) derive their sense in part from their (perhaps tacit) acceptance of realism. On this view, the world pre-exists and is a condition of the possibility of Sutton's antirealist efforts to do away with that world.

(4) Lastly, I will assume that nothing is ever known directly. Rather, all of our contact with the world is mediated by thoughts, words, and percepts construed as signs-signs, moreover, having referents distinct from themselves. They are the natural signs of perception and the conventional ones of thought and language. Signs serve as the "vehicles of our thinking," while all of the things that they might signify (or represent) remain the proper objects of thought (Weissman 1989b, p. 160; see also Millikan 1984, Part II). We organize these signs into meaningful hypotheses as we speculate-in practice, science and metaphysics-about a world we have not made.

Realism, therefore, claims to "deliver the (objective) goods." Yet the speculative nature of these reality-testing behaviors is undeniable, as our every factual claim and perceptual judgment is a hypothesis addressing a possible state of affairs. "That rose is red" as much as "Sutton is a philosopher" are speculative in this way, as each assertion involves a factual claim about a possible state of affairs that may or may not obtain. A properly fallibilistic realism will confront directly the notorious over-determination of evidence by theory, since our claims about the world are forever exceeding the possible evidence for them. It is sensible, I suppose, to say that some reactions to this phenomenon are excessive, as the idea, popular among some philosophers, that we might all be "brains in vats" (or, more commonly, that some contingent fact about ourselves limits that which we can know) seems to be. But in the end, lacking all access to a detailed, a priori illumination of the world-or even of ourselves and our place there-we should admit the possibility that we could be wrong (or right) about anything. It is an objective feature of the real, according to this view, that our access to its features is forever mediate and fallible.

References

Bhaskar, Roy. (1978). A Realist Theory of Science. Sussex, Great Britain: Harvestor Press.

Carnap, Rudolf. (1967). The Logical Structure of the World. R. A. George, Trans. Berkeley: University of California Press.

Devitt, Michael. (1984). Realism and Truth. Princeton: Princeton University Press.

Dummett, Michael. (1978). Truth and Other Enigmas. Cambridge: Harvard University Press.

Engels, Frederick. (1978). Ludwig Feuerbach and the Outcome of Classical German Philosophy. New York: International Publishers.

Gadamer, Hans-Georg. (1975). Truth and Method. New York: The Seabury Press.

Goodman, Nelson. (1972). Problems and Projects. Indianapolis: Bobbs-Merrill.

Johnson, David K. (1993). "The Metaphysics of Constructivism." Cybernetics and Human Knowing, 1, 4.

Kordig, Carl. (1971). The Justification of Scientific Change. Boston: Reidel.

Maturana, Humberto. (1978). "Biology of Language: The Epistemology of Reality." In G. A. Miller & E. Lenneberg, Eds. Psychology and the Biology of Language and Thought. New York: Academic Press.

Millikan, Ruth G. (1984). Language, Thought, and Other Biological Categories: New Foundations for Realism. Cambridge: MIT Press.

Pols, Edward. (1982). The Acts of Our Being: A Reflection on Agency and Responsibility. Amherst: University of Massachusetts Press.

Putnam, Hilary. (1987). The Many Faces of Realism. LaSalle, Illinois: Open Court.

Rorty, Richard. (1981). The Consequences of Pragmatism. Minneapolis: University of Minnesota Press.

Russman, Thomas. (1987). A Prospectus for the Triumph of Realism. Macon, GA: Mercer University Press.

Sutton, Robert C. (1995). "Realism and Other Philosophical Mantras." Inquiry: Critical Thinking Across the Disciplines, XIV, 4.

von Glasersfeld, Ernst. (1987). The Construction of Knowledge: Contributions to a Conceptual Semantics. Seaside, CA: Intersystems Publications.

Weissman, David. (1989a). "Metaphysics After Pragmatism." The Review of Metaphysics, 42.

Weissman, David. (1989b). Hypothesis and the Spiral of Reflection. Albany: State University of New York Press.

Weissman, David. (1987). Intuition and Ideality. Albany: State University of New York Press.

Weissman, David. (1977). Eternal Possibilities: A Neutral Ground for Meaning and Existence. Carbondale: Southern Illinois University Press.

Endnotes

1 One clear example of what I oppose is (the later) Putnam's a prioristic emphasis on the necessary "internal" determination of all reference relations between the sentences of a theory and pieces of the world (Putnam 1987, pp. 23–32). See also Devitt 1984, Ch11.

2 In Contrast, the antirealist Richard Rorty claims that "modern science does not enable us to cope because it corresponds, it just enables us to cope" (Rorty 1981, p. xvii).

3 Common to many varieties of antirealism is the assumption that we cannot express in language what is extra–linguistically the case. Yet beneath this claim about language lies a "repressed" claim that conflicts with the relativistic strictures of the first, namely: "that we can know what the status of language -- its epistemic, its ontological status -- in fact is; that we know what is the case about language; that we know what the truth of the matter is" (Pols 1982, p. 89).

4 The main outlines of my defense and characterization of realism owe much to the work of David Weissman, especially Weissman 1989a; 1989b; 1987; and 1977.

5 I assume those first reflections, arising out of the urgencies of practical life, often involve the use of abductive (or "retroductive") inference. Those inferences, well known to science and practice, have the following form: (1) Some unexplained phenomenon P is observed; (2) P would be explicable as a matter of course if H were true; hence (3) there is reason to think that H is true (Kordig 1971, p. 12; see also Bhaskar 1978). Thought is speculative in this way when we infer from something thought or observed to its conditions: "Seeing smoke, we think of fire. Considering mind's various ways of behaving, we ask if these might be the activities of an exclusively physical system.... [And in all practical thinking] we infer from a result desired to the sequence of behaviors required for achieving it" (Weissman 1989b, p. 27).

6 Frederick Engels' "two great camps thesis" suggests that the history of philosophical thought centers on a series of disputes between materialism (or realism) and idealism (or antirealism) (1978, p. 21).

7 Hoping to secure their every knowledge claim against skeptical arguments, antirealists move ever closer to a constructivist idealism where mind has created every intelligible difference within the matters it thinks or describes (Weissman 1987, p. 104; Johnson 1993, p. 32). All versions of constructivism will have, like the realism I defend, an instrumental, pragmatic expression. These constructivist appropriations of pragmatism are not, however, theories of things and relations used (like a map or tool) in our dealings with an extra-linguistic or extra-mental world. While hypothetical realism relies on practice and inquiry as a test of truth, constructivist pragmatics sees language or thought as the mind's principal instruments in creating a thinkable experience: "mind," individually or corporately, creates a thinkable because coherent experience meant to satisfy our interests and needs.

Thanks to Kathleen R. Johnson and Matt Silliman for helpful comments on earlier drafts of this paper.