Metaphysics and the Representational Fallacy

Heather Dyke

• Publisher: Taylor & Francis
• Year: 2012
• Print ISBN: 9780415956697
• eBook ISBN: 9781135910297
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophical Metaphysics

This book is an investigation into metaphysics: its aims, scope, methodology and practice. Dyke argues that metaphysics should take itself to be concerned with investigating the fundamental nature of reality, and suggests that the ontological significance of language has been grossly exaggerated in the pursuit of that aim.

1 Metaphysics and the origins of the representational fallacy

The term “the representational fallacy” refers to a general philosophical tendency to place too much emphasis on the significance of language when doing ontology. We shall see, in the course of this book, that this tendency can manifest itself in various ways, and with different results. To begin this investigation, it will be instructive to examine some of the movements and revolutions in the recent history of philosophy that led to the representational fallacy getting a grip on current philosophical thinking.

The Kantian turn

The starting point for an investigation into the origins of the representational fallacy is the work of Immanuel Kant. One of the driving forces behind Kant’s philosophical program was the desire to rescue human knowledge from the perils of skepticism. Prior to Kant, the threat of skepticism was thought to be particularly acute, largely as a result of the view that humans do not directly perceive reality; what we have direct epistemic access to are our perceptions, or sense impressions. The human mind, philosophers reasoned, is presented with perceptual impressions picked up by the sense organs and not with reality itself. That being the case, what grounds have we for the claim that there really is a reality out there at all, let alone one that is causally responsible for our sense impressions, or that is adequately represented by them? Kant sought to undermine this skeptical threat by focusing attention not on reality but on our perceptual impressions themselves which, due to our direct access to them, are not similarly threatened by skepticism.

Another driving force behind Kant’s philosophy was a desire to avoid the excessive metaphysics of the rationalist philosophers. By anchoring human knowledge to the perceptual and conceptual features of human experience, which are universal, in the sense of being common to all humankind, and to which we have direct epistemic access, he intended to exclude from serious consideration the elaborate, speculative metaphysical systems like those of the rationalists. These were merely ad hoc philosophies; what Kant wanted was a philosophy grounded in certainty and supported by empirical evidence.

Unfortunately, Kant’s intention to avoid excessive metaphysical systems backfired, and spectacularly so. His strategy of anchoring human knowledge to the perceptual and conceptual features of human experience involved arguing that the objects of human knowledge are mere appearances, in part constituted by the knowing mind, and not to be confused with objects that are independent of human knowledge. He thus distinguished between the world of appearances, the “phenomenal” world, and the world of things-as-they-are-in-themselves, the “noumenal” world. The only world we can have knowledge of is the phenomenal world. Furthermore, the phenomenal world, being the object of our knowledge, is partly constituted by the imposition on our sensory input of our a priori, categorizing concepts.

There are two features of this strategy that were seized upon by those wishing to perpetuate the excessive metaphysical system-building that Kant was seeking to avoid. First, by drawing the distinction between the phenomenal and the noumenal worlds, Kant’s view permitted speculation about the transcendent realm of reality that lies beyond mere appearance. Thus, instead of undermining speculative metaphysics, Kant’s philosophy legitimized it as a discipline distinct from science, with its own subject matter and methods of enquiry.

The second feature of Kant’s strategy that caused it to backfire was his claim that phenomena, the objects of knowledge, are partly constituted by the knowing mind. Kant argued that we cannot have knowledge of anything without using our conceptual apparatus, and he concluded that the objects of our knowledge are the product of the operation of our conceptual apparatus upon the information received via our senses. Thus, far from being independent of humans, and part of the external world, the things we have knowledge of are, at least in part, the product of human conceptual activity. Now, for Kant, the conceptual apparatus employed in generating the objects of human knowledge is universal, so the products of this activity are the same for everyone. However, many neo-Kantians dropped this feature of Kant’s philosophy, while retaining the idea that the objects of knowledge are generated by the imposition of our concepts and categories on sensory input. The upshot of this deviation from Kantianism was that humans, or cultures, with different conceptual schemes would generate different objects of knowledge; different worlds. Thus, Kantianism gave way not just to elaborate metaphysical theorizing as to the nature of the noumenal world but also to a brand of relativist constructivism, according to which, in so far as human beings think about the world in different ways, they actually live in different phenomenal worlds.

Despite its popularity, one of the central arguments for Kantian and neo-Kantian metaphysics is deeply flawed. This is the argument from the premise that we cannot know things without using our conceptual apparatus, or employing our cognitive faculties, to the conclusion that we cannot know things-as-they-are-in-themselves; we can only know things-as-conceived-of-by-us. This argument was identified by David Stove (1991: 140) and labeled the Gem.¹ The premise of the Gem is a tautology. It states that we cannot conceive of things without employing our conceptual apparatus, or, alternatively, we cannot think of things without bringing them under the categories of our thought, or, what comes to the same thing, we cannot think of things without using our minds. Its conclusion, however, is not tautological. It is that we cannot conceive of things that exist independently of us, or things-as-they-are-in-themselves. The tautological premise concerns the human cognitive activity of knowing things, or conceiving of things. The conclusion concerns the objects of our knowledge, or conceptual activity. Since no tautological premise can, by itself, validly yield a non-tautological conclusion, the Gem is an invalid argument.²

The impact of the Kantian and neo-Kantian episode in the history of metaphysics on the development of the representational fallacy is most noticeable, as we shall see in the following section, in the way in which the logical positivists, and their successors, reacted against it. However, before moving on to that stage in the history of metaphysics, it is worth reflecting briefly on the role of representations in the Kantian conception of metaphysics. According to Kantians, the aim of metaphysics is not to investigate the nature of mind-independent reality; mind-independent reality is essentially (and almost by definition) inaccessible to human minds. Consequently, if that were the goal of metaphysics, it would be a futile enterprise. Instead, the object of metaphysical investigation is the human conceptual apparatus itself. Furthermore, since, for the Kantian, the world of appearances is partly constituted by the operation of the human conceptual apparatus, by studying the latter we will, in effect, be studying a world of some sort: the phenomenal world.

The only legitimate object of metaphysical enquiry, then, according to the Kantians, is the human conceptual scheme, and the representations it consists of. Such an enquiry cannot provide us with knowledge of the reality that these representations are representations of, even for those who accept that there is such a thing. The representations thus constitute a barrier between us and the world, making it essentially inaccessible to us. Thus, the Kantian is, in a sense, immune from committing the representational fallacy. The fallacy is only committed when one claims to derive ontological conclusions from premises about representations of reality. The Kantian, however, denies that any ontological conclusions can be drawn from any premises. Her study of our representations is purely intended to achieve a picture of the conceptual scheme which we employ to represent, and partially construct, the phenomenal world. Given the Kantian’s denial that we can have any knowledge of mind-independent reality, it is natural for her to focus her attention on the nature and structure of our representations, since that is all we can have knowledge of, according to her. It is possible that this focus on representations and conceptual schemes within the Kantian tradition has encouraged a general acceptance of the view that representations are a legitimate object of metaphysical enquiry. However, I do not think that this is the principal source of the representational fallacy. We shall move a step closer to identifying that by examining how the logical positivists reacted against neo-Kantian metaphysics.

The linguistic turn

Just as Kant had tried (and failed) to rescue philosophy from the speculative metaphysical system-building of the rationalists, the logical positivists attempted to rescue philosophy from the speculative metaphysical system-building of the neo-Kantians. As we have seen, Kant’s distinction between the phenomenal and noumenal worlds had opened up new possibilities for extravagant metaphysical theories about the ultimate nature of reality. Furthermore, these theories were not subject to any empirical standards for assessing their truth, since empirical standards applied only to investigations into the phenomenal world. All this horrified the scientifically minded positivists, who sought to find a way to throw out the entire neo-Kantian metaphysical enterprise. They focused their attention on the nature of the philosophical project itself.

The positivists concentrated on language, and on its logical analysis. There are a number of reasons for this. One significant one was that the method of analysis had already delivered promising results when deployed against certain unattractive metaphysical theories. Russell, for example, had shown that Meinong’s ontology of nonexistent objects could be avoided by adopting his account of the logical form of sentences involving definite descriptions that have no referent (Russell 1905). His theory of types (1903: 523–8) was seen to offer an even more devastating weapon against unsavory metaphysical systems. Russell employed that theory to argue that certain set-theoretic statements, which appeared to be grammatically correct, could be shown, by logical analysis, to be meaningless strings of symbols.

The positivists leapt upon this strategy of using logical analysis to demonstrate not merely the falsity but also the meaninglessness of whole classes of assertions they deemed to be unsavory, and it became central to their overall aim of undermining metaphysics. Russell’s deployment of his theory of types showed that it was possible for statements that appeared to be perfectly coherent and meaningful to be shown to be, as a matter of fact, incoherent and so meaningless. The positivists were thus able to target entire realms of discourse with which they took issue and to argue that they were no more than meaningless nonsense.

Another reason for, or perhaps an outcome of, the shift in philosophical focus to questions of language was that it gave the positivists away of securing a legitimate domain of enquiry for philosophy, distinct from that of science, while being able to avoid the speculative metaphysics they so deplored. As we have seen, the neo-Kantians saw philosophy as prior to science because it investigated the ultimate nature of reality, while science merely investigated the nature of phenomenal reality; a distinctly second-class activity. The positivists denied that philosophy was prior to science in this sense, upholding science as the privileged intellectual discipline, but they still wanted to recognize a role for the philosopher. They found a means for distinguishing the two domains by appealing to the idea that before one is in a position to determine whether a statement is true or false, one must at least have understood what it means. They took this distinction between meaning and truth to reflect the different domains of investigation of the philosopher and scientist. The philosopher, by her investigation into which statements are meaningful and which are not, defines the limits of scientific investigation. Only those statements deemed meaningful by the philosopher are worthy of investigation into their truth or falsity by the scientist.

Whatever the reasons for this “linguistic turn” – the shift in philosophical focus to questions of language – it is important to note that it did not merely generate a single view of the philosophical significance of language. Many of the positivists, as we have seen, used linguistic and logical analysis to argue that metaphysical claims were meaningless. Other philosophers, for example Russell, used it not to attack metaphysics, but to do metaphysics; to argue against one ontological view and in favor of another. The later Wittgenstein had different motives again for focusing his attention on language. He thought that the role of philosophy was to uncover the true logical form of ordinary language statements and that this would have the effect of dissolving all philosophical problems, which were thus revealed to be merely “pseudo-problems” (Wittgenstein 1953). Other philosophers, for example, Ayer, rather than rejecting all philosophical claims as meaningless assertions about the extralinguistic world, attempted to reconstrue what he took to be the legitimate ones as significant assertions about language itself (Ayer 1946). He says, for example, “What has contributed as much as anything to the prevalent misunderstanding of the nature of philosophical analysis is the fact that propositions and questions which are really linguistic are often expressed in such a way that they appear to be factual” (Ayer 1946: 57–8).

Not only were there, in the aftermath of the linguistic turn, multiple views of the philosophical significance of language, and of what we could establish by focusing our attention on it, there were also two different views of the relationship between language and extralinguistic reality. The official line of the positivists was that philosophical assertions about the nature of extralinguistic reality were meaningless. The only meaningful philosophical assertions were those about language itself. Thus, philosophical investigation into the world as it really is was systematically replaced by philosophical investigation into the language we use to talk about the world.

The positivist view on the relationship between language and reality is captured by the idea that all knowledge of the world is relative to language. Any claim apparently about the world is either meaningless or, if meaningful, is in fact a claim about how we use language. This idea was most fully developed by Carnap, and his views on this matter received their most developed presentation in his paper “Empiricism, Semantics and Ontology” (1950). In that paper, Carnap argued that apparently ontological questions, such as “Are there numbers?” are not to be understood as questions about the nature of extralinguistic reality. Understood in that way, they would be meaningless. Instead, they can be given two possible interpretations. They can either be interpreted as internal questions, or as external questions. If they are treated as internal questions, then they are questions that are asked from within some linguistic framework. Treated in this way, the answer to them will be determined by appeal to the framework itself. So, from within the linguistic framework of mathematics, the question “Are there numbers?” receives an affirmative answer simply because the linguistic framework itself employs number terms. If they are treated as external questions, they are pragmatic questions about whether or not we ought to adopt the linguistic framework of mathematics. So, metaphysical assertions, such as “There are numbers,” which appear to be about reality, are treated either as assertions about the rules of particular linguistic frameworks or as pragmatic proposals for adopting those linguistic frameworks.

For the positivists, then, language constitutes a barrier between us and extralinguistic reality. Any claims we might wish to make about the world are, in fact, claims about language, and any knowledge available to us about the world is relative to a linguistic framework. There is no sense in which we can find out about the world as it is independently of any linguistic framework. Given this picture, the positivists, just like Kant, are effectively immune from committing the representational fallacy. If we simply cannot find out about the world as it is independently of any linguistic framework, then we cannot find out about the world by examining language.

The alternative view of the relationship between language and extralinguistic reality that emerged from the aftermath of the linguistic turn was that, although the proper object of philosophical investigation was language, this investigation could yield results about the nature of extralinguistic reality. For example, Russell and the early Wittgenstein thought that there was a correspondence between the ideal language (the language obtained by applying the methods of logical analysis to ordinary language) and the reality that it represented, such that it was legitimate to infer the nature or structure of the world from that of the ideal language. For them, the logical structure of the propositions of the ideal language correspond to, or “picture,” the ontological structure of the facts or states of affairs they represent. For Wittgenstein, for example, the propositions of the ideal language picture possible states of affairs, and those that are true picture actual states of affairs. By examining the true propositions of the ideal language, then, we are effectively examining pictures of reality.

Russell, too, thought that we could infer the nature or structure of the world from that of language. In An Inquiry into Meaning and Truth, for example, he writes, “Some modern philosophers hold that we know much about language, but nothing about anything else.…For my part, I believe that, partly by means of the study of syntax, we can arrive at considerable knowledge concerning the structure of the world” (Russell 1940: 347). Not only does Russell here explicitly deny the positivist idea that the extra-linguistic world is inaccessible to us, he actually endorses the contrary idea that there is a definite relation between the logical form of sentences and the ontological form of the facts they represent. He further thinks that by examining the logical structure of sentences we can learn about the ontological structure of the facts represented.

Another philosopher who endorsed the idea that there is an ideal language was Bergmann. He thought that the ideal language is the one that provides a true representation of the extralinguistic world. Bergmann’s view was, thus, in stark contrast to that of Carnap. According to Bergmann, one does not choose the linguistic framework of the ideal language on purely pragmatic grounds. Instead, one discovers that it is the one language that provides a true representation of reality. His view as to the relationship between the ideal language and the extralinguistic world was that, “philosophical discourse is not just about the ideal language but rather, by means of it, about the world” (Bergmann 1959: 93).

According to this brand of philosopher, whom we might label “picture theorists,” language provides a picture of the world. The picture provided by ordinary language is imperfect, and it is the job of the philosopher to clarify it until a perfectly accurate one is arrived at. This is done by reformulating ordinary language, by means of logical analysis, into the ideal language, which is the one that presents the one true picture of reality. The true picture of reality is one that, by its logical form, represents the ontological form of reality. These philosophers thought that the ideal language mapped so perfectly onto reality itself that the ontological features of reality could be ascertained by examining the logical features of the ideal language.

The positivists were unhappy with the claims to metaphysical knowledge of the picture theorists, and also with the idea that there is only one true, or ideal, language. The picture theorists were, in turn, unhappy with the positivist conclusions that there can be no knowledge of reality independent of any linguistic framework, and the inevitable relativism that this generated. They were also unhappy with the positivist idea that there was no privileged linguistic framework. Just like the positivists, they were keen to banish the speculative metaphysical system-building of the neo-Kantian tradition and return philosophy to a more...

Table of contents

• Cover
• Title
• Copyright
• Contents
• Preface
• Introduction
• 1 Metaphysics and the origins of the representational fallacy
• 2 A new metaphysical strategy: lessons learned from the philosophy of time
• 3 The representational fallacy: or how not to do ontology
• 4 The relationship between language and reality
• 5 The methodological map
• 6 The overlooked strategy in practice: moral discourse
• 7 Some further applications of the overlooked strategy
• Notes
• References
• Index