Referring to the World
An Opinionated Introduction to the Theory of Reference
KENNETH A. TAYLOR
3
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2021
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data
Names: Taylor, Kenneth Allen, 1954–2019, author.
Title: Referring to the world : an opinionated introduction to the theory of reference / Kenneth A. Taylor.
Description: New York, NY, United States of America : Oxford University Press, [2021] | Includes bibliographical references and index.
Identifiers: LCCN 2020049975 (print) | LCCN 2020049976 (ebook) | ISBN 9780195144741 (hb) | ISBN 9780197537343 (epub) | ISBN 9780197537350
Subjects: LCSH: Reference (Philosophy)
Classification: LCC B105 .R25 T395 2021 (print) | LCC B105. R25 (ebook) | DDC 121/.68–dc23
LC record available at https://lccn.loc.gov/2020049975
LC ebook record available at https://lccn.loc.gov/2020049976
DOI: 10.1093/oso/9780195144741.001.0001
1 3 5 7 9 8 6 4 2
Printed by Integrated Books International, United States of America
5.3.
5.
7.
2.
The slow miracles of thought take shape through patience into grace.
Elinor Wylie,
“O Virtuous Light”
Foreword
On December 2, 2019, our beloved colleague and friend, Ken Taylor, announced to all of his Facebook friends that the book he had been working on for years, Referring to the World, “finally existed in an almost complete draft.” Completing references, eliminating redundancies, and writing a conclusion was all that remained. But that same day, while at home in the evening of December 2, Ken died unexpectedly.
Claire Yoshida and Kiyoshi Taylor located the computer files for Ken’s book for us. The book you have before you consists of those files. We have corrected typos, added some extra commas in some of Ken’s long sentences and some extra quotation marks for the many terms he introduces, made numbering of examples and such things consistent, and filled in the references as best we could. Some of the footnotes were notes from Ken to himself about references to add, or points to elaborate on, without enough detail for us to confidently fill in the missing information. We left these as they were written and they are set in small caps to indicate this. The index was compiled by Ken Perry of CSLI Publications. Krista Lawlor and Debra Satz helped in a number of ways.
The book abounds, as one would expect, with insights, examples, trenchant analyses of criticisms of scores of philosophers, and subtle theorizing that weaves this all together in a new and exciting framework for thinking about reference and many related issues. Sadly, it lacks a concluding chapter.
Anna-Sara Malmgren, John Perry, Mark Crimmins, Robin Jeshion
Ken Taylor’s Facebook Post, December 2, 2019.
Now this calls for a minor celebration
My long delayed book, Referring to the World, finally exists in an almost complete draft. It just needs a preface and a conclusion and a beefed up scholarly apparatus of notes and citations
This creature started out its life years and years and years ago as a completely different book. It was first commissioned and conceived of as an opinionated introduction to the theory of reference. it was supposed to be a SHORT
introduction too . . . of no more than 60,000 words. I thought I could write such a thing pretty quickly
Maybe I could have. But the sprawling book that I actually wrote became more and more opinionated and much more focused on arguments designed to back up those often philosophically controversial opinions. Somewhere along the line—I don’t know when exactly—it gradually became less and less of an introduction to anything except my own views. Some sort of Sorites Paradox lies therein perhaps
To be fair to me, it still does contain substantial traces of its original introductory design. Because throughout most of it, it tries hard to be scrupulously dialectally fair to competing points of view—all of which, though plausible and appealing, are ultimately wrong
The opinions and the supporting arguments have been developed over way too many years, in a series of articles, some of which were written with the book in mind others of which were not. But now they are all gathered together in a single if somewhat sprawling argumentative thread. Maybe some will find them more convincing that way. But maybe not. We shall see . . . I hope.
It will still take some time to spit and polish it all up into a form suitable for shipping off to the publisher. But god it feels good to get this way too long delayed book project in close to final form. I think I’ll pour a glass of wine to mark the occasion, before plunging back into the work that is still to be done.
One of my favorite poems includes the line “the slow miracles of thought take shape through patience into grace.”
Amen to that!
The Mystery of Reference and Objective
Representational Content
This book is an investigation of the mystery of reference in both private thought and public talk. Just what is it for some bits of either our shared public language or our inner thoughts to refer to or stand for bits of the world? In virtue of what does the relation of reference obtain between some bit of the world and some bit of either outer language or inner thought? What about apparent reference to putatively non-existent objects, like Santa Claus or Sherlock Holmes? We appear to think and talk about objects that do not exist. But there are no such objects. So just how do we manage to think and talk about them? Or consider abstract objects, like numbers, that are thought by some to exist outside the spatial-temporal order. We appear to think and talk about such objects as well. But it is a mystery how, if at all, the reach of our thought can extend beyond even the bounds of space and time. Such questions are tied up with some of the deepest, most challenging, most enduring problems of philosophy. They concern how it is possible for the mind and language to make contact with an apparently mind-independent reality. In one form or another, this problem has bedeviled the imaginations of philosophers as diverse as Aristotle, Plato, Descartes, Hume, Berkeley, Kant, Russell, Frege, Heidegger, Husserl, and Wittgenstein. To be sure, the commonality of concern is sometime masked by stark differences of philosophical idiom. The philosophers of the early modern period, for example, worried incessantly about our “ideas” and about their “resemblance,” or a lack thereof, to a possibly mind-independent reality. Partly because an idea is more likely to resemble some other idea than it is to resemble any mindindependent reality, the idiom of ideas proved not to be an altogether stable and fruitful means of explaining the relationship between mind and world. Moreover, partly because of the temptation to see ideas as essentially private, it was hard for the moderns to offer any convincing account of how ideas could serve as the basis of shared contact with a mind-independent world.
Referring to the World. Kenneth A. Taylor, Oxford University Press (2021). © Oxford University Press. DOI: 10.1093/oso/9780195144741.003.0001
Early in the twentieth century, with the beginning of the so-called linguistic turn in philosophy, the idiom of ideas was supplanted. The idiom of language, of words and sentences, sometimes with a stress on their logical forms, sometimes with a stress on their everyday use and abuse, came to occupy center stage. This mere shift of idiom was widely advertised as a great and clarifying revolution in philosophy. Though this revolutionary shift did not on its own suffice to instantly render the mind’s power to think and talk about a mind-independent world more conceptually transparent, it did help to dispel any lingering temptation to explain the mind’s ability to think and talk about the world in terms of the resemblance between its representations and its representeds. Words and sentences are clearly the vehicles or instruments of the mind’s thought and talk. They are the mind’s representations. They are not the mind’s objects, not its representeds. Once this insight has been gained, it is a short step to seeing that there need not be any deep resemblance between the vehicles of thought and talk and the objects of thought and talk. Taking that short step does not, however, settle the question of how possibly words and thoughts manage to be about objects in a mind-independent world.
Though this book addresses a number of discrete questions about reference, the answers I proffer to them stem from a single underlying view about the source and nature of what I call “objective representational content.” By that I mean the property that our words and our thoughts have of being “semantically answerable” to how things are by objects in the world. Consider the following example. There is a person Smith with whom I am acquainted. I have many beliefs about her. One such belief is well-expressed by my utterance of the form of words, "Smith has a generous spirit and a quick and incisive intellect." Now the truth or falsity of my words depends entirely on how things are by Smith and her properties. My words are true just in case Smith, and no one else, has just those qualities, and no other, of spirit and intellect. How things are by Smith with respect to her qualities of spirit and intellect determines how things are in the way of truth and falsity by my thoughts and my words. I mean to point only to unremarkable facts of this kind when I say that our thoughts and our words are semantically answerable to how things are by objects and their properties.
A major goal of this book is to explain how possibly both sentences in our public language and the representational vehicles of private thought, whatever they are, achieve this remarkable property of semantic answerability to the world.
My explanation of how our thoughts and words manage to be semantically answerable to the world will turn, in part, on the fact that the vehicles of thought and talk are structured wholes built out of parts that are themselves semantically valued. For example, the sentence “Smith has a generous spirit and a quick and decisive intellect” contains as constituent parts the name “Smith” and the predicate “. . . has a generous spirit and a quick and decisive intellect.” Intuitively, the name has the semantic function of standing for or referring to a certain object, viz., Smith, while the predicate has the semantic function of expressing a certain property. Moreover, the grammar of our language entails that a sentence containing a name and a predicate arrayed in just the way that this name and this predicate are arrayed may be used to make a statement and express a belief about the referred to object to the effect that it has the expressed property. Such an approach proceeds in three steps. First, it appeals to a theory of what we might call “semantic value” for the basic constituents of our language. The theory of semantic values should include at least a theory of reference for the referring expressions of our language and a theory of “expression” for the predicates of the language. Second, it should include principles of what we might call syntactic combination. Such a theory would explain how less complex expressions can be combined to yield more complex expressions. Finally, the theory would include principles of semantic composition. Principles of semantic composition would determine the semantic values of whole sentences as functions of the semantic values of the constituent parts of those sentences and the way those parts are organized to yield wholes.
Though the approach just gestured at will strike many as highly intuitively compelling, there are those who take it to be misguided from the very start. One goal of this book is to assess both the considerations that weigh in favor of my own preferred approach and also those considerations that weigh against it. As such, though the book sets out to defend a peculiar and novel approach to the problem of objective representational content, it should also be serviceable as something of an introduction to the foundational issues facing any theory of reference. My aim is not, however, to provide a survey of the ins and outs of various alternative approaches to the problem of objective representational content. My aim is, rather, to think through, in a systematic way, the range of problems with which any theory of objective representational content must deal and to give a variety of reasons for preferring one particular approach—what I call two-factor referentialism—to its main
rivals. While I do attempt to give rival approaches a fair hearing, I make no claim to neutrality.
Prominent among those who reject the kind of explanatory approach defended in this book is Donald Davidson. Davidson readily admits that our beliefs and our utterances are true or false of a world that is mostly independent of mind. And to that extent he agrees that our thought and talk have what I call objective representational content. But he claims that concepts such as reference play no essential role in explaining the relationship between mind and/or language, on the one hand, and the world, on the other. As he puts it, “We don’t need the concept of reference; neither do we need reference itself, whatever that may be.” (Davidson, 1977, 256) Similarly, Robert Brandom claims that the supposed reference relation is “a philosopher’s fiction, generated by grammatical misunderstandings.” (Brandom, 1994, 324) Though Brandom affords what he calls an “expressive role” to what he calls “reference talk,” he, like Davidson, denies that there is a genuine reference relation. Moreover, he denies as a consequence that any such relation has any role in explaining the relationship between mind and/or language and the world.
Views that seek to explain where objective representational contents come from, but without appeal to anything like a reference relation, pose a far-reaching challenge to the explanatory strategy favored in this book. Our strategy takes reference to be both a real relation in nature—though admittedly one that is itself in need of explanation—and a relation that can be used, once it has itself been explained, to shed further explanatory light on the nature of objective representational content. This approach presupposes that although reference is itself neither an explanatorily nor metaphysically primitive notion, it is more metaphysically and explanatorily basic than the objective representational contents of whole sentences or entire thoughts. Call a theory with this character a “bottom-up” explanatory theory of reference. In Chapter 3, we shall dig more deeply into Brandom and Davidson style arguments for thinking there can be no bottom-up explanatory theory of reference. For now, it is enough to note that one major part of what is at issue between friend and foe of a bottom-up explanatory theory of reference is a dispute about the significance of what might be called the priority of the sentence. It is hard to deny that names and predicates perform their semantic work primarily in the context of the production of sentences. Except in very special circumstances, the utterance of an isolated name or an isolated predicate would not yet constitute the staking out of a determinate claim about an
object and its properties.1 Strikingly, this seemingly trivial observation leads some to the consequential conclusion that a referring expression, taken on its own, outside the context of a sentence in which it is deployed, is entirely semantically idle.
Wittgenstein (1953) famously put the point as follows:
Naming is so far not a move in the language-game—any more than putting a piece on a chessboard is a move in chess. We may say: nothing has so far been done, when a thing has been named. It has not even got a name except in a language game.2
Wittgenstein apparently thinks that an expression functions as a name or referring expression only if it plays a (yet to be specified) role in an already constituted “language game.” If we take a language game to be something like a totality of significant linguistic moves or speech acts, as they are often called—where these include acts of asserting, promising, requesting, to name just a few—then Wittgenstein’s claim would seem to be twofold. First, acts of referring happen only in the performance of already significant speech acts—paradigmatically, perhaps, acts of assertion. Wittgenstein seems to want to say, that is, that referring is not some separable act, somehow antecedent to any speech act. It is not that one first refers to Socrates and then makes an assertion about him. Rather, one refers to Socrates only by and in the course of asserting something or other about him.3
Now many are tempted to conclude, on the basis of such considerations, that there can therefore be no separable reference relation, specifiable independently of the specifying of a space of actual and possible speech acts.4 On this sort of view, talk of the reference of a name must be understood entirely derivatively, entirely in terms of the role played by the name in a totality of sentences, the potential utterances of which constitute a space of already linguistically significant speech acts. Reference is not, in other words, explanatorily prior to asserting and other speech acts, but is, at best,
1 Editorial note: Some of the footnotes were notes from Ken to himself about references to add, or points to elaborate on, without enough detail for us to confidently fill in the missing information. We left these as they were written and they are set in SMALL CAPS to indicate this. (See Foreword.)
Reference John Perry’s “Davidson’s Sentences and Wittgenstein’s Builders” (1994) and the primitive language games described therein. Note that when there is an articulate background of prior “issues” (Who just entered the room is a question in the air), the utterance of a name does count as a significant linguistic performance.
2 Wittgenstein, 1953, §4.
3 Reference to Searle on referring vs. predicative speech acts.
4 Give examples of those who endorse the view. Quine, Davidson, Brandom, others?
explanatorily dependent on them and, at worst, an altogether explanatorily idle notion.
In Chapter 3, I shall say what can and can’t be said for the principle of the priority of the sentence and for the correlative claim that reference is at best a derivative and at worst an entirely idle notion. I shall grant that there is a sense in which the principle is true, perhaps even trivially so. It seems exactly right to say, for example, that neither we theorists nor the child acquiring her first language can gather direct evidence about the reference of a name, say, or the properties expressed by a predicate, say, except by witnessing the use of the predicate or name in the context of sentences of various sorts. This fact points to what I call the dynamic priority of the sentence. The dynamic priority of the sentence is a straightforward consequence of the hardly unexpected fact that significant linguistic performances typically involve the production of whole sentences. But I shall show that the dynamic priority of the sentence does not support skepticism about the bottom-up explanatory utility of reference in an account of objective representational content.
Arguments against bottom-up explanatory theories of reference from the supposed priority of the sentence purport to show that not just reference but any other semantic property attaching in the first instance to mere constituents of sentences rather than to whole sentences themselves must be devoid of both independent metaphysical standing and foundational explanatory utility. My main argument against the priority of the sentence is a blocking argument. That is, I attempt to show that to the extent that the priority of the sentence is true, it does not, in fact, entail that reference has no foundational explanatory utility. To be sure, even if my blocking argument succeeds in keeping open the question of whether reference has some foundational explanatory work to do in the theory of objective representational content, that will not suffice to solve the more fundamental puzzle of just what in the order of things the reference relation could possibly be. Solving that puzzle requires us to explain just how names and other referring expressions could possibly manage to have the property of standing for objects (and the correlative puzzle of how predicates manage to express properties) to begin with. Pointing the way toward the unraveling of that puzzle is one of the main burdens of this book, though I do not flatter myself that I have penetrated to the bottom of the mystery of reference.
Before launching head-on into the main arguments of this book, I spend much of the remainder of the current chapter dwelling on just how puzzling the fact of objective representational content, and with it the reference relation, really is. Viewed in one way, our words are just marks on the paper
or perturbations in the air or patterns of nerve firings in our brains. Why should such marks and perturbations be semantically answerable to any independent existents at all? What endows different kinds of expressions with their distinctive roles in bringing about this semantic answerability? What exactly is a predicate, for example, such that it has the semantic role of expressing a property? What exactly is a name such that it has the semantic role of standing for an object? And what exactly is a sentence such that its truth or falsity determinately depends on how things are by this or that object and its properties.
Standing in the philosophical wake of Kripke’s Naming and Necessity, one’s first thought may be that the key to answering such questions lies with matters causal and/or informational.5 Our linguistic output is enmeshed in myriad causal and informational chains. At one extreme of such chains stand thinking and speaking agents—stocked with their beliefs, desires, and intentions. At the other extreme stand events in the world. Worldly events initiate an inward rush of energy upon the portals of sensation. This inward rush gives rise to perception, which eventually occasions vast torrents of language, thought, and action in response. And it is not an entirely unpromising thought—one that lies at the foundation of all forms of empiricism—that it is in the causal interaction of mind and world via the portals of sensation that objective representational thought is somehow first constituted. But this suggestive initial thought carries us only so far. That is because the inward rush is more occasion than direct cause of either outer linguistic production or inner content bearing thought episodes. Outer linguistic production is not directly caused by events in the outer world but by our own inner intentions, desires, and beliefs. Our episode beliefs and judgments do not, as such, directly “encode” information about the world. A speaker may say that it is raining outside because she believes that it is raining outside and intends to inform her hearer of that fact. And she may want to inform her hearer that it is raining so that the hearer will act on that information by, for example, grabbing her umbrella or putting on a raincoat. But the crucial point is that she may say, believe, and intend all these things even in the absence of rain.
Now many philosophers hold that inner episodes of believing, intending, or desiring themselves involve a play of inner representations, connected to the outer world via intricate causal and informational chains, ultimately anchored in perception of and action in and upon an external world. Because of the causal role of beliefs, intentions, and desires in bringing about
5 Kripke (1980).
utterances, many a philosopher has held that the semantic answerability of sentences in a public language depends on their being instruments for expressing certain already contentful inner thoughts. This approach makes thought prior to language in the semantic order. There is, I think, something deeply right about this approach—though as always in philosophy, there are those who intensely disagree.6 But let us bracket such disagreements for the moment. If for the nonce we take it to be given that thought is prior to language in the semantic order, it follows immediately that if we could but tell a systematic story about how the association between sentences, on the one hand, and already semantically answerable thoughts, on the other, is achieved, we would have gone part way toward explaining how sentences, at least, manage to achieve objective representational content.
But telling this much of the story would still leave us far from our ultimate goal of explaining the ultimate source and nature of semantic answerability, for reasons that will become clear shortly. Yet telling even this incomplete fragment of the ultimate story would itself be no trivial matter. For one thing, the number of thoughts potentially thinkable by minds like ours is infinite. So too is the number of sentences whose objective representational contents we potentially understand. Yet our minds are finite. The mapping between thought and language could not, therefore, be given by a mere list. Such a list would have to be infinite. No merely finite mind could grasp such a list via the exercise of merely finite cognitive powers. If the infinite mapping between thought and language is to be graspable by a finite mind, there must be a finite principle, graspable by us, that somehow generates and explains that infinite mapping.
A promising strategy for dealing with the infinity problem was articulated by Frege early on and given fuller expression in more recent times in the work of Jerry Fodor.7 That strategy posits a language of thought. The idea is that to think a thought is to deploy an inner sentence of so-called mentalese, the inner idiom of thought. Like the sentences of our shared language, the “sentences” of mentalese are supposed to be structured wholes built out of discrete and semantically significant parts. If there is at least a rough correspondence between the constituents of the sentences of mentalese and the sentences of the public language, and if rules of composition at the level of
6 Examples. Connection to those who accept the priority of the sentence
7 Reference to Frege. Reference to Fodor. Quote passage where Frege seems to argue for something like LOT as a contingent psychological hypothesis about cognizers like us
thought correspond to the rules of composition at the level of language, then, perhaps, we can give a systematic account of why a given sentence expresses just the thought it expresses and no other.8
Though it is not without its detractors, the language of thought hypothesis has, on my view, a great deal to recommend it and for much of the course of the argument of this book I will help myself to the assumption that there must be something like a language of thought. To be sure, this assumption does not solve our hardest problems. For even if we could uncover some finite principle that generates and explains the potentially infinite mapping between outer public languages and an inner mentalese, we would have succeeded only in pushing the deeper problem back one step. We would still be left with the unanswered question of what renders our inner thoughts themselves, rather than just the outer sentences that express those thoughts, semantically answerable to a mind-independent reality. This book proposes some features of the language of thought that arguably make it possible for our thoughts and our words to be semantically answerable to a mind-independent world of objects and their properties. I don’t pretend to have a knock down argument in favor of the story I tell in this book. But I shall argue that my story is superior to the competing alternatives.9
Now the problem of objective representational content has so thoroughly bedeviled the philosophical imagination that some philosophers have rejected the very idea that thought and language are semantically answerable to a mind-independent reality. The most explicit and prominent advocates of such views are idealists of various stripes.10 Idealists typically find the idea that our thoughts are answerable to something lying outside of the mind nearly incoherent. The most immediate object of thought, the idealist maintains, must be something like our own ideas or our own sense data. To be sure, on some versions of this tale, we may have an indirect hold, via the ideas on which we have a direct hold, on that which lies outside the mind. But no matter the precise form it takes, idealism is an extreme and desperate doctrine. Though many a philosopher has claimed to have decisively refuted it, I myself suspect that it is very likely with us to stay, in some form or other, at least as a permanently tempting philosophical option, available
8 More on the language of thought hypothesis
9 That, in a way, is small praise. Explaining where objective representational content comes from and how it fits in to the order of things, turns out to be a very hard problem indeed. As a consequence, even moderately convincing hypotheses are thin on the ground.
10 Berkeley and descendants
when the going gets rough. I do not pretend to offer a decisive refutation of all forms of idealism. Because I doubt that idealism can be decisively taken off the table of dialectically available options, I will be content if I manage to diminish whatever dark allure idealism continues to enjoy. Even if I cannot demonstrate the outright falsity of idealism, it will be enough for my dialectical purposes if we can show that nothing in the philosophical common ground between idealists and non-idealists is threatening to the claim that our words and thoughts are semantically answerable to a world largely independent of the mind. There is, I shall argue, a coherent story to tell about how possibly mind and language achieve semantic answerability to a mindindependent world. That story is not threatened by any philosophical or empirical premises plausibly available to both idealists and non-idealists alike. Moreover, I shall argue that that coherent story is plausibly true. Showing that nothing in the common ground between idealists and non-idealists threatens the idea that our thought and talk are semantically answerable to a mind-independent world does not constitute a direct refutation of idealism. But it does show that nothing in the philosophical common ground makes idealism inevitable. And since idealism is a corner into which no one should willingly retreat, merely stripping it of any claim to inevitability should suffice to diminish its dark allure.
One need not go all the way over to idealism, however, to experience a sense of puzzlement about the mind’s ability to reach beyond itself and hold itself answerable to an external reality. Viewed in one way, episodes of thinking amount to nothing but the energized sloshing of a vat of chemicals locked up in our skulls. How possibly could such energized goings-on confined to our skulls be semantically “answerable” to anything at all, let alone to realities beyond themselves? Of course, viewed in one way, the mind is surely more than just a random vat of energized chemicals sloshing about. The mind is, I think, rightly and informatively characterized as a field of inner representations. Its inner states are symbolic in that they have both (syntactic) form and (semantic) content. Understood in this way, standing for objects, representing or being about mind-independent facts, expressing mind-independent properties are of the very essence of both the mental and the linguistic. Though this is all arguably correct as far as it goes, it only goes so far. Without some further explanation of how configurations of neuronal states and processes constitute themselves as symbolic and representational, the mere insistence that they do it somehow or other has all the philosophical benefits of theft over honest toil.
To acknowledge the need for honest toil in our philosophizing about the representational powers of mind and language is not to deny those very powers, however. The mind’s representational powers are at the very core of our cognitive and social lives. Without the ability to cognize the world via inner representations, all knowledge would be impossible. Without our shared ability to represent a common external world, and to meta-represent our own representations of that common world, human social life as we know it would not be possible at all. We could never communicate with one another about the world, never engage in rational collective action on a common world.
Conceiving of a brain as a mind, rather than as a mere vat of energized chemicals is, in part, conceiving of it as the seat of representations and representational capacities.11 As a thoroughly modern naturalist, I endorse the view that the mind’s representations and representational capacities must ultimately arise out of merely material processes and states of our brains. But qua philosopher, rather than neuroscientist, it’s not my job to tell you the details of how actually the brain manages to constitute the mind. Philosophy is typically more concerned with the how possibly than with the how actually. Moreover, in the particular case at hand, there are good reasons to postpone a frontal assault on the how actually questions for another day. That’s because the mystery of objective representational content isn’t fully ripe for a frontal assault by those concerned with how the actual processes and states of the brain constitute the representational capacities of the mind. This is not to deny that there is, in our times, much revealing scientific, as opposed to philosophical, work currently being done on the mind and its representational powers. But for better or for worse, at our stage of the history of inquiry, problems about the source and nature of objective representational content remain pretty much a matter of determining “how possibly at all.” As such, addressing them is very much the business of philosophy.
The problem of objective representational content remains mostly a how possibly problem despite the fact that the cognitive revolution of the last century has opened up an entirely new conceptual space. That revolution was in part a rediscovery of the lost wisdom of an earlier day. It amounted to a reclaiming of modern philosophy’s foundational insight that the mind is a field of inner representations, with real causes and effects in the world. As strange as it now seems, for a large chunk of the last century, especially
during the period when both philosophical and scientific behaviorism reigned supreme, this simple but crucial insight was widely derided and dismissed. It was thought both to be a source of deep philosophical confusion and to be a bar to scientific progress. But the cognitive revolution has achieved much more than simply reclaiming the lost wisdom of earlier centuries. It has conceptually augmented and refined that ancient wisdom. The best fruits of this revolution have been not just the actual detailed theories and models of the cognitive workings of the mind-brain, but also the generation of whole new families of heretofore unimagined—perhaps even unimaginable—hypotheses. Still, even with the ever-increasing consolidation of the explanatory gains of the cognitive revolution, there remains considerable conceptual distance between the concepts with which we cognize the biochemistry, functional anatomy, and even computational architecture of the working brain and the concepts with which we cognize the representational and semantic capacities of mind and language. Partly as a consequence, we still don’t know how to re-identify the brain states and processes that we pick out via our neurochemical, neuroanatomical and even neurocomputational concepts as the contentful mental states that we pick out via our ordinary mentalistic concepts. As a consequence, we are more or less stuck with philosophizing, with asking how possibly at all, at least when it comes to explaining how both mental and linguistic content arises out of facts about the intact working brain.
The key to making philosophical progress toward the day when a more frontal assault on the details of the “how actually” question is in order is to reduce conceptual distances. We can do so by self-consciously and reflectively evolving a set of intermediate or “bridging” concepts. With proper deployment of an appropriate set of bridging concepts, we may hope to gain fuller imaginative acquaintance with the possibility that what we have heretofore conceived of as two metaphysically independent domains is really a single domain, antecedently conceived in two distinct ways, with two distinct and heretofore uncoordinated sets of concepts. To that end, this book articulates a set of concepts specifically designed to more fully enable us to imagine what in the natural order reference and related semantic relations might be. We do not initially conceive semantic notions like “standing for,” “being about,” “expressing,” and the like in terms that readily allow us to see how the semantic arises out of or reduces to the natural. Because we lack antecedently available, ready-to-hand concepts that bridge the gap between the semantic and the natural, we simply don’t initially know how to re-identify reference
and other semantic properties as aspects of the natural order. If we are to bridge that gap, we need some new concepts. I hope to provide a few such concepts in this book. By providing such concepts and showing that they plausibly have application to possible worlds not entirely unlike our own, I hope to convince you at least that objective representational content really could arise of out of the natural order, with no supernatural or non-natural intervention or additions. Even if I succeed in convincing you that the semantic could arise out of the natural order, I won’t have told you the details of how it actually does so. But it is not unreasonable to hope that the stock of concepts I develop in this book may ultimately prove to have application not just to nearby possible worlds but to our very own as well. And if that is so, they may also prove to be of use in answering the “how actually” question about objective representational content. But that is an argument for another day. For the nonce, I will be satisfied if the reader gains fuller imaginative acquaintance with the possibility that reference and related semantic properties and relations are a real part of the natural order. She should thereby be left with no lingering temptation to appeal to certain misbegotten doctrines to explain how objective representational content fits into the natural order of things.
This last point bears stressing. Precisely because of the distance and lack of systematic coordination between our concepts of material nature, including the material nature of the brain, on the one hand, and our concepts of mind, language, and their contents, on the other, philosophers have sometimes felt less than fully confident about the places of mind and language in the natural order. Such qualms are most loudly and explicitly voiced by self-professed dualists, like Descartes, who famously thought that “mental substances” and their states were metaphysically distinct from material substances. Nothing with the properties characteristic of thought and thinking, he argued, could possibly be resident in a merely material substance. These days, Cartesian dualism is a distinctly minority position. But even some self-professed materialists, who have no truck with Cartesian substance dualism, sometimes deny the possibility of illuminating, non-question-begging explanations of the objective representational contents of our thoughts.
Consider, for example, the family of philosophers we might call the interpretationists, including Quine, Dennett, and Davidson. Each of these philosophers is deeply committed to the metaphysical doctrine of materialism in one way or another. Nothing non-natural has any place in any of their fundamental ontologies. But each has argued that there is really no
independent fact of the matter about the contents of our thoughts and/or words. Each does allow that there is an interpretive practice of ascribing contents to thoughts and words. They also insist that our interpretive practices are well-nigh indispensable. Dennett and Davidson, for example, see our interpretive practices as inextricably tied up with our cognition of ourselves as rational cognizers and actors. Rational cognizers are pushed and pulled not merely by the blind forces of material nature but by the force of the better reason. Absent our interpretative practices, we could appeal to only the blind operations of nature in explaining human behavior. If we would cognize human cognition and action as rational, as responsive to the force of the better reason, we must, the interpretationist will say, interpret. That is, we must assign (objective) representational contents to our thoughts and our words. But for all their insistence on the indispensability of interpretations, interpretationists typically maintain that interpretations are infected with a deep and ineliminable observer relativity. Different interpreters may be fully justified in ascribing incompatible mental contents to one and the same subject. And this supposed observer relativity and indeterminacy of all interpretations is supposed to render interpretations if not quite fictional or illusory, nonetheless not quite part of the “objective” furniture of the universe either.
There are other self-professed materialists who, though conceding both the determinacy and objective reality of objective representational content, seem, nonetheless, to regard the objective representational contents of our thoughts as all but inexplicable. John Searle is the perhaps unintended paradigm here. Searle has argued that what is distinctive about thought is that it enjoys what he calls intrinsic, as opposed to derived intentionality. As such, thoughts are supposed to stand in sharp contrast to sentences. Unlike sentences, thoughts do not have their representational content derivatively, in virtue of being associated with some other already contentful things. This claim amounts to an assertion of the priority of thought over language. About this much, Searle seems to me to be entirely correct. But he also holds something stronger and more questionable. He apparently believes that just because thoughts enjoy intrinsic rather than derived intentionality, there can be no illuminating explanation of how objective representational content arises out of that which is not yet intentional. Contra the dualist, however, Searle does not claim that intentionality sits outside natural order. Indeed, on Searle’s view, the brain literally secretes intentionality and consciousness. Moreover, unlike interpretationists like Quine and Davidson, Searle insists
that intentionality is a genuine and determinate causal power, somehow resident in or at least caused by the brain.
But it seems to me that Searle’s confidence that he has located intentionality in nature is misplaced. He is insufficiently puzzled about just how the vast vat of neurotransmitters, electrochemical impulses, and synaptic junctions that make up our brains could have the startling property of secreting states with objective representational content. Since Searle is famous for arguing that all other supposedly naturalistic theories of mind— save his own—fail to respect the intrinsic intentionality of the mental, that charge may seem somewhat surprising. But despite Searle’s self-declared naturalism, his insistence that the mind is intrinsically intentional misleads him into treating the intentionality of thought as if it were a metaphysical surd, one neither subject to nor in need of deeper explanation and explication in other terms.12
Searle’s notion of “intrinsic intentionality” in fact brings to mind an important and philosophically challenging view first articulated by Brentano. Brentano was no materialist. He was, rather, a kind of idealist and also a kind of dualist. (Brentano, 1995/1874) Brentano’s dualism is shown by the fact that he took the mental to be fundamentally and categorically distinct from the physical. As a consequence, he also took the mental to be irreducible to the physical. In particular, he claimed that the objects of our thoughts are defined by the property of what he called “intentional inexistence”:
Every mental phenomenon is characterized by what the Scholastics of the Middle Ages called the intentional (and also mental) inexistence of an object, and what we would call, although not in entirely unambiguous terms, the reference to a content, a direction upon an object (by which we are not to understand a reality . . .) (emphasis added) or an immanent objectivity. Each one includes something as an object within itself, although not always in the same way. In presentation something is presented, in judgment something is affirmed or denied, in love [something is] loved, in hate [something] is hated, in desire something is desired, etc. This intentional inexistence is exclusively characteristic of mental phenomena.
(1995/1874, 68)
12 This is in the main because of his views about the connection between intentionality and consciousness. It is really consciousness, in particular the first-person point of view, that is the primary metaphysical surd for Searle. Insert quote from Searle and explicate quote
