The Twin Earth Chronicles

Twenty Years of Reflection on Hilary Putnam's the "Meaning of Meaning"

Andrew Pessin, Sanford Goldberg

• Publisher: Routledge
• Year: 2016
• eBook ISBN:
  9781315284798
• Topic: Philosophie
• Subtopic: Weltgeschichte

In 1975, Putnam published a paper called The Meaning of 'Meaning', which challenged the orthodox view in the philosophies of language and mind. The article's Twin Earth conclusions about meaning, thought and knowledge were shocking. This work contains writings on the subject of Twin Earth.

Part III
Mental Content and Mental Causation

Introduction

IN PART III, WE MOVE away from the philosophy of language and focus more directly on some of the implications of the Twin Earth stories for the philosophy of psychology. These fall into two main, related categories.

The first set of implications concern mental contents. Many mental states, such as beliefs, desires, and thoughts, have the interesting property of being about things. To believe that water quenches thirst is to be in a state that is, loosely, “about” at least water, thirst, and quenching; to desire that your spouse will be rewarded with a promotion is to be in a state that is “about” your spouse, promotions, and being rewarded. We typically ascribe such states (called “intentional states”) to agents by using sentences of the form “X believes (desires, thinks, etc.) that p,” where X is an agent and p is the proposition that X believes (desires, etc.). This proposition is known as the “content” of the state, and, according to many, is essential to that state: two agents (or one agent at two different times) will be in the same intentional state if and only if they are in states with the same content. The Twin Earth stories, then, raise many issues important for developing an understanding of the nature of that content.

Perhaps the most controversial issue is whether, or to what degree, the agent’s external environment is relevant to the determination of the content of his intentional states. On the one hand, the environment is relevant in at least the sense that the entities that intentional states are about are frequently external to the agent; on the other hand, the environment is irrelevant in at least the sense that which mental state(s) an agent is in, including intentional state(s), seem to depend on nothing more than what is inside the agent’s head. Given these competing constraints, many have found it necessary to distinguish different kinds of content. “Wide” (or “broad”) content is highly sensitive to environmental context: one can change an agent’s wide content by changing (in appropriate ways) only her external context. “Narrow” content, to the contrary, is sensitive to, or dependent on, only the agent’s internal context: one can change an agent’s narrow content only by changing, for example, her brain. Applied to the Twin Earth stories, then, two Twins, each saying to herself, “I could use some water,” will differ in wide mental contents and so in wide mental states (since, in their different external contexts, what the Twins desire is different, viz. H₂O v XYZ), while they will be identical in narrow mental contents and thus in narrow mental states (since the Twins are molecular duplicates and as such share all inner states). The key question, then, is this: Which of these competing views of content (if either) is most appropriate for psychology?

Those who answer in support of some version of narrow content are called “internalists” or “individualists,” while those who support versions of wide content are called “externalists” or “anti-individualists.” In the selections that follow, we will see much discussion of this key question (and its relatives) by members of both camps, as well as by at least one author who resists the very dichotomy itself. Must we indeed make the wide-narrow distinction, or can we get by without it? If we must make the distinction, how will we go about defining adequate notions of either kind? Are there, perhaps, different sorts of wide content, or of narrow content, each with its own merits and disadvantages? Does narrow content even qualify as “content”? How reliably do the “that-clauses” in sentences ascribing mental states track mental content? How do the different conceptions of content cohere with other important notions, including physicalism, supervenience, explanation, and especially, as we will see in a moment, causation? And what, if anything, does commonsense, “folk” psychology, have to say about all these issues?

The second set of Twin Earth’s implications for the philosophy of psychology concerns the issue of mental causation. One common intuition is that mental states have causal powers. That is, many authors maintain that we typically behave as we do because we have the beliefs and desires we do. More specifically, we behave as we do in large measure because our beliefs and desires have the contents they do. The problem is, however, that it is no simple matter to make sense of how our mental states and their contents can have these causal powers in our seemingly very physical world.

The idea that they do have causal powers is problematic for a number of reasons. Foremost amongst these is the intuition that if we are to avoid an intolerable dualism, then there simply is no room for the mental to be causally efficacious since physical states and properties seem to do all the causing there is; the physical world is causally “closed.” In other words, in what way can the content of an agent’s intentional state be at all relevant to the state’s causal powers when the neurons appear to do all the causing within the agent? Indeed, the problem of accounting for mental causation is even more imposing for externalists since wide content, with its sensitivity to external factors, invokes more than what is in the agent’s head—and if mental states are to have any causal powers at all, it would seem, they could plausibly do so only as mediated by the agent’s brain, which is comfortably nestled well inside the agent’s head. In short, the problem for externalists isn’t limited to merely the apparent superfluousness of mental content, but includes as well the fact that (wide) mental content is not exclusively dependent upon the neural.

In the selections below, then, we’ll also see discussion of the following sorts of questions: What role, if any, should causal powers play in individuating mental states? How does the answer to this question bear on which notion of mental content is most appropriate for psychology? Is wide content even capable of being causally efficacious? How does the notion of causation cohere with those other notions mentioned above, i.e., physicalism, supervenience, and explanation? Is there any way to reconcile the causal relevance of the mental without merely identifying it with or reducing it to the physical? [For more on these and related issues, cf. Heil and Mele (1993).]

Having sketched some of the relevant general issues in Part III, we now turn to a brief overview of the articles.

In Tyler Burge’s first selection in this volume, excerpted from his classic article “Individualism and the Mental” (1979), he argues against “individualism,” here understood as the thesis that mental content depends on nothing more than what is inside the agent’s head. At the heart of the argument is a three-step thought experiment. We are first asked to imagine an agent with a large number of beliefs about “arthritis,” (as we would ascribe them), one of which is false—viz., his belief that he has developed arthritis in his thigh (since, in fact, arthritis is a disease of the joints only). Second, we are, asked to imagine a counterfactual situation in which we have just the same agent, with the same physical history and properties, but with a different social environment: in his community, “arthritis” is used to denote a disease that can occur not only in the joints, but also in the thigh. Finally, we are invited to join in an interpretation of these two steps, according to which the counterfactual agent lacks some or all of the arthritis beliefs of the original agent, due to his lack of the appropriate concept arthritis. (Although the counterfactual agent lacks the concept arthritis, note that he does possess a similar concept—say, tharthritis—of a disease that can occur in the joints and in the thigh.) If, as Burge argues, this interpretation is correct, then individualism must be false because “social context infects … mentalistic attributions.”

While Burge’s thought experiment has much in common with Putnam’s Twin Earth thought experiments, Burge notes that his is more general. Putnam’s analyses apply primarily to natural-kind terms (and concepts); Burge’s apply also to artifact terms, color adjectives, social-role terms, abstract nouns, action verbs, legal terms, etc., and their corresponding concepts. Similarly, while Putnam is primarily concerned with differences between Twins’ physical environments, Burge explores the implications of differences in social/linguistic environments as well. Putnam and Burge, in short, reach conclusions roughly similar in general externalist tone, but they do so via very different considerations and with very different emphases.

Burge addresses Putnam’s Twin Earth analysis more directly in his “Other Bodies” (1982), our next selection, and in so doing further develops the anti-individualism sketched in “Individualism and the Mental.” Burge notes that in “The Meaning of ‘Meaning’,” Putnam interprets the Twins’ different uses of “water” purely as a difference in extension (cf. Crane’s selection below—eds.); further, he notes that Putnam ascribes the Twins the very same thoughts. On Burge’s externalism, to the contrary, the Twins differ not merely in the extensions of their terms, but in their thoughts as well. Declining to speculate “why Putnam did not draw a conclusion so close to the source of his main argument,” Burge instead criticizes those aspects of Putnam’s discussion that obscure that conclusion. These include Putnam’s claims that natural-kind terms are implicitly indexical and that mental states “narrowly individuated” do not “fix” the extensions of relevant agents’ terms.

Regarding the first claim, Burge protests that the word “water” does not act like ordinary indexicals, such as “here.” If it did, then just as “here” shifts its extension with context, so also would “water.” But then when an Earthling visited Twin Earth and said of some XYZ that it was “water,” he would be speaking truly—even though there is no water on Twin Earth! Thus, Burge concludes, any difference between the Twins is not due to the context-relativity of indexicals.

Regarding the second claim, Burge first suggests that Putnam’s “narrow mental states”—those that don’t presuppose the existence of anything other than the individual in that state—are equivalent to de dicto, nonrelational intentional states. But then on this construal, he argues, the Twin Earth examples fail to show that an agent’s narrow states don’t fix the extensions of his terms since (in effect) a distinct narrow state corresponds to each distinct extension. After further analysis, he concludes that the contents of all of an agent’s intentional states involving natural kinds do indeed presuppose entities other than the agent. If to be narrow is not to so presuppose, then there just aren’t any narrow intentional states.

Daniel Dennett, in a selection excerpted from his “Beyond Belief” (1982), disagrees with Burge’s conclusion here. To “characterize psychological states ‘in the narrow sense’,” he writes, is to answer the question, “What is the organismic contribution to the fixation of propositional attitudes?” Inner syntax, he argues, can’t be it because different syntactic states can support the same “contribution”; nor are propositional attitudes themselves basic since different propositional attitudes may rest on the same “contribution” (as in the case of indexical beliefs). Thus, something else is needed, something “halfway between syntax and semantics”—what Dennett calls “notional-attitude psychology.”

Like phenomenology, such a psychology is concerned with the individual’s subjective world; unlike phenomenology, it analyzes that world from outside, from a third-person perspective. This psychology tries to construct a model of the subject’s representations, or, more accurately, of that which is represented thereby. This model, this notional world, is a fictional world, “but the inhabitants of the fictional world are treated as the notional referents of the subject’s representations.” After exploring constraints on the construction of notional worlds, Dennett observes that if notional attitudes are to play the intermediary roles described above, then it should follow that, roughly, notional attitudes plus environment yield propositional attitudes. But do they? In reply to the objection that they don’t, he presents and analyzes a short tale called “The Ballad of Shakey’s Pizza Parlor.”

In the next selection, “Social Content and Psychological Content” (1985), Brian Loar observes that a strategy common to many of the authors arguing about “psychological” (i.e., mental) content is to criticize a given thesis about mental content by showing that the thesis fails to individuate intentional states in accordance with ordinary uses of that-clauses (such as “believes that p”). The thesis to be criticized might suggest, for example, that in a given case, two different beliefs are at play even though only a single that-clause applies to the believer. It might otherwise suggest that two individuals have the same belief even though we apply two different that-clauses to them. Loar notes that these two versions of the critical strategy share two assumptions: paraphrased roughly, that (A) sameness of that-clause application implies sameness of mental content; and (B) differences in that-clause application imply differences in mental content. Loar then argues that both (A) and (B) are false.

Regarding (A), Loar varies Kripke’s well-known example about “Pierre,” who grows up in France and comes to believe, via stories, a belief expressible by the sentence “Londres est jolie,” and who later moves to London, not realizing it is also called “Londres,” and comes to believe a belief expressible by the sentence “London is not pretty.” In Loar’s variation, Pierre’s new belief is, in fact, one ascribable by “London is pretty.” However, Loar argues, though “Londres est jolie” is synonymous with “London is pretty” (so that the corresponding that-clauses count as the same), Pierre actually has two different beliefs. This must be the case because those beliefs interact differently with Pierre’s other beliefs. In France, for example, Pierre believed that if he ever lived in London (“Londres”) he would live in the same city that Oscar Wilde once lived in. But though Pierre now believes he lives in London, he doesn’t believe that he lives in the same city Oscar Wilde once lived in! Sameness of that-clause application, in short, does not imply sameness of belief.

Regarding (B), Loar argues that if we were to be given a diary and told that it belongs to either an Earthling or a Twin Earthling, but not told which, we would have no trouble providing adequate psychological explanations of the behaviors described therein, despite not knowing the proper referents (H₂O or XYZ) of the relevant terms—and, therefore, which that-clauses properly apply to the agent. This underscores the fact, Loar claims, that two Twins indeed share the relevant mental content even though different that-clauses apply to them. Differences in appropriate that-clauses do not, then, imply differences of belief.

Free of assumptions (A) and (B), the Twin Earth e...

Table of contents

• Cover Page
• Half Title Page
• Title Page
• Copyright Page
• Dedication
• Contents
• Acknowledgments
• Preface
• Where It All Began
• Natural Kinds and Philosophy of Language
• Mental Content and Mental Causation
• Self-Knowledge
• Bibliography
• Index