Common Sense Metaphysics

Essays in Honor of Lynne Rudder Baker

Luis Oliveira, Kevin Corcoran, Luis R.G. Oliveira, Kevin J. Corcoran

• Publisher: Routledge
• Year: 2020
• ISBN:
  9781000330564
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophical Metaphysics

This book celebrates the research career of Lynne Rudder Baker by presenting sixteen new and critical essays from admiring students, colleagues, interlocutors, and friends. Baker was a trenchant critic of physicalist conceptions of the universe. She was a staunch defender of a kind of practical realism, what she sometimes called a metaphysics of everyday life. It was this general "common sense" philosophical outlook that underwrote her famous constitution view of reality. Whereas most of her contemporaries were in general given to metaphysical reductionism and eliminativism, Baker was unapologetic and philosophically deft in her defense of ontological pluralism. The essays in this book engage with all aspects of her unique and influential work: practical realism about the mind; the constitution view of human persons; the first-person perspective; and God, Christianity, and naturalism.

Common Sense Metaphysics will be of interest to scholars of Baker's work, as well as scholars and advanced students engaged in research on various topics in metaphysics, philosophy of mind, philosophy of language, and philosophy of religion.

Part I
On Practical Realism about the Mind

1 What Is a Concept?¹

Christopher Hill

Lynne Baker wrote two books (1987, 1995) and a number of articles defending a very strong version of intentional realism – that is, realism concerning beliefs and other intentional states. One of her principal claims was that statements ascribing intentional states are objectively true or objectively false, no less than statements about the density of fluids, the orbits of planets, and the charges of particles. In other words, according to Baker, there are objective facts that serve as truth-makers for statements ascribing intentional states. Another principal claim was that the reality of intentional states is not relative to any interests, stances, or explanatory practices of human beings. For example, the statement that Columbus believed that Earth is round is true simpliciter. There is no need to prefix an operator such as “According to folk psychology” or “According to the intentional stance.” These two doctrines define a position I will call hardcore intentional realism. The present paper is a defense of this view.

Hardcore intentional realism is just one component of a family of views that Baker developed and collected under the label practical realism. Although I fully share her commitment to hardcore realism, I am less enthusiastic about the other components, so I will not be concerned with them here. In my view, and perhaps also in Baker’s, hardcore realism is the most important component, and it can easily be pried loose from the rest of her position.

Baker’s defense of hardcore realism covers a lot of territory, but there are three objections to the view that to my knowledge she never addressed. My defense will be complementary to hers in that it will focus on these three challenges.

The first is Quine’s complaint that there is no principled basis for individuating propositions (Quine 1960). If legitimate, this complaint would also pose a threat to propositional attitudes, since propositions are the objects on which intentional states are directed.

Is the proposition that a fortnight is a period of 14 days identical with the proposition that a fortnight is a fortnight? If not, then in virtue of what are the propositions different? Unless there is a principled basis for answering questions of this sort, Quine maintained, statements attributing propositional attitudes are too vague to play a substantial role in science. Now it can seem that Quine was presupposing a general principle to the effect that realism about a domain is permissible only insofar as the domain is governed by a precise principle of individuation for its members. If that were true, his objection to propositions could be challenged, for it is by no means obvious that the principle in question holds in all cases. In fact, however, Quine’s objection can be seen as depending on the considerably less tendentious idea that the domains that figure in the designs of scientific experiments require reasonably precise principles of individuation. So far from being trivially false, reflection shows that this idea has considerable merit. Without some way of individuating the entities that define an experiment, it would be unclear what would count as different trials of the experiment, and unclear also what would count as a replication of it. To appreciate the relevance of this point to intentional realism, consider an experiment concerning problem-solving that is being conducted by a cognitive scientist. It is clearly desirable for the scientist to have a solid basis for determining whether different subjects are bringing the same or different beliefs to bear on the experimental task, and a solid basis for determining whether a single subject is bringing the same beliefs to bear on different trials of the task. If the basis for making such determinations were as slender and soupy as Quine maintained – if the determinations were as empirically unconstrained as he maintained – there would be correspondingly slender grounds for thinking that the subjects had been involved in a single experiment. Equally, claims that the experiment had been replicated would become problematic.

1 I am indebted to Anil Gupta and Riki Heck for countless helpful discussions

One of my main goals in the paper is to show that there is a promising proposal for a principle of individuation for propositions. The proposal I have in mind is due to Chisholm (1981, 1989) and is the topic of the next section.

The second objection to hardcore realism is also associated with Quine (1960), though it has many echoes in contemporary literature. This is the claim that attributions of intentional states are radically underdetermined by all possible evidence. Quine argues for this view at length in Word and Object, and then goes on to infer the following antirealist conclusion:

If we are limning the true and ultimate structure of reality, the canonical scheme for us is the austere scheme that knows no quotation but direct quotation and no propositional attitudes but only the physical constitution and behavior of organisms.

(1960, p. 221)

Quine also expressed this conclusion by saying that there is no fact of the matter as to whether attributions of attitudes are true or false. The truth values of such attributions are indeterminate. Chomsky responded to Quine by pointing out that underdetermination by evidence does not entail indeterminacy of truth values (Chomsky 1969). All theoretical claims in the sciences are underdetermined by the evidence that is respectively relevant to them, but we hardly regard this as a ground for rejecting scientific realism. This objection is prima facie legitimate. In fact, however, a number of passages make it clear that Quine was not inferring indeterminacy from underdetermination. Rather, he was inferring it from a presumed radical underdetermination – an underdetermination so extreme as to amount to a qualitative difference between our folk practice of attributing attitudes and scientific theorizing.

My second goal is to make a case that there is no qualitative difference here. Our folk heuristics for attributing propositional attitudes and for testing such attributions are quite powerful – perhaps less so than the methods of hypothesis formation and testing in mature sciences, but nonetheless sufficient to be compatible with intentional realism.

The third objection to hardcore realism comes from arguments for intentional holism, which asserts that all of the intentional states possessed by an individual are interdependent. For example, according to intentional holism, if a given belief is removed from a system of propositional attitudes in which it is embedded, the agent ceases to possess all of the other members of the system as well. They are automatically replaced by attitudes with different essential natures. Versions of holism have been defended by Davidson (1984), Stich (1985), Dennett (1989), and Churchland (1991), among others.

Holism poses a serious threat to intentional realism because it implies (i) that few if any individuals share intentional states, and also (ii) that few if any individuals have intentional states that persist across time. The first consequence is alarming because it would be difficult to maintain that different individuals attach the same meanings to their words if the individuals never use their words to express the same beliefs. That is, (i) threatens to undercut our understanding of communication. Further, (ii) is alarming because it challenges our understanding of diachronic rationality. Diachronic rationality is what occurs when individuals work out plans and make choices in advance of acting, and then at appropriate later times perform the actions that have been preselected. In such cases, the actions are made rational by the reasoning that produced the earlier plans. According to holism, however, at the time of acting, the individual agent may no longer possess the intentional states that figured in the original reasoning and may no longer possess the resulting plan.

Various philosophers have responded to these problems by invoking similarity, maintaining that communication and diachronic rationality are not threatened as long as it makes sense to speak of whole systems of attitudes as similar to other whole systems. But this is not nearly enough. Any two systems of attitudes will be similar in a number of respects and also dissimilar in a number of respects. To make the suggestion work, it would be necessary to specify the respects of similarity that are relevant to communication and diachronic rationality, and to describe a scheme for pairing up individual attitudes on the basis of these respects. There have been proposals for doing this (e.g., Churchland 1998, Schroeder 2007), but they have run into difficulties. This is not surprising, since, as Fodor and Lepore pointed out (1992), there is a serious underlying problem. If one were to explain the relevant form of similarity, one would presumably want to say that two beliefs B1 and B2 in different systems are relevantly similar if they stand in similar inferential relations to other beliefs. But these other beliefs cannot be beliefs that are shared by the different systems, for by hypothesis there are no such beliefs. So, what must be said is that B1 and B2 are relevantly similar if they bear similar inferential relations to other members of their respective systems that are relevantly similar. Clearly, we are on our way to a regress or a circle. In view of this, it is not hard to see that on the present approach, similarity of individual beliefs would collapse into overall similarity of total systems of attitudes – a fatal flaw, since overall similarity of total systems is comparatively rare, except in individual agents across short intervals of time. (Each of us has countless beliefs that others lack – about our upbringing, about the literature we have read, about our acquaintances, about what we did yesterday, about what we are seeing now, and so on.)

To simplify the discussion, let us focus for the moment on the part of holism that is concerned with beliefs. This is the thesis that an agent would perforce cease to hold every member of a system of beliefs if the agent ceased to hold any one member of that system. The most natural way of opposing this thesis is to embrace the following pair of claims:

• (C1) In order to hold a given belief, B, an agent need hold no beliefs other than the ones that are constitutive of the concepts that figure in B.
• (C2) In most cases, the beliefs that are constitutive of a concept are comparatively few in number.

In combination, these claims provide a sufficient rationale for rejecting holism. But are the claims acceptable?

(C1) is initially plausible and appears to survive reflective vettings, but as Fodor and Lepore have emphasized, (C2) faces a problem that is on the face of it quite serious. The claim presupposes that for any concept C, there is a principled basis for distinguishing between the C-involving beliefs that are constitutive of C and C-involving beliefs that are merely adventitious or nonessential. As Fodor and Lepore pointed out, this presupposition is strongly analogous to, and therefore stands or falls with, the claim that there is a principled basis for distinguishing between the statements containing a word that are constitutive of the meaning of the word and the statements containing the word that play no role in fixing its meaning. In other words, the presupposition stands or falls with the idea that there is a well-motivated analytic/synthetic distinction. It follows that (C2) is called into question by Quine’s critique of analyticity (or, more precisely, by a counterpart of Quine’s critique that is concerned with concepts rather than meanings of words). Since they were persuaded by Quine’s critique, Fodor and Lepore rejected (C2) and also the defense against holism that is based on it.

Believing that the defense fails, but appropriately horrified by the prospect of having to embrace holism, Fodor developed an alternative approach to the problem of holism that involves (i) reducing concepts to words in a representational system that he calls the language of thought, and (ii) reducing beliefs to computational relations to sentences composed of such words.² Given these reductions, he maintained, we can see how it is possible to possess one belief without possessing many others. To possess a certain belief, one need only have performed some computations on a specific sentence in the language of thought. Computation being what it is, it is possible to have performed the relevant computations on one sentence without having performed those same computations on others.

Unfortunately, Fodor ran into serious problems when he turned to the task of developing his language-of-thought proposal. The proposal depends crucially on the claim that there is a universal language of thought – one that is shared at least by all human beings at all times of their lives. Otherwise the proposal could not explain communication or diachronic rationality. Now to vindicate the claim of universality, it is necessary to individuate words in the language of thought in a way that is independent of intersubjective differences in human cognitive powers and faculties, and also in a way that is independent of intersubjective changes in powers and faculties across time. In fact, however, Fodor was unable to do this. He tried to accomplish it by claiming that identity of a word in the language of thought is determined by its effects on all of the high-level processing mechanisms in the human cognitive architecture. (Fodor 1995) But it is clear that not all humans have the same mechanisms, and also that, when two individuals do happen to have the same mechanisms, there can be significant differences in the mechanisms’ manners of operation. Similar points apply to the mechanisms possessed by a single individual at different times. In effect, then, Fodor’s proposal for individuating words in the language of thought confines words to individual minds and significantly restricts the temporal extent of words within individual agents. (For an extended treatment of these points, see Schneider 2011.)

2 See, e.g., Fodor (2008, p. 25, p. 69). I am simpli...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• Introduction
• Part I On Practical Realism about the Mind
• Part II On the Constitution View
• Part III On the First-Person Perspective
• Part IV On God, Christianity, and Naturalism
• Notes on Contributors
• Index