Hilary Putnam, Ethics Without Ontology (Cambridge, MA, Harvard U. P., 2004), pp. ix + 129.

Ethics Without Ontology deserves to be read. It masterfully weaves many disparate philosophical concepts, arguments, and traditions, goes for analytic philosophy's jugular by pummeling some of its fundamental assumptions, and reminds us that ethics' ultimate goal is to provide solutions to practical problems. EWO is about ethics' appropriate concerns and the appropriate application of intelligence to the solution of practical ethical problems. Putnam views ethics as a system of an 'indefinitely large number of [interrelated] concerns' (p. 23) that are 'mutually supporting but also in partial tension' (p. 22). Chief among these are an impulse to immediately feel, upon encountering one who is suffering, an obligation to help that person (p. 23-24), universal moral equality (pp. 2426), and the nature of an admirable human life (pp. 26-27). Putnam hopes that we are embarking on a 'Third Enlightenment' in which we learn to use our intelligence to solve ethical problems by sophisticatingly balancing the myriad of ethical concerns, just as engineers and judges solve practical engineering and legal problems by sophisticatingly balancing a myriad of engineering and legal concerns. EWO's central focus, however, is a sustained attack on the following line of thought, prominent in analytic philosophy, that blinds us to ethics' appropriate concerns and causes us to misuse our moral intelligence: (1) Statements, including ethical statements, are objective, truth-evaluable, and true only if they describe 'the world'; (2) Statements can describe the world only if there can be objects in the world (e.g., properties, facts, propositions) that 'correspond' to the statements, thereby 'making' the statements true;

(3)

Therefore, statements, including ethical statements, are objective, truth-evaluable, true, and descriptive only if there can be objects in the world that correspond to and make them true.

Accepting these claims leads to a form of 'monism' (p. 19) by reducing the myriad of ethical concerns to just one: the presence or absence of objects that purport to correspond to or make ethical statements true. The consequence is to also reduce our use of intelligence in solving moral problems to the sole, barren job of determining whether such objects are present in concrete ethical contexts. Putnam provides putative counterexamples to (1) and defends a thesis of 'conceptual relativity', which appears to render (2) unjustified. Logical, conceptual, mathematical, and 'thin' ethical truths are not even descriptive; and though many value judgments, including 'thick' ethical judgments, are descriptive and true, they are not true because they are descriptive. Thus, (1) is false. Conceptual relativity, as I understand it, is the conjunction of four linguistic claims: (i) some important words in natural languages, such as 'exist', have a core use (and, hence, a core meaning); (ii) use of these words is continually extended in various ways, consistent with their core use, thereby creating divergent conventions for their use, i.e., divergent 'optional languages' (p. 43); (iii) the core use, and hence the core meaning, of these words underdetermines which of these divergent conventions is correct; and, therefore, (iv) speakers of a language may decide which conventions, consistent with a word's core use, to adopt. The important metaphysical upshot of conceptual relativity is that 'in certain cases what exists may depend on which of various conventions we adopt' (p. 39). Thus, if conceptual relativity is true, there is no 'fact of the matter' about what objects exist and, hence, little justification for thinking that there can be objects in the world that correspond to or make descriptive statements true. Therefore, (2) is

unjustified, and the 'reductive' ethical concern and use of moral intelligence it helps to engender is 'deeply misguided' (p. 3). We can provide neither 'an Ontological explanation of the objectivity of ethics' (p. 3) nor 'reasons which are not part of ethics for the truth of ethical statements' (p. 3). Central to Putnam's defense of conceptual relativity is the claim that two people can use 'exist' differently, yet still be disagreeing about something important. For example, one who wishes to use 'exist' such that it becomes permissible to assert that mereological sums exist, and one who rejects such a use, disagree about something important when the former asserts that mereological sums exist and the latter asserts that mereological sums do not exist. They do not, and should not, disagree about whether mereological sums really exist, for doing so is 'stupid' (p. 37). Rather, they disagree about whether to extend the use of 'exist' in the way the former would like; they disagree about whether to adopt a particular optional language. Many will find the thesis and defense of conceptual relativity the most controversial part of EWO. About the defense, for example, it certainly seems that the disputants disagree about whether mereological sums really exist, which suggests that the two are not using 'exist' differently. However, for Putnam, asking whether mereological sums really exist 'is asking a question to which Ontologists have not succeeded in giving a sense' (p. 3). But there is a sense to this question that can be given using Putnam's terminology: asking whether mereological sums really exist is asking whether the use of 'exist', when asserting that mereological sums exist, is consistent with its core use; it is to ask whether this use of 'exist' is really an extension of its core use, or a different use altogether. Mereological sums appear to be so unlike people, swans, trees, and rocksobjects that paradigmatically existthat it seems difficult at best to extend the use of 'exist', consistently with this core use, in a way that would permit one to assert that mereological

sums exist. Thus, it appears that the disputants do not disagree about whether to adopt a particular optional language; rather, they disagree about whether the particular language is an optional language in Putnam's sense, for they disagree about whether the use of 'exist' in this particular language is consistent with its core use. And if the two disagree about whether this use of 'exist' is consistent with its core use, they disagree about whether mereological sums really exist. As a 129-page collection of Putnam's 2001 Hermes and 2001 Spinoza lectures, EWO is admittedly (p. 3) short on some important details that Putnam covers elsewhere, especially details about Putnam's current views about what objectivity and truth are. For those unfamiliar with his recent work, I recommend reading EWO alongside The Collapse of the Fact/Value Dichotomy and Other Essays (2002) for discussion of Putnam's notions of objectivity and truth, The Many Faces of Realism (1987) for discussion of truth and conceptual relativity, and 'How Not To Solve Ethical Problems' (in Realism with a Human Face (1992)) for an example of the appropriate application of intelligence to the solution of practical problems.

Daniel R. Boisvert California State University, Bakersfield