Kent Bach



Minimalism about truth-aptitude, if correct, would undercut expressivism about moral discourse. Indeed, it would undercut nonfactualism about any area of discourse. But it cannot be correct, for there are areas, about which people hold beliefs and make statements, to which nonfactualism uncontroversially applies. Or so I will argue. I will be thereby challenging John Divers and Alexander Miller’s [3] appeal to minimalism about truth-aptitude in defending a certain argument against expressivism about value. But I will not be defending expressivism. For what is wrong with minimalism about truth-aptitude is, in my view, also what is wrong with expressivism: both mistakenly assume that for an utterance to qualify as a statement or a psychological state as a belief, it must be capable of being true or false.


What is minimalism about truth-aptitude (MTA)? It is the thesis that utterances that have the form and function of statements and psychological states that have the form and function of beliefs are genuinely true or false. That is, being assessable as ‘true or false according to the standards of truth-aptitude imposed by disciplined syntacticism’ is enough to make these utterances and states qualify as full-fledged statements and beliefs. MTA claims, in effect, that ostensible truth-aptitude is real truth-aptitude. Now MTA is not to be confused with minimalism about truth itself (thus the titles of [2] and [8], the papers that started this Analysis thread, are misleading). Minimalism about truth (MT), which has been defended forcefully in a recent book by Paul Horwich [6], says that there is nothing more to truth than what is captured by the disquotational schema, ‘“S” is true if and only if S’. And, as Hartry Field [4] has painstakingly explained in a recent article aptly entitled ‘Disquotational Truth and Factually Defective Discourse’, MT has the resources to make sense of debates between factualists and nonfactualists in each of a number of areas of discourse. His discussion makes clear that MT and MTA are mutally independent theses. Our topic is MTA, which is a minimalist view about what it takes to be true (or false), not MT, which is a minimalist view about what truth is. In discussing MTA, it will be convenient to put asterisks around the words ‘statement’ and ‘belief’ so as to leave it open whether *statements* and *beliefs* in a given area really are true or false and qualify as genuine statements and beliefs.


Divers and Miller contend that to vindicate MTA it is enough to show that *statements* and *beliefs* satisfy certain platitudes, provided these are ‘read in such a way as to respect their platitudinous nature’. That is, breaches of a strengthened version of a platitude don’t count. In regard to the platitude that belief is a state designed to fit the facts, for example, one cannot refute MTA with respect to a certain subject matter by showing that the ‘robust’ notion of truth as correspondence does not apply to that area. Divers and Miller argue that any apparent failure of minimalist belief to satisfy a platitude about belief depends on a non-minimalist reading on which the platitude is not a platitude. So, they conclude, the conception of belief required by a certain MTA-based argument for expressivism is not fatally attenuated. (For their discussion of this argument and related issues see [3] and references there.)


Two of the three platitudes Divers and Miller take up pertain to the practical function of belief, and I do not take them to be problematic. I am concerned with the platitude that belief is a state designed to fit the facts, especially with the phrase ‘fit the facts’. Let’s call it the ‘fact-fitting platitude’. As noted above, it must be so read as to require no more than a platitudinous interpretation of the idea of correspondence. This, Divers and Miller suggest, is captured by ‘Things are as the belief that P says they are if and only if P’. Any weightier construal of correspondence would render this platitude no longer platitudinous. If the platitude is interpreted as intended, it is satisfied by minimalist belief, for which the following schema holds: ‘The belief that P is true if and only if P’. This schema is analogous to the disquotational schema for the truth of sentences: ‘“S” is true if and only if S’. Divers and Millers are not here endorsing minimalism about truth itself, only minimalism about truth-aptitude. Their point is that any commitment to a more robust conception of truth would compromise the platitudinous character of the fact-fitting platitude. They aim to show that ‘the minimalist conception of truth-aptitude has the resources to sustain what is genuinely platitudinous about the idea of belief as a state designed to fit the facts’.


How are we to read this platitude? Presumably nothing weighty is meant by ‘fact’: it is a fact that P if and if only P. Still, we may ask, if a state is designed to fit the facts, must there be a fact for it to fit (or fail to fit, in the case of error)? If there are such facts, then *beliefs* (and *statements*) are genuine beliefs (or statements) because they have truth-valuable contents. But if there are no such facts, then the contents of these states (or utterances) are not truth-valuable. Even so, they could be ‘designed to fit the facts’ in a weaker sense, for they are still treated as true or false. So, for example, people having *beliefs* in that area try to avoid *error* and often accept *correction*. But how could these *beliefs* be genuine beliefs, and utterances expressing them be genuine statements (expressions of genuine beliefs)? MTA cannot count the fact-fitting platitude as satisfied if there are no facts of the relevant sort. For then it would count states (or utterances) as truth-apt which are incapable of being true of false. On the other hand, if counts that platitude as satisfied only if there are facts of the relevant sort, then it is of no help in combatting nonfactualism. In particular, it cannot help in the fight against expressivism about values if it presupposes that matters of value are matters of fact, for that is precisely what expressivism denies.


Before continuing on the question of value, let us consider less controversial areas. There are many everyday *beliefs* and *statements* that are not true or false, or at least not straightforwardly true or false (to borrow a phrase from Field [4]). There are many common one-place predicates, such as ‘large’, ‘poisonous’, ‘tasty’, ‘interesting’, ‘disgusting’, and ‘illegal’, which do not express one-place properties but which we often use as if they do. We use them in ordinary subject-predicate sentences to make statements and thereby express beliefs. Although we say ‘Fido is large’ to mean that Fido is large for a dog, ‘That mushroom is poisonous’ to mean that it is poisonous to humans (we certainly do not mean that it is poisonous to all creatures), and ‘Anchovies are tasty’ to mean that they are tasty to oneself, these utterances, taken strictly and literally, are neither true or false. They are true or false only relative to something, something that these utterances do not make explicit (there are many sorts of relativity, e.g., category-, argument place-, location-, time-, reference frame-, and norm-relativity). They would be straightforwardly true only if that something were made explicit.


Now consider how the dilemma described above in connection with value arises in an uncontroversial kind of case. Our *statements* and *beliefs* regarding time of day are relative to time zone. In spite of that, there are people who are ignorant of time zones and unaware of this relativity (there are even some who, though able to tell time, are not cognitively capable of understanding this relativity). These people never explicitly relativize their utterances about time of day, as we sometimes do. They treat *statements* and *beliefs* regarding time of day as straightforwardly true or false. Nevertheless, these *statements* and *beliefs* are true or false only relative to time zone. Accordingly, the minimalist fact-fitting platitude is not satisfied, at least not if that platitude is interpreted as entailing that there are facts of the relevant sort. On the other hand, if the fact-fitting platitude is read as not requiring that there be facts of the sort in question, then these *statements* and *beliefs* do satisfy that platitude but are nevertheless incapable of being true or false. So MTA is either not minimalist or not correct. Either it assumes that there are facts of the relevant sort, or it entails that *statements* and *beliefs* can be true or false even if there are no facts for them to be true or false about.


We may now consider whether *statements* and *beliefs* that are not truth-apt are genuine statements and beliefs anyway, albeit ‘factually defective’ ones (to borrow a phrase from the title of [4]). We should distinguish types of case. (i) Some cases are based on grammar. For example, anyone who uses sentences like ‘Jack has finished’ and ‘Jill is ready’ would be making an implicit reference to what Jack has presumably finished or to what Jill is presumably ready for; they would not mean that Jack has finished simpliciter or that Jill is ready simpliciter. Not relativizing them is, where grammatical (compare ‘Jack has completed’ with ‘Jack has finished’), done just for convenience and efficiency. Clearly such utterances are genuine statements and expressions of genuine belief; it is just that their contents are not being made fully explicit (for discussion of the pragmatic processes involved in the use of syntactically complete but semantically incomplete sentences such as these, see [1]). (ii) There are many cases besides the ones based on grammar about which relativism is so obviously correct that everyone is a relativist. Almost everyone realizes that largeness is relative to comparison class and that poisonousness is relative to kind of creature. Here again, it is convenient not to bother to relativize. (iii) Then there are cases about which relativism is not blatantly obvious, such as relativity to time zone. There are people who do not realize that weight is relative to a gravitational field, that illegality is relative to a legal system, and that disgustingness is in the eye of the beholder. They may not have philosophized about such matters, but in practice they are absolutists. In speaking about such matters, they do not intend any implicit relativization. Even so, such people are, it seems, making statements and thereby expressing beliefs -- not with relativized contents but with contents as presented. (iv) Then there are the philosophically controversial cases, such as discourse about values, about which it is disputed whether or not people’s *statements* and *beliefs* in that area are capable of being true or false. In these cases, it is an open question whether or not *statements* and *beliefs* are capable of being straightforwardly true or false (or even true or false at all). If Divers and Miller’s MTA-based defense of the argument against expressivism were correct, then the status of *statements* and *beliefs* about values would depend on how this issue is settled. Whether or not these utterances and these psychological states count, respectively, as statement and beliefs would depend on whether or not matters of value are matters of fact. Their status would depend on the ontological status of values.


(An interesting question arises, as Field points out, of how to characterize disagreement in a given area of discourse. As far as values are concerned, ostensible disagreement about cases may actually be disagreement about norms for evaluating cases. Disputants might agree about values with respect to a given set of norms, and yet still disagree. Consider predicates like ‘funny’, ‘holy’, and ‘obscene’. Many people apply such predicates as if they express monadic properties of things, when in fact they do not. As Field points out, in characterizing what is going on here, there are two options. We can treat the predicates as used by such people as genuinely monadic but as failing of reference; or we can treat them as implicitly relative, even though the people in question do not regard them as such and might not acknowledge that they are, even after being given good arguments to that effect. Field thinks there is no determinate basis for choosing one option over the other.)


Let me sum up the argument. There are sentences of the form ‘a is F’ which are true or false only with respect to some parameter. The disquotational schema for such sentences takes the form ‘“a is F” is true relative to R if and only if a is F relative to R’. So, the *statements* made in uttering such sentences, and the *beliefs* thereby expressed, are not true or false simpliciter. But there are people who treat them, and for whom they function, just as if they were true or false simpliciter. This suffices to qualify them as genuine statements and beliefs. All that is lacking are facts for them to fit (or fail to fit). Even so, they satisfy the fact-fitting platitude of MTA, at least if it not so interpreted as to entail that there are facts of the required sort (if it is so interpreted, then it assumes factualism and, in the case of values, it begs the question against expressivism). But if there are no such facts, then these statements and beliefs cannot be truth-apt; only their relativizations can be. So MTA is false.


It might be objected that I have implicitly equated relativism about values with expressivism. I agree that there is a difference, but my point may be extended to expressivism proper. According to Divers and Miller, ‘the central claim of expressivism is that declarative sentences of moral discourse are not apt to be evaluated as true or false.’ Since this formulation does not exclude relativism, let us call ‘strong’ expressivism the view not only that *statements* and *beliefs* about values are incapable of being true or false but that they are really not statements and beliefs and are merely expressions of subjective (evaluative or affective) attitudes. These could be preferences, emotional responses, or personal norms. Weak expressivism is the view that *statements* and *beliefs* in a given area do have truth conditions but their truth conditions advert to subjective attitudes (Alan Gibbard’s view in [5], according to which moral judgments are relative to personal norms, is an important recent example of weak expressivism). Weak expressivism is thus a special case of relativism, in which the implicit reference is to something subjective. Only strong expressivism implies that *statements* and *beliefs* are not statements and beliefs at all. A third view, projectivism, does not go quite this far. It allows that there are genuine moral statements and beliefs but says that they do not have contents of the sort suggested by their form; although they superficially appear to ascribe properties to the items they are about, such as acts and traits, they really ascribe properties (relational properties), to agents. Subjectivism is the version of projectivism that says that statements about values express beliefs about one’s desires. It is not to be confused with strong expressivism, which says that *statements* about values are not statements at all, ie.,not expressions of beliefs about one’s desires, but expressions of one’s desires.


All three views, strong expressivism, relativism (including weak expressivism), and projectivism, are error theories of sorts. They all deny that *statements* and *beliefs* about values are straightforwardly true or false. I cannot assess their comparative merits of these views here, but what have I argued does conflict with strong expressivism, precisely because it moves directly (and dubiously) from a claim about the ontological status of values to a claim about the status of *statements* and *beliefs* about them. It entails that if *statements* and *beliefs* about values are not straightforwardly true or false, they are not statements and beliefs at all. However, expressivism, as Divers and Miller define it, does not entail this. Against them I have argued that *statements* and *beliefs* about values qualify as genuine statements and beliefs not because they are true or false, if only in the attenuated sense of MTA, but independendently of that. They count as genuine statements and beliefs simply because they have the form and function of statements of beliefs.






[1] Bach, Kent, ‘Conversational Impliciture’, Mind & Language 9 (1994) 124-162.

[2] Divers, John and Alexander Miller, ‘Why the Expressivist about Value Should Not Love Minimalism about Truth’, Analysis 54 (1994) 12-19.

[3] Divers, John and Alexander Miller, ‘Platitudes and Attitudes: A Minimalist Conception of Belief’, Analyst  Preprint #2 (1994).

[4] Field, Hartry, ‘Disquotational Truth and Factually Defective Discourse’, Philosophical Review 103 (1994) 405-2.

[5] Gibbard, Alan, Wise Choices, Apt Feelings (Cambridge, MA: Harvard University Press, 1990).

[6] Horwich, Paul, Truth (Oxford: Blackwell, 1990).

[7] Jackson, Frank and Philip Pettit, ‘Moral Functionalism and Moral Motivation’, Philosophical Quarterly 45 (1995) 20-40.

[8] Smith, Michael, ‘Why Expressivists about Value Should Love Minimalism about Truth’, Analysis 54 (1994) 1-11.