The Embedding Objection: Part I, “What is Expressivism?”

Many people think that the problem of embedding ethical sentences like (1)

(1) Intentionally flying airplanes into tall buildings is wrong

within more complex sentences like (2)

(2) If intentionally flying airplanes into tall buildings is wrong, then I won’t do it

is one of the most challenging difficulties facing expressivists. This difficulty is often called “The Embedding Objection” or “The Frege-Geach Problem.” I also think The Embedding Objection is one of the most challenging difficulties facing expressivists, though I think the difficulties are not often fully understood as well as they might be. One reason, I think, is that ‘The Embedding Objection’ is an exceedingly misleading definite description: despite the occurrence of the definite article, there are, as far as I can tell, at least eight different problems that fall under the label ‘The Embedding Objection’; despite the occurrence of ‘objection’, not all of these problems are intended by those who mention them to be objections (for example, Dreier’s (1996) insistence that expressivist theories must account for compositionality is not an objection that they cannot); and despite the occurrence of ’embedding’, some of the problems have nothing fundamentally to do with embedding (for example, Sinnott Armstrong’s (2000) “Deepest Problem of Embedding”). Another reason is that there are several different kinds of expressivism, and so it is not always clear toward which kind of expressivism each of the eight problems is most forcefully directed. So, in a series of entries, I’ll try to make more transparent the different problems falling under the label ‘The Embedding Objection’ and toward which kind of expressivism each problem is most forcefully directed.

The embedding problems themselves will be the heart of the series, but I’ll first state as precisely as I can what I take expressivism to be (a task that is surprisingly difficult) in order to prevent some confusion that may result later on. In Part II, I’ll distinguish five kinds of expressivism toward which each embedding problem may be directed. This first entry is a bit longer and more detailed than I’d like for a blog entry, but I’m hoping the detail will be worth preventing misunderstandings in later entries.

The main idea behind expressivism seems simple enough: if a person utters an ethical sentence, like (1), then the speaker expresses a pro- or con-attitude–in this case, a con-attitude toward intentionally flying airplanes into tall buildings. But there are numerous ways one could “express an attitude” (I count at least eight–maybe a future blog entry?). We might first think that one “expresses an attitude” in uttering an ethical sentence by thereby performing an expressive illocutionary act, as one does when uttering “Hooray!” There are several difficulties with this characterization, but, for purposes of discussing the embedding problems, the most important difficulty is that it does not capture the sense of “express an attitude” according to which Hare is counted by some to be an expressivist (and the embedding problems are also supposed to present problems for Hare’s theory). So, what we want is a characterization of expressivism that captures the main idea behind expressivism, while also making sure the characterization is broad enough to cover all ethical theories that historically have been counted as expressivist. Here’s my attempt.

Expressivism is the view that (E) is true:

(E): If a speaker utters an ethical sentence, such as (1), and uses all the words in the sentence properly and literally, then the speaker performs a direct illocutionary act whose sincerity condition requires the speaker to have a pro- or con-attitude whose directedness is toward a suitable object of intentionality whose identity is made clear by the conventions governing the use of the sentence or its constituent terms.

A speaker “utters” a sentence S of a language L when she intends (i) to produce a meaningful sentence S of L, (ii) that S be a sentence of L, and (iii) that S’s constituent terms refer to or mean what the speaker believes the terms refer to or mean in at least one sense these have in L. Condition (i) ensures that the meaningful string of symbols be intentionally produced, condition (ii) ensures that the speaker intends to produce a sentence of L (so that she intends to be speaking L), and condition (iii) ensures that the speaker has some understanding of the constituent terms of L (so that she is not just producing, for example, the sentence ‘La neige est blanche’ as a sentence of French, without understanding any of the constituent terms). A speaker utters a sentence of L “properly” when she utters a sentence of L and uses the sentence’s constituent terms correctly, i.e., she does not misuse any of the sentence’s constituent terms. A speaker utters a sentence S of L “literally” when she performs all and only the direct illocutionary act(s) that is (are) appropriate for S given its meaning in L. For example, a speaker’s utterance of ‘The cat is on the mat’ in English will be literal when she intends to perform a direct assertive illocutionary act (an act of describing the world) and only that direct assertive; a speaker’s utterance of ‘Go home’ in English will be literal when she intends to perform a direct directive illocutionary act and only that direct directive.

What is a “direct assertive,” “direct directive,” and, more generally, a “direct illocutionary act”? It is also surprisingly difficult to explain what an illocutionary act is (for example, one quarter of William Alston’s Illocutionary Acts and Sentence Meaning (2000) is devoted to sorting out issues about the nature of an illocutionary act), so I’ll just hope that Austin’s stock definition is intuitively clear enough: an illocutionary act is “the performance of an act in saying something.” For example, in saying ‘You are on my foot’, I may be describing a state of affairs in which you are on my foot, suggesting that you get off my foot, commanding that you get off my foot, advising you to get off my foot, and so on. The acts of describing, suggesting, commanding, and advising would all be illocutionary acts, since they would all be acts that I would be performing in saying ‘You are on my foot’. There are literally hundreds of distinct illocutionary acts, but there have been different taxonomies that try to sort out the different illocutionary acts into more basic kinds. I’ll follow Searle’s taxonomy which sorts illocutionary acts into one of five basic types according to their illocutionary point. The most important of these basic types for our purposes are: assertives, whose point is to describe the world as being a certain way; expressives, whose point is to express one’s pro- or con-attitude; and directives, whose point is to direct one’s hearer to do things at a time later than the time of utterance. A direct illocutionary act is an illocutionary act that is not performed by way of another illocutionary act (which would be an indirect illocutionary act). For example, if I perform a directive directing you to get off my foot by uttering ‘You are on my foot’, the directive would be an indirect illocutionary act that is performed by way of performing a direct assertive illocutionary act, which describes the world as being such that you are on my foot.

According to Searle, each of the five basic types of illocutionary act has a sincerity condition. That is, each token of a basic type of illocutionary act requires the speaker to have a certain psychological attitude in order to count as a sincere act. For example, in order to perform a sincere assertive, the speaker must believe that the world is the way it is being described as being. Most importantly for us, in order to perform a sincere expressive, the speaker must have the pro- or con-attitude that he or she is expressing (for example, the speaker must have some kind of pro-attitude if she is to perform a sincere expressive in uttering ‘Hooray!’), and in order to perform a sincere directive, the speaker must desire (a kind of pro-attitude) that the chargee of the directive (usually the speaker) perform the act he or she is being directed to perform. It is by characterizing expressivism in terms of the performance of direct illocutionary acts whose sincerity condition requires the speaker to have some kind of pro- or con-attitude that we are able to count Hare as an expressivist, along with Blackburn, Gibbard, and Ayer.

Of course, the speaker cannot just have any pro- or con-attitude if his or her illocutionary act is count as sincere. The attitude must also be directed toward some suitable object of intentionality. Intuitively, one’s pro- or con-attitude may be directed toward individuals, groups of individuals, act tokens, act types, properties, systems of norms, businesses, and so on, so any of these is a “suitable object of intentionality.” Moreover, the object of intentionality must somehow be made clear by the sentence being uttered. For example, if a speaker performs an direct expressive in properly and literally uttering ‘Hooray, John, for winning the race!’, we would not count the expressive as sincere simply because she has a pro-attitude toward some state of affairs other than John’s winning the race; in order to count as sincere, the attitude must be directed towards John’s winning of the race.

The upshot to this characterization of expressivism is this: in order for a metaethical theory to count as expressivist, it must hold that a proper, literal utterance of an ethical sentence like (1) is either a direct expressive or a direct directive (at least in part), and the attitude the speaker must have in order for the expressive or directive to be sincere must be directed towards a suitable object of intentionality that the conventions for the use of the sentence or its constituent terms make clear. Using (1) as an example, Ayer’s (1952, p. 21) expressivist theory would hold that the sincerity condition for the expressive performed in properly and literally uttering (1) is disapproval, which is directed toward the act type of intentionally flying airplanes into tall buildings, which is made clear by the grammatical subject. According to Blackburn’s (1998, pp. 68-69) most recent articulation of his theory, he would hold that the sincerity condition for the expressive performed (he calls it an “avowal”) is disvaluing intentional acts of flying airplanes into tall buildings, which is made clear by the conventions governing the use of ethical predicates. Gibbard (1990, p. 7) would hold that the sincerity condition for the expressive performed is nonacceptance of any system of norms that permits intentional acts of flying airplanes into tall buildings, which is made clear by the conventions governing the use of the ethical predicate ‘is wrong’. Hare (1952, p. 176) would hold that the sincerity condition for the directive performed is a desire directed toward everyone who is in the circumstance of flying airplanes to the effect that they not fly their planes into tall buildings, which is made clear by the conventions governing the use of the ethical predicate ‘is wrong’; I (2003, p. 24; 2004, pp. 9-10) would hold that the sincerity condition for the expressive performed is some kind of con-attitude toward any act that is wrong (and not specifically toward acts of flying airplanes into tall buildings), which is made clear by the conventions governing the use of the ethical predicate ‘is wrong’.

In the next installment, I’ll distinguish five different kinds of expressivism.

Ayer, A. J. 1952. Language, Truth, and Logic. New York: Dover Publications, Inc.

Blackburn, Simon. 1998. Ruling Passions. Oxford: Clarendon Press.

Boisvert, Daniel R. 2004. “Expressive-Assertivism and ‘The Embedding Objection’.” Paper presented to the 2004 Meeting of the American Philosophical Association, Pacific Division, Pasadena, CA, March, 2004.

Boisvert, Daniel R. 2003. Expressive-Assertivism: A Dual-Use Solution to The Moral Problem. Doctoral Dissertation, University of Florida, Department of Philosophy, Gainesville, FL.

Dreier, James. 1996. “Expressivist Embeddings and Minimalist Truth.” Philosophical Studies 83 (1): 29-51.

Gibbard, Alan. 1990. Wise Choices, Apt Feelings: A Theory of Normative Judgment. Cambridge, MA: Harvard University Press.

Hare, R. M. 1952. The Language of Morals. Oxford: Clarendon Press.

Sinnott Armstrong, Walter. 2000. “Expressivism and Embedding.” Philosophy and Phenomenological Research 61 (3): 677-693.