How to Distinguish Between Theoretical and Practical Attitudes? Avery Archer ABSTRACT: Many philosophers draw a distinction between theoretical attitudes
(which represent the world as being a certain way) and practical attitudes (which
represent the world as to be made a certain way). But giving a detailed account of
the difference between theoretical and practical attitudes turns out to be no easy
matter. One common strategy involves the claim that theoretical and practical
attitudes display different “directions of fit”. However, I argue that the notion of
direction of fit, as commonly understood, is poorly suited for this purpose. To this
end, I consider the two most influential versions of the “direction of fit” analysis:
(1) the revision-conditions analysis inspired by Elizabeth Anscombe, and (2) the
dispositional analysis due to Michael Smith. I claim that both offer an inadequate
basis for distinguishing between theoretical and practical attitudes. I then propose
an alternative account, which I call the two-content analysis. According to the
two-content analysis, the difference between theoretical and practical attitudes is
best conceived of in terms of the kinds of deductive inferences in which each may
feature: indicative inferences in the case of the former, and imperative inferences
in the case of the latter. 1. Introduction Many philosophers draw a distinction between theoretical attitudes (which represent the world as being a certain way) and practical attitudes (which represent the world as to be made a certain way). But giving a detailed account of the difference between theoretical and practical attitudes—i.e., one that includes a specification of the necessary and sufficient conditions for an attitude to be theoretical or practical—turns out to be no easy matter. One of the most common ways of doing so is to claim that theoretical and practical attitudes display different directions of fit (henceforth, DOF). According to this approach, an attitude is theoretical just in case it displays mind-­‐‑to-­‐‑world DOF and practical just in case it displays world-­‐‑to-­‐‑mind DOF. However, the notion of DOF is itself fraught with difficulty. Accounts of DOF typically fall under two broad headings: normative and descriptive. In this paper, I consider the most influential examples of each type of approach: (1) the analysis in terms of revision-­‐‑conditions inspired by Elizabeth Anscombe, and (2) the dispositional analysis due to Michael Smith, respectively. I claim that both fail to provide an adequate basis for distinguishing between practical and theoretical attitudes. I then propose an alternative account of the difference between theoretical and practical attitudes, which I call the two-­‐‑content analysis. 1
2. Direction of Fit as Psychological Taxonomy The standard starting place for discussions of direction of fit is Elizabeth Anscombe’s shopping example: Let us consider a man going round a town with a shopping list in hand. Now it is clear
that the relation of this list to things he actually buys is one and the same whether his wife
gave him the list or it is his own list; and that there is a different relation when a list is
made by a detective following him about. If he made the list itself, it was an expression
of an intention; if his wife gave it to him, it has the role of an order. What then is the
identical relation to what happens, in the order and the intention, which is not shared by
the record? It is precisely this: if the list and the things that the man actually buys do not
agree, and if this and this alone constitutes a mistake, then the mistake is not in the list
but in the man's performance...; whereas if the detective's record and what the man
actually buys do not agree, then the mistake is in the record.1
Anscombe’s shopping example illustrates two different ways in which our words (written and spoken) may relate to the world. In the case of the shopping list, the aim is to get the world (the items purchased) to match our words (the shopping list), while in the case of the detective'ʹs record, the aim is to get the word (the detective'ʹs record) to match the world (the items purchased). In the terminology of speech act theorists, the former is an example of world-­‐‑to-­‐‑
word fit, and the latter of word-­‐‑to-­‐‑world fit.2 John Searle (and several philosophers following him) took the preceding observations about the two different ways in which our words may relate to the world as also applying to certain psychological attitudes. This has led to a distinction between two broad classes of attitudes: ones that display mind-­‐‑to-­‐‑world fit and ones that display world-­‐‑to-­‐‑mind fit. Huw Price summarises the central intuition behind the DOF analysis of attitudes as follows: Beliefs have a 'mind-to-world' direction of fit; they aim at fitting the world, at being true.
Desires, on the other hand have a 'world to mind' direction of fit; we aim to change the
world to fit desires, and not vice versa.3
The notion of DOF is typically applied to belief and desire, with belief being the paradigmatic example of an attitude with mind-­‐‑to-­‐‑world DOF and desire being the paradigmatic example of an attitude with world-­‐‑to-­‐‑mind DOF. Indeed, some discussions of DOF treat the distinction as applying exclusively to belief and desire.4 However, the distinction is also commonly applied to a range of psychological phenomenon. For example, Searle employs it to contrast perception 1
Anscombe (2000: 56). It remains a matter of some debate whether Anscombe herself should be considered a DOF
theorist (See Frost [2014]). However, it is clear that she remains a major inspiration for the revision-conditions
analysis of DOF. (See and Cf. Humberstone [1992: 59]; Sobel and Copp [2001: 45].)
2
The expression “direction of fit” was never employed by Anscombe herself, and was actually popularised by John
Searle, who employed it to describe the two different ways in which words may relate to the world (See: Searle
[1985]). It is worth noting that Searle thought that there were a total of four possible directions of fit a speech-act
may display: (1) word-to-world DOF, (2) world-to-word DOF, (3) double DOF, and (4) null or empty DOF. (Searle
and Vanderveken, [1985: 52-53].)
3
Price (1989: 120).
4
See: Smith (1987) and Milliken (2008).
2
(which he claims displays mind-­‐‑to-­‐‑world DOF) and intentional action (which he claims displays world-­‐‑to-­‐‑mind DOF).5 Richard Moran uses it to contrast intention (which he claims displays world-­‐‑to-­‐‑mind DOF) and prediction (which he claims displays mind-­‐‑to-­‐‑world DOF).6 David Velleman posits that there are many other attitudes besides belief—such as hypothesizing, assuming, and fantasizing—that display mind-­‐‑to-­‐‑world DOF.7 Raimo Tuomela observes that promises, intentions, directives, and wants all display world-­‐‑to-­‐‑mind DOF. Hence, DOF is widely taken to have applications beyond the singular contrast between belief and desire. The distinction between attitudes with mind-­‐‑to-­‐‑world and world-­‐‑to-­‐‑mind DOF is often taken to correspond with the distinction between cognitive and conative attitudes, respectively (to use David Velleman’s preferred terminology)8, or the distinction between thetic and telic attitudes, respectively (to use Lloyd Humberstone’s preferred terminology), or the distinction between theoretical and practical attitudes, respectively (to use my preferred terminology). Each of these dichotomies comes with their own subtle set of associations. But I will not attempt to settle the question of which (if any) is to be preferred. As the title of my paper suggests, I am interested in the use of DOF as a way of drawing a distinction between theoretical and practical attitudes. However, the arguments in this paper will be limned at a level of generality that will allow them to be applied, mutatis mutandis, to the cognitive/conative and thetic/telic distinction as well. Hence, the present investigation may be seen as offering not only an account of the difference between theoretical and practical attitudes, but also an account of the difference between cognitive and conative attitudes and the difference between thetic and telic attitudes.9 3. The Revision-­‐‑Conditions Analysis According to what I shall call the revision-­‐‑conditions analysis, the DOF of an attitude depends on whether the mind or the world is subject to revision when there is a lack of correspondence between the two.10 Mark Platts summarises the view as follows: The distinction is in terms of direction of fit of mental states to the world. Beliefs aim at
being true, and their being true is their fitting the world; falsity is a decisive failing in a
belief, and false beliefs should be discarded; beliefs should be changed to fit the world,
not vice versa. Desires aim at realization, and their realization is the world fitting with
5
Searle (1983: 88).
Moran (2004: 46-47).
7
Velleman (1992: 12). Velleman actually uses the label “cognitive attitudes” rather than the traditional “mind-toworld” terminology. However, it is clear from the discussion, which makes explicit appeal to the notion of
“direction of fit”, that Velleman (or at least his imagined rhetorical interlocutor) takes the distinction between mindto-world and world-to-mind DOF to track the contrast between “cognitive” and “conative” attitudes, respectively.
8
Velleman (1992).
9
This means that the positive account of theoretical and practical attitudes limned in this paper may be seen as an
alternative and competitor to the distinctions found in Velleman (1992) and Humberstone (1992).
10
Many theorists prefer to use the label “correctness-conditions” to “revision-conditions”. One reason to prefer the
latter is that the former leaves the notion of world-to-mind fit somewhat mysterious since it remains unclear in what
sense the world can be said to be incorrect, when there is a mismatch between the mind and the world. By contrast,
it seems easier to get a handle on the idea of the world being subject to revision when there is a mismatch between
the world and the mind. However, as we shall see shortly, the “revision-conditions” label has its drawbacks as well.
Fortunately, it will not matter, as far as the arguments in this paper are concerned, whether we adopt the
“correctness-conditions” or “revision-conditions” conception. See and Cf. Platts (1979: 257).
6
3
them; the fact that the indicative content of a desire is not realised in the world is not yet
a failing in the desire, and not yet any reason to discard the desire; the world, crudely,
should be changed to fit with our desires, not vice versa.11
According to this approach, saying that a belief has mind-­‐‑to-­‐‑world DOF entails that when the world and the belief fail to line up, the burden is on the believing agent to alter her mental representations so as to bring them into line with the world. By contrast, to say that a desire has world-­‐‑to-­‐‑mind DOF entails that when the world and the desire fail to line up, the burden on the desiring agent is to alter the world so as to bring it into line with her mental representations. Strictly speaking, it would be incorrect to say that a belief is subject to revision when there is a mismatch between the belief and the world. After all, if one were to switch from believing P to not believing P (not to be confused with believing ¬P), one has essentially given up one’s belief, not revised it. What is subject to revision is not the particular token belief but rather the agent’s total doxastic state. Whereas the agent’s doxastic state initially included the belief that P, it is updated so as to no longer include the belief that P. However, if we understand the locution “revising a belief” in a broad (albeit somewhat Pickwickian) sense that includes giving it up, then it seems harmless to say that a belief is subject to revision when it fails to match the world. Hence, I will continue to use the revision-­‐‑conditions label, if only on stylistic grounds. (Talking about the “revision-­‐‑conditions” of an attitude seems easier on the ear than talking about the “giving up-­‐‑conditions” of an attitude.) Unfortunately, the revision-­‐‑conditions analysis fails to provide an adequate basis for distinguishing between theoretical and practical attitudes. This is because it is unable to preserve the intuition that there are theoretical attitudes—such as assuming and imagining—
which fail to share the same revision-­‐‑conditions as belief. Recall, according to the revision-­‐‑
conditions analysis, what it means for an attitude to be theoretical is that it is subject to revision when there is a mismatch between the attitude and the world. However, while believing that P and hypothesising that P are plausibly thought of as subject to revision if it is not true that P, the same cannot be said of the attitudes of assuming that P and imagining that P. Although assuming that P and imagining that P both represent things as being a certain way, they are not necessarily subject to revision if things are not as they are represented. For example, we often assume that a certain premise is true for purposes of reductio. Of course, we do not typically believe the assumed premise, particularly if the reductio is successful at generating a contradiction. However, we must continue to assume the premise if our reductio is going to constitute an argument at all. If we took the fact that premise were false as grounds for giving up the assumption, qua assumption, then we would at that point cease to have an argument. A similar point may be made with regards to imagining that P. If I were to imagine that I lived on the moon, the fact that I did not live on the moon would not constitute grounds for revising my imagining. The upshot is that if we assume, in keeping with the revision-­‐‑conditions analysis, that an attitude is theoretical only if it is subject to revision when there is a mismatch between 11
Platts (1979: 257). Richard Moran employs a version of the revision-conditions analysis to distinguish between
the DOF of an intention and a prediction: “When intention and action fail to match, it is said, the mistake is in the
action, whereas if the prediction or report fails to match the action then it is the report which is mistaken and must be
corrected…”(Moran [2004: 47] Italics mine). See: Price (1989) and Humberstone (1992) for discussion.
4
the attitude and the world, then we seem forced to deny that assuming and imagining are theoretical attitudes.12 The preceding observation suggests a possible diagnosis of why the revision-­‐‑conditions analysis provides an insufficient basis for distinguishing between theoretical and practical attitudes. To say that a particular attitude is theoretical is not yet to say anything about when that attitude is subject to revision. Whether or not an attitude is subject to revision seems tied to its aim, and not to the fact that it is theoretical or practical. Hence, a belief is subject to revision if the proposition believed is false because belief aims at truth.13 Since, by contrast, the attitudes of assuming that P and imagining that P do not aim at the truth, they are not subject to revision if the proposition assumed or imagined is false. The lesson is that if an account of the difference between theoretical and practical attitudes is going to be satisfactory, then it must preserve the distinction between saying that an attitude has a certain aim (and by extension, certain revision-­‐‑
conditions) and saying that it is theoretical. Unfortunately, the revision-­‐‑conditions analysis falls short in precisely this respect. 4. The Dispositional Analysis An alternative account of DOF is that offered by Michael Smith, who advocated the following dispositional analysis: [A] dispositional conception of desires enables us to cash the metaphor characterising
beliefs and desires in terms of their direction of fit . . . . [T]he difference between beliefs
and desires in terms of direction of fit comes down to a difference between the
counterfactual dependence of a belief and a desire that P, on a perception that not P:
roughly, a belief that P is a state that tends to go out of existence in the presence of a
perception that not P, whereas a desire that P is a state that tends to endure, disposing the
subject in the state to bring it about that P. Thus, we may say, attributions of beliefs and
desires require that different kinds of counterfactuals are true of the subject to whom they
are attributed. We may say that this is what a difference in their direction of fit is.14
One distinctive feature of Smith’s dispositional analysis is that it renders it impossible for the same attitude to have both directions of fit. After all, this would require that the attitude in question have the tendency to both go out of existence in presence of a perception that not P and endure in the presence of a perception that not P.15 Unfortunately, the dispositional analysis is inadequate for the very same reason that the revision-­‐‑conditions analysis is. Recall, in order to serve as a basis for distinguishing between theoretical and practical attitudes, an account of DOF must not only preserve the intuition that desires and beliefs have different DOF. It must also preserve the intuition that believing that P, 12
One response to the present objection would be to switch from a revision-condition to a correctness-condition
analysis. While a premise assumed for reductio may not be subject to revision, it may still be deemed incorrect.
However, this strategy does not work in the case of imagining that P since imagining that P is not plausibly thought
of as incorrect if P is false.
13
While there have been challenges to the claim that belief aims at truth (see Owens [2003]), it remains the orthodox
view among philosophers. For a helpful overview of the topic, see Engel (2004) and Chan (2013).
14
Smith (1987: 54).
15
Smith (1987: 56).
5
assuming that P and imagining that P all share the same DOF. However, by Smith’s lights, assuming and imagining turn out to have a different DOF to believing, owing to the fact that assuming that P (for purposes of reductio) and imagining that P both tend to persist in the presence of the perception that ¬P while believing that P tends to desist in the presence of the perception that ¬P. Additionally, it is not clear that the dispositional analysis is even able to offer us a satisfactory analysis of the difference between belief and desire.16 The dispositional analysis takes it to be a brute psychological fact about us that our desires tend to endure in the presence of a perception that ¬P and that our beliefs tend desist in the presence of a perception that ¬P. However, it is entirely conceivable that there may be exceptions to the above rule. For example, there may be a belief that does not tend desist in the presence of the perception that ¬P, simply because it is dogmatically held. (Consider, for example, the dogmatic beliefs of a religious fanatic.) However, The dispositional analysis forces us to regard such a belief as a desire. Thus, Smith'ʹs framework seems to define out of existence what many would regard as a real possibility; namely, a belief that is so dogmatically held that it is immune to contravening perceptual evidence. What makes the belief of the religious fanatic a belief is the fact that it involves the commitment to the truth of a proposition, not the fact that it displays a tendency to desist in the presence of a contrary perception. This is why the belief of the religious fanatic is subject to criticism when the fanatic ignores contrary evidence. The fanatic is committed to a proposition being true, but is not appropriately responsive to evidence showing the proposition to be false. The dispositional analysis is unable to preserve this basic idea. By Smith’s lights, if we were to encounter a religious fanatic who held such a dogmatic belief, we would be left without grounds for criticising them since their dogmatic “belief” would not constitute a belief at all, but would instead constitute a desire. Hence, if we were to adopt the dispositional analysis, we would be unable to criticise an agent who held a belief so dogmatically that it tended to persist in the face of contravening perceptual evidence. These shortcomings of the dispositional analysis do not only render it inadequate for our present purposes, but also suggests that it is generally implausible. 5. The Two-­‐‑Content Analysis The remainder of this paper will be devoted to articulating an alternative account of the distinction between theoretical and practical attitudes. Here, in brief, is my proposal: an attitude is theoretical just in case it displays the same logical properties as an indicative speech-­‐‑
act and practical just in case it displays the same logical properties as an imperative speech-­‐‑act. I will unpack this proposal at present. I begin with the contrast between indicative and imperative speech-­‐‑acts. Consider the following pair of utterances: (A): “The office door is shut.” (B): “Shut the office door!” 16
For an attempt to impugn the adequacy of Smith’s analysis of DOF, see: Sobel and Copp (2001).
6
In most ordinary conversational contexts, someone who utters (A) is performing an indicative speech-­‐‑act (e.g., making an assertion), and someone who utters (B) is performing an imperative speech-­‐‑act (e.g., giving a command). Intuitively, (A) and (B) share something in common. They are both, in some sense, about the office door and its being shut. Some philosophers have attempted to capture this intuition by suggesting that what (A) and (B) have in common is their propositional content; namely, the proposition: [the office door is shut].17 The difference between the two speech-­‐‑acts has to do with the illocutionary force with which this single proposition is expressed, with the proposition being asserted in (A) and commanded in (B). Since, according to this view, it is possible to vary the force of a speech-­‐‑act while keeping its content fixed, it follows that content and force are independent features of a speech-­‐‑act. I will refer to this view as the independence thesis.18 It is worth noting that the independence thesis is not free from controversy.19 Indeed, while I believe that (A) and (B) have something in common, I am sceptical about the thesis that what they have in common is a proposition—i.e., something truth-­‐‑evaluable. Hence, I am disinclined to accept the independence thesis. Even so, the discussion below will assume (if only for the sake of argument) that the independence thesis is true. My motivation for this is twofold. First, something like the independence thesis continues to enjoy widespread acceptance among philosophers and may even be considered the orthodox view. Second, presupposing the truth of the independence thesis will make defending the two-­‐‑content analysis more difficult. Hence, by assuming the independence thesis I hope to forestall the accusation that I am making my work too easy. If I am able to mount a successful defence of the two-­‐‑content analysis given the assumption that the independence thesis is true, then (a fortiori) I should be able to offer a successful defence of it when the truth of the independence thesis is not assumed. In the concluding section of this paper, I will briefly sketch what my proposal would look like when the independence thesis is not assumed. Significantly, the independence thesis relies on a notion of content that leaves the logical properties of a speech-­‐‑act—i.e., the kinds of deductive inferences a particular speech-­‐‑act may be used to perform—unspecified. This is because the logical properties of a speech-­‐‑act are at least partly determined by its illocutionary force. Recall, according to the independence thesis, (A) and (B) both have the same propositional content: the proposition [the office door is shut]. If we assume that this content is sufficient to determine their logical properties, then it would follow that (A) and (B) should be logically interchangeable, despite their contrasting illocutionary force. But this is clearly not the case. For example, the following is a valid inference: (A1): “The office door is shut.” (A2): “If the office door is shut, then professor Smith is away.” (A3): “Therefore, professor Smith is away.” 17
In order to minimize confusion, I will (here and henceforth) use double quotation marks to indicate when I’m
referring to a particular token utterance used to perform a speech-act, and square brackets to indicate when I’m
referring to the abstract proposition that constitutes the content of such an utterance.
18
See Frege (1879; 1979) and also Searle and Vanderveken (1985: 1).
19
For a defence of the independence thesis, see Stenius (1967). For criticisms of the independence thesis, see Hanks
(2007) and Mastop (2011).
7
However, the following putative inference is not valid: (B1): “Shut the office door!” (B2): “If the office door is shut, then professor Smith is away.” (B3): “Therefore, professor Smith is away.” The problem with (B1)-­‐‑(B3) is that the initial premise fails to satisfy the antecedent of the conditional specified in (B2). This leaves us without any basis for inferring the consequent of the conditional, as specified in (B3). Moreover, the failure of (B1) to satisfy the antecedent of (B2) is directly due to its illocutionary force. (B1) fails to satisfy the antecedent of (B2) precisely because it fails to depict the proposition [the office door is shut] as true. In fact, (B1) is perfectly consistent with the falsity of the proposition [the office door is shut]. The same, of course, cannot be said of (A1), which is clearly inconsistent with the falsity of the proposition [the office door is shut]. Although, according to the independence thesis, (A1) and (B1) share the same content, (A1) is logically inconsistent with the negation of the proposition [the office door is shut], while (B1) is logically consistent with the negation of the proposition [the office door is shut]. Given that the only difference between (A1) and (B1) has to do with their illocutionary force, it follows that illocutionary force makes a difference with regards to the kinds of valid inferences a speech-­‐‑act may be used to perform. The takeaway of the preceding discussion is that if we are interested in the logical properties of a speech-­‐‑act, we cannot simply consider its propositional content. We must also consider its illocutionary force. Given that the logical properties of a speech-­‐‑act are determined by a combination of its propositional content and illocutionary force, it will be helpful to have a label for the combination of these two features of a speech-­‐‑act. I will use the label “illocutionary content” for this purpose. In short, the illocutionary content of a speech-­‐‑act is the combination of its illocutionary force and propositional content. Hence, we may now say that the types of deductive inferences a speech-­‐‑act may be used to perform is determined by its illocutionary content. While a speech-­‐‑act is typically conceived of as involving a propositional content and illocutionary force, an attitude is typically conceived of as involving a propositional content and psychological mode of presentation.20 I will use the label “psychological content” to refer to the combination of the propositional content and psychological mode of presentation of an attitude. In short, psychological content is to an attitude what illocutionary content is to a speech-­‐‑act. Moreover, just as we may distinguish between speech-­‐‑acts with indicative and imperative illocutionary contents, I claim that we may distinguish between attitudes with indicative and imperative psychological contents. According to the present suggestion, if I believe that Hermione shut the door, then my belief displays indicative psychological content, and if I desire that Hermione shut the door, then my desire displays imperative psychological content. Hence, I hold that there is a type of parity between indicative speech-­‐‑acts (e.g., assertions) and attitudes with indicative psychological content (e.g., beliefs), and a type of parity between imperative speech-­‐‑acts (e.g., commands) and attitudes with imperative psychological content (e.g., desires). 20
See: Kissine (2009: 126).
8
Indeed, we often express beliefs (such as the belief that Hermione attends Hogwarts) via an assertion (such as the assertion: “Hermione shut the door.”) and we often express desires (such as the desire that Hermione shut the door) via a command (such as the command, “Hermione, shut the door!”). While I do not take the fact that we often express beliefs via assertions and desires via commands to be conclusive grounds for holding that the former has indicative psychological content and the latter has imperative psychological content, I do not think this is merely a coincidence. But this is not the line of argument I wish to pursue. In section 6, I explain why I believe that we should conceive of attitudes like believing that P, assuming that P, and imagining that P as having indicative psychological content, and in section 7 I explain why we should conceive of practical attitudes—like desiring to ϕ—as having imperative psychological content. It will turn out that the reasons for thinking of beliefs and desires as indicative and imperative attitudes, respectively, run deeper than the mere observation that we often express them via certain types of speech-­‐‑acts. But let us table this point for the time being. Assuming that my analogy between psychological and illocutionary content is apt, it follows that the logical properties of an attitude—i.e., that which determines the kinds of deductive inferences in which an attitude may feature—is determined not by the attitude’s propositional content, but by its psychological content. For example, one can validly infer from the belief that the door is closed and the belief that if the door is closed, then professor Smith is away that professor Smith is away. However, the same is not true of the desire to close the door, even though (according to the independence thesis) the belief and desire share the same propositional content. Hence, if we are interested in the types of deductive inferences in which an attitude may validly feature, then we must pay attention to the attitude’s psychological content, rather than its propositional content. With the preceding theoretical vocabulary in place, we may now summarise my proposal as follows: to describe an attitude as theoretical or practical is to make a claim about its psychological content. According to the present suggestion, to describe believing as a theoretical attitude is to say that it has indicative psychological content. Similarly, to describe desiring as a practical attitude is to say that it has imperative psychological content. Let us call attitudes that display indicative psychological content indicative attitudes, and attitudes that display imperative psychological content imperative attitudes. Given this way of putting things, the distinction between theoretical and practical attitudes corresponds with the distinction between indicative and imperative attitudes, respectively. 6. The Analysis of Theoretical Attitudes Why should we conceive of believing that P, assuming that P, and imagining that P as indicative attitudes? The answer is that (like believing that P) assuming that P, imagining that P and pretending that P can all feature in classically valid inferences. This is obvious in the case of assumptions, since we often use assumptions in the premises of arguments. However, it is also true of such attitudes as imagining that P and fantasizing that P. I will focus on the attitude of 9
fantasizing that P since it is most controversial case.21 Suppose that I am pretending to be Harry Potter (say, while attending a science fiction and fantasy convention), and that my pretence prompts me to make the following inference: (C1): “I am Harry Potter.” (C2): “If I am Harry Potter, then Voldemort killed my parents.” (C3): “Therefore, Voldemort killed my parents.” By stipulation, (C1) and (C3) are not things I believe. I do not believe I am Harry Potter; nor do I believe that Voldemort killed my parents. I only fantasize that these propositions are true. However, (C1)-­‐‑(C3) remains a perfectly respectable deductive inference for all that. This is because (C1)-­‐‑(C3) conforms to the classical conception of validity; the truth of its premises guarantees the truth of its conclusion. In sum, believing that P, assuming that P, imagining that P, and fantasizing that P are indicative attitudes because they all share the same psychological content, and they all share the same psychological content because they may all feature in classically valid inferences: inferences in which the truth of the premises guarantees the truth of the conclusion. Practical attitudes, by contrast, cannot feature in classically valid inferences. This follows from the conjunction of the following two claims: (i) practical attitudes are not truth-­‐‑
evaluable and (ii) the classical conception of validity only applies to truth-­‐‑evaluable items. In the next section, I explain why we should conceive of practical attitudes as having imperative psychological content. Assuming that the arguments I offer in the next section are sound, it follows that the psychological content of a practical attitude, like the illocutionary content of an imperative speech-­‐‑act, is not truth-­‐‑evaluable. As such, practical attitudes cannot feature in deductive inferences that involve truth-­‐‑evaluable premises and conclusions (i.e., classically valid inferences). It is worth emphasising that the present line of argument does not assume that practical attitudes cannot feature in inferences of their own. For example, the following also appears to be an acceptable inference: (D1): “Fight for freedom or acquiesce to authoritarianism!” (D2): “Do not acquiesce to authoritarianism!” (D3): “Therefore, Fight for freedom!” (D1)-­‐‑(D3) appears to be a “valid” inference. However, it cannot be valid in the classical sense, according to which an argument is valid just in case the truth of its premises guarantees the truth of its conclusion. This is because both the premises and conclusion of (D1)-­‐‑(D3)—being imperatives—are not truth-­‐‑evaluable. The sense in which (D1)-­‐‑(D3) is “valid” seems tied to the fact that the satisfaction of (D1) and (D2) guarantees the satisfaction of (D3), in a sense 21
While David Velleman considers fantasizing that P to be an example of an attitude with mind-to-world fit,
Elizabeth Camp appears to classify fantasizing that P (or what she calls “make-believe”) as having world-to-mind
fit. (See: Velleman [1992: 12; 2000: 110] and Camp [2009]). By focusing on the controversial case of fantasizing
that P, I hope to display some of the power of my account to offer a principled answer on a disputed case.
10
analogous to how the truth of (C1) and (C2) guarantees the truth of (C3). However, this would, at best, constitute a non-­‐‑classical sense of validity.22 And therein lays the point of contrast between a desire (conceived of as an imperative attitude) and an assuming, imagining, or a pretending (conceived of as indicative attitudes). We do not need to resort to a non-­‐‑classical notion of validity in order to make sense of the fact that the psychological content of assuming that P, imagining that P, or pretending that P may feature in a valid inference. The apparatus of a classical truth-­‐‑conditional logic is perfectly adequate for this task. Certainly, a point so fundamental deserves to be marked, and the present suggestion is that the claim that assuming that P, imagining that P, and pretending that P are all theoretical attitudes is best conceived of as doing just that. 7. The Analysis of Practical Attitudes Why should we conceive of practical attitudes as imperative attitudes? My answer is that we can only make sense of the special connection between practical attitudes and action (what we may call the practicalness of practical attitudes) by conceiving of them as having imperative illocutionary content. Consider the command mentioned earlier: (B): “Shut the office door!” (B) is a call to action. It is not a call to judgement. One has only responded to (B) in the way it calls for when one has formed the intention to shut the office door. Merely forming the belief that the door is, will, or should be shut is not enough. Let us call the act of responding to a speech-­‐‑act in the way that the speech-­‐‑act calls for assenting to that speech-­‐‑act. According to the present suggestion, one assents to an imperative by forming an intention, not by merely adopting a belief. Contrast this with the making of a sincere assertion: (A): “The office door is shut.” (A) is not a call to action. At most, it is a call to believe. Hence, assent to (A) does not require the forming of an intention. The takeaway is that imperative speech-­‐‑acts require action in a way that indicative speech-­‐‑acts do not; one has only responded to an imperative in the way it calls for when one forms the intention to do what the imperative says. If this is right, then we can only preserve the practicalness of practical attitudes by holding that they have an imperative logical structure. 22
I include the qualification “at best” because it still remains controversial whether imperative inferences display
their own kind of validity. This is the problem that Ross (1944) has dubbed as “Jørgensen’s dilemma”: how can
imperatives (which are not truth-evaluable) feature in what appears to be valid inferences? Hence, we may
characterize practical attitudes (i.e., attitudes with imperative psychological content) as those to which Jørgensen’s
dilemma applies. Theoretical attitudes, by contrast, do not face this problem, and are therefore not attitudes to which
Jørgensen’s dilemma applies. For an in-depth discussion of the semantics of imperatives, including arguments that
an assertive interpretation of imperatives is inappropriate, see: Kaufmann (2012). Henceforth, I will leave the “at
best” qualification implicit.
11
One possible objection to the preceding line of thought is that there are some indicative speech-­‐‑acts that do call for action. The standard example is indicative speech-­‐‑acts that describe moral obligations. For example, consider the following sincere assertion: (E): “You are morally obligated to defend the innocent.” According to some theorists, one has not truly assented to (E) until one has formed the intention to defend the innocent. Hence, it might be urged, it is not true that imperative speech-­‐‑acts are connected to action in a way that indicative speech-­‐‑acts are not. Insofar as (E) is a genuine indicative speech-­‐‑act, I am inclined to deny that one can only genuinely assent to it by shutting the office door. But I will not press the point here. Even if we assume that an agent can only assent to (E) by adopting an intention, it would not follow that indicative speech-­‐‑acts require action in the same way that imperatives do. This is because imperative speech-­‐‑acts like (B) require action in virtue of their logical structure. By contrast, if indicative speech-­‐‑acts like (E) require action at all, it is purely in virtue of their subject matter or content. This point can be seen by noting that an imperative speech-­‐‑act demands action irrespective of its content, while an indicative speech-­‐‑act only demands action when its content is evaluative in nature. Hence, it remains true that imperative speech-­‐‑acts require action in a way that indicative speech-­‐‑acts do not. Moreover, insofar as our goal is to characterise the distinction between theoretical and practical attitudes, we want to say that the attitude of desire demands action, not merely its content. In other words, we do violence to the distinction between theoretical and practical attitudes if we conceive of a practical attitude like desire as demanding action in the very same way that a theoretical attitude like belief might. That would be to offer an account of the difference between theoretical and practical contents, not an account of the difference between theoretical and practical attitudes. One final objection to my claim that desires are imperative attitudes is worth considering. According to the two-­‐‑content analysis, a desire is a practical attitude because it has imperative psychological content. However, this seems inconsistent with the observation that desires may sometimes feature in inferences involving truth evaluable premises and conclusions. For example, consider the following inference: (F1): “I desire to make an omelette.” (F2): “If I desire to make an omelette, then I must break some eggs.” (F3): “Therefore, I must break some eggs.” (F1) appears to express a desire. And yet, (F1)-­‐‑(F3) remains a classically valid inference. Hence, it appears as though desires may feature in classically valid inferences after all. Fortunately, this objection may be met. Where the preceding line of reasoning goes wrong is that it equates (F1) with the psychological content of a desire when it actually corresponds with the illocutionary content of an assertion reporting a desire or the psychological content of a belief about a desire. I will unpack this reply at present. Let us begin by observing that we may express the belief that I desire to make an omelette via the assertion “I desire to make an omelette”. Moreover, if I assert that I desire to make an omelette, then the 12
illocutionary content of my assertion corresponds with (F1), and if I believe that I desire to make an omelette, then the psychological content of my belief also corresponds with (F1). Now, we may ask, what is the intentional object of my belief that I desire to make an omelette? What is my belief about? The answer is that my belief is about my desire to make an omelette. This means that my belief is about a certain desire I have. In short, my belief is a higher order attitude; it is an attitude about an attitude. Now, consider my desire to make an omelette. Let us suppose that (F1) corresponds with the psychological content of my desire. This would mean that my desire to make an omelette has the same psychological content as my belief that I desire to make an omelette. But then, by parity of reasoning, my desire to make an omelette must also be about my desire to make an omelette. This follows from the fact that one cannot have two attitudes that share the same psychological content but which differ in their intentional object. If my belief that I desire to make an omelette (which has psychological content corresponding with (F1)) is about my desire to make an omelette, then the same must also be true of my desire to make an omelette (when the desire is conceived of as having psychological content corresponding with (F1)). But if my desire to make an omelette is about my desire to make an omelette, this makes my desire to make an omelette a higher order desire. But this would constitute a mistake. For comparison, suppose I did not have the desire to make an omelette, but that I desired to have the desire to make an omelette. This would be an example of a genuine higher order desire. However, the desire to make an omelette is not like this. It is about my making an omelette, not about my desire to do so. The takeaway of the preceding discussion is that although I may report my desire to make an omelette by an assertion that corresponds with (F1), it does not follow that (F1) corresponds with the psychological content of my desire. Holding that it does leads to the false conclusion that my desire to make an omelette is a higher order desire—i.e., a desire whose intentional object is a desire. Instead, (F1) corresponds with illocutionary content of my assertion that I desire to make an omelette, or with the psychological content of my belief that I desire to make an omelette. Finally, consider the alternative picture entailed by my proposed account. According to the two-­‐‑content analysis, the psychological content of my desire to make an omelette is the self-­‐‑directed command: “Make an omelette!” On this view, the psychological content of my desire to make an omelette makes no mention of a desire. My proposal therefore avoids the erroneous conclusion that my desire to make an omelette is a higher order desire. 8. Conclusion The aim of this paper has been to offer necessary and sufficient conditions for an attitude to be theoretical or practical. I began by considering one common strategy for drawing the distinction between theoretical and practical attitudes—namely, via an appeal to the notion of direction of fit. I have argued that the two most widely discuss analyses of direction of fit are poorly suited for this task. In so doing, I go some distance towards showing that there is a need for an alternative analysis of the difference between theoretical and practical attitudes. The second half of the paper provides such an analysis. My discussion of the two-­‐‑content analysis, in the last three sections, proceeded under the assumption that the independence thesis is true. I observed that even if one were committed to the claim that indicative and imperative speech-­‐‑
13
acts shared the same propositional content, one would still need to posit another kind of content (what I have called “illocutionary content”) if one wanted to have a notion of content that corresponds with the kinds of deductive inferences a speech-­‐‑act may be validly used to perform. Similarly, I have argued that if we wish to arrive at a complete logical characterisation of an attitude—one that determines the kinds of deductive inferences in which it may validly feature—one cannot simply consider the propositional content of that attitude. One must consider the conjunction of its propositional content and psychological mode of presentation (i.e., its psychological content). Given such a picture, the two-­‐‑content analysis may be seen as highlighting a distinction between two kinds of psychological content an attitude may display: the kind of psychological content that determines if an attitude may feature in (classically valid) indicative inferences or (non-­‐‑classically valid) imperative inferences. For those who reject the independence thesis (i.e., the claim that an indicative and an imperative speech-­‐‑act may share the same propositional content), articulating the two-­‐‑content analysis becomes much more straightforward. Simply put, the two-­‐‑content analysis proposes that we conceive of theoretical and practical attitudes as those that display indicative and imperative content, respectively, where indicative and imperative content are seen as logical primitives (as opposed to a composite consisting of propositional content and psychological mode of presentation). Whether one accepts or rejects the independence thesis, the two-­‐‑content analysis amounts to the following suggestion: that we conceive of the contrast between theoretical and practical attitudes as tracking the kinds of deductive inferences in which an attitude may feature. An attitude is theoretical just in case it may feature in (classically valid) indicative inferences, and practical just in case it may feature in (non-­‐‑classically valid) imperative inferences. References Anscombe, Elizabeth. (2000) Intention. Cambridge: Harvard University Press. Camp, Elisabeth. (2009) Two Varieties of Literary Imagination: Metaphor, Fiction, and Thought Experiments. Midwest Studies in Philosophy: Poetry and Philosophy 33: 107-­‐‑130. Chan, Timothy (2013) The Aim of Belief. Oxford: Oxford University Press. Engel, Pascal. (2004) Truth and the Aim of Belief. in D. Gillies (ed.), Laws and Models in Science. (pp. 77-­‐‑97). London: King'ʹs College Publications. Frege, Gottlob. (1879) Begriffschrift. in P. Geach and M. Black (eds.), Translations from the Philosophical Writtings of Gottlob Frege (pp. 1-­‐‑20). New York: Philosophical Library. Frege, Gottlob. (1979) Posthumous Writings. Chicago: University of Chicago. Frost, Kim. (2014) On the Very Idea of Direction of Fit. Philosophical Review 123 (4): 429-­‐‑484. 14
Hanks, Peter. (2007) The Content-­‐‑Force Distinction. Philosophical Studies 134: 141-­‐‑164. Humberstone, Lloyd. (1992) Direction of Fit. Mind 101 (401): 60-­‐‑83. Jørgensen, Jørgen. (1937) Imperatives and Logic. Erkenntnis 7 (1): 288-­‐‑296. Kaufmann, Magdalena. (2012) Interpreting Imperatives. New York: Springer. Kissine, Mikhail. (2009) Illocutionary Forces and What is Said. Mind and Language 24 (1): 122-­‐‑
138. Mastop, Rosja. (2011) Imperatives as Semantic Primitives. Linguistics and Philosophy 34: 305-­‐‑340. Milliken, John. (2008) In a Fitter Direction: Moving Beyond the Direction of Fit Picture of Belief and Desire. Ethical Theory and Moral Practice 11: 563-­‐‑571. Moran, Richard. (2004) Anscombe on ‘Practical Knowledge.’ in J. Hyman & H. Steward (eds.), Royal Institute of Philosophy Supplement Volume 55: Agency and Action (pp. 43-­‐‑68). Cambridge: Cambridge University Press. Owens, David. (2003) Does Belief Have and Aim? Philosophical Studies 115: 283-­‐‑305. Platts, Mark. (1979) Ways of Meaning. London: Routledge and Kegan Paul. Price, Huw. (1989) Defending Desire-­‐‑as-­‐‑Belief. Mind 98 (389): 119-­‐‑127. Ross, Alf. (1944) Imperatives and Logic. Philosophy of Science 11(1): 30-­‐‑46. Searle, John. (1983) Intentionality: An Essay in the Philosophy of Mind. Cambridge: Cambridge University Press. Searle, John and Daniel Vanderveken. (1985) Foundations of Illocutionary Logic. Cambridge: Cambridge University Press. Smith, Michael. (1987) The Humean Theory of Motivation. Mind 96: 36-­‐‑61. Sobel, David and David Copp. (2001) Against Direction of Fit Accounts of Belief and Desire. Analysis 61 (1): 44-­‐‑53. Stalnaker, Robert. (1998) On the representation of context. Journal of Logic, Language and Information 7: 3–19. Stenius, Erik. (1967) Mood and Language-­‐‑Game. Synthese 17: 254-­‐‑274. 15
Tuomela, Raimo. (2007) The Philosophy of Sociality: The Shared Point of View. Oxford: Oxford University Press. Velleman, David. (1992) The Guise of the Good. Noûs 26 (1): 3-­‐‑26. 16