Emotion and The Will: Anscombe on Brentano 

by  Steve Bayne
(January 2003)
Throughout our discussion we have invoked an important distinction between 'resolve' and 'volition'. Two decades following the publication of Intention, Elizabeth Anscombe wrote a curious essay, "Will and Emotion" (Grazier Philosophische Studien, 5. 1978, reprinted in Collected Papers vol. I. pp. 100-07), which went far towards clarifying her position with respect to a number of issues related to this distinction. The essay is somewhat curious because of the attention paid to two crucial chapters in Brentano's groundbreaking work, Psychology from an Empirical Standpoint, chapters which largely concern another distinction, one which contrasts judgment and presentation, which Brentano sees as being distinct categories of mental phenomena, and feeling and will, which Brentano regards as falling under the same category.
    While Brentano, exercised perhaps the greatest single influence on philosophy of mind since Descartes, he says very little about volition or action as such. However, we can say this much: for Brentano volition is not to be considered an action preceding another action, which confers upon this other action its character of voluntariness. Indeed, Brentano, following Kant, is clear in maintaining that a volition involves a desire for something to happen as a result of the desire itself (PES. p. 257). A volition may turn out to be a "mere intention" or what we have called a "resolve," something very close to what Davidson sought to account for in his discussion of "pure intentions." But that this should turn out to be the case is doubtful since more is required of a volition, according to Brentano, than a desire. Additionally, the agent must realize that that desire (or love or hate) brings about the desired object. It is not this alone that draws Brentano closer to what we have been calling the "classical action theorists" and further away from Anscombe's view; it is also Brentano's conflation (perhaps a correct one at that) of "striving" (conativity) and volition, and it is primarily this that argues against his having identified volition as desire or knowledge of a desire's probable consequences.
    It will be my contention that intentionality figures in a very significant way in Brentano's theory of the will. I will say more about this, but for the moment let us refresh our memory a bit as to how Brentano viewed intentionality as criterial of mental phenomena. Shortly we will pursue another issue which concerns classification, namely, that of the viability, in light of Anscombe's criticisms, of Brentano's claim that will and emotion (or feeling) belong to the same category of mental phenomena.


Brentano introduces intentionality this way:

Every mental phenomenon is characterized by what the Scholastics of the Middle Ages called the intentional (or mental) inexistence of an object, and what we might call, though not wholly unambiguously, reference to a content, direction toward an object (which is not to be understood as meaning a thing), or immanent objectivity. (Psychology from an Empirical Standpoint.[PES] ed. L. McAlister, Humanities Press. 1973 [1874]. p. 88.)
Precisely because intentionality characterizes all mental phenomena, something besides intentionality must be considered in classifying them, or at least some differentiation of the intentional nexus itself must be discussed. What this something else is, according to Brentano, has to do more with "fundamental differences in the way phenomena refer to an object than with any other difference" (op. cit. p. 197). The primary classification of mental phenomena will be in terms of the various "ways phenomena refer to an object." It will divide the mental into three primary categories: presentation, judgment, and emotion - where emotion includes intention, interest, love etc. A presentation directs us to an object by way of appearance, while judgment puts us in a relation of acceptance or rejection to what is judged. Emotion, the more vague of the three categories, can be thought of conveniently as whatever doesn't belong to the other categories. Crucial to Anscombe's discussion of Brentano is the fact that, according to him, just as a judgment takes an "object" as being true or false the will (think 'love') takes objects as being good or bad. "But," one might ask, "how are we to know when the intentional relation to an object in numerically different mental phenomena are the same?" Brentano tells us (op. cit. p, 200) that we can only know for sure by inner experience. The claim he makes, which Anscombe singles out for extended consideration, is that feeling and will are to be assimilated within a single classification, but she will also criticize Brentano's claim that presentation (what she calls - unfortunately - "imagination") and judgment sharply differ. His position, she avers, is owing to a failure to get clear on the difference between assertion and predication. Because Brentano's procedure in evaluating the relation of feelings and the will is very close to being that which he makes use of in discussing the relation of judgment and presentation, some preliminary comments on this latter relation will serve to lay the ground not only for Anscombe's criticism but also for a correct understanding of Brentano on the relation of emotion and the will.


Brentano maintains that an object must be brought before the mind before it can serve as subject of a judgment. To a degree this presages what Gareth Evans, citing Russell's Problems of Philosophy (p. 58), calls "Russell's Principle" viz. "that a subject cannot make a judgement about something unless he knows which object his judgement is about" (Varieties of Reference. Oxford. 1982. p. 89).
    This bringing to the attention of the mind is what Brentano calls "presentation." Judgment, while differing from presentation, presupposes it. Something similar might be said of desire, although desire differs categorically from judgment. It is important to keep in mind what approach is being rejected in the attempt to classify mental phenomena on the basis of the character of their intentional relations ("relation" here being a term of mere convenience). What is being excluded is categorization based on consideration of the content of those mental phenomena. Describing contents and states in terms of their relations is not a new idea, even for Brentano. Brentano entertains Alexander Bain's suggestion that the right way to go about characterizing mental phenomena is in terms of their consequences (op. cit. p. 203). Brentano's reply to this suggestion is that it is the nature of the states having the consequences of which Bain speaks that constitutes the basis for explaining them, and that even if some classification on the basis of consequences were possible there would be no resulting account of the differences between categories of mental phenomena. However, an appeal to content is not logically incompatible with taking content cashable in currency of relations. What is important to get clear about is that other appeals to relations other than those which rely on the character of the intentional "relation" are often, if not always, covert appeals to "content." Intentionality belongs to the "nature" of a mental state but not as part of what has traditionally been considered (e.g. Twardowski) its content. When materialists like David Lewis argue that "the definitive characteristic of any (sort of) experience as such is its causal role, its syndrome of most typical causes and effects," (Philosophical Papers vol. 1. 1983. p. 100) what they are suggesting is that the "basis" to which Brentano alludes is in fact purely material and that the "mental" sense of 'content' relates to cause and effect relations in ways largely consistent with Bain's original proposal. Plausible, even though there may be some sleight of hand in Lewis's use of 'typical'. Bain's view as well as Lewis's functionalism are variations on the thematic primacy of "content," despite its seeming elimination in terms of causal relations. Nevertheless, Brentano has some rather persuasive arguments in support of the belief that no difference between presentation and judgment is possible on the basis of their differing content alone. We now turn to these arguments.
    First suppose someone acquainted with existent red trees asks, "Do you believe there are red trees?" Not being acquainted with any myself and being unfamiliar with the variety of trees there are, I withhold judgment. In this case there is no judgment on my part, only presentation. In the case of my interlocutor, however, there is both judgment and presentation. In my case what I entertain without judgment is the idea of red trees; what my interlocutor judges is that red trees exist. But the important thing is that both of us have before us the same idea, the same "content," thus presentation and judgment cannot be distinguished on the basis of a difference in the complexity of their respective contents. The reader must keep in mind that this particular argument is devised under the assumption that the difference being alleged is one of difference in contents, a matter of their complexity or connectivity - in particular, that is, a presentation is understood as lacking a connection between characteristics that judgments possess. The main point here is that many presentations, like all judgments, can be combinations of constituent ideas or presentations, and thus complexity, alone, will not allow for the alleged distinction between judgments and presentations. Although Anscombe is primarily interested in Brentano's bringing together emotion (feeling) and will under a single category, the discussion is difficult to follow and it will expedite an understanding of her arguments if we take a brief look at what Brentano has to say about love and hate in making the case for distinguishing presentation and judgment. We now proceed to a second argument for distinguishing presentation and judgment.


What presentations present are objects, and it is from these objects alone that the basis for asserting of any two presentations that they are contradictory derives. Contrariety in other words is not a characteristic of presentations unless their objects are contradictory. It is the contrariety of warm and cold that makes their presentations contradictory, not some contrariety in the acts of presentations themselves. If we add love to presentation, however, then we have the basis for a new sort of contrariety, one independent of objects - for identical objects can be both loved and hated. Most important for our discussion is that judgment shares with love (and hate) this feature; that is, if we add affirmation and denial to presentations, as we did love and hate, then we emerge with a new concept of contrariety which owes its reality to something besides presentations alone. In addition to the contrast between judgment and presentation with respect to contrariety there is a further difference exposed by attending to a similarity between judgment and love (and hate). This has to do with the fact that before adding love and hate to presentations the only sense in which we can compare presentations with respect to their intensity is in terms of something like vivacity; but if we add love and hate to presentations 'intensity' acquires a new use in describing them, for now we can speak of the act component as intense and not just of the object component as being vivacious; we can speak of degree of vehemence for example, noting as we do that acts not presentations can be "vehement." Once again, judgment parallels love and hate in this regard, for if we add to presentations affirmation (denial) we are then able to speak of degrees of certainty, whereas presentations, themselves, are not describable as certain to any degree. There is, however, a further application of the argument of more direct relevance to Anscombe's discussion of Brentano.
    Presentations can be "bad," but only because - according to Brentano - lovers of the objects of such presentations would sin in so doing; similarly, judgments only because to affirm them would be to err. This is of some importance for understanding Anscombe's take on Brentano, because we see that the badness of a presentation depends on the sinfulness of loving its object. For Anscombe, by contrast (Intention Sec. 40. p. 76) the goodness of desire belongs primarily to the object of such a desire. For Brentano, it is neither act nor object which is taken to be good (or bad). Rather what is taken to be good belongs to the "way in which a mental act refers to a content" (p. 240). This first relation is echoed, albeit imperfectly, by subsequent thinkers, such as C. Broad and Ewing, who locate goodness not in a property but, rather, the 'fittingness' of a fact (for an interesting discussion of this view see Everett Hall's What is Value. chpt. I. Humanities Press. 1952). Brentano's effort is subtle but unmistakable, namely, to link ethical predicates to the reality of the intentional "relation," which just is the way an act refers to content, making it mental in the first place. This should remind us of Kant's idea that the only thing good in itself is a good will. Although we shall not pursue the matter, it is worth noting that there is an interesting, but less striking, parallel, also, between what Brentano is saying about the good and what Frege said about the "sense" of an expression. Sense for Frege is not an object, it is the *way* in which a word refers to an object, much like 'good' describes one way in which a mental act refers to a content.
    It is essential to Brentano's position that being good and being loved in some way become identified (PES. 247). The reason should be clear: a good action is good owing at least in part to the will of the agent whose motive is love (without sin). In the absence of will no object, nor presentation, is good; as far as judgments are concerned 'affirmation', not 'good' expresses the appropriate felicity. Indeed, 'good' belongs essentially to the will, and so is not simple and unanalyzable but requires for its description some reference to desire or love.
    Much of our attention has been devoted to ways one might argue for a distinction between representations and judgments, but most turned out to be based on erroneous comparisons. So how is it that Brentano comes to accept the distinction? Love and hate are not governed by such laws as relate presentations on the basis of cause and effect relations, but rather by such principles as that we like those who like those we love. Analogously judgments operate according to principles of logic, principles that are disanalogous to those governing presentations, inasmuch as they are not causal.
    Briefly stated, Brentano's procedure for distinguishing presentation and judgment is done in connection with love and hate - and therefore the emotions. It is to show that for every significant difference between presentation and love, for example, there is a property analogous to the one belonging to love but which belongs to judgment and not presentation. From this he concludes that presentation and judgment must differ. (PES. pp. 222-225) We are now in a position to discuss the claim Brentano makes that feeling and will belong to the same category - neither judgment nor presentation - and Anscombe's denial of this assertion.


    In Psychology from an Empirical Standpoint Brentano remarked:

Consider the following series, for example: sadness - yearning for the absent good - hope that it will be ours - the desire to bring it about - the courage to make the attempt - the decision to act. The one extreme is a feeling, the other an act of will; and they may seem to be quite remote from one another. But if we attend to the intermediate members and compare only the adjacent ones, we find the closest connections and almost imperceptible transitions throughout. - If we wished to classify them as feelings or strivings, to which of the two basic classes should we assign each case? - We say,"I feel yearning," "I feel hope," I feel a desire to bring this about for myself,"I feel courageous enough to attempt this," - the only thing which no one would say is that he feels a decision.(PES. pp. 236-37).
Brentano is especially concerned to point out that the elements of this series possess a certain inhomogeneity. "Act of will" appears to refer to an act of mind, a decision, and differs from the rest in in not being able to serve as an object of feeling. This suggests that a possible categorical difference between feeling and the will is to be located at this element of the series, a position he rejects.
    Anscombe (op. cit. p. 100) treats the last element of the series as an act of will involving the body, not a mental act such as making a decision. By adding an element to the series, strictly speaking, she is no longer talking about Brentano's original "continuum." This has the result of creating confusion from which the reader never entirely escapes, as she begins her discussion by distinguishing emotion from will as sharply as soap from washing, a contrast between a non-action (soap) and a bodily action (washing), whereas the contrast Brentano would have us attend to is that between one state (soap) and another (water, say) or, at best approximation to Anscombe's understanding, between a state (soap) and another state (resolve to wash, say). Keep in mind as we proceed to discuss the difference between emotion and will that while Anscombe alleges that for Brentano, unlike emotions, bodily sensations do not rely on presentations, but such is not the case (PES. p. 82). Such discrepancies are made all the more significant by Brentano's having indicated that the members of the series constituting the continuum are all states (PES. p. 247).
    We next briefly consider a long but highly instructive quote from Anscombe dealing with the very passage above. Here she attempts to restate Brentano's claim using a contrast between two carefully contrived examples.
In illustration, imagine a young person standing outside the door of someone alarming, whom he is summoning up the courage to beard. He has just nerved himself to walk in, he has arrived at the state of 'Mut'. Now consider the *next* thing, before he actually pushes the door open and steps forward. If we can insert something psychological, something inner, in there at all - something which belongs in the *development* which is to culminate in action, won't it be *almost* the same as the 'Mut' itself, only *more committed* to the action? To see that we might do so, consider that he might summon up the 'Mut' and then realize that the action was impossible - he perceives that the swing door is locked. He physically can't push it open. Now if that's what happens, he hasn't even tried to do it. In just the same situation, in which however he doesn't notice the metal tongue of the door in position, given that little extra, the act of will itself, he won't indeed push the door open (for he can't) but he will have tried. So there is a difference between this last term and the 'Mut', but how small! And aren't they obviously the same in kind? If there is that last term there at all, it clearly belongs to the same class as the 'Mut', and hence in the same class all the rest. And so we have will assimilated to emotion. ("Will and Emotion." in Collected Papers vol. 1. Oxford. pp. 100-01.
Let us begin by noting that contrary to what Brentano believes, and apparently Anscombe as well, the series described is not continuous in the mathematical sense. Every element of Brentano's has an immediate predecessor and strictly speaking a continuous series in the mathematical sense has the property that none of its elements has an immediate predecessor or successor.
    One difficulty with the two scenarios portrayed by Anscombe is that the first introduces a judgment ("then realizes that the action was impossible"), terminating the series by breaking its continuity at the point of 'Mut' (courage). Now the contrast Anscombe draws arises from a comparison of the series which introduces trying and the original series but only up to the point of 'Mut'. Trying in the second scenario follows the last element of what was Brentano's original series, the decision to act, but may precede the ensuing action which involves certain basic actions, such as clasping the door knob before making the effort to turn it. In the case of the second scenario it is unclear whether the last element of the series is a decision to act or a bodily action. Since the important difference involves the addition of trying, we must consider whether trying precedes the action or not. In the case of Brentano's original series it would make no sense to insert trying between 'Mut' and the decision to act. One does not first make the effort to act and then decide to act; and yet this is where the trying present in Anscombe's second scenario is to be placed, if, as she says, "the difference between this last term and the 'Mut'" is so slight. To be sure, the trying precedes the bodily action, but it is also made to precede the decision, unless of course the decision is more like 'Mut' than is the trying. Indeed, this may be the case, but it is something Anscombe fails to take into account.
    If trying follows the last element of Brentano's original series and we merely tack trying onto the original series' last element, Anscombe's intent is stymied since the comparison will be between decision and trying rather than between trying and 'Mut'. What she attempts is to describe an unfolding series of circumstances coinciding with Brentano's purported continuum but which can go either of two ways: one where there is no actual trying, and one where there is. The difference that trying makes is so small that adding it to the series following 'Mut' ("nerving") merely adds to the series making up the continuum but which insofar as it (trying) is an act of will becomes "assimilated to emotion."
    Notwithstanding Anscombe's exclamation of the smallness of the difference between the two scenarios she describes, there is in fact quite a significant difference. In the first case, where the lock is seen to be in place, there is no basic action forthcoming; the intention to open the door is cancelled and the prospective agent doesn't even try to open the door. But in the second case, not only is there a similarity to the previous case, prior to the prospective agent's seeing that the door is locked, there is the additional fact that the agent tried, that is, actually moved his hand in an effort to open the door. Yet, this leaves less certain whether or not the difference "belongs to the same class" as 'Mut'("nerving" or "courage"). Might not the agent in this second case try without bodily action (i.e., "basic action") of any sort? Let us ask: "Does such *trying* "belong to the same class" as 'Mut'? This is unlikely but if this can be successfully maintained, and I shall argue in due course that it can, then Anscombe's primary objection can be sustained. That is, Brentano's assertion to the contrary notwithstanding the difference between emotions and the will turns out to be categorical and intuitively sound.


So far, we have concentrated on the two cases Anscombe has constructed in an effort to "insert" an element into Brentano's original continuum between 'Mut' (nerving, or courage) and the decision to act (or, perhaps, the act itself). In one case, the agent fails to see that the door is locked and so does not, as in the first case, withhold his action in light of this realization; but there is another case Anscombe might have considered but did not, one which suggests the possibility of another element between 'Mut' and either the deciding to act or trying to act. Suppose that the agent sees that the door is latched and yet unlike the agent in the first scenario Anscombe describes does not withhold trying to open it. In such a case there is no deciding to do, only a deciding to try to do. If, perchance, he succeeds, his deciding to try will occupy a place in the continuum between 'Mut' and the action. But now the following question arises: Is the decision to try as much like 'Mut' as the decision to act? Any answer will be controversial but the right answer seems to be yes. When I decide to open a window I do not decide to try to open the window, although should I try and fail I might retrospectively describe the effort as following a decision to try. But now the context has changed, calling forth different semantical features.
    What I now want to suggest is this: a volitional act has a beginning that entails no change has taken place. This is the first occasion where we part company with Bradley.     If we consider a nonvolitional agent - air pressure for example - then it is clear that

        The air pressure began to raise my arm


        My arm began to move.

But if we consider a volitional agent in a minimally contrasting sentence, we see that from

        I began to raise my arm

we cannot infer

        My arm began to move

    One lesson to be learned to be learned from this linguistic data is that if trying is included in the act itself then part of any such action will lack a public criterion for its ascription. If I have not succeeded in moving my arm whether I tried can only be determined with certainty by the agent. I do not say: "I think I tried but I cannot be sure"; I do say: "I think he tried but I cannot be sure."
    Now that we have established that there is a clear difference between deciding to do and deciding to try let's return to Brentano's resemblance continuum. Brentano's argument relies on the idea that because there is a resemblance continuum from emotion ('Mut') to deciding to act emotion and will belong within the same category of mental phenomena. If we take the continuum argument as fundamental then the decision to try and the decision to act cannot be shown to belong to the same mental category since no continuum includes both. But deciding to try and deciding to do resemble one another more than either resembles 'Mut'. The resemblance continuum including deciding to try may be indistinguishable from the resemblance continuum including deciding to do, even though no continuum includes both and here I mean that no one decides to try if he has already decided to do and vice versa. But this indistinguishability provides the basis for maintaining that deciding to try, like deciding to do, belongs to the same category as emotion, and so belongs to the same category. But we have already established that deciding to try and deciding to do cannot be shown to belong to the same category, as long. The continuum leads us, therefore, to a paradox of sorts, and Brentano's argument in the form it takes must be rejected. One possible objection to this line of reasoning is that having distinguished the causal continuum and the resemblance continuum deciding to try and deciding to do are excluded from occurring together only in the physical continuum - not the resemblance continuum. I will postpone, for now, further consideration of these and related matters.


If, as Brentano maintains, mental phenomena come either as judgments, presentatations, or feelings, then there is some question as to where bodily sensations fit into the picture, especially since emotions are feelings of a sort. Anscombe believes bodily feelings must be distinguished from emotions somehow. Before proceeding to examine why she doesn't believe Brentano can make this distinction in a non-question begging way two points need to be made. In the first place there have been some very good philosophers who have rejected the idea that ultimately there is any categorical distinction at all between these two, no more at least than Brentano would admit between sadness and courage. One such philosopher was Spinoza who accepted pain, pleasure and desire as primary emotions on the basis of which the others could be derived (Ethics III. xi). The notion that a pain in the foot belongs in the same general category as courage might sound quite natural to a radical mind-body identity theorist like Spinoza. But this is certainly less true of those who reject the thesis that "minds" are just bodily states. Anscombe's insistence on the distinction may attest her inclination towards dualism despite the anti-Cartesian views she inherited from Wittgenstein. I will attempt no decision on this matter. I will now turn to the relation of sensations and emotions.
    Just as affirmation and negation are involved in judgment, so too, in the case of emotion affirmation (e.g. in loving) and denial (e.g. in hating) come into play. Concentrating, momentarily on negativity with respect to feelings and emotions, we say that one "finds bad," or regrets that someone is ill - the case of feeling; similarly, there is negativity of the will insofar as it may incline us to *avoid* illness; and while this is a similarity between emotion (regret) and will - something suggesting, contrary to Anscombe that the emotions and will are to be assimilated - she claims such facts actually to be of the utmost significance in drawing the distinction between bodily feelings and emotions. This is so because there are feelings that must be distinct from emotions, feelings which cannot be assimilated to the will, given that love is related conceptually to our taking something to be good or evil. A bodily sensation involves no such relation to good or evil and so must be regarded as distinct from emotion.
    One way of describing the point Anscombe makes (op. cit. 104) is to say that if we have to make use of the notions of good and evil in distinguishing emotions and bodily sensations, then we cannot appeal to emotions in distinguishing good and evil. There are, however, a number of questions that require answering before we accept the conclusion she draws from this. One such question is this: Don't we make use of the fact that propositions are true or false in order to distinguish them from other entities, such as names and, yet, make use of the idea of a *proposition's* corresponding to the facts in order to *explain* what we mean by truth rather than simply distinguishing truth from falsity? Isn't something quite analogous involved in explaining and not merely distinguishing good and evil? It is important to keep in mind that whether truth is correspondence is the right theory of truth is not the issue, only whether it is question begging on condition that we distinguish propositions from other things in the way suggested. A more fitting description of what Anscombe appears to be saying is that according to Brentano to distinguish bodily feelings and emotions we need the ideas of good and evil, but to explain good and evil we presuppose the notion of the emotions. Anscombe has not shown that one cannot use x to identify y and then use y to explain x. But even if we assume that the reasoning behind her claim is that Brentano is committed to circularity it is essential to her argument that Brentano must make use of good and evil in distinguishing emotions and bodily feelings. In fact such is not the case. According to Brentano (PES. p. 267) love depends on judgment, and while bodily sensations involve presentation they do not involve judgment. So contra Anscombe, Brentano doesn't need the notions of good and evil to distinguish emotion and bodily sensation. Anscombe makes a further point in commenting on Brentano that relates to issues we have touched upon already in our dealings with classical action theory, and James in particular.
    Anscombe suggests, as we have already noted, that Brentano seems embarrassed by his own admission that will is not a feeling (op. cit. 105). She goes on to point out, correctly, that had he spoken of willful actions such as opening a door instead of actions that follow having to muster up courage etc. then Brentano's continuum of feelings and emotions would not be available. That this is an objection, and if so what its force may be is not entirely clear, but there are a number of things Brentano might say in his own defense.
    Brentano will speak of feeling hope,not just hope; or he will speak of feeling courageous enough to do something and not just being courageous enough to do something. The point in other words is that courage is an emotion but the feeling of courage is something different even though both the feeling of courage and courage are emotions. In case of many if not all emotions there is this distinction: the emotion and the feeling of that emotion. However, it makes no sense to speak feeling a decision, a fact that, as we have seen, does not escape Brentano's attention. But this in and of itself is insufficient to establish that the will and emotion don't belong to the same class of mental phenomena; besides we have already seen reason to suspect Brentano's continuum argument to be flawed. Consider, also, however that our consciousness of anger and our being angry are not easily distinguished; but a decision and our consciousness of that decision are markedly different. This ties in with Anscombe's other point that had Brentano selected an action such as opening a window his continuum would not have been in evidence. Brentano, however, was no doubt well aware of this and his choice of actions was guided by a need to lay out a case where we are aware of the action associated, somehow, with our own will. Similarly, James elected to use getting out of bed on a cold morning as his paradigm, owing to our consciousness in such a case of the presence of the will.
    Anscombe is not the only one to detect embarrassment in Brentano. In one of the probing examinations of Brentano's ethics offered in English Everett Hall calls attention to an apparent problem in Brentano. But Hall's interest is not so much in the admitted distinction between feeling and the will, but rather in the distinction between feeling and judgment. These difficulties are related inasmuch as if it should be as Brentano argues that feeling and the will belong in the same category then any difference between will and judgment will be as problematic as a difference between judgment and feeling. Recall much of Brentano's argument for distinguishing presentation and judgement depends on analogies between feeling and judgment. All the more reason to be concerned with any striking differences between feeling and judgment. On the face of it this is at least as dangerous to Brentano as is posed by the cause of his embarrassment over the distinction between feeling and the will. Indeed, Hall's criticisms are more than the brief mention afforded by Anscombe and they are well worth considering in some detail. Here in part is what he says.

Actually Brentano does point out a distinction between feelings and judgments that is of semantical significance, but it is obviously a source of embarrassment, which he wishes not to emphasize but to play down. The objects of any two right assertions equally exist, but the objects of any two right loves need not be equally good - one may be better than the other. Brentano sees that this difference is not psychological as it would be if, e.g. it were a difference in the intensities of the feelings involved. Brentano's solution seems obviously ad hoc Feelings embrace a peculiar species of phenomenon not to be found in either ideas or judgments, namely, choice or preference. And in terms of right preferences that Brentano that Brentano defines 'better'. My basic criticism of this is simple. Brentano does not really connect his account of comparative value with that of absolute value; he does not define 'better than' in terms of 'good' or conversely. (What is Value Humanities Press. 1952 p. 104-105)
for every significant difference between presentation and love, for example, there is a property analogous to the one belonging to love but which belongs to judgment and not presentation
    Alexander Shand (MIND 1894) once made the point that we are made *aware of* (that is we "feel") the will only when we encounter some sort of resistance. This is sufficient to explain James's paradigm of willful action, as well as Brentano's (nerving). It is also relevant to the case we encountered above allowing us to distinguish deciding to try and deciding to do. We anticipate an "effort" of will when we act on a decision to try but not always when we decide to do. Furthermore, the conditions required for experiencing an act of will as such are not one and the same with those of experiencing shame (an emotion), for example. The "continuum" leading from feeling to will is presented, then, only under conditions where resistance to a certain action makes available awareness of the will's presence; and such conditions are manifest only in cases where the continuum is in evidence, not in cases such as shutting a window, which as Anscombe notes does not require any courage or nerving (typically). Here is a rare occasion where James and Brentano can be seen to be on the same side with respect to the experiencing of our own will.
    It would be consistent for Brentano to hold that a voluntary action is one performed with a certain intention and that there is a continuum in going from a resolve (or "mere intention") to what Bradley called an "incomplete volition" and finally to a completed action. Moreover, in neither case would Brentano be subject to Anscombe's criticism (p. 105) based as it is on understanding his theory as one where the voluntariness of an action is to be located in the decision which precedes it. No more is voluntariness inherited from the decision which precedes it than the courage of an action depends on the courage of the decision which precedes it. Similarity does obtain between the fact that we are made aware of the voluntariness of the action by attending to overcoming obstacles to the action and the fact that that feeling courage, not having courage, depends on overcoming fear. Rather, advantage can be taken of our now familiar locution 'with the intention that', so that any action done "with a (certain) intention" would be voluntary with respect to that intention. In this way intentionality becomes not only the hallmark of mental phenomena but a feature of all voluntary physical actions.
    It is productive to speculate that emotions, such as anger, may promote (cause) an action, such as knocking over the coffee cup, and that such an action becomes a true act of will only when the feature of intentionality is what results from the motive. It becomes a matter of necessity that an emotion can never serve as an intention. One would never say, "I did it with the intention of anger." And it appears we can extend our judgment to saying that an intention is never a motive. We conclude our discussion with an imaginary exchange between Anscombe and Brentano, one that seems to favor Brentano on the matter of the relation of decision and the voluntariness of an action.

BRENTANO: The willfulness of the act is in the decision preceding the act.

ANSCOMBE: This cannot be since I may ask of the action following the decision whether it was voluntary. I may, for example, decide to go into the pool headfirst. But at that very moment you push me headfirst into the pool. Here I can ask: "Was going into the pool headfirst a voluntary action?" The answer being no, the decision is insufficient for the voluntariness of my going headfirst into the pool.

BRENTANO: The decision to act is itself an act of will. Whether the action following the decision to act is voluntary is something we needn't decide; but even if we should decide that the decision is necessary for the action to be voluntary we need not make the stronger claim that it is sufficient.

    What is one to say about this hypothetical exchange? We must distinguish a volition from an act of will. We do do this much in the same way as we would go about distinguishing the difference between saying something is actual and saying it is possible. A volition entails and is entailed by some attempt at a basic action. A volition entails trying; an act of will, which may be a decision to act, need not. To do something I must begin to do something; but when I begin to do something and nothing is done there is an act of will but no volition. Volition is what distinguishes an voluntary action from a mere act of will. Decision, an act of will, does not entail volition. To say a volition is an act of will is like saying the actual is possible. When actuality is added to an act of will it is a volition in Bradley's sense. An act of will remains a "resolve" or, perhaps, a "pure intention" only when no change follows from the act of will. An action's voluntariness cannot be located in a *preceding decision*; for if it could, once the decision had passed either the subsequent action would not be willful, or, if it were, an action would be considered voluntary when without some act of will it would not have occurred, an unnecessary supposition.