Return to Owen

Locke and Hume on Belief, Judgment and Assent*

David Owen

Part One: Introduction

In 1.3.5, when Hume launches his investigation of causal reasoning, and his account of how we come to believe in unobserved objects, on the basis of observed ones, he subdivides what needs to be explained into three parts:

Hume raises the first issue only to set it aside. We do not and cannot know the ultimate causes of our impressions of the senses. The second issue is a matter of explaining how we get from the observed impression to the unobserved idea. Hume explains this transition by appeal to the association of ideas in Section vi, "Of the inference from the impression to the idea". The third issue remains: AThe nature and qualities of that idea@, i.e., Hume must explain what it is to believe something rather than simply have the idea occur in the imagination. Hume=s answer is disarmingly simple. Just as impressions and ideas Adiffer from each other only in their different degrees of force and vivacity@, so too a belief differs from a mere idea only because of its Aadditional force and vivacity.@ The extra force and vivacity of a belief comes from the impression with which the idea is associated, so the official definition of belief becomes: AA lively idea related to or associated with a present impression.@ (T 96)

Hume=s account has been much reviled, especially when it is considered, as Hume intended it to be, as an account of judgment and assent as well as belief. Consider the following important footnote:

There are several remarkable things about this passage, but for the moment let us just note the following. Judgment, especially judgments about existence, need not be a matter of predicating one idea of another. AGod exists@ appears to be Aa proposition, which contains only one idea@. Thus Hume wants to resolve judgment into conception, though he recognizes that there is Aa remarkable difference@ between what happens when we merely conceive of something, and Awhen we join belief to the conception, and are perswaded of the truth of what we conceive.@ For Hume, to believe something is to assent to it as true, and to give an account of belief is to give an account of judgment. Hume wants to discuss Athe nature of belief, or the qualities of those ideas we assent to@ (T 94). Although a belief is an idea conceived of in a certain way, it is clear that Hume frequently uses the term >belief= to pick out what it is about an idea, when it is believed, that distinguishes it from ideas that are merely conceived. Used in this way, >belief= means just what >assent= means.

A colleague of mine, when giving a course on >the judgment problem=, started off with what he called AHume=s spectacularly inadequate account,@ while another described Hume=s account of belief as Aappalling.@ Barry Stroud, referring to the long footnote at T 96-97, just quoted, claims that because Hume Asees judging as just a special case of an object=s being present to the mind... he does not see the special difficulties that creates. And he does not see that without an account of how ideas combine to make a judgment or a complete thought he can never explain the different roles or functions various ideas perform in the multifarious judgments we make, or in what might be called the >propositional= thoughts we have.@ It is clear what is held to be inadequate in Hume=s account. No >single idea= theory can hope to account for what it is to predicate one idea of another, to form a complete thought with propositional structure. A fortiori, no such theory can account for what it is to assent to such a proposition. On the other hand, as we have seen in the long footnote, Hume is perfectly aware that the standard account of judgment involves Athe separating or uniting of different ideas,@ or what we would now call predication. He explicitly rejects this account, in favour of his own. This is not some simple oversight or unawareness of the problems that it will bring. In fact, Hume recognizes that by reducing judgment to conception, he is left with the difficulty of accounting for the Aremarkable difference... when we join belief to the conception, and are perswaded of the truth of what we conceive.@ I will argue that this problem, of accounting for assent, arises for Hume precisely because he has abandoned an account of judgment as predication. Furthermore, Hume asserts that the problem is new (AThis act of mind has never yet been explain=d by any philosopher@). Again, perhaps it is new because no one before had given an account of judgment that rejected predication, and thus the problem of assent, as Hume sees it, had never before arisen. In this paper, I will explore these themes.

Part Two: Locke and Hume on knowledge and belief

One fruitful way to understand Hume on these matters is to compare him to Locke. Locke thought that knowledge was the perception of the agreement or disagreement of ideas. But knowledge, Locke thought, was Avery short and scanty@ (IV.xiv.1). Fortunately, the AMind has two Faculties, conversant about Truth and Falshood.@ The mind not only has the faculty of knowledge, Awhereby it certainly perceives, and is undoubtedly satisfied of the Agreement or Disagreement of any Ideas.@ It also has judgment, Awhich is the putting Ideas together, or separating them from one another in the mind, when their certain Agreement or Disagreement is not perceived, but presumed to be so@ (IV.xiv.4). Knowledge is the perception of agreement or disagreement of ideas, while belief, judgment or assent is the presumption or supposition that the ideas agree or disagree. Belief seems to be an approximation to knowledge; where we cannot or do not perceive agreement, we make do with supposing it. In the index to the Essay, added to the second edition, Locke directs us to sections 1 and 3 of Book IV, chapter xv >Of Probability=, for an account of what assent is. He assents to a proposition who Areceives it for true.@ What Acauses his Assent to this Proposition@ is Athe Probability of the thing@. Just as in the previous section Locke introduced judgment by analogy with knowledge, so here Locke introduces probability by analogy with demonstration:

Probability is the appearance of agreement or disagreement of ideas, and it causes our assent. That is, it causes us to presume the ideas to agree or disagree, where we don=t perceive that agreement. Furthermore, Locke makes it clear in this section that judgment is not to be understood by analogy with intuition. We don=t judge two ideas to agree or disagree the way we intuitively perceive their connection; judgment here is compared to the indirect perception that results from demonstrative reason, and it requires proofs or intermediate ideas. There is no immediate belief: belief or assent comes from probable reasoning. These points are repeated in IV.xv.3. AProbaility is likeliness to be true..., signifying such a Proposition, for which there be Arguments or Proofs, to make it pass or be received for true. The entertainment the Mind gives this sort of Propositions, is called Belief, Assent, or Opinion, which is the admitting or receiving any Proposition for true, without certain Knowledge that it is so.@ All the ingredients I want to emphasize are here: the comparison with knowledge, the dependence of beliefs on reasoning, arguments and proofs, and the fact that entertaining a proposition is not something we do prior to judging it to be true; entertaining the proposition is judging it to be true. Assent is not a separate act from proposition formation.

Locke sometimes speaks of this judgment as assent. According to Locke, believing or assenting to a proposition, is a sort of pale imitation of knowing it, a presuming rather than a perceiving. Both knowledge and belief involve proposition formation; predication is a form of affirmation or denial. Assent is not an attitude one takes towards a proposition already formed. So when Locke speaks of a self-evident proposition as something one Aassents to at first sight@ (4.7.2), he is not saying that we believe or assent to something known; he is just saying we come to know it. Hume disagrees with much of this. He doesn=t think of belief as an approximation to, or a pale imitation of, knowledge. He treats belief both as the idea believed and as that feature of the idea that constitutes our assent to it. He allows that we can conceive of a (probable) proposition without affirming or denying it. Furthermore, Hume treats this assent, with respect to beliefs, as something quite distinct from whatever it is that compels our assent to things that we know. Hume=s account of belief is quite distinctive, and this fact helps explain why he thinks of his account as so new and original.

The crucial points are these. Both Locke and Hume consider judgment to be limited to belief formation, and not to extend to proposition formation in general. Locke thinks that belief is a pale imitation of knowledge, and in both cases the predicative process by which the proposition known or believed is formed just is the process of affirming or denying one idea of another. By and large, Hume accepts this account of knowledge, to the extent that he is interested in it at all. But he rejects any predicative account of belief formation, at least for the central case of beliefs that involve existence. But in rejecting this, the Lockean account of the assent we give to beliefs is no longer available to him.

It is almost impossible, for us post-Russellians immersed in propositional attitude psychology, to resist distinguishing between forming or understanding a proposition, and judging it to be true. But we must guard against bringing this distinction to Locke: it is a fundamental error in understanding Locke=s account of knowledge and belief. For Locke, to perceive, or judge, that two ideas agree or disagree simply is to know, or believe, the proposition formed by the agreement or disagreement of ideas. And to know, or believe, such a proposition is to know it with certainty, or believe or assent to it with relevant degree of assurance. Predication isn=t distinct from affirmation or denial; understanding propositional content isn=t distinct from knowing or judging to be true. Propositions just are ideas Aput together, or separated by the Mind, perceiving, or judging of their Agreement, or Disagreement.@ (IV.v.5) But perceiving or judging the agreement of ideas is to know (or believe) the relevant proposition, to take it as true, and to be aware of the relevant degree of certainty (or assurance). There is only one act of the mind here (perception in the case of knowledge; presumption, supposition or judgment in the case of belief) and only one object (the agreement or disagreement of ideas.) As a result, Locke frequently says such things as the following: ACertainty of Knowledge is, to perceive the agreement or disagreement of Ideas, as expressed in any Proposition. This we usually call knowing, or being certain of the Truth of any Proposition.@ (IV.vi.3) To form a proposition, to know, to believe, to assent to, to take as true, to consider certain or probable, all these are just to perceive or presume the agreement or disagreement of ideas.

One motivation for resisting this interpretation, and to treat Locke as distinguishing between understanding a proposition, on the one hand, and assenting to it so that one either knows or believes it, on the other, is as follows. If the assent we give to a proposition is not distinct from the formation of that proposition, it would seem impossible to withhold assent from a proposition. But Locke explicitly allows for the witholding of assent, especially in IV.xx. So the assent, it appears, must be distinct from proposition formation. There is no doubt that Locke allows for the withholding of assent, and that this presents a problem for anyone who puts forward the interpretation I have concerning Locke and the perception or presumption of the agreement or disagreement of ideas. This is a problem with a history. Passmore thought that Locke must have changed his mind between the early account of belief in the central chapters of Book IV, and chapter 22. Ayers tries to show how the position taken in IV.22 is, by and large, consistent with the main theory of belief. I lean towards the latter interpretation. In Book IV.xx.16, Locke reaffirms that as knowledge is the perception of agreement or disagreement of ideas, we cannot stop ourselves from knowing when we perceive the agreement. He then makes, as he has before, the same point about belief or assent: Awhat upon full Examination I find the most probable, I cannot deny my Assent to.@ Nonetheless, an agent can, for better or for worse, stop the formation of knowledge or belief, by not putting one=s faculties into gear, as it were. If one hasn=t got to the stage where one perceives the agreement, or supposes it after considering the evidence, Awe can prevent or suspend our Assent.@ This is important for Locke, because A[i]f it were not so, Ignorance, Error, or Infidelity could not in any case be a Fault.@ We can, Locke suggests, suspend our assent Aby stopping our Enquiry, and not imploying our Faculties in the search of any Truth.@ Nonetheless, Locke concludes the section with a ringing affirmation of his original position: Aa Man can no more avoid assenting, or taking it to be true, where he perceives the greater Probability, then he can avoid knowing it to be true, where he perceives the Agreement or Disagreement of any two Ideas.@ (IV.xx.16)

Let us return to a further examination of Locke=s view that beliefs are never immediate, but always the result of reasoning. Locke thinks that the process via which we come to have beliefs on the basis of evidence, or judge rather than perceive propositions to be true, is probable reasoning. His account of how such reasoning issues into judgment is usually presented by analogy with demonstrative reasoning. The analogy occurs in many places, including the beginning of his discussion of probability in IV.xv.1, from which I have already quoted extensively, and in IV.xvii.17, whose description is >Intuition, Demonstration, Judgment=. The section reads:

Beliefs are formed as a result of probable reasoning, where the probability or evidence causes us to take two ideas to agree, where that agreement is not perceived. The proofs or intermediate ideas induce us to judge the proposition to be true, to presume that the ideas agree. Knowledge, where not immediate and intuitive, is formed as a result of demonstrative reasoning, where the intermediate ideas are such that each can be perceived immediately to agree with its neighbour. The two ideas at the extremes are perceived mediately to agree, via the intervention of the intermediate ideas. Knowledge essentially involves perception, while belief involves only presumption or supposition, of the agreement of ideas. Reason is the discursive faculty (IV.vii.14), the faculty whose characteristic activity is reasoning, both demonstrative and probable. Such reasoning is needed Aboth for the enlargement of our Knowledge, and regulating our Assent= because >Sense and Intuition reach but a very little way.@ (IV.xvii.2)

To sum up, Locke thought of belief or assent as a substitute for knowledge when we could only presume, not perceive, the agreement or disagreement of ideas. But such perceiving or presuming just was a matter of forming some proposition by affirming or denying it. There is no distinction between forming a proposition, and assenting to it. Predication is a matter of affirmation or denial. And unlike knowledge, all beliefs are formed as a result of reasoning.

Hume seems to accept something analogous to this for knowledge. Consider the following passage:

Conceiving two ideas Aaccording to the proposition@ just is seeing that they stand in a certain relation. Furthermore, they must stand in that relation, as long as the ideas remain the same. And this is all there is to knowledge; nothing further is needed. If the same account would explain the assent we give belief, then Hume could argue, like Locke, that belief was a pale imitation of knowledge. But he does not think the same account will do. Since concerning belief

Hume thought that Locke=s account of predication as affirmation or denial would account for the assent that attaches to knowledge only because we can=t imagine affirming what we deny or denying what we affirm in the case of knowledge. Such imaginative possibilities obtain in the case of belief, so the assent that attaches to belief must be explained in a different way.

Judgment about unobserved matters of fact does not produce assent in the way knowledge does. Some other account must be given. Hume sees only two possibilities. Either Abelief joins some new idea to those which we may conceive without assenting to them@ (T 653) or it doesn=t. If it doesn=t, then since Abelief implies a conception, and yet is something more; and since it adds no new idea to the conception; it follows, that it is a different manner of conceiving an object@ (T 653). Hume=s main argument that belief does not involve a separate idea involves some claims about existence. This argument is crucial to Hume=s account of belief in the Treatise. It is repeated in the Appendix and the Abstract, though, interestingly, it is not repeated in the first Enquiry. We will examine it in the next section.

Part Three: Belief and Existence

Hume advances a range of arguments intended to establish that belief does not involve the addition of a separate idea. The most important of these arguments turns on his claim that we have no separate idea of existence. Hume=s first example, which occurs both at the beginning of Section vii, AOf the nature of the idea or belief@, and in the footnote at T 96-97, involves the existence of God. What is the difference between merely conceiving of God and believing God to exist? We might think that existence is a distinct idea Awhich we join to the idea of his other qualities, and can again separate and distinguish from them.@ (T 94) But this is hopeless: the idea of God, and the idea of God existing, must be the very same idea; otherwise it would not be possible to distinguish between believing in, on the one hand, and merely conceiving of, on the other hand, the very same thing. If the difference involved a change in what was conceived, then we would not have the same thing now conceived, and then believed: Athe belief of the existence joins no new ideas to those, which compose the idea of the object. When I think of God, when I think of him as existent, and when I believe him to be existent, my idea of him neither encreases nor diminishes.@ (T 94)

This is a good argument within Hume=s system. It shares, indeed expresses within the radical constraints of Hume=s system, Kant=s insight that existence is not a predicate. Hume says:

Existence is not some quality that may or may not attach itself to some object. It is in a different category, and needs special treatment. One might argue that Hume did not really have a theory of judgment, or predication, and was thus systematically prevented from giving a reasonable account of belief and existence. It is true that, as Hume thinks relations are just complex ideas, he has, unlike Locke, no account of the mental equivalent of predication. This is not simply an oversight on Hume=s part; it is a crucial part of his theory. It is expressing an awareness that a picture of belief, like Locke=s, will not work, especially when that belief concerns existence. One can=t explain what it is to believe in God by predicating or affirming existence of God. Believing that God exists is not a matter of perceiving (or presuming) the agreement between the idea of God and the idea of existence. On Hume=s account of thought, this would amount to a different idea than that had by one who merely conceived of God. Furthermore, it is arguable that Hume=s refusal to accept a picture of our mental life as consisting of a language of ideas represents a determination to keep within the confines of his method. Public language has its grammatical structures, sentences, propositions and predications. But at the level of ideas, there are only associations and relations. Beliefs in existents are to be explained with these minimal tools. Furthermore, Hume thought these explanations could be carried over to animals such as dogs and birds. He thought that the fact that his account of belief was the only one that could explain both human and animal behaviour was decisive evidence in its favour:

The first part of Hume=s argument showed that belief cannot be a matter of predicating or annexing a separate idea of existence to the idea of any object. The second part of Hume=s argument attempts to show that any alteration to an idea, apart from the manner of conceiving it, i.e., apart from changing its degree of force and vivacity, will change the idea into an idea of something else:

Once again, the point concerns retaining the identity of the idea both as merely conceived and as believed. Hume makes the strong claim here that any change, apart from a change in force and vivacity, will change the idea. The result is that the difference between belief and mere conception is to be accounted for solely in terms of force and vivacity, that is to say, solely in terms of what he used originally to distinguish ideas and impressions. Ideas become beliefs by becoming more like impressions. And beliefs in the unobserved get their extra force and vivacity from their association with a present impression.

In the next section, Hume offers other considerations drawn from experience, which he calls Aexperiments@, to confirm this theory of belief and belief formation. He reiterates the main features of his account as follows:

A few pages later he sums up his account of probable reasoning as belief formation in the following, deliberately provocative, way:

Hume=s account of belief is essentially an account of believing that something unobserved, i.e., something not present to the senses or memory, exists. This helps explain an otherwise puzzling feature of his official definition of a belief as a lively idea related to or associated with a present impression. David Norton has argued that, in these same sections, Hume=s use of the phrase Amatters of fact@ typically means Aunobserved matters of fact@. I want to suggest that belief in an unobserved matter of fact is primarily belief in something=s existing or an event=s occurring, and only secondarily belief that something has a certain quality or property. The original sort of inference that Hume set out to explain was the inference from the impression of one object or event to the belief in the existence or occurrence of another object or event. By and large, he avoids the question of what it is to believe, and why it is we do believe, in observed existents. That is a question deferred until later. Impressions are more forceful and vivacious, and beliefs are explained as being approximations of impressions in this regard. The status of impressions is held to be unproblematic at this stage.

It will be remembered that Hume=s first example of belief was belief in the existence of God, and that Hume=s main argument against belief as a separate idea involved the denial that we had a separate idea of existence. The central case of belief as belief in unobserved existents helps explain why Hume did not simply say that belief is a more lively and forceful idea, but added the claim that it must be related to a present impression. Why is an association with a present impression necessary for this sort of belief? Of course, some impressions had to occur at some time for me to have this belief, but that is almost trivially true, given Hume=s account of impressions and ideas. His central case of belief is one where, upon having an impression of something existing or occurring, one comes to believe in the existence or occurrence of something else. But the explanation of this sort of belief essentially requires the present impression; it is that impression that provides the force and vivacity which when transmitted to the subsequent idea turns it from something merely conceived into a belief. Hume says

Once we realize beliefs in unobserved existents are the central case for Hume, his official definition, and the role Apresent impression@ plays in it, is no longer puzzling.

Part 4: Why is Hume=s account new?

Hume=s account of belief is quite extraordinary, and Hume remained very proud of it, in spite of the difficulties he saw and tried to remedy in the Appendix. In the Abstract, written before the Appendix, he not only takes pride in the importance of his treatment of probabilities, but he also singles out his treatment of belief, not just as an example of a new solution, but as an instance of the discovery of a new problem: AWhat then is this belief? And how does it differ from the simple conception of any thing? Here is a new question unthought of by philosophers.@ (T 652) He is only slightly more cautious in a footnote in the Treatise at T 97: AThis act of the mind has never yet been explain=d by any philosopher; and therefore I am at liberty to propose my hypothesis concerning it; which is, that >tis only a strong and steady conception of any idea, and such as approaches in some measure to an immediate impression.@

Why does Hume think that his account of belief is, not just a new answer, but Aa new question unthought of by philosophers@ (T 652)? There are, I think, three reasons for this. First, let us remember that while philosophers like Locke thought of belief as a pale imitation of knowledge, Hume thought of it as a separate cognitive state that required a different treatment. In part, this is because Locke thought that predication just was affirming or denying that two ideas stand in a certain relation (i.e., perceiving the agreement or disagreement between ideas). Hume thought something like this worked for knowledge, but not for belief, at least belief in existents. Believing that God exists is not a matter of affirming the idea of existence of the idea of God. For Locke, the question of assent is entirely absorbed into the question of proposition formation and understanding. But Hume thought that the question of the nature of our assent to beliefs remained, even after the question of the content of what was believed was resolved. As a result, the question of assent is a new question. It is not simply a matter of giving a new answer to an old question.

Secondly, Hume thought that beliefs in unobserved existents, his central case, were the result of probable or causal reasoning. Suppose one thinks, as Locke did, that probable reason functions much like demonstrative reason, providing rational support for its results. One might think that the results are also analogous: demonstrative reasoning produces knowledge, which has a very high degree of certainty, while probable reasoning produces opinion, assent or belief, which have lower degrees of certainty. So one might think that the latter states are just like knowledge, only less certain. But this depends on seeing both knowledge and belief, when they are inferentially accessed, as based on reason.

Hume accepts that in demonstration we are determined by reason. What we know as a result of demonstration is based on reason, because we are certain that the relevant connection holds between ideas. We do not intuitively and immediately perceive that connection, but thanks to intermediate ideas, and the intuitive connection they have one with another, we can come to see that it holds. Hume=s account of intuition and demonstration itself provides an account of why we accept their dictates: knowledge carries its own conviction. If our inferences from the observed to the unobserved were based on reason, we would reason via something intermediate, such as the idea of necessary connection or the Uniformity Principle. And the results of such inferences would carry with themselves the appropriate degree of assurance. But no such intermediaries are available to us prior to our engaging in probable reasoning, and Hume has explained such inferences as based, not on reason, but on custom or habit. This explanation explains only the occurrence of an idea; the question of the assurance we place in such an idea remains unanswered, prior to Hume=s account of belief. That question could not even have been raised, let alone answered, until one rejected reason as the basis of probable reasoning. To sum up points one and two, I have suggested that Hume might think that his question about belief is new because he thinks that belief is not analogous to knowledge, and is to be explained as the outcome of custom, not reason.

Thirdly, Hume=s account of belief, at least in the Treatise, is quasi-sensory. Hume attempts to account for a whole range of beliefs, including memories as well as causal beliefs of unobserved existents, by treating them as approximating to sense impressions. This unified account of diverse cognitive states that are distinct from knowledge is, I think, new. Hume explains the assent we give to these memories and beliefs, not by treating them as instances or imitations of knowledge, but by treating them as imitations of sense impressions. What transforms a merely conceived idea into a belief is precisely the same thing as distinguishes impressions from ideas, and memories from ideas of the imagination, i.e., force and vivacity. The quasi-sensory account of beliefs arises, in part, because one of the points of differentiating beliefs from mere ideas is that beliefs, like impressions of sensation and ideas or impressions of memory, and unlike mere ideas, have an influence on our passions, volitions and actions. Both these points are laid out explicitly in Section x, AOf the influence of belief@:

Beliefs differ from mere ideas by their extra force and vivacity. This is explained by beliefs (or, at least, those beliefs which are the product of causal or probable reasoning which typically are Aideas of those objects, which we believe either are or will be existent@) being related to present impressions. And it explains how it is possible that beliefs can have an influence on passions, volitions and actions similar to the influence had by impressions. It is, of course, absolutely crucial that beliefs have this influence:

Hume regards all this, not only Aas an additional argument for the present system@, i.e., the system of belief and belief formation, but also as a way of giving Aus a notion after what manner our reasonings from causation are able to operate on the will and passions.@

We have argued that Hume=s central case of belief is belief in an unobserved existent, hence the official definition of belief as a more lively idea related to or associated with a present impressions. But there are other beliefs, characterized by extra force and vivacity, that are not the result of causal reasoning, formed by an association with a present impression. Force and vivacity characterize not only causal beliefs but also the belief or assent that attends ideas of memory, and the judgments that issue from probable reasoning based on an imperfect or mixed past experience (cf. AThe probability of causes@, 1.3.12). Hume provides a summary of his account on T 153-54, which starts AThus it appears upon the whole, that every kind of opinion or judgment, which amounts not to knowledge, is deriv=d entirely from the force and vivacity of the perception, and that these qualities constitute in the mind, what we call belief of the existence of any object.@ (T 154) Hume considers three instances, in descending order of the degree of force and vivacity. 1) Memory, where the force and vivacity are Amost conspicuous,@ so that Aour confidence in the veracity of that faculty is the greatest imaginable, and equals in many respects the assurance of a demonstration.@ 2) The Anext degree@ of force and vivacity is found in causal judgments based on a uniform past experience, i.e., where Athe conjunction is found by experience to be perfectly constant, and when the object, which is present to us, exactly resembles those, of which we have had experience.@ 3) @below this degree of evidence@, there are other sorts of judgments with a lower degree of assent, i.e., Awhich have an influence on the passions and imagination, proportion=d to that degree of force and vivacity, which they communicate to the ideas.@ These are the sorts of judgments discussed in AOf the probability of causes@ and AOf unphilosophical probability@, which Hume summarizes on T 154. Of them, he says AIn all these cases the evidence diminishes by the diminution of the force and intenseness of the idea. This therefore is the nature of the judgment and probability.@ (T 154)

Memory seems to be the odd one out in this summary. Causal judgments and judgments of probability are inferential; they are judgments produced by reasoning. Memory is not produced by reasoning, though ideas of memory are derivative from the original impression. Indeed in one place Hume describes a memory as Aa repetition of that impression in the memory@ (T 86). Later, he argues:

Hume includes memory here because what distinguishes ideas of memory from ideas that are merely imagined is the very same thing as the assent we give to causal judgments: force and vivacity.

Because memories, sense impressions and beliefs have this unifying feature of assent due to force and vivacity, they are form a unified, cognitive system. Our world is not limited to what we remember and perceive; it also includes what we believe:

Our world, or Areality@, would be quite impoverished if it were limited to impressions and memories. By extending it with beliefs, we populate it, not only with the objects of our immediate causal judgments, but also with objects far removed from us in space and time. Hume uses the example of Rome, both contemporary and ancient, but he might just as well have talked about the astronomical investigations of distant galaxies:

The world that we believe in, transcending that revealed by the present testimony of the senses and memory, includes the unobserved objects we believe to exist. If beliefs did not approximate to impressions, this world would not be a unified one, nor could we reason from impressions to ideas in a smooth and effortless manner. But as beliefs are experienced much as sense impressions are, and are distinguished from ideas of the imagination in the very same way as are ideas of memory and impressions, it is almost true to say that beliefs are sensed, or something very close to that. Beliefs are believed, not because they are the conclusions of reason, but because they approximate to the experience of having sense impressions. This is the import of Hume=s comparison of probable reasoning with sensation:

The details of Hume=s account of belief formation and causal and probable reasoning are difficult and subtle, but the general picture is straightforward enough. Beliefs, like memories, differ from mere ideas by being more like impressions of sensation, i.e., by being more forceful and vivacious. The central case of a belief formed by causal reasoning is a single idea, enlivened so as to approximate an impression. This central case typically involves the belief in the existence of something unobserved, though such beliefs are assented to in just the same way as are impressions of sensation and memory. Hume=s account of belief has seemed radically defective because it is a >single idea= account. By leaving out predication, it seemed impossible for Hume=s theory to account for judgment and assent. Locke=s theory absorbed judgment and assent into predication and proposition formation. Hume saw the problems such a theory faced concerning existential beliefs. By ceasing to treat existence as a predicate, Hume was confronted with the issue of what it was to judge something to be true, or to assent to something. This issue had to be solved independently of the question of what it was to conceive something, or understand the content of a proposition. Hume thought this problem was new. He should be looked at, not as giving a bad answer to an important question, but rather as being the first in the early modern period to recognize that there was an important question here to be answered.

University of Arizona

References

Ayers, Michael: 1993, Locke, 2 vols. London, Routledge.

Ayers, Michael: 1997, critical notice Chappell (1994) The Locke Newsletter 28, 157-158.

Bennet, Jonathan: 1994, >Locke=s philosophy of mind=, in Chappell (1994), 89-114.

Chappell, Vere: 1994 ed., The Cambridge Companion to Locke, Cambridge, Cambridge University Press.

Cohon, Rachel, and Owen, David: 1997, >Hume on Representation, Reason and Motivation=, Manuscrito 20, 47-76.

Falkenstein, Lorne: 1997, >What Happens when Beliefs Conflict=, read at the Hume Society Conference, Monterey.

Hume, David: 1970, Enquiries Concerning the Human Understanding and Concerning the Principles of Morals, ed. L.A. Selby-Bigge, Oxford, Clarendon Press.

Hume, David: 1987, A Treatise of Human Nature, ed. L.A. Selby-Bigge and P.H. Nidditch, Oxford, Clarendon Press.

Kenny, Anthony: 1986 ed., Rationalism, Empiricism, and Idealism: British Academy Lectures on the History of Philosophy, Oxford, Clarendon Press.

Locke, John: 1975, An Essay Concerning Human Understanding, ed. P.H. Nidditch, Oxford, Clarendon Press.

Nuchelmans, Gabriel: 1983, Judgment and Proposition from Descartes to Kant, Amsterdam, North-Holland Publishing Company.

Norton, David, and Norton, Mary: 2000 eds., David Hume A Treatise of Human Nature, Oxford, Oxford University Press.

Owen, David: 1999a, Hume=s Reason, Oxford, Oxford University Press.

Owen, David: 1999b, critical notice of Wolterstorff (1996), The Locke Newsletter 30, 102-127.

Passmore, John: 1986, >Locke and the Ethics of Belief=, in Kenny (1986), 23-46.

Stroud, Barry: 1978, >Hume and the Idea of Causal Necessity=, Philosophical Studies 33, 39-59.

Wolterstorff, Nicholas: 1994, >Locke=s Philosophy of Religion=, in Chappell (1994), 172-198.

Wolterstorff, Nicholas: 1996, John Locke and the Ethics of Belief, Cambridge, Cambridge University Press.

Notes

*This paper arose from my discussion of related topics in Hume=s Reason (Owen, 1999a), chapters 3 and 7. Earlier versions have been given at the Hume Society Conference, Stirling, July 1998, The California Conference in Early Modern Philosophy, Long Beach, June 1999, University of California San Diego and University of Massachusetts Amherst departmental colloquia, The California Conference in Early Modern Philosophy, Berkeley, October 2000, and the conference Humean Readings, Rome, June 2000. I am especially grateful to Ted Morris, Paul Hoffman and Gideon Yaffe, who commented on the paper in its various versions at conferences, and to Wayne Waxman, Don Garrett, Vere Chappell, João Monteiro and David Norton, who have given me many helpful comments and criticisms.

Return to Owen