G. J. Mattey’s Notes on Hume’s Treatise, Book 1, Part 4, Section 5

Notes on Hume’s Treatise

by G. J. Mattey

Book 1
Of the UNDERSTANDING
PART 4
Of the sceptical and other systems of philosophy.

Sect. 5. Of the immateriality of the soul.

The preceding two Sections concerned systems that purport to describe the nature of material objects. The present Section and the next treat other systems that attempt to explain certain aspects of the human spirit or soul. This Section is devoted to the system that takes the soul to be an immaterial substance.

In the Scholastic philosophy derived from Aristotle, the soul was taken to be a substantial form, which is responsible for acts of the intellect. (Substantial forms are part of what the author called in Section 3 the antient philosophy.) The soul is not material, though during the life of the human it informs the matter of the body. According to Aquinas, the soul, as an immaterial form, continues to subsist after the death of the body (Summa Theolgicæ Question 75, Article 6).

Descartes rejected substantial forms and instead considered the human soul to be a substance, which he called a thinking thing. The other kind of substance Descartes acknowledged was body or extended substances. He lays out the distinction in the Principles of Philosophy, Part I, Article XLVIII:

Locke also rejected substantial forms and held that there are two kinds of substances, spirits and bodies. The idea of a substance is that of a collection of ideas inhering in an unknown substance or substratum (An Essay concerning Human Understanding, Book II, Chapter 8).

Berkeley held as well that the soul is an immaterial substance, but his reason was that matter does not exist at all, so that everything is immaterial. (See Principles of Human Knowledge, where Berkeley argues against the existence of matter and contemplates the consequences of his conclusion.)

Spinoza held that the minds of finite spirits, as well as all bodies, are modes (or ways of existing) of the attributes of thought and extension, respectively, belonging to a single monolithic substance he called God. A substance is defined as that which is in itself, and is conceived through itself. In the Corollary to Proposition X of Part II of the Ethics, Spinoza claims to have proved that the essence of man is constituted by certain modifications of the attributes of God. In the Corollary to the subsequent Proposition XI, he elaborates:

1. The result of the previous two Sections was that every system (i.e., that of the ancient and that of the moderns) attempting to describe the nature of external objects and the idea of matter is fraught with contradictions and difficulties. Since it seems that we understand these external things better than we understand the mind, it appears that its nature is so much the more obscure, and uncertain. However, although there are many things that are obscure about the nature of the mind, the intellectual world can be described without any such contradictions, as those we have discover’d in the natural world. What is known concerning it, agrees with itself; and what is unknown, we must be contented to leave so. [However, in the Appendix, the author confesses that he cannot escape contradictions in his account of the nature of the mind.]

2. Some philosophers are not content to leave the unknown alone, but in so doing, they run into contradictions that are due to the nature of their accounts, rather than the nature of the mind. These philosophers suppose that our perceptions inhere in either material or immaterial substances. [Thus, the view of those who say that our perceptions inhere in material substances contradict those who claim that they inhere in immaterial substances.] The author wishes to put a stop to the endless disputes between these two sides, and he proposes to do so by asking both camps the question, What they mean by substance and inhesion. When, and only when, that question has been answered will it be reasonable to enter seriously into the dispute.

3. In Section 3, it was shown that this question cannot be answered with regard to matter and body. [The alleged identity and simplicity of substance is shown to be a mere fiction of the imagination, as is the alleged relation of inhesion of qualities in a body on which the qualities are said to depend.] When applied to mind, the question of the meaning of substance and inhesion faces all the same problems as with body, and others peculiar to mind itself. [The subject of the identity of the mind will be discussed in detail in the next Section.] The author once again deploys his principle from Part I, Section I, that every idea is a copy of some preceding impression. So, if we have an idea of mental substance, we must have had an impression of it. But this poses a difficult, if not impossible problem. If an impression of substance represents it, presumably it would have to resemble the substance. But by hypothesis, an impression is a perception that depends on, or inheres in, a substance. Since it is not a substance and has none of the particular qualities or characteristics of a substance, how could it represent a substance by resembling it? [Note the similarity with the argument of Berkeley given above.]

4. We now turn from the modal question of how an impression could represent a substance to the factual question of how it does represent a substance. The author challenges anyone claiming that we have an idea of mental substance to point out the impression that gives rise to it, explain how the impression operates, and show its origin. This would require answering a number of questions that the author presumes cannot be answered:

5. One might wish to avoid answering these embarrassing questions by claiming that we do not need an impression to get an idea of a substance, but that a definition will do. Such a definition might be: “something which may exist by itself.” “ [A definition of this sort was offered by Descartes and Spinoza, as noted above under Background.] The author argues that this definition cannot be adequate because it fails to respect the distinctions between substance and accident and specifically between the soul and its perceptions. He appeals to two principles. The first is that what can be conceived clearly to exist can exist, and what can be conceived clearly to exist in a certain way can exist in that way. The second is the familiar principle that whatever is different can be distinguished, and what can be distinguished can be separated in the imagination. Now take the case of an individual perception. It is different from other perceptions and from everything else, for that matter. So, it can be separated in the imagination from everything else. Thus, it can be conceived clearly to exist separately from everything else, and have no need of any thing else to support [its] existence. But in that case, the perception is a substance, according to the definition. [The consequence that a perception is a substance completely undermines the distinction between perception as an accident of a substance in which it inheres.]

7. Beginning in this paragraph, and continuing through paragraph 16, the author takes up the topic of whether the mind can be locally conjoined with matter. The question is said to arise in the context of a common argument for the immateriality of the soul. What is extended consists of parts, and any extended thing is divisible in the imagination if not in reality. Thought or perception, by contrast, is inseparable and indivisible. The claim is that what is separable and divisible cannot be conjoined with what is inseparable and indivisible. The argument is by reductio ad absurdum. Suppose that a thought is conjoined with a divisible object. Then the thought must be located at some place within the dimensions of the divisible object: either on its right or left side, top or bottom, surface or interior. The location could be in one part of the divisible body or in the whole body. If the thought is located in one part of the body, then since the thought is indivisible, the part must be indivisible as well. In that case, the thought is not conjoined with the extension, but rather with an unextended (because indivisible) part of the extended body. If the thought is located in the whole body, then the thought is in fact extended and divisible just as the body is. This result is absurd and contradictory, since it is inconceivable that thoughts such as passions have measurable dimensions. Therefore, a thought cannot be conjoined with a divisible object.

8. The author notes that this argument concerns only the possibility of the joining of thought and matter in one place, not whether the substance of thought is material. He takes this as an entrée into a discussion of in general what objects are, or are not susceptible of a local conjunction. He speculates that this discussion will generate some significant results.

9. The ideas we have of space and extension are said by the author to be derived from impressions of sight and touch alone. Sight conveys colored objects that are disposed in a certain way toward one another, as does touch with respect to solid objects. On the other hand, a sense such as taste does not convey an idea of space or extension. Consider the burning taste of a hot pepper, which increases in intensity over a short period of time. This increase is nothing like the increase in the size of a visible object. The same holds of several sounds which combine to make a louder sound, as with the simultaneous playing of several instruments of the same kind. We may indeed make a judgment that the component sounds are a certain distance away and are projected from certain angles. But this is the result of custom and reflection alone. [Not only do no other impressions of sense convey an idea of space and extension, but neither do any impressions of reflection, such as desires.] Suppose we wish to locate a desire at a certain place. If the place is extended, then the desire will have the shape of the place, say as being triangular. No desire, and no taste, sound, or smell, is triangular or any other shape. We might try to locate the desire at a mathematical point, even though both are indivisible. If we could do this, then, we could also increase the desire by addition of others located at adjacent points, so that it would have dimensions, which is evidently absurd.

10. Given this argument for the inability to locate most impressions of sensation and all impressions of reflection, the author recognizes that it will be of no surprise that he endorses a maxim that is rejected by several metaphysicians. The maxim is that an object may exist, and yet be no where. [Thus, tastes, sounds, desires, etc. exist but have no location.] I assert, that this is not only possible, but that the greatest part of beings do and must exist after this manner. Existing nowhere is understood as follows. First, the situation of parts does not generate a figure or quantity. Second, the whole object cannot be said to be contiguous with or distant from other objects. All our perceptions except sight and touch, and all the objects of those remaining perceptions, are of this kind. As an example of perceptions, moral reflections cannot be placed on either side of a passion. As an example of the objects of a perception, one cannot smell the circularity or squareness of a body. These objects and perceptions, so far from requiring any particular place, are absolutely incompatible with it, and even the imagination cannot attribute it to them. It might be held in response that it is absurd to say that these things exist nowhere. But the author has shown that the passions and sentiments do not appear to the perception to exist in any place, since if they did so appear, the idea of extension could be derived from them. If, on the other hand, they appear not to be located, it is possible that they are not located, since anything which we can conceive is possible.

11. The next claim that the author makes is that simple perceptions that exist nowhere cannot be conjoined locally with objects that exist somewhere, i.e., extended, divisible bodies. This is not necessary to prove, given that in order for there to be a relation, there must be some common quality of what is related [and specifically, there is no common quality of location that can be the basis of the relation of the two].

[Footnote. The author refers to Part I, Section 5, where he had laid down the general principle that all relations require some degree of resemblance between what is related.]

What may be better worth remarking is that questions of local conjunction occur not only in metaphysics, but also in common life. The example given is that of a fig and an olive located at opposite sides of the table. The two fruits have tastes, one sweet and one bitter. We ordinarily think that the tastes are in the fruits, and since the fruits are separated by a distance, the tastes are separated by a distance as well and thus exist at some place. This is so notable and so natural an illusion, that it may be proper to consider the principles, from which it is deriv’d.

12. Although the tangible and visible qualities of the fruits (or other extended objects) are not conjoined spatially with their gustatory qualities, they do stand in other relations to each other. Specifically, there are causal relations between them (whichever is the cause or the effect) and contiguity in the time in which they appear to the senses. [We taste the fruits immediately after seeing and feeling them.] The effect of these two relations on the mind is that the thought of one quality gives rise to the thought of the other. Moreover, we try to impose a new relation, conjunction in place, on the visible or tangible qualities and the qualities of taste, in order to make the transition even easier and more natural. The author will later [in the next section] develop the claim that when two objects are related, we have a strong propensity to try to make their union complete by adding a new relation. When we arrange bodies [in our thought] we place resembling bodies in contiguity with one another, or at least in corresponding ways of seeing them [and vice-versa]. We do this only because we feel a satisfaction in joining the relation of contiguity to that of resemblance, or the resemblance of situation to that of qualities. This tendency has already been seen, in the case where we imagine a resemblance between our particular sensory impressions and the external causes of them.

[Footnote. The author refers to Section 2, toward the end. In paragraph 54, he explains our tendency to believe that external objects in general resemble our impressions by the quality of the imagination that all of its ideas are taken from preceding impressions. Since the only thing we can conceive is impressions, we must make every thing resemble them. In paragraph 55, he explains why we believe that our particular impressions resemble particular objects. This is due to the relation of cause and effect between particular impressions and their causes.]

Although there are other instances of the tendency of the imagination to feign a new relation, the present case in which an imaginary spatial relation is joined to those of causation and contiguity in time is as evident as any.

13. Reflection on these confused notions of a spatial union of the taste and other qualities of objects such as a fig reveals it to be something altogether unintelligible and contradictory. The author poses a question to which he argues no satisfactory answer can be given. The question is whether the taste, if it is located in the body, is in one part of the body or all the parts. It is not in one part, because we experience that the various parts of the object taste the same. But if we try to conceive the taste as being in all parts of the body, then we must think of it as being extended and having a shape, a result which is absurd and incomprehensible. There is, therefore, a conflict between our inclination to try to incorporate the taste with the extended object and reason, which shows us the impossibility of such an union. Rather than confronting the conflict head-on and giving up one of the two principles, we instead involve the subject in such confusion and obscurity, that we no longer perceive the opposition. When we suppose that the taste is in the fig, we think of it as existing in all parts of the fruit, yet without being extended or divisible into parts. In our familiar way of thinking about the taste, we adopt a principle of scholasticism, which, when crudely propos’d, appears so shocking: that the whole is in the whole and the whole is in every part of the whole. Which is much the same, as if we shoul’d say, that a thing is in a certain place, and yet is not there.

14. The absurdity just described is the consequence of our trying to assign a place in objects to what cannot exist in a place, and this is done because we are trying to effect a complete unity that is begun with the relations of causality and contiguity in time, by adding the relation of conjunction in space. This is as good a case as any for the triumph of reason over prejudice. Reason presents us with the following choices with regard to some beings (qualities such as taste). They:

15. This argument condemns the view of materialists who would conjoin perceptions with matter. But it tells equally against those who affirm the immateriality of the soul by conjoining all thought with a simple and indivisible substance. [The argument for this claim begins in this paragraph and is completed in the next.] We are informed by the most vulgar philosophy that objects can make themselves known to us only through the intermediary of an image or impression. An object such as a table that appears to me now is only a perception, and all its qualities are qualities of a perception. Extension is the most obvious of its qualities, so extension is a quality of a perception, which therefore has parts. The situation of the parts affords us ideas of distance and contiguity, length, breadth, and depth. Figure is the termination of length, breadth, and depth, and it can be moved, separable and divided. The latter two qualities are the distinguishing properties of extended objects. The idea of extension itself is copied from impressions, and so must agree perfectly with these impressions. And this is to say that the idea of extension is itself extended.

16. Given the conclusion that perceptions, impressions and ideas are extended, the free-thinker [not bound by religious dogma] may turn the argument just used by the theologians to disprove the materiality of the soul against their doctrine of its immateriality. The question now becomes how they can incorporate a simple and indivisible subject with an extended perception? The same questions can be asked as before: where exactly is the immaterial soul located on the extended perception? In one particular part, or in the whole (without thereby being extended)? Or is the whole substance in the whole perception and equally in all its parts? ’Tis impossible to give any answer to these questions, but will both be absurd in itself, and will account for the union of our indivisible perceptions with an extended substance.

17. Having completed his argument against the local conjunction of both unextended perception and extended qualities, and extended perception and unextended substance, the author returns to a question that he has already dismissed as meaningless, namely, what is the substance of the soul? [The original question was whether the substance of the soul is material or immaterial.] He will argue that the doctrine of the immateriality, simplicity, and indivisibility of a thinking substance is a true atheism. This doctrine in particular will justify the infamous view of Spinoza [that there is a single substance that has both material and immaterial attributes]. The advantage of this discussion is that it will prevent his opponents from attacking his own doctrine that perceptions exist nowhere, because their arguments for an immaterial unthinking substance can be so easily retorted upon them.

18. Spinoza’s view is that the universe is a simple substance in which both matter and thought inhere. There is only one substance, says he, in the world; and that substance is perfectly simple and indivisible, and exists every where, without any local presence. [Only the modes of the attribute of extension (finite bodies) are locally present, for Spinoza.] All that we discover both internally and externally is modes of the simple substance with no separate existence of their own. Every passion of the soul; every configuration of matter, however different and various, inhere in the same substance, and preserve in the themselves their characters of distinction, without communicating them to that substance, in which they inhere. There is a single substratum, so to speak, which supports all modifications but is not itself modified by them, and which is responsible for all variations, without becoming varied itself. Neither time, nor place, nor all the diversity of nature are able to produce any composition or change in its perfect simplicity and identity.

19. The author describes this system as hideous, while at the same time nearly identical to the hypothesis of the immateriality of the soul, which has become so popular. The argument for this claim relies on a previous thesis.

Our idea of an external existence and that of a perception can never be specifically different from each other because every idea is derived from a previous impression. We may try to think of the two as different, but the only idea we can have of an external object is that of relation [perception to external existence] without one of the relata [external existence] or simply that of a perception itself.

20. From this follows a conclusion that at first might seem fallacious, but upon the least examination will be found solid and satisfactory. We may suppose, but never conceive, a difference in kind (or specific difference) between perceptions and external objects. Upon this supposition, we cannot know for certain whether conclusions we draw about the agreement or disagreement of impressions apply to external objects. Conversely, we can know that any such claim about the objects will apply to the impressions. The reason is not difficult. On the one hand, if we suppose that impressions and objects are different, we cannot tell whether reasoning about the impression applies to the object. ’Tis still possible, that the object may differ from it in that particular. But any reasoning about the object must apply to the impression. This is the case because the idea we have of the object is derived from the impression, so whatever applies to the idea applies to the impression. The only way in which our the conclusions of our reasonings about objects might not extend to impressions is if they are irregular.

[Footnote. An example of irregular reasoning is the reasoning described in Section 2 [paragraphs 19 and 20], from the coherence of our [interrupted] impressions [to the uninterrupted existence of the object].

So, with the exception given, if we suppose that impressions and objects are distinct in kind, all reasonings about objects extend to impressions, but not all reasonings about impressions extend to objects.

22. Despite the evidence of the argument beyond all doubt and contradiction, its parts will be examined in more detail. The question is whether all the absurdities that have been found in Spinoza’s system apply to the theologians’ system of the immateriality of the soul.

[Footnote. The arguments are to be found in Bayle’s Dictionary, in the article Spinoza.]

23. The first criticism is couched in the terms of the scholastics—terms that the author has repudiated as meaningless in paragraph 6 above. Although he characterizes it as a way of talking rather than thinking, he takes the following argument to be just, as far as we can understand it. The extension of the universe is a mode of substance [and more specifically, of substance’s attribute of extension]. As such, the universe must be identified with substance, i.e., that simple, uncompounded essence, in which the universe is suppos’d to inhere. But if this is so, the uncompounded, indivisible substance would have to expand into an extended universe or the extended universe would have to contract into a single point so as to answer to the indivisible substance. [Yet neither is possible, so the notion that an extended universe inheres in an unextended substance is thereby impossible.] [This argument resembles the second argument given by Bayle above.] The author maintains that this argument can be applied to the relation between our extended perceptions and the simple essence of our soul, with only a change in terms. [That is, if extended perceptions are taken to be modifications of the soul, then the soul is extended. But then either the simple soul must be expanded to admit of extension, or the extended perception must be reduced to a single, simple, indivisible point.] In this way, the ideas of extended objects and of extended perceptions are in every respect the same, although we suppose them to be different in a way that is unknown and incomprehensible.

The second criticism follows pretty closely the first argument given by Bayle and quoted above. In the author’s terminology, any idea of a substance applies to matter, and every idea of a distinct substance applies to a distinct part of matter. So, we must say that matter is a substance. [For Spinoza, matter is a mode of the attribute of extension, and a substance is no different from its attributes.] Further, given that matter has distinct parts, it is composed of distinct substances. The author notes that he has proved already that we have no perfect idea of substance. But if we try to understand it with the words something, that can exist by itself, then every perception is a substance, because every perception can exist by itself, and every part of an extended perception is a distinct substance as well. And consequently the one hypothesis [that the soul is an immaterial, unextended substance] labours under the same difficulties in this respect as the other [that God is an unextended substance].

25. The third criticism follows, and uses the same example as, the third criticism by Bayle cited above. According to the Spinozistic doctrine, the same substance is modified into incompatible forms. But it is inconceivable how a substance can be modified such as to be both a round table and a square table at the same time. The same problem arises for the impressions of the two tables. [How can an two incompatible extended perceptions inhere in a single immaterial substance?] The finds that the answer is no more satisfactory in one case than in the other.

26. The upshot of this examination is that not only is the system of the immateriality of the soul subject to the same difficulties as the Spinozistic system of a universal substance, but reasons to support the former would support the latter, preparing the way for a dangerous and unrecoverable atheism. One way of resisting this conclusion is to claim that it rests on the description of thoughts as modifications of the soul. Thoughts might instead be described as actions of the soul, a term that the author calls more ancient than ‘modification’ but also more modern. [The ancient philosopher Aristotle in De Anima, Book I, Chapter 5, refers to the affections and actions of the soul, e.g. reasoning, sensation, pleasure, pain, &c. The modern philosopher Locke described perception, thinking, doubting, believing, reasoning, knowing, willing as actings of our own minds (An Essay concerning Human Understanding, Book II, Chapter I, Section 4. Berkeley described ideas as being in the mind in the sense of being perceived by the mind. The making and unmaking of ideas doth very properly denominate the mind active (Principles of Human Knowledge, Section 28).] An action is taken by the author to signify much the same thing, as what is commonly call’d an abstract mode; that is, something, which, properly, speaking, is neither distinguishable nor separable from its substance, and is only conceiv’d by a distinction of reason, or an abstraction. [Distinctions of reason were treated in the last two paragraphs of Part I, Section 7.] The next two paragraphs will show that this terminological shift will be of no help in defense against the criticisms of the system of an immaterial soul.

27. The first point is that the notion of action described in the preceding paragraph does not apply to perceptions. An action is an abstract mode that is not separable from its substance, but perceptions are in fact all really different and separable, and distinguishable from everything else, which we can imagine. It is commonly thought that we may describe the way in which perception is an action of the soul by appeal to the way that motion is an action of a body. However, the author takes this to confound us rather than instruct us. The problem is that motion to all appearance is not a change in a body per se, but rather only a change in the relation of one body to another. We might compare the situation of a person walking in the garden and another imprisoned in a dungeon. That of the former changes with respect to the earth, while that of the latter does not. However, their perceptions are not relative in this way when the first person is with agreeable company and the second is trembling in terror: the difference between the two is radical, and of quite another kind. To return to the main thread, we take objects to be separate existences when the ideas of them are distinguishable and separable. The same condition applies to the ideas themselves: given that they are distinguishable and separable from one another, we should conclude that they exist separately. But even if we do not adopt this line of reasoning, it is impossible to say how we can have different or even contrary perceptions in a single soul, given that we have no adequate idea of the soul’s substance. As a consequence, we can never tell in what sense perceptions are actions of that substance. [That is, we cannot tell how we can make a distinction of reason between contrary perceptions as we can between, say, the color and shape of a body.] The use, therefore, of the word action, unaccompany’d with any meaning, instead of that of modification, makes no addition to our knowledge, nor is of any advantage to the doctrine of the immateriality of the soul.

28. The second point is parallel to what was stated earlier. If the change in terminology is to be of advantage to the doctrine of immateriality, it must also be of advantage of atheism. The notion of an action cannot properly be confined to souls. The atheist could hold that plants, animals, humans, and perhaps other beings are nothing but particular actions of one simple universal substance, which exerts itself from a blind and absolute necessity. Although this appears absurd, and the author agrees that it is unintelligible, he adds that he has shown that any absurdity here is equally applicable to a like supposition concerning impressions and ideas.

29. Having dispensed with the hypothesis of the soul as a substance and that of the local conjunction of the soul and matter, the author turns to a hypothesis which is more intelligible than the former, and more important than the latter. The claim is that matter cannot be the cause of our perceptions. It is said in the schools that no matter how they are varied, the only result of changes in matter and motion is in the situation of bodies. [See, for example, the argument of Locke, cited above.] No matter how much a body is divided, re-configured or moved, it remains a body with a certain shape and state of motion. It would be absurd to single out some motions, such as an elliptical one, as being a passion or moral reflection, while another motion, say a circular one, is not a perception at all. The same holds for the notion that the collision of two spherical bodies becomes a pain, while that of two triangular ones becomes a pleasure. Now as these different shocks, and variations, and mixtures are the only changes, of which matter is susceptible, and as these never afford us any idea of thought and perception, ’tis concluded to be impossible, that thought can ever be caus’d by matter.

30. Although this argument is compelling to most, it is easily refuted. The basis of the refutation is the doctrine of the author that we do not perceive causality as such, but know of it only through our experience of constant conjunction. [See Part III, Section 6 for the argument for this thesis.] To this is added the claims that no real objects are contrary and that all objects which are not contrary may stand in the relation of constant conjunction. [For this claim, see Part I, Section 5.] The author has inferred from these two principles that, to consider the matter a priori, any thing may produce any thing, and that we shall never discover a reason, why any object may or may not be the cause of any other, however great, or however little the resemblance may be betwixt them.

[Footnote. This inference is made in Part III, Section 15. Because of its conclusion, the author found it necessary to lay out rules by which actual causes and effects can be discovered.]

This observation evidently destroys the precedent reasoning concerning the cause of thought or perception. First, we perceive no connection among bodies themselves that would allow us to know independently of experience how they will behave when situated relative to other bodies. If a weight is placed at one end of a lever and an equal weight at the other end, we cannot say without experience how the bodies will move. So the a priori claim that motion cannot cause thought can be turned around: we do not know a priori how bodies will move bodies. Yet we do not conclude that bodies cannot move one another, since there is no more apparent connexion in the one case than in the other. Second, we know from experience of constant conjunction that bodies move bodies. It may be that there is also constant conjunction between the motion of bodies and the existence of thoughts. Nay, ’tis not only possible we may have such an experience, but ’tis certain we have it; since every one may perceive, that the different dispositions of his body change his thoughts and sentiments. Changes in our bodies are constantly conjoined with changes in our thoughts, and so they should be recognized as causes of our thoughts. It might be claimed that the changes in our thoughts are due to the union of our soul and our body. But this claim is beside the point, since the subject here is the cause of thought, not of its substance. And when we consider that judgments of causality are based on constant conjunction, and that this is the only consideration when we pronounce on the cause of the motion of bodies, we may certainly conclude, that motion may be, and actually is, the cause of thought and perception.

31. This leaves us with an apparent dilemma. We could claim that the only causes are those whose connections can be perceived by the ideas of the relevant objects. Or we could allow that all objects which are constantly conjoined stand in causal relations. The author lists two consequences of choosing the first horn of the dilemma. The first consequence is that we banish causality entirely from the universe, and even from God. Our idea of God, like all other ideas, is derived from impressions, and we find nothing in these impressions of causal efficacy, or even any connexion with any other existence. One might object that we do find a necessary and inevitable connection between the idea of a being with infinite power and what that being wills. The author, however, professes not to have such an idea because he does not have an idea of any power. [See Part III, Section 14 for an extensive defense of this claim.] The only idea of power we have involves a connection between cause and effect. So if we change the expression, we can describe an idea of a being that is infinitely causally efficacious. But in that case, all we are really saying is that a being, whose volition is connected with every effect, is connected with every effect; which is an identical proposition, and gives us no insight into the nature of this power or connexion. Now the author turns to the second consequence of accepting the first horn of the dilemma. Suppose that to make up for the fact that we can perceive no connections in the ideas of natural objects, we suppose that a supreme being causes everything. Such a view, the author asserts, leads us into the grossest impieties and absurdities. If, because of our failure to perceive connections, we need to appeal to the supreme cause to explain the production of motion and thought, then we must apply this solution to all causes. In particular, the volitions and perceptions of individual human beings must be caused by the deity as well, since they have no more apparent connexion either with one another, or with the suppos’d but unknown substance of the soul. Several philosophers have attributed all agency except volition (or an inconsiderable part of volition) to God. [Malebranche denied that our volitions to produce ideas and sensations and to move our bodies are causally efficacious. See the for the former, the Illustration to Book I, Part I, Chapter 1 of The Search after Truth , and for the latter, Illustration to Book VI, Part II, Chapter 3. We have a natural inclination toward the good, which constitutes our will but is not caused by our willing. What we can cause by our willing is to direct our general inclination toward the good toward one specific object or another (Book I, Part I, Chapter 1).] It is this ability to channel our natural inclination toward the good in the direction of particular objects that the author calls an inconsiderable part of volition.

[Footnote. This view is attributed to Malebranche and other followers of Descartes.]

The exclusion of volition from divine causality is easily seen to be an attempt to cover up the impious consequences of the doctrine of causality by the supreme being. If nothing be active but what has an apparent power, thought is in no case any more active than matter; and if this inactivity must make us have recourse to a deity, the supreme being is the real cause of all our actions, bad as well as good, vicious as well as virtuous.

32. Since the first half of the dilemma must be rejected, we are left with the second half, viz. that all objects, which are found to be constantly conjoin’d, are upon that account only to be regarded as causes and effects. Because no objects are contrary, and every object that is not contrary may be conjoined constantly with any other, for all we can tell by considering mere ideas, any thing may be the cause or effect of any thing. This result is gives the advantage to the materialist [who holds that matter can cause thought, over those who would deny that it can].

33. We at last arrive at the final decision upon the whole. There are three main results of the discussion.

34. The author asserts that philosophy is a sovereign authority that should be autonomous in its researches and not apologize to every group that takes offense at its conclusions. To do so would be like a king defending himself against a charge of treason brought by his subjects. The only occasion where such justification is called for is when those conclusions seem to conflict with religion, whose rights are as dear to [philosophy] as her own, and are indeed the same. The author will in the next paragraph try to show that his arguments are in no way dangerous to religion.

35. There is no foundation for any conclusion formed a priori or independently of experience about either the operations or duration of any object of which we can conceive. The reason is that we can imagine any object not acting at all or being instantaneously annihilated. And it is an evident principle, that whatever we can imagine, is possible. [So it is possible for any object that we can conceive either not to act or not to continue to exist.] The lack of a foundation for reasoning a priori applies to objects, spiritual (simple and unextended) no less than material (compound and extended). Thus, we cannot justify a priori from metaphysical arguments the claim that the soul is immortal, either in the sense that it cannot cease to operate or that it cannot cease to exist. Arguments for either of these two conclusions are equally inconclusive. [See the argument from Berkeley, quoted above.] The argument can be extended to claim that because a spirit has no parts, its operations cannot be brought to an end by its dissolution. But it remains possible to imagine the discontinuation of the existence and operations of the soul.] There are foundations for arguing for the continued existence and operations of the soul, both from the observable nature of man and from the analogy between man and nature. The arguments for continued existence and continued operations are equally strong and convincing. [The author does not reveal what these arguments are.] Even if no addition to the argument for religion is made by the author’s philosophy, no subtraction is made either, and every thing remains precisely as before.

The topic of the immateriality of the soul is not discussed in the Enquiry. The thesis of the last two paragraphs of the Section, that philosophy neither adds to nor subtracts from the arguments for religion, is re-formulated in Section 10. Hume thinks that his reasoning that undermines the credibility of testimony about miracles may serve to confound those dangerous friends or disguised enemies to the Christian Religion, who have undertaken to defend it by the principles of human reason. Our most holy religion is founded on Faith, not on reason; and it is a sure method of exposing it to put it to such a trial as it is, by no means, fitted to endure. In Section 11, there is a speech made to the people of Athens in the name of Epicurus, arguing that philosophy is no threat to the state. The speaker asserts, The religious philosophers, not satisfied with the tradition of your forefathers, and doctrine of your priests (in which I willingly acquiesce), indulge a rash curiosity, in trying how far they can establish religion upon the principles of reason; and they thereby excite, instead of satisfying, the doubts, which naturally arise from a diligent and scrutinous enquiry. Thus, while philosophical investigations need not interfere with religious belief, if they are extended to religious matters, they have a harmful effect upon that kind of belief.