Shepherd on Hume’s Argument for the Possibility of Uncaused Existence

1. Shepherd’s Argument

As Shepherd herself emphasizes, the thesis at the foundation of her grand argument against Hume’s conclusions concerning necessary connection is that, ‘a Being cannot begin its existence of itself’.

Before I proceed any further, I wish my reader to grant the proposition, ‘That a Being cannot begin its existence of itself;’ because I mean to make use of it in my further reply to Mr. Hume’s doctrines; and, unless this step is allowed, I can make no further progress in this argument. (An Essay on the Relation of Cause and Effect [ERCE] 39)

Not only does Shepherd declare the importance of this thesis for all that follows it, but she frequently refers back to it later in the text, cites it as a premise in several arguments, and can plausibly be understood as relying on it precisely as much as she announces she will.² So, Shepherd certainly does take herself to need the conclusion that a being cannot begin to exist without a cause, but unfortunately it is less than obvious how to understand the argument that she gives in support of it. That argument takes the form of a reductio ad absurdum. Here is the imagined scenario that is supposed to represent the initial assumption, which will later be rejected.³

Let the object which we suppose to begin its existence of itself be imagined, abstracted from the nature of all objects we are acquainted with, saving in its capacity for existence; let us suppose it to be no effect; there shall be no prevening circumstances whatever that affect it, nor any existence in the universe: let it be so; let there be nought but a blank; and a mass of whatsoever can be supposed not to require a cause START FORTH into existence, and make the first breach on the wide nonentity around. (ERCE 34–35)

With a great deal of fanfare, detail, and care Shepherd describes what we imagine when we imagine a being coming into existence uncaused. It is precisely such a sequence that she will argue is impossible. Here is the argument.

Now, what is this starting forth, beginning, coming into existence, but an action, which is a quality of an object not yet in being, and so not possible to have its qualities determined nevertheless exhibiting its qualities? … But my adversary allows that, no existence being supposed previously in the universe, existence, in order to be, must begin to be and that the notion of beginning an action (the being that begins it not supposed yet in existence), involves a contradiction in terms; then this beginning to exist cannot appear but as a capacity some nature hath to alter the presupposed nonentity, and to act for itself, whilst itself is not in being. (ERCE 35–36)⁴

Prima facie, one striking problem with this argument is that it seems to beg the question against Hume. It appears as though Shepherd’s argument that no being can come into existence uncaused is that a description of such an event is a contradiction in terms! If true, that would certainly be a knock-down argument, but Shepherd appears to do very little here to demonstrate the truth of that claim. Her argument, such as it is, appears to be something like the following:

1. Beginning, or coming into existence, is an action.

2. An action is a quality of an object.

3. If an object can come into existence uncaused, then its beginning can only be a quality of that very object itself and no other.

4. If an object can come into existence uncaused, then its beginning is a quality of an object not yet in existence. (1, 2, 3)

5. An object not yet in existence cannot have its qualities determined.

6. Thus, a non-existent object must both have qualities (the quality of beginning to exist) and not have qualities (because it does not exist). (4, 5)

7. Therefore, an object cannot come into existence uncaused. (3, 6)

Suffice it to say that there is a great deal in that argument that would provide ample basis for an objection from Hume. (E.g., 1, 2, 4 and 5 would make for relatively straightforward targets.) Of course, some of these steps could receive support from Shepherd’s positive accounts of objects as loci of causal powers, of objects rather than events being the relata of causal relations, and of cause and effect as synchronous rather than successive. The problem with relying on those accounts, however, is that such support occurs in the text following this argument, which we have just seen Shepherd claims to be the foundation on which those later texts are grounded.

In light of these prima facie problems with Shepherd’s argument, it is not surprising to find scholars commenting on Shepherd evaluating her argument rather negatively. For example, Paoletti simply declares that Shepherd’s argument is question begging, without presenting any further evidence.

Shepherd was not afraid of Hume’s sceptical argument according to which a known effect cannot be inferred from an unknown cause. She was instead eager to insist that, as causality is a necessary connection, the appearance of the effect cannot but imply that its cause produced it; the appearance of an altered effect means that a dissimilar cause occurred. As a matter of fact, Shepherd did not demonstrate that cause and effect are necessarily connected; she rather took a deterministic view of physical phenomena for granted, which allows us to affirm that similar events are produced in similar circumstances and that this is true in past as well as in future situations. (Paoletti 2011: 51)

In the final sentence here, Paoletti accuses Shepherd of taking determinism for granted, but does not explain where she finds Shepherd doing so. Nor does she defend this as an interpretation of Shepherd.

Fantl, too, accuses Shepherd of begging the question, and while he does take more care in explicating the details of Shepherd’s view, he also finds Shepherd’s answer to Hume’s thought experiment insufficient.

Whereas the bundle view of objects and the causal view of properties seem like rather contemporary views (even if they do have historical antecedents), premise 1 [No object can begin to exist uncaused] and Shepherd’s reasoning for it seem rather old-fashioned. Her overall argument, then, is a curious mixture of the forward-looking and the nearly medieval. Why suppose that beginning to exist is an action of an object at all? For an object to begin to exist it need only be the case that one time there is no such object in the universe while at the next moment the object is present. Shepherd will worry that such a change needs to be caused. But that’s the conclusion of her argument; it’s premise 1. It’s not clear why the moving of the universe from a state of lacking such an object to having such an object requires the action of any object—the object itself or any other. (Fantl 2016: 12)

Fantl asks ‘Why suppose that beginning to exist is an action of an object at all?’ and then performs the same thought exercise that Shepherd rehearses from Hume. He imagines a time in which an object does not exist followed by a time in which it does, and does not also imagine that object having been caused to begin its existence by anything.

Notice that Fantl’s objection to Shepherd’s argument consists of his rehearsal of the very same example that she herself uses to introduce that argument, followed by the accusation that she is begging the question against Hume. It is striking, though, that Shepherd herself presents this very same example as part of her own argument. This suggests that she draws a different lesson from it than Hume and Fantl do. Shepherd follows Hume’s instructions to imagine an object, exactly as it is, but without also imagining anything to have preceded it; she concedes that we can so imagine an object; but she also claims, contra Hume and Fantl, that despite our being able to imagine such a thing, it is nonetheless inconceivable for it to actually occur. What is so striking about that argument form is that it suggests that Shepherd understands the role of imagination, meaning, and possibility in our mental lives very differently than does Hume. Whereas Hume concludes from this thought experiment that objects can begin to exist without a cause, Shepherd conducts the same experiment and is able to draw no such conclusion. Thus, there must be a something, which Shepherd rejects, that Hume takes to license the inference from the results of his thought experiment to his conclusion.

This way of understanding Shepherd’s argument develops a suggestion made by Bolton: that Shepherd’s argument here is undergirded by her implicit rejection of the theory of mental representation that leads Hume to his own conclusions with respect to uncaused existence and the conceivability test for necessity and possibility.⁵

Perhaps most important, although the reasoning sketched may seem to be circular, that is just to assume that a cause is necessary, this is not correct. It is based on analysis of the situations in which we experience things that begin to exist; there are things which pre-exist the effect, come together and immediately produce the effect. … However, the dispute is not a stand off, as this may suggest. The disagreement concerns competing theories of mind which stand or fall on empirical support and explanatory success. (Bolton 2019: 7–9)

As Bolton sees it, the argument at hand can appear to be circular, but only when one reads it in isolation from the broader dialectical situation. Understanding the place of that argument in its proper context as part of Shepherd’s more systematic approach to issues in the philosophy of mind and metaphysics reveals it to be an entirely appropriate response to what Shepherd takes to be the implausible and undefended dogma of Humeanism of the day.

More important, though, the misguided notion of the causal relation receives theoretical support from an incorrect account of the intentional content of sense perception, as Shepherd sees it. That is to deny that we perceive things as having causal connections. We are said to perceive the temporal order of singular events and note the regularity with which objects of one sort are succeeded by objects of some other sort. Then because the succession of any number of perceived events does not constitute a causal connection, belief in the existence of causes must have a source extrinsic to perception. … To Shepherd’s mind, these accounts fail to explain both the content and epistemic justification of causal beliefs, a lapse traceable to a deficient theory of the conceptual structure of our perceptions. … Intentional content to this effect is built into sensory perceptions as a result of a complex perceptual process. As she sees it, sense perception involves sensations which are subsumed under ideas of certain relations by latent operations of reason. (Bolton 2010: 245)

As I hope to show, Bolton’s explanation of Shepherd’s general line on Hume’s mistakes makes good sense of the passages at hand. Shepherd rehearses Hume’s thought experiment with great fanfare. She imagines a world in which first nothing exists, and then a moment later something does. Hume concludes from that experiment that it is possible that such a series of events could occur. Shepherd refuses to draw this inference. Why? Because the inference itself is only valid if one draws a very close connection between what we can imagine and what is possible. Shepherd, however, does not draw that connection precisely because she takes the imagination itself, or the senses, to be part and parcel of a more fundamental representational faculty, perception, and holds that pretending to separate the deliverances of the former from those of the latter is itself an illicit and misleading piece of philosophical sleight of hand. Or, as Bolton has it, idiotcy.

Hume and Berkeley are cited for advocating an especially impoverished theory of perception, on which we perceive nothing but sensations in spatial and temporal relations. They recognise no faculty of reason capable of providing structure to the sensory given, and thus capable of representing external objects and causal connections. As she sees it:

Idiotcy appears to be little else, than an incapacity for further perception than what resides in the immediate impressions created by the use of the five organs of sense, and the power of motion … and although I must allow that [children] do not, cannot argue formally on the subject; yet, … their understandings take notice of, (i.e. their latent powers of observation enable them to perceive,) certain simple relations in those ideas of sensation, which are determined to their minds by the organs of sense. (PWMS 2.315) (Bolton 2010: 245–46, ellipses Bolton’s)

Again, my plan in what follows is to show that Bolton’s general line on Shepherd can be put to use in understanding Shepherd’s argument for the thesis that no object can come into existence uncaused. Paoletti and Fantl read this argument in isolation from the rest of Shepherd’s systematic approach to these issues, and so take its entire content to be contained in a single short paragraph. Bolton’s focus thus far has been on using the philosophical system that Shepherd develops in Essays on the Perceptions of an External Universe (EPEU) to situate that paragraph in its broader context, as part of Shepherd’s attempt to provide a plausible alterative system to Hume’s. What I hope to do is show that a close reading of the text of ERCE itself reveals Shepherd’s argument there not only does not beg the question against Hume, but in fact contains plausible objections to his argument, and a plausible alternative to his conclusion.

To see that, we first need a sense of Hume’s theory of mental representation in place so that we can better understand Shepherd’s rejection of that theory. To that end, the following section will present an interpretation of Hume’s theory. It is not intended as a deep or thorough reckoning with the literature on this topic, but rather as a prima facie plausible reading of Hume’s theory of mental representation, one that does find support in the secondary literature on that topic, and that can be used to make sense of Shepherd’s apparent objections to it.⁶ It is this last desideratum that is, perhaps, most important: to present a tenable reading of Hume that can make sense of Shepherd’s argument against him. That is, my goal in the following section will be less to get Hume right (although I think I do), than it is to get a version of Hume that plausibly represents the Hume to which Shepherd takes herself to object.

2. Hume’s Argument

Before turning to the details of Hume’s argument against the thesis that, ‘whatever has a beginning, has also a cause of existence’ (T 1.3.3.2; SBN 79), it will be helpful to begin earlier in the Treatise with the theses that Hume will use as premises in that argument.⁷ Many of Hume’s arguments depend crucially on his deployment of his theory of mental representation, which itself is closely related, but not identical, to his famous Copy Principle: that every simple idea is a copy of some simple impression. The closely-related theory of mental representation begins with the thesis that all simple ideas represent that from which they are copied.⁸ So, since all simple ideas are copies of some simple impressions, simple ideas represent their corresponding simple impressions. Of course, complex ideas also represent, but are often not copies of anything at all, so this account of representation requires emendation. It becomes relatively clear how Hume would emend it in his discussions of the origins of our complex ideas of space and time.

The table in front of me is alone sufficient by its view to give me the idea of extension. This idea, then, is borrow’d from, and represents some impression, which this moment appears to the senses. But my senses convey to me only the impressions of colour’d points, dispos’d in a certain manner. If the eye is sensible of any thing farther, I desire it may be pointed out to me. But if it be impossible to show any thing farther, we may conclude with certainty, that the idea of extension is nothing but a copy of these colour’d points, and of the manner of their appearance. (T 1.2.3.4; SBN 34, emphasis added)

Our complex idea of a spatial complex comes to represent the spatial complex that it does by being a collection of simple ideas of colored points arranged in a way that exactly resembles the arrangements of the spatial complex being represented. We represent the relation that some simple impressions stand in to one another by arranging simple representations of each of these impressions into the same relation. We represent a as being next to b by placing an idea of a next to an idea of b. The idea of a spatial complex is nothing more than a spatial complex of ideas.

Hume is clear that our representation of temporal complexes works in the same manner.

The idea of time being deriv’d from the succession of our perceptions of every kind, ideas as well as impressions, and impressions of reflection as well as of sensation, will afford us an instance of an abstract idea, which comprehends a still greater variety than that of space, and yet is represented in the fancy by some particular individual idea of a determinate quantity and quality. (T 1.2.3.6; SBN 34)

Our idea of time is ‘deriv’d from the succession of our perceptions’. Hume’s thought is that we represent two items as being related, now temporally, by placing them in a temporal relation to one another. For example, we represent one thing as happening before another by having a representation of the former followed by a representation of the latter. So, whereas we represent a spatially complex state of affairs by forming a kind of picture before our mind’s eye, we represent a temporally complex state of affairs by forming a kind of movie there.

In general, complex ideas represent the simple impressions that their component simple ideas represent as being arranged in the way that those component ideas are arranged in the complex. We can generalize this account of complex representation using the general schema:

‘x’R‘y’ represents xRy.

A representation of x and y as related to one another in way R consists of a representation of x related in the same way, R, to a representation of y.⁹ A representation of a spatial complex is a spatial complex of representations, a mental picture. A representation of a temporal complex is a temporal complex of representations, a mental movie. Etc. As I have indicated, I believe that Hume deploys this theory of mental representation many times throughout the Treatise as an essential step in many of his most important arguments. It will be worthwhile to consider one particularly relevant example, namely, his use of it in his argument for the thesis that we can have no idea of necessary connection.

‘Tis easy to observe, that in tracing this relation, the inference we draw from cause to effect, is not deriv’d merely from a survey of these particular objects, and from such a penetration into their essences as may discover the dependance of the one upon the other. There is no object, which implies the existence of any other if we consider these objects in themselves, and never look beyond the ideas which we form of them. Such an inference wou’d amount to [demonstrable] knowledge, and wou’d imply the absolute contradiction and impossibility of conceiving any thing different. But as all distinct ideas are separable, ’tis evident there can be no impossibility of that kind. When we pass from a present impression to the idea of any object, we might possibly have separated the idea from the impression, and have substituted any other idea in its room. (T 1.3.6.1; SBN 86–87)

Hume’s first premise here is that if ‘the inference we draw from cause to effect [were] deriv’d merely from a survey of these particular objects,’ it would ‘imply the absolute contradiction and impossibility of conceiving any thing different.’ To ‘derive’ the inference from cause to effect from a survey of objects would be for that inference to copy the relations of those objects to each other. If the copied relation is to be a necessary connection, then to be a copy of it, the ideas representing each of the relata would likewise have to be necessarily connected to each other. That is, to represent a necessary connection between distinct objects, would require a necessary connection between the distinct ideas representing these. To plug these values into the above schema:

(‘x’ necessarily connected to ‘y’) represents (x necessarily connected to y).

To represent two objects as necessarily connected, the idea of each of the objects represented would themselves have to be necessarily connected. As Hume points out, though, no two distinct ideas are necessarily connected to each other. We can always entertain the idea of the cause without also thinking of the effect, and vice versa. Thus, since the representing simple ideas are not necessarily connected, the complex idea that they constitute cannot represent a necessary connection.

Of course, in addition to rejecting Hume’s thesis concerning the beginning of existence, Shepherd will also reject his claims here about necessary connection. Specifically, Shepherd holds that objects do imply the existence of other objects (e.g., ERCE 49), that an inference from one to the other does amount to demonstrable knowledge (e.g., ERCE 92–93), and that this does imply the absolute contradiction and impossibility of conceiving anything different (e.g., ERCE 55). What we are after here, though, is her grounds for doing so, and what I want to suggest is that the common root of these disagreements (and more besides) between Hume and Shepherd is their radically different theories of mental representation.

To demonstrate that last claim, we can now turn to the very specific use to which Hume puts the example that we have seen Shepherd cite from 1.3.3 ‘Why a cause is always necessary.’ Again, Hume leverages that example to refute the thesis that, ‘Whatever has a beginning has also a cause of existence’ (T 1.3.3.2; SBN 79). What we can see now is that that argument, like the one against the idea of necessary connection, also relies on Hume’s theory of mental representation.

But here is an argument, which proves at once, that the foregoing proposition is neither intuitively nor demonstrably certain. We can never demonstrate the necessity of a cause to every new existence, or new modification of existence, without shewing at the same time the impossibility there is, that any thing can ever begin to exist without some productive principle; and where the latter proposition cannot be prov’d, we must despair of ever being able to prove the former. Now that the latter proposition is utterly incapable of a demonstrative proof, we may satisfy ourselves by considering, that as all distinct ideas are separable from each other, and as the ideas of cause and effect are evidently distinct, ’twill be easy for us to conceive any object to be non-existent this moment, and existent the next, without conjoining to it the distinct idea of a cause or productive principle. (T 1.3.3.3; SBN 79–80, emphasis added)

Notice that Hume begins this argument by specifying that his target is those who would claim that we can demonstrate that every beginning of existence must have a cause. As we saw above with the case of necessary connection, the idea here is that to demonstrate that every beginning has a cause would imply that the opposite claim—that some beginnings do not have a cause—is a contradiction. Hume’s proof, then, takes the form of demonstrating that this latter claim is not, in fact, a contradiction. He does so by doing nothing more than presenting the example that Shepherd later cites. That is, he prompts us to notice that for any object of which can form an idea, we can also imagine that object being non-existent one moment, and then existing the next, without also having to imagine anything else.

Notice, though, that that imaginative exercise only shows that the thesis at hand is not a contradiction if there is a strong connection between what we can imagine and what we can represent. As Hume frames his argument, it is intended to show that the proposition that ‘whatever has a beginning has also a cause of existence’ cannot be given a demonstrative proof. What would such a proof look like if it could be given? Well, it would show that there is something in the very idea of a beginning that implies that such a thing must have a cause. That is, it would show that the idea of ‘beginning’ and ‘cause’ are necessarily connected. Or, to put it into the schema from above, it would be a representation of the form,

‘beginning’ necessarily connected to ‘cause’.

Thus, the way that Hume’s example operates is precisely to undermine the suggestion that we can form any such representation. Our ability to imagine a beginning without also imagining a cause shows that the ideas of ‘beginning’ and ‘cause’ are not, in fact, necessarily connected. Thus, the above schema is not instantiated, and so it cannot be demonstrated that whatever has a beginning also has a cause of existence. Hume’s argument here, as in the more straightforward case of the idea of necessary connection (and many others) is an application of his theory of mental representation.

3. Shepherd’s Rejection of Hume’s Conclusion

Now, back to Shepherd. In light of the way that we have seen Hume’s argument operate, what is so striking about Shepherd’s engagement with it is that she specifically deploys the very same example that Hume does, concedes to him that the example succeeds, but denies that Hume’s conclusion follows from that fact. Here is that text again.

Let the object which we suppose to begin its existence of itself be imagined, abstracted from the nature of all objects we are acquainted with, saving in its capacity for existence; let us suppose it to be no effect; there shall be no prevening circumstances whatever that affect it, nor any existence in the universe: let it be so; let there be nought but a blank; and a mass of whatsoever can be supposed not to require a cause START FORTH into existence, and make the first breach on the wide nonentity around. (ERCE 34–35)

Here we see Shepherd absolutely belabor the point that she can concede to Hume that we can imagine an object without also imagining what precedes it, and even that we can imagine an object coming to exist in an utter blank. (By my count she says as much six times in that single sentence.) She goes on, however, to ask,

Now, what is this starting forth, beginning, coming into existence, but an action, which is a quality of an object not yet in being, and so not possible to have its qualities determined, nevertheless exhibiting its qualities? (ERCE 35)

Shepherd accepts that we can imagine an object beginning to exist without also imagining its cause of existence, but then immediately (in the very same sentence) wonders what such a beginning could be other than an action of an object not yet in being, etc. That Shepherd accepts that we can imagine such a thing, and yet also rejects that this suffices to demonstrate that the proposition, ‘whatever has a beginning has also a cause of existence’, cannot be proven implies that Shepherd also rejects Hume’s theory of mental representation. That is, the dialectic to this point proceeds as follows. Hume proposes his theory of mental representation: that a complex idea represents the objects of its simple components as being arranged as those simples are in that complex. Hume combines this theory with the fact that we can imagine an object beginning to exist without also imagining that beginning to have a cause, to conclude that we cannot represent that the beginning of an object must have a cause. Shepherd agrees that we can imagine an object beginning without also imagining that beginning to have a cause, but rejects Hume’s conclusion that this shows that we cannot represent that the beginning of an object must have a cause. On what grounds might she do so? My proposal is that she rejects Hume’s theory of mental representation. That is, Shepherd rejects the very close tie that Hume supposes there is between imagination and representation.

As further evidence for this understanding of Shepherd’s position, notice that the very next point that she makes after rejecting Hume’s use of this example, is to offer an alternative account of how that example operates, and to use that account to derive precisely the opposite of Hume’s conclusion: that the idea of a beginning without a cause contains a contradiction.

But, my adversary allows that, no existence being supposed previously in the universe, existence, in order to be, must begin to be, and that the notion of beginning an activity (the being that begins is not supposed yet in existence), involves a contradiction in terms; then this beginning to exist cannot appear but as a capacity some nature hath to alter the presupposed nonentity, and to act for itself, whilst itself is not in being.—The original assumption may deny, as much as it pleases, all cause of existence; but, whilst in its very idea, the commencement of existence is an effect predicated of some supposed cause, (because the quality of an object which must be in existence to possess it,) we must conclude that there is no object which begins to exist, but must owe its existence to some cause. (ERCE 35–36, emphasis mine)

Notice that Shepherd’s argument against Hume is that ‘in its very idea’ the commencement of existence is an effect predicated of some supposed cause, and therefore, it is ‘a contradiction in terms’ to suppose that an object begins to exist without a cause. For the very idea of the commencement of existence to be an effect predicated of some cause, given that Shepherd accepts that we can imagine a beginning of existence without imagining its cause, it must be that the idea of a commencement of existence is something over and above whatever it is that we imagine when considering such an example.¹⁰ For Hume, however, it is not. For Hume, the idea of a beginning just is the imagining of a beginning. Shepherd rejects that account of mental representation, suggests how an alternative theory should account for the same example, and is thus not beholden to the conclusion that Hume derives from his own theory.

Of course, one might object on Hume’s behalf that while Shepherd clearly does reject Hume’s theory of mental representation, and the conclusion that he derives from it, she nonetheless fails to cite the reasons that ground that rejection. That objection would be hold more water if either (a) Hume’s theory was particularly plausible,¹¹ or (b) Hume himself had given reasons in support of it.¹² Nevertheless, Shepherd does at least point out that the targets of Hume’s argument themselves would not have accepted his theory of mental representation, and for that reason did not bother to address themselves to his potential objection.

It is undoubtedly true, that these authors [Dr. Clarke and Mr. Locke] assumed that which was in question; namely, that every existence must have a cause; but, as every thing not yet in existence, to exist at all, must begin, and the beginning of any thing must always be supposed, by the nature of the action, to be a quality of something in existence, which existence is yet DENIED by the statement of the question, these philosophers felt the involved absurdity so great, that they passed over the first question as too ridiculous, probably, to consider formally; then showed, that the mind of man was forced to look upon all things which begin to exist as dependent QUALITIES; and thus, that an object could neither depend upon itself for existence, not yet upon nothing. (ERCE 37)

If Shepherd is right that Clarke and Locke understood our representation of beginning in a way that presupposes the existence of an object that undertakes this activity, then she is on firm footing in rejecting Hume’s argument to the contrary as begging the question against them (so long as she does not find in Hume an argument against their account of representation).

This dialectic comes to a head in the next paragraph where Shepherd returns to the limiting role that Hume assigns to the imagination.

Let it be remembered, too, that although Mr. Hume inveighs against this method [of Dr. Clarke and Mr. Locke] as sophistical, by conceiving it begs the question, yet his own argument, the whole way, consists in the possibility of imagining an effect ‘non-existent this minute,’ and ‘existing the next;’ and does not himself consider any other ‘sort of being’ possible; and has no other way of supporting his own notion of the beginning of existence by itself, except under the idea of an effect in suspense; which is still a relative term, and begs the question for the necessity of its correlative, i.e. its cause, just as much as he asserts his adversaries do, whom he declares to be illogical reasoners. (ERCE 37–38)

Shepherd reiterates her two related objections to Hume here. The first is that his entire argument ‘consists in the possibility of imagining an effect “non-existent this minute,” and “existing the next;” and he does not himself consider any other “sort of being” possible.’ That is, Hume’s entire argument consists of his citing our ability to imagine a beginning without imagining its cause. If, however, one is not convinced that the limits of our imagination are the limits of what we can represent, as Shepherd is clearly not, then this example is insufficient to demonstrate Hume’s conclusion. Shepherd’s second objection is that once one adopts her more plausible account of what we represent when we represent objects, or even just returns to the conception employed by Hume’s own opponents, the contradiction that Hume could not find using his own theory of mental representation is obvious: an effect is an effect of a cause, and so cannot exist without it.

In fact, Shepherd lays the foundation for this conclusion—that it is only by illicitly replacing the ordinary meanings of the words ‘cause’ and ‘effect’ with those in accordance with his own theory of mental representation that Hume’s argument can so much as get off the ground—before she presents her argument at all.¹³

But before examining this notion … it will be necessary to render the expressions in which it is conveyed more intelligible. This can in no way be done so long as the definition of the word effect presupposed a cause, for the supposition of the objection lies in its being possible for effects to be held in suspense: but in order that this should be possible, the meaning of the word effect must be altered.

Then, if the ideas are altered that lie under the term, according as the varied occasion seems to require, there can be no philosophy; and it never can be insisted on, that the effects, which are supposed to be conjoined with their causes at one period of time … can again, upon another occasion, not be effects. (ERCE 30–31)

‘Cause’ and ‘effect’ as they are ordinarily used are co-defined, and it is a straightforward contradiction in terms to suppose that a cause qua cause could exist without an effect, or vice versa. A cause is a cause of an effect, and an effect is the effect of some cause. It is only by changing the meaning of those terms that Hume’s argument does not fail immediately, but once one has noticed that this change of meaning has occurred, it is important to take account of the consequences of the change.

To make, therefore, anything like a rational meaning in this sentence of Mr. Hume’s, nothing more can be intended by it, than that we should imagine, those existences which we always observe conjoined with others in such a manner, that they appear to be their effects, properties, or qualities, to owe them no real existence or dependence; and therefore capable of being independent objects, and of beginning their own existence. (ERCE 31–32)

In changing the meaning of ‘cause’ and ‘effect’ in such a way that to claim that an effect can exist uncaused is not a contradiction in terms, Hume shifts the meaning of these terms from referring to actual causes and effects to referring to merely apparent causes and effects. Correspondingly, though, it is no surprise that he is able to deduce that merely apparent causes and effects can be coherently imagined as being independent of one another. Once one stipulates that these objects might not be causally related, no examination of their intrinsic properties or other relations to one another should be expected to yield a demonstration that they are, in fact, causally related.

Furthermore, this objection of Shepherd’s to Hume’s procedure is supported by aspects of her own philosophical system that run deeper than what she takes to be Hume’s illicit linguistic machinations. To see this, we need examine Shepherd’s own, more plausible, account of the representation of cause of and effect, which I mentioned a few moments ago. Unfortunately, no such account is presented at this point in the text, nor in any detail in the remainder of ERCE. Shepherd appears content there to have shown that Hume begs the question against Locke and Clarke, and that she is therefore licensed to abandon his conclusion and pursue her own account of what our idea of cause and effect really is. She does, however, give a more detailed account elsewhere, namely in EPEU.¹⁴ To fully understand her argument against Hume’s analysis of cause and effect, then, it is worth delving into that text at least a bit.

We can begin that exploration with a passage, like the one regarding the beginning of existence above, that appears to be specifically designed to engage Hume’s account of mental representation on its own terms. This time Shepherd adopts Hume’s example of our idea of a table, which he casts as being nothing more than what his senses convey to him: ‘only the impressions of colour’d points, dispos’d in a certain manner.’ Here is Shepherd’s account of the representation of the same table.

Now the truth is, that no real table is formed, no image of a table is formed, unless the whole united mass of the unknown objects in nature exterior to, and independant of the instruments of sense, (not yet worthy of the name of ‘TABLE,’) unite with the mechanical action of these, and by their means with the sentient principle, in order to create such an union that object which alone can properly be termed ‘TABLE.’ Yet after experience, the outward OBJECTS, the CONTINUALLY EXISTING PARTS of the whole causes necessary to the creation of a table, must be named by the name by which the whole is named; for there is no other name whereby they can be called, nor any other ideas by which the memory of them can be introduced into the mind, save by the appearance of ‘the faint images of those sensible qualities’ which their presence originally created. (EPEU 148–49)

Shepherd’s account of the representation of a table seems to be something like the following. First, draw a distinction between what we see, a table, and what we see of what we see, say, the facing surface of the top of the table. While ‘the faint images of [the] sensible qualities’ of the table in some sense represent just the sensible qualities that correspond to the seen parts of the table, say a rectangular patch of brown, what our idea of the table represents, is precisely that such sensible qualities are the result of our perspectival encounter with a table, which comprises many other possible and actual sensible qualities united by the idea of the table as their common cause. ‘Table’ is the name of a complete object, not just the name of certain of the sensible qualities that such an object excites in the mind of someone perceiving it.¹⁵

Thus, while Hume attempts to construct the idea of a table from just these faint images of sensible qualities, that is a deliberately impoverished approach, which Shepherd clearly rejects, and rejects for the very reason that Hume himself embraces!¹⁶ That is, Shepherd and Hume agree that one cannot construct the notion of the table as a continual existent locus of causal powers from the sensible qualities alone. Hume concludes that we have no such idea of the table. Shepherd concludes that we must have representative resources over and above those afforded by the table’s sensible qualities.

It is thus by a union of observation and reason, coalescing with the conscious use of the senses, that we are enabled justly to affirm, that ‘outward objects are perceived immediately by sense.’ (EPEU 57)

The table, qua table, is represented not by just the senses, but by a union of observation and reason. The faint images of the sensible qualities of the table, for example, its color, might be perceived by the senses alone, but the table as the object of one’s perception, can only be represented as such via a combination of senses and reason. The table as the object of one’s perception is represented as that which causes those faint images, and so the images themselves, which do not themselves contain the idea of a cause, must be supplemented by this contribution from reason. In fact, even that strong connection between these two faculties undersells Shepherd’s account. As Shepherd sees it, there is actually only a single faculty at work in perception, and ‘the senses’ and ‘reason’ are abstractions of different aspects of this faculty.¹⁷

Now I repeat there is one sense in which it may be said that objects are perceived immediately, as existing outwardly, by the senses. It is this; the conscious powers of the understanding, and the senses, are blended together in man; we are analysing them, but in nature they are united as intimately as are the prismatic colors in one uniform mass of light. (EPEU 67)

A representation of a table as a table, as existing outwardly by the senses, is the result of the faculty of perception, which itself already contains the understanding and the senses as parts ‘united as intimately as are the prismatic colors in one uniform mass of light.’

Hume thinks of his theory of mental representation as bringing philosophers ‘back to basics’ (cf. T 1.1.1.1n2; SBN 2). What we can think is derived from impressions, and impressions consist in nothing other than sensible qualities. As Shepherd sees it, though, it is Hume who has imported into his theory certain confounding theoretical commitments. Specifically, while perception can be analyzed in terms of the contributions of ‘the senses’ and ‘reason’, in its most basic form, it is a unity of these faculties. ‘The senses’ themselves are an abstraction from a more basic representational faculty, and Hume’s separating off this faculty and deeming it the only legitimate source of representational content is a perversion.

Now objects in our conscious apprehensions … are not only blue or red, sweet or sour, hard or soft, beautiful or ugly, warm or cold, loud or low; but the ideas of their causes are included in their names as continually existing, and that even when the organs of sense are shut. (EPEU 135)

The name ‘table’ stands for a certain object, which is for Shepherd, first and foremost, a certain locus of causal powers, including but not limited to the power to cause certain sensible qualities in a perceiver.¹⁸ In very rough outline, that is Shepherd’s account of the mental representation of objects, including her account of where Hume (and Berkeley and others) go wrong. It might have made her case against Hume in ERCE clearer if she had included it there, but as I pointed out earlier, this account is not strictly speaking dialectically necessary there, because all she needs there is grounds to reject Hume’s conclusion that there can be a beginning of existence without a cause. She does that by showing that Hume’s argument for that conclusion is question begging, and does not also need, at that moment in that context, to offer a detailed alternative to what she more-or-less rightly takes to be his implausible and unsupported theory of mental representation. That she presents such an alternative elsewhere (in a book that was originally intended as an extension of ERCE) is so much dialectical and philosophical icing on the cake.

4. The Status of Shepherd’s Argument

By way of conclusion, I would like to revisit the quick-and-dirty reconstruction of Shepherd’s positive argument for the conclusion that an object cannot come into existence uncaused that I presented earlier. Here is that reconstruction again.

1. Beginning, or coming into existence, is an action.

2. An action is a quality of an object.

3. If an object can come into existence uncaused, then its beginning can only be a quality of that very object itself and no other.

4. If an object can come into existence uncaused, then its beginning is a quality of an object not yet in existence. (1, 2, 3)

5. An object not yet in existence cannot have its qualities determined.

6. Thus, a non-existent object must both have qualities (the quality of beginning to exist) and not have qualities (because it does not exist). (4, 5)

7. Therefore, an object cannot come into existence uncaused. (3, 6)

Earlier I noted that that there appears to be a great deal in this argument to which Hume could object. I also noted that while some of these steps could receive support from Shepherd’s positive accounts of objects from sections farther along in ERCE, such resources appear to be disallowed at this point in the text, insofar as Shepherd indicates that it is her conclusion here that will provide the foundation for those later theses, not the other way around. What we can now see is that the dialectic surrounding this argument is more subtle than it first appears. While Hume can raise objections to some of Shepherd’s premises here, if those objections depend on his theory of mental representation, and if the foregoing account of Shepherd’s rejection of that theory is roughly correct, then Shepherd is free to employ those premises, despite these objections, because she takes herself to have good reason to reject their underpinnings.

For example, consider 1 and 2: the beginning or coming into existence is an action or quality of some object. On what grounds might Hume object to that premise? Well, one such ground might be that we can imagine a beginning or coming into existence without also imagining ‘an object’ of which that beginning is an action. Suppose, however, that Shepherd is justified in rejecting Hume’s imagistic theory of mental representation, and also justified in replacing it with her own conception of representation according to which Hume’s ‘impressions’ and ‘ideas’ are mere abstractions from a more fundamental representational faculty that represents the world as consisting of objects existing in space, persisting through time, and necessarily connected to one another and to the perceiver. In that case, Shepherd would be further justified in holding that Hume’s objection rests on illicit grounds, and that the notions of a ‘beginning’ or an ‘action’ or a ‘quality’ are themselves essential components of the idea of an object, which object is fundamentally conceived of as a locus of necessary causal relations. The details of such a move, or corresponding moves that might be made with respect to her other premises here, must be left for another occasion, but notice more broadly that if these kinds of moves are available to Shepherd, then her argument is in better dialectical standing than scholars have previously recognized.

So, the key to understanding Shepherd’s positive argument here is to understand that dialectical context. Shepherd takes Hume’s argument to beg to the question against his own opponents insofar as it rests on his undefended and implausible theory of mental representation. Having rejected that theory of mental representation, Shepherd is free to make her argument relying on what she takes to be a more plausible account. Thus, Shepherd’s argument is not question begging: she rejects the assumptions on which Hume’s objections to it would be based and proceeds to argue unhindered by these assumptions. Her argument also has a claim to prima facie plausibility insofar as it is grounded in our ordinary usage of the words ‘cause’, ‘effect’, ‘beginning’, ‘action’, etc. and a comparatively plausible account of the content of the ideas that correspond to those words. Thus, while I will not render a verdict on the ultimate success of Shepherd’s argument, what I hope to have shown is that her argument is not the obvious failure that scholars have taken it to be.

Competing Interests

This paper was written with the support of a San Francisco State University Marcus Research Grant.