Shepherd on Causal Necessity and Human Agency

  • Louise Daoust orcid logo (Eckerd College)


Shepherd defends an account of the universe founded on two causal principles: that effects necessarily have causes, and that like causes have like effects. Folding mind into the class of natural phenomena governed by these principles, Shepherd naturalizes the mind, but in doing so she sets herself the challenge of explaining how, within a deterministic universe, agents can be necessary causes of their own actions. With special attention to Shepherd’s resistance to materialism and to any reduction of the mental, the paper argues that we can read Shepherd as leveraging her original theory of causation to develop a distinctive compatibilist view of the psychology of intention, one that makes agents the necessary causal sources of their own actions.

Keywords: Mary Shepherd, causal necessity, agency, materialism, compatibilism

How to Cite:

Daoust, L., (2022) “Shepherd on Causal Necessity and Human Agency”, Journal of Modern Philosophy 4: 15. doi:



Published on
28 Oct 2022
Peer Reviewed

The most widely discussed aspect of Shepherd’s philosophical system is her unconventional view of cause and effect, a view she details in her 1824 work, An Essay Upon the Relation of Cause and Effect (ERCE). In that work, Shepherd denies Hume’s conjecture that the existence of cause and effect is a psychological conclusion that results from habit and custom. She develops an account according to which facts about the nature of cause and effect can reasonably be inferred on the basis of phenomenal experience. On her view, we can infer that the universe we inhabit is governed by strict causal relationships, and that the relations of cause and effect are foundational to our understanding of the natural world.

In the main essay in Shepherd’s 1827 work, Essays on the Perception of an External Universe (EPEU), Shepherd extends her account of cause and effect to offer a refutation of external world skepticism. In doing so, she situates minds within the causal framework described in ERCE. Minds, for Shepherd, are natural phenomena, just as are the external objects responsible for our sensations. And just as the independent, continuously existing, external objects causally responsible for our experiences are known by reason, the existence of mind is inferred from the sensations and relations among sensations of which we are conscious.

Shepherd discusses at length her views regarding how external objects cause changes in our experiences, but she says much less about the sense in which mind can work as a necessary causal force within the deterministic world she envisages, or how human actions relate to the broader causal view she defends. By interpreting mental phenomena as among the effects brought about by the initial uncaused cause (God), Shepherd finds continuity between mind and matter, naturalizing the mind. Yet, her account raises questions about how agents can be necessary causes of their own actions in a universe that is bound by causal necessity.

In what follows, I begin by sketching Shepherd’s unconventional view of cause and effect, and then turn to some of her more explicit commitments about the nature of mind and its relation to other causal phenomena. Read alongside one another, these two facets of her system—her understanding of the nature of mind and her view of cause and effect—seem to raise a challenge for Shepherd about the possibility of genuine agency. With special attention to Shepherd’s resistance to materialism and to any reduction of the mental, the paper argues that we can read Shepherd as leveraging her idiosyncratic theory of causation to develop a novel account of human agency. Agency for Shepherd is not merely unincumbered or unimpeded action, as it is on prominent classical compatibilist views of the early modern period. Instead, Shepherd pioneers an account of the psychology of intention that challenges the standard incompatibilist tenet that an agent cannot be the causal source of her own actions if determinism is true. In this sense, Shepherd sketches a compatiblism that is unusual and decidedly modern.1


Shepherd’s causal system aims to provide foundations for ‘scientific research, … practical knowledge, and … belief in a creating and presiding Deity’ (ERCE 194). Finding that Hume’s philosophical views about our epistemic relation with the world ‘lead directly to a skepticism of an atheistical tendency’ (ERCE 4), Shepherd targets Hume’s doctrine of the relation of cause and effect as the central culprit. According to Hume, the relation of cause and effect is not known by either experience or reason, and must be a psychological posit founded in custom and habit. When I have eaten bread in the past, for instance, that bread has nourished me. From this evidence, I reason to the conclusion that, in the future, bread will nourish me again. But in reasoning in this way about the future, I rely on the assumption that the future will be like the past. This assumption cannot come from reason, according to Hume, because reason can only take facts about the past and present and find correlations, for example, between bread and nourishment, correlations that may, for all we know, fail to hold in the future. The imagination has no trouble separating the idea of bread and the idea of nourishment.2 If we can conceive of one without the other, then we cannot know by reason that one entails the other. Nor can the assumption that things in the future will be like things in the past be based in experience, according to Hume, since in experience we encounter merely the present (the taste of the bread, the feeling of nourishment). As is well-known, Hume’s problem of induction challenged scientific inquiry and prediction, and much of the philosophy of science in ensuing centuries has taken the problem as central. In Shepherd’s day, it persisted as a highly contested issue.3

Shepherd claims that Hume is wrong to think we cannot know the necessity of a cause for every beginning of existence. Contra Hume, reason, understood as ‘the observation of the relation of our simple sensations’ (EPEU 19), yields the causal principle: everything which begins must have a cause. If we suppose that an object could begin its own existence (a claim Shepherd takes to be entailed by Hume’s account), then that beginning would be an action or a quality of that object, a quality that the object has only after it has come to exist. But by the principle of non-contradiction, an object cannot both exist and not exist: ‘All that experience has to do, is to show us, by what passes within ourselves, that there is a contradiction in the supposition of qualities beginning their own existence; and a contradiction is never admitted in the relation of any ideas that present themselves’ (ERCE 143). Shepherd takes her causal principle, that everything which begins must have a cause, to be a necessary truth.4 From this principle, she professes to derive another necessary truth: the causal likeness principle, according to which similar causes must necessarily produce similar effects. When sourness results from eating a slice of bread when in the past slices from the same loaf have always been sweet, we infer that the causes at play in the current scenario (in which the bread tastes sour) must be different from the causes that were at play in the earlier scenarios (in which the bread tasted sweet). By Shepherd’s first causal principle, we know that the new quality (sourness) had to be caused, for it would be impossible for a difference to ‘begin of itself,’ (ERCE 101) uncaused. Shepherd appeals to analogies to make her understanding of the causal likeness principle precise: ‘If one added to one, bear out the result two, once; it must ever do so; and if a certain proportion of blue and yellow particles, form a mixture termed green, once; green in like manner shall ever thence result’ (ERCE 101). If an unexpected or unpredicted property emerges, we can reason to the certain conclusion that some additional cause must have been involved.

Equipped with these principles, Shepherd develops a complex set of claims about the history and nature of the universe. Because cause and effect are connected necessarily, the sequence of events unfolding in the universe is bound by necessity. There is a dependability to the order of nature, such that we can begin to describe it and make predictions.5 Causes come together and interfere with one another; and, assuming no additional causes intervene, they necessarily cause certain effects.6 Moreover, because causes are necessary to their effects, causes and their effects should be understood as synchronous. As Shepherd puts it, ‘antecedency’ and ‘subsequency’, concepts central to Hume’s account of cause and effect, are

immaterial to the proper definition of Cause and Effect; on the contrary, although an object, in order to act as a Cause, must be in Being antecedently to such action; yet when it acts as a Cause, its Effects are synchronous with that action, and are included in it: which a close inspection into the nature of cause will prove. For effects are no more than the new qualities, of newly formed objects. Each conjunction of bodies, (now separately in existence, and of certain defined qualities,) produces upon their union those new natures, whose qualities must necessarily be in, and with them, in the very moment of their formation. (ERCE 49–50)

An effect is thus best understood as a necessary change in qualities resulting from the interactions of the qualities of different objects already in existence. The beginning of each new quality ‘is but a change of that which is already in existence’ (EPEU 170). Correspondingly, a cause is ‘such action of an object, as shall enable it, in conjunction with another, to form a new nature, capable of exhibiting qualities varying from those of either of the objects unconjoined. This is really to be a producer of a new being’ (ERCE 63).

On this account, all changes are necessary effects resulting from the interactions of existing causes. Shepherd has God—the one uncaused cause—initiating the universe. She contends that the universe, with all its variations, and powers of life and motion, must have ‘come out’ (ERCE 97) from God in such a way as to establish this causal order. Because nothing begins its own existence, minds and matter may be ‘considered as having existed eternally, coming forth from him [God], living in him, and supported by him; whilst an analogous state of being must be expected to continue eternally, in like manner’ (ERCE 98). Consequently, the present and future are ‘included in the past’ (ERCE 142). On the basis of our body of experiences interacting with the world, we gain practical knowledge of cause and effect, eventually becoming able to make predictions about complicated events, such as weather patterns. Further scientific study of this sort will engender further understanding of ourselves and our environments, and may even allow for us to control and ‘imitate nature, better than we have hitherto done’ (EPEU 308). This deterministic view of the universe thus purports to ground scientific knowledge, and our confidence in predictions about the future.7

Much of Shepherd’s EPEU is dedicated to an investigation into the existence and nature of the external objects that act as causes for our sensations. Applying the causal principle and the causal likeness principle, Shepherd views phenomenal experience, and the complex relations among our sensations, as provoking in us inferences to the existence of continuous, external, and independent causes for our sensations. From the patterns we observe phenomenologically, we can tell that necessarily there must be external causes isomorphic to the sensations we experience, that ‘certain definite varieties of mind, must be occasioned by equal varieties in external nature’ (EPEU 307). An external object can only be contemplated under the form of the union of the sensible impression it causes and the idea of it as a cause (EPEU 20–21). So, to be an external object is to be a capacity to cause certain mental effects. It is to these causes that we refer when we use words such as ‘table’ or ‘apple’ in ordinary discourse. Shepherd is clear that we can know almost nothing about the natures of these external objects or capacities to cause sensations in us, save for what can be inferred from the sensations and patterns of sensations they cause: that they are various, that they are independent and continuous, that they exist, and that they have identities that can persist across time.8

Shepherd argues in a parallel way for the existence of mind as an external, continuous capacity that acts as a cause for sensations:

if it should be asked, whence the mind knows itself to be exterior to each sensation in particular, and continued in its existence, I answer from the same principle which enables it to judge other things as exterior to itself; namely, from that perception of the understanding which forces upon it the conclusion, that because each sensation in its turn vanishes, and new changes spring up, so there must necessarily be some continued existence the subject matter of these changes; otherwise, ‘each change would begin of itself.’ (EPEU 56)

All sensations depend upon mind, and the causes of particular sensations cannot be the same as those that cause the general power of sensation (EPEU 153). The general power of sensation, mind, is a theoretical posit according to Shepherd, just as external objects are necessary theoretical posits (EPEU 98). This is because sensations are interrupted; they are effects. The underlying mental cause of sensations must be ‘uninterrupted; and such an uninterrupted cause as is equal to keep up the life of the body, or mass deemed our body, and to unite it under that form with the powers of memory and sense’ (EPEU 154). Shepherd acknowledges that we might worry that the cause that unites impressions of the present sense and ideas of memory might itself be interrupted, and thus that it could not exist as continuous in the way we assume our selves do. However, she thinks that we can avoid this worry by reasoning to what must be an uninterrupted cause removed by some (unknown) degree:

Should it be objected that the causes for such an union [a union of the ideas of memory and the impressions of present sense] might be interrupted; then as these would ‘begin their existences,’ and would only be effects, the mind would go backwards till it reposed in some uninterrupted cause, and would consider such, and such only, as an independent capacity in nature, fitted to excite the union of memory with the present sense, and as the complicate being self; which when conscious, could take notice of its existence, and when unconscious, (as in sound sleep) could exist independently of its own observation. (EPEU 154–55)

Mind—whether a human’s, a worm’s (ERCE 174), a bird’s (EPEU 360), or a barnacle’s (EPEU 377)—is an uninterrupted existence, an ‘inward sentient principle’ (EPEU 15), that is, a ‘capacity or cause, for sensation in general’ (EPEU 155).9 And as with ordinary external objects, a mind cannot change ‘unless interfered with’ (EPEU 43).

Each of us has a sense of the self or mind, a sense of continuous existence, which, when analyzed, is known to be a union of the ideas of memory with the present impressions of sense (EPEU 154). Still, it is worth underscoring that mind, like the external objects that act as causes for our sensations, is known by reason.10 Reason’s inferences yield ideas about the mind, and it is in terms of these that we are able to represent the mind. And, again as with the external objects that act as causes for our sensations, not much can be known about the mind independently of these representations. As Shepherd writes, ‘the real essences of matter and mind we know not; we only know our sensations, as real beings, very essences: these are the very things themselves. We know of other things which must ‘needs exist’ by our sensations, but cannot conceive the nature of any essence not in our experience’ (EPEU 244). Because various effects should have proportional causes, ‘there must be some extraneous reason for sentiency, beyond what is absolutely necessary for mere insentient extension’ (EPEU 158). We can know that mind exists, and we can know it is a simple capacity to cause sensations different from causes that are insentient. But we know very little beyond this.

With Shepherd’s unconventional views about cause and effect in view, we can begin to see how Shepherd ‘naturalizes’ the mind by positioning it within the set of natural phenomena governed by the necessary unfolding of causal processes.11 In doing so, however, Shepherd sets herself the challenge of explaining how agents can be necessary causes of their own actions. Typically, when philosophers discuss the problem of causal determinism, they mean to raise questions about whether minds are the causal sources of intentional behaviors, or whether the mental has any real causal power. In the case of Shepherd’s philosophical system, minds clearly do have causal force—Shepherd insists that minds are simple capacities to cause sensations, causes that, like any causes on Shepherd’s view, necessitate the effects they are involved in bringing about. How, then, does Shepherd understand the causal relationship between mind and body? Are mental causes reducible to physical causes, and if not, in what sense can mental and physical causes be thought to mix to produce new effects? Can Shepherd endorse a notion of agency that is at once satisfying and also compatible with the tenet that actions are determined by prior causal histories? In the following section, I explore passages that bear on Shepherd’s psychology of intention, and I argue that Shepherd leverages her rejection of materialism and her innovative causal account to recommend a non-reductive view of intention according to which mind operates as a necessary causal source of action within a determined universe.


It is clear that Shepherd takes intentional action seriously,12 and yet intentional action is by no means a kind of uncaused intervention on Shepherd’s view. On the contrary, in the fourth of the shorter essays published in the EPEU, Shepherd reemphasizes the significance of having analyzed body and mind in similar ways:

there must always be a natural necessity in the interchange of qualities according to their original formation; so that the contradiction would be to imagine them otherwise than they are, when once experience informs us of their appearances: therefore, muscular action, nervous influence, and in short, all actions of the human frame; all the actions of nature, are to be explained after one and the same method, namely, by conceiving cause and effect as synchronous in each step of the series of actions which take place, from the first junction or mutual affection of the external senses, with the particles of external bodies, to the last sensation of animated consciousness. (EPEU 311–12)

Once we have learned from experience that some ‘interchange of qualities’ is manifest in the world, we can know that this interchange was necessary, given ‘the original formulation’, that is, the ordered development of the universe from the original uncaused cause, God. All behaviors can be explained using the same method: in the causal universe Shepherd limns, agential action will always be the effect of causes intermixing in ordered and, in principle, predictable ways. Human action is conceived of as a mere aspect of the ‘progress arising from successive changes’ (EPEU 240).

Like Hume, Shepherd thinks that it’s possible to take behaviors as signs of reliable underlying natures in people, just as we take apparent qualities of objects to be dependable guides to their underlying natures. As Shepherd puts it in ERCE,

had I a friend whose absence might suggest a dread, lest the powers of his friendship had been weakened; if upon his return I observed the same sensible manifestations of regard as heretofore I should have very reasonable ground to judge, that they were the symptoms of a heart, as true to me as ever, whose faith was always found to shew itself in similar demonstrations of kindness. (125–26)

Shepherd suggests here that patterns in human behavior are subject to empirical study just as are other facets of the natural world. In a more challenging passage, Shepherd comments on the conditions under which humans develop. Action in humans, Shepherd writes,

is as multifarious as food, medicine, and climate; the circulation of blood, the passions, the habits of education, and the notions of individuals, can render it. They are wrong, therefore, who, ignorantly taking no notice of these things, expect the human will, to be in all circumstances equal to self-command. Men make excuse for their actions in dreams and insanity, saying, the essences of things are then different; but never consider, that every degree and variety of their state of mind depends upon analogous laws and causes, which wisdom acting in time might alter with advantage, but which afterwards may lie beyond any human power to ameliorate. (EPEU 265)13

Many factors—cultural, environmental, physiological, psychological, emotional, educational, and cognitive—can impact which actions we undertake, according to Shepherd. Because of the various roles of these causal factors in generating dispositions to act, we are wrong to think of the human will as ‘equal to self-command’, as wielding uncaused control over which actions we perform. Instead, we must be vigilant against factors that will degrade us over time, and cultivate those that will improve our characters, for example by obtaining a good education. Dispositions to act are controlled by prior causal circumstances.

Shepherd endorses intentional self-improvement,14 so she takes intentions to be compatible with her deterministic framework. But how is this supposed to work? In response to classical forms of compatibilism, incompatibilists argued that in a deterministic universe, there simply cannot be a satisfying sense in which agents are the necessary causal sources of their own actions. For Shepherd, though, an alternative form of compatibilism is viable. Shepherd rejects a materialist metaphysics, taking mental causes to be explanatorily irreducible, and then appeals to her idiosyncratic account of causation to make minds causal sources of action.

The irreducibility of the mental is apparent in how Shepherd discusses intentions and the causal processes underlying those intentions. In arguing for the presence of final causes in the universe in short essay 9, an essay clearly influenced by Paley, Shepherd asks us to reason by analogy to what must be the designs of God by beginning from our own experiences with final causes in the generation of human-made artifacts. There is evidence of human intention all around us, for physical causes alone cannot adequately explain artifacts such as a clock, steam engine, or sculpture. In each of these objects there is the ‘appearance of contrivance’—something that could not result from ‘a chance coincidence of effects, arising out of a determination of motion that had no end in view’ (EPEU 346–47). It is too unlikely that something as contrived as a steam engine could exist without there having been an intention to bring it about; thus, we need to refer to mental causes in explaining the artifacts around us.

Shepherd uses the example of a well-built ship in full sail to illustrate her claim that mental causes are necessary over and above physical causes. Upon seeing such a ship, a materialist might attempt to treat each part of it as the accidental culmination of a long chain of necessary causal processes, the ‘necessary physical, mechanical actions of matter’. Indeed, we might even imagine attempting to trace these causal chains all the way back to the raw materials out of which the ship was built, along with the ‘other actions of matter, viz. of the muscles, the nerves, and the brains of the human beings concerned in the arrangement.’ Given even the most detailed physical description imaginable, however, we would still lack a full explanation of the phenomenon, according to Shepherd: ‘we know by experience, this will not explain the whole objects which have been in action’ (EPEU 348–49). In other words, appealing to the brain and muscle activity of the ship’s architect, or to any other mechanical or physical ‘actions of mere matter’ will never be sufficient to explain the design of the ship—the ‘mental quality of design’ (EPEU 353). What is required additionally for an adequate explanation of the ship is discussion of its necessary mental causes. No reductive physical account could be satisfying here. In recent accounts of the relationship between mental and physical phenomena, mental concepts such as those of will and intention are sometimes discussed as explanatory on a different level than are physical phenomena such as neural underpinnings. But, for Shepherd, mental and physical causes do not operate on different explanatory levels. Rather, mind and body mix to generate new qualities such as directed bodily actions, actions that can give rise to physical artifacts. The physical causes involved certainly are among the causes necessary to bring about some intended end, but alone they are not sufficient. Physical causes impact the mind, and the mind also affects the physical. Both are necessary if we want to account for the well-built ship.15

Intentions to bring about particular states of affairs necessitate their effects, on Shepherd’s view. This irreducibility of the mental allows us to distinguish Shepherd’s account of agency from earlier classical compatibilist views according to which agency amounts to unincumbered or unimpeded behavior. Shepherd insists that we must recognize mental causes just as we do physical causes, and that we cannot explain the artifacts around us without appealing to mental causes, no matter how complete of an account we might have of the physical causes involved.

Shepherd describes the mental causes involved in intentional behavior as follows: if one perceives an end or state of affairs one knows one could bring about through action, and then joins to that an intention to create or bring about that state of affairs, the result is an efficient cause of the direction of motion.16 When we perceive the utility of some action x (say, the action of kneading dough), that perception, and the intention to follow through, are united with the brain and other physiological systems, and through this union move the body to act (to knead the dough). It is ultimately the direction of motion that is ‘in our power’ (EPEU 347), but it is the will (the intention), rather than the perception of a future quality or state of affairs, that is the immediate cause of the direction of motion of the body (EPEU 353–54). The final cause, the perception of an attainable end, should be conceived of as the efficient cause for the will to act, which, in turn, is the efficient cause of or means for the direction of motion (EPEU 359). Shepherd tells us, moreover, that the whole cause made up of these various causes—(1) the perception of an attainable end; (2) the will to act to bring it about; and (3) the direction of motion upon matter—forms ‘one compound physical efficient cause’ (EPEU 360). But in what sense is this compound efficient cause of change a ‘physical’ cause, and how can this be consistent with Shepherd’s clear commitment to the irreducibility of mental causes?

It is easy to misread Shepherd as suggesting a kind of reducibility of the mental to the physical in these passages, and it is worth considering in further detail how Shepherd thinks of the relation between the initial mental phenomenon—the perception of an attainable end—and the proximate physical changes involved in the direction of motion, such as the actions of the brain, nerves, and muscles, especially because Shepherd takes herself to have succeeded in improving upon prior philosophical attempts to understand this relation, such as those made by Bacon and Newton (EPEU 358). Shepherd uses the notion of an intimate ‘unity’ to characterize the relation: ‘the union of sentient and insentient qualities is so intimate as to coalesce, and together to form the physical efficient cause of the beginning and direction of motion amidst the powers of nature’ (EPEU 388). Accordingly, the powers of mind are ‘one with the visible affections of matter, they inhere as one physical cause along with them’ (EPEU 349). The mind ‘interferes’ with the mechanical actions of the powers of matter, but in such a way that it is ‘perfectly one’ with them (EPEU 388). Thus, a final cause becomes an efficient physical cause for the beginning and direction of motion. It becomes ‘identical with those which are efficient’ (EPEU 359). In fact, the mental is not only ‘one’ with the physical causes with which it, or the will to act, interferes; it is one with the initial physical causes that interfere with the mind in order to generate as an effect the perception of an attainable end. In inventorying the physical factors at play in a bird building his nest, for instance, we must note not only those that are caused by the perception of the attainable end, but also those that are causes of that perception, including those that are most proximate, those that ‘co-exist with those affections of mind’ (EPEU 404). In an important sense, then, Shepherd takes ‘final’ to be ‘nothing more than a name for the compound set of physical efficient causes’ (EPEU 361).

It is Shepherd’s metaphysics that underpins and illuminates the sense of intimate and total unity between mental and physical causes that is supposed to account for intentional action. Recall that, for Shepherd, causes exist synchronously with their effects, and the existence of each new quality is but a change of that which is already in existence. In the context of discussing final causes, Shepherd reminds us that ‘every object would remain as it existed at any given moment unless it were interfered with; and an interference cannot be either before or after itself; but must be in and with the same moment of the change occasioned by it’ (EPEU 312n). In the case of mental causation, a mental cause, such as the will to pursue an attainable end, will be synchronous with its physical effect, the direction of motion. The direction of motion will be a quality that emerges through the interaction of its proximate causes, mental and physical, and those causes will exist alongside it—it will be a part of them. In this sense, we might talk of the causes as being ‘equal to’ the effect (EPEU 313). At each step of the development of the universe, cause and effect are synchronous, whether there is mental causation involved or not. Shepherd sees her theory of cause and effect as making plausible the tight connection she envisages between final and efficient causes, mental and physical phenomena. There is no mystery in these unions, for Shepherd, in part because there is no mystery in any union: ‘all things are united, and form one whole in their mutual interactions according to their natures’ (EPEU 406–7). Because body and mind can be so intimately united, even identified, on Shepherd’s view, we can see why Shepherd thinks that to address the body is to address the mind, that we must cultivate mind and body as a package (EPEU 264).17

When an agent forms an intention to act—an intention to raise a hand, for instance—that effect is, on Shepherd’s way of understanding things, one with the causes that mix to create it, including the mental causes. That intention can also become a new causal power, one that mixes with further causes to create the motion of raising a hand. In so far as something necessary for the agent’s action originates within the agent herself—in so far as the perception is an irreducible and necessary component of the mixture that results in the intention—an agent may be said to be a causal source of her own actions. Shepherd thus develops an account of intention that allows agents to be irreducible causal sources of their own actions within a deterministic framework.18 To further recommend Shepherd’s unusual account of causation, then, she has embedded within it a remarkably innovative compatibilist account of human agency.


A richer understanding of how final causes become efficient is essential to our understanding of how nature operates, according to Shepherd (EPEU 359). Of course, we can’t easily detect the exact ways in which the mind impacts the physical realm (EPEU 349); mental powers cannot be discovered by sense or instruments of detection like changes in matter can be. Prior to the existence of contrivance, mental powers are only known phenomenologically, by ‘experience of what passes within ourselves’ (EPEU 350). Were it not for this phenomenological proof of intention, available to each of us, ‘our modern atheists might deny its [the mind’s] perception of ends, and its direction of means, as final and efficient causes amidst the motions they witness’ (EPEU 388–89). Nor are we conscious of the physical components of the efficient cause of the direction of motion, causes that are, for instance, neurological (EPEU 406).19 Whatever happens to manifest as the ‘first sensible propellant’—the first physical, detectable change in the relevant causal sequence—mind may yet be the ‘final, i.e. the only efficient cause’ (EPEU 347) of the motion leading to some apparent contrivance.

The existence of mental causes in others can be posited, ‘by reason and analogy’ (EPEU 361). In such cases we must begin from a posteriori comparisons with objects known to be designed, and reason in terms of probabilities. If ‘reflection, determination of reason, or passion’ (EPEU 350) have interfered with material and accidental causes to yield some object, then a comparison between that object and other things that we know to be contrived should steer our conclusions correctly. That which reason ‘after examination admits to be the appearance of design’ is the only real ‘proof’ of human contrivance, on this view (EPEU 354–55). Doubt regarding our comparisons will always be possible (EPEU 174), especially because ‘to judge properly in many cases, whether intellect has been at work or not, requires extraordinary understanding,—higher faculties of mind than the abstract sciences stand in need of’ (EPEU 350–51).20 It is easy to be wrong about whether something was designed, and in cases in which intention was a cause, it will often be difficult to tell what exactly the design was. But, if an object is designed, then ‘no mechanical, or physical actions of mere matter will account for the mental quality of design’ (EPEU 353). Consequently, we will be able to reliably posit when mental powers have played an irreducible causal role in bringing about some physical effect.

More substantial doubts about the possibility of the mind’s causal power are likely to come from a history of dealing with views of mind-body interaction that begin from an assumption about two ‘essentially different natures’ (EPEU 310).21 Shepherd argues that because ‘the qualities of body and mind are equally unknown … then there appears no more contradiction to me, that they should thus act in, and with each other, than that any one event or object in nature should take place according to the condition of its essence’ (EPEU 311). For Shepherd, it is our acknowledgment of our own epistemic limitations regarding the universe that helps us to overcome the problem of mind-body interaction: if we are not able to know much about the nature of mind or body, then there is no special problem worth discussing when it comes to how mind and body interact.

Philosophical errors regarding the positing of false or insufficient causes are often the most dangerous, because these are the errors that are most liable to escape our notice:

the idleness of the mind which prosecutes with reluctance difficult researches into remote proofs; its impatience which eagerly grasps at the readiest solution of a doubt; and its pride, so prone to triumph indiscreetly at the glimpse of a discovery supposed to be complete; for ever occasion it to be guilty of that mode of sophistry scholastically termed non causa—pro causa. (ERCE 167)

Shepherd encourages us to embrace our considerable epistemic limitations in this context: ‘the whole of the matter is, I repeat, a mystery; an “unknown language” is not that in which to think, with much ease and satisfaction’ (EPEU 268). Recognizing our epistemic limitations regarding the universe and its underlying causal dynamics may give us pause, and encourage us to see ourselves, as Shepherd does, as one with that universe in a profound sense.


  1. ‘Classical compatibilism’ is a term that is meant to underscore commonalities across compatibilist views, beginning with figures in the early modern and modern periods, such as Thomas Hobbes, David Hume (though see Russell 2015), and John Stuart Mill (see Berofsky 2017 for further discussion). [^]
  2. ‘Objects have no discoverable connexion together, nor is it from any other principle, but custom operating on the imagination, that we can draw any inference from the appearance of one, to the existence of the other’ (Hume [1739] 2005: 1.3.8). [^]
  3. See, for instance, Stewart (1805) and Brown (1818). For discussion, see Boyle (2018: 5–6). For a twentieth century response to Hume’s problem of induction, see Goodman (1983). [^]
  4. For a thorough analysis of Shepherd’s disagreement with Hume, see Landy (2020). [^]
  5. See Martha Bolton’s (2011), where she notes that it follows from Shepherd’s two causal principles that it is ‘impossible that the course of nature should change’ (251). [^]
  6. Shepherd describes the interactions that occur between causes in a range of ways. She claims, for instance, that causes come together to form a ‘union’ (ERCE 50, 57), a ‘mutual mixture’ (ERCE 170), or a ‘junction’ (ERCE 171); she claims, also, that it is ‘not meant that qualities must always unite, but that they mutually affect each other’ (EPEU 313). [^]
  7. It is worth noting that Shepherd’s causal account resists a simplistic deterministic treatment. God isn’t necessitated by anything, on Shepherd’s view (God is uncaused), and Shepherd tells us that God can choose to mix with other causes to produce religious miracles (Shepherd points to Biblical miracles as examples of this having occurred in the past). There is more to say on this topic; for the purposes of this paper I intend my use of ‘determinism’ to be consistent with the possibility of divine interference. [^]
  8. Shepherd follows Berkeley in emphasizing that the causes of our sensations do not resemble those sensations in any straightforward sense. For instance, our sensations of extension cannot be said to resemble the real cause we call ‘extension,’ about which we know little. What can be known is gleaned from the proportional relations among ideas:

    we know not what extension unperceived is, although I am willing to concede a mite cannot be the same as the globe, not only with respect to that condition of being which, when exhibited upon the eye or touch, yields the notion of extension, but which, when subjected to calculation, manifests that in its unknown state, it must be liable to that variety, which when perceived, is called size or figure, and becomes altered in its dimensions (EPEU 165).

    Rejecting resemblance, Shepherd endorses a kind of isomorphic relation between sensations and the objects that are their causes. [^]
  9. For a more detailed overview of how Shepherd’s argument for a continuous mind is analogous to her argument for the continuous existences of external objects, see Boyle (2020). [^]
  10. Again, Shepherd defines reason as ‘the observation of the relation of our simple sensations’ (EPEU 19). [^]
  11. Shepherd is somewhat ambivalent about the term ‘law,’ in that she rejects the idea that a law is ‘an arbitrary rule which matter would observe without there being a necessity for it in any physical cause’ (EPEU 313n). She does, however, tell us that objects are governed by laws (ERCE 185), and describes her causal principles in a number of places as laws (EPEU 290, 329, 373). [^]
  12. Shepherd defines knowledge of human nature as ‘the penetration which enables us to discover the intentions that govern the motions of ourselves and others’ (EPEU 351). [^]
  13. Shepherd agrees with Hume in thinking that a range of factors matter for which mental states an agent ends up with. For Hume, we habitually infer that there is a necessary connection between psychological and bodily states, on one hand, and subsequent actions, on the other (see Hume [1739] 2005:–12). Shepherd likewise agrees with Hume that we sometimes struggle to predict what someone else will do (or what we ourselves will do) because we fail to understand what causes are at play. [^]
  14. Moral education, or ‘moral treatment’ as Shepherd calls it, should be understood to be entirely suitable whether or not the mind is material, as William Lawrence supposed that it is. (See ERCE 171–72 for Shepherd’s response to Lawrence on moral education.) [^]
  15. Shepherd’s focus on the relationship between final causes and explanations, here, is Aristotelian; for Aristotle, ‘causes are not ways in which we explain things, except derivatively, in virtue of the fact that they are ways in which some elements of the natural world explain others’ (Stein 2012: 705). [^]
  16. Shepherd refers to the perception of a state of affairs one knows one could bring about in terms of becoming a final cause. Becoming a final cause is thus to ‘perceive a future possible quality, capable of being gained by that means in our power, called the direction of motion’ (EPEU 347). Shepherd defines final causes in a number of closely related ways. Later in the same essay, she writes that a final cause ‘properly signifies the mental perception of an attainable end; the contemplation of a certain number of qualities, the determination of whose existence is known to be in the power of the efficient agent, by his voluntary direction of the motion of those already present to him’ (EPEU 360). [^]
  17. Despite the tight relationship between the mental and the physical on Shepherd’s view, Shepherd considers it possible that the mind could exist without body. Admitting that it is unclear whether we could have sensations without the neural underpinnings that ordinarily support them, she notes that the worm never anticipated turning into a butterfly, and we should not limit ourselves in what we are willing to acknowledge as a possibility (EPEU 158–59). [^]
  18. Some compatibilists would suggest that we can conceive of this kind of causal sourcehood as ultimate or original causal sourcehood. Such a compatibilist account of ultimate originating sourcehood would not require a deterministic break or initiation of a causal sequence (see McKenna 2008: esp. 198–99). [^]
  19. Shepherd refers to the synchronous physical causes of mental phenomena in terms of a ‘mysterious law’: ‘the mysterious law, or natural power which is a material property and executes the motion, is hidden from its [the mind’s] observation, although it should react upon it, whether by pain or pleasure, in each conceivable variety’ (EPEU 405–6). In theory, though, we might detect relevant brain activity (EPEU 348). [^]
  20. Shepherd thinks that in cases of ‘higher order’ designs, or ‘very involved operations’, some ‘meaner capacities’ will not be able to detect design. Disturbingly, Shepherd describes the reaction of an ‘Esquimeaux Indian’ who stares ‘with an undefined astonishment’ when presented with ‘master pieces, for instance, of music, sculpture, or painting’ (see EPEU 351). [^]
  21. Shepherd has in mind Descartes’s dualism, and the types of concerns raised by Elisabeth about how interaction could be possible between two different types of substances, one material and one immaterial (see Shapiro 2007). [^]


Thanks to Allauren Forbes, Michaela Manson, Charlotte Sabourin, and two anonymous referees at the Journal of Modern Philosophy for helpful feedback on versions of this essay. Thanks also to David Landy, Antonia LoLordo, and colleagues who participated in the TEMPO 2020 conference meeting.

Competing Interests

The author has no competing interests to declare.


1 Berofsky, Bernard. 2017. “Classical Compatibilism.” In The Routledge Companion to Free Will, edited by Kevin Timpe, Meghan Griffith, and Neil Levy, 41–51. New York: Routledge.

2 Bolton, Martha Brandt. 2011. “Causality and Causal Induction: The Necessitarian Theory of Lady Mary Shepherd.” In Causation and Modern Philosophy, edited by K. Allen and T. Stoneham, 242–61. New York: Routledge.

3 Boyle, Deborah. 2018. “Introduction.” In Lady Mary Shepherd: Selected Writings, edited by Deborah Boyle, 1–26. Exeter, UK: Imprint Academic.

4 Boyle, Deborah. 2020. “Mary Shepherd on Mind, Soul, or Self.” Journal of the History of Philosophy 58 (1): 93–112. DOI:

5 Brown, Thomas. 1818. Inquiry into the Relation of Cause and Effect. Edinburgh: Archibald Constable.

6 Goodman, Nelson. 1983. “The New Problem of Induction.” In Fact, Fiction, and Forecast, 4th ed., 59–84. Cambridge, MA: Harvard University Press.

7 Hume, David. A Treatise of Human Nature. (1739) 2005. Edited by David Fate Norton and Mary J. Norton. Oxford: Oxford University Press. DOI:

8 Landy, David. 2020. “Shepherd on Hume’s Argument for the Possibility of Uncaused Existence.” Journal of Modern Philosophy 2 (1): 1–14. DOI:

9 McKenna, Michael. 2008. “Ultimacy and Sweet Jane.” In Essays on Free Will and Moral Responsibility, edited by Nick Trakakis and Daniel Cohen, 186–208. Newcastle upon Tyne: Cambridge Scholars Publishing.

10 Russell, Paul. 2015. “Hume’s Lengthy Digression: Free Will in the Treatise.” In The Cambridge Companion to Hume’s Treatise, edited by Donald Ainslie and Annemarie Butler, 230–51. Cambridge: Cambridge University Press. DOI:

11 Shapiro, Lisa, ed. and trans. 2007. The Correspondence between Princess Elisabeth of Bohemia and René Descartes. Chicago: University of Chicago Press. DOI:

12 Shepherd, Mary. 1824. An Essay Upon the Relation of Cause and Effect, Controverting the Doctrine of Mr. Hume, Concerning the Nature of that Relation; with Observations upon the Opinions of Dr. Brown and Mr. Lawrence, Connected with the Same Subject. London: Printed for T. Hookham.

13 Shepherd, Mary. 1827. Essays on the Perception of an External Universe, and Other Subjects Connected with the Doctrine of Causation. London: John Hatchard and Son.

14 Stein, Nathanael. 2012. “Causation and Explanation in Aristotle.” Philosophy Compass 6 (10): 699–707. DOI:

15 Stewart, Dugald. 1805. A Short Statement of Some Important Facts, Relative to the Late Election of a Mathematical Professor in the University of Edinburgh. Edinburgh.