The moral philosophy John Locke outlines in his Essay concerning Human Understanding, according to which all moral concepts are “mixed modes,” has aroused perplexity from the date of its publication. It is not easy to understand what exactly Locke means by “mixed modes,” or why he believes they represent a viable basis of moral philosophy. It is not clear, at the outset, why Locke thinks a radical new basis for morality is necessary in the first place. Mixed modes, Locke tells us, are constructs of the mind, composed “very arbitrarily, made without patterns or reference to any real existence.”Footnote ¹ This has led, understandably, to a line of interpretation according to which Locke is a moral relativist.Footnote ² Yet as a founder of liberalism, Locke is also known as a champion of natural law and natural right, taken as universal and absolute moral standards. In the Essay itself, Locke speaks of moral vice as a violation of “the eternal law and nature of things” (1.21.56), and asserts that this law is a product of divine legislation (1.3.6, 12; 2.28.8). How can this be consistent with the notion that morality rests on a set of “arbitrary” mental constructs?

This paradox has raised doubts concerning Locke's consistency, and his sincerity. Some interpreters have maintained that the Second Treatise of Government, Locke's signature argument for liberal natural law and natural right, is incompatible with the philosophical foundations provided by the Essay concerning Human Understanding.Footnote ³ Yet the publication histories of the two works overlap, making it difficult to argue that the apparent discrepancy reflects a change of opinion on Locke's part. Moreover, as we have just seen, the Essay is apparently inconsistent with itself on this issue. This has led others to argue that at least some of the inconsistencies are deliberate, the product of Locke putting a religious veneer on his teaching, for purely rhetorical purposes.Footnote ⁴ In the past generation of scholarship, a consensus has emerged that Locke is fundamentally consistent in arguing for an objective morality, though without agreement as to whether this morality relies on a divinity. Locke's consistency is often asserted in only a vague or general way, however, without a clear account of how natural law and the theory of mixed modes are to be reconciled.Footnote ⁵

Locke was not so blind or careless as to be unaware of the problem his theory of mixed modes creates. He calls attention to it himself: if moral terms are arbitrary constructs of the mind, he asks in the Essay, will this not lead to complete “confusion of virtues and vices” (4.4.9)? His reply is that it will not, any more than it does in mathematics, which is also based on “modes,” or constructions of mind. It is precisely this grounding in modes, Locke proclaims, that allows both mathematics and moral philosophy to be precise or “demonstrative” bodies of knowledge (3.11.16; 4.3.18–20; 4.4; 4.12.8). What has remained mysterious to a great many readers is how exactly Locke understands the philosophy of modes to have made this possible.

We can fully vindicate Locke's consistency, and make progress on the religious question, only if we reexamine, from the ground up as it were, Locke's concept of modes, including mixed modes. This will help us to place Locke's argument within its proper philosophic and theological context. In the Essay, Locke relentlessly attacks the old philosophy—the Scholastic philosophy rooted ultimately in Aristotle—from what we may characterize as a “nominalist” position.Footnote ⁶ The only things that exist are particulars; “General and universal are creatures of the understanding” (3.3.11). The older philosophy held that universal categories, such as species, reflected intelligible, immaterial forms. Locke maintains that no such “metaphysical” entities exist.

Locke takes this position partly in support of the new natural science. He advertises the Essay concerning Human Understanding as an assist to the work of the scientists Boyle, Sydenham, Huygens, and “the incomparable Mr. Newton” (Essay, “Epistle to the reader”). Important portions of the Essay's argument rest on the atomic or “corpuscular” hypothesis of his friend and fellow Royal Society member Robert Boyle (e.g., 2.8.18–19, 21–22; 3.6.3; 3.23.11, 26; 4.3.25). Boyle makes clearer than Locke (though Locke makes it clear enough) that the materialist, corpuscular view of nature is incompatible with the old, Scholastic metaphysics. The new science could not prevail, then, unless the old metaphysics were disposed of. One of the aims of his Essay, Locke tells us, is to clear away the “rubbish that lies in the way to knowledge,” especially the “uncouth, affected, or unintelligible terms” foisted by Scholasticism on the educated public of his day (“Epistle to the reader”; 3.6.24, 4.7.11). Significant portions of the Essay's argument cannot be properly understood unless we keep this destructive task in mind.

Clearing away the Scholastic rubbish will facilitate the advancement of science, but its consequences will not end there. Along with its distinctive understanding of physical nature, the older metaphysics supported a distinctive grounding for morality. This had anchored Western moral and political philosophy for millennia. The new materialist science fashioned what Locke and his colleagues regarded as a more than adequate replacement for the natural philosophy of Scholasticism. It was much less clear (as it remains unclear today) what kind of moral theory is compatible with this science. A good deal of the controversy surrounding the new science, and Locke's Essay, stemmed from the fact that the repudiation of the old metaphysics also destroyed the accepted foundation of morals.Footnote ⁷ Locke assigns himself the task, after removing the Scholastic rubbish, of building a new moral theory on principles compatible with the new view of the world. This is what his “mixed modes” are intended to accomplish. Where the older theory gave moral ideas the kind of being or existence associated with Aristotelian form, Locke insists that moral categories have no substantial existence, are not “beings” of any kind. He also tells us they are objectively, indeed eternally, valid. We can only hope to make sense of this by taking up Locke's argument from the beginning.

I

“Mixed modes” in Locke's presentation are a subset of the broader class of “modes.” Modes in turn are a subset of complex ideas, that is, ideas assembled by the mind (2.12.3). The mind takes the simple ideas of brown, rough, and cylindrical, for example, and assembles them into the complex idea of tree trunk. All ideas of “substances,” that is, of physical objects, are such assemblages (ibid.). “Relations” constitute another type of complex idea, and “modes” are a third. Modes, Locke explains, are ideas of attributes that are not seen as existing independently, but “dependencies of or affections of substances” (2.12.4). Number is a principal class of simple modes, created by the mind through mere repetition of the idea of unity. Ideas of extension of space and duration of time are similar quantitative modes (2.13–17). Other simple modes include ideas of motion, such as rolling or sliding (2.18.1–82). These ideas are distinct from any object to which we may apply them, but have no existence apart from objects.

Locke did not pioneer this use of “mode.” By his time, it had become common currency among modern, anti-Scholastic thinkers. Descartes used it in a sense comparable to Locke's, as did the authors of the influential Port-Royal Logic.Footnote ⁸ It will profit us most to look at Robert Boyle's use of the term, because of both Locke's close association with Boyle, and Boyle's use of “mode” explicitly as a weapon against Scholasticism. On the basis of his corpuscular theory, Boyle argues that all the qualities possessed by material objects are only “Modes of Matter.”Footnote ⁹ This, he tells us, displaces the Aristotelian-Scholastic use of “form.” In Considerations and Experiments concerning the Origin of Forms and Qualities, one of his book-length polemics against the Scholastics or “Modern Aristotelians,” Boyle puts it this way: “The Controversie betwixt us and the Schools is this, Whether the Forms of Natural Bodies be educed out of the Power of the Matter, and whether they be Substantial Entities distinct from Matter?” (FQ 48). Boyle's account is purely materialistic: all the qualities of matter, from shape to weight to color, are mere modes of the matter's existence, generated from the qualities or “texture” of the corpuscles, and the way those corpuscles combine and interact in compounds.Footnote ¹⁰

The Scholastic theory that Boyle and Locke are so keen to rebut was inspired by Aristotle, but differed in key respects from Aristotle's original. Aristotle had proposed that the order of nature, its division into regular species or kinds of things, was to be explained by the existence of “forms” and “formal causes.” In Aristotle's view, the existence of horses or human beings, as species or as classes of things produced regularly by nature, cannot be accounted for by the bare action of matter and motion. In addition to the material causes of such things, there must be formal causes. The forms of horse and man, immaterial and intelligible, must in some way cause individual horses and men to come into being. If these forms do not have being in the strong sense postulated by Plato, they at least have independent being of a kind.Footnote ¹¹ There exists, then, a class of intelligible beings that collectively form a template, so to speak, for the order we observe in material nature.

Over the centuries of Scholasticism's development, this general understanding of nature was modified or augmented. Perhaps in an effort to increase the explanatory power of the Aristotelian approach, the later Scholastics multiplied the number of forms, or created a hierarchy of forms. Shapes became separate forms (sphere, cube); colors were held to be produced by their own forms; heft or taste could be forms, as could motion and rest.Footnote ¹² “Quality” and “form” were assimilated, as virtually every quality was to be attributed to its own “substantial form.” The species forms of compound beings, then, consisted of a specific collection of these “subordinate forms.” Forms of yellow, ductile, and heavy were thus part of the form of gold (see Boyle, FQ, 38, 168; Locke, Essay, 2.31.6, 3.10.16–17).

Strange as this doctrine may seem to children of the scientific age, it is important to understand that it was not moribund or beneath serious notice in Boyle's or Locke's day. To the contrary, it dominated both pulpit and university. Boyle, in one of his polemic titles, calls the neo-Aristotelian view the “vulgarly received notion of nature.”Footnote ¹³ The vehemence and even indignation with which Locke rebuts this view sufficiently reveal the power he thought it still had in intellectual circles. The polemics in which the Essay became embroiled, some rooted in the idea that the Christian doctrines of the Trinity and of the Eucharist depended on the old idea of substance, serve as confirmation of the continuing power of Scholasticism at the time of the work's publication.Footnote ¹⁴

When Boyle maintains, in his Origin of Forms and Qualities, that the “form” of a body is merely a “mode,” “educed out of the power of the matter,” he emphasizes the incompatibility of this view with neo-Aristotelianism. Form results from the combined characteristics of an object's material components, and is not the expression of a metaphysical or immaterial form (FQ, 12, 28–36). Changes of form are to be explained by changes in objects' physical components, the qualities, texture, arrangement, and motion of those components. When an object changes color, becomes warmer, or turns to liquid, these are not superventions of neo-Aristotelian “forms,” but mere changes of “mode” in the matter. Gold is yellow and heavy owing to the combined characteristics of its corpuscles, not because it participates in certain forms. When asked to explain why gold has the qualities it does, the Scholastics, Boyle charges, would offer the circular or intellectually sterile answer that it is because of the influence of the respective forms: it is yellow owing to the presence of the form of yellow, and so on. Similarly, if a piece of brass is made into a sphere, the Scholastics take this to show that the form of sphere was in it, in potentia (FQ, 30). If blue and yellow powders are mixed and the product is green, they attribute this to the supervention of the form of green on it (FQ, 43). Boyle finds such explanations fruitless, indeed “incomprehensible” (FQ, 29). Spherical shape, green color, and all other qualities of material objects are products of material causes alone, are mere modes of matter.

Like Boyle, Locke finds the Scholastic use of form to be useless and “unintelligible” (Essay, “Epistle”; 2.31.6; 3.10). He adopts the concept of “mode” as a partial corrective. Locke's emphasis is somewhat different from Boyle's, inasmuch as he is primarily interested in the ideas we have of modes, and how we get these ideas, rather than the scientific investigation of them.Footnote ¹⁵ A mode, then, to repeat, is an idea of an attribute that is separable from ideas of objects, but not conceived as having any existence independent of objects, ideas like dozen, or rolling. In Locke's analysis, the mind assembles or abstracts these ideas from its experience (1.13–18). Indeed, he tells us, ideas of modes are “voluntary collections of simple ideas which the mind puts together, without reference to any real archetypes or standing patterns existing anywhere” (2.31.3). This formulation is striking, and anticipates Locke's language concerning the “arbitrary” character of mixed modes such as moral ideas. What he is saying, though, is less radical than it seems. Locke does not mean that numbers, for example, are wholly arbitrary constructs, or that it is not possible objectively to determine whether we have a dozen pencils here or not. This, of course, would make the science of Newton and Boyle impossible. What he is denying is the Scholastic view that these ideas refer to real, intelligible entities—forms—whose influence accounts for the existence of the trait in the objects (e.g., 2.31.3, 6). To say that ideas of number refer to nothing “existing” is only to deny the Scholastic theory of substantial forms, and to reiterate the nominalist premise that the only things that exist are particulars. It does not entail a claim that number does not correspond to reality, or may not be used to describe reality. Similarly, to say that “green does not exist” is only to say that green does not exist as a separate being or entity, whose supervention makes green objects green. It is not to say that there are no green things in the world, or that “green” is a purely arbitrary mental construct (though to a degree it is—where exactly is the line between green and blue on the color spectrum?).

Locke quarrels with Scholasticism regarding the status of such qualities, but his quarrel does not stop there. Simple modes are not even his most important disagreement with the Scholastics. “Subordinate forms” such as sphere or green are less worrisome for Locke, and for Boyle, than are the forms attributed by the Scholastics to whole beings, or objects in the world. This is the sense of “form” that is true to Aristotle's own teaching, the sense in which the being of a tree or a horse or a man is to be explained (in part) by the presence of the form of tree or horse or man. One of the strengths of the Aristotelian theory of forms was that it explained the existence, and persistence, of species or kinds in nature. It also allowed a distinction between the “essential” traits of a species, those included in and owing to the species form, and traits that are merely “accidental.” Being quadrupedal is essential to horses, while being of a certain color is not. Being rational is essential to man, but being bald or non-bald is not.

According to both Boyle and Locke, if the centuries-long career of Aristotelianism demonstrated anything, it was the inability of the theory to discover such essences. Learned doctors simply asserted that certain attributes were essential to a species of objects, on the authority of other doctors. Thereafter, “science” consisted of little more than wrangling over the consequences of the stipulated ideas. Observations contrary to the accepted principles were explained away by overly subtle distinctions, or discounted as “nature failing to achieve its intended end.” Boyle ridicules the neo-Aristotelians for debating whether water and ice have different essences (FQ, 49, 57–58; see also Locke, Essay, 3.6.13). Locke is equally scathing: could the Scholastics ever tell us for certain which qualities of gold are essential, and which are not? If we found a piece of metal containing all the “essential” traits of gold but one, would their hypothesis help us in categorizing it? They cannot tell us, he asserts, whether “monsters” or unnatural births throughout the animal kingdom constitute separate species, or which human “changelings” might share in the human essence and so qualify as truly “human” (3.6.3, 17, 22; 2.31.6, 10).

Locke, who pursues the philosophical implications of this position much further than Boyle, makes two incisive and interrelated challenges to the Aristotelian-Scholastic tradition. One is epistemological, the other more broadly ontological. The epistemology of Aristotle and the Scholastics, based on forms, promised certainty. If forms play the role postulated by their philosophy, true knowledge can consist only of knowledge of these forms, for a grasp of the shifting and evanescent particulars is not worthy of the name of “knowledge.”Footnote ¹⁶ Propositions regarding particular objects may be true, but they could be otherwise. Aristotle places these in the category of “opinion,” as Plato had. Knowledge begins with sense experience for Aristotle, but requires that the mind at some point transcend the particulars of sense, to gain a grasp of the intelligible form that stands behind the particulars and accounts for what they are (horse, man, etc.).Footnote ¹⁷ As Aristotle says in De anima, mind must have a special capacity to receive the intelligible objects of knowledge, like the capacity of the senses to receive sensation.Footnote ¹⁸ This direct grasp or intuition of the intelligibles constituted knowledge, and brought certainty. Once this knowledge was secured, science could become deductive. Indeed, it might resemble the Scholasticism that Boyle and Locke describe, drawing consequences from notions fixed in advance, with empirical observation playing a minimal role (e.g., Essay, 2.31.6, 4.12.1).

Locke's epistemology denies this process at every point. Universal ideas, such as species, are constructed by the mind on the basis of its experience of particulars (2.32.6–7; 3.3.8). They do not, nor can they ever, represent a grasp or intuition of an intelligible form that constitutes the pattern for that species. Accordingly, Locke abandons the promise of certainty that knowledge of forms held out. If “science” entails certainty, a “science of bodies” will be forever beyond our reach, he concedes, for empirical “knowledge” can attain to no more than probability (4.3.26). Our ideas of species, which the Aristotelians attributed to a grasp of form, are rooted in nothing more than observed similarities and correlations; like all empirical discoveries, they can be generalized only probabilistically (4.3.11–14, 25; 4.6.4, 15). Our species idea of gold, for example, depends on the attributes of gold that we are aware of at present, that we have found in all samples of gold that we have seen to date, and that we choose to make definitive of the species. We cannot claim that this complex idea is the “real essence” of the thing, for that would mean that “the properties we discover in that body would depend on that complex idea” (2.31.6). It would entail the existence of a species form that causes each member of the species to have these traits. Locke asserts to the contrary that our species notions represent only the “nominal essences” of classes of things, those characteristics that we find convenient at present to set off a given group of objects. We may later revise our species concept, or even decide to divide nature into different species, for “general and universal are creatures of the understanding” (3.3.11). We devise such notions for our own utilitarian purposes, and they may be perfectly adequate to those purposes. They may facilitate the advancement of useful knowledge, without representing the master template of nature (3.1.3; 3.3.4, 10; 3.6.2, 7; 4.6.4; 4.12.10).

One mistake we must never make is to claim that these notions do represent an objective template of reality. Here Locke's epistemology becomes ontology, an anti-Scholastic metaphysic. In book 3 of the Essay, Locke identifies two ways of understanding species. One belongs to those who “suppose a certain number of essences, which serve as patterns for all natural beings, and account for those beings' belonging to one species or another” (3.3.17; 4.4.13). This we recognize as the Aristotelian and Scholastic view. “The other and more rational opinion,” Locke says, belongs to those who look on all natural things to have a real but unknown constitution of their insensible parts, from which flow those sensible qualities which serve us to distinguish them one from another, according as we have occasion to rank them into sorts under common denominations (3.3.17).

The “more rational” view rests on the corpuscular theory of nature (“insensible parts”), and issues in a nominalist view of species (“according as we have occasion to rank”). Particulars are the only things that exist; species have no existence, as species. There are no “essences” of species, only of particular objects. As to those essences, we do not know, and probably will never know, the “real essence” of any physical object—not even of a pebble, Locke tells us elsewhere, and certainly not of ourselves (3.23.35; 3.6.3; 3.10.21). We cannot know all the attributes of all its corpuscular parts, nor how they combine to produce the sensible impressions we receive of them (e.g., 2.23.10; 2.30.2; 3.6.3; 4.2.13; 4.3.11, 25). Every object is a unique combination of such parts, of which we can have only an imperfect idea. When we group objects into species, we do so on the basis of characteristics of our choosing. Nature does indeed make things alike, such as races of animals and plants (3.3.12–13). But Locke is adamant that nature does not produce things by species. The “frequent production of monsters,” he insists, represents nothing less than a refutation of the Scholastic-Aristotelian hypothesis (3.3.17). Such phenomena show that it is a mistake even to presume that nature's production is fundamentally “regular” (3.10.20; 3.6.23). There is often greater variation among individuals of a (nominal) species than between two species.

This applies as well to those we place in the human species (4.16.12; 4.6.4; 4.20.5). Locke does not shrink from this consequence of his philosophical premises: our notion of the human species is just as much a mental construct, just as tenuous or arbitrary, as all our other species concepts. Indeed, humanity might be the Essay's principal example of a constructed and variable species concept. We may assign humanity a “nominal essence,” as we do with other species. Whether “naturals” are to be considered human depends on how one has defined this nominal essence; nature itself provides no way to decide the issue (3.6.22; 4.4.15–16). The common view of the species, Locke tells us, focuses crudely on appearance (2.27.8; 3.6.3). If one were to define “white” as part of the human essence, Negroes would be excluded from humanity (4.7.16). One may define man as a “featherless biped,” “laughing creature,” or (like the Scholastics) as a “rational animal.” Each of these is correct, and all might be useful, depending on one's purpose or interests (3.10.17; 4.8). People “bred up in the learning taught in this part of the world” will think that their idea of man refers to a “real essence,” but they are mistaken (2.31.6; 3.6.24). In fact, each human being (and each natural) is a discrete being, with a separate “real essence.” There is no species essence of man in the Aristotelian-Scholastic sense, no human form. Putting himself forward for inspection, Locke notes that he has a distinctive appearance, and a distinctive set of acuities, compared to others put under the same species, “but there is nothing I have is essential to me. An accident or disease may very much alter my colour or shape; a fever or fall may take away my reason or memory, or both”; yet the remaining lump will still be John Locke, with a real essence of its own (3.6.4). If we wish to decide whether that lump is “human,” we will be thrown back on some nominal essence we assign to humanity for, to repeat, humanity has no species essence by nature.

This argument undercuts Aristotelian and Scholastic epistemology and metaphysics, obviously, but it does much more. For people educated in Locke's part of the world, it destroys the very basis of morality. The Aristotelian-Scholastic theory of form imparted a teleology to nature, inasmuch as each natural being's form was also a perfection toward which it tended or aspired.Footnote ¹⁹ An acorn's nature and perfection is to become a stately oak. In the case of humanity, the form and the perfection were the anchor of morality: it is the duty of each of us to exert ourselves in the direction of human perfection, the perfection implicit in our species form. The Scholastic view of this perfection and the character of the striving were different from Aristotle's, but their moral theory otherwise operated in the same way. When Locke says that there are no species forms, including very emphatically the human species, then, he not only dissolves the old moral theory, he forecloses the possibility of building morality on any similar foundation.

Locke can take such an uncompromising position on the old metaphysics of morals because he believes he has discovered a wholly different and better way to ground morality. This is his theory of moral “mixed modes.” But how can this theory possibly rescue morality? If species concepts like “humanity” have an element of arbitrariness, Locke tells us that mixed modes are arbitrary to a still greater degree (3.6.28). If the constructed nature of species concepts makes them ineligible as bases of morality, how can mixed modes possibly substitute for them? How can such arbitrary constructs ground duty, and the certain or “demonstrable” science of morality that Locke promises? We can make significant progress toward solving that mystery, I believe, if we look at the parallel theory of one of Locke's preferred authorities in moral philosophy, Samuel Pufendorf.

II

Samuel Pufendorf was an exact contemporary of Locke, both being born in 1632. Pufendorf's magisterial, eight-volume De Jure Naturae et Gentium (On the Law of Nature and Nations), was published in 1672, relatively early in Locke's intellectual career. One year later, Pufendorf published a short compendium of his philosophy for use in schools, De Officio Hominis et Civis Juxta Legem Naturalem (On the Duty of Man and Citizen According to Natural Law), which made his teaching more widely accessible.Footnote ²⁰ Pufendorf was a legal and philosophical authority of the highest stature in Locke's day. Locke's mature judgment on Pufendorf may be gathered from the fact that, in Some Thoughts concerning Education, there is no author he recommends for a child's education so highly as Pufendorf (§186). In Some Thoughts concerning Reading and Study for a Gentleman, perhaps the only place where Locke publicly recommends his own (anonymous) Two Treatises of Government, he simultaneously recommends Pufendorf's full-length work as “the best book of that kind.”Footnote ²¹

If Robert Boyle provided a model for the use of “modes” in describing attributes of physical objects, Pufendorf seems to have pioneered the similar use of “modes” in moral philosophy. He explains this theory at the beginning of On the Law of Nature and Nations (the theory is not explicitly stated in the shorter Duty of Man and Citizen). As with Boyle, Pufendorf adopts this conceptualization as a deliberate substitute for Scholasticism: “Substance” and “mode” are to replace “substance” and “accident.”Footnote ²² We have seen Boyle's application of this paradigm to the study of nature. When Pufendorf extends this line of thought into the moral realm, making moral entities into “modes,” it entails, as Locke also concluded, that “moral ideas clearly do not exist of themselves” (JNG, 1.1.3; Locke, Essay, 2.22; 2.31.3; 3.5.3). Instead, Pufendorf says, they are ideas imposed or superimposed on material reality by mind (JNG, 1.1.4). In his account of the “nature of moral entities,” Pufendorf describes them as “certain modes, added to physical things or motions, by intelligent beings,” to direct man and “secure a certain orderliness and decorum in civilized life” (JNG, 1.1.3). A mode of this type is clearly different from a mode of Boyle's type, or a Lockean “simple mode,” which merely portrays an affect of a material object. A moral mode, by contrast, may or may not correspond to actual human behavior. It is not a description, but a prescription, a command, a law, an ought.

Simple natural or physical modes, if we may call them that, arise simply and ineluctably from the material properties of objects. Pufendorf traces them ultimately to God, as creator of those objects and their properties (1.1.2, 4). Knowledge of these, Pufendorf says, has been developed to an admirable degree by others. Moral entities, however, have gone largely uninvestigated, though study of their nature behooves man most of all, “to whom has been given the power to produce them” (1.1.1). Now, it can hardly be said that earlier writers had neglected the study of morality. To say nothing of classical and medieval writers, Pufendorf's own theory relies heavily and explicitly on both Hugo Grotius and (often as a negative foil) Thomas Hobbes. What he means to say, then, is that earlier authors have failed to grasp the nature of moral entities as “modes.”

Property is one of Pufendorf's principal examples of a moral mode. Like Grotius and unlike Locke, Pufendorf holds that property does not exist by nature. Human intelligence creates the moral mode of property and imposes or superimposes it on the world.Footnote ²³ As a mode, it refers to nothing “real.” Calling something “my property” designates no real attribute of the object. When I transfer ownership to someone else, nothing in the object changes. What changes is merely the “mode” in which it is viewed by human beings. Nature, we might say, is oblivious to this and all moral modes. Such modes bear a different relationship to material reality than do the simple or physical modes of which Boyle and Locke speak. Modes such as rolling or three describe aspects of concrete experience. Although they are in themselves abstractions, with no “real existence,” their relationship to material reality is relatively straightforward. A moral mode such as property is much more a free creation of mind. It describes no aspect of material reality. There was a time when it did not exist. When it is brought into being, it is imposed on reality as a prescription, and a prescription that applies only to certain types of beings. In reality, Pufendorf points out, “property” designates relations among human beings rather than objects, for the objects, to repeat, are unaffected by it (1.1.16). The moral mode of property is, from beginning to end, a human construct.

The discovery that moral entities are “modes” of this type is a great breakthrough, according to Pufendorf, because it allows morality to become a science capable of “demonstration,” yielding “a certain and pure knowledge, in every way and at all times constant and free from error” (1.2.3). In an extended repartee with Aristotle, he maintains that Aristotle correctly identified certainty with sciences that proceed deductively from axioms, but incorrectly identified ethics or morality as a field incapable of such precision (1.2). There is a part of moral theory, to be sure, which is prudential and variable. This concerns the management of one's own and the common affairs, and Pufendorf agrees with Aristotle that this part of moral philosophy is not demonstrative or precise (1.2.4; 1.3.6–7). The foundation of morality, however, and its demonstrative or certain part are the modes devised by intelligence. Since these modes are not empirical but are purely ideal, knowledge of them, and of the relations among them, can attain to a certainty akin to that of mathematics. “‘Where there is no property, there is no injustice,’ is a proposition as certain as any demonstration of Euclid,” as Locke was to put it, for, like geometry, it relies on the relations among ideas exclusively (Essay, 4.3.18).

Since moral modes, as opposed to mathematical modes, are prescriptive and imposed on reality by intelligent agents, their imposition is in effect an act of legislation. Pufendorf accordingly asserts that all morality takes the form of law, a binding rule that serves as the measure of right and wrong (1.2.4; 1.6). For such modes to be binding, however, the legislating intelligence must have specific authority to obligate (1.6). Hobbes saw clearly enough that the essence of morality is law, says Pufendorf, but he wrongly took all law to originate in human consent.Footnote ²⁴ He properly saw that morality could become a demonstrative science like mathematics, but he mistakenly thought that this was because human beings have created its principles. Grotius, like Hobbes, saw that morality had the character of law; he saw that human consent could generate moral law, but he realized in addition that this human power could only exist against the backdrop of a wider moral order. Where Grotius stumbled, according to Pufendorf, was in following the ancients who held that this law was somehow intrinsic to or ingrained in nature (JNG, 1.2.6; Grotius, JBP, Prolegomena 11). Pufendorf charges that this Aristotelian and neo-Aristotelian view is incompatible with divine omnipotence, as it postulates a “co-eternal extrinsic principle” that constrains God in creation (1.2.6; 2.3.2). Pufendorf takes a “voluntarist” position: the moral law is not an eternal and unalterable principle, but a free creation of God, who could have ordained that law differently. So, although human beings create some moral rules, for example through the institution of property and of political sovereignty, human consent is not the ultimate source of their authority. The duty of sociability, laid upon man by God, provides the true grounding for these human institutions (1.2.5; 2.3.5; 3.4). This background principle, rather than the consensual precepts, provides the ultimate foundation of morality (1.1.3; 1.2.5).

Pufendorf's voluntarism is motivated by more than the desire to vindicate divine power and freedom. His rejection of the Aristotelian-Scholastic, “naturalistic” metaphysic has left him with divine legislation as the only possible origin of morality. Nature is morally vacuous, bereft of the formal causes spoken of by Scholasticism, or the immanent moral principles relied on by Grotius. Morality can arise only by the imposition of moral modes upon nature, by intelligent agents. So far as nature is concerned, no actions are good or evil in themselves. This, says Pufendorf, is the truth in Hobbes's claim that good and evil do not exist by nature. If Hobbes argued that the bare impulse of self-preservation is the foundation of right, however, he was mistaken, for right cannot emerge from brute natural facts. Facts gain moral significance only with reference to a moral mode, a law (1.2.6). The only agent with the capacity, and the authority, to impose moral modes upon humanity in a universally valid way, is God (1.1.4). God is “the author of natural law” (1.6.7). God's authority to impose this law comes not simply from his overwhelming power, as Hobbes had supposed (1.6.10). It does not derive from his superior perfection, as Cicero had supposed. Mere superiority of nature, Pufendorf believes, is not enough to overcome the freedom of a creature such as man (1.6.11). Authority over human beings comes only from their consent, or in debt for some “special service” done them (1.6.12). God's creation of the world and of the human race is service enough to incur this debt, Pufendorf finds (1.1.4). Even then, he says, enforcement is necessary, including punishments for violation, in order to give law “stability,” to make it “efficacious,” indeed to make it law (1.6.12, 14; 1.1.4; 1.6.5; 1.8.3). Again, only God could effectively impose such sanctions upon the whole of mankind.

The fundamental moral law is thus divine law. In order for it to qualify also as “natural law,” of course, it must be knowable by the unassisted efforts of reason. Anyone who applies his reason, Pufendorf assures us, will discover that the author of the universe is also the author of natural law (1.6.13). But how do we determine the content of this law? Pufendorf denies that knowledge of the moral law is innate; we must reason our way to it (2.3.13; 2.4.2). The way we do so, he says, is by considering our nature, which comports with the natural law (1.2.6). The nature God has given us is rational, and sociable. The natural law therefore bids us care for our own preservation, but also to respect the equal rights of others (2.4.1; 3.2). Sociability requires us not only to respect, but actively to assist other human beings (2.2.9; 3.3). Virtues such as “kindness, humanity, fidelity, gratitude, and the like” are necessary to our survival, according to Pufendorf, and our nature is attracted to them—this shows us that they are a law and duty to us (2.3.5).

We learn these things, according to Pufendorf, by reflecting on experience. This, however, creates a difficulty that is relevant to our consideration of Locke. In Pufendorf's philosophy (and Locke's), moral modes are essentially, even radically, nonempirical. They are not part of nature, but intellectual constructs imposed upon nature. Yet Pufendorf claims that our knowledge of them stems, at least indirectly, from empirical observation, observation of human nature and of the requirements of human society. If nature is morally vacuous, if even human actions have no intrinsic moral significance, how can observation of nature or of human behavior teach us of the moral law? How can we pursue knowledge of the law empirically, if our quarry is essentially nonempirical? We can do so only if we regard the empirical facts as mere signposts, pointing to something outside the empirical realm. It is not that our sociable nature is a law to us, as Grotius or Cicero held; rather, that nature is an indication of the law God has imposed on us (2.3; 3.4.4). God gave us a sociable nature and imposed a corresponding law upon us, but his imposing this law and giving us that nature were two different, metaphysically different, acts. Only the second has moral significance. The distinction may seem overly subtle, but it is pivotal to Pufendorf's argument, because it creates the separation between material nature and moral law that he needs in order to make morality a “demonstrative” science. The only certain knowledge is knowledge of the relations among ideas, without specific empirical referent. Pufendorf's argument does rest on an assumption that God was consistent in the fashioning of our natures and of the moral law. It even seems to impose a rule of consistency that constrains God in creation. If God had intended a different natural law, Pufendorf insists, he would necessarily have given us a different nature.Footnote ²⁵ This assumption, or set of assumptions, would have warranted more notice than Pufendorf gives it, particularly since it seems at odds with his defense of divine omnipotence, and since it seems to imply a heterodox position on original sin.Footnote ²⁶ For our purposes, what we need to notice is that the voluntarist and nominalist position, so radical in its metaphysical suppositions, issues in a moral teaching that fits within familiar parameters. Though Cicero and Grotius were metaphysically in error for believing that the moral law is part of nature, their moral teachings, and even the way they derived those teachings, were to a large degree correct. Like them, Pufendorf argues from the rationality and sociability of human nature to the moral law. He frequently uses both Cicero and Grotius as authorities in developing his own argument. Though Pufendorf's argument is philosophically “voluntarist,” he is able to proceed essentially as a rationalist, or naturalist, in deriving and studying the natural law. Natural reason is sufficient to the task. This model must be borne in mind as we turn to Locke.

III

Locke is not a simple disciple of Pufendorf. He disagrees with Pufendorf on such fundamental issues as the orgins of property and the limits of political authority. Nonetheless, the foregoing synopsis is enough, I believe, to indicate that Locke embraced Pufendorf's approach to morals in its fundamentals.Footnote ²⁷ Morality consists of modes—a subset of “mixed modes,” in Locke's terminology. They are creations of intellect, which superimposes them on the material world, but they do not refer to anything real or any pattern existing in that world. Though we do not have innate knowledge of morality, we can reason our way to it (Essay, 1.3; Second Treatise, §6). Morality can become a demonstrative science, capable of true certainty, only because it is nonempirical, dealing with relations among mental constructs (3.11.16; 4.3.18–20; 4.4.6–7; 4.12.8). Since they are mental constructs, moral modes can be objectively valid or binding only if they come ultimately from God. Locke grounds this argument, much more explicitly than does Pufendorf, in modern science, which (he argues) has destroyed the notion of species forms and shown brute nature to be morally vacuous. The resulting theory, as both Pufendorf and Locke emphasize, cannot dispense with divine legislation.

Seeing Locke as a follower of Pufendorf in these specific respects goes far, I believe, toward clearing up the controversies in Locke interpretation mentioned at the beginning. One of these controversies concerns the compatibility of Locke's nominalism, as revealed in the “mixed modes” teaching of the Essay, with the natural law teaching of the Two Treatises of Government. The second concerns the sincerity of Locke's appeals to God in developing his moral theory. One basis for doubts on the first score is the “naturalism” of the Two Treatises, where Locke moves from empirical natural facts to conclusions about natural law, which seems incompatible with the Essay's claim that moral principles are arbitrary and nonempirical mixed modes.Footnote ²⁸ It also seems to obviate any divine grounding for these principles, opening the possibility that Locke's gestures in that direction are superfluous if not insincere. In the First Treatise, for example, he infers a universal human right of appropriation from our appetites and the availability of provisions in nature (§86). He infers parental duty from the desire of procreation (§88). In the Second Treatise, his argument for property similarly rests on human appetites and needs (§§25–26). Here, Locke seems to be closer to Grotius, who argued that moral principles are immanent in nature and independent even of God, than to Pufendorf. But we have just seen Pufendorf argue from empirical nature while rejecting Grotius's view of the source and grounding of moral modes. Close attention to Locke's argument in the passages just cited reveals, I believe, that Locke is following Pufendorf: the empirical facts he relies on are not the source of the moral principles, but are signposts or indications of divine intent. The duty of parents, for example, does not derive from their appetite to procreate, considered simply as a natural fact. That desire is rather an indication of a divine plan whose fulfillment requires this duty.Footnote ²⁹

Locke's argument is not “naturalistic” in the strong sense, then, but in the sense of Pufendorf. Yet this renders divine grounding necessary to his argument. Locke could have advanced a more secular and naturalistic argument without controversy. He could have relied not only on Grotius, but on Aristotle, Cicero, and indeed the entire “rationalist” as opposed to “voluntarist” tradition of Christianity (and we may put Richard Hooker too in this camp).Footnote ³⁰ He could have sided with Leibniz, or with Ralph Cudworth and the “Cambridge Platonists,” who strove to maintain a teleological view of nature in the face of modern science.Footnote ³¹ He does not, because his rejection of Aristotelian Scholasticism has foreclosed this possibility to him. He has demolished the metaphysical underpinnings of the old approach to morals. He follows Pufendorf rather than Grotius because he has been convinced by modern science that nature is, as Boyle described it, a “cosmical mechanism,” a “great automaton” moving in accordance with the “mechanical hypothesis” (VN, 39, 40, 160; FQ, 16). In and of itself, it is devoid of moral content. Nature is not ordered by species or forms; no teleology is inherent in it. In particular, there is no human form or essence in the Scholastic sense, to give humanity a perfection that stands as a duty to us. More humbly, our physical desires for preservation or for procreation cannot in themselves be the basis of any moral right or law. Locke's embrace of Pufendorf's theory of moral modes is conditioned by the whole of this epistemology, even his metaphysics. If (objective and universal) morality is to exist, then, it must come from outside of nature, from supernature. Locke says, repeatedly and in many contexts, that the essence of morality is conformity to a rule or law, and that the lawgiver is God. The “true ground of morality,” he says in one of these passages, “can only be the will and law of a God, who sees men in the dark, has in his hand rewards and punishments, and power enough to call to account the proudest offender”; or again, “what duty is, cannot be understood without a law; nor a law be known, or supposed, without a lawmaker.”Footnote ³² This is the only way Locke can combine his very prominent account of moral concepts as “arbitrary” mixed modes, with his equally prominent account of the moral law as “the eternal law and nature of things” (2.21.56). Mixed modes are all arbitrary from the point of view of nature, and can become eternal law only by legislation. Natural facts gain moral significance only by being brought under a rule, a moral mixed mode, by one with the authority to impose such a rule (2.28.15). Drunkenness is one of the Essay's principal examples of a violation of the “eternal law.” Drunkenness, Locke indicates in one passage, is no more remarkable than the drinking of a horse, considered as a mere collection of ideas, or as a phenomenon in nature (2.28.15). It is only by constructing “drunkenness” as a mixed mode separate from “drinking,” one imbued with moral meaning, that drunkenness is made a vice. “Dueling” is a mixed mode denoting certain types of combat; according to “the law of fashion, in some countries,” Locke notes, it is regarded as a sign of “valour and virtue.” If he left it at that, Locke would be a “constructivist” in the contemporary sense of the word. But he does not leave it at that: “when considered in relation to the law of God,” he informs us, dueling “will deserve the name sin” (2.28.15). This is the only way that mixed modes can become eternal rules of right and wrong.

IV

This will serve to illustrate at a general level how Locke takes his theory of mixed modes to have solved the problem of moral theory in the era of modern science, and how he believes it may escape some of the difficulties with which it is beset. Yet the way Locke elaborates portions of his theory creates other difficulties that demand some attention. He argues that “humanity,” like all universal or species categories, refers to nothing real. From the point of view of brute nature, all the creatures we call human beings are distinct collections of corpuscles, sharing no species essence or form or natural perfection, being brought into being by no metaphysical pattern. This is why the older moral theory, based on the metaphysics of species and their inherent perfections, must be dispensed with. Yet Locke speaks, in the Essay itself, of a “perfection” that somehow stands as a duty to all of us (2.6.3–4; 2.21.48, 51; 4.14.2; 4.18.6). This inconsistency does not seem to be resolvable through the theory of mixed modes, in and of itself. The perfection Locke has in mind is something of which a dog, for example, would be incapable, no matter who tried to devise a mixed mode to that effect. This perfection would seem to be based on the notion that certain “excellencies” are part of our makeup, which is exactly how Locke characterizes our faculty of thinking, abstracting, and understanding (2.11.10; 2.1.15; 4.18.5). This would seem to be a reversion to something like the old, Aristotelian-Scholastic type of theory.

The best way to get at Locke's understanding of this matter might be by looking at his conception of the moral “demonstration” that his philosophy of mixed modes will make possible. Locke reserves the word “demonstration” for those types of argumentation that are capable of certainty, something that empirical observation or conclusions rooted in empirical observation—including all of natural science—can never achieve (3.9.7; 3.11.17; 4.3) Moral philosophy can become “demonstrative” because it deals only with relations among ideas, abstaining from claims about empirical reality. This is why mathematics is certain, and this, according to Locke and Pufendorf, is why moral theory based on modes can become certain (3.11.16; 4.3.18–20; 4.4.16; 4.12.8).

In the Essay, Locke describes his projected moral demonstration this way:

This demonstration, or projected demonstration, follows the model Locke has laid out. Like geometric proofs, it is constructed entirely from abstractions. God is essential to it, and our (imperfect) idea of God, Locke has told us elsewhere, is formed by abstraction and extrapolation from certain observed qualities and perfections (2.17; 2.23.33; 3.6.11). According to the demonstration, this God stands in a certain relationship to us: “workmanship” on the one side, dependency on the other. But the crucial point, for our purposes, is that we are subject to the moral law not as “human beings,” but as “understanding, rational beings.” That is, man is subject to God's law not as the natural being, “man,” but as he falls under these abstract categories. Any being in these categories, human or not, standing in the specified relationship to a God with the specified nature, will be under the moral authority of that God. As Locke says at one point, a monkey with the requisite traits would be subject to God and moral law just as we are (3.11.16). The same would be true of the created intelligences on other planets about whom Locke speculates (4.3.24).

Locke's précis works as a “demonstration” because it does not rely on the supposition of a morally freighted and empirically grounded “human essence,” but on isolated traits that carry moral weight. Thus, as Locke is careful to point out, “when we say that ‘man is subject to law,’ we mean nothing by ‘man’ but a corporeal, rational creature: what the real essence or other qualities of that creature are in this case, is no way considered” (3.11.16). Locke may then without contradiction invoke these traits as “perfections” that serve as the basis of morality, since they are not perfections of an Aristotelian species, but abstract qualities that make any beings who possess them into moral beings. To the extent that moral beings are a superior kind of being (not according to nature, of course, but in accord with a standard above nature), these traits are appropriately described as perfections. The perfections are not “human” perfections, but perfections pertaining to “intellectual nature,” which might apply to any number of creatures (2.21.51; 1.1.1; 2.21.48; 4.18.5). The duty this perfection imposes on us is the same that would accrue to a moral monkey.

The formal consistency of Locke's demonstration, then, is unimpeachable. Yet other problems present themselves. We cannot help noticing the tentativeness or diffidence in Locke's description of his proposed demonstration, quoted above. The fact that Locke never produced the demonstration, despite the prodding of friend and enemy alike, suggests that it will not be quite so easy as a geometric proof.Footnote ³³ One obvious difficulty is the specification of its terms. Locke characterizes them somewhat differently even in the few passages we have just reviewed. In one, the moral subject is an “understanding, rational” being; in another, a “corporeal, rational creature.” When discussing his hypothetical moral monkey, Locke identifies the key trait as the ability to “understand general signs, and to deduce consequences about general ideas” (3.11.16)—that is, the ability to abstract and to reason abstractly, which Locke elsewhere describes as the “excellency” that sets man apart from beast.Footnote ³⁴ Now, these different formulations clearly converge on rationality, but they are not simply the same, and the differences might produce significant divergences in a moral demonstration. Locke never fully clarifies his meaning.

A potentially more serious difficulty has to do with the theology required by the projected moral demonstration. The demonstration stipulates a divine lawgiver, and we have already seen why Locke needs this figure to ground his morality. For the demonstration to work as a grounding for morality, obviously, the existence of such a god would have to be established. For although Locke takes it as a cardinal point in favor of his demonstrative approach to morality that it rests on abstraction rather than (less-than-certain) empirical knowledge, moral creatures must ultimately be assured that its terms correspond with empirical reality.Footnote ³⁵ We may take the existence of corporeal, rational creatures as a given.Footnote ³⁶ But for these creatures to know that they are in fact under a moral law, they must know that the divine legislator really does exist. More precisely, they must know that this legislator metes out rewards and punishments based on the moral law, in another life (1.3.6). Now, Locke does insist that we know of the existence of a god with demonstrative certainty. The proof is in ourselves: the mechanism of nature, as the new science has uncovered it, is incapable of producing a conscious, “cogitative” being. Any such beings must therefore be God's workmanship.Footnote ³⁷ But Locke evidently did not believe that he could establish with certainty that this god has the providential attributes that morality requires. At least, he never presents such a proof, though it would have been most opportune.Footnote ³⁸ When he discusses the afterlife and other aspects of God's providence in the Essay, Locke speaks only probabilistically.Footnote ³⁹

Probability will obviously not rise to the level required for “demonstrative certainty.” Before we dismiss it entirely, however, we should consider that probability, in Locke's accounting, may attain to virtual certainty. We can know only with probability that a man named Julius Caesar lived in ancient Rome, Locke notes, but no one can reasonably doubt it either (4.20.16). Locke tells us that probability is all that we have to guide us in most of our concernment, and that that is sufficient for the purposes of human life (1.1.5). Probability is all that empirical natural science can offer. Most of us are content with the “probability” that a stone will fall to the ground if I drop it, and might well be content with a moral theory having the same degree of probability. Yet Locke promises us certainty in moral demonstration.

Or are we misconstruing the moral certainty that Locke promises? Is it his point that certainty may obtain in moral demonstration, while probability must suffice for claims about the correspondence of the demonstration to reality? A bridge engineer may warrant the absolute certainty of his mathematical calculations, since they reflect no more than relationships among abstract, ideal entities; but (in Locke's scheme) he is justified in certifying only with (a high degree of) probability that the resulting bridge will stand. The second claim depends on empirical facts that as such are incapable of certainty. Is the same combination of certainty and probability to be characteristic of moral demonstration? Locke suggests as much in at least one passage, positing that the probability of rewards and punishments in the afterlife is enough to determine the behavior of a rational agent (2.21.70). Moral demonstration would then be certain, based on the suppositions of a rewarding and punishing god and corporeal, rational creatures, and it would suffice for moral action that the probable existence of that god could be established. Locke at any rate proceeds with his philosophical project as though it has a foundation solid enough to support it.Footnote ⁴⁰

A final difficulty with Locke's moral demonstration concerns the actual content of the moral law. What does the providential god demand of his rational creatures? What moral mixed modes has he made into eternal law? The Essay is remarkably sketchy on this question. When friends who had read the Essay pressed Locke to fill out the details of his moral demonstration, this is perhaps the principal thing they had in mind. Locke's reply, that his project in the Essay was to lay out the form and origin of moral knowledge rather than its content, did not satisfy them.Footnote ⁴¹ Of course, we get a fairly thorough sense of Locke's view of morality from his other works, such as the Two Treatises of Government (since this work was published anonymously, Locke may not have felt free to refer readers to it). Still, Locke did not in this or any other work produce a detailed or systematic account of natural law, as for example Hobbes had done.Footnote ⁴²

We do not have time here to canvass Locke's oeuvre in order to get a complete picture of his moral thought. What we can do is identify certain problems of Lockean morality that are linked to the philosophical underpinnings of the Essay, problems that are reflected in the teaching of the Two Treatises. When Locke speaks in the Essay of the perfection we have a duty to pursue, he describes it simply as the exercise of “due consideration” in pursuing true as opposed to merely apparent happiness (2.21.46; 2.21.43, 47, 50, 51, 59). Due consideration consists of the screening of pleasures and pains with a view to happiness, and exercising rational control over the appetites in order to secure it (2.21.48–53). It is in this context that Locke offers his extended treatment of the vice of drunkenness (2.21.35–38). Now, it certainly makes sense to view this as a perfection of “intellectual nature,” inasmuch as it involves rational deliberation and rational self-control. But it is tied as well to nonintellectual parts of our nature, our bodies and our appetites, which partially dictate the substance of morality. Morality is the pursuit of proper happiness, and happiness, Locke tells us, consists in securing pleasure and avoiding pain (2.21.41–42). He is careful to stipulate that he understands pleasure and pain broadly, not simply in a carnal sense, but his discussion makes clear that bodily pleasure and unease are key components of our moral pursuits, as the example of drunkenness shows (2.21.31–2, 34–41). Locke's moral philosophy is “hedonistic.”Footnote ⁴³

Hedonism is rooted deeply in Locke's philosophy. Happiness and pleasure are the concerns of morality because they are necessary preoccupations and motivations of any creature endowed with consciousness, and consciousness is inseparable from the intelligence that makes us moral creatures. “I lay it for a certain ground, that every intelligent being really seeks happiness, which consists in the enjoyment of pleasure, without any considerable mixture of uneasiness” (2.21.62). Self-concern is ncessarily at the center of every intelligent creature's motivation (2.27.9–10, 17, 62). What this means, paradoxically perhaps, is that all moral creatures will also be self-interested, indeed hedonistic, creatures. It goes without saying that the God who created them will be aware of this fact, and will tailor his moral law accordingly. For one thing, rewards and punishments will be attached to the law, without which obedience would not be rational, indeed the law would not be binding (Essay, 1.3.6; 2.28.6; Pufendorf, JNG, 1.6.12, 14; 1.1.4; 1.6.5, 8.3). Locke's moral law also serves happiness in the here and now. Apart from the example of drunkenness in the Essay, we may think of the way that natural law and natural right are linked to human needs such as food, drink, and clothing in the Two Treatises (1.86, 2.25–30, 57). In that work, it is also our right, if not our duty, to labor for “conveniences of life” beyond mere need (1.33, 41, 2.26, 34, 37–49). But exactly what degree of pleasure or convenience is sanctioned by moral law? It would seem to be simply that degree that maximizes our happiness, but what degree is that? For rational creatures with different bodies and different needs, would there be a different moral law? Would rational monkeys be under the same moral law, or would corporeal rational creatures on other planets, who may be “infinitely different” from us in other respects (Essay, 4.3.24)? Locke discusses the possible traits of angels in the Essay, as rational creatures superior to us, and perhaps without bodies (2.23.28; 3.6.3, 12; 3.11.23; 4.11.12; 4.17.14). What he never clarifies is how (or indeed why) corporeity and its needs condition moral law, when it is rational nature that makes us moral creatures.

V

Readers familiar with the thought of Immanuel Kant will recognize this set of problems. They may be struck by the similarity between the philosophy of Locke, on the foregoing account, and that of Kant. Kant bases morality on abstract, legislating reason, which he separates strictly from empiricism and the realm of natural necessity. We are moral as “noumenal,” rational beings, not as “phenomenal,” material beings.Footnote ⁴⁴ Locke says we are moral not as we are human beings per se, but as we are rational creatures. Nonhuman creatures would be subject to moral law as we are, so long as they were rational. The source of this law must be located outside of material nature, and moral science is strictly separated from empirical science.

It would not be hard to show that Locke had an influence on Kant. Locke's Essay was prominent throughout the eighteenth century, Kant was a lifelong reader of Locke, and references to Locke are found across Kant's works.Footnote ⁴⁵ I would point, however, to a different and deeper reason for the similarities between the two. Locke and Kant (and a host of other philosophers, beginning at least with Pufendorf) are confronting the same problem, the problem created for moral philosophy by empirical modern science and its view of nature. Modern science rests on an account of nature fundamentally incompatible with the old moral philosophy, perhaps with any “naturalistic” moral philosophy. Science finds nature to be matter and motion, rather than matter and Aristotelian form. Species have a tenuous existence; species perfections in the Aristotelian-Scholastic mold seem to be ruled out entirely. Nature thus understood is devoid of moral content; purely natural events have no moral significance. Human beings are obviously natural creatures, part of this natural world, whatever else we might be. Human action, considered simply as empirical fact, also has no moral significance—so at least Locke and Pufendorf and (later) David Hume tell us.Footnote ⁴⁶

Under these circumstances, it may well be that the only way to rescue morality is to find a grounding for it outside of material nature. This is the purport of Pufendorf's reformulation of natural law as a “mode” superimposed on material reality by (divine) intelligence, and of Locke's theory of morality as “mixed modes.” Thus was launched a line of thought that found a particular kind of nonempirical “metaphysic of morals” to be necessary. Kant's division between the material-phenomenal realm and a rational-noumenal realm is a more radical solution, but formally not dissimilar. I do not intend to portray Locke as a (mere) proto-Kantian; a Kantian might be more impressed with the differences between the two than with their similarities. A Kantian would attribute the problems discussed in the last section to Locke's failure or refusal cleanly to separate the empirical and the moral realms. Locke gives moral law a “transcendent” origin, but insists that it must serve the happiness of the individual, a happiness that has material as well as intellectual components. Locke stipulates indeed that the moral law must serve individual happiness, if it is to be binding on the individual. Kant would charge that the motivation of Locke's moral agent is hopelessly “heteronomous,” that a transcendent moral law cannot be mixed in that way with material concerns. Locke would retort that self-concern is inseparable from reason or rational consciousness, that for creatures that are corporeal as well as rational, material happiness is a necessary, and fully legitimate, concern. He might well charge Kant with failing to expel happiness from his own account of morality in any case.

This dispute, or imagined dispute, cannot be played out here, obviously. It does, however, serve to highlight some problems or unanswered questions in Locke's Essay. It is difficult to conceive exacly how Locke understands the relationship between humanity as a collection of natural beings, with no Aristotelian form, no species essence, and no perfection, on one hand, and humanity as part of the abstract category “corporal, rational creature,” subject to moral law, having a purpose and a perfection, on the other. It is paradoxical that Locke should deny that rationality is any part of his own essence, for example (3.6.4), and yet assert that his perfection, and ours, lies in a certain cultivation of the rational faculty. Locke has opened a metaphysical divide between material nature and morality, a divide that rational (or conscious, abstracting) creatures straddle. But how are we to understand this straddling? It is not clear that Locke worked out a fully consistent answer. Looking at the line of thought that goes from Pufendorf to Kant, it is not clear how the disparate threads of this approach can be pulled together consistently.