The ancient Greeks divided philosophy into three sciences: physics, ethics, and logic. This division is quite fitting for our purposes, and we can perfect it simply by determining the basis for such a division, so that we may be sure we have included everything, and at the same time have the necessary subdivisions correctly defined.
Kant uses a division of philosophy employed by the Stoics. He calls these divisions “sciences,” meaning by “science” a systematic body of knowledge. This threefold division is based upon a principle of distinction which applies to all knowledge.
All rational knowledge is either material, about some object, or formal, about the forms of understanding and reason themselves, and about the universal canons of thinking, without reference to particular objects of thinking. Formal philosophy is called logic.
Material knowledge includes not only scientific knowledge, as we commonly understand it, but also aesthetics, psychology, and metaphysics. Even ethical principles are listed as material knowledge. Anything which may be classified as “what we think about” is to be labeled material knowledge. Formal knowledge, on the other hand, we might describe as the way we think about the objects of material knowledge. Formal knowledge is not knowledge of the thoughts or emotions which at any given time comprise our intellectual states, but rather it is knowledge of the procedural patterns which the mind follows when it thinks about anything at all.
Kant mentions two kinds of logic: (1) a logic concerned with the “forms of understanding and reason themselves,” and (2) a logic concerned with the “universal canons of thinking.” Kant does not emphasize this distinction in the first part of the Foundation, but in Section III this distinction becomes important, and we must take a brief note of it here. We have already seen Kant’s concern with the structure or framework of understanding.1 In the Critique of Pure Reason, Kant gave the title “Transcendental Logic” to his investigation of the elements of this a priori structure. Transcendental logic is the examination of the pure a priori concepts with which the mind molds raw sensations into meaningful experiences.
The other logic, normally called formal logic and concerned with the universal rules of thinking, Kant calls “general logic.” General logic is a study of the formal structure of argument, that is, with whether an argument is in valid or invalid form. What both these logics have in common is that neither is concerned with any particular item of knowledge. In other words, logic is a formal kind of knowledge. Kant plays on two meanings of “formal": the formal aspects of argument, and the formal structure underlying experience.
Material philosophy, on the other hand, directed at particular objects as well as at the laws which rule these objects, is subdivided into two parts. These parts correspond to the different kinds of laws. Material philosophy dealing with the laws of nature we call physics, while that concerned with the laws of freedom we call ethics. Some refer to the former as natural philosophy, and to the latter as moral philosophy.
We all are familiar with laws of nature, but what could Kant mean by “laws of freedom"? At first glance this seems self-contradictory, since whatever happens according to law will be determined, while a free choice or action should be free from any determination by law. The solution to this puzzle will become clear when Kant discusses the nature of moral law. Here we need note only that Kant makes a precise, though undefined, distinction between laws of nature and laws of freedom (i.e., morals). True enough, everything subject to a law of nature is inflexibly determined. But a law of freedom is not a law of nature. It does not describe what always happens, as do laws of nature; rather it prescribes what an agent must do in order to fulfill his moral duty. Laws of freedom are no more determining than are the laws enacted by a legislature but they are truly laws for all that.
Logic cannot derive its universal and necessary (i.e., a priori) laws of thinking from experience. If it did, it simply would not be logic at all, since logic is the ultimate guideline for understanding and reason; its principles must be valid for every activity of thought, and thus provable by reason alone. Natural and moral philosophy, on the other hand, do indeed rely on experience. Natural philosophy seeks knowledge of laws of nature based on experience of what actually takes place in nature. Moral philosophy seeks to know the laws of human choice, even when such choice is influenced by nature. Thus while moral philosophy tries to learn the laws of what ought to happen, it must at the same time consider those situations in which what ought to happen often fails to happen.
Kant, like others before and after him, holds the laws of logic to be a priori, grounded in reason. Experience cannot provide the necessity characteristic of these laws. But both physics and ethics are to some extent based on experience (literally, they each "have an empirical part"). The empirical part of physics includes what we usually call the natural sciences. The empirical part of ethics would include anthropology, psychology, sociology, or the social sciences in general. As empirical studies, both physics and ethics investigate how antecedent factors of heredity, ancestry, environment, and society affect human choice and activity. As such, these sciences formulate descriptive laws, general and universal statements of the ways events have been observed to take place. These laws do not state commands given to nature. If an unsupported stone did not fall, we would not say that it had in any sense disobeyed the law of gravity. Rather, we would hunt for a purely natural explanation of what had taken place; or we even might want to restate the law of gravity to account for this new datum in experience.
Moral laws, however, do command; they prescribe; and they are frequently disobeyed. What is more, we are able to account for this disobedience without having to restate the law; the explanation would be found in the laws of human inclination. One of Kant’s insights is that everything that happens happens in accordance with some law, either a law of nature or a law of freedom.
Any philosophy which relies on experience is empirical philosophy; but if a philosophy proposes a theory derived solely from a priori principles, we call it pure philosophy. Logic, having no particular object derived from experience, is pure formal philosophy. Pure material philosophy, while it does have an object of investigation, does not find this object in experience. The name for such a pure material philosophy is metaphysics.
Both physics and ethics have an empirical side: physics (properly called) and psychology, respectively. But they each have an a priori side. So two kinds of metaphysics are possible: a metaphysics of nature and a metaphysics of morals.
So far Kant has distinguished between what we generally call science and philosophy. Science (the a posteriori or empirical "philosophy") is derived from perceptual evidence. “Pure” philosophy (a priori or rational) is a purely intellectual investigation. When this pure philosophy is merely formal, it is logic. Pure philosophy about the content of knowledge Kant calls metaphysics. And since pure material philosophy deals with two different kinds of objects, physical and moral, we have two kinds of metaphysics: a metaphysics of nature, which is a priori knowledge of the a priori laws of nature; and a metaphysics of morals, a priori knowledge of a priori moral laws.
Throughout the history of philosophy, the term “metaphysics” has referred principally to the metaphysics of nature. Metaphysics traditionally embodied the philosophical examination of the ideas of substance, qualities, causality, and personal identity; it sought to prove the existence of God, freedom of the will, and immortality of the soul; it theorized about universals, relationships, and the nature of reality. Kant argued that some of these ideas are pure a priori concepts: they are forms of knowledge. Substance and causality belong to this group. Other ideas, such as God, freedom, and immortality, do not fit into categories of experience at all. Since these ideas refer beyond the range of human experience, Kant denies that we can know whether such ideas refer to anything real at all. (See Chapter I, Section 1.) In brief, Kant claims that a metaphysics of nature is impossible: without experience we cannot have any knowledge of natural or supernatural objects or events. If there are any a priori laws of nature, they must be purely formal, contained in a transcendental logic, not in a metaphysics of nature. If traditional metaphysics is not to be an empty discipline, it will have to be either a transcendental logic or a metaphysics of morals.
At this point, by certain principles of division, Kant has arrived at a definition of a “metaphysics of morals.” He has not yet tried to prove that such a metaphysics is possible. So the question still remains, can we have a priori knowledge of a priori moral laws? This question, of course, is only another way of asking whether we can discover an ultimate principle of morality. But even so, we can now see how such a metaphysics of morals would be related to the rest of philosophy. A graphic summary may be helpful.
We are now able to define a metaphysics of morals as an a priori investigation of the knowledge of moral laws governing human conduct. Consequently, a “Foundation for a Metaphysics of Morals” will be an investigation into the ultimate basis for all a priori knowledge of moral law; that is, a search for the ultimate foundation for rational morality.
All trades, arts, and crafts benefit from a division of labor. Instead of performing all tasks, each person specializes in one particular area which differs from all the others in the special skills it requires. By doing so, he achieves a higher level of performance and greater facility. On the other hand, where everyone is a jack-of-all-trades, doing all kinds of work indiscriminately, performance will remain at the crude level of the beginner.
Thus we might give some thought to whether the distinct areas of pure philosophy would likewise benefit from specialization. Would it not be better for the whole house of intellect to censure those who pander to the public taste with a thoughtless mixture of experience and rational principle? These so-called savants have no idea of proper procedure, yet they consider themselves scholars, and deride as “idealists” those who devote themselves to purely rational pursuits. In trying to do two things at once, things which require both a distinct technique and a special talent, they bungle the whole job.
I merely want to ask whether the nature of science does not demand that we carefully separate the empirical from the rational part. Should not a metaphysics of nature be prologue to natural science, and a metaphysics of morals serve as a ground for social science? If so, we must make sure to keep the meta-physical foundations of the sciences pure of anything derived from experience, so that we can determine in each case just how far reason can proceed by itself and where it finds its a priori principles. Also, we might discover whether this search should be pursued by all moralists (whose name is legion), or only by those who have the knack for it.
This passage is a brief plea for the same critical approach with which Kant began the Critique of Pure Reason. Kant objects not so much to psychological and moral investigation by one individual-he himself does both-but to the simultaneous, “mixed,” empirical-rational treatment of morality. He insists, as a fundamental principle of his critical approach, that it is intellectually dangerous to attempt an investigation of rational knowledge unless we first have outlined clearly just how far reason can legitimately proceed. The limits of rational inquiry must be carefully determined before we can develop an adequate theory of knowledge. For obviously, if the limits are surpassed, the resulting so-called “principles of knowledge” will be illegitimate and will produce contradictory beliefs. The first step, then, is to see “how far reason can proceed by itself and where it finds its a priori principles.”
Since my immediate concern here is with moral philosophy, I narrow the question: Are we not constrained by the gravest necessity to erect a pure moral philosophy which is wholly free from the merely empirical descriptions of social science? The common ideas of duty and moral law make obvious the possibility of such a philosophy.
Since we do have an idea of duty and recognize some moral laws to bind us, the awareness of these duties and laws must be based upon a priori principles. The obligation or moral necessity commanded by these laws cannot be found in experience (i.e., in any anthropological, psychological, or sociological investigation), but must be based upon principles of pure practical reason. When Kant says that such a philosophy is possible, he does not mean that someone has already developed it or that the possibility is obvious to everyone who understands the common ideas of duty and moral law. He means that it is indeed possible to develop a valid, pure moral philosophy and that our common ideas of duty and moral law are sufficient grounds upon which to begin the investigation.
Everyone must agree that a moral law must bind with unqualified necessity if it is to be the basis for obligation. The command, “Thou shalt not lie,” does not apply solely to men, as though other rational beings did not have to obey it. The same holds for all precepts which are truly moral commands. Thus we cannot expect to find the basis for obligation in the nature of a human being, nor in any set of human circumstances; rather, we must seek this basis a priori in the concepts of pure reason itself. Those commands which are based on principles derived from experience, no matter how slight the influence of this experience—even if it affects nothing more than the motive—are not moral laws. They may even be universally applicable and thus serve as practical rules; but lacking the unqualified necessity which rests upon a purely a priori foundation, they cannot be moral commands.
If a moral law is to bind under moral obligation, then it must bind absolutely. Kant insists that this absolute moral necessity must be grounded in reason, not in the nature of man. In the first place, we can know man’s human nature only through experience. Even though this experience be introspective, it is still a posteriori knowledge and cannot give us necessary truth.
Which of us can say that man is the only being who reasons? As a fanciful example, suppose we discover that dolphins have intellects comparable to ours and we learn to communicate with them. Dolphins are surely not human beings, even if they should think like we do. (For one thing, men and dolphins cannot interbreed.) Now suppose one day a dolphin lied to us. Couldn't we blame the dolphin, accuse it of immoral conduct, and say that it had done wrong? Where a cat cannot lie and so cannot do moral wrong, this dolphin intentionally said something it knew was false. So we could judge the statement as a lie and accuse the dolphin of wrongdoing, not because its conduct was “undolphinly”—dolphins may be natural-born liars—but because lying is contrary to principles of rationality and this dolphin is a rational dolphin. Obligation, or moral necessity, must be based upon rationality, not upon human nature.
General rules of prudence, such as “Honesty is the best policy,” are not moral laws, since they appeal to some psychological motive, such as desire for profit and good standing in the community. Kant calls such rules “practical,” but since they lack the element of moral necessity, they cannot be moral laws. Should a man not wish to follow the best business policy, he still has a moral obligation to be honest.
Thus moral laws and their principles are essentially different from all principles of action which we derive from empirical knowledge. Moral philosophy must have an entirely pure foundation. Without using any anthropological knowledge of human nature, moral philosophy imposes a priori laws on man because he is a rational being. Yet experience helps us to sharpen our powers of judgment: we must know the situation in order to determine which law is applicable to it, and we must know ourselves in order to assist the laws in exerting their force upon our will and our actions. For even though we are able to conceive of pure practical reason, we are influenced by so many natural inclinations that we find it difficult to put these laws into everyday practice.
Kant makes here a distinction which is often overlooked by his critics. “Moral philosophy,” he says, “must have an entirely pure foundation” Note that he does not say that all of moral philosophy is pure a priori philosophy. As he already emphasized, the necessity inherent in moral obligation must find its ultimate basis in pure practical reason, but this does not mean that from a knowledge of this ultimate principle alone we can deduce our particular obligations. Yet this is how some philosophers interpret Kant and then accuse him of being too rigid, too formalistic. We must have the ultimate principle and the a priori laws of morality, but we need experience just as well, for we must know a particular empirical situation in order to know what must be done and which law applies to that situation. The moral law forbidding adultery does not apply to taking a final examination. A student found cheating on his examination could hardly absolve himself by proclaiming his innocence of adultery.
A metaphysics of morals then is truly necessary if we wish to find in reason the basis for the a priori principles of activity. But more importantly, so long as we lack a supreme norm for the correct application of these principles, morality itself is in danger of all forms of corruption. In order for an action to be morally good, it is not enough that it conform to the moral law: it must be done because the law commands it. If we act from some other motive, the conformity is merely accidental. Our nonmoral motive may, in this case, produce activity which happens to conform to law, but in other cases it might just as well lead to unlawful activity.
The history of ethics gives us clear examples of what can happen when morality is based on a natural inclination. Epicurus' basic principle is that pleasure is the ultimate goal of all human endeavor. To a moral philosopher, this principle may serve as a foundation for a respectable moral system. Epicurus warns: “Since pleasure is the first good and natural to us, for this very reason we do not choose every pleasure, but sometimes pass over many pleasures, when greater discomfort accrues to us as the result of them.” And he continues: “When we maintain that pleasure is the end, we do not mean the pleasures of profligates and those that consist in sensuality . . . but freedom from pain in the body and trouble in the mind.”2 Whatever leads to these modified pleasures is good, while what causes pain is bad.3 As it is presented to us, this is an admirable moral system and whoever follows carefully the dictates of the Master will live a moral life indeed. But our own word epicurean suggests only too well the decadence which can result from the search for a life of pleasure.
Since the distinction between acting in conformity with the law and acting for the sake of the law, “because the law commands it,” is the essential issue of Section I (Chapters 3-4), discussion of this point may be postponed until then.
Only in pure philosophy can we find the pure, genuine moral law (and in matters of practice, this is the basic issue). Unless we first build a metaphysics of morals, we will have no moral philosophy whatsoever. Since philosophy distinguishes rational from empirical knowledge, while everyday scientific knowledge confuses the one with the other, no theory which tangles a priori principles with principles of experience deserves the name of philosophy. Even less does such a mishmash deserve the name of moral philosophy, for it spoils the purity of morality itself and even confounds its own purposes.
Although Kant insists that we must have a separate study of pure moral philosophy, i.e., a metaphysics of morals, he does not intend to disparage everyday rational knowledge. On the contrary, he considers commonsense rational knowledge of moral duties to be an excellent starting point leading to a philosophical knowledge of moral laws. He emphasizes here that everyday rational knowledge is not itself the foundation. It is a confused knowledge, mixing the empirical with the rational. We may well begin with such knowledge since we must begin somewhere, but our first job will be to untangle the empirical from the rational elements. It is the a priori basis for this mixed knowledge of moral laws that we are looking for, not any particular system erected upon this commonsense knowledge.
Do not jump to the conclusion that the celebrated Wolff, in his Universal Practical Philosophy, has already done what we are talking about. No, this is an entirely new field. Wolff’s work was a universal practical philosophy, and so investigated willing in general along with all those activities and circumstances which relate to willing in general. But he did not examine specifically that will which is determined solely by a priori principles, without empirical motives. His work differs from a true metaphysics of morals in the same way that general logic differs from transcendental logic: general logic deals with the acts and rules of thinking in general, while transcendental logic investigates the acts and rules of pure thinking, that is, thinking which knows things purely a priori. A true metaphysics of morals will search for the idea and principles of a possibly pure will, and not the activities and circumstances of specifically human willing (which is derived mostly from psychology ).
Kant’s early philosophical training had been in the Wolffian tradition. Christian Wolff (1686-1754) had written a large and popular compendium of moral principles, entitled the Universal Practical Philosophy. Kant rejected Wolff’s theories because, as Kant says, Wolff’s ethical theories were based upon observations of how moral agents acted rather than upon pure practical philosophy independent of any reference to “specifically human willing.” In other words, Wolff made the very error Kant warns against: he mixed a rational analysis with an empirical inquiry, with no clear distinction between the two.
The fact that laws and duty are discussed in ordinary practical philosophy, even though incorrectly, carries no weight against my argument. Even here the authors make the same error of thinking: they fail to distinguish the motives derived a priori from reason itself (the only truly moral motive) from those motives which the understanding has established as universal ideals on the basis of compared experiences. They lump them all together, without noticing any difference in origin, and distinguish them merely on the basis of relative applicability. In this context they define obligation, yet this definition will be anything but adequate for morality. What can you expect from a philosophy which ignores whether the source of all the possible ideas of moral activity is a priori or only a posteriori?
Kant continues his criticism of Wolff’s approach. Only one motive will give our actions true moral value, he says, and this he calls the moral motive. Wolff, on the other hand, examined the motives from which men do in fact act; finding some to be universal, or almost so, he maintained that such motives are the basis of human, and hence moral, action. John Stuart Mill (18061873), the noted British Utilitarian, argued from a similar position. All men do act from the pleasure motive, Mill said, and concluded that morality must then be based upon the proper pursuit of pleasure.
Vigorously opposing such procedures, Kant insists that the moral motive can be determined only by a pure philosophy of practical reason, that is, a metaphysics of morals. He thus agreed with Hume, who half a century earlier had asserted that from an investigation of what is so we cannot directly conclude what ought to he so. Moral principles must ultimately be derived from some source other than a study of facts.
Although Kant has argued the need for a metaphysics of morals, that is, for a complete examination of the a priori principles of morality and how they are derived from reason alone, he does not intend to pursue such a lengthy investigation in the Foundation.
As early as 1767, fourteen years before the first Critique, Kant had contemplated writing a “Metaphysics of Morals.” Such a work, however, demanded an a priori foundation valid for speculative as well as practical reason. So Kant spent the years 1767-1781 working out his critical examination of pure speculative reason. Having published his results as the Critique of Pure Reason, he hoped then to begin work on the long delayed “Metaphysics of Morals.” But more and more he came to recognize that there had to be a correspondingly thorough critical examination of pure practical reason, even though he had included a sketch of such an examination in the first Critique. And so it went. Kant did finally publish his Metaphysics of Morals in 1797, thirty years after he first conceived it. Ironically, the “preliminaries” constitute his tremendous contribution to philosophical thought, whereas the Metaphysics of Morals receives relatively scant attention.4
Someday I will publish a metaphysics of morals, for which this Foundation is the first step. The only sure foundation for such a metaphysics of morals is a critical investigation of pure practical reason, just as the only sure foundation for a metaphysics of nature is the (already published) critical investigation of pure speculative reason.
However, on the one hand, a critical investigation of pure practical reason is not so urgently important as that of pure speculative reason, because human reason, even in minds of ordinary ability, can be directed to a high level of accuracy and completeness in moral thinking, whereas in pure theoretical thinking human reason invariably goes astray.
Still another reason for postponing the critique of practical reason in favor of the Foundation is that such a foundation is immediately necessary for our moral system, since we are always involved in moral situations. We may sit on the sidelines watching the play of speculative philosophers without being much affected by the outcome. But we cannot tolerate such noninvolvement in moral matters. We must have the rules now, and we demand that they be the right rules. Even though the Foundation presupposes a comprehensive critique of practical reason, such a critique need not be complete in order for the Foundation to have immediate practical value.
On the other hand, if a critical investigation of pure practical reason is to be complete, we must be able to prove that it is united to pure speculative reason by some common principle. Ultimately, reason is one; only its functions are different. Yet in order to prove this unity, I would have to include considerations outside the present context and these would confuse the reader. So I call this book Foundation for a Metaphysics of Morals rather than Critique of Pure Practical Reason.
Again, a metaphysics of morals, despite its distasteful name, might become quite popular and be revised for ordinary readers. Thus it is better to set the groundwork apart (along with its inescapable subtleties), allowing the later book to be simpler.
Speculating that his “Metaphysics of Morals” might become something of a best seller, Kant wished to prepare the way for such a popular work by treating the “subtleties” in the separate Foundation. For instance, in the Metaphsics of Morals, Kant distinguishes between Duties of Right and Duties of Virtue. But such a discussion must rest upon an analysis and justification of the objective status of duty in general, which Kant provides in the Foundation.
The Foundation, then, has one purpose only: to discover and justify the supreme principle of morality. Such a study is complete all by itself and ought to be carried out all by itself.
Undoubtedly my findings in this basic study (which has as yet received scant attention) would appear much clearer if the supreme principle were to be applied to all of morality; and the success of this application would be further proof of their correctness. However, I must forego this advantage for such a course would have little real value, even though it would satisfy me personally. The fact that a principle is easy to apply and seems to cover all situations is no guarantee that it is correct. On the contrary, we might become biased in favor of such a principle and so be prevented from pursuing our search and critique objectively, without considering consequences.
To be precise, Kant’s purpose is twofold: to discover the supreme principle and to justify its objective validity for the whole of morality. Sections I and II are devoted to the discovery; Section III is his attempt to justify the principle as objectively valid.
I have adopted a method which seems to me most appropriate for our purpose. First, we will analyze ordinary moral knowledge in order to determine its supreme principle; then, knowing the principle and its source, we will reapply it to ordinary moral knowledge. I have divided the work as follows :
Section I: Transition from ordinary rational knowledge of morality to philosophical knowledge.
Section II: Transition from popular moral philosophy to a metaphysics of morals.
Section III: Final step from a metaphysics of morals to a critique of pure practical reason.
There is a discrepancy between Kant’s explanation of his procedure and his three-part outline of the contents. He says that he will first analyze ordinary moral thinking in order to find the principle underlying such thought; then, having found this principle, he will consider whether ordinary moral thinking is consistent with this principle. Yet the three sections of the Foundation do not correspond with this plan. Sections I and II are both analytical but Section III is a critical examination of freedom, not applied ethics. Kant never returns to ordinary moral thinking (except in his four examples in Section II, which hardly comprise a major section of the book). At the close of Section II, at that point where he had proposed to reverse his procedure, he apparently changed his mind. He says: “If we are to prove that morality is no mere fantasy ... we must show that a synthetical use of pure practical reason is possible. However, we cannot do this without first undertaking a critical investigation of this faculty of reason.” Consequently, in Section III he outlines the path to a full critique of practical reason. The synthetical return had to wait for the Metaphysics of Morals.
A possible explanation for this discrepancy is that Kant wrote the Preface before he completed the Foundation, adding the outline as an afterthought. Perhaps his first draft did include both the analytical and synthetical approach he mentions. Then, realizing that he could not complete his original project without first examining the limits of practical reason more thoroughly, he postponed the synthetical application to common moral thinking. In its place he examined freedom of the will as the “foundation” for a critical examination of pure practical reason. Having changed his outline, he naturally would have revised the title of Section III in the preface. The three titles given above do correctly summarize the contents of the three sections. We can imagine Kant revising the section titles in the preface, but through some oversight failing to change his prose description of his procedure.
The titles for the three sections sound somewhat technical, yet they are precise. Section I begins with ordinary rational knowledge of morality. This knowledge is not “ordinary” in a man-in-the-street sense. In Kant’s day the prevailing moral philosophy was an adaptation of Ciceronean stoicism,5 and the moral propositions Kant uses are representative of this moral philosophy. Kant obviously believes them true, but he does not argue for them. He is interested in finding the underlying principle that presumably would justify any other consistent set of moral propositions.
Section II begins with an analysis of the concepts of law and the moral imperative. Through these concepts—taken from “popular moral philosophy”—he again advances to the supreme principle. Then he outlines some ways in which this principle may be applied to particular moral situations, providing us with a vest pocket “metaphysics of morals.”
In Section III Kant attempts to justify the objective validity of the supreme principle. He offers a critical examination in capsule form of the limits of moral knowledge and of the autonomy of the will. This section contains some of the most compact writing Kant ever produced and it will require careful attention. Yet, paradoxically, its very compactness forced Kant to a clarity of expression which we might wish were characteristic of his other writings.
The reader would do well at this point to pause and reread the text of the Foundation without the commentary. A philosophical commentary of this kind has the disadvantage of breaking the text up into bits and pieces in order to analyze and discuss issues in detail. But the text itself should afterwards be read straight through in order to obtain a full view of the argument as the author presented it. Such a rereading is strongly recommended at the end of each chapter.