PHENOMENOLOGY

"Phenomenology" is a term that has been used in as many widely varying senses in modern philosophy as has the term that names the subject matter of this science, "phenomena."

Johann Heinrich Lambert, a German philosopher contemporary with Immanuel Kant, first spoke of a discipline that he called "phenomenology" in his Neues Organon (Leipzig, 1764). He took "phenomenon" to refer to the illusory features of human experience and hence defined phenomenology as the "theory of illusion." Kant himself used "phenomenology" only twice, but he gave a new and broader sense to "phenomenon" that, in turn, resulted in a redefinition of "phenomenology." Kant distinguished objects and events as they appear in our experience from objects and events as they are in themselves, independently of the forms imposed on them by our cognitive faculties. The former he called "phenomena"; the latter, "noumena," or "things-in-themselves." All we can ever know, Kant thought, are phenomena.

The next generation of philosophers, notably G. W. F. Hegel, was at great pains to show that this was a mistake. Hegel's first major work, Phenomenology of the Spirit (1807), traced the development of Spirit (or Mind) through various stages, in which it apprehends itself as phenomenon, to the point of full development, where it is aware of itself as it is in itself—as noumenon. Phenomenology is the science in which we come to know mind as it is in itself through the study of the ways in which it appears to us.

In the middle of the nineteenth century, the definition of "phenomenon" was further extended until it became synonymous with "fact" or "whatever is observed to be the case." As a consequence, "phenomenology" acquired the meaning that it possesses most frequently in contemporary uses—a purely descriptive study of any given subject matter. In this sense, Sir William Hamilton, in his Lectures on Metaphysics (1858), spoke of phenomenology as a purely descriptive study of mind. Similar was Eduard von Hartmann's use of the word in the title of his book Phenomenology of Moral Consciousness (1878), which had as its task a complete description of moral consciousness. When the American philosopher C. S. Peirce used the term phenomenology, he had in mind not only a descriptive study of all that is observed to be real but also of whatever is before the mind—perceptions of the real, illusory perceptions, imaginations, or dreams. It was the task of phenomenology to develop a list of categories embracing whatever can be included in the widest possible meaning of "to be." Peirce introduced this sense of the term in 1902.

The changes described so far are all due to extensions of the meaning of "phenomenon," but phenomenology, the science of phenomena in these different senses, remained one field of study among others, having a relation to philosophy as a whole comparable to those of logic, ethics, and aesthetics. Frequently it was recommended as a descriptive study that was to precede any attempt to provide explanations of the phenomena. But since Edmund Husserl employed the term in the early 1900s, it has become the name of a way of doing philosophy—by using the phenomenological method. For the phenomenologists, who regard their method as the only correct way of proceeding in philosophy, phenomenology is therefore the best and perhaps the only legitimate way of philosophizing today. For other philosophers, phenomenology is one school or movement in philosophy today. At the same time, however, the older sense of the term persists. "Phenomenology" is therefore used in two distinct senses. In its wider sense it refers to any descriptive study of a given subject. In the narrower sense it is the name of a philosophical movement. This entry will deal with phenomenology in the second sense.

The Movement and Its Origins

"Phenomenology" became the name of a school of philosophy whose first members were found in several German universities in the years before World War I, notably at Göttingen and Munich. Between 1913 and 1930 this group published a series of volumes of phenomenological studies titled Jahrbuch für Philosophie und phänomenologische Forschung, whose editor in chief was Husserl, the most original and most influential thinker of the group. Most of the better-known members of the phenomenological movement—Moritz Geiger, Alexander Pfänder, Max Scheler, and Oscar Becker—were coeditors, at least for a time. Martin Heidegger was another coeditor, but he cannot be counted among the phenomenologists without serious qualifications. Other major figures in the movement were Adolf Reinach and Hedwig Conrad-Martius.

The contributions to the Jahrbuch ranged from Husserl's writings about the foundations of phenomenology, to essays in the philosophy of mind and Scheler's major work on ethics, to pieces on the nature of analytic judgments and the paradoxes in set theory. As the interests of the various phenomenologists differed, so did their conceptions of phenomenology. These disagreements emerged only gradually, as Husserl developed the theory of the phenomenological method further and encountered a progressively more critical reception among his fellow phenomenologists. At the outset, there was general agreement that phenomenology was to be descriptive and that it was to describe phenomena by means of direct awareness (Anschauung ). It is best to begin to clarify these terms by showing what they could, but do not, mean.

Description

The terms descriptive, phenomenon, and direct awareness all suggest that phenomenology is here used in its wider sense as a purely descriptive science of observable phenomena. But this wider sense of the term does not include what for the phenomenologists is the most important feature of phenomenology—that it is a nonempirical science. From the very beginnings of the phenomenological movement, when the conception of phenomenology was otherwise still quite vague, there was general agreement that phenomenology does not describe empirically observable matters of fact. Insisting on this, the early phenomenologists took a stand in opposition to philosophical views then in vogue.

Kant had distinguished three kinds of statements: empirical statements, statements true by definition (which he called "analytic"), and a third kind that he called "synthetic a priori." After being temporarily eclipsed by the German idealism of the early nineteenth century, Kant found many vigorous adherents in the later decades of that century. But there were also many philosophers who found Kant's account of the third type of statement—the statements that are neither empirical nor analytic—profoundly unsatisfactory and who, instead of attempting to supply an alternative account, rejected the tripartite classification altogether. This was done, for instance, by the German positivists Ernst Mach and Richard Avenarius, who insisted that there are no nonempirical statements that are not analytic. Of equal, if not greater, importance were those philosophers who regarded all statements as empirical. Analytic statements seemed to them clearly to rest on "the artful manipulation of language" (Mill's phrase), and they thought it therefore implausible that the statements of logic and/or mathematics should be analytic, that they should be true, and, more important, that they should be applicable to objects of everyday experience and science merely by virtue of an arbitrary choice of definitions. Accordingly, John Stuart Mill in England and Christoph Sigwart in Germany, among others, sought to show that statements in logic and mathematics are no less empirical than statements in the sciences.

In the case of logic, the most plausible argument for such a view begins with the observation that logic deals with correct and incorrect thinking. Thinking is a mental or psychological activity and must, therefore, be studied in psychology just as any other mental or psychological activity. It seems to follow, then, that logic is either a special field within empirical psychology or a practical discipline whose theoretical foundations are supplied by empirical psychology. In the former case, the relation of logic to psychology is comparable to that of learning theory or abnormal psychology to psychology as a whole. In the latter view, logic is related to psychology as surveying is to geometry or accounting to arithmetic.

opposition to psychologism

The phenomenologists were not the first to question the identification of logical with psychological statements—a view they called "psychologism." But while some other philosophers had approached the issue by distinguishing logic from psychology in terms of the distinction between theoretical and practical disciplines, the phenomenologists attacked the identification of logical with psychological statements on the grounds that the latter are empirical statements and the former are not. The most sustained and painstaking critique of psychologism is contained in the first volume of Husserl's Logische Untersuchungen (Logical Investigations; Halle, 1900–1901), and the arguments in that book served as a first rallying point for phenomenologists.

Husserl's attack on psychologism had a special edge to it because his Philosophie der Arithmetik (Philosophy of Arithmetic; Vol. I, Halle, 1891; the projected second volume was never published) had been a frankly psychologistic account of arithmetic. His change of heart was in part occasioned by a controversy with the German mathematician and philosopher of mathematics Gottlob Frege, in which Frege had insisted that a sharp line be drawn between psychological statements, on the one hand, and logical and/or mathematical ones, on the other.

Husserl devoted an entire book to the detailed examination and refutation of every variety of psychologistic doctrine, taking careful account of each view and trying to show its inadequacy. Underlying all his arguments, however, were a few general principles to which he appealed again and again in the course of his discussion:

1. Psychology deals with facts; therefore its statements are empirical. It has not, until now, produced any precise scientific laws, and its generalizations are vague. The rules of logic, on the other hand, are precise. Hence, psychological generalizations can neither be identical with logical laws nor be premises from which they may be derived.
2. Empirical statements are probable, at best, for there is always a real possibility that further evidence will show them to be false. Logical truths are necessary truths. A logical principle such as modus ponens ("Given that 'If p, then q ' is true and that 'p ' is true, 'q ' is true") is not probable; it is necessarily valid.
3. Closely connected with (1) and (2) is the argument that empirical generalizations rest on induction; they are derived from a number of individual cases. This is not true of logical rules.

Both (2) and (3) are supported by pointing out that where there is a conflict between a logical principle and an empirical generalization, the logical principle will always emerge victorious because necessary truth is not to be refuted by a probable statement and logical truth cannot be shown to be false by an inductive generalization.

• (4) The empirical generalizations of psychology produce, at best, causal laws, and logical principles are not causal laws. Premises and conclusions of an argument are not related as cause and effect; the truth of a conclusion is not the effect of the truth of the premises. Causal relations hold between events, and events happen at definite times in definite places. But the premises of an argument do not "happen," nor does the conclusion; they are either true or false. In a valid argument the truth of the conclusion "follows" from the premises; it is not the effect of events called premises.
• (5) Empirical laws imply matters of fact; logical rules do not. Since empirical laws are, presumably, derived from the observation of particulars, the existence of such particulars in some place and at some time can be inferred from the truth of the empirical law. Modus ponens, on the other hand, does not imply that there exists, in a particular place and at a particular time, a pair of statements of the form "If p, then q " and "p." Nor are any corresponding facts implied by any other logical law. This point is sometimes stated in a phrase, borrowed from Gottfried Wilhelm Leibniz, that empirical laws are true only for this actual world; logical laws are true "for all possible worlds."

The upshot of these arguments is that logical and empirical statements differ in kind. Logical statements are precise, necessarily true, and not derived inductively from particulars. They are, or give rise to, logical rules, not causal laws, and they do not imply matters of fact. Empirical statements, on the other hand, are vague, probably (but not necessarily) true, and based on inductive generalizations. They are, or give rise to, causal laws and imply the existence of matters of fact. Quite clearly, in the refutation of psychologism, the decisive argument, for Husserl, consisted in showing that there are two kinds of statements: empirical and nonempirical. Phenomenological statements are to be nonempirical.

To deny that phenomenological statements are empirical is to deny that their truth or falsity depends on sensory observation. But if not on sensory observation, on what does their truth depend? Some philosophers might be inclined to say that phenomenological statements are analytic. Insofar as only those statements are analytic that are true by virtue of explicit definition of terms, phenomenologists deny that their statements are analytic. We shall have abundant evidence that they are right in this, for phenomenological statements are not true by virtue of stipulation of meaning. But insofar as "analytic" is used in some other sense, it is not helpful either to assert or to deny that phenomenological statements are analytic; the meaning of the term analytic is much debated in contemporary philosophy and has therefore become extremely obscure. It is more profitable to ask the phenomenologists about the truth conditions of their statements. Their preliminary answer to this question consists in introducing the term phenomenon by saying that phenomenological statements are true if they accurately describe phenomena. This answer, however, remains merely a verbal maneuver unless phenomenon can be shown to have a clear and definite meaning.

Phenomena

We have seen that phenomenon is a technical philosophical term that different philosophers have used in very different senses. The phenomenologists sometimes say that "phenomenon" is their name for whatever appears to us in "immediate experience." By "immediate experience" they do not mean sensory observations that have not been interpreted or classified under general concepts ("raw sense data"). Like many other contemporary philosophers, the phenomenologists are not at all sure that there are for us any sensory observations that are not interpreted or classified under general concepts. The appeal to phenomena or to immediate experience is therefore not an appeal to simple, uninterpreted data of sensory experience. Furthermore, the appeal to phenomena does not presuppose the existence of a special class of objects called "phenomena." The phenomenologists do not claim to have discovered that besides all the kinds of entities found in this world (physical objects, thoughts, numbers, feelings, poems, etc.) there is one other class, phenomena. Any object is a phenomenon if looked at or considered in a particular way. This particular way of looking at all kinds of objects is recommended in the slogan "Zu den Sachen!"

Literally translated, this slogan means "To the things!" where "things" must be taken in the widest possible sense to embrace all possible kinds of objects. Like other slogans, moreover, this one gains its force from having more than one meaning. If a German says to someone, "Zur Sache!" he is exhorting him, as we would say, "to get down to business." "Zu den Sachen!" admonishes one to get down to the proper business of the philosopher by examining and describing all kinds of objects in the particular way that reveals them as phenomena.

This explication of "phenomenon" is, so far, circular. To clarify what is meant by that term, we must therefore explain what alternative ways of doing philosophy are excluded by telling us to examine and describe phenomena. We must explain the polemical import of the slogan "Zu den Sachen!" Once this is done, we must pursue the concept of phenomenon further by attempting to clarify the nature of the examination and description that shows all kinds of objects as phenomena.

opposition to reductionism

The polemical import of "Zu den Sachen!" is readily made clear. In it the phenomenologists expressed their opposition to all reductionism, or, as Reinach called them, "nothing-but philosophies." Such philosophies are couched in sentences like "Logical laws are nothing but psychological laws," "Moral laws are nothing but the expressions of the mores of a given society," and "Aesthetic judgments are nothing but expressions of personal taste." To oppose all views of this sort would seem dogmatic. Some "nothing-but" statements may be false, but perhaps others are true; and one would think that each would have to be examined on its merits rather than be rejected summarily as an example of reductionism. However, the phenomenologists did not attack these "nothing-but" views on the grounds that they are false but on the grounds that the philosophers who held them, held them for the wrong kinds of reasons.

Psychologism, which is just one example of reductionism, did not assert that logical laws are nothing but psychological laws in the light of a thorough examination of the nature of logical laws that proved that they are identical with psychological ones. If someone challenged the psychologistic philosopher's views, he was not invited to examine for himself the nature of logical laws and to discover that they did not differ from those in psychology. Instead, he was given an argument from which it followed that logical laws "must" be psychological ones. Psychologistic assertions about logical and psychological laws do not result from an examination of laws in logic and psychology but are the logical consequences of certain more general assumptions. These assumptions themselves are not examined but are taken as self-evident.

Reductionism as attacked by the phenomenologist is the outcome of accepting certain statements that have not been examined carefully. If the implications of these assumptions are shown to conflict with facts about the world, the reductionist does not, the phenomenologists say, reexamine his original assumptions. Instead, he redefines the terms used to describe the facts about the world in such a way that the contradictions between these descriptions of facts and the implications of the original assumptions disappear. The redefinitions necessitated by the conflict between assumptions and facts are expressed in the "nothing-but" statements.

Opposition to phenomenalism

An example of a specific reductionist view attacked by the phenomenologists will clarify the process. David Hume's empiricism was attacked for its phenomenalism, that is, for its view that physical objects, as well as human beings, are no more than collections of their observable properties. ("Phenomenology" must not be confused with "phenomenalism.") "Observable properties" in this context refers exclusively to sensory qualities like shape, color, sound, etc. This view of Hume's did not issue from a careful examination of the nature of physical objects. Instead, it was a product of his psychological theories about the origin and meaning of concepts and words. Hume held that all concepts are either derived directly from sensory experience or are complex collections of such concepts. He regarded it as a consequence of this view that all concepts refer either to sensory qualities like shape, color, and sound or to complex collections of these. He also thought that all nouns are the names of concepts. It follows from this that all nouns naming physical objects refer to concepts that can be completely analyzed into simple concepts referring to sensory qualities. Hence physical objects—what is named by physical object nouns—are no more than complex collections of sensory qualities. However, this view is not supported by a careful examination of physical objects themselves but follows from, and hence "must" be true in the light of, Hume's psychology and views on the meanings of words.

Opposition to psychological atomism

Another target of the antireductionist polemic was the then popular attempt by philosophical psychologists like Wilhelm Wundt to define consciousness as a set of contents—sensations, feelings, affects—on which operations—association and apperception—are performed. This view was not the product of careful examination and description of the series of phenomena that we call consciousness but was a logical consequence of more general assumptions about the world. It missed, the phenomenologists maintained, the essential characteristic of consciousness that they, following Franz Brentano, called "intentionality."

Opposition to scientism

Also objectionable was the so-called scientism of the positivists Mach and Avenarius. Scientism regarded scientific statements as premises in philosophical arguments such that the truth of statements in philosophy depends on the truth of scientific statements. This view was a direct consequence of two assumptions: that all statements are either empirical or analytic, and that all empirical statements are, at least ideally, statements in science. Given these assumptions, there is a choice between restricting philosophy to the practice of logic, in which statements are often thought to be analytic, or saying that philosophical truths are empirical. If we choose the latter alternative, philosophical statements "must" have scientific premises.

But this conclusion, phenomenologists held, was drawn without paying careful attention to actual and possible functions of philosophy, which, they held, is independent of science. In this they were not motivated by any hostility toward science; on the contrary, their aim was to establish philosophy as a "rigorous science" by means of the phenomenological method. Husserl had discussed this aim at some length in his article "Philosophie als strenge Wissenschaft" ("Philosophy as Rigorous Science," in Logos, Vol. I, 1910–1911, 289–341). This phenomenological and rigorously scientific philosophy was expected to provide the foundations for the existing sciences by providing clear explications of the concepts that the sciences use but do not themselves explicate. For instance, the definition of number, in which Reinach was interested, was considered a task for phenomenology. Husserl was concerned with clarifying epistemological terms such as meaning and truth. So conceived, phenomenology had to be independent of the existing sciences because it was to explicate the concepts and procedures presupposed by them. To consider philosophy a branch or subsidiary of existing science was one more example of "nothing-but" philosophy.

Presuppositionless inquiry

Here it must be asked whether philosophers must not make certain assumptions. We cannot, it would seem, show that all statements are true by reference to the truth of other statements; some we must merely assume to be true. But phenomenologists are unconvinced by this sort of argumentation. Statements in phenomenology are not true because certain other statements are true; they are true because they describe the phenomena correctly. In order to achieve true description, the phenomenologist must resist the temptation to make assumptions and, afterward, to define his terms in such a way as to make the descriptions of facts consistent with the assumptions and what must be inferred from them. The phenomenologist does not frame theories; he merely examines and then describes phenomena as they present themselves to his unprejudiced view. Having no theoretical commitment and only one practical one—to examine all phenomena carefully and to take none of them as familiar or understood until they have been carefully explicated and described—the phenomenologist says that his science is descriptive and that it is presuppositionless.

This obviously does not mean that at any given time the phenomenologist may not be operating with certain unexamined assumptions—this can always happen. The claim of presuppositionlessness expresses the resolution to eschew all unexamined assumptions and the belief that such assumptions are unnecessary; No statement must be taken as true without examination. Phenomenology does not need any true but unexamined premises; the truth of all its premises can be tested by examining the phenomena.

This sheds some light on the second, affirmative sense of the slogan "Zu den Sachen!"—an exhortation to examine phenomena and to make them the sole touchstone of the truth of philosophical statements. But the precise import of this exhortation remains unclear until the meaning of "phenomenon" has been explicated, so this is a pressing question. It is also a question fraught with particular difficulties. Phenomena, as was stated, are those aspects of objects of every kind that are revealed by a particular way of looking at objects. The phenomenal aspects of objects are not revealed by ordinary empirical observation but only by looking at them as phenomena. The meaning of "phenomenon," on the other hand, cannot merely be stipulated in analytic statements. Hence, explications of "phenomenon" must result from using the phenomenological method and must be couched in phenomenological statements. But what these statements are cannot be made clear until it is clear what a phenomenon is, nor do we know what the phenomenological method is until we know what a phenomenon is.

"methodological circle"

The entire phenomenological enterprise is involved in a circle that can be called the "methodological circle." This methodological circle does not differ formally from the circle involved in any kind of logical investigation where the rules of inference, for instance, which the completed investigation hopes to formulate and justify must be employed during the course of the investigation itself so that its result, the logical rules, is the product of the application of the rules to themselves. The existence of this circle does not prove that logic is an impossible or unjustifiable discipline, nor does its presence in phenomenology support an analogous argument against it.

The occurrence of this circle should, however, put one on his guard against taking for completed analyses statements made by phenomenologists that are, in fact, merely gropings toward and anticipations of what phenomenology, its method, and the completed theory of method will be like in some indefinitely remote future. Phenomenology does not exist as a set of doctrines but at best as a method—and this method is to be developed by applying phenomenology to itself. Hence, even the phenomenological method is still in the process of being clarified, properly described, and elaborated; it is, at least to date, quite incomplete.

Husserl liked to refer to himself as a "perpetual beginner," an expression that meant several things to him. In one of its senses, it expressed what was just said about phenomenology: It is a method that can only be progressively developed by applying it to itself. Accordingly, most of Husserl's published works are discussions of the phenomenological method. This has sometimes been taken as a symptom of an excessive fondness for writing manifestoes, but discussions of phenomenological method are not of the nature of manifestoes prior to doing phenomenology, nor are they propaedeutics. Only while doing phenomenology can we clarify its method. To write about it was, in Husserl's case, to do phenomenology.

The Intuition of Essences

The preceding discussion has brought to light three properties of phenomenological statements:

1. Phenomenological statements are nonempirical.
2. Phenomenological statements are descriptive.
3. Phenomenological statements describe phenomena.

These leave the task of making clear what phenomena are, a matter of disagreement among phenomenologists: Most of the schisms within the phenomenological movement originate in disagreements about the set of conditions necessary for anything to be a phenomenon. We shall examine a variety of conditions proposed, beginning with the most simple and proceeding to more complex sets as the simpler ones turn out to be incomplete. The criterion of completeness for this set of necessary conditions is that any set of conditions required for anything to be a phenomenon must at least be consistent with the first requirement for phenomenological statements—that they be nonempirical. Hence, the set of conditions laid down for anything to be a phenomenon must clearly rule out any possibility that phenomena can be described in empirical statements.

The simplest specification of phenomenon, given by some early phenomenologists, contains only two conditions:

1. Phenomena are essences.
2. Phenomena are intuited.

The reason for identifying phenomena with essences is instructive. As we saw, it was claimed that there are some entities by virtue of which statements in phenomenology are said to be true or false. These entities (or phenomena) are not particular observable objects by reference to which empirical statements are confirmed or disconfirmed. Instead, the phenomenologists say, they are the necessary and invariant features of objects. Phenomenology explicates those features of any given object without which it could not truly be said to be the object that it is. These most general, necessary, and invariant features of objects have been called "essences" by other philosophers, and, following that terminological tradition, the phenomenologists also talk about essences.

Many philosophers in the past have held that statements about essences are empirical statements, arrived at by comparison of many examples of a type of object and extracting from the descriptions of all these examples the common features by means of some kind of generalization. Such a process has often been called abstraction. Abstract statements, since they are logically dependent on empirical descriptions of particular cases, are themselves empirical statements. Phenomenological statements, on the other hand, are, for the reasons given, not empirical statements. Hence, phenomenological statements are not reached by abstraction. They are, phenomenologists say, derived from a scrutiny of particular cases by seeing, intuition, or intuition of essences (Wesensschau ).

The identification of phenomena as essences brings us one step closer to the goal of clarifying the particular way of looking at objects that reveals objects as phenomena. It turns out to be a species of intuition. Phenomenology is a form of intuitionism and has, accordingly, acquired the ill repute of all intuitionisms of being no more than a veiled refusal to provide evidence for one's philosophical statements. But sometimes such a refusal can be justified. Intuitionism is objectionable only if the philosopher is not willing to argue either about the nature of his intuition or about the justification for appealing to it in this case—if his appeal to intuition is merely intended to terminate philosophical debate. The phenomenologists' appeal to intuition is not of this kind. Hence more can, and must, be said about intuition.

Intuition seems to be a psychological term. Its German counterpart, Anschauung, often means no more than "seeing." The objects of seeing, in its ordinary sense, are empirical objects. Essences are not empirical objects, so they cannot be seen in any ordinary sense of that term. Hence, intuition must be seeing of some extraordinary kind. One might suggest that the phenomenologists claim to have discovered one more human cognitive faculty than had been known before, but such a discovery of an actual human faculty would have to be couched in empirical statements. Phenomenologists do not make empirical statements, so they cannot claim—nor do they—to discover previously unknown cognitive faculties.

The point of introducing intuition is not psychological but epistemological. To appeal to intuition is not to make a psychological statement about the causal origins of certain statements but an epistemological one about the sort of evidence that will be relevant to them. To say that we know essences by intuition is to say, negatively, that the truth or falsity of statements about essences is not dependent on the truth about empirical statements.

The appeal to intuition makes another positive, epistemological point: Our acquaintance with essences possesses an epistemological feature also possessed by our sensory acquaintance with empirical objects. This logical feature is sometimes described by saying that what we see is described in self-validating statements. A statement, "P," about particular objects is self-validating if the strongest evidence that we can adduce for it is a statement like "I have seen that P " or "I have observed that P." We cannot, therefore, claim that "P " is true because there is some other true statement, "Q," from which "P " can be inferred and that is not equivalent to "P." Statements about essences are self-validating in the same sense. Given any statement, "E," of the form "________ is the essence of ________," we cannot claim that "E " is true because there is some other true statement, "F," which is not equivalent to "E " and from which we can infer "E." Of course, some statements about the existence of particular objects may be deducible from other statements, and it is similarly true that some statements about essences may be deducible from other statements. But such a deduction does not provide stronger evidence for statements about empirical existence or about essences than do self-validating statements.

Phenomenological statements are not derived by means of abstraction from particular statements, since, if they were so derived, they would not be self-validating. But they are not the only self-validating statements; empirical statements are also self-validating. An adequate account of phenomena must state more than that phenomena are revealed in the intuition of essences; it must specify this intuition to clarify in what respects it differs from the simple seeing of objects of sensory observation.

Bracketing Existence: Free Imaginative Variation of Examples

In the light of the problem about the meaning of intuition, the reason for introducing a further condition defining "phenomenon" becomes clearer. This condition is not accepted by all phenomenologists but was regarded as necessary by Husserl, Pfänder, Reinach, and Scheler. We are in a position, they said, to describe objects as phenomena only after we have "bracketed existence" or "suspended our belief in the existence of objects." Husserl calls this the "phenomenological epoche" or the "phenomenological reduction." Epoche was borrowed from the Skeptics, but Husserl's use of it differed from theirs.

These references to "bracketing" or "suspending belief in existence," together with the talk about essences, led to the view that phenomenology is a kind of essentialism and, as such, is diametrically opposed to existentialism. There is no room here to bring out all the confusions that produced this fairly common interpretation; suffice it to say that the phenomenological epoche is not achieved by resolving to make no more statements about existence or what exists. To bracket existence is not to eliminate existence in general or existing entities in particular from the list of possible objects for phenomenological study.

In the light of Husserl's repeated insistence on the close similarities between his phenomenology and René Descartes's methodical doubt, the phrase "suspending belief in the existence of objects" is often taken as a description of Cartesian doubt. But this is a misunderstanding, for Husserl insisted on distinguishing suspending belief in existence from doubting existence. This distinction cannot, therefore, be simply ignored.

Suppose a young woman states that she has direct evidence that she is terribly attractive to red-haired men. Her statement is not derived from a psychological law about the preferences of red-haired men or from a physiological one about their exceptional susceptibility to her figure and coloring. Her statement, a direct inductive generalization, is the result of her own experiences with red-haired men and tells us something about many or all of the members of the class of red-haired men. Besides all being red-haired and male, they have one further property: They cannot resist the charms of this young woman. In order to substantiate such a statement, she would have to cite cases of a number of red-haired men who at various times, under various circumstances, have given indubitable proof of their devotion. Two things are important here: that the red-haired men really exist and that their devotion to her is real. The truth of the inductive generalization depends at least on those two conditions. On the other hand, if the generalizations are correct, it follows that there exist (or existed) several red-haired men in this particular condition. If, however, the red-haired men do not exist or if their attachment is a figment of this young woman's imagination, then the general statement is false (unless evidence of a different kind can be found).

The story of this young woman was told in order to exemplify the relation of empirical generalizations to particular empirical statements—of "I am irresistible to red-haired men" to, for instance, "A red-haired matinee idol in New York committed suicide over me," and of both of these to the facts of the case. These relations were exemplified with an imaginary example, for it is quite unimportant that I do not know any young woman of this description. Where a description serves as an example in this sense (example is an ambiguous word), it is quite irrelevant whether the object described exists or not. If, on the contrary, I am interested in making a general statement about objects observed, it makes all the difference in the world whether the particular objects covered in my generalization exist and exist as described.

This is one sense of "bracketing existence." When existence has been bracketed in this sense, the descriptions of objects or situations do not serve as premises for an inductive generalization (or an abstraction), but as examples. But "example" is used in several senses. Sometimes it is used to designate one instance of an empirical generalization, but this is not the sense used here. At other times, examples serve a merely pedagogical function. I might have told my story about the young woman merely to provide a concrete illustration of abstract truth about empirical generalizations, in order to make the abstract statement easier to understand. In a third sense—"example" is used in phenomenology in this sense—the example both serves as an illustration and has evidential functions. In that case, the truth of the statement about empirical generalizations depends on the accuracy of the description of the example. I claim that my general statement is true because the description of the particular example is accurate, but how do I know whether a description is accurate so that it can have evidential force as an example? Since we have bracketed existence, I cannot say that the description is accurate because the case described has actually been observed to exist in a particular place and at a particular time, for examples need not be actual existents.

In order to understand this sense of bracketing existence, we must be able to answer two questions: (1) When can the description of an example rightly be said to be accurate? (2) How is a phenomenological statement to be derived from an example?

In this context Husserl talked about a procedure that he called "free imaginative variation," comparable to what Anglo American philosophers call the method of "counter-examples." Here we describe an example and then transform the description by adding or deleting one of the predicates contained in the description. With each addition or deletion, we ask whether the amended description can still be said to describe an example of the same kind of object as that which the example originally described was said to exemplify. Sometimes we shall have to say that if we add this predicate to the description or take that one away, what is then described is an example of a different kind of object from that exemplified by the original example. At other times the additions or deletions will not affect the essential features of the kind of object exemplified by the different examples.

In this way we discover the necessary and invariant features of a given kind of thing that the example must possess in order to be an example of that kind of thing. We also discover which features are accidental and hence irrelevant to the question whether this object, as described, is or is not an example of a certain kind of thing. What we discover is what phenomenologists call the "essence" of objects.

For example, let us suppose that we meet someone who does not have the usual five senses but only three: sight, touch, and hearing. We might be perplexed, but we should still call him a person. The same would hold if he had three more senses than normal persons. But suppose we met someone who looked like a person but seemed to be deaf and blind, and without any tactile, olfactory, or gustatory sensations. He would still be regarded as a person, although as a seriously defective one. But suppose further that we find that this creature looks like a human being except that it has no sense organs at all. Would he nonetheless be called a person? No. An animal? No. A plant? Not really. We have no word in our language for such a being. We would not know what to say about it.

Here we have varied in imagination an example of a person with reference to one predicate, "possessing sense organs." We find that in order for anything to be a person, it must have sense organs of some kind; there is an essential (necessary and invariant) relation between "person" and "possessing sense organs." The results of free imaginative variation are statements of such essential connections. Since statements about phenomena are one kind of statement about essences (and vice versa), the statements resulting from this procedure are phenomenological statements.

"epistemological circle."

Phenomenological statements are made while existence is and remains bracketed. If true, they are so not because they describe something that we have directly observed. Nor are they true because they are warranted by a series of observations of particular objects or events. Hence, they do not imply the past or present existence of particular objects in just the way in which empirical generalizations imply it. All that is asserted in the phenomenological statement is that if any being is an example of a person, then it must have sense organs. We are, therefore, making an assertion about the necessary relations of properties: Whatever has the property of being a person must also possess the property of having sense organs.

This is the method of free imaginative variation. It would seem to provide an answer to the second question raised earlier—how a phenomenological statement is to be derived from an example. But the same procedure can also be said to provide an answer to the first question, how we decide whether an example is described accurately—whether the description contains all the essential predicates so that the thing described may rightly be said to be an example of a certain kind of object. For, once we have made clear the invariant features of the sort of thing exemplified, we are in a position to say whether the example contains all those necessary features. But to use free imaginative variation to answer both questions is, of course, circular; we derive the phenomenological statement from any given example by means of free imaginative variation and then confirm that the original example was accurately described because it possesses the invariant features expressed in the phenomenological statement. It would seem that we need an independent criterion for deciding the accuracy of the description of any given example, but there is no discussion of such an independent test in the writings of the phenomenologists. The phenomenological method appears, therefore, to be circular in a second sense that might be called the "epistemological circle."

Phenomenology, as we saw, is circular because it clarifies its own method while using it (the methodological circle); it is also circular, we see now, because it confirms its statements by reference to examples and then attests the accuracy of the descriptions of these examples by reference to the statements derived from them (the epistemological circle). We must now show that what we claimed earlier for the methodological circle—that its presence cannot be construed as an argument against phenomenology—is true for the epistemological circle as well. This will be argued for by an examination of a second sense of "bracketing existence." In this second sense, "bracketing existence" refers to the transition from nonreflective to reflective thinking.

Bracketing Existence: Phenomenology and Reflection

In free imaginative variation we ask ourselves about any given property of an example, "Is this a necessary feature for being a such and such? Is that?" For our answer we do not appeal to empirical observation. Neither do we give an answer simply by deciding to regard some particular feature as essential. We do not define our terms arbitrarily; instead, with each variation, we ask ourselves whether the example described could still be recognizable as an example of the same sort of thing as that exemplified before. We ask ourselves what features an object must have in order to be recognized as an example of a certain kind of object. What we discover are necessary conditions for recognizing a certain kind of thing.

But recognition presupposes previous acquaintance. I cannot recognize someone whom I meet for the first time, unless I have seen pictures of him or have been given his description or perhaps dreamed of him before. But if we can recognize only what we know already, then we must already know the necessary features of the objects that we are able to recognize. In that case, there would seem to be no need to bracket existence and to vary the examples freely in imagination in order to discover their essential features, since the entire procedure presupposes that we know these essential features all along.

The resolution of this difficulty comes when we consider that the word know has two radically different senses, which some English philosophers have called, respectively, "knowing how" and "knowing that." The latter refers to knowledge expressed in statements. To "know that" something is the case is to be able to put what is known into words. I can show that I know a person by describing his looks; however, it is of course also possible that I should know a person and yet be quite unable to give any sort of adequate description of his looks. It is often very difficult to give a good description of those persons whom we know very well. I know them, not in the sense that I can describe them but that I could recognize them anywhere. I can pick them out of a crowd without hesitation. I can identify them by their voice or their walk, although I might be hard put either to describe in words or to imitate them. This second kind of knowledge is "knowing how"; in the example, I know how to recognize a person.

These two kinds of knowledge are independent of each other. It is not a necessary condition for being able to do something, such as recognize someone, that I should be able to say that he is a person of a certain description. Conversely, it is not necessary that I should be able to do a certain kind of action, such as ride a bicycle, in order to be able to give a detailed and accurate description of riding a bicycle. It is, furthermore, possible that for certain kinds of knowing how there is no corresponding knowledge that.

Of some performances I can say: This time I did it right; last time I did it badly. Therefore, I possess criteria for proper performance. If asked what these criteria are, I may not be able to put them into words, but I know them in the sense that I use them and, in many cases, I can, upon reflection, state what they are. I have then, by means of reflection, produced knowledge that ________ corresponding to the knowledge how ________ which I possessed all along. This is what happens when I vary an example freely in imagination: I am always able to discriminate between the thing that I would recognize as a certain object and the thing that I would either take as a different kind of object or about which I would not know what to say. But only upon reflection can I verbalize the criteria implicit in such a recognition by stating the essential features of any given kind of object.

reflective thinking

When I vary examples freely in imagination, I reflect about the criteria implicit in my ability to recognize examples of the given sort of object; I now put into words the criteria that previously were merely implicit in my performances. This description of the two sides of the process called "bracketing existence" accords perfectly with Husserl's explanations of it. Phenomenology, he stated, is a reflective enterprise. In its reflection it brings to light what was previously "anonymous" or "latent" in our "performances" (Leistungen ). But phenomenological reflection is a very special kind of reflection. In phenomenology we do not reflect about facts ("Did I see right? Was that really Jones lying in the gutter?") or about specific actions ("Should I have lectured Jones on the evils of drink?"). Phenomenological reflection does not produce any factual statements, nor does it employ factual statements as premises or as the starting points of reflection. In phenomenology we reflect about examples, in the sense explained; the result of such a reflection is not a factual statement or an empirical generalization but a statement about the necessary conditions for any object's being an example of the sort of thing considered in our reflection.

"Bracketing existence" and the other phrases applied in this context are used ambiguously. Why did Husserl fail to distinguish these two senses? We have already uncovered one source of this ambiguity by showing that we can employ the method of free imaginative variation of an arbitrarily chosen example in order to clarify the essential feature of any object only if we reflect about the example. Hence, treating a given case merely as an example (bracketing in the first sense) presupposes that we have made the transition from nonreflective to reflective thinking (bracketing in the second sense). Although the two kinds of bracketing are distinct, they must occur together.

But there is a second source of the ambiguous use of all these phrases. "Bracketing existence" and "suspending our belief in the existence" of an object seem to be particularly apt in describing important features of the transition from nonreflective to reflective thinking. Reflection involves questioning—more specifically, questioning something that I believed before or regarded as properly done. When I reflect, I ask, "Was that really Jones in the gutter?" or "Should I have helped him up?" Such questioning requires awareness that there are questions to be asked in this situation and that they are not pointless. Before I can reflectively question my earlier belief that it was Jones whom I saw lying in the gutter, I must be open to the possibility that it was not Jones. Hence, as I begin to reflect, I suspend my belief in the existence of Jones in that condition in that place, or I put his existence in brackets. "Bracketing" in this sense means that I become aware of the possibility that something which I believed to exist does not exist as I thought it did, that a statement which I considered true is not, or that some act which I considered right when I did it might have been wrong. Once I have become aware of that possibility, I am ready to reflect.

The insight that phenomenological statements are the product of reflection resolves the methodological and the epistemological circles. The methodological circle arises because the method must be used to clarify what the method itself consists of. It seems, therefore, that we can use the method only if we know what it consists of, but we can know what it consists of only if we have already used it. Therefore it would seem that we can never get started. But since phenomenology is reflective, it does not presuppose knowledge that the phenomenological method consists of certain procedures; it only presupposes that we know how to use it (to reflect about the essential features of arbitrarily chosen examples), even if we cannot describe it. Such a description is not a necessary condition for using the method, so there is no problem here.

The epistemological circle is resolved in a similar manner. In the method of free imaginative variation, it seemed that we could know that a given phenomenological statement, "P," is true only if we know that the description, "E " of the corresponding example is accurate. But we can know that "E " is accurate only if we know that "P " is true. Hence, it would seem that we cannot know either that "P " is true or that "E " is accurate. But phenomenological reflection begins with my being able to recognize the example described in "E." I know that I describe the example accurately to the extent that I recognize the object in my description of it. Both the accuracy of "E " and the truth of "P " are tested by the criteria implicit in my ability to recognize the object. Hence, there is no difficulty in this case either.

nonempirical status of phenomena

In the search for a complete definition of phenomenon we have now discovered three conditions defining phenomena: (1) phenomena are essences, (2) phenomena are intuited, (3) phenomena are revealed by bracketing existence. The third requirement is twofold: Phenomena are known only upon reflection of a specific sort, namely, reflection about the essential features of arbitrarily chosen examples. Once again the question must be raised whether this definition of phenomena is complete. The criterion of completeness used earlier was that a definition of "phenomenon" is complete only if it is consistent with the first of the three requirements for phenomenological statements—that they are nonempirical. We must ask, therefore, whether phenomena as defined can be described in empirical statements or whether our definition has ruled out that possibility.

It may seem obvious that the definition of phenomenon is complete by this criterion because it seems impossible that phenomena as defined—as being revealed only by bracketing existence—could be described in empirical statements, for statements about phenomena are not statements about single, observed particulars or about series of such single, observed particulars. They are, rather, statements about the necessary relations between the properties of some example of a certain kind of thing in which we do not consider whether the description of our example refers to an actually existing object.

But can we really conclude from this fact, namely, that no observation of actually existent objects is consulted in phenomenological reflection, that the truth of phenomenological statements is independent of the truth of empirical observation statements? We must distinguish between the description of the process by which we arrive at phenomenological statements and the logical conditions that these statements must fulfill in order to be true. The former merely describes how I discover certain statements, but it reveals nothing about the truth conditions of my statements. It is said, for instance, that some Greek geometers discovered certain statements about plane figures by measuring and weighing actual plane figures of tin. They arrived at their statements by means of observations; they were able to make certain statements in geometry after observing actual physical objects, but their statements are no more empirically true (or false) than are the same statements when they appear as theorems in Euclid's Elements.

This example presents a case in which statements whose truth or falsity is independent of empirical observation are discovered through empirical observations. It is possible that statements about phenomena constitute a converse case where empirical statements are discovered without explicitly consulting observation of sensory particulars. For instance, it was stated in the preceding section that the phenomenologist does not necessarily consult actual observations when he describes phenomena; his example may be purely imaginary. But it is possible that the statements that he is thus able to make are nevertheless empirical statements. All that was said was that the making of a phenomenological statement is not immediately preceded by observations of existent objects.

Perhaps, however, this is not necessary, since we know the necessary conditions for anything to be an example of a certain kind of thing because we have observed examples of this kind of thing many times and have, as it were, performed an unconscious induction all along. If this is true, then phenomenological statements may still be empirical statements. That they are not empirical statements has not been proved by stating that they are not discovered by means of explicit and deliberate observation of existing objects. The description of "bracketing existence" and of the subsequent reflection has revealed something about the method of discovering statements in phenomenology, but it has not shown that the statement so discovered may not nevertheless be empirical in the sense of being either verifiable by reference to observations of particulars or confirmable or at least refutable by reference to such observations.

There is reason to suspect that the phenomenologists who required that existence be bracketed in phenomenology thought that this requirement assured them that the statements so discovered would not be empirical in any of the senses mentioned. But, as has been shown, they have no such assurance. Hence, they can have no assurance that what is discovered once we have bracketed existence is a phenomenon, in the sense of being the referent of a nonempirical statement. We need further argument to show that bracketing existence does reveal phenomena in the required sense, in all or at least in some cases. Some of the phenomenologists, notably Husserl, have brought forward a number of considerations that provide the arguments needed here. These considerations can best be approached by considering intentionality.

Intentionality

It was said earlier that reflection undertaken after we have bracketed existence yields, if successful, descriptions of activities that we perform with ease in everyday life but are not able at the same time to describe. Concerning such activities we also know when they miscarry, when they are performed incorrectly or in an improper context, or when someone mistakes such an activity for a different one. We possess criteria for correct and appropriate performance and identification of such activities but are, ordinarily, unable to formulate them. Reflection subject to bracketing of existence yields formulations of these criteria. The phenomenologists regarded all statements resulting from such reflection as nonempirical, but there is no ground for thinking that this is true. These phenomenologists also believed that all the activities that are reflectively described and clarified after bracketing existence are intentional activities. This view can also be shown to be open to objections, but from these two doubtful assertions we can extract a more defensible characterization of phenomena than the one reached so far. So far three necessary conditions for phenomena have been listed: (1) They must be essences that are (2) intuited (3) as the result of the exemplary reflection that requires bracketing existence. We now add a fourth condition, namely, that statements about phenomena must be limited to statements about intentional acts.

The noun intentionality does not refer to a thing (as does, for instance, sodality ) but to the state of an entity—the state of being intentional. Although Husserl used intentional in all kinds of contexts, in its primary sense it is an adjective modifying "act"; being intentional is a characteristic of acts. In this employment, "intentional" has an ordinary meaning as a synonym for "deliberate" or "done on purpose," and a philosophical meaning different from, although related to, its ordinary, nonphilosophical meaning. The philosophical use of the term dates back to scholastic philosophy. Later, it completely disappeared from the philosophical vocabulary until it was reintroduced in 1874 by the Austrian philosopher Franz Brentano. Husserl, a student of Brentano's, gives credit to Brentano for reintroducing intentionality into philosophical discussion but adds that intentionality became a fruitful philosophical concept only in phenomenology.

Intentional acts have four aspects, and there are four distinct questions we can ask about them. The sentence "Luther thought that the devil was in his cell" is the complete description of an intentional act. We can ask who is performing an intentional act, and the answer consists of a proper name ("Luther"). It could also be a personal pronoun ("I" or "we") or a definite description ("the father of the Reformation"). We can, in the second place, ask what this person is doing, and the answer will consist of the inflected form of a verb ("thinks," "thought"). The third question concerns the intentional object of the act, what the act is about. In the example, Luther is thinking about the devil. Finally, we can ask in what manner or under what description the intentional object is object of the act; in the example, what is Luther thinking about the devil? "The devil is in my cell."

The intentional act, having four elements, is a tetradic relation. So, for instance, is the relation described in the sentence "I place the book on the table." Here also there are four elements: the subject or agent (myself), my action (placing), what I place (the book), and the table on which I place it. There is, however, an important difference between the two cases. The second statement is false unless there is a table on which I place the book. If the statement as a whole is true, the final of the four terms in the tetradic relation must also exist. It would be self-contradictory to say "I place the book on the table … but there is no table."

We can therefore infer the existence of the table from the truth of the statement "I place the book on the table." This is not so in the case of intentional acts. If it is true that Luther thought that the devil was in his cell, it is not therefore true that the devil exists, let alone that he was in Luther's cell. Luther might have had hallucinations; he might have been the victim of religious madness; or he might have been drunk. All three of these are situations in which we are inclined to see things that are not there or to believe that things exist which in fact do not. Nor can we conclude from the truth of the original sentence that the devil does not exist or was not in Luther's cell. The same holds of whatever is thought or believed to be the case. A belief that my wallet was stolen or that there are leprechauns does not allow the inference that there was a thief who stole my wallet or that there are leprechauns. The same is true of perceiving, of hoping, expecting, doubting, fearing, and all similar activities. The truth of a statement describing someone's intentional act does not allow the inference of either the existence or the nonexistence of what the act is about. This distinguishes intentional acts and their four elements from genuine tetradic relations, where the existence of all four elements can be inferred from the truth of a description of the relation.

the noninference criterion

The usual discussion of Husserl's doctrine of intentionality presents intentionality as (1) the defining characteristic of consciousness in the ordinary sense of that term, which (2) consists in the fact that all consciousness is consciousness of something. The first point is false; the latter is true but trivial. It merely asserts that to be conscious is to be related to something. But I am also related to something if, for instance, I own property. In that case I am the owner of something. But being the owner of something is not an intentional act because the existence of the object owned can be inferred from the fact that I own it. The existence or nonexistence of the object of the intentional act, however, cannot be inferred from the true description of that act. (We shall call this the "noninference criterion"). This, rather than merely being related to an object, is the property of intentional acts that distinguishes them from all other kinds of tetradic relations. Hence, it is a defining characteristic.

Two examples will show that intentionality is not the exclusive property of consciousness. Consider the sentences "Luther threw an inkwell in order to injure the devil" and "The rat pushes the lever in order to obtain food." Both sentences express tetradic relations: the agent (Luther, the rat), what he does (throwing, pushing), what he does it with (the inkwell, the lever), and the object of the activity (injuring the devil, obtaining food). It may be said that these are not intentional acts because the object in each case is not what the act is about but is, rather, an aim or a purpose. The acts described in these two sentences are intentional in the ordinary sense of being purposive, but according to the noninference criterion, they are also intentional in the philosophical sense because we cannot infer from the first sentence that the devil was injured and hence we cannot infer that the devil exists or does not exist, nor can we infer from the second that food was obtained by pushing the lever.

The acts described in the two sentences are not acts of consciousness or mental acts in the traditional sense. Throwing and pushing have traditionally been regarded as physical acts, but they differ according to the purpose served. When throwing something at a person in order to injure, one throws differently (much harder, for instance) than when one throws someone a cigarette in order to be helpful. Although physical, both of these acts are intentional in the philosophical sense. Hence intentionality is not, as Brentano thought and Husserl thought at certain times, the defining characteristic of consciousness in the ordinary sense. Husserl became aware of this and redefined "consciousness," in his later writings, by extending the term beyond its ordinary meaning to apply not only to mental acts but also to all kinds of activities, even to those usually regarded as physical, as long as they are intentional. Here intentionality became the defining characteristic of consciousness because this was how consciousness was defined. Husserl would perhaps not have wanted to apply "consciousness" to the behavior of animals, but his views on this point are not well known.

Inference

The verb "to infer" is used in a variety of senses in English, so it must be made clear in what sense it is used in the formulation of the noninference criterion. Suppose I see my foot as it sticks out unshod from my trouser leg and I say, "There's my foot." If someone asks me why I think that my foot is there (exists), I answer, "Because I see it" (or "Because I see something that looks like my foot"). In a loose sense of infer, I may be said to infer the existence of my foot from the fact that I see it. In this sense of infer, therefore, the correct description of an intentional act ("I see what looks like my foot") allows me to infer the existence of what I see (my foot). But this is inference in a loose sense. The conclusion does not follow necessarily from the premises. It is possible that the premise should be true and the conclusion false, as happens, for instance, when I am having hallucinations. There I see what looks like my foot, but the foot is not there. Common examples of this are the so-called phantom feelings—an amputee feels his foot long after it has been amputated. It is true that he feels his foot, but it is false that his foot is there. But if I say that I know my foot is under the table because I feel it, the inference (in the loose sense) is correct.

The sense of "to infer" used in the noninference criterion is stricter. In this sense we say that something is inferred from a premise or set of premises if the falsity of the conclusion is incompatible with the truth of the premise(s). In this sense it was said earlier that we can infer from the truth of "I place the book on the table" that there is a table. It would be self-contradictory to say "I place the book on the table … but there is no table" and to claim that both parts of this compound statement are true. It is in this stringent sense of "to infer" that the noninference criterion denies that we can infer the existence of the object of the intentional act from a true description of the act itself. The noninference criterion does not deny that feeling my foot, for instance, is often sufficient ground for saying that my foot is there. But it does deny that my foot must exist necessarily if I feel it. Intentional acts differ from other tetradic relations in that it is not inconsistent in the case of intentional acts to deny the existence of the final term of the four-term relation and to assert that the relation is described truly, but it is inconsistent to do this in the case of all nonintentional four-term relations.

Criterion is nonempirical

It is now easy to show that a statement of the noninference criterion is a nonempirical statement in the sense that no empirical statement can show it to be false. In this sense mathematical statements are nonempirical—no measurement of angles or lines in a triangle can show that geometrical statements about triangles are false. If there does appear to be a conflict between actual measurements and measurements predicted on the basis of certain geometrical propositions, we do not reject the geometrical proposition underlying our prediction; rather, we conclude that the measurements are false. The reason for this is, of course, that the procedures used in measuring presuppose the truth of the pertinent statements in geometry. In order to show that the statement of the noninference criterion is false, there must be at least one intentional act in which the existence of what the act is about or aims at follows with necessity from a true description of the act. But philosophers agree that no necessary relations are observed, or can be inferred from observations, so no statement about a necessary relation can be an empirical statement. Hence, the case needed to refute the noninference criterion cannot be described in empirical statements. It follows that the statement of the noninference criterion, not being refutable by means of an empirical statement, is not itself an empirical statement.

intentionality as a phenomenon

The statement of the noninference criterion satisfies the fourth condition laid down for phenomena: It is a statement about intentional acts. It is easy to show that it also satisfies the other three conditions for phenomena: (1) The preceding analysis consisted of reflection subject to bracketing of existence. (2) It brought to light certain essential features of intentional acts. (3) The truth of the statements rests on intuition, in the sense discussed earlier. Intentionality is, therefore, not only one mark of phenomena but is also itself a phenomenon. It has also been shown that the description of this phenomenon contains at least one nonempirical statement, namely, the noninference criterion. There is, then, at least one statement about phenomena, as now defined, that is nonempirical. This suggests that the four conditions for phenomena constitute a complete definition. However, the four conditions for phenomena are not sufficient for a complete definition, so a fifth condition must be added—that, with respect to intentional acts, phenomena serve as criteria of coherence.

Criteria of Coherence

Intentional acts are of two kinds; they are either purposive or about something. Purposive acts may be said to be adequate to their intentional object if the means chosen accomplish their purpose. Acts that are about some intentional object may be said to be adequate if what is believed or asserted about an object is really true, if what is questioned is questionable, if what is doubted is doubtful. Whether a given purpose is pursued correctly by using certain means depends on the nature of the purpose and of the means chosen, and on the way the means are used. Whether Luther throws the inkwell correctly at the devil depends on the weight of the inkwell, the distance between him and the devil, and how he throws. There are correct and incorrect ways of throwing inkwells or anything else. Which ways are correct and which are not is a matter of empirical fact, to be discovered by empirical study. Hence, rules about correct performance of this kind of intentional act are empirical rules. Similarly, it is in many cases an empirical question whether my beliefs are true, whether what I question is questionable, or whether what I doubt is doubtful. It can be shown that at least some of these rules satisfy all four defining conditions for phenomena; hence, they can be regarded as statements about phenomena, as defined so far. This, in turn, shows that the four conditions laid down do not constitute a complete definition of "phenomenon," for phenomena, under this definition, are capable of being described in empirical statements. We need a fifth condition.

The following consideration will yield the required fifth condition for a complete definition of "phenomenon." Before we can ask whether any given intentional act is correctly performed—whether it is adequate to its intentional object—we must be certain that what we are asking about is a genuine intentional act. Since intentional acts have four elements—the subject (or agent), the action, the intentional object, and either the means used or what is asserted about the intentional object—we need certain rules to determine which subject can be combined with what actions, which intentional objects, and which means or assertions to form coherent intentional acts. Not just any member of each of these four classes of elements can be combined with any other to form a coherent and intelligible intentional act.

coherence and intelligibility

The meanings of "coherent" and "intelligible" are best indicated by examples of their opposites, intentional acts that are incoherent or unintelligible. Purposive acts are not coherent and not intelligible (they "make no sense"), for instance, where the action and the means used are inappropriate to the intentional object. Someone might have said to Luther that it made no sense to throw anything at the devil because the devil is not a person but merely a symbol of evil. Not being a person, the devil has no body—and hence no location—and therefore cannot be made the target of any physical missile. A different case of an incoherent purposive act is that in which the means are inappropriate to the action. "Killing a person with kindness" is a metaphorical expression precisely because it literally makes no sense; the means chosen for killing a person are utterly inappropriate. They are not inappropriate merely in the sense that someone might try to use kindness as a murder weapon and discover that it does not do the job. It is not at all clear how one would proceed literally to try to kill someone with kindness. "Killing a person with kindness" is therefore not an intelligible or coherent intentional act. Similar incoherences can be found in the other relations among the four elements of intentional acts.

Corresponding incoherences appear in intentional acts that are about something. If what I believe about something is utterly inappropriate to its intentional object, such as "The Pythagorean theorem is mellifluous and sweet-smelling," there is no way of telling or even of finding out whether the statement is true. Asserting this sentence is not an intelligible intentional act, and hence the assertion is neither true nor false. Similar incoherences can occur between the action (for instance, "I predict") and its intentional object (for instance, some past event) or what is being predicted (that something happened yesterday).

So far the notions of coherence and incoherence, of intelligibility and unintelligibility, have been exhibited within single intentional acts. Husserl pointed out that there is also coherence and intelligibility of series of acts.

Suppose that Luther, rage suffusing his face, threw an inkwell at the devil with all his might and the very next moment rushed up to him, saying, "My dear fellow, I am so sorry. How very clumsy of me. Here, let me help you." This would be very surprising because the first action seemed clearly intended to injure, the second to placate. The change between the two is unmistakable and can be described by saying that the second act has a different intentional object from the first. As juxtaposed, the two acts make no sense because they seem to be members of two incompatible series of acts. The first act seems part of a series intended to enrage or injure the devil, and the second seems part of a different series aimed at mollifying the devil. The first action clearly leads to the expectation of another angry action. The second one disappoints that expectation, so the two actions make no sense, although each by itself makes sense. As single acts they are intelligible or coherent, but they do not make sense when they come in the above order. No one can understand what Luther is up to. We know what a man is up to if we understand a sequence of his actions and have correct expectations about what he is going to do next. If our expectations are disappointed, we may conclude that the agent has changed his mind or that we did not understand him to begin with. We understand or do not understand what someone is up to if his purposive actions form a coherent or incoherent series, respectively.

All this is true irrespective of whether the series of acts is performed well or badly. Hence, there are two sets of rules governing series of acts that correspond to the two sorts of rules governing individual acts: those which govern the coherence of act series and those which govern the adequacy of the act series to its collective purpose. What a man is up to in a series of acts can be inferred only from the sequence of acts performed. But not all sequences of acts are coherent. There are, therefore, rules about intentional acts determining the conditions for coherence of any series of intentional acts. Only if a series is coherent corresponding to the rules governing coherence can the question whether the actions and the means chosen are adequate to the aim pursued in the whole series be answered in the light of the relevant facts. Empirical statements about the adequacy of actions and means to their collective end are to be distinguished from statements about the coherence of such collections of acts.

It is not necessary to cite more examples to show that a series of acts which are about something are coherent or incoherent, intelligible or unintelligible, in analogous ways. A single act of belief, assertion, or questioning may be perfectly coherent and intelligible by itself but may be entirely out of context with what precedes or follows, and it is not understood what this person, in this act, is talking about, what he is trying to say.

horizon

Husserl used the term horizon to refer to the relations of coherence and incoherence of intentional acts. Horizon was not intended to refer to the place where sky and earth meet but to the edge of the perceptual field, which moves and changes with movements of the head or of the entire body. The horizon metaphor suggests that as the edge of the perceptual field (the horizon) leads us to expect a continuation of what lies before us, so any given intentional act suggests further acts that would be continuous or coherent with it. What is said in one act or done in one purposive action leads one to expect a second assertion or a second action continuous with the first. The second statement is continuous with the first if it is about the same object as the first; if in the second action one is up to the same thing as in the first, the two are continuous. I know what you are talking about or what you are up to when I know what sort of thing you will say or do next.

The horizon metaphor also implies that these relations between intentional acts are necessary conditions for any act being intentional, just as it is a necessary condition for the existence of a perceptual field that it have a horizon. Something is an act of asserting, for instance, if and only if I can repeat what I said in another way; if I can amplify, clarify, explain what I said; or if I can confuse, muddle up, and utterly obfuscate what my assertions are about. It is impossible that an intentional act should be without horizons, that is, unrelated to any other intentional act.

Criteria of coherence

As the horizons of the perceptual field are to some degree indefinite, so are the horizons of intentional acts. I cannot infer from any given assertion or activity of yours that you will next assert one particular statement or do some particular action and no other. When I see a church steeple on the horizon, I know that, when I come closer, I will not see a hippopotamus at its base. But there is definitely a point in coming closer to discover what the church or the building that resembled a church from a distance looks like.

Similarly, there is a point in listening to you to find out what your next statement is going to be or in watching what you are going to do. If I understand what you are talking about or what you are up to, I have some idea of what you are going to say or do next. I know the minimum conditions for your next statement and action; I know the limits beyond which your next action will not be continuous with the last or your next statement will not be about the same object as the last. Horizons are the necessary conditions for any series of assertions or activities to be intelligible. Different kinds of intentional acts have different kinds of horizons. Linguistic acts are related in terms of their meaning; purposive activities, by reference to the purpose. It is the task of phenomenology to clarify the different sorts of horizons (conditions for intelligibility) and to put into words what the horizons of individual examples of each kind of act are. Husserl called the clarification and formulation of horizons "intentional analysis." The results of such intentional analyses are statements of the criteria for the coherence of intentional act series.

Having understood what Husserl meant by "horizon" and that there are criteria for the coherence of single acts corresponding to the horizons in act series, we have found the fifth condition defining "phenomenon." Statements about phenomena must, besides satisfying the first four conditions, be about the criteria of coherence of single intentional acts or of sequences of intentional acts. When we look at any object as a phenomenon, we are trying to discover the criteria for coherence of those intentional acts in which the object (or its name or description) can figure.

Are Phenomenological Statements a Priori?

Traditionally philosophers have called statements "a priori" if they are (1) nonempirical and (2) necessarily true. Phenomenologists have always held that their statements are a priori. The two parts of this claim must be examined separately.

It has been shown that phenomenologists agree that their statements are nonempirical, although they disagree about the description of phenomena. Some phenomenologists were content to describe them as essences intuited, but others regarded this as insufficient and added that phenomenological descriptions must be preceded by bracketing existence. But bracketing existence also turned out to be an inadequate guarantee that phenomenological statements are nonempirical. Therefore some members of the phenomenological movement, notably Husserl, added further requirements for statements about phenomena. The preceding discussion can be summarized by stating the five conditions that any statement must satisfy if it is to be a statement about phenomena:

1. It must be about essences.
2. It must be self-validating (intuitive).
3. It must be the result of bracketing existence.
4. It must be about intentional acts.
5. It must lay down the criteria of coherence (or intelligibility) of intentional acts.

We must now, once again, ask: Are statements of this kind nonempirical?

the senses of empirical

The above question is not easy to answer because the term empirical has several meanings. We must examine some of them.

Statements asserting particular matters of fact, such as "There is a fire burning in the fireplace," are true if observation shows them to be true and false if observation shows them to be false (for instance, that the fire has gone out). They are empirical because one observation will show them to be true or false.

General statements, such as "Continuous nervous tension produces high blood pressure," are neither confirmed nor refuted by one observation or even by a few observations but only by a series of carefully controlled observations. This case concerns generalizations about observable connections.

There is a further sense of "empirical" that applies to statements about objects which are in principle nonobservable, such as "ideal gases" or "perfectly elastic bodies." Such entities cannot be observed because they do not exist, and hence we cannot frame empirical statements about them in either the particular or the general sense of "empirical." These entities, which cannot be described in observation statements, are instead defined in a series of statements constituting a scientific theory. From such a theory statements can be deduced that can be tested by reference to direct experience. If observation shows the deduced statements to be false, we must reject the theory, and hence our theoretical statements about the unobservable entities are indirectly refuted by observation. These statements are therefore, in this indirect way, empirical because observations can serve to show them to be false.

phenomenological statements

Phenomenological statements, as described in the preceding sections, are not empirical in the first two senses of the term. They are not empirical in the first sense because they are never statements about individual existing intentional acts but only about the criteria governing types of acts; only particular statements are empirical in the first sense. Empirical in the second sense are generalizations derived by induction from a series of observations of particulars. Such inductive generalizations presuppose that we know what particulars belong to the class of objects to be observed. If we want to make a generalization about the relation between nervous tension and high blood pressure, we must have a very precise idea of what must count as examples of nervous tension and what blood pressure counts as "high" blood pressure. Similarly, we cannot inductively arrive at statements about intentional acts unless we are already able to differentiate a coherent intentional act from an incoherent collection of each of the four kinds of elements of intentional acts.

The same applies to generalizations about coherent series of intentional acts. Nothing said by the phenomenologists should exclude the possibility of framing empirical (in the general sense) statements about intentional acts. All that is argued is that the criteria of coherence of individual acts as well as of series of acts are presupposed and therefore are not established by such inductive generalizations. Therefore, statements formulating these criteria cannot themselves be empirical generalizations.

It is undoubtedly a task for phenomenology to differentiate the different senses of "empirical," that is, to describe the different kinds of intentional acts involved in what we call experience and the criteria of coherence belonging to each kind of act. Oddly enough, the phenomenologists so far have barely begun to undertake such an examination, and hence their conviction that statements about phenomena, as now defined, are nonempirical is not supported by adequate phenomenological analyses. This important shortcoming in the theory of the phenomenological method is all the more serious because there are good reasons for thinking that there is one perfectly good sense of the words experience and empirical in which statements about phenomena, as defined, are empirical.

empirical phenomena statements

In a scientific theory, the terms are defined in relation to one another in such a way that if we alter the definition of one term, the definitions of some of the other terms are also changed. The effect of such a set of interrelated definitions is to limit the contexts in which these terms may be applied. A set of phenomenological statements has a similar function; it limits the contexts in which given intentional acts may be performed. The limits imposed on these intentional acts in the phenomenological statements are interrelated as the definitions in a theory are. If we alter the limits of one intentional act, those of other acts are also changed. History and ethnology provide many examples of such changes.

Among the Trobriand Islanders, for instance, successful gardening requires the use of magic. Before seedlings are planted, a spell must be spoken over them. It is very important that the magician's mouth be as close as possible to the seedlings, for otherwise some of the power of the spell will be dissipated. The power of the spell resides not in the sound waves produced by the magician but in the meaning of the terms used, something that we would not regard as a physical phenomenon. Yet the power of the magical words is here treated as if it were a physical force that varies with the distance from the object it affects. It is clear that the Trobriander does not draw a distinction between the physical and the mental, so it makes perfect sense for him to say something that makes no sense to us—that the spell must be spoken as close as possible to the seedlings in order to be effective. He imposes different limits on his intentional acts—what makes literal sense to him is to us at best symptomatic of the confusions of the "primitive" mind—and these various limits are interrelated. We can formulate them in a set of phenomenological statements that we regard as false and he regards as true. This example shows the analogy between the limitations imposed on theoretical terms by their implicit definitions in a scientific theory and the mutual limitations imposed on intentional acts and expressed in phenomenological statements.

Statements in a scientific theory limit the application of the terms. If the limits imposed allow the use of the terms in false factual statements, these limits must be altered; the theory is invalid. In analogous ways phenomenological statements may be invalidated by experience. Phenomenological statements express the limits imposed on intentional acts, and if these limits are such that we cannot distinguish true factual statements from false ones, the limits must be altered; the phenomenological statements are invalid.

In order to make true generalizations about gardening and distinguish them from false ones, we need a clear notion of causation. Causal relations as discussed in science exist only between spatiotemporally contiguous events, and this implies that only spatiotemporally located events can be either causes or effects. A clear notion of causal relations, therefore, presupposes a clear distinction between events that are and those that are not spatiotemporally located, or between physical and mental events. Where such a distinction is not drawn, no clear understanding of causal relations is possible. The Trobriander does not differentiate physical events from mental events (and forces); hence he cannot clearly differentiate causal relations from noncausal relations. As a result, he cannot make general statements about gardening that are always true or always false as tested by the information available to us. They may, of course, be always true (or false) tested by what he knows. His generalizations are about classes containing very heterogeneous types of relations, both causal and noncausal. Statements about the causal are true under very different conditions from statements about the noncausal, so his generalizations are sometimes true and sometimes false, and he does not have the vocabulary necessary to reformulate them in such a way that they are always true or always false. This shows that the Trobriander's lack of scientific information about biology is not accidental. It is impossible for him to do natural science because his language lacks the requisite distinctions. Scientific statements cannot be made in his language, which is clear proof that it is inadequate and that the phenomenological statements describing his linguistic acts as well as the nonlinguistic ones, such as those associated with garden magic, are therefore invalid.

This argument as stated is not conclusive, but it can be strengthened to make a rather formidable case for holding that the phenomenologists are mistaken in their claim that their statements about phenomena are nonempirical in all senses of that term. This conclusion shows that the question asked at the very outset—what are the truth conditions of phenomenological statements—remains unanswered. In the preceding a good deal has been said about these truth conditions, but it has been shown that that answer is incomplete. The phenomenologists' account of their method not only lacks a complete theory of experience in its different forms but also a complete theory of truth, at least as that term applies to the statements in phenomenology.

the senses of "necessary"

The second aspect of a priori statements is their necessity. A priori statements are necessary because they are nonempirical; if they are true at all, they are true independently of facts about the world. Even if all the statements about this world that are now true were false, and if, therefore, our world were very different from what it is now the a priori statements would still be true. They are true whatever happens to be the case in the world. Hence we may say that, if true at all, they must be true regardless of any facts. For this reason the term necessary has often been explicated as "true for all possible worlds." A different world from ours is one whose description requires factual statements to be true that are false of our world. Since necessary statements are true whatever factual statements may or may not be true, they are true for all possible worlds. A statement is necessary, therefore, to the extent that its truth is logically independent of the truth or falsity of empirical statements. It follows that there are different senses of "necessity" to correspond to the different senses of "empirical." There are, therefore, also different senses of "a priori." Hence, phenomenological statements are clearly a priori insofar as they are not empirical in the first two senses of that term. But phenomenological statements are empirical in a third sense and are therefore not a priori in that sense of "a priori" that contrasts with this third sense of "empirical."

necessary phenomena statements

In the sense explained, statements are necessary if they are true necessarily. But if statements about phenomena are a priori—necessarily true and nonempirical—they are necessary in a second sense: Their truth is a necessary condition for any empirical statement to be capable of being either true or false. An empirical statement can be either true or false only if it is meaningful, and that depends on the coherence of the intentional act and of the intentional act series in which it is asserted. But as was seen, the coherence of such acts and act series is presupposed by any question about the adequacy of intentional acts to their intentional objects. Hence the statements that lay down the criteria for coherence of all kinds of intentional acts, including acts of asserting, must be true if we are to be able to decide whether any given intentional act is adequate to its intentional object—for instance, whether an assertion is true or a purposive action is successful. Insofar as phenomenological statements are a priori, they are, therefore, necessary in this second sense; they are presuppositions for the adequacy or inadequacy of any intentional act to its intentional object. The truth of phenomenological statements is logically prior to the truth or falsity of all empirical statements and to the correctness of all purposive actions.

Contemporary Phenomenology

Political events in Europe and the shifting winds of doctrine caused the phenomenological movement to lose much of its original momentum after Husserl's death in 1938. The best-known twentieth-century philosophers who used the term phenomenology in descriptions of their own work were Martin Heidegger in Germany and Jean-Paul Sartre and Maurice Merleau-Ponty in France. All three used the term phenomenology in appreciably different senses from the phenomenologists previously discussed.

heidegger

Heidegger was a student of Husserl's and at one time was a coeditor of the Jahrbuch. In that journal (Vol. 8, 1927) appeared his first major work, Sein und Zeit (translated by J. Macquarrie and E. Robinson as Being and Time, New York, 1962). The phenomenologists so far discussed all agreed that it is the task of phenomenology reflectively to bring to light the criteria implicit in the intentional acts we perform in everyday life, in which we act in, get to know about, and learn to master that everyday world which Husserl christened the Lebenswelt ("world in which we live"). The emphasis here is on putting into words what is commonly and familiarly done without one's knowing how to describe accurately what he is doing. Heidegger also regarded phenomenology as a sort of reflection but not a reflection designed to put into words what is familiar in performance.

On the contrary, Heidegger's brand of phenomenology tried to open the way back to what had, he thought, become completely unfamiliar, what he calls Sein (being). He recognized that "being" had become a philosophically empty word. Hence we cannot gain a better understanding of being by reflecting only about the world insofar as it is familiar to us, for in that world "being" has become almost meaningless; there are very few contexts in which it makes sense to talk about "being." Thus, reflection about the criteria of intelligibility, which we use now, will not reveal much about being. Rather than reflect on these criteria, Heidegger proposes to ask why "being" has become almost meaningless to us. But since a question is intelligible only to the extent that we can specify the sort of answer we expect, and since an answer to Heidegger's question would require a language in which "being" is meaningful, even an intelligible formulation of his question involves him in the attempt to re-create a very different language, in which "being," far from being an empty word, is the richest and most important concept. This language, he believed, is the language used by the pre-Socratic philosophers. Heidegger's phenomenology thus led him into an enterprise utterly unfamiliar to the other phenomenologists, the attempt to develop a new philosophical language by re-creating that of the pre-Socratic philosophers.

sartre

Sartre's major work, L'être et le néant (Paris, 1943; translated by H. E. Barnes as Being and Nothingness, New York, 1957), bears the subtitle An Essay in Phenomenological Ontology. The work does not, however, contain any explicit discussion of phenomenology, nor did Sartre explain his conception of phenomenology at length in any other work. More than once he differentiated phenomenology from science by saying that phenomenology makes statements about essences; science, about facts. In one long essay, "La transcendence de l'égo" (Recherches Philosophiques, Vol. 6, 1936–1937; translated by F. Williams and R. Kirkpatrick and published in book form as The Transcendence of the Ego, New York, 1958), he takes sharp issue with Husserl's transcendental phenomenology, particularly with the claim that in phenomenology we discover that there is a transcendental ego.

It would seem, then, that Sartre was a phenomenologist who, like many others, adopts the descriptive approach to essences but refuses to follow Husserl in his later developments of the theory of the phenomenological method. But Sartre differs radically insofar as he was not averse to constructing philosophical theories. His major work is an example of constructive philosophy in precisely that sense in which phenomenologists attacked it in their polemic against reductionism. Sartre's conception of phenomenology is no clearer if we look at his actual practice of the method than if we consider his sparse statements about it. If Sartre practiced phenomenology at all, the term as used by him and as applicable to his procedures has a different meaning from the one explicated in this discussion.

merleau-ponty

Merleau-Ponty's major work bears the title Phénoménologie de la perception (Paris, 1945; translated by Colin Smith as Phenomenology of Perception, London, 1962). Unlike Sartre, he includes an introduction devoted to a clarification of "phenomenology." The clear and explicit result of this discussion is that Merleau-Ponty has interpreted the notion of phenomenology in a sense rather different from that subscribed to wholly or partly by members of the phenomenological movement, as well as from that used by either Heidegger or Sartre.

These three philosophers used "phenomenology" in appreciably different ways from those in which it has been used by the phenomenologists discussed. To be sure, there were also radical and profound disagreements among the latter about the nature and presuppositions of the phenomenological method, but they regarded these differences as different results arrived at by applying the same method. In this sense these philosophers—Husserl, Pfänder, Geiger, Becker, and Reinach, among others—can be regarded as belonging to one school of philosophy. All of them shared certain common views at the outset, and they believed that they were using the same method. But Heidegger, Sartre, and Merleau-Ponty began doing their respective brands of phenomenology by explaining what they considered phenomenology to be and how their conception differed from that of Husserl. They did not begin with the same common views, as did the earlier phenomenologists; and they did not regard their method as identical with that of Husserl and the other phenomenologists. For this reason they do not belong to the same school of philosophy.

See also Binswanger, Ludwig; Brentano, Franz; Existentialism; Existential Psychoanalysis; Heidegger, Martin; Intentionality; Life, Meaning and Value of; Psychologism; Sartre, Jean-Paul; Scheler, Max; Time, Consciousness of.

Bibliography

original works in german

Husserl, Edmund. Husserliana. Vol. I: Cartesianische Meditationen. The Hague: Nijhoff, 1950. The shortest, though not always easy, introduction to Husserl's mature conception of phenomenology.

Husserl, Edmund. Husserliana. Vol. VI: Die Krisis der europaischen Wissenschaften und die transcendentale Phänomenologie. The Hague: Nijhoff, 1954. A very late work of Husserl's; introduces the important notion of the Lebenswelt, the world of everyday life.

Husserl, Edmund. Husserliana. Vols. VII and VIII: Erste Philosophie. The Hague: Nijhoff, 1956–1959. Husserl's lectures on the history of philosophy and phenomenology; more accessible than many of his other writings.

Husserl, Edmund et al., eds. Jahrbuch für Philosophie und phänomenologische Forschung. 11 vols. Halle: Niemeyer, 1913–1930. Contains representative writings of all the major phenomenologists.

Reinach, Adolf. "Über Phänomenologie." In his Gesammelte Schriften, edited by Hedwig Conrad-Martius. Munich, 1921. A brief and very lucid statement of an early conception of phenomenology.

original works in translation

Husserl, Edmund. Cartesian Meditations. Translated by Dorion Cairns. The Hague: Nijhoff, 1960.

Husserl, Edmund. The Idea of Phenomenology. Translated by William P. Alston and George Nakhnikian. The Hague: Nijhoff, 1964. Translates a series of lectures given in 1907 in which the transcendental-phenomenological reduction is introduced.

Husserl, Edmund. Ideas. Translated by W. R. Boyce Gibson. New York: Macmillan, 1931; New York: Collier, 1963.

Husserl, Edmund. "Phenomenology." Translated by C. V. Solomon. In Encyclopaedia Britannica, 14th ed., Vol. XVII, 699–702. 1927.

Husserl, Edmund. Phenomenology and the Crisis of Philosophy: Philosophy as Science and Philosophy and the Crisis of European Man. Translated with notes and introduction by Quentin Lauer. New York: Harper, 1965. Contains a long essay, published in 1910, in which Husserl provides an expanded version of his earlier polemics against psychologism. This book also translates portions of Husserl's late work, Krisis.

books on phenomenology

Bachelard, Suzanne. La logique de Husserl. Paris, 1951. A lucid and detailed discussion of Husserl's Formale und Transcendentale Logik (1929).

Berger, Gaston. Le cogito dans la philosophie de Husserl. Paris: Aubier, 1941. One of the earliest and still one of the best monographs on one aspect of Husserl's thought.

Brand, Gerd. Welt, Ich und Zeit. The Hague: Nijhoff, 1955. Attempts a summary of Husserlian phenomenology on the basis of unpublished manuscripts. Rather general but often illuminating.

Brentano, F. Psychology from an Empirical Standpoint. Translated by A. C. Rancurello, D. B. Terrel, and L. L. McAlister; edited by L. L. McAlister. London: Routledge, 1969.

Cooper, D. E. Heidegger. London: Claridge Press, 1996.

Derrida, Jacques. Speech and Phenomena. Translated by Gayatri Chakravorty Spivak. Baltimore: Johns Hopkins Press, 1976.

Dreyfus, H. L. Being-in-the-World: A Commentary on Heidegger's Being and Time, Division I. Cambridge, MA: MIT Press, 1991.

Farber, Marvin, ed. Philosophical Essays in Memory of Edmund Husserl. Cambridge, MA: Harvard University Press for University of Buffalo, 1940. A collection of stimulating and sometimes informative essays on different aspects of Husserl's work.

Heidegger, M. Being and Time. Translated by J. Macquarrie and E. Robinson. London: SCM Press, 1962.

Merleau-Ponty, M. Phenomenology of Perception. Translated by C. Smith. London: Routledge and Kegan Paul, 1981.

Mohanty, J. N. Edmund Husserl's Theory of Meaning. The Hague: Nijhoff, 1964. An interesting discussion of Husserl's early work in the light of the treatment given the same questions by Bertrand Russell, G. E. Moore, and others.

Mohanty, J. N. Transcendental Phenomenology: An Analytic Account. Oxford: Blackwell, 1989.

Moran, D. Introduction to Phenomenology. London: Routledge, 2000.

Schrag, C. O. Experience and Being. Evanston, IL: Northwestern University Press, 1969.

Smith, B., and D. W. Smith, eds. The Cambridge Companion to Husserl. Cambridge, U.K.: Cambridge University Press, 1995.

Smith, D. W. Husserl and Intentionality: A Study of Mind, Meaning, and Language. Dordrecht: Reidel, 1982.

Sokolowski, R. Introduction to Phenomenology. Cambridge, U.K.: Cambridge University Press, 2000.

Spiegelberg, Herbert. The Phenomenological Movement. 2 vols. The Hague: Nijhoff, 1960; 3rd ed., The Hague: Kluwer Academic, 1982. Discusses phenomenology in general as well as individual contributors to the movement. Extremely informative.

articles on phenomenology

Ayer, A. J., and Charles Taylor. "Phenomenology and Linguistic Analysis." PAS, Supp. 33 (1959).

Boehm, Rudolf. "Basic Reflection on Husserl's Phenomenological Reduction." International Philosophical Quarterly 5 (1965): 183–202.

Downes, Chauncey. "On Husserl's Approach to Necessary Truth." Monist 19 (1965): 87–106.

Findlay, J. N. "Phenomenology." In Encyclopaedia Britannica, 1965 ed.

Ryle, Gilbert. "Phenomenology." PAS, Supp. 11 (1932).

Smith, D. W. "Phenomenology." Stanford Encyclopedia of Philosophy. Available from http://www.plato.stanford.com.

Spiegelberg, Herbert. "Toward a Phenomenology of Experience." American Philosophical Quarterly 1 (1964): 1–8.

Richard Schmitt (1967)

Bibliography updated by Benjamin Fiedor (2005)