1. Berkeley

In treating the indirect arguments for sensations I shall begin with the British empiricists and then move back to Descartes. The British arguments can be dealt with more briefly; and I wish to spend more time with Descartes because of his influence on Husserl and Phenomenology.

The British arguments are basically two in number, the one more articulate than the other. The less articulate argument was diffused in the new intellectual atmosphere and did not require to be brought to utterance. The new atmosphere is that of the sensation theory. Hume referred to it, typically, as a matter of the “slightest philosophy,” as trite as the complaints which in all ages have been raised against the senses. Everybody knows how dependent we are on our sense organs, how they vary from person to person, even from time to time with the same person. Everybody knows how all things sensory are thus “relative,” hence “subjective,” and that this subjectivity roots in our “sense impressions.” Plainly, the direct argument has done its work; it has nailed sensations firmly to the framework of modern thought and even infused them into some of our most familiar locutions.

What makes this atmosphere an indirect argument for sensations is the tendency to regard “relative” and “subjective” as though they were equivalent. This equivalence is distinctively “modern”; it did not obtain prior to Descartes. Although the relativity of the senses was well known before Descartes, it was only after him that this relativity came to signify also subjectivity. Thus to show that hot and cold are relative was for Locke tantamount to showing that they must be subjective impressions of sense, or “ideas of sensation,” as he called them. Similarly, when Berkeley showed that primary qualities are equally “relative” with the secondary, it was evident at once that they must also be equally “subjective/’ i.e., mere ideas of sensation—or impressions of sense.* Brought to utterance, then, the indirect argument involved here is simply this: the relativity of sensory qualities is prima facie evidence of their subjectivity as impressions of sense.

As an argument this is quite untenable; there is no such connection between relative and subjective as is here assumed. The inference from the relativity of sense qualities to their subjectivity as sensations is possible only on the prior assumption of sensations. But in this event the argument is no longer an argument for sensations. It is no argument at all; it is but a philoSophie predilection; and such indeed it was.

The more articulate argument is best set forth perhaps by George Berkeley. In the first of his Three Dialogues between Hylas and Philonous Berkeley argues, much like Locke before him, that “the most vehement and intense degree of heat [is] avery great pain”* But, he goes on, if heat of this intensity is a subjective sensation, then heat of every intensity must also be a subjective sensation. Similarly with every sensible quality; as “immediately perceived” or felt it must be but a sense impression in the perceiving consciousness—an impression, moreover, which is indistinguishable from the pain or pleasure which it is commonly said to “cause.”

The pivot of this argument is plainly the identity of heat and pain. This identity is admittedly a conclusion drawn from premises. “Seeing therefore they are both immediately perceived at the same time, and the fire affects you only with one simple, or uncompounded idea, it follows that this same simple idea is both the intense heat immediately perceived, and the pain; and consequently, that the intense heat immediately perceived, is nothing distinct from a particular sort of pain” (p. 176).

There are two premises here. The first is that “the fire affects you on!y with one simple, or uncompounded idea.” In saying, “the fire affects you,” Berkeley would seem to be reviving the discredited causal theory of perception. This he would disclaim, however, insisting that the locution was only a concession to common usage, an instance of “speaking with the vulgar.” The fire itself and its affecting us are not “immediately perceived.” They are “inferences.” And the senses “make no inferences” (p. 174). Leaving this aside, then, let us go on to the remainder of the statement, that we are affected “only with one simple, or uncompounded, idea” when we experience the fire and its heat. In this premise Berkeley is invoking the doctrine of simple and complex ideas, which he shared with both Locke and Hume, a doctrine sometimes referred to as “psychological atomism.” Remember, all “ideas of sensation” are simple. The conclusion is evident; there being but one simple idea involved, which is admittedly a subjective idea of pain, this same idea must also be the heat which we “immediately perceive.” Therefore, heat and all secondary qualities must be subjective “ideas of sensation,” or more briefly, “sensations.”

The second premise is not openly stated. It lies implicit in the term “idea” and hence underlies the first premise. We have already noted that by “idea” Berkeley means what we “immediately perceive,” and what we “immediately perceive” is nothing but our own ideas or sensations. Notice in the above brief quotation that he uses “immediately perceived” three times. This is obviously his principal premise: not that the fire affects us with one simple idea, but rather that the heat and the pain “are both immediately perceived.” For to be immediately perceived is to be forthwith a subjective idea of sensation. This is the bare bones of Berkeley’s argument. It is superfluous to say that the heat is perceived at the same time as the pain, that the two are but one simple idea; it only obscures the fact that the argument is a petitio principii, that it has covertly assumed what it would ostensibly prove-—sensations.

2. Descartes

Descartes’s indirect argument for sensations is quite different from those of the English empiricists. It is characteristically more elaborate. Its general purport is that since experience may be but a consistent dream or hallucination, it must be a process of the same ilk operating on sensory materials of the same kind, namely, subjective sensations. As an argument for sensations it too will turn out to prove only what it has already assumed—sensations. Let us now follow this classic argument in the Meditations of Descartes.

Consider, says Descartes, the phenomena of dreams, the delusions of madness and delirium, hallucinations, and the like, and reflect on the power and vividness with which they frequently occur. “Dwelling carefully on this reflection,” he continues, “I see so manifestly that there are no certain indications by which we may clearly distinguish wakefulness from sleep that I am lost in astonishment. And my astonishment is such that it is almost capable of persuading me that I now dream” (Meditation I). The pivot of this argument is “that there are no certain indications by which we may clearly distinguish wakefulness from sleep,” whence it is “manifestly” possible that they arc not actually distinct at all, but that experience is a kind of dream (or hallucination) and the world a grandiose illusion.

Descartes, it may be well to note, does not assume, let alone prove, that experience is a dream: nor does he believe it for a minute. And he does not deny the existence of an external world; he proves this in the sixth Meditation. His concern at this juncture is merely to avoid accepting for “certain” or “indubitable” anything which he finds subject to a “possible doubt.” The fact that experience may be a dream, hence that the world may not exist, plainly renders these matters subject to a possible doubt and warrants laying them aside for the time being as though they were false. Descartes is content, therefore, with the bare possibility that experience may be a kind of dream or hallucination, and that the world may not exist.

This bare possibility, however, that experience may be a dream, speaks volumes. Notice its onesidedness: experience may be a kind of dream, not dream a kind of experience. On the sole basis that the two are not clearly distinguishable why should the one be said to be a version of the other, not the other a version of the one? The answer, I think, can easily be surmised. It is clear to Descartes that dreaming and hallucinating are fabricating or constituting modes of awareness, not discovering modes—to use my own terms, not DescartesY It is not clear to Descartes that experience and perception are discovering modes of awareness. All that is clear to him about experience and perception is that they possess a sensory content, the same content we find in dreams and hallucinations. And since this content is patently subjective in the one instance, it must also be subjective in the other. The conclusion is unavoidable: experience must be a kind of dream, not dream a kind of experience, because the content is subjective sensations.

3. External Experience As Operation and Idea

But if experience is possibly a kind of dream or hallucination, must it not be actually a constituting or fabricating mode of awareness? For if it were actually a discovering mode of awareness, it could not possibly be a constituting mode and hence not even possibly a kind of dream or hallucination. Experience, in short, can possibly be a kind of dream or hallucination only if experience is actually a constituting, not a discovering, mode of awareness.

The question now boils down to what external experience and perception actually are. Either they are discovering modes of awareness or fabricating modes of awareness. For the natural attitude they are discovering modes, not fabricating (or constituting), and discovery is their intrinsic mark just as fabrication is the intrinsic mark of dreams and hallucinations. By virtue of these criteria the two modes of awareness are so clearly distinct that neither one can be said to be even possibly a version of the other.

But if this is evident to the natural attitude of common sense, it was not evident to Descartes. To him the contrary was evident; there are no distinguishing marks. “I see so manifestly that there are no certain indications by which we may clearly distinguish wakefulness from sleep that I am lost in astonishment.” Before giving way to astonishment, however, notice that what is here manifest to Descartes is 1) the absence of distinguishing marks on the side of perception and experience, and 2) the presence of distinguishing marks on the side of dreams, delusions and hallucinations. The presence of these latter marks is revealed in the fact that Descartes has not the slightest doubt about the fabricating or constituting nature of dreams and hallucinations. Only the nature of sense perception and experience is not manifest.

This is a curious disparity, because both perceiving and dreaming are operations of consciousness, cogitationes, and fall equally within the purview of the indubitable certainty which I have of myself as a thinking being, or res cogitans. Among the operations, or cogitationes, which constitute my nature as a res cogitans Descartes repeatedly mentions perception (which he does not distinguish from experience), adding that “although the things which I perceive and imagine are perhaps nothing at all apart from me and in themselves, I am nevertheless assured that these modes of thought (cogitationes) which I call perceptions and imaginations, inasmuch only as they are modes of thought, certainly reside and are met with in me” (Med. Ill and passim.). That this holds for all the operations of consciousness is explicitly stated; “by the word ՝thought’ (cogitatio) I understand all that of which we are conscious as operating within us” (Principles, IX). Manifestly, then, perception (experience) and imagination, being operations of consciousness, are matters of indubitable certainty, possessing intrinsic marks by which they can be clearly distinguished both from each other and from all other cogitationes of consciousness. The nature of sense perception and imagination must be as manifest as the nature of dreams and hallucinations.

Descartes is here plainly at odds with himself. The trouble is the theory of sensation and ideas. For if the “adventitious ideas” of perception are but subjective sensations, then perception and experience do indeed have the same sensory content as dreams and hallucinations. This, however, is just what is to be proved, not assumed. Descartes assumed it nonetheless, and in so doing he quietly overlooked the certain indications by which we can—and actually do—distinguish between outer perception and dreams. In removing these indications Descartes also misconstrued the nature of outer perception and misassessed the quality of its evidence. Let us take a closer look at experience and its evidence.

In turning to external experience it may be well to remind ourselves at the outset of one obvious point. Both external experience and its warrant, the quality of its evidence, are “objects” only for internal experience, i.e., introspection or reflection. As an object of internal experience, external experience is a mode of awareness, an operation of consciousness, a cogitatio, clearly distinct both from its (outer) object and from all other (inner) operations of consciousness. Similarly, its warrant, the evidential quality of its disclosures, is an object for reflection alone, an object moreover inseparable from external experience itself.

Reflecting on this warrant we call to mind the familiar complaints which in all ages have been raised against the senses. Sense experience, they say, is quite unreliable; its warrant is hopelessly prejudiced by the fallibility of its constituent perceptions. Now sense perceptions are undeniably fallible; they abound in errors, distortions, illusions. This is an incontrovertible fact. But on whose authority? Is it not experience itself which shows up these errors of sense; and does it not expose these errors in the very process of correcting them? Consider for a minute this inherent power of self-correction.

It is exercised in every waking instant, in every one of the myriad adjustments, accommodations we are constantly making and have to make because perception confronts us only with the perspectives of things. Every object we perceive is beheld at a given distance, from a certain angle, under these or those conditions of visibility, audibility, and so on. Never do we perceive an object otherwise than through these perspectivai арpearances. It is to these that we adjust in perceiving the things thus appearing. Without this adjustment we should be quite deceived, sooner or later fatally deceived. As adults we have learned to allow for most of these perspectives almost without thinking. A few however are so striking that we have to discount them more or less thinkingly as “illusions,” optical illusions like the wet spot on the road ahead, the converging of the railroad tracks, the oar bent in water, mirages, and the like. There are also auditory and tactual illusions, such as echos, the Doppler effect, and others. In precisely the same way we discern and allow for fancies, dreams, and more rarely hallucinations. If at times we are “taken in” by these illusory appearances, or apparitions, still it is the normal course of experience to expose these errors in the process of correcting them. And external experience is “normal” only to the extent that it is thus selfcorrecting.

In this power of self-correction lies the warrant of external experience. By this power experience redeems itself, so to speak, from the errors of sense. By this power, which exposes as it corrects, experience neutralizes without extinguishing the fallibility of perception and acquires thereby that authenticity which it has for us all as we live it from cradle to grave. This is not to say that experience is thus rendered infallible. But it is to say that experience is not so utterly fallible as to be a possible tissue of illusions—a grand hallucination.

Being neither infallible nor yet merely fallible, the warrant of external experience lies somewhere in between. Were experience infallible its deliverances would be beyond doubt, indubitable; their denial would involve us in inconsistency if not outright contradiction; they would be apodictic. Were it on the other hand merely fallible, its deliverances would be utterly dubious, indecisive, admitting equally of affirmation or denial, in a word problematic. Between these two lies what has been called assertorie; the deliverances of external experience are assertoric in quality, that is, they give rise to assertions of “fact.” These assertions of fact are unlike apodictic statements in that they can be denied without contradiction and hence are ever open to future confirmation and correction. They are unlike problematic statements in that their denial is contrary to fact and hence not equally possible with their affirmation. Lacking the high finality of the apodictic and the utter indecisiveness of the problematic, the assertorie evidence of external experience has its own unique force and value, that of the empirical “facts” of life and science.

But if the warrant of external experience is assertorie, that of isolated perceptions is not; it is merely problematic. An isolated perception, ripped from the fabric of experience, with its connections severed and its edges so frayed that we cannot tell where it begins or ends, such a perception like a mounted specimen can yield only an inert evidence, wholly inconclusive, problematic. Whether this evidence is to be affirmed or denied, in whole or in part, no one can possibly say. Only subsequent perceptions can decide this. But no subsequent perception, similarly excised, is in a position to do the deciding. For the same disability attaches to every subsequent perception as attaches to the given perception; each is so utterly problematic as to require confirmation before it can confirm. Hence no isolated perception can possibly confirm another; confirmation is not only a prospect indefinitely deferred; it is an impossibility in principle. In other words, no sum of problematic evidences can yield an assertorie evidence; the two are as different from each other as from apodietie evidence.

What is amiss here is patently the artificial isolation. Abandon this artifice, restore perceptions to the living fabric of experience, allow them to function normally in its authenticating process, and the whole picture clears up at once. No longer merely fallible and requiring a validation that can never come off, perceptions are now by virtue of this living context both validating, and validated, the springs of assertorie evidence. Such is the authenticity of external experience, its assertorie warrant.

Assertorie evidence has as a companion piece assertorie certainty. Most of the certainties we live by are of this quality. Consider but one of them, one of the most important of all, the living certainty we all have of the existence of the sensible world in which we are born and live and die. I call this a living certainty because it is so nearly one with human life that without it our mortal existence would be unimaginable. Notice too how impregnable it is to all assaults of theory. Challenge, even deride it as one will, its imperious hold remains unshaken in the allimportant business of living. We cannot live without it; we can only philosophize without it.

This certainty, being assertorie, not apodictic, does not leave open as a genuine possibility that the sensible world does not exist. This possibility is not a “real” possibility, it is only a “logical” possibility, i.e., it can be thought without contradiction. To regard this empty, merely logical possibility (that the world does not exist) as of equal weight with the assertorie actuality (that the world does in fact exist) is to deny altogether the assertoric and to reduce the assertorie to the status of the probIernatic. This is what Descartes did when he admitted only two kinds of evidence, apodictic and problematic.

Descartes was moved to this view of evidence by his mathematical predilection for “clear and distinct ideas.” Mathematical “ideas” (objects?) are clear and distinct; their evidence is apodietie and yields indubitable certainty. All other evidence, being unclear and indistinct, is merely problematic and cannot yield true certainty. The evidence of inner experience Descartes found to be apodictic like that of mathematical “ideas.” The evidence of outer experience he found to be not apodictic, hence merely problematic and totally unfit for the secure founding of philosophy. No mention of assertorie evidence or assertorie certainty.

Descartes’s penchant for the apodictic and his consequent neglect of the assertorie were widely shared in the ensuing development of modern philosophy. Even today and in spite of much professed empiricism the inclination is to regard but one certainty as worthy of the name, a certainty that will brook no doubt. Rarely is certainty construed otherwise than as apodictic or indubitable. In philosophic discussions the usual test is: Are you absolutely sure? Are your criteria “adequate”? Can your statement be denied without contradiction? If it can be so denied, it is not genuinely certain and cannot be acceptable to reason; it is merely problematic—as though there were no such thing as assertorie evidence and certainty, or as though they were not philosophically respectable.

The scientific tradition was not as adamant in its preference for the apodictic. For a while this preference did prevail and was quite exclusive. It reached its apogee, perhaps, in Laplace’s famous “ideal” of mechanical determinism. But this ideal was admittedly one which only an “infinite intellect” could possibly realize. From here on it began to recede. With the advent of relativity, quantum theory, and the indeterminacy principle it was finally superseded by “probability.” In probability the classical notion of the assertorie has returned under a new name and under new auspices and is rapidly regaining its respectability.

But Descartes had a second reason for overlooking the assertoric. Besides his mathematical penchant for the formal and apodictic he had also a firm commitment to the theory of sensation. According to this theory sense perception is in effect nothing but the having and manipulating of our own sensations or “ideas.” It follows from this that sense perception cannot possibly reveal to us anything about an existence other than its own. We may “infer” or “posit” such an other and outer existence on the basis of what we thus perceive within; but this inference has no immediate warrant in perception itself, and the senses do not infer. Perception itself, being an “idea” and as such both object and awareness, can bear witness only to itself as an operation or “content of consciousness.” In this respect it is apodictically evident—as an object of introspection. In every other respect it is problematic. Should it correspond perchance to an object outside the mind this correspondence would be for the idea something wholly adventitious and for the mind unknowable.

Whether for these two reasons or not, Descartes recognized only two kinds of evidence, apodictic and problematic, thus virtually denying the assertorie. In denying the assertorie he reduced the evidence of sense experience to the level of the probIernatic and construed the empirical evidence of the world’s existence as a mere possibility no greater than the possibility of the world’s nonexistence. Only by thus depriving sense experiåïñå of its assertorie warrant could Descartes hold as equally and “really” possible 1) that the world does not exist and 2) that the world does exist.

4. Internal Experience

Fortunately the theory of sensation did not similarly threaten internal experience and its apodictic warrant. Descartes saw clearly that on turning inward our attention falls on “objects” which are immediately present and evident beyond doubt. These inner objects are the acts and contents of our “stream of consciousness”—an expression which Descartes did not use. He saw, too, that these acts are grasped apodictically as cogitationes, that is, as individual instances of general operations, the “operations of consciousness.” This conjunction of particularity and universality in inner perception is worthy of some comment.

Ask anybody who seems to be lost in thought what he is thinking about and he will invariably reply, “I was looking at the carving on that table,” or “I was listening to the music next door,” or, “I was imagining what it would be like to be in Mexico,” or, “1 was thinking about my income tax.” And so on. In every instance his report will contain some such general term as “looking” “listening,” “imagining,” “thinking,” each of which signifies not an individual act, but a kind of activity, a general operation,—or cogitatio, as Descartes called it. Obviously, such a “kind” cannot possibly occur; it is not an individual event in time, but a timeless generality under which individual events may fall as “instances.” To report merely that 1 was looking, therefore, would convey only the general nature of what I was doing. There are myriad acts of looking. Only when I add, “looking at the carving on that table,” do 1 begin to convey something of the individuality of the unique act I was actually performing. Even so, the report falls short of conveying the full individuality of the act. No report can do this; the most it can do is describe certain general features of the act in such detail as will suffice for the purpose in hand. In nearly every instance the first of these general features is the kind or species of the act, the operation of which it is an instance; the second is the object of the act, the “content” of the act as it is often unfelicitously called.

I am quite aware that the term “act of consciousness” is not in good odor. But I see no convenient way of avoiding it, or any advantage in doing so. My conscious life, if seems to me, is an “activity” in which I perform countless “acts.” These acts are all individual and unique and vary greatly in complexity, some containing others as parts. This suggests that there may be acts so utterly simple that like atoms they possess no distinguishable parts at all. If there are such atomic acts, I have not been able to discover them. All acts, 1 find, are complex, to the extent, at least, of having distinguishable parts, as we shall see more fully later on. Acts can combine to form more complex acts, and these in turn still higher acts, and so on up to the stream of consciousness itself. In a way the whole life of an individual may be said to be one act. This entails that no act can be isolated except in thought, and that every act, embedded as it is in the stream of consciousness, merges at both ends into other acts and above into higher acts. Below, it may have simpler acts as constituent parts. “Act” in short is an indefinite term, even arbitrary, like the term “object.” But it is not for this reason useless. No other term, I believe, signifies quite as poignantly what is going on inside us, so far at least as this inner activity is an articulated process with discernible segments.

Turning, then, to the acts which were reported above, my point is that these acts, and all acts, can be grasped and expressed only as individual instances of general operations, cogitationes. In every instance an act of which I am aware is grasped as an act of seeing or hearing or touching, of remembering, imagining, thinking, judging, feeling, willing, or the like, always as an individua! instance of an “operation,” or kind of activity. Were this not so my inner gaze would be but a mute awareness—if, indeed, an awareness at all—of my flowing stream of consciousness with its constant coming and going of nameless events, some together, others overlapping, most in sequence. To this inert gaze each event for the duration of its occurrence would be present or “now,” actual only this once. Once over and done with, it would vanish into an oblivion beyond recall. In this event introspection would be as shifting as the stream it contemplates, a fleeting and ineffable awareness hardly worthy of the name.

Plainly this is not what introspection grasps. What it grasps is not a mere Heraclitean flux, but an articulated process with discernible segments, each an instance of a timeless operation. If this circumstance is not a mere accident of language, nor a feature imposed on the stream by introspection itself, then it must lie in the nature of consciousness that its acts actually embody operations. And since these operations are actually embodied or incorporated, they are not “classes” but real “species” or “essences.’’ For individuals do not instance or embody classes; they are only members of classes, of numberless classes at that, whereas each act embodies but a single essence.

Be this as it may, it is only as instances of operations that we grasp acts in introspection, identify them for what they are, bring them to utterance, and repeat them in essence, i.e., in specie, not individualiter, and make them “objects” of philoSophie reflection.

This conjunction or union of particular and universal, of act and operation, of instance and essence, is more widely recognized than one might at first assume. I have already adduced common usage in reporting what goes on within us. Empirical psychology, even when it eschews the language and evidence of introspection, is Still preoccupied with operations or processes—the “learning process” for example. Curiously, this operational grasp is nowhere more conspicuous than in mathematics. An arithmetic operation—addition, for example—is not a psychological act or event in time. It may be exemplified by psychological acts in time, any number of them, as when we actually add a column of figures. But addition itself as a mathematical operation can never thus occur. It is a timeless entity possessing only such properties as commutation, association, and the like. Psychologica! acts of addition, on the other hand, are not timeless entities, but actual events in time, each and all exemplifying or instancing the one operation of addition. Similarly with every other operation in mathematics. It is not surprising, then, that Descartes the mathematician should view the acts of consciousness as operations, or cogitationes, and should take them in this respect as apodictically evident. That individual acts, too, may be apodictically evident to introspection he neither affirms nor denies. He is interested in acts only in their essentia! character as operations and the apodictic certainty with which they can be grasped. Even his interest in them as operations, however, did not extend much beyond their being items in the primal certainty which I have of myself as a res cogitans. Hence he was content in the main simply to list them in his several inventories (all of them partial) of the cogitationes of consciousness.

In each of these inventories he lists external experience or sense perception (sentire, sensus) and distinguishes it without further ado from imagining, thinking, willing, and presumably from all the other operations of consciousness, including dreams, delusions, and hallucinations. This can only mean that there are in fact “certain indications” by which we can “clearly distinguish” external experience from dreams and hallucinations, that these indications are indubitably clear and distinct, and that they delineate external experience as a discovering mode of awareness yielding assertorie evidence.

All of this follows on Descartes’s own showing and refutes, once again, his argument that because it lacks intrinsic marks external experience may be a kind of dream or hallucination. Since it has intrinsic marks apodictically evident to reflection, external experience cannot possibly be a kind of dream or hallucination. But if it is impossible that experience should be a dream or hallucination, is it still possible that the world does not exist, that it is an illusion? By the assertorie warrant of experience the existence of the world is an empirical fact. The possibility that the world does not exist is therefore contrary to fact, not a “real” possibility. It is nonetheless logical, since the evidence of experience is not apodictic. As such it is quite empty and quite unfit to serve as a premise in a serious philosophic argument, especially when there is an appeal to empirical fact.

This is not Descartes’s view because, as we noted in the preceding section, he construed the warrant of experience as probIernatic—as the sensation theory would seem to entail—and so could regard as equally real 1) the possibility that the world does not exist, that it is an illusion, and 2) the possibility that the world does exist, that it is not an illusion. Inasmuch as Descartes bequeathed this view, along with the sensation theory, to a number of his successors, Husserl among them, it is fitting that we take a closer look at the view and at some of its consequences.

5. Possible and Contingent Existence

Assuming with Descartes that the two possibilities are equally real, we must then regard them as suspended in a kind of balance in which neither possibility outweighs the other. For if either one were to outweigh the other, the balance would be destroyed, the preponderant member would be not merely possible, but actual or necessary, and the outweighed member wou!d be no longer possible (in the initial sense), but contrary to fact or impossible. In order for either side to be possible in this “problematic” sense, both sides must be equally possible; they must form a balanced pair. Only then is the situation strictly problematic.

Taking now the balanced pair, the possibility that the world exists and the possibility that the world does not exist, we may then compress this pair into the single statement that the existence of the world is not necessary. For if the existence of the world were necessary there would be no possibility that it does not exist; and if there were no possibility that the world does not exist, there would be no balancing possibility that it does exist, its existence would be actual or necessary, not possible. Both possibilities vanish if the world exists necessarily; they obtain only if the world does not exist necessarily. Hence the balance of possibilities can be boiled down to the statement that the existence of the world is not necessary.

This statement, however, may be construed in two very different ways. So far we have construed it as signifying the possibility-pair that the world may exist or not exist; so far nothing more. But there is another way of construing it and that is to take it as implying the classical doctrine of “contingency,” which also holds that the existence of the world is not necessary. This contingency doctrine, however, has nothing to do with our possibility doctrine; and to assume that it does is to fall into a serious confusion. It is imperative that the two doctrines be held carefully apart.

The classical doctrine of contingency holds that the world does not exist “necessarily” but only “contingently,” that its contents are not “necessary beings” but “contingent beings.” They are said to be contingent because their essence does not involve existence. A momenťs reflection will make this clear. Existence is involved in our perception of things; since, as the natural attitude holds, we can perceive only what is “there” (existent) to be perceived. But conception is not thus limited by existence; we can think and talk of what is nonexistent as well as of what exists, of things that no longer exist or do not yet exist or cannot exist at all. The reason for this is that in merely conceiving things—as distinct from perceiving them—what we grasp of them is their nature or essence, “what” they are. Of this nature or essence existence is not a part; it is not a “real predicate,” as Kant put it; for it adds nothing one way or the other to “what” a thing is. In this respect all things are equally capable of existing or not existing. If they do exist, therefore, their existence must derive from something other than their essence, something that must be already existing, since existence can come only from existence, not from essence. Now if this other existing thing is similarly capable of not existing, then its existence must in turn derive from yet a third existing thing. And so on without end so long as we are dealing with things whose existence is derived or “contingent.”

All things temporal are in this sense contingent and so is the whole order and assemblage of things in time, the world. There would be no difficulty here were it not for the fact that this contingent existence is also actual. We are assured of this actuality by the assertorie witness of experience. Being actual this contingent existence can be adequately accounted for only by assuming a “sufficient reason or ground,” namely, another existence, not contingent, not capable of not existing, but “necessary”—a thing, in other words, which exists necessarily because its essence precludes the possibility of nonexistence; or, what is the same thing, whose essence involves or includes existence.

Descartes’s version of this argument is somewhat different though not conflicting. Instead of focusing on the essence of things existing in time, he focuses instead on the nature of temporai existence itself, more precisely on the order of instants in the duration of anything existing in time. In the successive course of its existence, no matter how long or short its duration, a thing actually exists at only one instant at a time, never at all instants at once, simultaneously or tota simul. Consider one such instant or “now” in the duration of a thing’s existence. Existing “now” it exists no longer at past instants and not yet at future instants. Hence neither its past existence nor its future existence, since both are now nonexistent, could possibly be the sufficient cause of its present existence. Nor could anything existing simultaneously with the thing in the same “now” cause its existence, for there is no time in the “now” to exercise this causality; all empirical causality “takes time,” requires duration, and presupposes existence in time—the very thing that is here in question. There is, in short, no sufficient cause in time for the existence of anything at any instant of its duration in time. If there be such a sufficient cause for existence in time, this cause must exist not seriatim in time but tota simul outside or beyond time.

A corollary of this argument might be called the “hazard of existence.” At any given instant in its duration the existence of a thing precludes its nonexistence for that instant. But the respite is only momentary; for the very next instant and all future instants hang in a quivering balance of being or not being which nothing in the “now,” still less in the past or future, can possibly tilt. Thus even while existing a thing confronts nonexistence as an ever-present possibility, a constant threat, as it were, menacing every instant in its duration. In conscious beings this threat or hazard may arouse a kind of “existential anxiety.” But this is not an explicit tenet of the classical doctrine of contingency.

Whatever you may think of this argument for contingency and for postulating a necessary existence, one thing is unmistakably clear: what is said to be contingent is not a possible existence but an actual existence, the actual existence of things in time and the world. Only actual existence can be contingent or necessary. A possible existence is so far only possible, not even contingent, let alone necessary. In holding, therefore, that the existence of the world is not necessary, the contingency doctrine takes this existence of the world as actual (assertorie), not as merely possible (problematic). Correlatively, the lurking possibility which renders this existence contingent is the (threatening) possibility of not existing, not the mere possibility of existing.

This understanding of the proposition, “the existence of the world is not necessary,” is plainly different from our former understanding of the proposition as signifying simply the possibility-pair, the world may or may not exist. This difference should be sharpened and made verbally manifest.

This can be done simply by recalling the possibility-pair and noting how the possibility that the world does not exist is virtually one with the possibility that experience is a kind of dream or hallucination. For if experience is a dream or hallucination, then the world is an illusion and its existence is not necessary to our experience of if, we can experience a world whether the world exists or not. The italicized words make clear at once what is meant here by “not necessary”; it means not necessary in relation to experience. Of “necessary” as an intrinsic quality, opposed to “contingent,” it says absolutely nothing. It completely ignores actual existence, as it ignores assertorie evidence, and hence does not even hint at the contingency doctrine. What it does assert refers primarily to the nature of experience, not existence, and it is in relation to experience alone that the existence of the world is said to be not necessary. “Not necessary” is thus an ellipsis for the phrase “not necessary to experience.” On this understanding the proposition should read, “the existence of the world is not necessary to experience.”

The corresponding reading for the contingency doctrine should be, “the existence of the world is not necessary in itself.” The slight difference in wording makes clear at once the vast differenee in meaning. It makes the word “necessary” refer in the one case to a relative property, and in the other to an intrinsic property of existence itself. It reveals at the same time that the opposite of necessary, “not necessary,” is equivalent in the one case to “possible” and in the other to “contingent,” two very distinct terms. Clearly the important difference here is that the one proposition—or doctrine—is principally about experience, the other about existence.

6. Experience and Existence

According to Descartes’s argument the possibility that the world does not exist is a consequence of the possibility that experience may be an illusion. For if it is really possible that experience is an illusion, then plainly experience does not depend in any way on the existence of a world; it can take place without a world; the “world” in this instance would be simply the “ideas” we directly perceive. Obviously the existence of such a world is not necessary to experience. Curiously, Descartes never said this. He implied it plainly enough; but he did not say it in so many words. Some of his successors did say it, among them Husserl, and in saying it went on to equate “not necessary” with “contingent” as though the nature of existence were determined by the nature of the experience in which it is given, contingent existence in problematic experience, necessary existence in apodictic experience. Descartes was spared this confusion by holding to the classical doctrine of contingency.

According to this classical doctrine the contingency of existence does not follow from the assertorie quality of experience, nor does the assertorie quality of experience follow from the contingency of existence. The two are independent, at least to the extent that contingency and necessity are intrinsic properties of existence without regard to experience, whereas assertorie and apodictic are qualities of experience without regard to existence. Because they are thus independent Descartes could hold to the one (contingency) while abandoning the other (assertoric). Also, because they are independent they can be combined in four different ways. We can have

1) assertorie experience of contingent existence,

2) assertorie experience of necessary existence,

3) apodictic experience of contingent existence, and

4) apodictic experience of necessary existence.

If there be such a thing as experience of a necessary being, it would fall under 2) or 4), possibly both. But I set these aside in order to focus on 1) and 3), since 1) refers to our external experience of the outer world, and 3) refers to our internal experience of our inner conscious selves.

I am particularly interested in 3), since it brings out the point that apodictic experience does not entail necessary existence. Although inner experience yields apodictic evidence of consciousness and its operations, it does not make their existence necessary. Descartes was perfectly clear on this point. What is “necessary” here is not the existence of my psychic operations or acts, or—what is the same thing—my existence as a conscious (as Descartes put it, “thinking”) being. What is “necessary” is solely the proposition (pronunciatimi) which asserts this existence. This proposition, moreover, is necessary only in the sense of being “necessarily true.” And it is necessarily true because the apodictic evidence on which it is based cannot, in this instance at least, be denied without contradiction. For if every act of thinking is an implicit assertion of my existence, then I cannot possibly deny my existence, since the very act of denying it would implicitly assert it and thus involve me in contradiction. What is necessary, therefore, is not my existence but only the truth of the proposition which asserts my existence.

Even this logical necessity, however, is not unqualified. It is subject to two limitations. 1) The proposition in question must be phrased solely in the first person singular: I must assert my existence. Only for me is my existence apodictically evident; only my reflections can grasp my acts with indubitable immediacy. For nobody else is my existence thus evident; and nobody else’s existence is thus evident to me. Only for me in relation to myself is the evidence apodictic and the proposition undeniable. Expressed in the second or third person the proposition loses completely its necessity; it ceases to be a necessary truth. And all because my existence is not necessary in itself, but only contingent. 2) Not only must this proposition be couched in the first person singular, it must also be restricted to the present tense. On this too Descartes was perfectly clear. “Having refleeted well and carefully examined all things, we must come to the definite conclusion that this proposition: I am, I exist, is necessarily true each time that I pronounce it, or that I mentally conceive it . . . I am, I exist, this is certain. But how often? Just when I think; for it might possibly be the case if I ceased entirely to think, that I should likewise cease altogether to exist” (Meditation II). The proposition holds, in short, only so long as 1 am thinking or am actively conscious. It says nothing of my existence when I may not be thinking, say before birth or after death. Hence I can think of myself as not existing at times when I am not thinking, and this without a semblance of contradiction. The thought of my coming to be and passing away, the thought of my existence as contingent, is thoroughly consistent with the cogito ergo sum and with the apodictic evidence on which it is based.

What this all signifies is not alone that the proposition which asserts my existence is necessarily true only when uttered in the first person singular and in the present tense. It also signifies that this proposition is subject to these limitations because my existence is contingent, not necessary. The important thing here is that I have apodictic experience of (my own) contingent existence. This is important because it shows at once that apodictic evidence and the necessary truths to which it gives rise do not imply or entail necessary existence. This means that we cannot infer from the apodictic warrant of inner experience to the neeessary existence of consciousness. There is no such connection between inner experience and its object and none between outer experience and its object. Descartes did not assume such a connection; but Husserl did, as we shall see in Chapter VII.

7. Summary

Descartes’s indirect argument for sensations begins by assuming that sense experience possesses no intrinsic marks to distinguish it clearly from dream or hallucination. This gives rise at once to the real possibility that sense experience is a dream or hallucination and to the accompanying possibility that the world does not exist, that it is an illusion. This is tantamount to saying that the existence of the world is not necessary to experience, that as what we dream or hallucinate need not exist, so what we experience need not exist.

This view, I believe, rests on one initial assumption: the theory of sensation. Only if we first assume that the immediate objects of sense perception are sensations or ideas, can we then liken experience to a dream or hallucination and take as real the possibility that it may really be such. In this event the argument would invalidate itself by covertly assuming what it would prove; like most of the others, it would be an argument from sensations, not for sensations.

Misled by the theory of sensation Descartes took for “real” a possibility which is in fact “unreal,” merely logical, empty. This involved reducing the warrant of sense experience from assertorie to problematic. Husserl followed Descartes in this and went one step farther; he construed the apodictic and assertoric warrants of inner and outer experience as definite indices of necessary and contingent existence.

We have now completed our survey of the arguments for sensations, direct and indirect, and have found them all untenable. Let us now return to the theory of sensation and ideas and follow its further development, especially that phase of its development in which the successors of Descartes attempt to show how from sensations we get experience and knowledge of an empirical world. We can see from the outset that this process will be the reverse of the subjectivizing process by which sensations were originally got—a process of “production” rather than a process of “reduction.” Let us examine the principal steps in this process of “production,” not so much in historical detail as rather schematically with our eyes fixed on the exigencies of the problem itself.

*It was only the word “idea” that made it possible to avoid the embarrassment of having to speak, and think, of primary sensations or impressions as well as secondary!

*P. 176. References to the Three Dialogues will be given by the page number of the Works of George Berkeley, Bishop of Cloyne, edited by A. A. Luce and ò. E. Jessop, Thomas Nelson and Sons, Ltd., London, New York, 1948-57, Vol. 2.