The philosopher's first duty is in clear language to declare his starting-point, with what a mathematician would call the "tangent to the origin" of the path along which he is travelling, as afterwards the critic's first duty is to describe this initial attitude. I have therefore first of all to indicate the directing idea of the new philosophy. But it is not a question of extracting a quintessence, or of fencing the soul of doctrine within a few summary formulae. A system is not to be resumed in a phrase, for every proposition isolated is a proposition falsified. I wish merely to elucidate the methodical principle which inspires the beginning of Mr Bergson's philosophy.

To philosophy itself falls the task and belongs the right to define itself gradually as it becomes constituted. On this point, an anticipation of experience seems hardly possible; here, as elsewhere, the finding of a synthetic formula is a final rather than preliminary question. However, we are obliged from the outset of the work to determine the programme of the inquiry, if only to direct our research. It is the same on the threshold of every science. There, it is true, the analogy ceases. For in any science properly speaking the determination of beginning consists in the indication of an object, and a matter, and beyond that, to each new object a new science reciprocally corresponds, the existence of the one involving the legitimacy of the other. But if the various sciences—I mean the positive sciences—divide different objects thus between them, philosophy cannot, in its turn, come forward as a particular science, having a distinct object, the designation of which would be sufficient to characterise and circumscribe it. Such was always the traditional conception: such will ours continue to be. For, as a matter of fact, every object has a philosophy and all matter can be regarded philosophically. In short, philosophy is chiefly a way of perceiving and thinking, an attitude and a proceeding: the peculiar and specific in it is more an intuition than a content, a spirit rather than a domain.

What, then, is the characteristic function of philosophy, at least its initial function, that which marks its opening?

To criticise the works of knowledge spontaneously effected; that is to say, to scrutinise their direction, reach, and conditions: that is today the unanimous answer of philosophers when questioned about the goal of their labours. In other terms, what they study is not so much such and such a particular "thing" as the relation of mind to each of the realities to be studied. Their object, if we must employ the word, is knowledge itself, it is the act of knowing regarded from the point of view of its meaning and value. Philosophy thus appears as a new "order" of knowledge, co-extensive with what is knowable, as a kind of knowledge of the second degree, in which it is less a question of learning than of understanding, in which we aim at progressing in depth rather than in extent; not effort to extend the quantity of knowledge, but reflection on the quality of this knowledge. Spontaneous thought—vulgar or scientific—is a direct, simple, and practical thought turned towards things and partial to useful results; seeking what is formulable rather than what is true, or at least so fond of formulae which can be handled, manipulated, or transmitted, that it is always tempted to see the truth in them; a thought which, moreover, sets out from more or less unguarded postulates, abandons itself to the motive impulses of habits contracted, and goes straight on indefinitely without self-examination. Philosophy, on the contrary, desires to be thought about thought, thought retracing its life and work, knowledge labouring to know itself, fact which aspires to fact about itself, mental effort to become free, to become entirely transparent and luminous in its own eyes, and, if need be, to effect self-reform by dissipating its natural illusions. What we have before our eyes then are the initial postulates themselves, the first spontaneous thoughts, the obscure origins of reason; and we are proceeding towards a point of departure rather than arrival.

The new philosophy does not refuse to carry out this first critical task; but it carries it out in its own way after determining more precisely the real conditions of the problem. At the hour when methodical research begins, the philosopher's mind is not clean-swept; and it would be chimerical to wish to place oneself from the beginning, by some act of transcendence, outside common thought. This thought cannot be inspected and judged from outside. It constitutes, whether we wish it or no, the sole concrete and positive point of departure. Let us add that common-sense constitutes also our sole point of insertion into reality. It can only then be a question of purifying it, not in any way of replacing it. But we must distinguish in it what is pure fact, and what is ulterior arrangement, in order to see what are the problems which really are presented, and what are, on the contrary, the false problems, the illusory problems, those which relate only to our artifices of language.

The search for facts is then the first necessary moment of all philosophy.

But common thought comes before us at the outset as a piece of very composite alluvial ground. It is a beginning of positive science, and also a residue of all philosophical opinions which have had some vogue. That, however, is not its primary basis. Primum vivere, deinde philosophari, says the proverb. In certain respects, "speculation is a luxury, whilst action is a necessity." ("Creative Evolution", page 47.) But "life requires us to apprehend things in the relation they have to our needs." ("Laughter", page 154.) Hence comes the fundamental utilitarianism of common-sense. Therefore if we wish to define it in itself and for itself, and no longer as a first approximation of such and such a system of metaphysics, it appears to us no longer as rudimentary science and philosophy, but as an organisation of thought in view of practical life. Thus it is that outside all speculative opinion it is effectively lived by all. Its proper language, we may say, is the language of customary perception and mechanical fabrication, therefore a language relative to action, made to express action, modelled upon action, translating things by the relations they maintain to our action; I mean our corporal and synthetic action, which very evidently implies thought, since it is a question of the action of a reasonable being, but which thus contains a thought which is itself eminently practical.

However, we are here regarding common-sense considered as a source of fact. Its utilitarianism then becomes a kind of spontaneous metaphysics from which we must detach ourselves. But is it not the very task of positive science to execute this work of purification? Nothing of the kind, despite appearances and despite intentions. Let us examine more closely. The general categories of common thought, according to Mr Bergson, ("Philosophic Intuition" in the "Metaphysical and Moral Review", November 1911, page 825.) remain those of science; the main roads traced by our senses through the continuity of reality are still those along which science will pass; perception is an infant science and science an adult perception; so much so that customary knowledge and scientific knowledge, both of them destined to prepare our action upon things, are of necessity two visions of the same kind, though of unequal precision and reach. It does not follow that science does not practise a certain disinterestedness as far as immediate mechanical utility is concerned; it does not follow that it has no value as knowledge. But it does not set itself genuinely free from the habits contracted in common experience, and to inform its research it preserves the postulates of common-sense; so that it always grasps things by their "actable" side, by their point of contact with our faculty for action, under the forms by which we handle them conceptually or practically, and all it attains of reality is that by which nature is a possible object of language or industry.

Let us turn now towards another aspect of natural thought, to discover in it the germ of the necessary criticism. By the side of "common-sense," which is the first rough-draft of positive science, there is "good sense," which differs from it profoundly, and marks the beginning of what we shall later on call philosophic intuition. (Cf. an address on "Good Sense and Classical Studies", delivered by Mr Bergson at the Concours general prize distribution, 30th July 1895.) It is a sense of what is real, concrete, original, living, an art of equilibrium and precision, a fine touch for complexities, continually feeling like the antennae of some insects. It contains a certain distrust of the logical faculty in respect of itself; it wages incessant war upon intellectual automatism, upon ready-made ideas and linear deduction; above all, it is anxious to locate and to weigh, without any oversights; it arrests the development of every principle and every method at the precise point where too brutal an application would offend the delicacy of reality; at every moment it collects the whole of our experience and organises it in view of the present. It is, in a word, thought which keeps its freedom, activity which remains awake, suppleness of attitude, attention to life, an ever-renewed adjustment to suit ever-new situations.

Its revealing virtue is derived from this moving contact with fact, and this living effort of sympathy. This is what we must tend to transpose from the practical to the speculative order.

What, then, will be for us the beginning of philosophy? After taking cognisance of common utilitarianism, and to emerge from the relativity in which it buries us, we seek a departure-point, a criterion, something which decides the raising of inquiry. Where are we to find such a principle, except in the very action of thought; I mean, this time, its action of profound life independent of all practical aim? We shall thus only be imitating the example of Descartes when solving the problem of temporary doubt. What we shall term return to the immediate, the primitive, the pure fact, will be the taking of each perception considered as an act lived, a coloured moment of the Cogito, and this will be for us a criterion and departure-point.

Let us specify this point. Immediate data or primitive data or pure data are apprehended by us under forms of disinterested action; I mean that they are first of all lived rather than conceived, that before becoming material for science, they appear as moments of life; in brief, that perception of them precedes their use.

It is at this stage previous to language that we are by these pure data in intimate communion with reality itself, and the whole of our critical task is to return to them through a regressive analysis, the goal of which is gradually to make our clear intelligence equal to our primordial intuition. The latter already constitutes a thought, a preconceptual thought which is the intrinsic light of action, which is action itself so far as it is luminous. Thus there is no question here of restricting in any degree the part played by thought, but only of distinguishing between the perceptive and theoretic functions of mind.

What is "the image" of which Mr Bergson speaks at the beginning of "Matter and Mind" except, when grasped in its first movement, the flash of conscious existence "in which the act of knowledge coincides with the generating act of reality"? ("Report of the French Philosophical Society", philosophical vocabulary, article "Immediate".)

Let us forget all philosophical controversies about realism and idealism; let us try to reconstruct for ourselves a simplicity, a virginal and candid glance, freeing us from the habits contracted in the course of practical life. These then are our "images": not things presented externally, nor states felt internally, not portraits of exterior beings nor projections of internal moods, but appearances, in the etymological sense of the word, appearances lived simply, without our being distinguished from them, as yet neither subjective nor objective, marking a moment of consciousness previous to the work of reflection, from which proceeds the duality of subject and object. And such also, in every order, appear the "immediate feelings"; as action in birth, previous to language. (Cf. "Matter and Memory", Foreword to the 7th edition.)

Why depart from the immediate thus conceived as action and life? Because it is quite impossible to do otherwise, for every initial fact can be only such a pulsation of consciousness in its lived act, and the fundamental and primitive direction of the least word, were it in an enunciation of a problem or a doubt, can only be such a direction of life and action. And we must certainly accord to this immediacy a value of absolute knowledge, since it realises the coincidence of being and knowledge.

But let us not think that the perception of immediacy is simple passive perception, that it is sufficient to open our eyes to obtain it, today when our utilitarian education is completed and has passed into the state of habit. There is a difference between common experience and the initial action of life; the first is a practical limitation of the second. Hence it follows that a previous criticism is necessary to return from one to the other, a criticism always in activity, always open as a way of progressive investigation, always ready for the reiteration and the renewal of effort.

In this task of purification there is doubtless always to be feared an illusion of remaining in the primitive stage. By what criteria, by what signs can we recognise that we have touched the goal? Pure fact is shown to be such on the one hand because it remains independent of all theoretical symbolism, because the critique of language allows it to exist thus as an indissoluble residue, because we are unable not to "live" it, even when we free ourselves from the anxiety of utility; on the other hand, because it dominates all systems, and imposes itself equally upon them all as the common source from which they derive by diverging analyses, and in which they become reconciled. Assuredly, to attain it, to extricate it, we must appeal to the revelations of science, to the exercise of deliberate thought. But this employment of analysis against analysis does not in any way constitute a circle, for it tends only to destroy prejudices which have become unconscious: it is a simple artifice destined to break off habits and to scatter illusions by changing the points of view. Once set free, once again become capable of direct and simple view, what we accept as fact is what bears no trace of synthetic elaboration. It is true that here a last objection presents itself: how shall we think this limit, purely given, to any degree at all in fact, if it must precede all language?

The answer is easy. Why speak thus of limit? This word has two senses: at one time it designates a last term in a series of approximations, and at another a certain internal character of convergence, a certain quality of progression.

Now, it is the second sense only which suits the case before us. Immediacy contains no matter statically defined, and no thing. The notion of fact is quite relative. What is fact in one case may become construction in another. For example, the percepts of common experience are facts for the physicist, and constructions for the philosopher; the same applies to a table of numerical results, for the scholar who is trying to establish a theory, or for the observer and the psychologist. We may then conceive a series in which each term is fact in relation to those which follow it, and constructed in relation to those which precede it. The expression "primitive fact" then determines not so much a final object as a direction of thought, a movement of critical retrogression, a journey from the most to the least elaborate, and the "contact with pure immediacy" is only the effort, more and more prolonged, to convert the elements of experience into real and profound action.