An essential difference between Kant's method and that which I follow is to be found in the fact that he starts from indirect, reflected knowledge, whereas I start from direct and intuitive knowledge. He is comparable to a person who measures the height of a tower from its shadow; but I am like one who applies the measuring-rod directly to the tower itself. Philosophy, therefore, is for him a science of concepts, but for me a science in concepts, drawn from knowledge of perception, the only source of all evidence, and set down and fixed in universal concepts. He skips over this whole world of perception which surrounds us, and which is so multifarious and rich in significance, and he sticks to the forms of abstract thinking. Although he never states the fact, this procedure is founded on the assumption that reflection is the ectype of all perception, and that everything essential to perception must therefore be expressed in reflection, and indeed in very contracted, and therefore easily comprehensible, forms and outlines. Accordingly, what is essential and conformable to law in abstract knowledge would place in our hands all the threads that set in motion before our eyes the many-coloured puppet-show of the world of perception. If only Kant had expressed this highest principle of his method plainly, and had then followed it consistently, he would at least have been obliged clearly to separate the intuitive from the abstract, and we would not have had to contend with inextricable contradictions and confusions. But from the way in which he has solved his problem we see that that fundamental principle of his method was only very indistinctly present in his mind, and thus we still have to guess at it, even after a thorough study of his philosophy.
Now as regards the method stated and the fundamental maxim itself, there is much to be said for it, and it is a brilliant idea. The real nature of all science consists indeed in our comprehending the endless manifold of the phenomena of perception under comparatively few abstract concepts, and arranging out of these a system from which we have all those phenomena wholly in the power of our knowledge, can explain the past and determine the future. The sciences, however, divide among themselves the extensive sphere of phenomena according to the special and manifold classes of these latter. It was a bold and happy idea to isolate what is absolutely essential to the concepts as such and apart from their content, in order to see from the forms of all thinking, found in this way, what is also essential to all intuitive knowledge, and consequently to the world as phenomenon in general. Now since this would be found a priori on account of the necessity of those forms of thought, it would be of subjective origin, and would lead exactly to the ends Kant had in view. Then before going farther, what the relation of reflection to knowledge of perception is should have been investigated (and this naturally presupposes the clear separation of the two, which Kant neglected); in what way reflection really repro duces and represents knowledge of perception. It should have been investigated whether such reflection remains quite pure, or is changed and partially disguised by assimilation into its own (reflection's) forms, whether the form of abstract reflective knowledge becomes more definite through the form of knowledge of perception, or through the nature or quality that unalterably belongs to itself, i.e., to reflective knowledge. In this way, even what is very heterogeneous in intuitive knowledge can no longer be distinguished, the moment it has entered reflective knowledge; and conversely, many distinctions observed by us in the reflective method of knowledge have also sprung from this knowledge itself, and in no way indicate corresponding differences in intuitive knowledge. As a result of this investigation, however, it would have been seen that knowledge of perception, on being taken up into reflection, undergoes nearly as much change as food does when assimilated into the animal organism, whose forms and combinations are determined by itself, so that from their composition the nature and quality of the food can no longer be recognized at all. Or (for this is saying a little too much) at any rate, it would have appeared that reflection is in no way related to knowledge of perception as a reflection in water is to the objects reflected, and hardly even as the shadow of these objects is to the objects themselves. Such a shadow reproduces only a few external outlines, but it also unites the most manifold into the same form, and presents the most varied through the same outline. Thus, starting from it, we could not possibly construct the shapes or forms of things with completeness and certainty.
The whole of reflective knowledge, or reason (Vernunft), has only one main form, and that is the abstract concept. It is peculiar to our faculty of reason itself, and has no direct necessary connexion with the world of perception. This world of perception, therefore, exists for the animals entirely without reflective knowledge, and even if it were to be a totally different world, that form of reflection would nevertheless suit it just as well. But the combination of concepts for judging has certain definite and regular forms which, found by induction, constitute the table of judgements. For the most part, these forms can be derived from the nature of reflective knowledge itself, and hence directly from the faculty of reason, especially in so far as they spring from the four laws of thought (which I call metalogical truths) and from the dictum de omni et nullo. [18] Others of these forms, however, have their ground in the nature of knowledge of perception, and hence in the understanding; yet they do not by any means point to an equal number of special forms of the understanding, but can be deduced wholly and entirely from the sole function that the understanding has, namely direct knowledge of cause and effect. Finally, still others of these forms have sprung from the concurrence and combination of the reflective and intuitive methods of knowledge, or really from the taking up of the latter into the former. I shall now go through the moments of the judgement individually, and demonstrate the origin of each from the sources mentioned. From this it follows automatically that a deduction of categories from them falls to the ground, and that the assumption thereof is just as groundless as its exposition has been found to be confused and self-conflicting.
(1) The so-called quantity of judgements springs from the essential nature of concepts as such. It therefore has its ground solely in our faculty of reason, and has absolutely no direct connexion with the understanding and with knowledge of perception. As explained in the first book, it is in fact essential to concepts as such that they have a range, a sphere, and that the wider and less definite concept includes the narrower and more definite. The latter can therefore be separated out, and this can be done in two ways; either we express the narrower concept merely as an indefinite part of the wider concept in general, or we define it and completely separate it by means of the addition of a special name. The judgement that is the carrying out of this operation is called in the first case a particular, in the second case a universal judgement. For example, one and the same part of the sphere of the concept "tree" can be isolated through a particular and through a universal judgement, thus: "Some trees bear gall-nuts," or "All oaks bear gall-nuts." We see that the difference of the two operations is very slight, in fact that its possibility depends on the richness of the language. Nevertheless, Kant has declared that this difference reveals two fundamentally different actions, functions, categories of the pure understanding that just through these determines experience a priori.
Finally, we can also use a concept in order to arrive by its means at a definite, particular representation of perception, from which, and at the same time from many others, this concept itself is drawn off; this is done through the singular judgement. Such a judgement indicates only the boundary between abstract knowledge and knowledge of perception, and passes directly over to the latter: "This tree here bears gall-nuts." Kant has made a special category of this also.
After all that has been said, there is no need here of further polemic.
(2) In the same way, the quality of judgements lies entirely within the province of our faculty of reason, and is not an adumbration of any law of the understanding that makes perception possible; in other words, it does not point or refer thereto. The nature of abstract concepts, which is just the inner nature of our faculty of reason itself objectively comprehended, entails the possibility of uniting and separating their spheres, as already explained in the first book, and on this possibility, as their presupposition, rest the universal laws of thought, the laws of identity and of contradiction. Since they spring purely from our faculty of reason, and cannot be further explained, I have attributed to them metalogical truth. They determine that what is united must remain united, and what is separated must remain separated, and hence that what is settled and established cannot at the same time be again eliminated. Thus they presuppose the possibility of the combination and separation of spheres, in other words, judgement. But according to the form, this lies simply and solely in our faculty of reason, and this form has not, like the content of the judgements, been taken over from the perceptible knowledge of the understanding, and therefore no correlative or analogue of it is there to be looked for. After perception has arisen through the understanding and for the understanding, it exists complete, subject to no doubt or error; accordingly it knows neither affirmation nor denial. For it expresses itself, and has not, like the abstract knowledge of our faculty of reason, its value and content in the mere relation to something outside it, according to the principle of the ground of knowing. It is therefore nothing but reality; all negation is foreign to its nature; that can be added in thought only through reflection, but on this very account it always remains in the province of abstract thinking.
To the affirmative and negative Kant adds the infinite judgements, making use of a fad of the old scholastics, a cunningly contrived stop-gap not even requiring an explanation, a blind window, like many others employed by him for the sake of his architectonic symmetry.
(3) Under the very wide concept of relation Kant has brought three entirely different properties of judgements, which we must therefore examine individually in order to recognize their origin.
(a) The hypothetical judgement in general is the abstract expression of that most universal form of all our knowledge, the principle of sufficient reason. In my essay on this principle, I showed in 1813 that it has four entirely different meanings, and that in each of these it originates primarily from a different faculty of knowledge, just as it also concerns a different class of representations. From this it is sufficiently clear that the origin of the hypothetical judgement in general, of this universal form of thought, cannot be, as Kant would have it, merely the understanding and its category of causality; but that the law of causality, the only form of knowledge of the pure understanding according to my description, is only one of the forms of the principle of sufficient reason embracing all pure or a priori knowledge. This principle, on the other hand, has in each of its meanings this hypothetical form of judgement as its expression. Here we see quite clearly how kinds of knowledge quite different in their origin and significance nevertheless appear, when thought by our faculty of reason in abstracto, in one and the same form of combination of concepts and judgements. In this form they can no longer be distinguished at all, but in order to distinguish them we must go back to knowledge of perception, leaving abstract knowledge altogether. Therefore the path followed by Kant for finding the elements and also the inner mechanism of intuitive knowledge from the standpoint of abstract knowledge was quite the wrong one. Moreover, the whole of my introductory essay On the Principle of Sufficient Reason is to be regarded to a certain extent merely as a thorough discussion of the significance of the hypothetical form of judgement; I shall therefore not dwell on it any more here.
(b) The form of the categorical judgement is nothing but the form of the judgement in general, in the strictest sense. For, strictly speaking, judging simply means thinking the combination, or the irreconcilability, of the spheres of concepts. Therefore, the hypothetical and disjunctive combinations are not really special forms of the judgement, for they are applied only to judgements already completed, in which the combination of the concepts remains unchanged, namely the categorical. But they again connect these judgements, since the hypothetical form expresses their dependence on one another, and the disjunctive their incompatibility. But mere concepts have only one kind of relation to one another, namely those relations expressed in the categorical judgement. The fuller determination, or the subspecies of this relation, are the intersection and the complete separateness of the concept-spheres, and thus affirmation and negation. Out of these Kant has made special categories under quite a different title, that of quality. Intersection and separateness again have subspecies, according as the spheres lie within one another completely or only partially, a determination constituting the quantity of the judgements. Out of these Kant has again made a quite special title of categories. Thus he separated what is quite closely related and even identical, namely the easily surveyed modifications of the only possible relations of mere concepts to one another; on the other hand, he united under this title of relation that which is very different.
Categorical judgements have as their metalogical principle the laws of thought of identity and contradiction. But the ground of the connexion of concept-spheres giving truth to the judgement, that is nothing but this connexion, can be of a very varied nature, and, as a result of this, the truth of the judgement is either logical, or empirical, or transcendental, or metalogical. This has already been discussed in the introductory essay, §§ 30-33, and need not here be repeated. But it follows from this how very different the immediate kinds of knowledge can be, all of which exhibit themselves in the abstract through the combination of the spheres of two concepts as subject and predicate, and that we cannot by any means set up a single function of the understanding as corresponding to and producing it. For example, the judgements: "Water boils"; "The sine measures the angle"; "The will decides"; "Employment distracts"; "Distinction is difficult," express through the same logical form the most varied kinds of relations. From this we obtain once more the sanction, however wrong the beginning, to place ourselves at the standpoint of abstract knowledge, in order to analyse direct, intuitive knowledge. For the rest, the categorical judgement springs from a knowledge of the understanding proper, in my sense, only where a causality is expressed through it; but this is the case also with all judgements expressing a physical quality. For if I say: "This body is heavy, hard, fluid, green, sour, alkaline, organic," and so on, this always expresses its action or effect, and thus a knowledge that is possible only through the pure understanding. Now after this knowledge, like much that is quite different from it (e.g., the subordination of highly abstract concepts), has been expressed in the abstract through subject and predicate, these mere relations of concepts have been transferred back to knowledge of perception, and it has been supposed that the subject and predicate of the judgement must have a special correlative of their own in perception, namely substance and accident. But later on I shall clearly show that the concept "substance" has no other true content than that of the concept "matter." Accidents, however, are quite synonymous with kinds of effects, so that the supposed knowledge of substance and accident is still always that of the pure understanding of cause and effect. But how the representation of matter really arises is discussed partly in our first book, § 4, and still more clearly in the essay On the Principle of Sufficient Reason at the end of § 21. To some extent we shall see it still more closely when we investigate the principle that substance is permanent.
(c) The disjunctive judgements spring from the law of thought of the excluded middle, which is a metalogical truth; they are therefore entirely the property of pure reason, and do not have their origin in the understanding. The deduction of the category of community or reciprocal effect from them, however, is a really glaring example of the acts of violence on truth which Kant ventures to commit, merely in order to satisfy his love for architectonic symmetry. The inadmissibility of that deduction has already often been rightly censured, and has been demonstrated on various grounds, especially by G. E. Schulze in his Kritik der theoretischen Philosophie and by Berg in his Epikritik der Philosophie. What actual analogy is there in fact between the problematical determination of a concept by predicates that exclude one another, and the idea of reciprocal effect? The two indeed are quite opposed, for in the disjunctive judgement the actual statement of one of the two terms of division is necessarily at the same time an elimination of the other. On the other hand, if we imagine two things in the relation of reciprocal effect, the statement of the one is necessarily the statement of the other also, and vice versa. Therefore the actual logical analogue of reciprocal effect is unquestionably the circulus vitiosus, for in it, just as ostensibly in the case of reciprocal effect, what is established is also the ground, and conversely. And just as logic rejects the circulus vitiosus, so also is the concept of reciprocal effect to be banished from metaphysics. For I now intend quite seriously to prove that there is no reciprocal effect at all in the proper sense, and that this concept, so extremely popular precisely on account of the indefiniteness of the idea, appears on closer consideration to be empty, false, and invalid. First of all, let us recall what causality in general is, and, to assist in this, let us look up my discussion about it in the introductory essay, § 20, also in my essay On the Freedom of the Will, chap. 3, pp. 27 seq. (2nd ed., pp. 26 seq.), and finally in the fourth chapter of the second volume of the present work. Causality is the law according to which the states or conditions of matter that appear determine their positions in time. With causality it is a question merely of states or conditions, in fact, really only of changes, and not of matter as such or of persistence without change. Matter as such is not under the law of causality, for it neither comes into being nor passes away; thus the whole thing, as we commonly say, does not come under this law, but only the states or conditions of matter. Further, the law of causality has nothing to do with permanence, for where nothing changes there is no producing of effects and no causality, but a continuing state of rest. If such a state or condition is changed, then the newly arisen state is again either permanent, or it is not, and it at once produces a third condition or state. The necessity with which this happens is just the law of causality, which is a form of the principle of sufficient reason, and thus cannot be further explained, since the principle of sufficient reason is the very principle of all explanation and all necessity. From this it is clear that the existence of cause and effect is closely connected with, and necessarily related to, the sequence of time. Only in so far as state A precedes state B in time, but their succession is necessary and not an accidental one, in other words, is no mere sequence but a consequence -- only to this extent is state A the cause and state B the effect. But the concept of reciprocal effect contains this, that each is cause and each is effect of the other; but this is equivalent to saying that each of the two is the earlier and the later at the same time, which is absurd. For that both states are simultaneous, and indeed necessarily simultaneous, cannot be accepted, since, as they necessarily belong together and are simultaneous, they constitute only one state. The enduring presence of all its determinations is certainly required for the persistence of this state, but then there is no longer any question of change and causality, but of duration and rest. Nothing is said except that, if one determination of the whole state is changed, the resultant new state cannot continue, but becomes the cause of the change of all the other determinations of the first state also, whereby a new, third state appears. All this happens merely in accordance with the simple law of causality, and does not establish a new law, that of reciprocal effect.
I also positively assert that the concept of reciprocal effect cannot be illustrated by a single example. All that we should like to pass off as such is either a state of rest, to which the concept of causality, having significance only in regard to changes, finds no application whatever; or it is an alternating succession of states of the same name that condition one another, for the explanation of which simple causality is quite sufficient. An example of the first class is afforded by a pair of scales brought to rest by equal weights. There is no effect at all here, for there is no change; it is a state of rest; gravity acts, uniformly distributed, as it does in every body supported at its centre of gravity, but it cannot manifest its force through any effect. That the taking away of one weight produces a second state that at once becomes the cause of a third, namely the sinking of the other scale, happens according to the simple law of cause and effect. It requires no special category of the understanding, not even a special name. An example of the other class is the continuous burning of a fire. The combination of oxygen with the combustible body is the cause of the heat, and the heat again is the cause of the renewed occurrence of that chemical combination. But this is nothing but a chain of causes and effects, the alternate links of which, however, bear the same name. The burning A produces free heat B; this produces a new burning C (i.e., a new effect having the same name as the cause A, but not individually the same with it); this produces a new heat D (which is not really identical with the effect B, but is the same only according to the concept, in other words, it has the same name as B), and so on indefinitely. A good example of what in ordinary life is called reciprocal effect is afforded by a theory of deserts given by Humboldt (Ansichten der Natur, second edition, vol. II, p. 79). In sandy deserts it does not rain, but it rains on the wooded mountains that border them. The cause is not the attraction of the clouds by the mountains, but the column of heated air, rising from the sandy plain, which prevents the particles of vapour from disintegrating, and drives the clouds upwards. On the mountain range the vertically rising current of air is weaker, the clouds descend, and the rainfall ensues in the cooler air. Thus want of rain and the absence of plants in the desert stand in the relation of reciprocal effect. It does not rain, because the heated surface of sand radiates more heat; the desert does not become a steppe or prairie, because it does not rain. But obviously we have again here, as in the above example, only a succession of causes and effects of the same names, and absolutely nothing essentially different from simple causality. It is just the same with the swinging of a pendulum, and even, in fact, with the self-maintenance of the organic body, where every state likewise produces a new one. This state is of the same kind as the one by which it was itself brought about, but individually it is new. Only here the matter is more complicated, since the chain no longer consists of links of two kinds, but of links of many kinds, so that a link of the same name recurs only after several others have intervened. However, we always see before us only an application of the single and simple law of causality which affords the rule of the sequence of states or conditions, but not something that needs to be comprehended by a new and special function of the understanding.
Or will it be said as a proof of the concept of reciprocal effect that action and reaction are equal to each other? But this is to be found precisely in what I urge so strongly, and have discussed at length, in the essay On the Principle of Sufficient Reason, namely that the cause and the effect are not two bodies, but two successive states of bodies. Consequently, each of the two states also implicates all the bodies concerned, and hence the effect, i.e., the newly appearing state, e.g., in the case of impact, extends to both bodies in the same proportion; therefore the impelled body undergoes just as great a change as does the impelling body (each in proportion to its mass and velocity). If we choose to call this reciprocal effect, then absolutely every effect is a reciprocal effect, and no new concept arises on this account, still less a new function of the understanding for it, but we have only a superfluous synonym for causality. Kant, however, thoughtlessly expresses just this view in the Metaphysical Rudiments of Natural Science, where the proof of the fourth proposition of mechanics begins: "All external effect in the world is reciprocal effect." Then how are different functions to lie a priori in the understanding for simple causality and for reciprocal effect; in fact, how is the real succession of things to be possible and knowable only by means of causality, and their coexistence only by means of reciprocal effect? Accordingly, if all effect is reciprocal effect, succession and simultaneity would be the same thing, and consequently everything in the world would be simultaneous. If there were true reciprocal effect, then the perpetuum mobile would also be possible, and even a priori certain. On the other hand, the a priori conviction that there is no true reciprocal effect and no form of the understanding for such an effect, is the basis for asserting that perpetual motion is impossible.
Aristotle also denies reciprocal effect in the strict sense, for he remarks that two things can indeed be reciprocally causes of each other, but only in so far as we understand this in a different sense of each, for example, that one thing acts on the other as motive, but the latter acts on the former as the cause of its movement. Thus we find the same words in two passages: Physics, Bk. ii, c. 3, and Metaphysics, Bk. v, c. 2. [x]. (Sunt praeterea quae sibi sunt mutuo causae, ut exercitium bonae habitudinis, et haec exercitii: at non eodem modo, sed haec ut finis, illud ut principium motus.) [19] Moreover, if he assumed a reciprocal effect proper, he would introduce it here, for in both passages he is concerned with enumerating all the possible kinds of causes. In the Posterior Analytics, Bk. ii, c. 11, he speaks of a rotation of causes and effects, but not of a reciprocal effect.
(4) The categories of modality have the advantage over all the others, since what is expressed through each of them actually corresponds to the form of judgement from which it is derived. With the other categories this is hardly ever the case, since they are usually deduced from the forms of judgement with the most arbitrary violence.
Therefore, that the concepts of the possible, of the actual, and of the necessary give rise to the problematical, the assertory, and the apodictic forms of judgement, is perfectly true; but that those concepts are special, original cognitive forms of the understanding incapable of further derivation, is not true. On the contrary, they spring from the single form of all knowledge, which is original and therefore known to us a priori, namely the principle of sufficient reason; and in fact knowledge of necessity springs directly from this. On the other hand, only by applying reflection to this do the concepts of contingency, possibility, impossibility, and actuality arise. Therefore all these do not in any way originate from one faculty of the mind, the understanding, but arise through the conflict of abstract knowledge with intuitive, as will be seen in a moment.
I maintain that to be necessary and to be consequent from a given ground or reason are absolutely reciprocal concepts, and completely identical. We can never know or even think anything as necessary, except in so far as we regard it as the consequent from a given ground or reason. The concept of necessity contains absolutely nothing more than this dependence, this being established through another thing, and this inevitably following from it. Thus it arises and exists simply and solely by applying the principle of sufficient reason. Therefore, according to the different forms of this principle, there are a physically necessary (the effect from the cause), a logically necessary (through the ground of knowing, in analytical judgements, syllogisms, and so on), a mathematically necessary (according to the ground of being in space and time), and finally a practically necessary. With this last we wish to express not some determination through a so-called categorical imperative, but the necessarily appearing action with the given empirical character according to the motives presented to it. But everything necessary is so only relatively, namely on the presupposition of the ground or reason from which it follows; therefore absolute necessity is a contradiction. For the rest, I refer to § 49 of the essay On the Principle of Sufficient Reason.
The contradictory opposite, in other words, the denial of necessity, is contingency. The content of this concept is therefore negative, and so nothing more than absence of the connexion expressed by the principle of sufficient reason. Consequently even the contingent is always only relative; thus it is contingent in relation to something that is not its ground or reason. Every object, of whatever kind it be, e.g., every event in the actual world, is always at the same time both necessary and contingent; necessary in reference to the one thing that is its cause; contingent in reference to everything else. For its contact in time and space with everything else is a mere coincidence without necessary connexion; hence also the words chance, contingency, [x] , contingens. Therefore an absolute contingency is just as inconceivable as an absolute necessity, for the former would be just an object that did not stand to any other in the relation of consequent to ground. The inconceivability of such a thing, however, is precisely the content of the principle of sufficient reason negatively expressed. This principle, therefore, would first have to be overthrown if we were to conceive an absolute contingency. But then this itself also would have lost all meaning, for the concept of the contingent has meaning only in reference to that principle, and signifies that two objects do not stand to each other in the relation of ground to consequent.
In nature, in so far as this is representation of perception, everything that happens is necessary, for it proceeds from its cause. If, however, we consider this individual thing in relation to everything else that is not its cause, we recognize it as contingent; but this is already an abstract reflection. Now if further, in the case of an object of nature, we abstract entirely from its causal relation to everything else, and hence from its necessity and contingency, then the concept of the actual comprehends this kind of knowledge. In the case of this concept we consider only the effect, without looking about for the cause, in reference to which we should otherwise have to call it necessary, and in reference to everything else contingent. All this rests ultimately on the fact that the modality of the judgement indicates not so much the objective quality of things as the relation of our knowledge to that quality. But as in nature everything proceeds from a cause, everything actual is also necessary; yet only in so far as it is at this time, in this place; for only thus far does determination through the law of causality extend. But if we leave nature of perception, and pass over to abstract thinking, we can in reflection represent to ourselves all the laws of nature, known to us partly a priori, partly only a posteriori. This abstract representation contains all that is in nature at any time, in any place, but with abstraction from every definite place and time; and in just this way, through such reflection, we have entered the wide realm of possibility. But what finds no place even here is the impossible. It is obvious that possibility and impossibility exist only for reflection, for the abstract knowledge of our faculty of reason, not for the knowledge of perception, although it is the pure forms of such knowledge which suggest to our reason determination of the possible and the impossible. According as the laws of nature, from which we start when thinking of the possible and the impossible, are known a priori or a posteriori, is the possibility or impossibility metaphysical or only physical.
From this exposition, which requires no proof because it rests directly on knowledge of the principle of sufficient reason and on the development of the concepts of the necessary, the actual, and the possible, it is clear enough how entirely groundless is Kant's assumption of three special functions of the understanding for those three concepts; here again we see that he did not let himself be disturbed by any scruple in achieving his architectonic symmetry.
In addition to this, however, there is also the very great mistake, namely his confusion with each other of the concepts of necessary and contingent, of course after the example of previous philosophy. This earlier philosophy misused abstraction in the following way. It was obvious that that of which the ground is set, follows inevitably, in other words, cannot fail to be, and so necessarily is. But men held to this last determination alone, and said that that is necessary which cannot be otherwise, or whose opposite is impossible. But they disregarded the ground and the root of such necessity, overlooked the relativity of all necessity that results therefrom, and thus made the utterly inconceivable fiction of an absolutely necessary, in other words, of something whose existence would be as inevitable as the consequent from the reason or ground, yet which would not be consequent from a ground, and would thus depend on nothing. This addition is just an absurd petitio principii, since it is contrary to the principle of sufficient reason. Now starting from this fiction they declared, in diametrical opposition to the truth, that everything established through a ground or reason was contingent, since they looked at the relative nature of its necessity, and compared this with the entirely fictitious absolute necessity that is self-contradictory in its concept. [20] Now Kant also retains this fundamentally perverse definition of the contingent, and gives it as explanation: Critique of Pure Reason, V, pp. 289-291; 243 (V, 301); 419, 458, 460 (V, 447, 486, 488). Here indeed he falls into the most obvious contradiction with himself, since he says on p. 301: "Everything contingent has a cause," and adds: "That is contingent, of which the non-existence is possible." But whatever has a cause cannot possibly not be; therefore it is necessary. For the rest, the origin of the whole of this false explanation of the necessary and the contingent is to be found in Aristotle in De Generatione et Corruptione, Bk. ii, chaps. 9 and 11, where the necessary is declared to be that of which the non-existence is impossible; opposed to it is that of which the existence is impossible. And between these two lies that which can be and also not be -- hence that which arises and passes away, and this would then be the contingent. According to what has been said above, it is clear that this explanation, like so many of Aristotle's, has resulted from sticking to abstract concepts without going back to the concrete and perceptible, in which, however, lies the source of all abstract concepts, and by which they must therefore always be controlled. "Something of which the non-existence is impossible" can certainly be thought in the abstract, but if we go with it to the concrete, the real, the perceptible, we find nothing to illustrate the thought, even only as something possible -- as merely the aforesaid consequent of a given ground, whose necessity, however, is relative and conditioned.
I take this opportunity to add a few more remarks on these concepts of modality. As all necessity rests on the principle of sufficient reason, and on this very account is relative, all apodictic judgements are originally, and in their ultimate significance, hypothetical. They become categorical only by the introduction of an assertory minor, hence in the consequent of a syllogism. If this minor is still undecided, and this indecision is expressed, this gives the problematical judgement.
What in general (as rule) is apodictic (a law of nature), is always in reference to a particular case only problematical, since first the condition which puts the case under the rule must actually appear. Conversely, what in the particular as such is necessary (apodictic) (every particular change necessary through its cause), is again in general, and expressed universally, only problematical, since the cause that appears concerns only the particular case, and the apodictic, always hypothetical, judgement invariably states only universal laws, not particular cases directly. All this has its ground in the fact that the possible exists only in the province of reflection and for our faculty of reason, the actual in the province of perception and for our understanding, the necessary for both. In fact, the distinction between necessary, actual, and possible really exists only in the abstract and according to the concept; in the real world all three coincide in one. For all that happens, happens necessarily, because it happens from causes, but these themselves in turn have causes, so that the whole course of events in the world, great as well as small, is a strict concatenation of what necessarily takes place. Accordingly, everything actual is at the same time something necessary, and in reality there is no difference between actuality and necessity. In just the same way there is no difference between actuality and possibility, for what has not happened, in other words has not become actual, was also not possible, since the causes without which it could never take place have themselves not happened, nor could they happen, in the great concatenation of causes; thus it was an impossibility. Accordingly, every event is either necessary or impossible. All this holds good merely of the empirically real world, in other words, of the complex of individual things, and thus of the wholly particular or individual as such. On the other hand, if by means of our faculty of reason we consider things in general, comprehending them in the abstract, then necessity, actuality, and possibility are again separated. We then know everything as generally possible according to a priori laws belonging to our intellect, and that which corresponds to the empirical laws of nature as possible in this world, even if it has never become actual; thus we clearly distinguish the possible from the actual. The actual is in itself always also necessary, but it is understood as being such only by the man who knows its cause; apart from this, it is and is called contingent. This consideration also gives us the key to that contentio [x] [21] between the Megaric Diodorus and Chrysippus the Stoic, which Cicero mentions in his book De Fato. Diodorus says: "Only what becomes actual has been possible, and all that is actual is also necessary," On the other hand, Chrysippus says: "Much that is possible never becomes actual, for only the necessary becomes actual." We can explain this as follows: Actuality is the conclusion of a syllogism for which possibility provides the premisses. Yet for it not only the major, but also the minor is required; only the two give complete possibility. Thus the major gives a merely theoretical, general possibility in abstracto; but this in itself still does not make anything possible at all, in other words, capable of becoming actual. For this the minor is still needed, which gives the possibility for the particular case, since it brings the case under the rule. Precisely in this way the case at once becomes actuality. For example:
Maj. All houses (consequently mine also) can be destroyed by fire.
Min. My house is catching fire.
Concl. My house is being destroyed by fire.
For every general proposition, and hence every major, establishes things with regard to actuality only under a presupposition, and consequently hypothetically; for example, the ability to be destroyed by fire has the catching fire as a presupposition. This presupposition is brought out in the minor. The major always loads the gun, but only when the minor applies the fuse does the shot, i.e., the conclusion, follow. This holds good everywhere of the relation of possibility to actuality. Now as the conclusion, which is the assertion of actuality, follows necessarily, it is clear from this that everything that is actual is also necessary; this can also be seen from the fact that necessity means simply being consequent of a given ground or reason. With the actual this ground is a cause; hence everything actual is necessary. Accordingly, we see the concepts of the possible, the actual, and the necessary coincide, and not merely the last presuppose the first, but also vice versa. What keeps them apart is the limitation of our intellect through the form of time; for time is the mediator between possibility and actuality. The necessity of the individual event can be seen perfectly from the knowledge of all its causes, but the coincidence of all these different causes, independent of one another, seems to us to be contingent; in fact their independence of one another is just the concept of contingency. However, as each of them was the necessary consequence of its cause, and the chain of causes is beginningless, it is clear that contingency is a merely subjective phenomenon, arising out of the limitation of the horizon of our understanding, and is just as subjective as is the optical horizon in which the heavens touch the earth.