CHAPTER VII

THE CONTINUANT

§ 1. We have found in our analysis of the nature and determination of occurrences that some link, besides mere temporal and spatial connection, must exist between one occurrence and another in order that the first may be conceived as determinative of the second. This link I have called the substantive continuant, and in this chapter we shall examine the notion in detail, and show how it differs from the traditional conception of substance.

The simplest and most obvious illustration of the continuant is the case of the moving particle: thus, if two movements, defined in character by direction and velocity, and defined also by reference to the period of time and region of space within which each takes place, are to be conceived as connected, in the sense that the character, date and location of the one is determinative of the character, date and location of the other, then such a connection can only be presumed if the same material continuant is existentially manifested in the two movements. Apart from the introduction of the continuant, this simple example serves to illustrate the way in which identity and difference is involved in causality. We speak of two movements, and briefly call the one cause and the other effect. Inasmuch as the movements are two, they cannot be identical; so that it may be laid down as the first and most indubitable [79] principle of causality that, whatever other subtle relations there may be between cause and effect, the relationof non-identity is to be unequivocally asserted. Hence, before the movements in question are connected as cause and effect, they must first be distinguished as other or two; and since time and place are the only conditions of otherness which have been conceived by the human mind in regard to physical phenomena, the movements in order to be conceived as two, must occupy either different periods of time and the same region of space, or different regions of space and the same period of time, or different periods of time and different regions of space. We will suppose that the two movements, connected as cause and effect, are referred to different periods of time and to different regions of space, and proceed to consider their characterisation as regards direction and velocity. In the very simplest case afforded by science of causal relation between movements, the direction and velocity of the movement called cause is identical with the direction and velocity of the movement called effect; in this case, therefore, cause and effect are non-identical as regards temporal and spatial reference, but identical as regards characterisation. Turning now from the adjectival characterisation of the occurrences to their substantival connection, our illustration may be expressed in terms of the first law of motion, as follows: -- So far as the movement of a particle within one period of time is causally determinative of its movements within another period of time, the direction and velocity of movement is the same within these two periods. Here the two movements which are causally connected are movements of one and the same particle; so that [80] substantival identity is a notion essential to the understanding of the formula. It is to this substantival identity that I refer when I speak of a continuant.

§ 2. For logical purposes it replaces the term substance, familiar in metaphysics; but the various unfounded or a priori characteristics which philosophers have attributed to substance must be carefully separated from the essential logical residuum, and rejected from the notion of the continuant. Thus, in the first place, the conception of continuance has been extended into the infinite future and the infinite past. In my view, on the other hand, the application of continuance must be strictly limited to the periods of time in reference to which we can speak of change; that is, so far as we are justified in speaking of a state or condition as changing when we pass in thought from one period of time to another, so far are we justified in conceiving of the same entity or continuant as preserving its existence throughout the two given periods. This does not warrant us in asserting its existence either before or after these two periods. In physics, it is true that scientists have found it convenient to postulate an indefinitely prolonged existence into the past and future of the ultimate atoms which constitute matter; but this, has no general logical or philosophical warrant, any more than there is philosophical or logical warrant for immortality

§ 3. The next way in which metaphysicians have characterised any continuant or substance in an unwarrantable fashion, is by maintaining that amid all the alterations of state or condition which the substance undergoes, there are some one or more characters which [81] continue to be manifested unchanged. This position may be held in a more or less crude form. From my point of view, what is important to point out, is merely that substantival continuance does not necessarily imply any adjectival changelessness. When philosophers like Locke and Hume sought for significance in the conception of identity, as substantive continuance used to be called, they were continually guilty of confusing continued existence of the same substantive entity with qualitative or adjectival identity of character or state. Failing to find this qualitative identity, Hume explicitly rejected substantival identity; and those who opposed Hume held equally with him that substantive identity could not be maintained except in so far as qualitative identity could be established. I repeat then, that the conception of substantive continuance does not by itself carry with it the implication of unchanged character through the period of time to which the substantive continuance applies. As in the matter of absolute temporal permanence, so also in this question of unchanged character, the physicists have found it convenient to postulate in various forms an unchanged continuance of character in the atoms or compound bodies which constitute the matter of the universe. Though I have here called these postulates of the physicist assumptions, I do not wish to deny that some of them may have inductive warrant; to this we shall have to return when we consider scientific induction in detail.

§ 4. The third and last a priori attitude towards the notion of the continuant must be briefly treated. This is the contention that the ultimate continuant is simple and not compound. On this subject it is perhaps of [82] greatest importance at the present day to distinguish between compound, in the sense of a whole consisting of parts, and compound in the sense of involving inner causal or dynamic interaction. The former conception raises no serious problem, the continued identity of the whole being obviously involved in the continued identities of the parts. It is possible however to conceive of a compound entity which continues to preserve its identity through change of time, although none of the parts, which appear from time to time to constitute the whole, can be said to preserve their several identities. This may conceivably be explained by exhibiting a law or principle in accordance with which the compound continuant develops a changing character by means of the instrumentality of the dynamic interactions amongst the parts or components which from time to time constitute so to speak the substantival material of which the compound continuant is composed. Thus the law or principle according to which the character of the continuant at one time can be exhibited as depending upon its character at another time, may be the ground for asserting continued existential identity, although the material components of this continuant are not themselves continuant.

§ 5. We began our exposition of the continuant by an illustration from physical science, showing how the physical continuant is involved in the simple formula known as the first law of motion. We shall now bring forward an illustration of approximately equal simplicity from the psychical sphere. In the physical illustration we included reference to space as well as to time; in our psychical illustration we shall drop, for the present, any reference to space. If a sensation characterised in some [83] way, and a thought process, also characterised in some way, occur one within some period of time, and the other within the same or a different period of time, then the character and date of the sensation can only be conceived as determinative of the character and date of the thought process if, in the simplest case, the same psychical continuant is existentially manifested both in the sensation and in the thought process. Precisely as in the case of physical phenomena, change or alteration in time does not mean the replacement of a sense-experience of red, say, referred to one period of time, by a sense-experience of blue referred to another period of time; for the mere reference of differently characterised experiences to different periods of time does not constitute what we call change or alteration. Here as in the case of the physical continuant, we can only speak of change or alteration by conceiving of an existent which continues to exist within both the periods of time to which the change refers; and it is for this reason that we call such an existent a continuant.

But the notion of change, when applied to the psychical continuant, raises a peculiar problem when we consider the different kinds of experience referable to one and the same continuant: thus we may put the question whether it is correct to speak of a change of state when, for example, we refer a sensation to one date, and a thought or volition to another date ; or when we refer say a colour sensation to one date and a sound sensation to another date. The mere fact that any colour sensation is by definition different from any sound sensation, and still more that any sensation is different from any thought or from any volition, does not appear to justify us in [84] speaking of change or alteration when such phases of experience are referred to different dates. On the other hand we should with less hesitation speak of change or alteration when the differing experiences come under the same determinable. A sensation of red followed by a sensation of blue -- blue and red being determinates under the same determinable, colour -- would appear to illustrate the notion of change of state more correctly than a sensation of red followed by a sensation of noise, or by a thought about geometrical relations, or again by a voluntary decision to get out of bed. Now the real reason why we apply the word change preferably to the first case and not to the second is because we suppose that the blue sensation has replaced the red sensation, so that at the time that the blue is manifested, the red has ceased to be manifested. It is thus the cessation of one character of our experience, and its replacement by another, that constitutes the essence of change.

§ 6. The above analysis helps towards a solution of the problem as to what it is that can be said to change. On the one hand, it cannot be the continuant itself, nor any of its properties, since these are asserted to be constant throughout the period of time to which the process of change is referred. Neither can it be the manifestations, dated at time-points, which can be said to change, since these merely replace one another from instant to instant. The clue to the problem is to be found in the theory of the determinable. The character of each dated manifestation is determinate, and a change implies always that the determinate character of the one manifestation at one instant is replaced at a [85] subsequent instant by a manifestation having a different determinate character under the same determinable. Thus we speak of temperature or colour or size or shape, etc., as changing or remaining constant during a certain period of time; it is therefore the manifestation -- not of a determinate -- but of a determinable that may be said to change. But further, the idea of change involves not only the adjectival determinable, but also the substantival determinandum; for change would have no meaning unless there were a continuant, which was necessarily manifested in a mode characterised by one or another determinate value of a determinable. Thus the substantival determinandum is conceived as continually manifesting one or another determinate character under the same determinable, and being potentially manifestable in a mode characterised by any value of the determinable. This aspect of the nature of change leads to the conception of that which determines this potentiality to become an actuality; in other words the conception of change brings with it the conception of causal determination.

To prevent one minor confusion it is necessary to point out that what holds of change proper holds also of the continuance of the manifestation unchanged; for the fact of continuance as well as of change requires the assignment of a cause. The fact that the popular mind demands only an explanation of change which will assign the event or occurrence operating as cause, is accounted for by familiarity with unchanged continuance in many manifestations. Actually the preceding unchanged continuance constitutes in such familiar cases the cause of the subsequent unchanged continuance; but it is only [86] when this continuance is interrupted that the question of the cause of interruption is generally raised. For this reason the conceptions of cause and of change are always supposed mutually to involve one another.

§ 7. The simple illustrations which we have brought forward of a physical continuant and a psychical continuant, have served to introduce the view that the two notions, familiarly known in philosophy as substance and causality, are mutually dependent the one upon the other. No adequate account of causality can be given without reference to the conception of substance, i.e. of an existent continuant, physical or psychical; and on the other hand, we can only assign properties to the substance or continuant by defining the modes according to which it is existentially manifested as a causal agent or re-agent. Thus what is called a property of a continuant is not an actually manifested character, but it defines what characters would be phenomenally manifested when certain assignable conditions occur. For example, the elasticity of an extensible string illustrates a property which we attribute to the string; it defines in general terms the degree of length which would be attained were the string exposed to a certain tensional force. A property, therefore, expresses a definable group of manifestations -- not as actual -- but as potential. The general formula for expressing the property of a continuant c assumes the shape: if a certain occurrence defined as p were to take place, in which the continuant c is patient, then a phenomenal manifestation defined as q would occur which is determined by the nature of c.

It should be observed that continuants -- i.e. in ordinary language things -- are classified according to their [87] properties; such familiar terms as solid, liquid and gaseous, for example, do not describe phenomena as actual but as potential, and they are typical of an innumerable host of terms in familiar use. It is not altogether easy to distinguish those adjectives in common use which denote respectively actualities and potentialities of manifestation ; in fact the problem of their distinction raises points of philosophical importance. As regards physical continuants, in predicating such adjectives as those of position, shape or size, or more generally of spatial configuration, I shall assume that we are predicating actualities; but these are the only adjectives descriptive of merely physical phenomena which are regarded unanimously by physicists as actually manifested. I myself hold that there is, besides spatial configuration and motion, another physically actualised mode, namely force as denned in statics. Physicists appear to me to maintain that, where equilibrium exists, what has been called force is merely an indication of potentiality for movement; so that only the energy of movement is actual, and in static condition force is held to be a myth. But in my view, static force represents a real condition of a body; e.g. when a heavy body is in equilibrium on a horizontal surface, the force called pressure actually exists, and is not a mere measure of what would take place if free motion were permitted. One evidence for this view is the recognised association of strain with stress; i.e. stress is a particular example of force which is correlated with strain; this latter being a geometrical conception. The formula according to which strain and stress are mutually connected and yet distinguished, so that they stand to one another as cause to effect or as [88] effect to cause, appears to me to place them both in the category of actualities, since a cause cannot be said to be in operation if we conceive it as a mere potentiality. I should therefore include in what used to be called the primary qualities of matter, besides spatial configuration and motion, resistant force, this phrase being preferable, in my opinion, as well as of wider application than the dubious term 'impenetrability.' My definition of the so-called primary qualities is, therefore, that they denote the adjectives or relations in terms of which actual physical phenomena can be described; whereas the so-called secondary qualities are properties, inasmuch as they denote potentialities for producing sensational effects. Thus in describing the colour of the surface of a body, we may be defining something physically actual, but we are also most certainly defining, besides, what is merely potential; viz., that if a luminous centre, such as the sun, is in such spatial relation as to radiate energy to the surface of the body in question, then assignable parts of this energy will be absorbed at the surface, and another assignable part emitted. Correlated with this physical potentiality of the body is a psychical potentiality, which must also be presented partly in spatial terms; viz., that if a living organism susceptible to light-impressions be in appropriate spatial relations to the body, there will be a visual sensation to which the name red primarily and properly applies.

The varied applications in physics of such terms as coefficient or index are obvious illustrations of what, from the logical standpoint, we regard as potentialities in contrast to actual physical process. Such terms denote what are commonly called constants; and the [89] common use of this term will serve to illustrate the notion of a property. Most so-called constants should more strictly be called 'variable invariables,' for only a few of them are absolutely invariable; they assume different values for different classes of bodies, although they may remain potentially invariable when applied to any one the same body. This illustrates the point that in attributing a property to a body, we imply that a certain formula can be asserted of the processes in which that body may be concerned, which formula remains unchanged on the different occasions in which the processes may take place. It further illustrates the point that bodies may be classified according to the different values of their determinable properties, as represented by the different values of the so-called constants. As an example of this, let us take the adjective 'ruminant.' So far as we predicate this adjective, we are certainly implying the existence of an organ of specific character, which could be denned anatomically or structurally in terms of spatial arrangement and, say, mechanical pressure. In addition to these elements of definition, we should further assume that the organ is persistently functioning, though the determinate mode in which it is functioning would be changed from time to time in accordance with a general formula defining the potentialities of the organ to function whenever assignable conditions may be actualised.

§ 8. So far we have been discussing the continuant chiefly from the point of view of deduction; but I propose now to treat it from the point of view of primitive induction, and to exhibit a constructive process by which, prior to anything that could be called classification of organised [90] or of unorganised bodies, mankind have put phenomena into groups. The symbols P_l, P₂, P₃ . . . will represent phenomena characterised by P, the suffixes being understood to indicate some kind of order -- either temporal or spatial or both -- which we shall speak of as a nexus. The first motive for grouping phenomena is the observation of some such nexus: thus when an order from P_l to P₂ to P₃ to P₄ . . . has been repeatedly observed to be maintained under a variety of circumstances which, in some of the different recurrences have been constant and in others have varied, then these phenomena have been grouped as manifestations of an existent agent. This observed uniformity in the temporal succession of phenomena is then inferentially extended to apply to other assumed phenomena, regarded as modes in which a single agent manifests its continued existence. The above conception of nexus therefore involves not only a preserved temporal order of phenomena, but also reference of these phenomena to a single continuant. In physical, as distinguished from psychical, phenomena there is, in addition to the temporal nexus, a spatial nexus, between phenomena presented in spatial contiguity, representing modes in which a single material body or occupant is spatially manifested. Thus when the spatial order of such characters as P_l, P₂, P₃ . . . is preserved throughout exterior processes which are changing or remaining unchanged, the group of characters is with special impressiveness taken to constitute a unity, and conceived as referable to a single occupant which maintains its form of spatial nexus, however exterior conditions may be altering.

The unity of the occupant is, however, not stably maintained with the same degree of permanence as [91] is attached to the temporal succession amongst the manifestations of the continuant. In this respect spatial occupants fall into different classes according to the degree in which they preserve the form of their spatial nexus: the bodies which preserve their spatial nexus in the highest degree are called solids; those that preserve it in less degree are called liquids; and those in which the spatial nexus can hardly be said to be preserved at all are called gases. These degrees of spatial nexus actually depend upon the power of bodies to sustain tension and pressure; bodies which can sustain both are called solids; liquids cannot sustain tension, but only pressure; and gases, which are also unable to sustain tension, would lose their spatial nexus altogether unless pressure, produced by external force, determined the space which they can be made to occupy. Bodies regarded as occupying spaces of defined shape and size can be divided into classes on many different fundamenta divisionis: for instance according as they are unorganised or organised in various different forms, or again according to their chemical composition. A body is classified according to the mode in which it maintains its unity as a whole, this unity consisting in the permanence of mode of presentation of its manifestations, while exterior conditions change or remain constant. When the spatial unity which leads us to conceive of a body as a single whole is dissolved by exterior force which the body is unable to resist, then each of the several parts into which the body becomes separated has its own constitution as a unit; and this constitution may or may not be generically the same as that of the whole into which the parts were previously united. [92]

§ 9. The consideration of the different degrees of spatial nexus exhibited by different spatial occupants, leads to the notion of a single continuant-occupant being at the same time a system of .^-continuants or atoms. These sub-continuants constitute a system, in the sense that they preserve a certain form of spatial nexus either unchanged, or else changing in accordance with a causal formula which expresses both the immanent causality to be attributed to the several sub-continuants, and the transeunt causality to be attributed to the interactions amongst these sub-continuants. For example, the gas that is contained in a vessel is a system of sub-continuants -- viz. the gaseous molecules -- which severally exhibit their own immanent causality, while their spatial nexus, manifested as mutual pressure, exhibits transeunt causality. Again each molecule may be a subsystem consisting of sub-continuants -- viz. atoms -- which taken severally exhibit immanent causality, and taken in combination, transeunt causality. Thus, any system of sub-continuants may be regarded either as a unity or whole, whose changes are determined under the form of immanent causality; or any such whole may be conceived as a system of parts which are themselves continuants, and whose processes are to be distinguished from the transeunt processes involved in their interactions in the larger system. In future, the phrase 'single continuant' is to be understood to include a single system of sub-continuants or sub-occupants. Such a system may be ranged in an order exhibiting higher and higher forms of unity; of which the lowest form is a mechanical system, and the highest a psychical continuant. In a mere mechanism, the system as [93] a whole can be defined in terms of the immanent char acter of its parts, combined with the transeunt causality under which the parts interact; but the higher form o unity exhibited in an organism entails a much more complex interrelationship of immanent and transeuni interactions; thus each part of an organism undergoes processes which follow chemical and mechanical formulae, which remain true even when the organism is dead.

§ 10. The forms of temporal and spatial nexus observed prior to experiment to have been maintained between groups of phenomena, lead to their constructive reference to a single continuant-occupant. Primitive in duction consists in the more or less tacit inference that where such forms of nexus have been found to obtain within the range of observation, the same form or de gree of nexus will be maintained in all cases that have not yet been observed. This may be shortly expressec in the proposition: 'All manifestations of a given single continuant will assume a specific form of nexus'; and the assignment of a substantive-name, therefore, to any manifestation in space and time, is tantamount to a statement of uniformity. It should be specially noted that the range of uniformity here asserted does not extend beyond the manifestations of a single individual continuant. Now this uniformity is not a mere statement of the invariability by which we can infer from certain phenomena other phenomena in spatial and temporal connection with these, but it involves also a statement of causal relation between the phenomena. The causality affirmed is immanent, and is conceived as exhibiting the nature of the individual continuant itself. The phenomena in question may be said to co-exist, the prefix co [94] indicating that they are manifestations of a single continuant; for, as I have elsewhere pointed out, the phrase co-existence of properties does not mean merely temporal simultaneity in the manifestations, but includes precedence and subsequence, and also in some cases spatial relations.

§11. The assertion of causality does not carry with it any implication of a uniformity of co-existence, affirming that the causal formula which applies to the continuant under consideration would hold of other continuants. The nature of any given individual is exhibited in a uniformity of its manifestations, which again, in my view, is a causal uniformity. But this species of uniformity is not one that can be extended in application to the manifestations of other individuals, except so far as we have inductive evidence that the same form of immanent causality that expresses the nature of one individual expresses also the nature of other individuals. The causal uniformity which obtains between the manifestations of a single individual continuant is established by more or less direct and simple problematic methods; whereas in order to establish a uniformity of co-existence, the criteria of probability employed are much more exact and logically elaborate. What I have now to maintain is that there are many conclusions reached by the first method which can be said to have been established with a very high degree of probability, amounting in some cases to practical certitude. They assert persistence in the form under which the manifestations of a single continuant may be subsumed. Induction, on the other hand, which attributes the same form of property to one and to another continuant, cannot attain a high [95] degree of certitude by simple problematic methods; for what enables us to construct a class of continuants so defined, is the employment of more or less scientific method.

§ 12. We will now pass to the consideration of the relation of the continuant or supreme substantive to its occurrents or modes of manifestation. These modes of manifestation have variously defined characters and are connected with one another by temporal and spatial relations. Just as the supreme substantive is called a continuant on the ground that it continues to be manifested throughout an indefinitely prolonged period of time, within which any one manifestation or occurrent is referred to a specific sub-period; so it may also be called an occupant, on the ground that its manifestations occupy an indefinitely extended region of space, within which any one manifestation is referred to a specific sub-region. The determinate sub-period and sub-region together define the relation of one occurrent to others; whereas the time and space, of which these sub-periods and sub-regions are respectively parts, may be predicated referentially to the continuant-occupant itself. Thus when we say that a continuant exists throughout the whole of time, or an occupant throughout the whole of space, we mean that its several different manifestations are to be referred to one or another of the sub-periods or sub-regions within this whole; and if we are concerned with an animate or psychical continuant, the whole of time or the whole of space within which it exists may be itself a sub-period or a sub-region within a larger whole; since there may be some part of time during which it is not manifested in any part of space, and [96] there may be some part of space in which it is not manifested in any part of time. This is one of the ways in which predications (viz. temporal and spatial), which are primarily attached to existent manifestations are transferred to an existent continuant or occupant. We are thus led to enquire what other adjectives can be predicated of a continuant which are derived from the characterisations of its several occurrents or manifestations. To explain this we must conceive of the unity of the continuant as exhibited in causal formulae sym-bolisable as follows:

P = f_P'X, Q = f_Q'X, R = f_R'X, . . . . . .

P = ψ_P(X, Y, Z, . . . T_XT_YT_Z . . .),

§ 13. From the above attempted explanation of the notion of property, it is but a short step to the concept of a continuant; for the main element in the notion of thing or continuant is the permanency of functioning that can be discerned in a series of characterised manifestations, presented in the course of time, as they may be observed in a temporally continuous, or discrete, series of acts. Thus the notion of a continuant is constructed in terms of temporal connection and causal determination, and my particular views on this subject may perhaps be best explained by comparing my account with Kant's exposition in the Critique of Pure Reason. Kant holds that there are certain categories, such as substance and causality, under which we objectify our sense-experiences in an order of time; whereas I prefer to treat substance and causality, not as two separate categories, but as two aspects of a single principle of construction. Again, instead of adding a third category -- namely reciprocity -- to substance and causality, as Kant does, I include reciprocity in my accountof immanent and transeunt causality. But such apparent differences between Kant's exposition and mine are not important, and readers of Kant, by putting together various parts of his exposition, would find at least hints of all that I have said. Thus, when he says with regard to his category of substance, that the idea of change involves the idea of permanence, and when this is supplemented by his schematism of causality under the form of time, his view is seen to be in close accord with my account of the way in which [99] temporal causality and permanency of functioning ente in the notion of a continuant; although the postulate of permanency refers in Kant's exposition to a 'quantum' rather than to the mode of functioning which I attribute to the continuant. In my view this permanency in the mode of functioning is inseparable from the property -- or form of causality -- this form being just that to which permanence is attributed; whereas Kant appears to affirm that the substance itself exists permanently, and that a second permanence is to be attributed to one of its modes of manifestation, namely to it's 'quantum.' This postulate of his was, in fact, an anticipation of the constancy of mass which is a special postulate in physics; but no similar quantitative constancy can be attributed to the higher substantive entities, such as the organism or the experient. I am inclined to attribute Kant's denial of the possibility of rationalising psychology to his rather exclusive consideration of the forms in which the principles of physics can be generalised and formulated in precise mathematical conceptions. Thus my account presents a more general conception of substantive continuance, which applies equally to the notion of a conscious experient on the one hand, and to a hypothetical physical atom on the other. The unity which I ascribe to the continuant is a causal unity of connection between its temporally or spatially separated manifestations; an observed or assumed causal formula, under which the character of these manifestations may be subsumed, is the sole ground for regarding them as manifestations of one and the same continuant. I have also attempted to render clear the difficult conception of the union of permanence with change. It is natural [100] to ascribe change to the modes of manifestation, and permanence to the substance to which these manifestations are referred; but this is an inadequate expression of the antithesis; for, to express the matter accurately, the only things which can be said temporally to exist are the manifestations themselves; thus our first definition of the continuant is that it is merely the sum of all the manifestations. This of course does not mean that manifestations of reality are taken indiscriminately, mentally added together, and their sum called a continuant; what is meant is that certain manifestations of reality, between which a unique kind of relation can be predicated, together constitute a genuine whole or unity, to which the name continuant may be given. This type of relation, which constitutes the unity of a single continuant, is conceived primarily as one of immanent causality, while it is transeunt causality that constitutes the ground for asserting a plurality of non-identical continuants whose manifestations can be said to belong to one universe of reality.

§ 14. All the conceptions expounded in this chapter are virtually denied by a school of philosophers to-day. In particular they regard the conception of change as fictitious, and substitute for it merely differently characterised phenomena referred to non-identical dates. Whenever there is a spatio-temporal nexus between phenomena, the locating and dating of the occurrents is such that these may be conceived as a whole. Such a whole is of the kind which we have described as extensional [Part II, Chapter VII, § 8.], and so far as extensional wholes are admitted by the scientist, no more transcendental conception than [101] that of a whole constituted by the binding relations of time and space is required; and hence the philosophers who reject the conception of a continuant are satisfied to replace it by the notion of such an extensional whole. But the stability of a spatio-temporal nexus cannot, I maintain, be explained without the conception of a continuant, which, in my view, is a priori in the Kantian sense, and not derived from the analysis of experimental data. Given the conception, however, it is a question of mere experience to what set of phenomena the a priori notion is to be applied. In attempting to avoid this conception, it appears to me that my opponents alternate between a purely physical and a supposititious perceptual account of the facts. Thus in one breath they shelve the physical continuant by supposing that the percipient is observing a continuity in the qualitative changes of the object perceived; and while in this way rejecting any physical continuant, they have recourse to a psychical continuant -- namely the percipient. Here I submit that the perception by any individual of certain processes offers no explanation whatever of what in objective reality determines the stability of any given nexus. Then again, on the other hand, when it is urged that the upholders of this view are all along assuming a psychical continuant -- viz. the percipient -- which from their standpoint must be repudiated, they, in effect, retort that it is quite unnecessary to postulate any psychical continuant, inasmuch as the nervous system itself will take the place of the ordinary conception of an ego. Here then they only eliminate the psychical continuant by reinstating the physical continuant.