Cursus Philosophicus Thomisticus, Secunda Pars Artis Logicae

WE ASSUME the distinction between a predicamental accident and a predicable accident. It consists in this: a predicamental accident is set against substance, which is the first category, and so stands only for an accident that inheres. A predicable accident, however, is set against an essential predicate; accordingly, whatever does not belong essentially or necessarily is called a predicable accident, whether it is real or a being of reason, inhering or subsisting.

Porphyry therefore gives three definitions of accident here.

The first is: An accident is “what is present and absent without the corruption of the subject.”

The second is: An accident is “what can belong and fail to belong to the same thing.”

The third is: An accident is “what is neither a genus, nor a species, nor a difference, nor a property, but is always in a subject.”

TWO THINGS MUST BE EXPLAINED about these definitions in this article: First, is accident defined here in the first intention, or as the second intention of this fifth predicable? Second, how are the words of the definition “what is present and absent without the corruption of the subject” to be understood?

On the first point, some think, as Martínez does here in his commentary on the chapter “On Accident,” that a second intention is being defined. For the definition describes something that belongs to accident through the intellect, not in reality. Although being present in one thing and absent from another belongs to an accident in reality, being present in and absent from one and the same subject belongs to it not in reality but by reason—especially in the case of an inseparable accident. Others argue for this view on the ground that, if accident were defined here in the first intention, the definition would not apply to every predicate that can be present and absent. It would not apply to beings of reason, which are not first intentions but second intentions; yet many of them are predicated under the fifth predicable, as when I say, “Animal is a genus.”

Nevertheless, the answer is that the definition given applies to the substrate, or first intention, of this fifth predicable, though it can be adapted to the second intention by adding something. By the substrate of this fifth predicable we mean not only a real first intention, but any accidental predicate, even if it is a being of reason.

The reason is clear: what these definitions explain is found also in an individual accident taken individually. But whatever is found in singulars and extends down to them does not belong only to the second intention of predicability and universality, since that intention does not extend down to singulars. Therefore, accident is defined here in the first intention.

The major premise is proved as follows. Being present in and absent from a subject also belongs to an individual accident, as is clear both in a separable accident, which is at one time present and at another absent in an individual without the subject’s corruption, and in an inseparable accident, on which the subject—that is, the nature in which it inheres—does not depend, even if the individual’s preservation sometimes does. Therefore, what these definitions explain also belongs to individual accidents.

If someone wishes to adapt the definition of accident to the second intention of accident, the phrase “of many” must be understood implicitly. One would then say that an accident is “what is predicated of many as a predicate, and is present and absent apart from the corruption of the subject.”

This makes clear the solution to the opposing argument. We say in reply to the first that being present in and absent from one and the same subject belongs to an accident in reality, both in a separable and in an inseparable accident, because it is connected with the subject contingently, not necessarily. Those words signify nothing other than the contingency of the way it belongs.

To the second argument we say that a first intention can also be found among beings of reason, as with many negations, privations, and extrinsic denominations. One second intention can also serve materially as the substrate of another second intention and receive an accidental denomination from it; in this way it takes on something like the mode of a first intention with respect to what it underlies.

You will object: In this definition, “being present and absent” is taken in the second intention, to mean being affirmed or denied without the corruption of the subject. Therefore, the definition is given through what belongs to a second intention, namely affirmation and negation.

The reply is, first, that this does not prove that the definition is given through the second intention of the universality of this fifth predicable. Rather, it is given through the second intention of affirming and denying, which belongs to the intellect’s second operation. Next, affirmation and negation can be taken in either of two ways: fundamentally, as something belonging to the first intention and found in every accident, even an individual one; or formally, as the second intention of affirming and denying, as we shall explain shortly.

On the second point, authors disagree about how the phrase “is present and absent, etc.” should be understood: does it refer to real presence and absence, or to affirmation and negation? The reason it seems impossible to understand it as real presence and absence is that an inseparable accident cannot really be present and absent, since otherwise it would be separable. But if it is said to be understood in the second intention, as being affirmed and denied, it would follow that every accident is separable, since there is nothing the intellect cannot separate, even a property, without which the subject can readily be understood.

This does not escape the difficulty; it presupposes real presence and absence, because affirmation and negation are either true or false. If false, then it is false that the accident is present and absent without the corruption of the subject. If true, then the matter stands in reality as affirmed: it is present and absent in reality. Thus real presence and absence of an accident are granted. If it is said that “being present and absent” is understood not as affirmation and negation by way of composition, but by way of simple abstraction—understanding one thing without the other—there are two objections. First, this directly contradicts St. Thomas, in the Question on Spiritual Creatures, art. last, ad 7,¹ and Opusculum 48, chapter “On Accident.”² Second, in this way a property too can be present and absent from a subject, by simple cognition that abstracts or understands one without the other.

For this reason, some have taken a distinctive path. They say that in this definition one and the same accident must be understood as capable of being present and absent with respect to different subjects, within the full range of subjects to which it can belong. This would apply to every accident, however inseparable. For example, whiteness has many bodies as its adequate subjects—snow, stone, a human being, and so on. Outside that whole range it cannot be absent, but within it is absent from one particular subject and present in another. So Martínez, in his commentary on the chapter “On Accident.”

This way of thinking is not correct. First, Porphyry speaks of separable and inseparable accidents as distinct kinds; on this account, every accident would be inseparable from the full range of its subjects, but separable from inadequate subjects. Second, the definition would then mean that an accident is partly present and partly absent without the corruption of the subject. But with respect to which subject? It can never be absent from its total, adequate subject; and with respect to a partial subject, there is sometimes a determinate subject from which it cannot be absent, as whiteness cannot be absent from snow. This does not avoid the difficulties; it increases them. Finally, there can be an accident separable from the whole range of its subject. At least with respect to such an accident, the same difficulties arise: how should the definition “is present and absent without the corruption of the subject” be understood? The premise is clear: being grammatical is an accident and can be separated from every human being, as would be clear if there were only two human beings in the world and neither was a grammarian. Likewise, sleeping is an accident that could in such a case be separated from both, if both were awake.

Therefore, we must say: The words “is present and absent without the corruption of the subject” signify nothing other than the contingency of the accident being predicated, together with the subject’s integrity in its quiddity despite the presence or absence of that accident.

This seems to be what authors generally hold when they explain the definition of the fifth predicable as “what is said of many in what sort, contingently.” The definition therefore does not consist in being present or absent through real presence, or simply in being affirmed and denied through intellectual predication. It consists in a foundation for such negation or affirmation: an accident can belong or fail to belong to a subject without dissolving the subject’s quiddity and nature.

This resolution and explanation should be taken from St. Thomas, On the Soul, q. 2, art. 12, ad 7,³ where he resolves the whole difficulty as follows: “There are,” he says, “three kinds of accidents. Some are caused by the principles of the species and are called properties, as the capacity for laughter in a human being. Others are caused by the principles of the individual; this is said either because they have a lasting cause in the subject—these are inseparable accidents, such as being male or female and similar attributes—or because they have a cause that does not last in the subject; these are separable accidents, such as sitting and walking. It is common to every accident that it does not belong to the essence of a thing and so does not enter into the thing’s definition. We therefore understand anything without understanding one of its accidents. But a species cannot be understood without the accidents that follow from the principle of the species; it can, however, be understood without the accidents of an individual, even inseparable ones. Not only the species but also the individual can exist without separable accidents.” So the Angelic Doctor.

From this teaching we gather that the phrase “is present and absent without the corruption of the subject” means: it belongs to the subject without dissolving its quiddity and nature as expressed by essential predicates. Whether an accident is separable or inseparable is taken not with respect to the subject’s species and nature, but with respect to individuals, in which an accident sometimes has a lasting cause and sometimes a passing one. Thus an accident can belong or be absent from a subject with the corruption of the individual, yet without corruption of the species, whose integrity is preserved. This is what distinguishes it from a property: if a property is absent—that is, if it does not belong to or connect with the species—the species is dissolved. It is therefore clear that an individual can depend for its preservation and existence, indeed for its very individuation, on accidents; but the species and quiddity depend on no accident. Every affirmation or negation, every belonging or failure to belong of an accident to a subject, is therefore without corruption but with the quiddity’s integrity preserved.

Note well that “being present and absent” is not used here to mean inhering or not inhering, which belongs only to a predicamental accident. It means belonging or failing to belong accidentally. Nor do we deny that “being present and absent” signifies being affirmed or denied; it signifies not only that, but also its foundation: belonging or not belonging to the specific nature without dissolving the essential connection among its predicates. It follows that an accident can be affirmed or denied while the subject’s quiddity remains intact; in this lies the contingency with which an accident belongs or fails to belong to a subject.

Against the opposing arguments: In reply to the first, an inseparable accident cannot be present and absent through real absence and presence with respect to an individual, which is the level at which the distinction between separable and inseparable is taken. But with respect to the species or quiddity in the subject itself, it can be present or absent—that is, it can belong or fail to belong without destroying the subject in its quiddity, even though the subject is destroyed in its existence.

To the second we say: let it be granted that every accident can be separated from the species or quiddity itself; but I deny that it can be separated from the individual and its existence, which follows individuation, since an accident can have a lasting cause in the individual. In accidents, “separable” and “inseparable” are said with respect to the subject as individuated; “present and absent without the subject’s corruption” is said with respect to its quiddity. Thus “present and absent” can be understood as real, and as truly affirmed and denied without corruption of the subject on the side of quiddity, though not without corruption on the side of individuation or of the existence that follows individuation. Or, secondly, “present and absent” is taken conditionally: whether an accident is present or absent, this will happen without corruption of the subject on the side of quiddity.

If you object: Separation or belonging with respect to the species, as distinguished from the individual, is a second intention of reason, since the species is distinguished from the individual only by reason. Therefore this definition is understood in the second intention.

The reply is that by “species” we do not mean the very state of abstraction from individuals, but the specific nature itself, which is also found in individuals. Hence we use the term “quiddity” instead, saying that an accident is present and absent without corruption of the quiddity, even when it is taken individually in a subject.

You will still ask how this distinguishes accident from property, since a property too can be understood and separated from the essence without corruption of the quiddity itself, because it lies outside that definition.

The reply is drawn from St. Thomas in the cited Question on Spiritual Creatures, art. last, ad 7,⁴ where he says: “There are two operations of the intellect. By one it understands what a thing is, and by this operation the essence of a thing can be understood without a property and without an accident, since neither enters into the essence. The other operation of the intellect divides and combines; by this operation substance can be understood without an accidental predicate, even when it is inseparable in reality, as when a white raven is understood. There is no repugnance between the intellects here, since the opposite of the predicate does not depend on the principles of the species signified by the name placed as subject. But by this operation the intellect cannot understand substance without a property. For it cannot understand that a human being is not capable of laughter, or that a triangle does not have three angles equal to two right angles. There is a repugnance between the intellects here, because the opposite of the predicate depends on the nature of the subject.” So St. Thomas. From this we gather not only the solution to the argument, but also that, although a proposition denying an inseparable accident of a subject and one denying a property are both false—since reality is not as they claim—the falsity of one destroys more than the falsity of the other. Denying that an inseparable accident belongs to the subject does not destroy its quiddity; denying that a proper attribute belongs to the subject and is rooted in its nature does destroy the essence. This does not rule out God’s actually separating a proper attribute from a subject, not as to its belonging, but as to its existence.

You will say: Death is an accidental predicate with respect to animal, and yet it corrupts not only the individual but also the nature, because the concept of animal nature is incompatible with the concept of death. Therefore, there is an accident that destroys the subject even as to its quiddity. I leave aside the predicate “existing,” which is said contingently of a created being and yet is not absent without its corruption. The absence of existence, however, does not bring about the corruption of quiddity, as the conjunction with death does when death is predicated.

The reply is that the predicate “death” can be taken for death in the process of coming to be, as when a person is giving up the soul and is said to be dying. Taken in this second way, it is an accidental predicate with respect to a human being and is verified for the time when that person is approaching death; it does not remove the person’s essence, but presupposes it and disposes it toward corruption. If taken in the first way, “A human being is dead” is not a predicate of the fifth predicable, because with respect to a human being it is a repugnant predicate, while with respect to a corpse it is essential. In that proposition, therefore, the term “human being” is shifted away from the predicate “dead” and stands for a nonliving human being—which is human only equivocally—so the proposition is essential. If it is not shifted, but “human being” retains its proper meaning, then the predicate is not verified of the subject but is incompatible with it, and so does not belong to the fifth predicable. If the term is taken by amplification, so that “A human being is dead” means “One who was a human being is dead,” the proposition is also essential, because what was a human being is now a corpse, and the predicate belongs to it essentially. Some also say that the proposition is verified by synecdoche, that is, with respect to a part, because the body that is dead is the same body that belonged to the human being, with respect to matter.

FINALLY, YOU WILL ASK whether every accident is a proper attribute with respect to some subject, which it follows. Some answer yes. But it is more correct that not all accidents are proper attributes or follow from the formal principles of a nature; some follow from material and individuating principles. This is the explicit teaching of St. Thomas in chapter 7 of On Being and Essence,⁵ in Sentences I, dist. 17, q. 1, art. 2, ad 2,⁶ and in Posterior Analytics I, lect. 14.⁷ This is manifest in certain accidents that do not follow immediately on substance but on other accidents, such as immanent acts, representative species, supernatural habits, and in bodies beauty, health, sickness, and the like. These are accidents that belong contingently, not necessarily or through the principles of nature. Such accidents belong to individuals before they belong to the species. Proper attributes, by contrast, belong to the species before individuals, although with respect to inherence they are first sustained by individuals, which are first substances.

AND IF YOU SAY that everything that belongs accidentally must be reduced to what belongs per se, the reply is that it is reduced to what is per se, not as to a necessary connection with some subject from which it flows, but as to efficient or final causality, because it is not accidentally caused by some cause but caused per se. It is not necessary that everything that belongs accidentally be reduced to every mode of what is per se; it suffices that it be reduced to one, such as per se causality, even if not to a per se connection with a subject.

THREE things are briefly to be treated in this article. First, how an accidental predicate is verified of a subject: by reason of the name or by reason of the thing inhering in it. Second, with respect to which inferiors it is called the fifth predicable: with respect to its singulars, as white with respect to this and that white thing, or with respect to subjects. Third, whether the fifth predicable is predicated only with respect to existing subjects or also with respect to possible ones.

Some affirm of the first point that, when an accident is predicated of a subject, only the bare name is predicated. For the Philosopher says in chapter 5 of the Categories Before the Categories¹ that there is nothing to prevent the name of things that are in a subject from being predicated of the subject, whereas it is impossible for their account to be predicated. The reason is that predication by the substantive verb “is” states the identity of predicate and subject, but there is no such identity between a subject, which is substance, and an accident.

Nevertheless, the exact opposite must be said: in these accidental predications it is not the name itself that is predicated, but the thing signified by the name, though denominatively.

This opinion is seriously deficient, because it does not distinguish two ways in which denomination occurs: first, from the side of the name, as white from whiteness; second, from the side of the form signified, even when the names themselves do not share in it, as the name officious is derived from office, not with respect to the name but with respect to the thing signified by the effect or diligence. Hence it does not predicate the name but the thing. It is plainly evident that when I say “Peter is white,” I verify this not by reason of the name but by reason of the thing signified. For if that name is verified of Peter, it must be verified by reason of what it signifies, not by reason of the word itself; it would be ridiculous to say that Peter is the word “white.” What is signified, however, is the thing itself, namely, something that has whiteness. Therefore it is this thing that is predicated and verified in these accidental predications. This is what we mean by these predications: that those accidents are present in and belong to those subjects, not that the names alone do so.

We have said more about these matters above, question 5,² when treating the predication of concrete accidentals.

THEREFORE, IN REPLY TO THE OPPOSITE ARGUMENT, the common exposition of Aristotle is that accidents are said of a subject according to their name and not according to their account—that is, not quidditatively but denominatively—because the term account there signifies essence, and the essence of an accident does not belong to the essence of a subject. Otherwise, if they were predicated according to the name alone and not according to the thing signified, they would be common only in name and not in reality, and would therefore be equivocal.

In reply to the argument from the notion, an accident is the same in reality as its subject by denominative and concrete identity, not by essential and intrinsic identity. For, as St. Thomas says, Summa theologiae I, q. 13, a. 12, “in a true affirmative proposition the predicate and subject must signify the same thing in reality in some way” (note: in reality, not in name), because even in accidental predicates “the predicate and subject are the same in supposit.” An accident in the concrete—for example, white, since it means “having whiteness”—is the same as its subject by reason of the term having. And because predication by identity is one thing and predication by denomination another, as St. Thomas says in Sentences III, d. 5, in his exposition of the text,³ we do grant that every predication requires some identity, but denominative and accidental identity, such as is found between an accident in the concrete and a subject, suffices; complete and essential identity is not required.

Concerning the second difficulty, the inferiors of the fifth predicable are the subjects of which it is said denominatively and accidentally, not things constituted from an accident and a subject, of which it is said essentially and quidditatively. For example, white is predicated essentially of this and that white thing, but as the fifth predicable it is predicated of the subjects in which it inheres, whether those subjects differ in species or not.

THE REASON IS CLEAR, because the fifth predicable is predicated accidentally in the category of quality. Therefore its inferiors must be those to which it is related accidentally in the category of quality. It is related in this way only to subjects, since it belongs to them accidentally and in the manner of a form actualizing and qualifying them. With respect to its proper inferiors—for example, white with respect to this and that white thing—it is predicated essentially, because it expresses their entire quiddity. Since white signifies whiteness in the concrete, it can refer either to the inferiors of whiteness, which are inferiors on the side of the form, or to inferiors on the side of the subject connoted by white. If it refers to the inferiors of the form, it is said essentially of them: just as this whiteness and that whiteness essentially participate in whiteness, so this white thing and that white thing essentially participate in white on the side of the form. But if the inferiors are considered on the side of the subject, they participate accidentally in the notion of white; consequently, with respect to them it is predicated accidentally in the category of quality, which is what the fifth predicable is.

But you will object: White, for example, cannot be a universal with respect to Peter and Paul unless it abstracts from them, not insofar as they are subjects but insofar as they are white. Therefore the subjects themselves, taken in themselves, are not inferiors of this predicable.

The consequence is clear, since those are inferiors from which it abstracts. Therefore, if it abstracts not insofar as they are subjects but insofar as they are white, they will be inferiors insofar as they are white. The antecedent is proved in two ways. First, abstraction of an accident from a subject insofar as it is a subject is formal abstraction, for a form is abstracted from matter; but abstraction of a universal must be total. Second, this universal, the fifth predicable, must abstract from its inferiors insofar as they are alike or agree in something, not insofar as they agree essentially; therefore it abstracts insofar as they agree accidentally. But they agree accidentally in being white insofar as they are white; therefore white in common abstracts from them insofar as they are white.

It is further confirmed as follows: the subjects themselves are accidentally inferior to white, because Peter and Paul are accidentally identified in whiteness. Therefore, with respect to these subjects, white is a universal and accidentally predicable; hence it is not one of the species of predicables, since predicables are per se universals.

THE REPLY is that white can abstract in either way: from this and that white thing, and from its subjects. But it can abstract from subjects in two ways. One is as a physical form from a subject, and as an abstract from a concrete; the other is as a denominative term from what is denominated. If it abstracts as a form from a subject, or as an abstract from a concrete, the abstraction is formal; in that case it is not white but whiteness that is abstracted, and this is not predicated of a subject, because it is signified as a part in relation to the whole, which is white. If instead it abstracts as a denominative term from what is denominated, then white is abstracted as the fifth predicable from this or that subject, which is inferior to white on the side of what is connoted, not on the side of the form. Thus formal abstraction requires that the accident be abstracted not in the concrete but in the abstract, as a form from matter. But if white is abstracted in the concrete, it functions by way of total abstraction and, when compared with subjects, is the fifth predicable; when compared with the inferiors of the form, it is quidditative.

You will likewise understand that what is present and absent according to physical inherence is not white but whiteness; what is present and absent according to denomination is white in the concrete.

To the second proof it is said that white is not abstracted from Peter and Paul or other subjects insofar as they are formally white (for in that respect they are inferiors of the form signified by white), but insofar as they are denominable—or rather, insofar as they can be denominated—by white. Insofar as they agree in white in this way, they agree accidentally or denominatively; and white is contained in them not formally but denominatively. For they are then considered not as subjects formally white, but as denominable by white. In this way they agree in something extrinsic to them, namely, in white, and consequently are accidentally inferior to it, as to the fifth predicable.

In reply to the confirmation, the inferiors are accidentally subjects on the side of the thing that is subject; but the notion of being subject and predicable is per se a true notion of predicating and being subjected, as universality requires. Thus the notion of the fifth predicable is per se the notion of a universal and per se requires inferiors, though not inferiors that have per se necessity in the nature and form being predicated, but contingent inferiors. Accordingly, what is accidentally predicated of many has universality, or a reference to many, per se—just as it belongs per se to opinion to concern contingent propositions, and to freedom to act contingently.

Concerning the third difficulty, IT IS CERTAIN that accidental predicates, in order actually to exercise affirmative predication, require the existence of the subject as a condition.

The reason is that in accidental predications the verb “is” is not free of time; therefore the predication is verified dependently on the existence of its terms, and if the subject does not exist, it is not verified. I mean that existence is required in the relevant time: a past predication for past things and a future one for future things.

THE DIFFICULTY, HOWEVER, IS whether not only the exercise of accidental predication but also universality itself—the aptitude for accidental predication—regards only existing inferiors. Some affirm this, moved by the reason that an accident can regard non-existing subjects only as possible subjects, in which it can possibly exist. But possible existence is not predicated accidentally, but essentially, as when one says: “Adam possibly walks” or “Adam is possibly white.” Therefore the universality of the fifth predicable does not regard possible subjects.

It is confirmed as follows: with respect to non-existing subjects, predication is impossible, since it is contingent and requires existence as a condition. But if predication is impossible, universality or predicability cannot obtain.

NEVERTHELESS, the opposite seems truer: the universality of the fifth predicable is not destroyed merely by the destruction of the subjects’ existence, because universality consists in an aptitude, on the part of the nature, to be predicated of many. The fact that those many are not actually given is no intrinsic impediment to that aptitude and is not opposed to it; it is opposed only to the exercise of predication itself. What is opposed to an act or exercise does not destroy an aptitude and power: opposition to act is not opposition to potency.

This is confirmed by a physical and real accident: its separation from a subject does not deprive it of its aptitude to inhere in that subject, since this follows from the nature of the accident. Likewise, an accident considered metaphysically in the concrete does not lose its power to be predicated contingently of many, because this capacity is not removed by the lack of existence in its subjects; only their application, so that it be in them, is removed.

You will say: The notion of the concrete is removed, because every subject that it could involve is destroyed; thus only the accident in the abstract remains.

But on the contrary: only singular subjects are destroyed by the lack of existence; a subject in common can still be conceived. The concrete of an accident in common is not a singular subject, but a subject taken in common, just as white as such does not have only this or that body as the subject of its concretion, but body in common.

IN REPLY TO THE OPPOSITE ARGUMENT, an accident has only a possible relation to non-existing subjects, so that it may belong to them. However, this possible existence and relation is not what is predicated; it is a condition on the side of the predicable thing. If possibility itself were predicated, it would be predicated of the subject essentially, not accidentally. For every modal proposition, if true, is an essential proposition. The examples offered proceed in this way. We, however, say that an accident in common, with the subjects’ existence set aside, has an aptitude to be predicated contingently of them—not by predicating the possibility itself, but with respect to a form that can belong contingently to them once the subjects exist.

Thus, in a word: possibility taken as a predicate makes a proposition essential; possibility on the side of the subjects of which an accidental form can be predicated does not make the predication essential, but leaves it with its aptitude and power to be predicated contingently, although its exercise cannot occur without the subjects’ existence.

In reply to the second argument, it is said that such predication is impossible of non-existing subjects, not because the accident itself is deficient, but because the subjects’ application is deficient. A separated accident likewise retains the possibility of inhering, although it does not actually inhere or become applied while it remains in that state.